Token | his |
Token ID | 465 |
Feature activation | +2.75e-03 |
Pos logit contributions | |
older | +0.043 |
stays | +0.038 |
sle | +0.037 |
younger | +0.037 |
bars | +0.037 |
Neg logit contributions | |
J | -0.025 |
anish | -0.024 |
bed | -0.024 |
Jack | -0.024 |
iv | -0.023 |
Token | book |
Token ID | 1492 |
Feature activation | +2.64e-02 |
Pos logit contributions | |
older | +0.005 |
heavier | +0.005 |
bars | +0.004 |
asks | +0.004 |
stays | +0.004 |
Neg logit contributions | |
anish | -0.004 |
iv | -0.004 |
iolet | -0.004 |
And | -0.004 |
bed | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | +4.39e-02 |
Pos logit contributions | |
older | +0.044 |
asks | +0.038 |
sle | +0.038 |
stays | +0.038 |
swe | +0.036 |
Neg logit contributions | |
anish | -0.046 |
bed | -0.043 |
room | -0.043 |
oon | -0.043 |
J | -0.042 |
Token | his |
Token ID | 465 |
Feature activation | +4.56e-03 |
Pos logit contributions | |
older | +0.065 |
heavier | +0.052 |
gs | +0.051 |
younger | +0.048 |
forward | +0.047 |
Neg logit contributions | |
iv | -0.076 |
anish | -0.076 |
And | -0.076 |
bed | -0.074 |
der | -0.069 |
Token | bin |
Token ID | 9874 |
Feature activation | +3.54e-02 |
Pos logit contributions | |
older | +0.008 |
heavier | +0.007 |
stays | +0.007 |
swe | +0.007 |
goes | +0.006 |
Neg logit contributions | |
anish | -0.008 |
bed | -0.007 |
iv | -0.007 |
And | -0.007 |
iolet | -0.007 |
Token | ocular |
Token ID | 37320 |
Feature activation | +8.88e-02 |
Pos logit contributions | |
older | +0.071 |
asks | +0.062 |
heavier | +0.060 |
swe | +0.059 |
younger | +0.059 |
Neg logit contributions | |
bed | -0.049 |
room | -0.047 |
anish | -0.046 |
edly | -0.044 |
iv | -0.043 |
Token | s |
Token ID | 82 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
older | +0.196 |
younger | +0.176 |
swe | +0.164 |
heavier | +0.162 |
brav | +0.154 |
Neg logit contributions | |
he | -0.094 |
room | -0.093 |
od | -0.090 |
were | -0.090 |
fully | -0.086 |
Token | in |
Token ID | 287 |
Feature activation | +1.84e-03 |
Pos logit contributions | |
older | +0.048 |
heavier | +0.041 |
swe | +0.040 |
sle | +0.040 |
younger | +0.040 |
Neg logit contributions | |
anish | -0.029 |
iv | -0.028 |
bed | -0.027 |
od | -0.026 |
And | -0.026 |
Token | his |
Token ID | 465 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
older | +0.003 |
stays | +0.003 |
swe | +0.002 |
heavier | +0.002 |
climbs | +0.002 |
Neg logit contributions | |
anish | -0.003 |
iv | -0.003 |
J | -0.003 |
Jack | -0.003 |
od | -0.003 |
Token | backpack |
Token ID | 22526 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
anish | +0.020 |
And | +0.019 |
iv | +0.018 |
bed | +0.018 |
room | +0.018 |
Neg logit contributions | |
older | -0.020 |
heavier | -0.017 |
bars | -0.015 |
stays | -0.015 |
younger | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
older | +0.044 |
swe | +0.038 |
stays | +0.038 |
heavier | +0.038 |
sle | +0.036 |
Neg logit contributions | |
anish | -0.027 |
iv | -0.026 |
od | -0.025 |
room | -0.024 |
oon | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +7.68e-03 |
Pos logit contributions | |
older | +0.015 |
swe | +0.013 |
heavier | +0.013 |
stays | +0.013 |
bigger | +0.012 |
Neg logit contributions | |
anish | -0.012 |
iv | -0.012 |
od | -0.012 |
iolet | -0.011 |
oon | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
older | +0.015 |
swe | +0.013 |
stays | +0.013 |
heavier | +0.013 |
sle | +0.012 |
Neg logit contributions | |
anish | -0.012 |
iv | -0.012 |
od | -0.012 |
And | -0.011 |
Jack | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +2.39e-03 |
Pos logit contributions | |
older | +0.042 |
heavier | +0.035 |
swe | +0.035 |
stays | +0.035 |
bigger | +0.035 |
Neg logit contributions | |
anish | -0.029 |
iv | -0.028 |
od | -0.027 |
iolet | -0.026 |
bed | -0.025 |
Token | " |
Token ID | 1 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
older | +0.005 |
heavier | +0.004 |
swe | +0.004 |
stays | +0.004 |
bigger | +0.004 |
Neg logit contributions | |
anish | -0.004 |
iv | -0.004 |
od | -0.004 |
iolet | -0.003 |
bed | -0.003 |
Token | O |
Token ID | 46 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
od | +0.022 |
anish | +0.021 |
iv | +0.021 |
bed | +0.019 |
oon | +0.019 |
Neg logit contributions | |
older | -0.035 |
swe | -0.031 |
goes | -0.031 |
lower | -0.031 |
heavier | -0.031 |
Token | ink |
Token ID | 676 |
Feature activation | +8.70e-02 |
Pos logit contributions | |
older | +0.038 |
stays | +0.034 |
heavier | +0.034 |
goes | +0.033 |
gs | +0.032 |
Neg logit contributions | |
anish | -0.027 |
iv | -0.026 |
od | -0.026 |
Sus | -0.024 |
And | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -1.42e-03 |
Pos logit contributions | |
older | +0.154 |
swe | +0.136 |
heavier | +0.133 |
sle | +0.127 |
younger | +0.125 |
Neg logit contributions | |
anish | -0.157 |
iv | -0.150 |
od | -0.150 |
oon | -0.146 |
iolet | -0.135 |
Token | o |
Token ID | 267 |
Feature activation | +4.41e-02 |
Pos logit contributions | |
iv | +0.002 |
od | +0.002 |
anish | +0.002 |
And | +0.002 |
Jack | +0.002 |
Neg logit contributions | |
older | -0.003 |
swe | -0.002 |
younger | -0.002 |
stays | -0.002 |
heavier | -0.002 |
Token | ink |
Token ID | 676 |
Feature activation | +8.25e-02 |
Pos logit contributions | |
older | +0.100 |
swe | +0.088 |
heavier | +0.084 |
younger | +0.083 |
stays | +0.083 |
Neg logit contributions | |
anish | -0.063 |
iv | -0.060 |
od | -0.057 |
iolet | -0.055 |
oon | -0.055 |
Token | ," |
Token ID | 553 |
Feature activation | -4.25e-03 |
Pos logit contributions | |
older | +0.169 |
swe | +0.149 |
heavier | +0.148 |
younger | +0.144 |
bars | +0.142 |
Neg logit contributions | |
iv | -0.119 |
anish | -0.119 |
oon | -0.119 |
od | -0.117 |
iolet | -0.099 |
Token | Tom |
Token ID | 4186 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
iv | +0.006 |
anish | +0.006 |
od | +0.006 |
Jack | +0.006 |
J | +0.006 |
Neg logit contributions | |
older | -0.009 |
stays | -0.007 |
swe | -0.007 |
heavier | -0.007 |
lower | -0.007 |
Token | baby |
Token ID | 5156 |
Feature activation | -3.50e-02 |
Pos logit contributions | |
older | +0.014 |
heavier | +0.011 |
swe | +0.010 |
stays | +0.010 |
bigger | +0.010 |
Neg logit contributions | |
anish | -0.019 |
iv | -0.019 |
od | -0.018 |
iolet | -0.018 |
Jack | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
anish | +0.054 |
iv | +0.053 |
od | +0.052 |
iolet | +0.049 |
And | +0.047 |
Neg logit contributions | |
older | -0.069 |
stays | -0.059 |
heavier | -0.059 |
swe | -0.058 |
lower | -0.056 |
Token | I |
Token ID | 314 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
older | +0.032 |
heavier | +0.027 |
swe | +0.026 |
stays | +0.026 |
bigger | +0.025 |
Neg logit contributions | |
anish | -0.028 |
iv | -0.027 |
od | -0.026 |
iolet | -0.025 |
And | -0.025 |
Token | am |
Token ID | 716 |
Feature activation | -1.40e-03 |
Pos logit contributions | |
anish | +0.022 |
iv | +0.021 |
od | +0.021 |
iolet | +0.020 |
And | +0.020 |
Neg logit contributions | |
older | -0.025 |
heavier | -0.020 |
swe | -0.020 |
sle | -0.020 |
bigger | -0.019 |
Token | three |
Token ID | 1115 |
Feature activation | +2.00e-03 |
Pos logit contributions | |
od | +0.002 |
iv | +0.002 |
And | +0.002 |
iolet | +0.002 |
oon | +0.002 |
Neg logit contributions | |
bars | -0.002 |
stays | -0.002 |
sle | -0.002 |
older | -0.002 |
gs | -0.002 |
Token | years |
Token ID | 812 |
Feature activation | +8.55e-02 |
Pos logit contributions | |
older | +0.004 |
stays | +0.004 |
heavier | +0.004 |
lower | +0.004 |
swe | +0.004 |
Neg logit contributions | |
iv | -0.003 |
od | -0.003 |
anish | -0.003 |
And | -0.002 |
oon | -0.002 |
Token | old |
Token ID | 1468 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
stays | +0.092 |
climbs | +0.079 |
reaches | +0.075 |
brakes | +0.074 |
carries | +0.073 |
Neg logit contributions | |
iv | -0.174 |
od | -0.170 |
Sus | -0.167 |
Em | -0.161 |
oon | -0.161 |
Token | . |
Token ID | 13 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
anish | +0.020 |
iv | +0.019 |
od | +0.019 |
iolet | +0.018 |
And | +0.018 |
Neg logit contributions | |
older | -0.023 |
heavier | -0.020 |
swe | -0.020 |
stays | -0.020 |
sle | -0.019 |
Token | You |
Token ID | 921 |
Feature activation | +2.40e-03 |
Pos logit contributions | |
older | +0.041 |
heavier | +0.035 |
swe | +0.034 |
stays | +0.034 |
cooler | +0.033 |
Neg logit contributions | |
anish | -0.035 |
iv | -0.034 |
od | -0.032 |
And | -0.032 |
iolet | -0.031 |
Token | are |
Token ID | 389 |
Feature activation | +5.74e-04 |
Pos logit contributions | |
older | +0.005 |
swe | +0.004 |
heavier | +0.004 |
sle | +0.004 |
bigger | +0.004 |
Neg logit contributions | |
anish | -0.004 |
iv | -0.004 |
od | -0.004 |
iolet | -0.004 |
And | -0.004 |
Token | mean |
Token ID | 1612 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
older | +0.001 |
bars | +0.001 |
sle | +0.001 |
stays | +0.001 |
gs | +0.001 |
Neg logit contributions | |
iv | -0.001 |
od | -0.001 |
anish | -0.001 |
iolet | -0.001 |
And | -0.001 |
Token | asked |
Token ID | 1965 |
Feature activation | +8.06e-03 |
Pos logit contributions | |
older | +0.013 |
stays | +0.011 |
heavier | +0.011 |
sle | +0.011 |
swe | +0.011 |
Neg logit contributions | |
iv | -0.014 |
od | -0.014 |
anish | -0.014 |
Jack | -0.014 |
iolet | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
older | +0.017 |
sle | +0.017 |
stays | +0.016 |
stable | +0.016 |
reins | +0.015 |
Neg logit contributions | |
J | -0.010 |
anish | -0.010 |
od | -0.009 |
iv | -0.009 |
Jane | -0.009 |
Token | He |
Token ID | 679 |
Feature activation | +2.31e-02 |
Pos logit contributions | |
older | +0.070 |
heavier | +0.059 |
swe | +0.059 |
stays | +0.059 |
sle | +0.056 |
Neg logit contributions | |
anish | -0.051 |
iv | -0.049 |
od | -0.048 |
And | -0.046 |
iolet | -0.046 |
Token | was |
Token ID | 373 |
Feature activation | -2.09e-03 |
Pos logit contributions | |
older | +0.046 |
sle | +0.039 |
heavier | +0.039 |
swe | +0.038 |
stays | +0.037 |
Neg logit contributions | |
iv | -0.035 |
anish | -0.034 |
od | -0.033 |
Jack | -0.033 |
And | -0.033 |
Token | three |
Token ID | 1115 |
Feature activation | -4.09e-03 |
Pos logit contributions | |
iv | +0.003 |
Jane | +0.003 |
And | +0.003 |
bed | +0.003 |
J | +0.003 |
Neg logit contributions | |
sle | -0.003 |
older | -0.003 |
goes | -0.003 |
stays | -0.003 |
passes | -0.003 |
Token | years |
Token ID | 812 |
Feature activation | +8.39e-02 |
Pos logit contributions | |
iv | +0.005 |
od | +0.005 |
anish | +0.005 |
And | +0.004 |
iolet | +0.004 |
Neg logit contributions | |
older | -0.009 |
stays | -0.008 |
heavier | -0.008 |
lower | -0.008 |
sle | -0.008 |
Token | old |
Token ID | 1468 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
stays | +0.107 |
passes | +0.088 |
leads | +0.087 |
fastest | +0.087 |
ishes | +0.086 |
Neg logit contributions | |
Sus | -0.160 |
edly | -0.155 |
Jane | -0.154 |
J | -0.152 |
iv | -0.148 |
Token | and |
Token ID | 290 |
Feature activation | +4.66e-02 |
Pos logit contributions | |
anish | +0.030 |
iv | +0.029 |
od | +0.029 |
iolet | +0.027 |
And | +0.027 |
Neg logit contributions | |
older | -0.033 |
swe | -0.028 |
heavier | -0.028 |
stays | -0.027 |
bigger | -0.026 |
Token | liked |
Token ID | 8288 |
Feature activation | +7.68e-03 |
Pos logit contributions | |
older | +0.097 |
sle | +0.085 |
swe | +0.081 |
heavier | +0.080 |
stays | +0.080 |
Neg logit contributions | |
anish | -0.060 |
iv | -0.060 |
Jack | -0.059 |
And | -0.059 |
bed | -0.058 |
Token | to |
Token ID | 284 |
Feature activation | +5.70e-03 |
Pos logit contributions | |
older | +0.014 |
stays | +0.012 |
heavier | +0.012 |
sle | +0.011 |
lower | +0.011 |
Neg logit contributions | |
od | -0.013 |
anish | -0.013 |
iv | -0.012 |
Jack | -0.012 |
J | -0.012 |
Token | copy |
Token ID | 4866 |
Feature activation | -6.10e-03 |
Pos logit contributions | |
older | +0.010 |
heavier | +0.009 |
swe | +0.008 |
stays | +0.008 |
sle | +0.008 |
Neg logit contributions | |
anish | -0.010 |
iv | -0.010 |
od | -0.009 |
iolet | -0.009 |
Jack | -0.009 |
Token | Ben |
Token ID | 3932 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
older | +0.004 |
stays | +0.004 |
swe | +0.004 |
heavier | +0.004 |
goes | +0.004 |
Neg logit contributions | |
od | -0.005 |
iv | -0.005 |
anish | -0.005 |
Sus | -0.004 |
Jack | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
older | +0.029 |
climbs | +0.026 |
brakes | +0.026 |
stays | +0.026 |
heavier | +0.026 |
Neg logit contributions | |
iv | -0.011 |
And | -0.010 |
od | -0.010 |
iolet | -0.010 |
anish | -0.010 |
Token | I |
Token ID | 314 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
older | +0.040 |
heavier | +0.033 |
swe | +0.033 |
stays | +0.032 |
bigger | +0.031 |
Neg logit contributions | |
anish | -0.037 |
iv | -0.035 |
od | -0.034 |
And | -0.033 |
Jack | -0.033 |
Token | 'm |
Token ID | 1101 |
Feature activation | +3.96e-03 |
Pos logit contributions | |
anish | +0.025 |
iv | +0.024 |
od | +0.024 |
iolet | +0.023 |
Jack | +0.022 |
Neg logit contributions | |
older | -0.033 |
heavier | -0.027 |
swe | -0.027 |
stays | -0.027 |
bigger | -0.026 |
Token | three |
Token ID | 1115 |
Feature activation | +2.45e-03 |
Pos logit contributions | |
older | +0.007 |
stays | +0.006 |
swe | +0.006 |
goes | +0.006 |
heavier | +0.006 |
Neg logit contributions | |
od | -0.006 |
iv | -0.006 |
Sus | -0.006 |
Jack | -0.006 |
anish | -0.006 |
Token | years |
Token ID | 812 |
Feature activation | +8.30e-02 |
Pos logit contributions | |
older | +0.005 |
stays | +0.005 |
heavier | +0.005 |
lower | +0.005 |
swe | +0.005 |
Neg logit contributions | |
iv | -0.003 |
od | -0.003 |
anish | -0.003 |
And | -0.003 |
iolet | -0.003 |
Token | old |
Token ID | 1468 |
Feature activation | -1.65e-02 |
Pos logit contributions | |
stays | +0.092 |
brakes | +0.071 |
climbs | +0.070 |
leads | +0.069 |
ches | +0.069 |
Neg logit contributions | |
Sus | -0.180 |
iv | -0.171 |
od | -0.168 |
J | -0.167 |
Jane | -0.167 |
Token | , |
Token ID | 11 |
Feature activation | -1.49e-03 |
Pos logit contributions | |
anish | +0.026 |
iv | +0.026 |
od | +0.024 |
iolet | +0.024 |
And | +0.023 |
Neg logit contributions | |
older | -0.033 |
heavier | -0.028 |
stays | -0.028 |
swe | -0.027 |
lower | -0.026 |
Token | too |
Token ID | 1165 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
anish | +0.002 |
And | +0.002 |
iv | +0.002 |
od | +0.002 |
Jack | +0.002 |
Neg logit contributions | |
older | -0.003 |
stays | -0.003 |
heavier | -0.003 |
swe | -0.002 |
gs | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +2.55e-02 |
Pos logit contributions | |
older | +0.057 |
stays | +0.053 |
swe | +0.051 |
heavier | +0.050 |
lower | +0.048 |
Neg logit contributions | |
iv | -0.037 |
od | -0.036 |
anish | -0.034 |
And | -0.032 |
oon | -0.032 |
Token | I |
Token ID | 314 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
older | +0.049 |
heavier | +0.042 |
swe | +0.041 |
stays | +0.040 |
bigger | +0.039 |
Neg logit contributions | |
anish | -0.042 |
iv | -0.040 |
And | -0.039 |
od | -0.038 |
Jack | -0.038 |
Token | " |
Token ID | 366 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
older | +0.027 |
stays | +0.025 |
sle | +0.025 |
swe | +0.024 |
heavier | +0.024 |
Neg logit contributions | |
anish | -0.025 |
iv | -0.024 |
iolet | -0.024 |
J | -0.024 |
od | -0.024 |
Token | M |
Token ID | 44 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
anish | +0.028 |
od | +0.026 |
bed | +0.026 |
iv | +0.026 |
iolet | +0.025 |
Neg logit contributions | |
older | -0.038 |
lower | -0.035 |
heavier | -0.034 |
bigger | -0.033 |
swe | -0.033 |
Token | oo |
Token ID | 2238 |
Feature activation | +3.97e-02 |
Pos logit contributions | |
older | +0.087 |
swe | +0.079 |
heavier | +0.079 |
stays | +0.078 |
lower | +0.077 |
Neg logit contributions | |
iv | -0.032 |
anish | -0.031 |
od | -0.028 |
iolet | -0.027 |
And | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | +2.62e-03 |
Pos logit contributions | |
older | +0.073 |
heavier | +0.060 |
swe | +0.058 |
sle | +0.057 |
stays | +0.056 |
Neg logit contributions | |
anish | -0.071 |
iv | -0.070 |
od | -0.068 |
oon | -0.064 |
iolet | -0.064 |
Token | o |
Token ID | 267 |
Feature activation | +3.71e-02 |
Pos logit contributions | |
older | +0.005 |
heavier | +0.004 |
swe | +0.004 |
lower | +0.004 |
gs | +0.004 |
Neg logit contributions | |
anish | -0.004 |
iv | -0.004 |
od | -0.004 |
Jack | -0.004 |
And | -0.004 |
Token | ink |
Token ID | 676 |
Feature activation | +8.29e-02 |
Pos logit contributions | |
older | +0.078 |
heavier | +0.066 |
swe | +0.066 |
stays | +0.065 |
bigger | +0.063 |
Neg logit contributions | |
anish | -0.055 |
iv | -0.053 |
od | -0.052 |
iolet | -0.049 |
And | -0.049 |
Token | , |
Token ID | 11 |
Feature activation | +2.06e-03 |
Pos logit contributions | |
older | +0.160 |
heavier | +0.143 |
swe | +0.138 |
sle | +0.135 |
younger | +0.133 |
Neg logit contributions | |
anish | -0.132 |
od | -0.129 |
oon | -0.127 |
iv | -0.127 |
iolet | -0.111 |
Token | b |
Token ID | 275 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
older | +0.004 |
swe | +0.003 |
heavier | +0.003 |
stays | +0.003 |
younger | +0.003 |
Neg logit contributions | |
anish | -0.003 |
iv | -0.003 |
od | -0.003 |
And | -0.003 |
oon | -0.003 |
Token | aa |
Token ID | 7252 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
older | +0.063 |
swe | +0.051 |
heavier | +0.050 |
bigger | +0.048 |
stays | +0.048 |
Neg logit contributions | |
iv | -0.090 |
od | -0.087 |
anish | -0.087 |
iolet | -0.083 |
oon | -0.079 |
Token | , |
Token ID | 11 |
Feature activation | +2.07e-03 |
Pos logit contributions | |
older | +0.043 |
heavier | +0.036 |
swe | +0.036 |
stays | +0.034 |
cooler | +0.034 |
Neg logit contributions | |
anish | -0.035 |
iv | -0.035 |
od | -0.033 |
iolet | -0.032 |
oon | -0.031 |
Token | cl |
Token ID | 537 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
older | +0.004 |
heavier | +0.003 |
swe | +0.003 |
younger | +0.003 |
stays | +0.003 |
Neg logit contributions | |
anish | -0.003 |
iv | -0.003 |
od | -0.003 |
And | -0.003 |
oon | -0.003 |
Token | : |
Token ID | 25 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
older | +0.027 |
heavier | +0.022 |
stays | +0.022 |
swe | +0.022 |
sle | +0.022 |
Neg logit contributions | |
anish | -0.025 |
od | -0.024 |
iv | -0.024 |
And | -0.022 |
iolet | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
older | +0.020 |
heavier | +0.017 |
swe | +0.017 |
stays | +0.016 |
bigger | +0.016 |
Neg logit contributions | |
anish | -0.019 |
iv | -0.018 |
od | -0.018 |
iolet | -0.017 |
Jack | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | +4.06e-03 |
Pos logit contributions | |
older | +0.033 |
heavier | +0.028 |
swe | +0.027 |
stays | +0.027 |
bigger | +0.027 |
Neg logit contributions | |
anish | -0.022 |
iv | -0.021 |
od | -0.021 |
iolet | -0.020 |
oon | -0.019 |
Token | " |
Token ID | 1 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
older | +0.007 |
heavier | +0.006 |
swe | +0.006 |
stays | +0.006 |
bigger | +0.006 |
Neg logit contributions | |
anish | -0.007 |
iv | -0.007 |
od | -0.007 |
iolet | -0.007 |
And | -0.007 |
Token | Dr |
Token ID | 6187 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
od | +0.034 |
anish | +0.032 |
bed | +0.031 |
iv | +0.030 |
oon | +0.029 |
Neg logit contributions | |
older | -0.043 |
swe | -0.039 |
lower | -0.037 |
bigger | -0.036 |
heavier | -0.036 |
Token | ink |
Token ID | 676 |
Feature activation | +8.26e-02 |
Pos logit contributions | |
older | +0.060 |
heavier | +0.049 |
swe | +0.049 |
stays | +0.048 |
bigger | +0.046 |
Neg logit contributions | |
anish | -0.056 |
iv | -0.054 |
od | -0.053 |
iolet | -0.051 |
And | -0.050 |
Token | some |
Token ID | 617 |
Feature activation | +3.63e-02 |
Pos logit contributions | |
older | +0.147 |
heavier | +0.134 |
swe | +0.129 |
cooler | +0.120 |
lower | +0.120 |
Neg logit contributions | |
anish | -0.145 |
iv | -0.140 |
od | -0.137 |
oon | -0.136 |
iolet | -0.126 |
Token | water |
Token ID | 1660 |
Feature activation | +5.23e-02 |
Pos logit contributions | |
older | +0.064 |
heavier | +0.055 |
gs | +0.054 |
stays | +0.053 |
swe | +0.053 |
Neg logit contributions | |
anish | -0.058 |
iv | -0.058 |
And | -0.052 |
od | -0.051 |
wards | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
older | +0.098 |
heavier | +0.085 |
swe | +0.082 |
younger | +0.080 |
stays | +0.078 |
Neg logit contributions | |
anish | -0.087 |
iv | -0.085 |
od | -0.079 |
oon | -0.078 |
room | -0.076 |
Token | It |
Token ID | 632 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
older | +0.040 |
swe | +0.034 |
heavier | +0.034 |
stays | +0.033 |
bigger | +0.032 |
Neg logit contributions | |
anish | -0.030 |
iv | -0.029 |
od | -0.028 |
And | -0.027 |
iolet | -0.026 |
Token | will |
Token ID | 481 |
Feature activation | +7.47e-04 |
Pos logit contributions | |
older | +0.042 |
reins | +0.034 |
heavier | +0.034 |
sle | +0.034 |
stable | +0.032 |
Neg logit contributions | |
anish | -0.050 |
iv | -0.050 |
od | -0.048 |
And | -0.047 |
oon | -0.047 |
Token | " |
Token ID | 1 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
older | +0.005 |
heavier | +0.004 |
swe | +0.004 |
stays | +0.004 |
bigger | +0.004 |
Neg logit contributions | |
anish | -0.004 |
iv | -0.004 |
od | -0.004 |
iolet | -0.003 |
bed | -0.003 |
Token | O |
Token ID | 46 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
od | +0.022 |
anish | +0.021 |
iv | +0.021 |
bed | +0.019 |
oon | +0.019 |
Neg logit contributions | |
older | -0.035 |
swe | -0.031 |
goes | -0.031 |
lower | -0.031 |
heavier | -0.031 |
Token | ink |
Token ID | 676 |
Feature activation | +8.70e-02 |
Pos logit contributions | |
older | +0.038 |
stays | +0.034 |
heavier | +0.034 |
goes | +0.033 |
gs | +0.032 |
Neg logit contributions | |
anish | -0.027 |
iv | -0.026 |
od | -0.026 |
Sus | -0.024 |
And | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -1.42e-03 |
Pos logit contributions | |
older | +0.154 |
swe | +0.136 |
heavier | +0.133 |
sle | +0.127 |
younger | +0.125 |
Neg logit contributions | |
anish | -0.157 |
iv | -0.150 |
od | -0.150 |
oon | -0.146 |
iolet | -0.135 |
Token | o |
Token ID | 267 |
Feature activation | +4.41e-02 |
Pos logit contributions | |
iv | +0.002 |
od | +0.002 |
anish | +0.002 |
And | +0.002 |
Jack | +0.002 |
Neg logit contributions | |
older | -0.003 |
swe | -0.002 |
younger | -0.002 |
stays | -0.002 |
heavier | -0.002 |
Token | ink |
Token ID | 676 |
Feature activation | +8.25e-02 |
Pos logit contributions | |
older | +0.100 |
swe | +0.088 |
heavier | +0.084 |
younger | +0.083 |
stays | +0.083 |
Neg logit contributions | |
anish | -0.063 |
iv | -0.060 |
od | -0.057 |
iolet | -0.055 |
oon | -0.055 |
Token | ," |
Token ID | 553 |
Feature activation | -4.25e-03 |
Pos logit contributions | |
older | +0.169 |
swe | +0.149 |
heavier | +0.148 |
younger | +0.144 |
bars | +0.142 |
Neg logit contributions | |
iv | -0.119 |
anish | -0.119 |
oon | -0.119 |
od | -0.117 |
iolet | -0.099 |
Token | Tom |
Token ID | 4186 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
iv | +0.006 |
anish | +0.006 |
od | +0.006 |
Jack | +0.006 |
J | +0.006 |
Neg logit contributions | |
older | -0.009 |
stays | -0.007 |
swe | -0.007 |
heavier | -0.007 |
lower | -0.007 |
Token | says |
Token ID | 1139 |
Feature activation | -5.63e-04 |
Pos logit contributions | |
older | +0.021 |
heavier | +0.019 |
swe | +0.017 |
sle | +0.017 |
stable | +0.016 |
Neg logit contributions | |
iv | -0.040 |
od | -0.039 |
anish | -0.039 |
iolet | -0.037 |
And | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
anish | +0.001 |
od | +0.001 |
iv | +0.001 |
bed | +0.001 |
iolet | +0.001 |
Neg logit contributions | |
older | -0.001 |
stays | -0.001 |
sle | -0.001 |
swe | -0.001 |
heavier | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
older | +0.054 |
stays | +0.049 |
swe | +0.048 |
heavier | +0.046 |
sle | +0.045 |
Neg logit contributions | |
anish | -0.038 |
And | -0.038 |
iv | -0.037 |
od | -0.036 |
J | -0.036 |
Token | " |
Token ID | 366 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
anish | +0.001 |
od | +0.001 |
iv | +0.001 |
Jack | +0.001 |
And | +0.001 |
Neg logit contributions | |
older | -0.001 |
stays | -0.001 |
heavier | -0.001 |
swe | -0.001 |
goes | -0.001 |
Token | But |
Token ID | 1537 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
anish | +0.023 |
iv | +0.023 |
od | +0.022 |
iolet | +0.020 |
oon | +0.020 |
Neg logit contributions | |
older | -0.035 |
swe | -0.031 |
lower | -0.030 |
stays | -0.030 |
heavier | -0.030 |
Token | it |
Token ID | 340 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
older | +0.042 |
swe | +0.036 |
stays | +0.036 |
heavier | +0.035 |
sle | +0.035 |
Neg logit contributions | |
anish | -0.036 |
iv | -0.035 |
od | -0.033 |
bed | -0.032 |
Jack | -0.031 |
Token | looks |
Token ID | 3073 |
Feature activation | +4.06e-02 |
Pos logit contributions | |
older | +0.043 |
reins | +0.040 |
sle | +0.039 |
heavier | +0.036 |
stable | +0.036 |
Neg logit contributions | |
anish | -0.042 |
od | -0.041 |
iv | -0.041 |
oon | -0.038 |
bed | -0.035 |
Token | so |
Token ID | 523 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
older | +0.077 |
bars | +0.076 |
sle | +0.076 |
reins | +0.076 |
climbs | +0.072 |
Neg logit contributions | |
And | -0.048 |
iv | -0.047 |
Jack | -0.047 |
od | -0.047 |
anish | -0.044 |
Token | com |
Token ID | 401 |
Feature activation | +8.21e-02 |
Pos logit contributions | |
older | +0.068 |
reins | +0.067 |
sle | +0.062 |
gs | +0.060 |
brakes | +0.056 |
Neg logit contributions | |
Jack | -0.061 |
iv | -0.061 |
Sus | -0.058 |
Jane | -0.058 |
anish | -0.057 |
Token | fy |
Token ID | 24928 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
older | +0.120 |
sle | +0.089 |
younger | +0.088 |
stable | +0.087 |
reins | +0.082 |
Neg logit contributions | |
iv | -0.166 |
anish | -0.156 |
And | -0.155 |
bed | -0.155 |
od | -0.155 |
Token | and |
Token ID | 290 |
Feature activation | +3.59e-02 |
Pos logit contributions | |
older | +0.046 |
stays | +0.039 |
swe | +0.038 |
heavier | +0.038 |
younger | +0.037 |
Neg logit contributions | |
iv | -0.035 |
anish | -0.035 |
od | -0.032 |
oon | -0.032 |
Jack | -0.032 |
Token | sweet |
Token ID | 6029 |
Feature activation | -6.42e-03 |
Pos logit contributions | |
older | +0.068 |
gs | +0.057 |
reins | +0.057 |
sle | +0.056 |
younger | +0.054 |
Neg logit contributions | |
Sus | -0.046 |
Jack | -0.045 |
iv | -0.044 |
Jane | -0.044 |
bed | -0.043 |
Token | ". |
Token ID | 1911 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
iv | +0.009 |
anish | +0.009 |
od | +0.009 |
iolet | +0.008 |
Jane | +0.008 |
Neg logit contributions | |
older | -0.014 |
heavier | -0.011 |
stays | -0.011 |
younger | -0.011 |
sle | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +2.58e-03 |
Pos logit contributions | |
anish | +0.020 |
iv | +0.019 |
od | +0.019 |
oon | +0.018 |
iolet | +0.018 |
Neg logit contributions | |
older | -0.024 |
swe | -0.020 |
heavier | -0.019 |
stays | -0.019 |
cooler | -0.019 |
Token | Anna |
Token ID | 11735 |
Feature activation | +4.64e-03 |
Pos logit contributions | |
od | +0.008 |
iv | +0.008 |
anish | +0.007 |
iolet | +0.007 |
Jack | +0.007 |
Neg logit contributions | |
older | -0.010 |
stays | -0.009 |
heavier | -0.009 |
lower | -0.008 |
swe | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
older | +0.011 |
heavier | +0.010 |
lower | +0.010 |
brakes | +0.009 |
gs | +0.009 |
Neg logit contributions | |
iv | -0.005 |
od | -0.005 |
And | -0.005 |
iolet | -0.005 |
oon | -0.005 |
Token | I |
Token ID | 314 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
older | +0.033 |
heavier | +0.028 |
swe | +0.027 |
stays | +0.027 |
bigger | +0.026 |
Neg logit contributions | |
anish | -0.027 |
iv | -0.026 |
od | -0.025 |
And | -0.025 |
iolet | -0.024 |
Token | am |
Token ID | 716 |
Feature activation | -3.86e-03 |
Pos logit contributions | |
anish | +0.021 |
iv | +0.020 |
od | +0.020 |
And | +0.019 |
iolet | +0.019 |
Neg logit contributions | |
older | -0.030 |
heavier | -0.025 |
swe | -0.025 |
sle | -0.024 |
bigger | -0.024 |
Token | three |
Token ID | 1115 |
Feature activation | -6.28e-03 |
Pos logit contributions | |
od | +0.006 |
iv | +0.006 |
Jack | +0.006 |
Sus | +0.006 |
And | +0.006 |
Neg logit contributions | |
older | -0.007 |
stays | -0.007 |
bars | -0.006 |
goes | -0.006 |
swe | -0.006 |
Token | years |
Token ID | 812 |
Feature activation | +8.19e-02 |
Pos logit contributions | |
iv | +0.007 |
od | +0.007 |
And | +0.007 |
anish | +0.007 |
Jane | +0.006 |
Neg logit contributions | |
older | -0.014 |
lower | -0.013 |
stays | -0.013 |
heavier | -0.013 |
swe | -0.012 |
Token | old |
Token ID | 1468 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
stays | +0.096 |
climbs | +0.078 |
brakes | +0.078 |
reaches | +0.075 |
leads | +0.074 |
Neg logit contributions | |
Sus | -0.164 |
And | -0.159 |
iv | -0.159 |
od | -0.158 |
Em | -0.153 |
Token | . |
Token ID | 13 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
iv | +0.034 |
anish | +0.034 |
od | +0.032 |
iolet | +0.031 |
And | +0.031 |
Neg logit contributions | |
older | -0.045 |
heavier | -0.039 |
stays | -0.039 |
swe | -0.039 |
lower | -0.038 |
Token | I |
Token ID | 314 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
older | +0.039 |
heavier | +0.033 |
swe | +0.032 |
stays | +0.032 |
bigger | +0.031 |
Neg logit contributions | |
anish | -0.033 |
iv | -0.032 |
od | -0.031 |
And | -0.031 |
iolet | -0.030 |
Token | like |
Token ID | 588 |
Feature activation | +6.78e-03 |
Pos logit contributions | |
iv | +0.019 |
anish | +0.019 |
od | +0.018 |
And | +0.018 |
iolet | +0.017 |
Neg logit contributions | |
older | -0.026 |
heavier | -0.022 |
swe | -0.022 |
sle | -0.021 |
younger | -0.021 |
Token | to |
Token ID | 284 |
Feature activation | -5.17e-03 |
Pos logit contributions | |
older | +0.011 |
heavier | +0.010 |
lower | +0.010 |
gs | +0.010 |
stays | +0.010 |
Neg logit contributions | |
od | -0.011 |
iv | -0.011 |
anish | -0.011 |
And | -0.011 |
Jack | -0.010 |
Token | his |
Token ID | 465 |
Feature activation | +3.92e-03 |
Pos logit contributions | |
older | +0.002 |
heavier | +0.002 |
stays | +0.002 |
shaky | +0.002 |
bars | +0.002 |
Neg logit contributions | |
od | -0.002 |
Jack | -0.001 |
J | -0.001 |
Jane | -0.001 |
Joe | -0.001 |
Token | book |
Token ID | 1492 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
older | +0.007 |
heavier | +0.006 |
swe | +0.006 |
stays | +0.006 |
lower | +0.006 |
Neg logit contributions | |
anish | -0.006 |
od | -0.006 |
iv | -0.006 |
iolet | -0.006 |
And | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +3.43e-02 |
Pos logit contributions | |
older | +0.020 |
stays | +0.017 |
swe | +0.017 |
heavier | +0.017 |
asks | +0.016 |
Neg logit contributions | |
anish | -0.017 |
od | -0.016 |
iv | -0.016 |
And | -0.016 |
room | -0.016 |
Token | his |
Token ID | 465 |
Feature activation | +3.71e-03 |
Pos logit contributions | |
older | +0.049 |
heavier | +0.042 |
swe | +0.039 |
gs | +0.037 |
lower | +0.036 |
Neg logit contributions | |
anish | -0.063 |
iv | -0.062 |
od | -0.062 |
bed | -0.059 |
Jack | -0.058 |
Token | bin |
Token ID | 9874 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
older | +0.006 |
heavier | +0.006 |
swe | +0.005 |
stays | +0.005 |
lower | +0.005 |
Neg logit contributions | |
anish | -0.007 |
od | -0.006 |
iv | -0.006 |
iolet | -0.006 |
And | -0.006 |
Token | ocular |
Token ID | 37320 |
Feature activation | +8.08e-02 |
Pos logit contributions | |
older | +0.057 |
swe | +0.054 |
heavier | +0.053 |
shaky | +0.049 |
sle | +0.049 |
Neg logit contributions | |
od | -0.040 |
room | -0.038 |
iv | -0.037 |
anish | -0.037 |
bed | -0.035 |
Token | s |
Token ID | 82 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
older | +0.169 |
younger | +0.150 |
swe | +0.146 |
heavier | +0.143 |
sle | +0.135 |
Neg logit contributions | |
od | -0.097 |
room | -0.094 |
he | -0.091 |
were | -0.087 |
fully | -0.083 |
Token | . |
Token ID | 13 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
older | +0.027 |
heavier | +0.023 |
swe | +0.023 |
stays | +0.022 |
sle | +0.022 |
Neg logit contributions | |
anish | -0.020 |
iv | -0.020 |
od | -0.019 |
And | -0.018 |
iolet | -0.018 |
Token | He |
Token ID | 679 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
older | +0.021 |
swe | +0.019 |
sle | +0.018 |
heavier | +0.018 |
stays | +0.018 |
Neg logit contributions | |
anish | -0.016 |
room | -0.015 |
J | -0.014 |
iv | -0.014 |
Jane | -0.014 |
Token | observes |
Token ID | 34526 |
Feature activation | -9.65e-03 |
Pos logit contributions | |
heavier | +0.027 |
older | +0.026 |
sle | +0.023 |
swe | +0.023 |
attention | +0.023 |
Neg logit contributions | |
iv | -0.035 |
od | -0.034 |
iolet | -0.032 |
And | -0.032 |
anish | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | -7.38e-04 |
Pos logit contributions | |
od | +0.016 |
anish | +0.016 |
iv | +0.016 |
Jack | +0.015 |
J | +0.015 |
Neg logit contributions | |
older | -0.018 |
stays | -0.015 |
swe | -0.015 |
heavier | -0.014 |
sle | -0.014 |
Token | pig |
Token ID | 12967 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
older | +0.007 |
stays | +0.006 |
bars | +0.006 |
heavier | +0.006 |
opens | +0.005 |
Neg logit contributions | |
And | -0.008 |
iv | -0.008 |
anish | -0.008 |
od | -0.007 |
iolet | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
older | +0.040 |
swe | +0.034 |
heavier | +0.034 |
stays | +0.033 |
younger | +0.033 |
Neg logit contributions | |
iv | -0.029 |
anish | -0.028 |
od | -0.028 |
oon | -0.026 |
iolet | -0.026 |
Token | He |
Token ID | 679 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
older | +0.035 |
heavier | +0.030 |
swe | +0.029 |
stays | +0.029 |
bigger | +0.029 |
Neg logit contributions | |
anish | -0.022 |
iv | -0.022 |
And | -0.021 |
od | -0.021 |
iolet | -0.020 |
Token | says |
Token ID | 1139 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
older | +0.041 |
swe | +0.034 |
heavier | +0.033 |
bars | +0.033 |
younger | +0.032 |
Neg logit contributions | |
iv | -0.044 |
And | -0.040 |
od | -0.040 |
oon | -0.038 |
anish | -0.038 |
Token | o |
Token ID | 267 |
Feature activation | +4.50e-02 |
Pos logit contributions | |
older | +0.033 |
lower | +0.028 |
tighter | +0.028 |
bars | +0.028 |
younger | +0.027 |
Neg logit contributions | |
od | -0.023 |
iv | -0.022 |
anish | -0.021 |
And | -0.020 |
wards | -0.019 |
Token | ink |
Token ID | 676 |
Feature activation | +8.05e-02 |
Pos logit contributions | |
older | +0.097 |
swe | +0.082 |
younger | +0.079 |
heavier | +0.079 |
stays | +0.078 |
Neg logit contributions | |
anish | -0.070 |
iv | -0.070 |
od | -0.065 |
iolet | -0.063 |
And | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
older | +0.145 |
heavier | +0.128 |
younger | +0.123 |
sle | +0.121 |
shaky | +0.115 |
Neg logit contributions | |
oon | -0.134 |
iv | -0.133 |
anish | -0.131 |
od | -0.131 |
And | -0.117 |
Token | And |
Token ID | 843 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
older | +0.041 |
stays | +0.035 |
heavier | +0.035 |
swe | +0.034 |
lower | +0.033 |
Neg logit contributions | |
anish | -0.033 |
iv | -0.032 |
And | -0.032 |
od | -0.031 |
oon | -0.030 |
Token | this |
Token ID | 428 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
older | +0.045 |
swe | +0.037 |
younger | +0.036 |
heavier | +0.036 |
bars | +0.036 |
Neg logit contributions | |
iv | -0.031 |
And | -0.028 |
anish | -0.028 |
od | -0.028 |
wards | -0.028 |
Token | is |
Token ID | 318 |
Feature activation | +4.26e-03 |
Pos logit contributions | |
older | +0.030 |
younger | +0.024 |
heavier | +0.024 |
bars | +0.022 |
swe | +0.022 |
Neg logit contributions | |
iv | -0.033 |
anish | -0.030 |
And | -0.029 |
od | -0.029 |
oon | -0.028 |
Token | a |
Token ID | 257 |
Feature activation | +2.18e-03 |
Pos logit contributions | |
bars | +0.008 |
older | +0.008 |
passes | +0.007 |
climbs | +0.007 |
reins | +0.007 |
Neg logit contributions | |
iv | -0.005 |
od | -0.005 |
And | -0.005 |
Jane | -0.004 |
wards | -0.004 |
Token | He |
Token ID | 679 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
older | +0.034 |
swe | +0.029 |
stays | +0.029 |
sle | +0.029 |
heavier | +0.028 |
Neg logit contributions | |
anish | -0.027 |
And | -0.026 |
iv | -0.026 |
Jack | -0.026 |
od | -0.026 |
Token | says |
Token ID | 1139 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
older | +0.032 |
forward | +0.028 |
heavier | +0.026 |
reins | +0.025 |
bars | +0.024 |
Neg logit contributions | |
iv | -0.040 |
And | -0.036 |
od | -0.035 |
anish | -0.034 |
bed | -0.034 |
Token | , |
Token ID | 11 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
older | +0.036 |
gs | +0.032 |
bars | +0.032 |
sle | +0.032 |
lower | +0.032 |
Neg logit contributions | |
anish | -0.022 |
od | -0.020 |
iv | -0.018 |
And | -0.017 |
iolet | -0.017 |
Token | " |
Token ID | 366 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
older | +0.032 |
stays | +0.028 |
swe | +0.028 |
heavier | +0.027 |
sle | +0.026 |
Neg logit contributions | |
anish | -0.019 |
And | -0.018 |
od | -0.017 |
iv | -0.017 |
bed | -0.017 |
Token | O |
Token ID | 46 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
anish | +0.026 |
od | +0.026 |
iv | +0.025 |
oon | +0.023 |
iolet | +0.023 |
Neg logit contributions | |
older | -0.042 |
swe | -0.037 |
heavier | -0.037 |
stays | -0.037 |
goes | -0.036 |
Token | ink |
Token ID | 676 |
Feature activation | +8.04e-02 |
Pos logit contributions | |
older | +0.022 |
heavier | +0.019 |
stays | +0.019 |
swe | +0.019 |
goes | +0.018 |
Neg logit contributions | |
anish | -0.014 |
od | -0.013 |
iv | -0.013 |
Jack | -0.012 |
iolet | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | -2.41e-03 |
Pos logit contributions | |
older | +0.146 |
heavier | +0.127 |
younger | +0.124 |
swe | +0.122 |
sle | +0.122 |
Neg logit contributions | |
anish | -0.144 |
iv | -0.139 |
oon | -0.139 |
od | -0.131 |
iolet | -0.124 |
Token | o |
Token ID | 267 |
Feature activation | +3.27e-02 |
Pos logit contributions | |
iv | +0.004 |
anish | +0.003 |
od | +0.003 |
And | +0.003 |
ocolate | +0.003 |
Neg logit contributions | |
older | -0.005 |
bars | -0.004 |
younger | -0.004 |
lower | -0.004 |
swe | -0.004 |
Token | ink |
Token ID | 676 |
Feature activation | +7.53e-02 |
Pos logit contributions | |
older | +0.073 |
swe | +0.062 |
heavier | +0.061 |
younger | +0.060 |
stays | +0.060 |
Neg logit contributions | |
anish | -0.048 |
iv | -0.047 |
od | -0.043 |
iolet | -0.043 |
oon | -0.042 |
Token | , |
Token ID | 11 |
Feature activation | -3.58e-03 |
Pos logit contributions | |
older | +0.161 |
heavier | +0.141 |
younger | +0.139 |
bars | +0.137 |
swe | +0.135 |
Neg logit contributions | |
anish | -0.108 |
iv | -0.107 |
oon | -0.106 |
od | -0.099 |
iolet | -0.090 |
Token | Tom |
Token ID | 4186 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
iv | +0.006 |
And | +0.005 |
anish | +0.005 |
od | +0.005 |
ocolate | +0.005 |
Neg logit contributions | |
older | -0.006 |
younger | -0.005 |
bars | -0.005 |
stays | -0.005 |
tighter | -0.005 |
Token | mom |
Token ID | 1995 |
Feature activation | +2.31e-02 |
Pos logit contributions | |
older | +0.023 |
heavier | +0.020 |
stays | +0.020 |
swe | +0.019 |
lower | +0.018 |
Neg logit contributions | |
iv | -0.029 |
anish | -0.029 |
od | -0.028 |
And | -0.027 |
iolet | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | -5.92e-03 |
Pos logit contributions | |
older | +0.039 |
stays | +0.038 |
brakes | +0.037 |
sle | +0.036 |
gs | +0.035 |
Neg logit contributions | |
od | -0.039 |
anish | -0.037 |
iv | -0.037 |
J | -0.034 |
Jack | -0.034 |
Token | " |
Token ID | 366 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
anish | +0.011 |
iv | +0.011 |
od | +0.011 |
iolet | +0.010 |
oon | +0.010 |
Neg logit contributions | |
older | -0.010 |
heavier | -0.008 |
swe | -0.008 |
stays | -0.008 |
bigger | -0.008 |
Token | What |
Token ID | 2061 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
od | +0.053 |
anish | +0.051 |
iv | +0.051 |
oon | +0.047 |
wards | +0.047 |
Neg logit contributions | |
older | -0.044 |
swe | -0.037 |
lower | -0.035 |
bigger | -0.035 |
stable | -0.034 |
Token | are |
Token ID | 389 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
anish | +0.044 |
iv | +0.043 |
od | +0.042 |
iolet | +0.041 |
And | +0.041 |
Neg logit contributions | |
older | -0.027 |
heavier | -0.021 |
swe | -0.021 |
stays | -0.020 |
bigger | -0.019 |
Token | years |
Token ID | 812 |
Feature activation | +8.02e-02 |
Pos logit contributions | |
older | +0.034 |
stays | +0.029 |
heavier | +0.029 |
swe | +0.028 |
lower | +0.028 |
Neg logit contributions | |
iv | -0.027 |
od | -0.027 |
anish | -0.026 |
Jack | -0.024 |
Sus | -0.024 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
stays | +0.086 |
swe | +0.070 |
older | +0.068 |
brakes | +0.067 |
tighter | +0.065 |
Neg logit contributions | |
iv | -0.182 |
od | -0.179 |
Sus | -0.177 |
oon | -0.171 |
anish | -0.165 |
Token | Her |
Token ID | 2332 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
iv | +0.031 |
anish | +0.030 |
od | +0.030 |
oon | +0.030 |
iolet | +0.027 |
Neg logit contributions | |
older | -0.023 |
swe | -0.018 |
asks | -0.018 |
old | -0.018 |
down | -0.018 |
Token | Mom |
Token ID | 11254 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
older | +0.015 |
heavier | +0.012 |
swe | +0.012 |
stays | +0.012 |
bigger | +0.011 |
Neg logit contributions | |
anish | -0.022 |
iv | -0.022 |
od | -0.021 |
iolet | -0.021 |
And | -0.021 |
Token | smiled |
Token ID | 13541 |
Feature activation | +5.91e-03 |
Pos logit contributions | |
older | +0.030 |
heavier | +0.027 |
stays | +0.026 |
swe | +0.026 |
sle | +0.025 |
Neg logit contributions | |
iv | -0.024 |
od | -0.024 |
anish | -0.024 |
Jack | -0.022 |
Sus | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
older | +0.012 |
stays | +0.011 |
heavier | +0.010 |
swe | +0.010 |
gs | +0.010 |
Neg logit contributions | |
anish | -0.008 |
od | -0.008 |
iv | -0.008 |
And | -0.008 |
bed | -0.008 |
Token | asked |
Token ID | 1965 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
older | +0.047 |
heavier | +0.041 |
sle | +0.041 |
swe | +0.039 |
gs | +0.039 |
Neg logit contributions | |
Jack | -0.037 |
iv | -0.036 |
od | -0.036 |
Jane | -0.035 |
J | -0.035 |
Token | , |
Token ID | 11 |
Feature activation | -1.98e-03 |
Pos logit contributions | |
older | +0.025 |
heavier | +0.022 |
stays | +0.022 |
sle | +0.021 |
swe | +0.021 |
Neg logit contributions | |
anish | -0.018 |
od | -0.017 |
iv | -0.017 |
iolet | -0.016 |
J | -0.016 |
Token | " |
Token ID | 366 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
anish | +0.003 |
iv | +0.003 |
od | +0.003 |
And | +0.003 |
Jack | +0.003 |
Neg logit contributions | |
older | -0.004 |
stays | -0.003 |
swe | -0.003 |
heavier | -0.003 |
goes | -0.003 |
Token | Can |
Token ID | 6090 |
Feature activation | -5.92e-03 |
Pos logit contributions | |
anish | +0.040 |
od | +0.039 |
iv | +0.038 |
iolet | +0.035 |
bed | +0.034 |
Neg logit contributions | |
older | -0.055 |
swe | -0.048 |
heavier | -0.047 |
bigger | -0.045 |
lower | -0.045 |
Token | I |
Token ID | 314 |
Feature activation | -2.12e-03 |
Pos logit contributions | |
anish | +0.009 |
iv | +0.009 |
Jack | +0.009 |
od | +0.008 |
And | +0.008 |
Neg logit contributions | |
older | -0.011 |
heavier | -0.010 |
swe | -0.010 |
sle | -0.010 |
stable | -0.010 |
Token | set |
Token ID | 900 |
Feature activation | +7.97e-02 |
Pos logit contributions | |
iv | +0.003 |
anish | +0.003 |
Jack | +0.003 |
od | +0.003 |
And | +0.003 |
Neg logit contributions | |
older | -0.004 |
stays | -0.004 |
heavier | -0.004 |
swe | -0.004 |
brakes | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
older | +0.159 |
stays | +0.146 |
swe | +0.137 |
heavier | +0.136 |
younger | +0.135 |
Neg logit contributions | |
iv | -0.113 |
od | -0.110 |
bed | -0.107 |
anish | -0.106 |
Jack | -0.105 |
Token | table |
Token ID | 3084 |
Feature activation | +4.83e-02 |
Pos logit contributions | |
older | +0.048 |
stays | +0.044 |
heavier | +0.043 |
sle | +0.041 |
brakes | +0.041 |
Neg logit contributions | |
iv | -0.035 |
anish | -0.035 |
od | -0.035 |
Jack | -0.033 |
And | -0.033 |
Token | with |
Token ID | 351 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
older | +0.088 |
stays | +0.080 |
brakes | +0.079 |
sle | +0.079 |
gs | +0.077 |
Neg logit contributions | |
anish | -0.078 |
room | -0.072 |
iv | -0.071 |
Jack | -0.069 |
iolet | -0.068 |
Token | you |
Token ID | 345 |
Feature activation | -7.00e-03 |
Pos logit contributions | |
older | +0.033 |
stays | +0.030 |
heavier | +0.028 |
gs | +0.028 |
sle | +0.028 |
Neg logit contributions | |
iv | -0.031 |
anish | -0.031 |
od | -0.030 |
Jack | -0.029 |
bed | -0.028 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
anish | +0.010 |
od | +0.010 |
iv | +0.010 |
oon | +0.009 |
Jack | +0.009 |
Neg logit contributions | |
stays | -0.013 |
older | -0.012 |
gs | -0.012 |
brakes | -0.012 |
sle | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
iv | +0.003 |
anish | +0.003 |
od | +0.002 |
bed | +0.002 |
And | +0.002 |
Neg logit contributions | |
older | -0.004 |
heavier | -0.003 |
stays | -0.003 |
swe | -0.003 |
sle | -0.003 |
Token | said |
Token ID | 531 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
older | +0.052 |
heavier | +0.044 |
sle | +0.043 |
gs | +0.043 |
swe | +0.041 |
Neg logit contributions | |
anish | -0.042 |
iv | -0.041 |
Jack | -0.040 |
od | -0.040 |
Jane | -0.039 |
Token | , |
Token ID | 11 |
Feature activation | +6.68e-03 |
Pos logit contributions | |
older | +0.025 |
stays | +0.021 |
heavier | +0.021 |
swe | +0.020 |
sle | +0.020 |
Neg logit contributions | |
anish | -0.023 |
od | -0.023 |
iv | -0.022 |
Jack | -0.021 |
iolet | -0.021 |
Token | " |
Token ID | 366 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
older | +0.014 |
stays | +0.012 |
swe | +0.012 |
heavier | +0.012 |
goes | +0.011 |
Neg logit contributions | |
anish | -0.009 |
od | -0.009 |
And | -0.009 |
iv | -0.009 |
Jack | -0.008 |
Token | Dr |
Token ID | 6187 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
od | +0.029 |
anish | +0.028 |
iv | +0.028 |
oon | +0.025 |
iolet | +0.025 |
Neg logit contributions | |
older | -0.047 |
swe | -0.042 |
stays | -0.041 |
lower | -0.041 |
heavier | -0.041 |
Token | ink |
Token ID | 676 |
Feature activation | +7.89e-02 |
Pos logit contributions | |
older | +0.063 |
swe | +0.053 |
heavier | +0.053 |
stays | +0.052 |
bigger | +0.050 |
Neg logit contributions | |
anish | -0.054 |
iv | -0.053 |
od | -0.051 |
iolet | -0.049 |
And | -0.049 |
Token | this |
Token ID | 428 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
older | +0.140 |
heavier | +0.125 |
swe | +0.122 |
sle | +0.116 |
cooler | +0.115 |
Neg logit contributions | |
anish | -0.139 |
iv | -0.133 |
od | -0.131 |
oon | -0.127 |
iolet | -0.121 |
Token | and |
Token ID | 290 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
older | +0.024 |
heavier | +0.019 |
younger | +0.018 |
swe | +0.018 |
tighter | +0.018 |
Neg logit contributions | |
iv | -0.028 |
anish | -0.027 |
od | -0.025 |
iolet | -0.024 |
oon | -0.024 |
Token | you |
Token ID | 345 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
older | +0.064 |
gs | +0.056 |
sle | +0.054 |
stays | +0.050 |
swe | +0.050 |
Neg logit contributions | |
anish | -0.068 |
iv | -0.063 |
Jack | -0.062 |
And | -0.061 |
All | -0.059 |
Token | will |
Token ID | 481 |
Feature activation | -7.38e-03 |
Pos logit contributions | |
iv | +0.032 |
anish | +0.032 |
od | +0.030 |
oon | +0.029 |
iolet | +0.029 |
Neg logit contributions | |
older | -0.033 |
swe | -0.027 |
heavier | -0.027 |
sle | -0.026 |
younger | -0.025 |
Token | feel |
Token ID | 1254 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
anish | +0.014 |
iv | +0.014 |
od | +0.013 |
iolet | +0.013 |
And | +0.013 |
Neg logit contributions | |
older | -0.012 |
heavier | -0.010 |
swe | -0.010 |
stays | -0.010 |
bigger | -0.009 |
Token | Sarah |
Token ID | 10490 |
Feature activation | +3.32e-03 |
Pos logit contributions | |
older | +0.001 |
swe | +0.001 |
stays | +0.001 |
bars | +0.001 |
sle | +0.001 |
Neg logit contributions | |
od | -0.001 |
iv | -0.001 |
wards | -0.001 |
And | -0.001 |
Sus | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
older | +0.007 |
stays | +0.006 |
heavier | +0.006 |
brakes | +0.006 |
climbs | +0.006 |
Neg logit contributions | |
od | -0.004 |
iv | -0.004 |
anish | -0.004 |
iolet | -0.004 |
Sus | -0.004 |
Token | I |
Token ID | 314 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
older | +0.073 |
swe | +0.064 |
heavier | +0.063 |
gs | +0.060 |
sle | +0.060 |
Neg logit contributions | |
anish | -0.064 |
iv | -0.063 |
od | -0.062 |
Jack | -0.059 |
oon | -0.056 |
Token | 'm |
Token ID | 1101 |
Feature activation | -2.56e-03 |
Pos logit contributions | |
anish | +0.032 |
iv | +0.031 |
od | +0.030 |
And | +0.030 |
oon | +0.029 |
Neg logit contributions | |
older | -0.034 |
swe | -0.029 |
heavier | -0.029 |
sle | -0.028 |
bigger | -0.027 |
Token | 3 |
Token ID | 513 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
od | +0.004 |
iv | +0.004 |
Sus | +0.003 |
And | +0.003 |
oon | +0.003 |
Neg logit contributions | |
older | -0.005 |
swe | -0.004 |
stays | -0.004 |
bars | -0.004 |
sle | -0.004 |
Token | years |
Token ID | 812 |
Feature activation | +7.89e-02 |
Pos logit contributions | |
anish | +0.021 |
iv | +0.021 |
od | +0.020 |
iolet | +0.019 |
And | +0.019 |
Neg logit contributions | |
older | -0.040 |
heavier | -0.034 |
stays | -0.034 |
swe | -0.034 |
cooler | -0.033 |
Token | old |
Token ID | 1468 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
stays | +0.073 |
climbs | +0.062 |
goes | +0.060 |
reaches | +0.058 |
leads | +0.058 |
Neg logit contributions | |
Sus | -0.170 |
iv | -0.167 |
od | -0.166 |
wards | -0.164 |
And | -0.163 |
Token | and |
Token ID | 290 |
Feature activation | +4.27e-02 |
Pos logit contributions | |
anish | +0.016 |
iv | +0.016 |
od | +0.016 |
iolet | +0.015 |
And | +0.015 |
Neg logit contributions | |
older | -0.021 |
stays | -0.019 |
heavier | -0.018 |
swe | -0.018 |
sle | -0.017 |
Token | I |
Token ID | 314 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
older | +0.077 |
swe | +0.066 |
heavier | +0.065 |
sle | +0.064 |
gs | +0.064 |
Neg logit contributions | |
iv | -0.068 |
anish | -0.067 |
od | -0.066 |
And | -0.063 |
Jack | -0.062 |
Token | 'm |
Token ID | 1101 |
Feature activation | -5.38e-03 |
Pos logit contributions | |
iv | +0.032 |
anish | +0.032 |
od | +0.031 |
And | +0.030 |
oon | +0.029 |
Neg logit contributions | |
older | -0.035 |
heavier | -0.029 |
swe | -0.029 |
sle | -0.028 |
bigger | -0.027 |
Token | very |
Token ID | 845 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
od | +0.008 |
iv | +0.008 |
Sus | +0.007 |
And | +0.007 |
wards | +0.007 |
Neg logit contributions | |
older | -0.010 |
swe | -0.010 |
stays | -0.009 |
sle | -0.009 |
goes | -0.009 |
Token | son |
Token ID | 3367 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
anish | +0.010 |
iv | +0.009 |
od | +0.009 |
room | +0.009 |
And | +0.009 |
Neg logit contributions | |
older | -0.014 |
heavier | -0.013 |
lower | -0.013 |
gs | -0.012 |
stays | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
od | +0.042 |
iv | +0.041 |
anish | +0.039 |
iolet | +0.037 |
room | +0.036 |
Neg logit contributions | |
older | -0.060 |
heavier | -0.054 |
lower | -0.052 |
sle | -0.051 |
brakes | -0.050 |
Token | He |
Token ID | 679 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
older | +0.026 |
heavier | +0.022 |
swe | +0.022 |
stays | +0.021 |
bigger | +0.021 |
Neg logit contributions | |
anish | -0.020 |
iv | -0.019 |
od | -0.019 |
iolet | -0.018 |
And | -0.018 |
Token | is |
Token ID | 318 |
Feature activation | -5.09e-03 |
Pos logit contributions | |
older | +0.036 |
heavier | +0.031 |
swe | +0.031 |
sle | +0.029 |
lower | +0.029 |
Neg logit contributions | |
iv | -0.045 |
od | -0.042 |
anish | -0.042 |
iolet | -0.040 |
And | -0.040 |
Token | three |
Token ID | 1115 |
Feature activation | -2.81e-03 |
Pos logit contributions | |
od | +0.008 |
iv | +0.007 |
sharing | +0.007 |
And | +0.007 |
creating | +0.007 |
Neg logit contributions | |
stays | -0.009 |
bars | -0.009 |
older | -0.008 |
reins | -0.008 |
climbs | -0.008 |
Token | years |
Token ID | 812 |
Feature activation | +7.86e-02 |
Pos logit contributions | |
iv | +0.003 |
od | +0.003 |
anish | +0.003 |
And | +0.003 |
iolet | +0.003 |
Neg logit contributions | |
older | -0.007 |
stays | -0.006 |
lower | -0.006 |
heavier | -0.006 |
climbs | -0.006 |
Token | old |
Token ID | 1468 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
stays | +0.103 |
brakes | +0.086 |
climbs | +0.086 |
reaches | +0.084 |
rises | +0.079 |
Neg logit contributions | |
od | -0.148 |
Sus | -0.147 |
iv | -0.143 |
J | -0.143 |
Em | -0.140 |
Token | , |
Token ID | 11 |
Feature activation | -6.75e-03 |
Pos logit contributions | |
anish | +0.036 |
iv | +0.035 |
od | +0.034 |
iolet | +0.033 |
And | +0.032 |
Neg logit contributions | |
older | -0.045 |
heavier | -0.038 |
swe | -0.038 |
stays | -0.038 |
lower | -0.036 |
Token | like |
Token ID | 588 |
Feature activation | -2.87e-03 |
Pos logit contributions | |
anish | +0.009 |
And | +0.008 |
iv | +0.008 |
od | +0.008 |
Jack | +0.008 |
Neg logit contributions | |
older | -0.015 |
stays | -0.013 |
heavier | -0.012 |
swe | -0.012 |
sle | -0.012 |
Token | you |
Token ID | 345 |
Feature activation | -3.24e-02 |
Pos logit contributions | |
anish | +0.004 |
iv | +0.004 |
od | +0.004 |
Jack | +0.004 |
And | +0.004 |
Neg logit contributions | |
older | -0.005 |
stays | -0.005 |
lower | -0.005 |
heavier | -0.004 |
gs | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
anish | +0.052 |
iv | +0.052 |
od | +0.050 |
iolet | +0.048 |
And | +0.047 |
Neg logit contributions | |
older | -0.062 |
swe | -0.054 |
heavier | -0.053 |
stays | -0.052 |
sle | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | +7.62e-03 |
Pos logit contributions | |
older | +0.020 |
stays | +0.018 |
heavier | +0.017 |
swe | +0.017 |
bars | +0.017 |
Neg logit contributions | |
iv | -0.013 |
anish | -0.012 |
od | -0.012 |
wards | -0.012 |
bed | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
older | +0.016 |
swe | +0.015 |
heavier | +0.014 |
stays | +0.014 |
sle | +0.013 |
Neg logit contributions | |
iv | -0.010 |
od | -0.010 |
anish | -0.010 |
And | -0.009 |
All | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -2.11e-03 |
Pos logit contributions | |
older | +0.034 |
heavier | +0.029 |
swe | +0.029 |
stays | +0.029 |
bigger | +0.028 |
Neg logit contributions | |
anish | -0.020 |
iv | -0.019 |
od | -0.019 |
iolet | -0.018 |
Jack | -0.018 |
Token | " |
Token ID | 1 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
iv | +0.003 |
anish | +0.003 |
od | +0.003 |
And | +0.003 |
J | +0.003 |
Neg logit contributions | |
older | -0.004 |
swe | -0.004 |
heavier | -0.004 |
stays | -0.004 |
younger | -0.003 |
Token | Spr |
Token ID | 38454 |
Feature activation | +6.15e-02 |
Pos logit contributions | |
anish | +0.036 |
iv | +0.034 |
od | +0.034 |
iolet | +0.032 |
And | +0.032 |
Neg logit contributions | |
older | -0.049 |
heavier | -0.042 |
swe | -0.041 |
stays | -0.041 |
bigger | -0.040 |
Token | ink |
Token ID | 676 |
Feature activation | +7.85e-02 |
Pos logit contributions | |
older | +0.103 |
heavier | +0.083 |
swe | +0.083 |
stays | +0.082 |
bigger | +0.078 |
Neg logit contributions | |
anish | -0.117 |
iv | -0.113 |
od | -0.112 |
iolet | -0.107 |
And | -0.106 |
Token | les |
Token ID | 829 |
Feature activation | +3.80e-02 |
Pos logit contributions | |
older | +0.141 |
heavier | +0.129 |
swe | +0.125 |
cheaper | +0.116 |
younger | +0.116 |
Neg logit contributions | |
od | -0.135 |
oon | -0.133 |
anish | -0.129 |
iv | -0.129 |
bed | -0.121 |
Token | , |
Token ID | 11 |
Feature activation | +4.58e-03 |
Pos logit contributions | |
older | +0.070 |
swe | +0.062 |
heavier | +0.061 |
stays | +0.057 |
younger | +0.056 |
Neg logit contributions | |
anish | -0.065 |
iv | -0.063 |
od | -0.062 |
oon | -0.059 |
iolet | -0.057 |
Token | no |
Token ID | 645 |
Feature activation | -3.58e-04 |
Pos logit contributions | |
older | +0.008 |
swe | +0.007 |
heavier | +0.007 |
stays | +0.007 |
tighter | +0.007 |
Neg logit contributions | |
iv | -0.008 |
od | -0.007 |
anish | -0.007 |
oon | -0.007 |
Sus | -0.007 |
Token | ! |
Token ID | 0 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
iv | +0.001 |
anish | +0.001 |
od | +0.001 |
Sus | +0.001 |
iolet | +0.001 |
Neg logit contributions | |
older | -0.001 |
stays | -0.001 |
heavier | -0.001 |
swe | -0.001 |
opens | -0.001 |
Token | Stop |
Token ID | 13707 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
older | +0.033 |
swe | +0.030 |
heavier | +0.030 |
stays | +0.029 |
lower | +0.028 |
Neg logit contributions | |
anish | -0.024 |
iv | -0.023 |
od | -0.022 |
iolet | -0.021 |
oon | -0.020 |
Token | girl |
Token ID | 2576 |
Feature activation | -5.01e-03 |
Pos logit contributions | |
anish | +0.017 |
iv | +0.016 |
od | +0.016 |
Jack | +0.015 |
Jane | +0.015 |
Neg logit contributions | |
older | -0.022 |
stays | -0.021 |
heavier | -0.020 |
swe | -0.019 |
sits | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | -5.89e-03 |
Pos logit contributions | |
od | +0.007 |
iv | +0.007 |
anish | +0.007 |
bed | +0.007 |
iolet | +0.006 |
Neg logit contributions | |
older | -0.010 |
stays | -0.009 |
swe | -0.009 |
heavier | -0.009 |
sle | -0.009 |
Token | who |
Token ID | 508 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
anish | +0.010 |
iv | +0.009 |
od | +0.009 |
iolet | +0.009 |
And | +0.008 |
Neg logit contributions | |
older | -0.012 |
heavier | -0.010 |
swe | -0.010 |
stays | -0.010 |
bigger | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
older | +0.031 |
heavier | +0.026 |
swe | +0.026 |
stays | +0.025 |
bigger | +0.025 |
Neg logit contributions | |
anish | -0.026 |
iv | -0.025 |
od | -0.025 |
iolet | -0.024 |
And | -0.024 |
Token | three |
Token ID | 1115 |
Feature activation | -3.87e-03 |
Pos logit contributions | |
iv | +0.019 |
od | +0.018 |
Jane | +0.017 |
anish | +0.017 |
wards | +0.017 |
Neg logit contributions | |
bars | -0.019 |
swe | -0.019 |
older | -0.019 |
stays | -0.019 |
sle | -0.018 |
Token | years |
Token ID | 812 |
Feature activation | +7.82e-02 |
Pos logit contributions | |
od | +0.004 |
iv | +0.004 |
anish | +0.004 |
And | +0.004 |
Jane | +0.004 |
Neg logit contributions | |
older | -0.009 |
stays | -0.008 |
heavier | -0.008 |
lower | -0.008 |
swe | -0.008 |
Token | old |
Token ID | 1468 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
stays | +0.106 |
sits | +0.088 |
carries | +0.086 |
climbs | +0.084 |
goes | +0.084 |
Neg logit contributions | |
Sus | -0.147 |
Jane | -0.143 |
edly | -0.134 |
Mill | -0.134 |
od | -0.133 |
Token | . |
Token ID | 13 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
anish | +0.038 |
iv | +0.036 |
od | +0.036 |
iolet | +0.034 |
And | +0.034 |
Neg logit contributions | |
older | -0.040 |
heavier | -0.033 |
swe | -0.033 |
stays | -0.033 |
bigger | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
older | +0.024 |
heavier | +0.021 |
swe | +0.020 |
stays | +0.020 |
cooler | +0.019 |
Neg logit contributions | |
anish | -0.018 |
iv | -0.017 |
od | -0.017 |
And | -0.016 |
Jack | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -9.32e-03 |
Pos logit contributions | |
older | +0.020 |
swe | +0.016 |
bigger | +0.016 |
old | +0.016 |
heavier | +0.016 |
Neg logit contributions | |
anish | -0.017 |
iv | -0.016 |
od | -0.015 |
iolet | -0.015 |
oon | -0.014 |
Token | " |
Token ID | 1 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
anish | +0.018 |
iv | +0.018 |
od | +0.018 |
iolet | +0.017 |
Sus | +0.016 |
Neg logit contributions | |
older | -0.015 |
swe | -0.012 |
heavier | -0.011 |
stays | -0.011 |
stable | -0.011 |
Token | 't |
Token ID | 470 |
Feature activation | -5.85e-03 |
Pos logit contributions | |
anish | +0.013 |
iv | +0.013 |
od | +0.013 |
iolet | +0.012 |
And | +0.012 |
Neg logit contributions | |
older | -0.022 |
swe | -0.019 |
heavier | -0.019 |
stays | -0.019 |
bigger | -0.018 |
Token | mind |
Token ID | 2000 |
Feature activation | -7.27e-03 |
Pos logit contributions | |
anish | +0.011 |
iv | +0.011 |
od | +0.011 |
And | +0.010 |
iolet | +0.010 |
Neg logit contributions | |
older | -0.010 |
swe | -0.008 |
heavier | -0.008 |
stays | -0.008 |
sle | -0.008 |
Token | eating |
Token ID | 6600 |
Feature activation | -6.38e-03 |
Pos logit contributions | |
od | +0.011 |
iv | +0.011 |
anish | +0.011 |
Sus | +0.010 |
iolet | +0.010 |
Neg logit contributions | |
older | -0.014 |
stays | -0.012 |
heavier | -0.012 |
sle | -0.012 |
swe | -0.012 |
Token | her |
Token ID | 607 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
anish | +0.011 |
iv | +0.011 |
od | +0.011 |
Jane | +0.010 |
iolet | +0.010 |
Neg logit contributions | |
older | -0.012 |
stays | -0.009 |
sle | -0.009 |
heavier | -0.009 |
younger | -0.009 |
Token | vegetables |
Token ID | 13701 |
Feature activation | -1.29e-03 |
Pos logit contributions | |
anish | +0.020 |
iv | +0.019 |
od | +0.019 |
bed | +0.017 |
iolet | +0.017 |
Neg logit contributions | |
older | -0.023 |
heavier | -0.020 |
sle | -0.019 |
gs | -0.019 |
lower | -0.019 |
Token | anymore |
Token ID | 7471 |
Feature activation | -8.26e-02 |
Pos logit contributions | |
anish | +0.002 |
iv | +0.002 |
od | +0.002 |
iolet | +0.002 |
Jack | +0.002 |
Neg logit contributions | |
older | -0.002 |
heavier | -0.002 |
swe | -0.002 |
stays | -0.002 |
bigger | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
anish | +0.192 |
iolet | +0.175 |
Sus | +0.172 |
And | +0.172 |
Jack | +0.171 |
Neg logit contributions | |
older | -0.100 |
swe | -0.081 |
old | -0.077 |
heavier | -0.076 |
goes | -0.075 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
anish | +0.022 |
iv | +0.022 |
od | +0.021 |
oon | +0.020 |
iolet | +0.020 |
Neg logit contributions | |
older | -0.021 |
swe | -0.018 |
goes | -0.017 |
old | -0.017 |
stays | -0.017 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
anish | +0.024 |
od | +0.023 |
iv | +0.023 |
Sus | +0.021 |
bed | +0.021 |
Neg logit contributions | |
older | -0.021 |
goes | -0.017 |
old | -0.017 |
cooler | -0.017 |
bigger | -0.017 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
older | +0.022 |
heavier | +0.019 |
stays | +0.019 |
swe | +0.019 |
sle | +0.018 |
Neg logit contributions | |
od | -0.031 |
anish | -0.031 |
Jack | -0.030 |
iv | -0.030 |
J | -0.030 |
Token | a |
Token ID | 257 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
anish | +0.036 |
iv | +0.034 |
od | +0.033 |
bed | +0.032 |
iolet | +0.032 |
Neg logit contributions | |
older | -0.015 |
bigger | -0.012 |
old | -0.012 |
heavy | -0.011 |
cooler | -0.011 |
Token | forgot |
Token ID | 16453 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
od | +0.002 |
anish | +0.002 |
J | +0.002 |
iv | +0.002 |
Jack | +0.002 |
Neg logit contributions | |
older | -0.003 |
sle | -0.002 |
stable | -0.002 |
heavier | -0.002 |
swe | -0.002 |
Token | about |
Token ID | 546 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
iv | +0.047 |
od | +0.046 |
anish | +0.045 |
Jack | +0.043 |
Sus | +0.043 |
Neg logit contributions | |
older | -0.058 |
sle | -0.051 |
heavier | -0.049 |
stays | -0.048 |
opens | -0.048 |
Token | the |
Token ID | 262 |
Feature activation | -3.25e-03 |
Pos logit contributions | |
anish | +0.050 |
iv | +0.048 |
od | +0.047 |
iolet | +0.045 |
And | +0.045 |
Neg logit contributions | |
older | -0.044 |
heavier | -0.035 |
swe | -0.035 |
stays | -0.035 |
bigger | -0.033 |
Token | girl |
Token ID | 2576 |
Feature activation | -6.28e-02 |
Pos logit contributions | |
anish | +0.006 |
iv | +0.006 |
od | +0.006 |
Jack | +0.005 |
J | +0.005 |
Neg logit contributions | |
older | -0.006 |
heavier | -0.005 |
stays | -0.004 |
swe | -0.004 |
sle | -0.004 |
Token | â |
Token ID | 22940 |
Feature activation | +4.28e-03 |
Pos logit contributions | |
od | +0.101 |
anish | +0.099 |
iv | +0.099 |
Jack | +0.092 |
And | +0.091 |
Neg logit contributions | |
older | -0.120 |
stays | -0.101 |
swe | -0.101 |
heavier | -0.100 |
sle | -0.097 |
Token | € |
Token ID | 26391 |
Feature activation | -8.16e-02 |
Pos logit contributions | |
older | +0.009 |
stays | +0.007 |
heavier | +0.007 |
swe | +0.007 |
goes | +0.007 |
Neg logit contributions | |
anish | -0.007 |
od | -0.006 |
iv | -0.006 |
iolet | -0.006 |
And | -0.006 |
Token | ™ |
Token ID | 8151 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
anish | +0.121 |
iv | +0.114 |
od | +0.106 |
iolet | +0.103 |
And | +0.101 |
Neg logit contributions | |
older | -0.167 |
stays | -0.142 |
swe | -0.140 |
goes | -0.139 |
heavier | -0.139 |
Token | s |
Token ID | 82 |
Feature activation | -9.33e-03 |
Pos logit contributions | |
od | +0.023 |
anish | +0.023 |
iv | +0.023 |
Jack | +0.022 |
iolet | +0.021 |
Neg logit contributions | |
older | -0.032 |
stays | -0.027 |
swe | -0.027 |
heavier | -0.027 |
sle | -0.026 |
Token | beautiful |
Token ID | 4950 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
anish | +0.016 |
iv | +0.015 |
od | +0.015 |
iolet | +0.014 |
And | +0.014 |
Neg logit contributions | |
older | -0.017 |
heavier | -0.014 |
swe | -0.014 |
stays | -0.014 |
sle | -0.014 |
Token | skin |
Token ID | 4168 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
anish | +0.053 |
iv | +0.051 |
od | +0.051 |
iolet | +0.047 |
Jack | +0.047 |
Neg logit contributions | |
older | -0.059 |
heavier | -0.049 |
stays | -0.047 |
swe | -0.047 |
sle | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
iv | +0.020 |
anish | +0.019 |
od | +0.019 |
oon | +0.017 |
iolet | +0.017 |
Neg logit contributions | |
older | -0.029 |
stays | -0.024 |
sle | -0.024 |
heavier | -0.024 |
swe | -0.024 |
Token | had |
Token ID | 550 |
Feature activation | -3.79e-02 |
Pos logit contributions | |
od | +0.002 |
anish | +0.002 |
iv | +0.001 |
Jack | +0.001 |
And | +0.001 |
Neg logit contributions | |
older | -0.002 |
heavier | -0.002 |
swe | -0.002 |
sle | -0.002 |
cooler | -0.002 |
Token | to |
Token ID | 284 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
anish | +0.060 |
od | +0.058 |
iv | +0.058 |
iolet | +0.055 |
Jack | +0.054 |
Neg logit contributions | |
older | -0.073 |
sle | -0.062 |
heavier | -0.062 |
swe | -0.060 |
stable | -0.059 |
Token | resist |
Token ID | 4180 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
anish | +0.070 |
iv | +0.068 |
od | +0.067 |
iolet | +0.064 |
And | +0.064 |
Neg logit contributions | |
older | -0.057 |
heavier | -0.045 |
swe | -0.045 |
stays | -0.044 |
bigger | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | +9.68e-03 |
Pos logit contributions | |
anish | +0.055 |
iv | +0.053 |
od | +0.052 |
And | +0.049 |
iolet | +0.049 |
Neg logit contributions | |
older | -0.064 |
heavier | -0.053 |
stays | -0.053 |
swe | -0.053 |
sle | -0.051 |
Token | other |
Token ID | 584 |
Feature activation | -3.50e-02 |
Pos logit contributions | |
older | +0.015 |
bars | +0.014 |
opens | +0.014 |
stays | +0.013 |
climbs | +0.013 |
Neg logit contributions | |
iv | -0.016 |
anish | -0.015 |
And | -0.015 |
od | -0.014 |
All | -0.014 |
Token | anymore |
Token ID | 7471 |
Feature activation | -8.10e-02 |
Pos logit contributions | |
iv | +0.048 |
od | +0.046 |
All | +0.045 |
And | +0.045 |
we | +0.044 |
Neg logit contributions | |
older | -0.066 |
bars | -0.058 |
stays | -0.057 |
lower | -0.057 |
climbs | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
anish | +0.152 |
iv | +0.143 |
od | +0.140 |
iolet | +0.137 |
Jack | +0.136 |
Neg logit contributions | |
older | -0.139 |
swe | -0.112 |
heavier | -0.112 |
stays | -0.110 |
goes | -0.106 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
anish | +0.036 |
iv | +0.034 |
oon | +0.032 |
od | +0.032 |
bed | +0.032 |
Neg logit contributions | |
older | -0.029 |
swe | -0.024 |
goes | -0.023 |
passes | -0.023 |
old | -0.023 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
anish | +0.023 |
iv | +0.022 |
od | +0.022 |
Sus | +0.020 |
And | +0.020 |
Neg logit contributions | |
older | -0.024 |
goes | -0.020 |
bigger | -0.019 |
heavier | -0.019 |
cooler | -0.019 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
older | +0.020 |
heavier | +0.017 |
swe | +0.017 |
stays | +0.017 |
sle | +0.016 |
Neg logit contributions | |
anish | -0.028 |
od | -0.028 |
Jack | -0.027 |
iv | -0.027 |
Jane | -0.027 |
Token | a |
Token ID | 257 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
anish | +0.047 |
iv | +0.043 |
od | +0.043 |
bed | +0.041 |
iolet | +0.041 |
Neg logit contributions | |
older | -0.018 |
bigger | -0.015 |
heavy | -0.014 |
cooler | -0.014 |
old | -0.014 |
Token | was |
Token ID | 373 |
Feature activation | +4.23e-03 |
Pos logit contributions | |
Jack | +0.019 |
od | +0.019 |
All | +0.018 |
anish | +0.018 |
iv | +0.017 |
Neg logit contributions | |
older | -0.030 |
reins | -0.026 |
lower | -0.026 |
sle | -0.026 |
ented | -0.026 |
Token | around |
Token ID | 1088 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
older | +0.008 |
swe | +0.008 |
sle | +0.008 |
stays | +0.008 |
reins | +0.008 |
Neg logit contributions | |
od | -0.005 |
iv | -0.005 |
Sus | -0.005 |
creating | -0.005 |
oon | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
od | +0.031 |
anish | +0.031 |
iv | +0.030 |
bed | +0.029 |
room | +0.028 |
Neg logit contributions | |
older | -0.052 |
swe | -0.048 |
heavier | -0.044 |
cooler | -0.044 |
stays | -0.044 |
Token | river |
Token ID | 7850 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
And | +0.024 |
All | +0.022 |
iv | +0.022 |
od | +0.021 |
Jack | +0.021 |
Neg logit contributions | |
older | -0.027 |
stays | -0.027 |
opens | -0.026 |
swe | -0.025 |
sits | -0.024 |
Token | â |
Token ID | 22940 |
Feature activation | +4.18e-03 |
Pos logit contributions | |
iv | +0.014 |
od | +0.014 |
anish | +0.013 |
room | +0.013 |
oon | +0.012 |
Neg logit contributions | |
older | -0.024 |
swe | -0.022 |
stays | -0.021 |
sle | -0.021 |
reins | -0.021 |
Token | € |
Token ID | 26391 |
Feature activation | -8.10e-02 |
Pos logit contributions | |
older | +0.009 |
heavier | +0.007 |
swe | +0.007 |
stays | +0.007 |
bigger | +0.007 |
Neg logit contributions | |
anish | -0.006 |
iv | -0.006 |
od | -0.006 |
And | -0.006 |
Jack | -0.006 |
Token | ™ |
Token ID | 8151 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
anish | +0.124 |
iv | +0.120 |
od | +0.113 |
iolet | +0.109 |
Sus | +0.106 |
Neg logit contributions | |
older | -0.163 |
stays | -0.137 |
heavier | -0.136 |
goes | -0.134 |
swe | -0.133 |
Token | s |
Token ID | 82 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
od | +0.024 |
iv | +0.023 |
anish | +0.022 |
Jack | +0.022 |
J | +0.022 |
Neg logit contributions | |
older | -0.031 |
swe | -0.028 |
stays | -0.028 |
sle | -0.027 |
heavier | -0.027 |
Token | bend |
Token ID | 19396 |
Feature activation | -7.92e-03 |
Pos logit contributions | |
anish | +0.020 |
od | +0.019 |
iv | +0.019 |
room | +0.019 |
bed | +0.018 |
Neg logit contributions | |
older | -0.026 |
swe | -0.024 |
stays | -0.022 |
heavier | -0.022 |
sle | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -9.35e-03 |
Pos logit contributions | |
anish | +0.009 |
iv | +0.009 |
od | +0.009 |
bed | +0.009 |
room | +0.008 |
Neg logit contributions | |
older | -0.018 |
sle | -0.016 |
swe | -0.016 |
heavier | -0.015 |
climbs | -0.015 |
Token | He |
Token ID | 679 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
anish | +0.010 |
iv | +0.010 |
And | +0.010 |
All | +0.009 |
od | +0.009 |
Neg logit contributions | |
older | -0.023 |
swe | -0.020 |
stays | -0.020 |
heavier | -0.020 |
bigger | -0.019 |
Token | spent |
Token ID | 3377 |
Feature activation | -4.42e-02 |
Pos logit contributions | |
anish | +0.039 |
iv | +0.038 |
od | +0.037 |
iolet | +0.035 |
Jack | +0.035 |
Neg logit contributions | |
older | -0.052 |
heavier | -0.044 |
swe | -0.044 |
stays | -0.043 |
bigger | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
anish | +0.109 |
iv | +0.103 |
oon | +0.102 |
od | +0.102 |
Sus | +0.100 |
Neg logit contributions | |
older | -0.050 |
old | -0.036 |
swe | -0.036 |
bigger | -0.035 |
heavier | -0.035 |
Token | rest |
Token ID | 1334 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
anish | +0.028 |
od | +0.026 |
iv | +0.026 |
iolet | +0.025 |
oon | +0.024 |
Neg logit contributions | |
older | -0.025 |
swe | -0.021 |
stays | -0.020 |
heavier | -0.020 |
goes | -0.020 |
Token | of |
Token ID | 286 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
anish | +0.079 |
iv | +0.076 |
od | +0.075 |
iolet | +0.073 |
Jack | +0.072 |
Neg logit contributions | |
older | -0.062 |
heavier | -0.049 |
stays | -0.048 |
swe | -0.047 |
bigger | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | -8.70e-03 |
Pos logit contributions | |
anish | +0.051 |
iv | +0.048 |
od | +0.047 |
oon | +0.047 |
Sus | +0.046 |
Neg logit contributions | |
older | -0.020 |
old | -0.013 |
heavier | -0.013 |
bigger | -0.013 |
swe | -0.013 |
Token | day |
Token ID | 1110 |
Feature activation | -8.05e-02 |
Pos logit contributions | |
anish | +0.013 |
iv | +0.013 |
od | +0.012 |
Jack | +0.012 |
And | +0.012 |
Neg logit contributions | |
older | -0.018 |
heavier | -0.015 |
swe | -0.015 |
stays | -0.015 |
bigger | -0.015 |
Token | enjoying |
Token ID | 13226 |
Feature activation | -4.30e-02 |
Pos logit contributions | |
anish | +0.114 |
iv | +0.107 |
od | +0.103 |
iolet | +0.100 |
And | +0.098 |
Neg logit contributions | |
older | -0.177 |
heavier | -0.150 |
swe | -0.149 |
stays | -0.147 |
bigger | -0.144 |
Token | themselves |
Token ID | 2405 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
anish | +0.103 |
oon | +0.095 |
iv | +0.094 |
od | +0.094 |
Sus | +0.094 |
Neg logit contributions | |
older | -0.053 |
swe | -0.041 |
heavier | -0.039 |
old | -0.039 |
stable | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | -4.63e-02 |
Pos logit contributions | |
anish | +0.057 |
iv | +0.055 |
od | +0.055 |
And | +0.051 |
iolet | +0.051 |
Neg logit contributions | |
older | -0.063 |
heavier | -0.052 |
swe | -0.052 |
stays | -0.051 |
sle | -0.049 |
Token | and |
Token ID | 290 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
anish | +0.063 |
od | +0.061 |
iv | +0.061 |
And | +0.058 |
Jack | +0.057 |
Neg logit contributions | |
older | -0.100 |
heavier | -0.085 |
swe | -0.085 |
stays | -0.085 |
goes | -0.081 |
Token | though |
Token ID | 996 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
od | +0.023 |
anish | +0.023 |
iv | +0.022 |
Jack | +0.021 |
And | +0.021 |
Neg logit contributions | |
older | -0.031 |
swe | -0.026 |
heavier | -0.026 |
stays | -0.026 |
cooler | -0.025 |
Token | she |
Token ID | 673 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
anish | +0.025 |
od | +0.024 |
iv | +0.024 |
Jack | +0.023 |
J | +0.023 |
Neg logit contributions | |
older | -0.045 |
heavier | -0.037 |
sle | -0.037 |
stays | -0.037 |
reins | -0.037 |
Token | didn |
Token ID | 1422 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
Jack | +0.017 |
od | +0.017 |
iv | +0.017 |
And | +0.016 |
J | +0.016 |
Neg logit contributions | |
older | -0.031 |
heavier | -0.028 |
sle | -0.028 |
swe | -0.027 |
reins | -0.026 |
Token | 't |
Token ID | 470 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
anish | +0.018 |
iv | +0.017 |
od | +0.017 |
iolet | +0.016 |
And | +0.015 |
Neg logit contributions | |
older | -0.033 |
heavier | -0.028 |
swe | -0.028 |
stays | -0.028 |
bigger | -0.027 |
Token | play |
Token ID | 711 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
anish | +0.030 |
iv | +0.029 |
od | +0.029 |
Jack | +0.028 |
And | +0.028 |
Neg logit contributions | |
older | -0.026 |
swe | -0.021 |
stays | -0.020 |
heavier | -0.020 |
sle | -0.019 |
Token | with |
Token ID | 351 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
anish | +0.036 |
iv | +0.035 |
od | +0.034 |
iolet | +0.033 |
Sus | +0.033 |
Neg logit contributions | |
older | -0.043 |
swe | -0.036 |
heavier | -0.036 |
cooler | -0.035 |
bigger | -0.035 |
Token | anymore |
Token ID | 7471 |
Feature activation | -8.01e-02 |
Pos logit contributions | |
anish | +0.043 |
iv | +0.042 |
od | +0.041 |
bed | +0.039 |
iolet | +0.038 |
Neg logit contributions | |
older | -0.060 |
stays | -0.051 |
sle | -0.050 |
heavier | -0.050 |
swe | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
anish | +0.168 |
iv | +0.156 |
Jack | +0.155 |
Sus | +0.154 |
iolet | +0.154 |
Neg logit contributions | |
older | -0.117 |
swe | -0.093 |
heavier | -0.091 |
old | -0.087 |
bigger | -0.086 |
Token | The |
Token ID | 383 |
Feature activation | +5.42e-03 |
Pos logit contributions | |
anish | +0.016 |
iv | +0.015 |
od | +0.015 |
And | +0.014 |
iolet | +0.014 |
Neg logit contributions | |
older | -0.021 |
heavier | -0.018 |
swe | -0.017 |
stays | -0.017 |
sle | -0.017 |
Token | next |
Token ID | 1306 |
Feature activation | -1.00e-03 |
Pos logit contributions | |
older | +0.009 |
stays | +0.008 |
heavier | +0.007 |
swe | +0.007 |
sle | +0.007 |
Neg logit contributions | |
anish | -0.010 |
iv | -0.010 |
Jack | -0.009 |
And | -0.009 |
od | -0.009 |
Token | day |
Token ID | 1110 |
Feature activation | -3.16e-02 |
Pos logit contributions | |
anish | +0.002 |
iv | +0.002 |
od | +0.002 |
iolet | +0.001 |
Jack | +0.001 |
Neg logit contributions | |
older | -0.002 |
heavier | -0.002 |
swe | -0.002 |
stays | -0.002 |
bigger | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
anish | +0.044 |
Jack | +0.044 |
J | +0.043 |
od | +0.043 |
iv | +0.041 |
Neg logit contributions | |
older | -0.063 |
sle | -0.059 |
reins | -0.058 |
stays | -0.058 |
heavier | -0.057 |
Token | play |
Token ID | 711 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
older | +0.030 |
swe | +0.024 |
stays | +0.024 |
heavier | +0.024 |
sle | +0.023 |
Neg logit contributions | |
anish | -0.033 |
od | -0.032 |
iv | -0.032 |
Jack | -0.030 |
And | -0.030 |
Token | with |
Token ID | 351 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
anish | +0.040 |
iv | +0.039 |
od | +0.038 |
iolet | +0.037 |
Jack | +0.037 |
Neg logit contributions | |
older | -0.043 |
swe | -0.036 |
heavier | -0.036 |
bigger | -0.036 |
old | -0.035 |
Token | her |
Token ID | 607 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
anish | +0.039 |
od | +0.037 |
iv | +0.037 |
iolet | +0.035 |
bed | +0.034 |
Neg logit contributions | |
older | -0.046 |
heavier | -0.039 |
swe | -0.038 |
stays | -0.038 |
sle | -0.037 |
Token | blue |
Token ID | 4171 |
Feature activation | -9.82e-03 |
Pos logit contributions | |
anish | +0.041 |
iv | +0.040 |
od | +0.039 |
iolet | +0.037 |
And | +0.037 |
Neg logit contributions | |
older | -0.043 |
heavier | -0.036 |
stays | -0.036 |
swe | -0.034 |
goes | -0.034 |
Token | ball |
Token ID | 2613 |
Feature activation | -9.84e-03 |
Pos logit contributions | |
anish | +0.018 |
iv | +0.017 |
od | +0.016 |
bed | +0.016 |
iolet | +0.016 |
Neg logit contributions | |
older | -0.017 |
heavier | -0.015 |
stays | -0.015 |
lower | -0.014 |
swe | -0.014 |
Token | anymore |
Token ID | 7471 |
Feature activation | -7.97e-02 |
Pos logit contributions | |
anish | +0.015 |
iv | +0.015 |
od | +0.015 |
iolet | +0.014 |
oon | +0.014 |
Neg logit contributions | |
older | -0.019 |
stays | -0.016 |
swe | -0.016 |
heavier | -0.016 |
sle | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
anish | +0.196 |
Sus | +0.186 |
Jack | +0.184 |
iolet | +0.181 |
Jane | +0.179 |
Neg logit contributions | |
older | -0.079 |
old | -0.061 |
goes | -0.058 |
swe | -0.058 |
down | -0.058 |
Token | But |
Token ID | 887 |
Feature activation | -3.82e-03 |
Pos logit contributions | |
iv | +0.026 |
anish | +0.026 |
oon | +0.025 |
od | +0.025 |
iolet | +0.024 |
Neg logit contributions | |
older | -0.021 |
goes | -0.018 |
swe | -0.018 |
old | -0.018 |
gets | -0.017 |
Token | she |
Token ID | 673 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
anish | +0.007 |
iv | +0.007 |
od | +0.007 |
oon | +0.006 |
iolet | +0.006 |
Neg logit contributions | |
older | -0.007 |
swe | -0.006 |
heavier | -0.006 |
bigger | -0.005 |
stays | -0.005 |
Token | learned |
Token ID | 4499 |
Feature activation | -3.10e-03 |
Pos logit contributions | |
anish | +0.024 |
od | +0.022 |
iv | +0.022 |
Sus | +0.021 |
iolet | +0.021 |
Neg logit contributions | |
older | -0.019 |
stays | -0.015 |
goes | -0.015 |
swe | -0.014 |
asks | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
anish | +0.006 |
od | +0.005 |
oon | +0.005 |
iv | +0.005 |
iolet | +0.005 |
Neg logit contributions | |
older | -0.006 |
swe | -0.005 |
old | -0.005 |
bigger | -0.005 |
heavier | -0.005 |
Token | special |
Token ID | 2041 |
Feature activation | -4.87e-03 |
Pos logit contributions | |
And | +0.001 |
anish | +0.001 |
Jack | +0.001 |
Da | +0.001 |
iv | +0.001 |
Neg logit contributions | |
older | -0.001 |
stays | -0.001 |
bars | -0.001 |
asks | -0.001 |
goes | -0.001 |
Token | place |
Token ID | 1295 |
Feature activation | +5.94e-03 |
Pos logit contributions | |
anish | +0.008 |
iv | +0.008 |
iolet | +0.007 |
And | +0.007 |
Jane | +0.007 |
Neg logit contributions | |
older | -0.009 |
stays | -0.008 |
heavier | -0.008 |
asks | -0.007 |
bars | -0.007 |
Token | in |
Token ID | 287 |
Feature activation | -8.51e-03 |
Pos logit contributions | |
older | +0.012 |
stays | +0.010 |
heavier | +0.010 |
swe | +0.010 |
sle | +0.010 |
Neg logit contributions | |
anish | -0.009 |
iv | -0.009 |
od | -0.008 |
iolet | -0.008 |
And | -0.008 |
Token | Lily |
Token ID | 20037 |
Feature activation | -5.56e-02 |
Pos logit contributions | |
anish | +0.016 |
iv | +0.016 |
od | +0.015 |
J | +0.015 |
Sus | +0.015 |
Neg logit contributions | |
older | -0.014 |
stays | -0.012 |
heavier | -0.011 |
swe | -0.011 |
sle | -0.011 |
Token | â |
Token ID | 22940 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
od | +0.092 |
iv | +0.091 |
anish | +0.090 |
bed | +0.086 |
iolet | +0.085 |
Neg logit contributions | |
older | -0.099 |
heavier | -0.092 |
stays | -0.082 |
brakes | -0.080 |
bigger | -0.079 |
Token | € |
Token ID | 26391 |
Feature activation | -7.93e-02 |
Pos logit contributions | |
older | +0.028 |
heavier | +0.023 |
swe | +0.023 |
stays | +0.023 |
bigger | +0.022 |
Neg logit contributions | |
anish | -0.020 |
iv | -0.019 |
od | -0.019 |
iolet | -0.018 |
And | -0.018 |
Token | ™ |
Token ID | 8151 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
anish | +0.132 |
iv | +0.126 |
od | +0.124 |
iolet | +0.117 |
Jack | +0.116 |
Neg logit contributions | |
older | -0.151 |
swe | -0.125 |
stays | -0.125 |
heavier | -0.123 |
goes | -0.120 |
Token | s |
Token ID | 82 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
anish | +0.018 |
iv | +0.017 |
od | +0.017 |
iolet | +0.016 |
Jack | +0.016 |
Neg logit contributions | |
older | -0.020 |
swe | -0.017 |
heavier | -0.017 |
stays | -0.017 |
sle | -0.016 |
Token | room |
Token ID | 2119 |
Feature activation | -9.55e-04 |
Pos logit contributions | |
anish | +0.022 |
iv | +0.021 |
bed | +0.020 |
And | +0.020 |
iolet | +0.019 |
Neg logit contributions | |
older | -0.018 |
heavier | -0.016 |
stays | -0.014 |
swe | -0.014 |
gs | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -6.14e-03 |
Pos logit contributions | |
anish | +0.001 |
iv | +0.001 |
od | +0.001 |
iolet | +0.001 |
And | +0.001 |
Neg logit contributions | |
older | -0.002 |
heavier | -0.002 |
swe | -0.002 |
stays | -0.002 |
sle | -0.002 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
anish | +0.009 |
iv | +0.009 |
od | +0.008 |
iolet | +0.008 |
And | +0.008 |
Neg logit contributions | |
older | -0.013 |
heavier | -0.011 |
swe | -0.011 |
stays | -0.011 |
bigger | -0.010 |
Token | not |
Token ID | 407 |
Feature activation | +8.80e-03 |
Pos logit contributions | |
anish | +0.039 |
iv | +0.039 |
Da | +0.039 |
oon | +0.038 |
Sus | +0.038 |
Neg logit contributions | |
older | -0.031 |
bigger | -0.026 |
heavier | -0.024 |
gets | -0.023 |
old | -0.023 |
Token | enjoy |
Token ID | 2883 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
older | +0.013 |
heavier | +0.010 |
bigger | +0.010 |
goes | +0.010 |
lower | +0.010 |
Neg logit contributions | |
anish | -0.019 |
iv | -0.018 |
iolet | -0.018 |
bed | -0.017 |
od | -0.017 |
Token | playing |
Token ID | 2712 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
anish | +0.034 |
iv | +0.032 |
od | +0.032 |
iolet | +0.031 |
Jack | +0.031 |
Neg logit contributions | |
older | -0.029 |
heavier | -0.023 |
swe | -0.023 |
stays | -0.023 |
bigger | -0.022 |
Token | with |
Token ID | 351 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
iv | +0.038 |
anish | +0.038 |
od | +0.037 |
Sus | +0.036 |
iolet | +0.036 |
Neg logit contributions | |
older | -0.039 |
heavier | -0.032 |
bigger | -0.032 |
swe | -0.031 |
old | -0.030 |
Token | blocks |
Token ID | 7021 |
Feature activation | +5.35e-03 |
Pos logit contributions | |
anish | +0.042 |
iv | +0.040 |
od | +0.040 |
Jack | +0.038 |
iolet | +0.038 |
Neg logit contributions | |
older | -0.052 |
stays | -0.043 |
heavier | -0.043 |
swe | -0.043 |
sle | -0.041 |
Token | anymore |
Token ID | 7471 |
Feature activation | -7.91e-02 |
Pos logit contributions | |
older | +0.010 |
heavier | +0.008 |
swe | +0.008 |
stays | +0.008 |
bigger | +0.008 |
Neg logit contributions | |
anish | -0.009 |
iv | -0.009 |
od | -0.009 |
iolet | -0.008 |
And | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +9.88e-03 |
Pos logit contributions | |
anish | +0.201 |
Sus | +0.197 |
iolet | +0.191 |
Jack | +0.189 |
Da | +0.189 |
Neg logit contributions | |
older | -0.068 |
goes | -0.049 |
asks | -0.048 |
old | -0.047 |
swe | -0.046 |
Token |
|
Token ID | 198 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
older | +0.018 |
swe | +0.015 |
old | +0.015 |
heavier | +0.015 |
bigger | +0.014 |
Neg logit contributions | |
od | -0.017 |
anish | -0.017 |
iv | -0.017 |
oon | -0.016 |
iolet | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | +1.87e-04 |
Pos logit contributions | |
older | +0.034 |
old | +0.029 |
bigger | +0.028 |
goes | +0.027 |
heavy | +0.026 |
Neg logit contributions | |
anish | -0.040 |
iv | -0.037 |
od | -0.037 |
iolet | -0.035 |
itive | -0.034 |
Token | S |
Token ID | 50 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
older | +0.000 |
lower | +0.000 |
swe | +0.000 |
heavier | +0.000 |
stable | +0.000 |
Neg logit contributions | |
od | -0.000 |
anish | -0.000 |
iv | -0.000 |
iolet | -0.000 |
oon | -0.000 |
Token | ara |
Token ID | 3301 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
older | +0.029 |
heavier | +0.026 |
swe | +0.025 |
stays | +0.024 |
sle | +0.024 |
Neg logit contributions | |
anish | -0.022 |
iv | -0.021 |
od | -0.020 |
iolet | -0.020 |
Jack | -0.020 |
Token | to |
Token ID | 284 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
od | +0.005 |
anish | +0.005 |
iv | +0.004 |
Sus | +0.004 |
Jane | +0.004 |
Neg logit contributions | |
older | -0.004 |
sle | -0.004 |
heavier | -0.004 |
gs | -0.004 |
stays | -0.004 |
Token | join |
Token ID | 4654 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
anish | +0.045 |
iv | +0.043 |
od | +0.043 |
iolet | +0.041 |
Jack | +0.041 |
Neg logit contributions | |
older | -0.036 |
heavier | -0.028 |
swe | -0.028 |
stays | -0.028 |
bigger | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
anish | +0.040 |
iv | +0.038 |
od | +0.038 |
iolet | +0.035 |
Jack | +0.035 |
Neg logit contributions | |
older | -0.046 |
heavier | -0.040 |
swe | -0.039 |
stays | -0.039 |
sle | -0.037 |
Token | tooth |
Token ID | 16162 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
older | +0.020 |
heavier | +0.017 |
stays | +0.017 |
swe | +0.016 |
lower | +0.016 |
Neg logit contributions | |
anish | -0.019 |
od | -0.019 |
iv | -0.018 |
Jack | -0.018 |
And | -0.017 |
Token | club |
Token ID | 3430 |
Feature activation | -8.71e-03 |
Pos logit contributions | |
anish | +0.028 |
od | +0.027 |
iv | +0.027 |
iolet | +0.026 |
oon | +0.025 |
Neg logit contributions | |
older | -0.032 |
heavier | -0.027 |
swe | -0.027 |
stays | -0.026 |
sle | -0.025 |
Token | anymore |
Token ID | 7471 |
Feature activation | -7.89e-02 |
Pos logit contributions | |
anish | +0.014 |
od | +0.014 |
iv | +0.014 |
iolet | +0.013 |
And | +0.013 |
Neg logit contributions | |
older | -0.017 |
heavier | -0.014 |
stays | -0.014 |
swe | -0.014 |
sle | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
anish | +0.159 |
iv | +0.147 |
Jack | +0.146 |
Sus | +0.145 |
iolet | +0.144 |
Neg logit contributions | |
older | -0.120 |
swe | -0.095 |
stays | -0.093 |
goes | -0.093 |
heavier | -0.092 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
iv | +0.045 |
anish | +0.044 |
od | +0.043 |
bed | +0.041 |
oon | +0.040 |
Neg logit contributions | |
older | -0.040 |
younger | -0.031 |
old | -0.031 |
swe | -0.031 |
goes | -0.031 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
anish | +0.035 |
iv | +0.034 |
od | +0.034 |
And | +0.032 |
iolet | +0.032 |
Neg logit contributions | |
older | -0.026 |
goes | -0.020 |
cooler | -0.020 |
stays | -0.020 |
heavier | -0.020 |
Token | upon |
Token ID | 2402 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
older | +0.017 |
heavier | +0.014 |
swe | +0.014 |
stays | +0.013 |
sle | +0.013 |
Neg logit contributions | |
anish | -0.024 |
od | -0.024 |
iv | -0.023 |
Jack | -0.023 |
iolet | -0.023 |
Token | a |
Token ID | 257 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
anish | +0.056 |
iv | +0.052 |
od | +0.051 |
iolet | +0.049 |
bed | +0.049 |
Neg logit contributions | |
older | -0.023 |
bigger | -0.018 |
old | -0.018 |
heavy | -0.017 |
cooler | -0.017 |
Token | 't |
Token ID | 470 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
anish | +0.012 |
iv | +0.011 |
od | +0.011 |
iolet | +0.011 |
J | +0.011 |
Neg logit contributions | |
older | -0.013 |
heavier | -0.010 |
bigger | -0.010 |
goes | -0.010 |
passes | -0.010 |
Token | end |
Token ID | 886 |
Feature activation | +6.42e-03 |
Pos logit contributions | |
anish | +0.032 |
iv | +0.032 |
oon | +0.031 |
iolet | +0.031 |
od | +0.031 |
Neg logit contributions | |
older | -0.019 |
heavier | -0.014 |
swe | -0.013 |
bigger | -0.013 |
goes | -0.013 |
Token | up |
Token ID | 510 |
Feature activation | -3.07e-04 |
Pos logit contributions | |
older | +0.010 |
swe | +0.008 |
heavier | +0.008 |
old | +0.007 |
bigger | +0.007 |
Neg logit contributions | |
anish | -0.014 |
iv | -0.013 |
iolet | -0.013 |
oon | -0.013 |
Jane | -0.012 |
Token | playing |
Token ID | 2712 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
anish | +0.001 |
iv | +0.001 |
oon | +0.001 |
od | +0.001 |
And | +0.001 |
Neg logit contributions | |
older | -0.001 |
heavier | -0.000 |
old | -0.000 |
bigger | -0.000 |
swe | -0.000 |
Token | together |
Token ID | 1978 |
Feature activation | -3.79e-02 |
Pos logit contributions | |
iv | +0.050 |
od | +0.047 |
anish | +0.046 |
oon | +0.046 |
Maria | +0.045 |
Neg logit contributions | |
older | -0.054 |
down | -0.047 |
heavier | -0.046 |
back | -0.046 |
bigger | -0.046 |
Token | anymore |
Token ID | 7471 |
Feature activation | -7.88e-02 |
Pos logit contributions | |
anish | +0.075 |
iv | +0.074 |
od | +0.071 |
Sus | +0.070 |
iolet | +0.070 |
Neg logit contributions | |
older | -0.060 |
heavier | -0.048 |
swe | -0.047 |
bigger | -0.046 |
stays | -0.045 |
Token | and |
Token ID | 290 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
anish | +0.214 |
Rachel | +0.212 |
iolet | +0.207 |
Ta | +0.206 |
Da | +0.205 |
Neg logit contributions | |
older | -0.037 |
gets | -0.031 |
down | -0.027 |
old | -0.027 |
comes | -0.025 |
Token | Tim |
Token ID | 5045 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
anish | +0.018 |
iv | +0.017 |
iolet | +0.016 |
oon | +0.016 |
od | +0.016 |
Neg logit contributions | |
older | -0.022 |
heavier | -0.019 |
swe | -0.018 |
stays | -0.018 |
goes | -0.018 |
Token | my |
Token ID | 1820 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
anish | +0.042 |
od | +0.041 |
iv | +0.040 |
Jack | +0.039 |
iolet | +0.039 |
Neg logit contributions | |
older | -0.031 |
heavier | -0.026 |
sle | -0.025 |
swe | -0.025 |
stable | -0.024 |
Token | learned |
Token ID | 4499 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
anish | +0.052 |
iv | +0.050 |
od | +0.048 |
iolet | +0.047 |
oon | +0.047 |
Neg logit contributions | |
older | -0.061 |
stays | -0.051 |
swe | -0.050 |
heavier | -0.050 |
goes | -0.049 |
Token | a |
Token ID | 257 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
anish | +0.027 |
oon | +0.027 |
od | +0.026 |
iolet | +0.025 |
iv | +0.025 |
Neg logit contributions | |
older | -0.019 |
old | -0.017 |
swe | -0.016 |
bigger | -0.016 |
heavier | -0.015 |
Token | special |
Token ID | 2041 |
Feature activation | -6.38e-03 |
Pos logit contributions | |
anish | +0.001 |
iv | +0.001 |
od | +0.001 |
And | +0.000 |
Jack | +0.000 |
Neg logit contributions | |
older | -0.000 |
stays | -0.000 |
heavier | -0.000 |
swe | -0.000 |
goes | -0.000 |
Token | spot |
Token ID | 4136 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
anish | +0.012 |
iv | +0.011 |
od | +0.011 |
iolet | +0.011 |
And | +0.011 |
Neg logit contributions | |
older | -0.011 |
heavier | -0.009 |
swe | -0.009 |
stays | -0.009 |
bigger | -0.008 |
Token | on |
Token ID | 319 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
older | +0.026 |
heavier | +0.021 |
swe | +0.020 |
stays | +0.020 |
bigger | +0.019 |
Neg logit contributions | |
anish | -0.028 |
iv | -0.027 |
od | -0.027 |
iolet | -0.026 |
And | -0.026 |
Token | Lily |
Token ID | 20037 |
Feature activation | -4.52e-02 |
Pos logit contributions | |
anish | +0.026 |
iv | +0.025 |
od | +0.025 |
iolet | +0.024 |
Jack | +0.024 |
Neg logit contributions | |
older | -0.019 |
heavier | -0.014 |
swe | -0.014 |
stays | -0.014 |
bigger | -0.013 |
Token | â |
Token ID | 22940 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
anish | +0.091 |
od | +0.090 |
iv | +0.090 |
bed | +0.086 |
Jack | +0.084 |
Neg logit contributions | |
older | -0.069 |
heavier | -0.056 |
swe | -0.054 |
stays | -0.054 |
bigger | -0.051 |
Token | € |
Token ID | 26391 |
Feature activation | -7.86e-02 |
Pos logit contributions | |
older | +0.026 |
swe | +0.022 |
heavier | +0.022 |
stays | +0.022 |
goes | +0.021 |
Neg logit contributions | |
anish | -0.020 |
iv | -0.019 |
od | -0.019 |
iolet | -0.018 |
And | -0.018 |
Token | ™ |
Token ID | 8151 |
Feature activation | -9.25e-03 |
Pos logit contributions | |
anish | +0.121 |
iv | +0.114 |
od | +0.110 |
Sus | +0.104 |
Maria | +0.102 |
Neg logit contributions | |
older | -0.152 |
stays | -0.134 |
goes | -0.131 |
passes | -0.128 |
opens | -0.128 |
Token | s |
Token ID | 82 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
anish | +0.016 |
iv | +0.016 |
od | +0.015 |
And | +0.015 |
iolet | +0.015 |
Neg logit contributions | |
older | -0.017 |
heavier | -0.014 |
swe | -0.014 |
stays | -0.013 |
bigger | -0.013 |
Token | shelf |
Token ID | 18316 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
anish | +0.023 |
iv | +0.022 |
od | +0.022 |
iolet | +0.021 |
And | +0.021 |
Neg logit contributions | |
older | -0.021 |
heavier | -0.017 |
swe | -0.017 |
stays | -0.016 |
goes | -0.016 |
Token | where |
Token ID | 810 |
Feature activation | +7.36e-03 |
Pos logit contributions | |
older | +0.023 |
swe | +0.020 |
stays | +0.020 |
heavier | +0.019 |
sle | +0.019 |
Neg logit contributions | |
anish | -0.018 |
iv | -0.018 |
od | -0.017 |
iolet | -0.016 |
oon | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
older | +0.016 |
sle | +0.015 |
heavier | +0.014 |
reins | +0.014 |
lower | +0.014 |
Neg logit contributions | |
iv | -0.009 |
od | -0.009 |
Sus | -0.009 |
anish | -0.008 |
J | -0.008 |
Token | 't |
Token ID | 470 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
anish | +0.020 |
iv | +0.020 |
od | +0.019 |
And | +0.018 |
Jack | +0.018 |
Neg logit contributions | |
older | -0.049 |
swe | -0.044 |
stays | -0.043 |
heavier | -0.043 |
sle | -0.042 |
Token | frightened |
Token ID | 24776 |
Feature activation | -9.74e-04 |
Pos logit contributions | |
od | +0.024 |
iv | +0.022 |
anish | +0.022 |
Sus | +0.022 |
Jane | +0.021 |
Neg logit contributions | |
older | -0.017 |
sle | -0.015 |
stays | -0.015 |
swe | -0.015 |
goes | -0.014 |
Token | of |
Token ID | 286 |
Feature activation | -8.00e-03 |
Pos logit contributions | |
od | +0.001 |
anish | +0.001 |
iv | +0.001 |
Jack | +0.001 |
Sus | +0.001 |
Neg logit contributions | |
older | -0.002 |
stays | -0.002 |
sle | -0.002 |
heavier | -0.002 |
swe | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | +3.14e-03 |
Pos logit contributions | |
od | +0.012 |
anish | +0.012 |
iv | +0.011 |
oon | +0.010 |
And | +0.010 |
Neg logit contributions | |
older | -0.015 |
swe | -0.013 |
sle | -0.013 |
stays | -0.013 |
heavier | -0.013 |
Token | dark |
Token ID | 3223 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
older | +0.005 |
heavier | +0.004 |
swe | +0.004 |
stays | +0.004 |
bigger | +0.004 |
Neg logit contributions | |
anish | -0.006 |
iv | -0.006 |
od | -0.006 |
iolet | -0.006 |
And | -0.006 |
Token | anymore |
Token ID | 7471 |
Feature activation | -7.86e-02 |
Pos logit contributions | |
anish | +0.073 |
iv | +0.071 |
iolet | +0.068 |
od | +0.068 |
And | +0.067 |
Neg logit contributions | |
older | -0.070 |
heavier | -0.056 |
old | -0.055 |
bigger | -0.055 |
stays | -0.055 |
Token | . |
Token ID | 13 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
anish | +0.169 |
Sus | +0.157 |
Jack | +0.155 |
iv | +0.155 |
iolet | +0.153 |
Neg logit contributions | |
older | -0.113 |
heavier | -0.085 |
swe | -0.085 |
old | -0.084 |
goes | -0.083 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
anish | +0.021 |
iv | +0.020 |
od | +0.020 |
iolet | +0.019 |
oon | +0.019 |
Neg logit contributions | |
older | -0.021 |
swe | -0.017 |
goes | -0.017 |
stays | -0.017 |
old | -0.017 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
anish | +0.024 |
od | +0.023 |
iv | +0.023 |
bed | +0.021 |
Sus | +0.021 |
Neg logit contributions | |
older | -0.022 |
goes | -0.018 |
cooler | -0.018 |
bigger | -0.018 |
old | -0.017 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
older | +0.026 |
heavier | +0.022 |
swe | +0.022 |
stays | +0.022 |
sle | +0.021 |
Neg logit contributions | |
anish | -0.037 |
od | -0.037 |
Jack | -0.035 |
iv | -0.035 |
J | -0.035 |
Token | a |
Token ID | 257 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
anish | +0.032 |
iv | +0.030 |
od | +0.029 |
iolet | +0.028 |
bed | +0.028 |
Neg logit contributions | |
older | -0.012 |
bigger | -0.010 |
heavy | -0.010 |
old | -0.010 |
cooler | -0.010 |
Token | 't |
Token ID | 470 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
anish | +0.005 |
iv | +0.005 |
od | +0.005 |
And | +0.005 |
Jack | +0.005 |
Neg logit contributions | |
older | -0.011 |
swe | -0.010 |
stays | -0.010 |
heavier | -0.010 |
sle | -0.009 |
Token | mind |
Token ID | 2000 |
Feature activation | -3.67e-03 |
Pos logit contributions | |
anish | +0.027 |
iv | +0.026 |
od | +0.026 |
Jack | +0.024 |
iolet | +0.024 |
Neg logit contributions | |
older | -0.025 |
swe | -0.020 |
heavier | -0.020 |
stays | -0.019 |
sle | -0.019 |
Token | exploring |
Token ID | 13504 |
Feature activation | -7.15e-03 |
Pos logit contributions | |
od | +0.005 |
iv | +0.005 |
anish | +0.005 |
Sus | +0.004 |
bed | +0.004 |
Neg logit contributions | |
older | -0.007 |
swe | -0.007 |
stays | -0.006 |
heavier | -0.006 |
sle | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | -9.07e-03 |
Pos logit contributions | |
anish | +0.011 |
iv | +0.011 |
od | +0.010 |
And | +0.010 |
Jack | +0.010 |
Neg logit contributions | |
older | -0.014 |
stays | -0.012 |
heavier | -0.011 |
swe | -0.011 |
sle | -0.011 |
Token | backyard |
Token ID | 24296 |
Feature activation | -9.81e-03 |
Pos logit contributions | |
And | +0.012 |
anish | +0.012 |
Jack | +0.012 |
iv | +0.011 |
All | +0.011 |
Neg logit contributions | |
older | -0.018 |
heavier | -0.016 |
stays | -0.016 |
swe | -0.015 |
sle | -0.015 |
Token | anymore |
Token ID | 7471 |
Feature activation | -7.85e-02 |
Pos logit contributions | |
anish | +0.015 |
od | +0.015 |
iv | +0.015 |
And | +0.014 |
iolet | +0.014 |
Neg logit contributions | |
older | -0.020 |
swe | -0.017 |
stays | -0.017 |
heavier | -0.016 |
sle | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
anish | +0.152 |
iv | +0.143 |
od | +0.139 |
iolet | +0.138 |
And | +0.137 |
Neg logit contributions | |
older | -0.130 |
heavier | -0.105 |
swe | -0.102 |
stays | -0.101 |
goes | -0.099 |
Token | He |
Token ID | 679 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
anish | +0.034 |
iv | +0.033 |
od | +0.032 |
iolet | +0.031 |
And | +0.031 |
Neg logit contributions | |
older | -0.029 |
heavier | -0.023 |
swe | -0.023 |
stays | -0.023 |
bigger | -0.022 |
Token | soon |
Token ID | 2582 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
iv | +0.029 |
anish | +0.029 |
od | +0.028 |
And | +0.028 |
Jack | +0.028 |
Neg logit contributions | |
older | -0.049 |
heavier | -0.042 |
swe | -0.041 |
sle | -0.040 |
younger | -0.039 |
Token | remembered |
Token ID | 12086 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
anish | +0.039 |
od | +0.039 |
iv | +0.038 |
Jack | +0.037 |
iolet | +0.036 |
Neg logit contributions | |
older | -0.052 |
heavier | -0.044 |
swe | -0.044 |
stays | -0.042 |
sle | -0.042 |
Token | that |
Token ID | 326 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
anish | +0.027 |
iv | +0.026 |
od | +0.026 |
Jack | +0.025 |
And | +0.024 |
Neg logit contributions | |
older | -0.030 |
heavier | -0.025 |
swe | -0.025 |
stays | -0.025 |
sle | -0.024 |
Token | celebrate |
Token ID | 10648 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
anish | +0.046 |
iv | +0.045 |
od | +0.044 |
iolet | +0.042 |
And | +0.042 |
Neg logit contributions | |
older | -0.053 |
heavier | -0.044 |
swe | -0.043 |
stays | -0.043 |
bigger | -0.041 |
Token | the |
Token ID | 262 |
Feature activation | -1.67e-03 |
Pos logit contributions | |
anish | +0.052 |
iv | +0.050 |
od | +0.049 |
iolet | +0.047 |
And | +0.046 |
Neg logit contributions | |
older | -0.059 |
heavier | -0.049 |
swe | -0.048 |
stays | -0.048 |
bigger | -0.046 |
Token | old |
Token ID | 1468 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
anish | +0.003 |
iv | +0.003 |
od | +0.003 |
Jack | +0.003 |
And | +0.003 |
Neg logit contributions | |
older | -0.003 |
stays | -0.003 |
heavier | -0.003 |
swe | -0.002 |
gs | -0.002 |
Token | woman |
Token ID | 2415 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
anish | +0.082 |
iv | +0.080 |
od | +0.080 |
iolet | +0.076 |
And | +0.076 |
Neg logit contributions | |
older | -0.052 |
heavier | -0.040 |
swe | -0.039 |
stays | -0.039 |
bigger | -0.036 |
Token | â |
Token ID | 22940 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
od | +0.050 |
iv | +0.048 |
anish | +0.047 |
And | +0.046 |
iolet | +0.044 |
Neg logit contributions | |
older | -0.057 |
stays | -0.048 |
heavier | -0.048 |
swe | -0.048 |
lower | -0.045 |
Token | € |
Token ID | 26391 |
Feature activation | -7.85e-02 |
Pos logit contributions | |
older | +0.023 |
heavier | +0.019 |
swe | +0.019 |
stays | +0.019 |
bigger | +0.019 |
Neg logit contributions | |
anish | -0.016 |
iv | -0.015 |
od | -0.015 |
iolet | -0.014 |
And | -0.014 |
Token | ™ |
Token ID | 8151 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
anish | +0.121 |
iv | +0.113 |
od | +0.108 |
iolet | +0.104 |
Jack | +0.101 |
Neg logit contributions | |
older | -0.159 |
stays | -0.133 |
swe | -0.132 |
heavier | -0.131 |
goes | -0.131 |
Token | s |
Token ID | 82 |
Feature activation | -4.92e-03 |
Pos logit contributions | |
anish | +0.025 |
iv | +0.025 |
od | +0.025 |
iolet | +0.023 |
Jack | +0.023 |
Neg logit contributions | |
older | -0.032 |
stays | -0.027 |
swe | -0.027 |
heavier | -0.027 |
sle | -0.026 |
Token | success |
Token ID | 1943 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
anish | +0.008 |
iv | +0.008 |
od | +0.008 |
And | +0.007 |
iolet | +0.007 |
Neg logit contributions | |
older | -0.009 |
heavier | -0.008 |
stays | -0.008 |
swe | -0.008 |
sle | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
anish | +0.039 |
iv | +0.039 |
od | +0.038 |
iolet | +0.035 |
And | +0.035 |
Neg logit contributions | |
older | -0.053 |
heavier | -0.045 |
swe | -0.044 |
stays | -0.044 |
sle | -0.043 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
anish | +0.014 |
iv | +0.014 |
od | +0.013 |
And | +0.013 |
Jack | +0.013 |
Neg logit contributions | |
older | -0.026 |
heavier | -0.023 |
swe | -0.022 |
stays | -0.022 |
bigger | -0.022 |
Token | be |
Token ID | 307 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
anish | +0.035 |
iv | +0.034 |
od | +0.034 |
iolet | +0.033 |
And | +0.032 |
Neg logit contributions | |
older | -0.028 |
heavier | -0.022 |
swe | -0.022 |
stays | -0.022 |
bigger | -0.021 |
Token | jealous |
Token ID | 19354 |
Feature activation | -2.55e-02 |
Pos logit contributions | |
od | +0.032 |
iv | +0.030 |
anish | +0.029 |
J | +0.029 |
Jack | +0.029 |
Neg logit contributions | |
sle | -0.034 |
bars | -0.033 |
reins | -0.031 |
older | -0.030 |
gs | -0.029 |
Token | of |
Token ID | 286 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
anish | +0.042 |
iv | +0.040 |
od | +0.040 |
iolet | +0.037 |
Jack | +0.036 |
Neg logit contributions | |
older | -0.047 |
sle | -0.041 |
heavier | -0.041 |
stays | -0.041 |
swe | -0.040 |
Token | other |
Token ID | 584 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
anish | +0.025 |
iv | +0.025 |
od | +0.025 |
Jack | +0.023 |
And | +0.022 |
Neg logit contributions | |
older | -0.027 |
swe | -0.023 |
stays | -0.023 |
heavier | -0.022 |
sle | -0.022 |
Token | kings |
Token ID | 21557 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
od | +0.037 |
anish | +0.034 |
endings | +0.031 |
iv | +0.031 |
J | +0.030 |
Neg logit contributions | |
older | -0.046 |
lower | -0.044 |
heavier | -0.043 |
stays | -0.042 |
shaky | -0.041 |
Token | anymore |
Token ID | 7471 |
Feature activation | -7.85e-02 |
Pos logit contributions | |
anish | +0.025 |
iv | +0.024 |
od | +0.024 |
iolet | +0.023 |
oon | +0.023 |
Neg logit contributions | |
older | -0.029 |
swe | -0.025 |
heavier | -0.024 |
stays | -0.024 |
bigger | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -6.84e-03 |
Pos logit contributions | |
anish | +0.158 |
Sus | +0.148 |
iv | +0.148 |
Jack | +0.146 |
J | +0.145 |
Neg logit contributions | |
older | -0.121 |
heavier | -0.094 |
goes | -0.092 |
stays | -0.091 |
swe | -0.090 |
Token | He |
Token ID | 679 |
Feature activation | -9.28e-03 |
Pos logit contributions | |
anish | +0.014 |
iv | +0.014 |
od | +0.014 |
iolet | +0.013 |
And | +0.013 |
Neg logit contributions | |
older | -0.010 |
heavier | -0.008 |
swe | -0.008 |
stays | -0.008 |
bigger | -0.008 |
Token | was |
Token ID | 373 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
iv | +0.012 |
anish | +0.012 |
od | +0.012 |
And | +0.012 |
Jack | +0.011 |
Neg logit contributions | |
older | -0.020 |
heavier | -0.017 |
swe | -0.017 |
sle | -0.017 |
stays | -0.016 |
Token | happy |
Token ID | 3772 |
Feature activation | -4.30e-02 |
Pos logit contributions | |
od | +0.036 |
iv | +0.034 |
And | +0.034 |
anish | +0.033 |
Sus | +0.033 |
Neg logit contributions | |
older | -0.045 |
swe | -0.042 |
sle | -0.042 |
stays | -0.042 |
gs | -0.041 |
Token | with |
Token ID | 351 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
anish | +0.089 |
iv | +0.085 |
od | +0.083 |
iolet | +0.081 |
Sus | +0.080 |
Neg logit contributions | |
older | -0.068 |
heavier | -0.053 |
swe | -0.052 |
bigger | -0.052 |
stays | -0.051 |
Token | her |
Token ID | 607 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
anish | +0.020 |
iv | +0.019 |
od | +0.019 |
iolet | +0.018 |
Jack | +0.018 |
Neg logit contributions | |
older | -0.025 |
heavier | -0.021 |
swe | -0.021 |
stays | -0.020 |
bigger | -0.020 |
Token | guitar |
Token ID | 10047 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
anish | +0.031 |
od | +0.030 |
iv | +0.029 |
And | +0.028 |
iolet | +0.028 |
Neg logit contributions | |
older | -0.062 |
heavier | -0.056 |
stays | -0.055 |
sle | -0.053 |
swe | -0.053 |
Token | in |
Token ID | 287 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
anish | +0.033 |
iv | +0.031 |
od | +0.031 |
iolet | +0.029 |
And | +0.029 |
Neg logit contributions | |
older | -0.043 |
heavier | -0.036 |
swe | -0.036 |
stays | -0.036 |
sle | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
anish | +0.045 |
iv | +0.043 |
od | +0.043 |
iolet | +0.041 |
And | +0.041 |
Neg logit contributions | |
older | -0.032 |
heavier | -0.025 |
swe | -0.025 |
stays | -0.025 |
bigger | -0.024 |
Token | forest |
Token ID | 8222 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
anish | +0.017 |
And | +0.017 |
iv | +0.017 |
Jack | +0.017 |
All | +0.016 |
Neg logit contributions | |
older | -0.025 |
stays | -0.022 |
heavier | -0.022 |
gs | -0.021 |
bars | -0.021 |
Token | anymore |
Token ID | 7471 |
Feature activation | -7.84e-02 |
Pos logit contributions | |
anish | +0.034 |
iv | +0.033 |
od | +0.033 |
iolet | +0.031 |
And | +0.031 |
Neg logit contributions | |
older | -0.042 |
swe | -0.035 |
stays | -0.035 |
heavier | -0.035 |
sle | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
anish | +0.182 |
Jack | +0.166 |
And | +0.164 |
Sus | +0.163 |
iolet | +0.162 |
Neg logit contributions | |
older | -0.100 |
old | -0.075 |
goes | -0.072 |
heavier | -0.071 |
swe | -0.071 |
Token | She |
Token ID | 1375 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
anish | +0.015 |
iv | +0.015 |
od | +0.014 |
iolet | +0.014 |
And | +0.014 |
Neg logit contributions | |
older | -0.022 |
heavier | -0.019 |
swe | -0.018 |
stays | -0.018 |
cooler | -0.018 |
Token | decided |
Token ID | 3066 |
Feature activation | +2.83e-02 |
Pos logit contributions | |
iv | +0.015 |
anish | +0.014 |
bed | +0.014 |
And | +0.014 |
od | +0.014 |
Neg logit contributions | |
older | -0.026 |
heavier | -0.023 |
swe | -0.022 |
sle | -0.022 |
gs | -0.021 |
Token | to |
Token ID | 284 |
Feature activation | -2.42e-03 |
Pos logit contributions | |
older | +0.047 |
sle | +0.039 |
heavier | +0.038 |
swe | +0.038 |
stays | +0.037 |
Neg logit contributions | |
anish | -0.050 |
iv | -0.049 |
od | -0.049 |
And | -0.048 |
Sus | -0.048 |
Token | take |
Token ID | 1011 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
anish | +0.005 |
iv | +0.005 |
od | +0.005 |
iolet | +0.005 |
And | +0.005 |
Neg logit contributions | |
older | -0.004 |
heavier | -0.003 |
swe | -0.003 |
stays | -0.003 |
bigger | -0.003 |
Token | stubborn |
Token ID | 24444 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
anish | +0.034 |
od | +0.033 |
iv | +0.033 |
iolet | +0.031 |
Jack | +0.031 |
Neg logit contributions | |
older | -0.023 |
stays | -0.018 |
swe | -0.018 |
heavier | -0.018 |
sle | -0.017 |
Token | about |
Token ID | 546 |
Feature activation | -2.42e-02 |
Pos logit contributions | |
anish | +0.030 |
iv | +0.029 |
od | +0.029 |
iolet | +0.028 |
Jack | +0.027 |
Neg logit contributions | |
older | -0.035 |
heavier | -0.029 |
swe | -0.029 |
stays | -0.028 |
bigger | -0.027 |
Token | learning |
Token ID | 4673 |
Feature activation | -8.83e-03 |
Pos logit contributions | |
anish | +0.038 |
iv | +0.037 |
od | +0.037 |
iolet | +0.034 |
Jack | +0.034 |
Neg logit contributions | |
older | -0.048 |
heavier | -0.040 |
swe | -0.040 |
stays | -0.040 |
sle | -0.038 |
Token | new |
Token ID | 649 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
anish | +0.015 |
iv | +0.015 |
od | +0.015 |
And | +0.014 |
iolet | +0.014 |
Neg logit contributions | |
older | -0.016 |
heavier | -0.013 |
stays | -0.013 |
sle | -0.013 |
swe | -0.013 |
Token | things |
Token ID | 1243 |
Feature activation | +6.30e-05 |
Pos logit contributions | |
anish | +0.044 |
iv | +0.043 |
od | +0.043 |
iolet | +0.041 |
And | +0.040 |
Neg logit contributions | |
older | -0.037 |
heavier | -0.030 |
stays | -0.028 |
sle | -0.028 |
swe | -0.028 |
Token | anymore |
Token ID | 7471 |
Feature activation | -7.83e-02 |
Pos logit contributions | |
older | +0.000 |
heavier | +0.000 |
swe | +0.000 |
stays | +0.000 |
sle | +0.000 |
Neg logit contributions | |
anish | -0.000 |
iv | -0.000 |
od | -0.000 |
iolet | -0.000 |
And | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
anish | +0.173 |
Sus | +0.159 |
iolet | +0.159 |
And | +0.157 |
Jack | +0.157 |
Neg logit contributions | |
older | -0.092 |
swe | -0.078 |
goes | -0.071 |
old | -0.070 |
heavier | -0.068 |
Token | She |
Token ID | 1375 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
anish | +0.033 |
iv | +0.031 |
oon | +0.030 |
bed | +0.029 |
iolet | +0.029 |
Neg logit contributions | |
older | -0.025 |
goes | -0.022 |
old | -0.021 |
swe | -0.021 |
gets | -0.020 |
Token | realized |
Token ID | 6939 |
Feature activation | -5.66e-03 |
Pos logit contributions | |
anish | +0.044 |
iv | +0.040 |
Sus | +0.039 |
iolet | +0.039 |
od | +0.039 |
Neg logit contributions | |
older | -0.044 |
goes | -0.037 |
stays | -0.037 |
swe | -0.036 |
bigger | -0.034 |
Token | that |
Token ID | 326 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
anish | +0.009 |
iv | +0.008 |
iolet | +0.008 |
od | +0.008 |
oon | +0.008 |
Neg logit contributions | |
older | -0.012 |
swe | -0.010 |
heavier | -0.010 |
old | -0.010 |
bigger | -0.010 |
Token | it |
Token ID | 340 |
Feature activation | -1.98e-03 |
Pos logit contributions | |
anish | +0.023 |
iv | +0.022 |
iolet | +0.021 |
oon | +0.020 |
od | +0.020 |
Neg logit contributions | |
older | -0.028 |
swe | -0.024 |
heavier | -0.023 |
bigger | -0.023 |
old | -0.022 |
Token | 't |
Token ID | 470 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
iv | +0.024 |
anish | +0.023 |
od | +0.023 |
Jack | +0.022 |
And | +0.022 |
Neg logit contributions | |
older | -0.054 |
swe | -0.048 |
stays | -0.048 |
heavier | -0.047 |
sle | -0.046 |
Token | upset |
Token ID | 9247 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
od | +0.029 |
iv | +0.028 |
anish | +0.028 |
Sus | +0.027 |
Jack | +0.026 |
Neg logit contributions | |
older | -0.022 |
stays | -0.019 |
swe | -0.018 |
sle | -0.018 |
goes | -0.018 |
Token | about |
Token ID | 546 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
anish | +0.041 |
iv | +0.040 |
od | +0.039 |
iolet | +0.037 |
And | +0.037 |
Neg logit contributions | |
older | -0.043 |
heavier | -0.035 |
swe | -0.035 |
stays | -0.034 |
bigger | -0.033 |
Token | eating |
Token ID | 6600 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
anish | +0.040 |
iv | +0.039 |
od | +0.038 |
iolet | +0.036 |
And | +0.036 |
Neg logit contributions | |
older | -0.050 |
heavier | -0.042 |
swe | -0.042 |
stays | -0.041 |
bigger | -0.040 |
Token | shrimp |
Token ID | 30060 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
anish | +0.022 |
iv | +0.021 |
od | +0.021 |
Jack | +0.020 |
And | +0.020 |
Neg logit contributions | |
older | -0.025 |
stays | -0.020 |
sle | -0.020 |
heavier | -0.020 |
younger | -0.019 |
Token | anymore |
Token ID | 7471 |
Feature activation | -7.82e-02 |
Pos logit contributions | |
anish | +0.034 |
iv | +0.033 |
od | +0.033 |
iolet | +0.031 |
And | +0.031 |
Neg logit contributions | |
older | -0.038 |
heavier | -0.032 |
swe | -0.031 |
stays | -0.031 |
bigger | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | -6.38e-03 |
Pos logit contributions | |
anish | +0.175 |
Sus | +0.162 |
J | +0.159 |
iolet | +0.158 |
Jack | +0.158 |
Neg logit contributions | |
older | -0.099 |
swe | -0.080 |
heavier | -0.076 |
old | -0.075 |
goes | -0.074 |
Token | she |
Token ID | 673 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
anish | +0.010 |
iv | +0.009 |
od | +0.009 |
iolet | +0.009 |
Jack | +0.009 |
Neg logit contributions | |
older | -0.013 |
heavier | -0.011 |
swe | -0.011 |
stays | -0.011 |
bigger | -0.010 |
Token | always |
Token ID | 1464 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
anish | +0.027 |
iv | +0.025 |
od | +0.025 |
iolet | +0.024 |
Sus | +0.024 |
Neg logit contributions | |
older | -0.030 |
stays | -0.026 |
swe | -0.025 |
heavier | -0.025 |
goes | -0.025 |
Token | remembered |
Token ID | 12086 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
anish | +0.018 |
iv | +0.017 |
od | +0.017 |
Jack | +0.016 |
iolet | +0.016 |
Neg logit contributions | |
older | -0.022 |
heavier | -0.018 |
swe | -0.018 |
stays | -0.017 |
sle | -0.017 |
Token | to |
Token ID | 284 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
anish | +0.034 |
iv | +0.033 |
od | +0.032 |
Jack | +0.030 |
iolet | +0.030 |
Neg logit contributions | |
older | -0.037 |
heavier | -0.031 |
stays | -0.030 |
swe | -0.030 |
sle | -0.029 |
Token | trees |
Token ID | 7150 |
Feature activation | -4.02e-03 |
Pos logit contributions | |
anish | +0.004 |
iv | +0.004 |
od | +0.004 |
And | +0.003 |
Jack | +0.003 |
Neg logit contributions | |
older | -0.006 |
heavier | -0.005 |
swe | -0.005 |
stays | -0.005 |
sle | -0.005 |
Token | in |
Token ID | 287 |
Feature activation | -6.56e-03 |
Pos logit contributions | |
anish | +0.006 |
iv | +0.006 |
od | +0.006 |
iolet | +0.006 |
And | +0.006 |
Neg logit contributions | |
older | -0.008 |
heavier | -0.007 |
swe | -0.007 |
stays | -0.007 |
bigger | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
anish | +0.013 |
iv | +0.013 |
od | +0.013 |
iolet | +0.012 |
Jack | +0.012 |
Neg logit contributions | |
older | -0.010 |
heavier | -0.008 |
swe | -0.008 |
stays | -0.008 |
bigger | -0.007 |
Token | sun |
Token ID | 4252 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
anish | +0.048 |
And | +0.048 |
iv | +0.047 |
All | +0.047 |
Jack | +0.047 |
Neg logit contributions | |
older | -0.054 |
heavier | -0.044 |
swe | -0.043 |
stays | -0.042 |
younger | -0.042 |
Token | â |
Token ID | 22940 |
Feature activation | +6.49e-03 |
Pos logit contributions | |
anish | +0.035 |
iv | +0.034 |
od | +0.034 |
iolet | +0.032 |
bed | +0.031 |
Neg logit contributions | |
older | -0.045 |
heavier | -0.037 |
swe | -0.037 |
stays | -0.037 |
sle | -0.035 |
Token | € |
Token ID | 26391 |
Feature activation | -7.76e-02 |
Pos logit contributions | |
older | +0.013 |
stays | +0.010 |
heavier | +0.010 |
swe | +0.010 |
goes | +0.010 |
Neg logit contributions | |
anish | -0.011 |
iv | -0.010 |
od | -0.010 |
iolet | -0.010 |
Jack | -0.010 |
Token | ™ |
Token ID | 8151 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
anish | +0.133 |
iv | +0.128 |
od | +0.119 |
iolet | +0.117 |
Sus | +0.115 |
Neg logit contributions | |
older | -0.139 |
stays | -0.121 |
goes | -0.118 |
opens | -0.114 |
stable | -0.114 |
Token | s |
Token ID | 82 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
anish | +0.030 |
iv | +0.029 |
od | +0.029 |
iolet | +0.028 |
And | +0.028 |
Neg logit contributions | |
older | -0.021 |
heavier | -0.016 |
swe | -0.016 |
stays | -0.015 |
bigger | -0.015 |
Token | light |
Token ID | 1657 |
Feature activation | -5.59e-03 |
Pos logit contributions | |
anish | +0.024 |
iv | +0.023 |
od | +0.023 |
bed | +0.022 |
iolet | +0.022 |
Neg logit contributions | |
older | -0.028 |
heavier | -0.024 |
swe | -0.023 |
stays | -0.023 |
sle | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -8.62e-03 |
Pos logit contributions | |
iv | +0.007 |
od | +0.007 |
anish | +0.007 |
oon | +0.007 |
bed | +0.007 |
Neg logit contributions | |
older | -0.012 |
heavier | -0.010 |
stays | -0.010 |
sle | -0.010 |
climbs | -0.010 |
Token | She |
Token ID | 1375 |
Feature activation | -7.81e-03 |
Pos logit contributions | |
anish | +0.010 |
iv | +0.010 |
od | +0.009 |
Jane | +0.009 |
Jack | +0.009 |
Neg logit contributions | |
older | -0.020 |
heavier | -0.018 |
stays | -0.017 |
swe | -0.017 |
sle | -0.017 |
Token | says |
Token ID | 1139 |
Feature activation | -8.60e-04 |
Pos logit contributions | |
older | +0.010 |
heavier | +0.009 |
swe | +0.008 |
sle | +0.008 |
stays | +0.008 |
Neg logit contributions | |
anish | -0.012 |
iv | -0.012 |
od | -0.012 |
iolet | -0.012 |
And | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
anish | +0.002 |
iv | +0.001 |
od | +0.001 |
iolet | +0.001 |
And | +0.001 |
Neg logit contributions | |
older | -0.002 |
heavier | -0.001 |
swe | -0.001 |
stays | -0.001 |
bigger | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
older | +0.039 |
heavier | +0.033 |
swe | +0.033 |
stays | +0.033 |
sle | +0.031 |
Neg logit contributions | |
anish | -0.027 |
iv | -0.025 |
od | -0.025 |
iolet | -0.024 |
And | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
older | +0.056 |
swe | +0.047 |
heavier | +0.046 |
stays | +0.046 |
bigger | +0.045 |
Neg logit contributions | |
anish | -0.042 |
iv | -0.039 |
od | -0.039 |
iolet | -0.036 |
oon | -0.035 |
Token | " |
Token ID | 1 |
Feature activation | -2.15e-03 |
Pos logit contributions | |
older | +0.026 |
swe | +0.022 |
heavier | +0.021 |
stays | +0.021 |
bigger | +0.021 |
Neg logit contributions | |
anish | -0.021 |
od | -0.020 |
iv | -0.020 |
iolet | -0.018 |
oon | -0.018 |
Token | Mom |
Token ID | 29252 |
Feature activation | +4.64e-02 |
Pos logit contributions | |
od | +0.004 |
anish | +0.004 |
iv | +0.003 |
bed | +0.003 |
oon | +0.003 |
Neg logit contributions | |
older | -0.004 |
swe | -0.004 |
lower | -0.004 |
stays | -0.003 |
heavier | -0.003 |
Token | my |
Token ID | 1820 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
older | +0.087 |
swe | +0.079 |
heavier | +0.077 |
sle | +0.077 |
stable | +0.075 |
Neg logit contributions | |
anish | -0.073 |
iv | -0.069 |
od | -0.068 |
iolet | -0.066 |
room | -0.063 |
Token | , |
Token ID | 11 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
older | +0.088 |
sle | +0.078 |
swe | +0.077 |
heavier | +0.076 |
stable | +0.075 |
Neg logit contributions | |
anish | -0.066 |
iv | -0.064 |
od | -0.062 |
iolet | -0.059 |
Jack | -0.056 |
Token | Anna |
Token ID | 11735 |
Feature activation | +3.43e-02 |
Pos logit contributions | |
older | +0.024 |
heavier | +0.020 |
lower | +0.018 |
swe | +0.018 |
gs | +0.018 |
Neg logit contributions | |
anish | -0.033 |
iv | -0.030 |
od | -0.029 |
And | -0.029 |
Jack | -0.029 |
Token | is |
Token ID | 318 |
Feature activation | +7.30e-03 |
Pos logit contributions | |
older | +0.072 |
heavier | +0.063 |
swe | +0.061 |
bars | +0.060 |
sle | +0.060 |
Neg logit contributions | |
anish | -0.049 |
iv | -0.049 |
od | -0.046 |
iolet | -0.044 |
oon | -0.042 |
Token | my |
Token ID | 616 |
Feature activation | -2.03e-03 |
Pos logit contributions | |
older | +0.013 |
stays | +0.011 |
swe | +0.011 |
sle | +0.011 |
goes | +0.011 |
Neg logit contributions | |
od | -0.012 |
iv | -0.012 |
anish | -0.012 |
And | -0.011 |
Jack | -0.011 |
Token | picnic |
Token ID | 35715 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
older | +0.002 |
stays | +0.002 |
heavier | +0.002 |
gs | +0.002 |
sle | +0.002 |
Neg logit contributions | |
anish | -0.001 |
iv | -0.001 |
iolet | -0.001 |
od | -0.001 |
And | -0.001 |
Token | ? |
Token ID | 30 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
older | +0.019 |
heavier | +0.017 |
brakes | +0.017 |
sle | +0.017 |
stays | +0.017 |
Neg logit contributions | |
anish | -0.017 |
iv | -0.017 |
room | -0.016 |
oon | -0.015 |
iolet | -0.015 |
Token | Should |
Token ID | 10358 |
Feature activation | -5.17e-03 |
Pos logit contributions | |
older | +0.038 |
swe | +0.033 |
heavier | +0.033 |
stays | +0.033 |
cooler | +0.031 |
Neg logit contributions | |
anish | -0.031 |
iv | -0.030 |
od | -0.029 |
And | -0.029 |
oon | -0.028 |
Token | I |
Token ID | 314 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
anish | +0.008 |
iv | +0.007 |
od | +0.007 |
Jack | +0.007 |
oon | +0.007 |
Neg logit contributions | |
older | -0.010 |
sle | -0.009 |
swe | -0.009 |
heavier | -0.009 |
lower | -0.009 |
Token | bring |
Token ID | 2222 |
Feature activation | +4.13e-02 |
Pos logit contributions | |
iv | +0.022 |
anish | +0.021 |
od | +0.020 |
And | +0.020 |
oon | +0.020 |
Neg logit contributions | |
older | -0.021 |
heavier | -0.018 |
swe | -0.018 |
sle | -0.018 |
stays | -0.017 |
Token | an |
Token ID | 281 |
Feature activation | +4.92e-02 |
Pos logit contributions | |
older | +0.085 |
heavier | +0.073 |
stays | +0.072 |
sle | +0.072 |
stable | +0.072 |
Neg logit contributions | |
anish | -0.058 |
iv | -0.057 |
Jack | -0.055 |
od | -0.055 |
bed | -0.054 |
Token | umbrella |
Token ID | 25510 |
Feature activation | +3.27e-02 |
Pos logit contributions | |
older | +0.075 |
stays | +0.064 |
heavier | +0.064 |
swe | +0.062 |
sle | +0.059 |
Neg logit contributions | |
iv | -0.094 |
anish | -0.094 |
od | -0.091 |
And | -0.086 |
oon | -0.086 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
older | +0.056 |
stays | +0.049 |
swe | +0.048 |
heavier | +0.047 |
sle | +0.046 |
Neg logit contributions | |
anish | -0.058 |
iv | -0.056 |
od | -0.053 |
room | -0.053 |
oon | -0.053 |
Token |
|
Token ID | 198 |
Feature activation | +1.00e-02 |
Pos logit contributions | |
iv | +0.035 |
anish | +0.035 |
oon | +0.033 |
od | +0.033 |
iolet | +0.031 |
Neg logit contributions | |
older | -0.035 |
asks | -0.027 |
old | -0.026 |
younger | -0.026 |
swe | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | -4.70e-03 |
Pos logit contributions | |
older | +0.021 |
swe | +0.017 |
heavier | +0.017 |
bigger | +0.016 |
old | +0.016 |
Neg logit contributions | |
anish | -0.016 |
iv | -0.015 |
od | -0.014 |
oon | -0.014 |
iolet | -0.013 |
Token | Her |
Token ID | 9360 |
Feature activation | +1.09e-03 |
Pos logit contributions | |
anish | +0.005 |
iv | +0.005 |
od | +0.005 |
oon | +0.005 |
iolet | +0.004 |
Neg logit contributions | |
older | -0.011 |
swe | -0.010 |
lower | -0.009 |
old | -0.009 |
goes | -0.009 |
Token | tell |
Token ID | 1560 |
Feature activation | +3.32e-03 |
Pos logit contributions | |
anish | +0.017 |
iv | +0.017 |
od | +0.017 |
iolet | +0.016 |
And | +0.016 |
Neg logit contributions | |
older | -0.015 |
heavier | -0.012 |
swe | -0.012 |
stays | -0.012 |
bigger | -0.011 |
Token | me |
Token ID | 502 |
Feature activation | +2.47e-02 |
Pos logit contributions | |
older | +0.004 |
heavier | +0.003 |
gs | +0.003 |
stays | +0.003 |
swe | +0.003 |
Neg logit contributions | |
anish | -0.007 |
iv | -0.007 |
od | -0.007 |
Jack | -0.007 |
iolet | -0.007 |
Token | where |
Token ID | 810 |
Feature activation | +4.13e-02 |
Pos logit contributions | |
older | +0.046 |
sle | +0.042 |
gs | +0.041 |
stays | +0.041 |
heavier | +0.041 |
Neg logit contributions | |
anish | -0.037 |
iv | -0.036 |
od | -0.035 |
oon | -0.033 |
iolet | -0.033 |
Token | to |
Token ID | 284 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
older | +0.083 |
gs | +0.083 |
lower | +0.078 |
heavier | +0.078 |
tighter | +0.076 |
Neg logit contributions | |
anish | -0.051 |
Jack | -0.046 |
od | -0.045 |
iv | -0.044 |
endings | -0.042 |
Token | find |
Token ID | 1064 |
Feature activation | +3.54e-02 |
Pos logit contributions | |
older | +0.015 |
heavier | +0.013 |
swe | +0.012 |
stays | +0.012 |
gs | +0.012 |
Neg logit contributions | |
iv | -0.020 |
anish | -0.020 |
od | -0.020 |
And | -0.019 |
oon | -0.019 |
Token | some |
Token ID | 617 |
Feature activation | +4.77e-02 |
Pos logit contributions | |
older | +0.064 |
stays | +0.055 |
heavier | +0.055 |
swe | +0.053 |
gs | +0.052 |
Neg logit contributions | |
anish | -0.057 |
iv | -0.056 |
od | -0.055 |
Jack | -0.053 |
And | -0.052 |
Token | sweet |
Token ID | 6029 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
older | +0.078 |
gs | +0.075 |
stays | +0.074 |
bars | +0.072 |
climbs | +0.068 |
Neg logit contributions | |
iv | -0.078 |
anish | -0.075 |
And | -0.068 |
od | -0.066 |
oon | -0.065 |
Token | berries |
Token ID | 36322 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
older | +0.019 |
heavier | +0.016 |
stays | +0.015 |
gs | +0.015 |
lower | +0.015 |
Neg logit contributions | |
anish | -0.017 |
iv | -0.016 |
od | -0.016 |
bed | -0.014 |
iolet | -0.014 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.65e-02 |
Pos logit contributions | |
older | +0.063 |
sle | +0.054 |
swe | +0.054 |
stays | +0.053 |
brakes | +0.053 |
Neg logit contributions | |
iv | -0.056 |
anish | -0.053 |
oon | -0.051 |
od | -0.051 |
And | -0.047 |
Token | The |
Token ID | 383 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
anish | +0.034 |
iv | +0.034 |
od | +0.033 |
iolet | +0.032 |
oon | +0.032 |
Neg logit contributions | |
older | -0.025 |
old | -0.019 |
heavier | -0.019 |
goes | -0.019 |
swe | -0.019 |
Token | mole |
Token ID | 9411 |
Feature activation | +6.00e-03 |
Pos logit contributions | |
older | +0.040 |
heavier | +0.033 |
swe | +0.033 |
stays | +0.032 |
bigger | +0.032 |
Neg logit contributions | |
anish | -0.034 |
iv | -0.032 |
od | -0.032 |
iolet | -0.031 |
And | -0.030 |
Token | Mia |
Token ID | 32189 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
anish | +0.004 |
iv | +0.004 |
od | +0.004 |
Jack | +0.004 |
iolet | +0.004 |
Neg logit contributions | |
older | -0.006 |
heavier | -0.005 |
swe | -0.005 |
stays | -0.005 |
sle | -0.005 |
Token | says |
Token ID | 1139 |
Feature activation | +3.54e-03 |
Pos logit contributions | |
older | +0.023 |
heavier | +0.020 |
sle | +0.020 |
swe | +0.019 |
stable | +0.018 |
Neg logit contributions | |
iv | -0.045 |
od | -0.044 |
anish | -0.044 |
And | -0.042 |
iolet | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | +2.58e-02 |
Pos logit contributions | |
older | +0.007 |
stays | +0.006 |
swe | +0.006 |
heavier | +0.006 |
sle | +0.006 |
Neg logit contributions | |
anish | -0.005 |
iv | -0.005 |
od | -0.005 |
bed | -0.004 |
And | -0.004 |
Token | " |
Token ID | 366 |
Feature activation | -2.22e-03 |
Pos logit contributions | |
older | +0.044 |
heavier | +0.035 |
swe | +0.035 |
stays | +0.035 |
bigger | +0.034 |
Neg logit contributions | |
anish | -0.049 |
iv | -0.047 |
od | -0.047 |
iolet | -0.045 |
Jack | -0.044 |
Token | The |
Token ID | 464 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
iv | +0.003 |
anish | +0.003 |
od | +0.003 |
iolet | +0.003 |
wards | +0.003 |
Neg logit contributions | |
older | -0.004 |
stays | -0.004 |
heavier | -0.004 |
lower | -0.004 |
goes | -0.004 |
Token | boat |
Token ID | 8848 |
Feature activation | +4.65e-02 |
Pos logit contributions | |
older | +0.038 |
swe | +0.031 |
heavier | +0.031 |
stays | +0.030 |
younger | +0.029 |
Neg logit contributions | |
anish | -0.051 |
iv | -0.049 |
And | -0.048 |
od | -0.048 |
J | -0.047 |
Token | is |
Token ID | 318 |
Feature activation | -3.18e-03 |
Pos logit contributions | |
older | +0.080 |
swe | +0.068 |
sle | +0.066 |
younger | +0.066 |
reins | +0.064 |
Neg logit contributions | |
anish | -0.079 |
iv | -0.076 |
And | -0.075 |
oon | -0.070 |
od | -0.069 |
Token | bad |
Token ID | 2089 |
Feature activation | -2.81e-02 |
Pos logit contributions | |
knowing | +0.003 |
sharing | +0.003 |
And | +0.003 |
celebrating | +0.003 |
creating | +0.003 |
Neg logit contributions | |
bars | -0.007 |
sle | -0.006 |
older | -0.006 |
opens | -0.006 |
reins | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
anish | +0.044 |
iv | +0.043 |
od | +0.041 |
And | +0.039 |
iolet | +0.039 |
Neg logit contributions | |
older | -0.056 |
heavier | -0.048 |
swe | -0.047 |
stays | -0.046 |
sle | -0.046 |
Token | The |
Token ID | 383 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
older | +0.051 |
heavier | +0.041 |
stays | +0.041 |
swe | +0.040 |
goes | +0.039 |
Neg logit contributions | |
anish | -0.047 |
od | -0.046 |
iv | -0.046 |
iolet | -0.044 |
Jack | -0.043 |
Token | man |
Token ID | 582 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
older | +0.035 |
swe | +0.030 |
heavier | +0.029 |
stays | +0.029 |
opens | +0.028 |
Neg logit contributions | |
anish | -0.038 |
iv | -0.036 |
And | -0.035 |
od | -0.034 |
J | -0.033 |
Token | ' |
Token ID | 6 |
Feature activation | -1.58e-03 |
Pos logit contributions | |
older | +0.013 |
swe | +0.011 |
stays | +0.011 |
heavier | +0.011 |
sle | +0.011 |
Neg logit contributions | |
anish | -0.009 |
iv | -0.008 |
od | -0.008 |
oon | -0.007 |
Jack | -0.007 |
Token | mean |
Token ID | 1612 |
Feature activation | +6.69e-03 |
Pos logit contributions | |
anish | +0.003 |
iv | +0.003 |
od | +0.003 |
Jack | +0.003 |
And | +0.002 |
Neg logit contributions | |
older | -0.003 |
swe | -0.002 |
stays | -0.002 |
sle | -0.002 |
heavier | -0.002 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
older | +0.014 |
heavier | +0.012 |
stays | +0.012 |
swe | +0.012 |
sle | +0.012 |
Neg logit contributions | |
anish | -0.011 |
iv | -0.010 |
od | -0.009 |
Jack | -0.009 |
iolet | -0.009 |
Token | asked |
Token ID | 1965 |
Feature activation | +6.43e-03 |
Pos logit contributions | |
anish | +0.021 |
iv | +0.020 |
od | +0.020 |
iolet | +0.019 |
And | +0.019 |
Neg logit contributions | |
older | -0.027 |
heavier | -0.022 |
swe | -0.022 |
stays | -0.022 |
bigger | -0.021 |
Token | Mar |
Token ID | 1526 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
older | +0.013 |
heavier | +0.011 |
swe | +0.011 |
stays | +0.010 |
bigger | +0.010 |
Neg logit contributions | |
anish | -0.010 |
iv | -0.010 |
od | -0.010 |
iolet | -0.009 |
Jack | -0.009 |
Token | iah |
Token ID | 9520 |
Feature activation | +6.66e-02 |
Pos logit contributions | |
older | +0.063 |
swe | +0.058 |
heavier | +0.058 |
bigger | +0.056 |
cooler | +0.055 |
Neg logit contributions | |
iv | -0.055 |
od | -0.053 |
Sus | -0.053 |
anish | -0.051 |
bed | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | -7.38e-03 |
Pos logit contributions | |
older | +0.129 |
heavier | +0.115 |
swe | +0.115 |
sle | +0.115 |
stays | +0.114 |
Neg logit contributions | |
od | -0.102 |
anish | -0.098 |
iv | -0.095 |
bed | -0.089 |
iolet | -0.089 |
Token |
|
Token ID | 198 |
Feature activation | +5.78e-03 |
Pos logit contributions | |
anish | +0.010 |
iv | +0.010 |
od | +0.009 |
And | +0.009 |
iolet | +0.009 |
Neg logit contributions | |
older | -0.016 |
heavier | -0.014 |
swe | -0.014 |
stays | -0.014 |
bigger | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
older | +0.013 |
heavier | +0.011 |
swe | +0.011 |
stays | +0.011 |
bigger | +0.011 |
Neg logit contributions | |
anish | -0.008 |
iv | -0.007 |
od | -0.007 |
iolet | -0.007 |
And | -0.007 |
Token | " |
Token ID | 1 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
anish | +0.017 |
iv | +0.016 |
od | +0.016 |
iolet | +0.015 |
And | +0.015 |
Neg logit contributions | |
older | -0.023 |
heavier | -0.019 |
swe | -0.019 |
stays | -0.019 |
bigger | -0.018 |
Token | It |
Token ID | 1026 |
Feature activation | +1.93e-02 |
Pos logit contributions | |
anish | +0.022 |
od | +0.020 |
iv | +0.020 |
iolet | +0.019 |
bed | +0.019 |
Neg logit contributions | |
older | -0.037 |
heavier | -0.032 |
swe | -0.032 |
stays | -0.031 |
lower | -0.031 |
Token | your |
Token ID | 534 |
Feature activation | +4.80e-03 |
Pos logit contributions | |
anish | +0.011 |
iv | +0.011 |
od | +0.011 |
And | +0.010 |
Jack | +0.010 |
Neg logit contributions | |
older | -0.013 |
heavier | -0.011 |
swe | -0.011 |
stays | -0.011 |
gs | -0.010 |
Token | teacher |
Token ID | 4701 |
Feature activation | -9.70e-03 |
Pos logit contributions | |
older | +0.009 |
heavier | +0.008 |
swe | +0.007 |
stays | +0.007 |
lower | +0.007 |
Neg logit contributions | |
iv | -0.008 |
anish | -0.008 |
od | -0.008 |
iolet | -0.008 |
And | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
iv | +0.014 |
od | +0.014 |
anish | +0.014 |
iolet | +0.013 |
oon | +0.013 |
Neg logit contributions | |
older | -0.020 |
heavier | -0.017 |
swe | -0.017 |
lower | -0.016 |
bigger | -0.016 |
Token | I |
Token ID | 314 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
older | +0.040 |
heavier | +0.035 |
swe | +0.034 |
stays | +0.033 |
bigger | +0.033 |
Neg logit contributions | |
anish | -0.031 |
iv | -0.030 |
od | -0.029 |
And | -0.028 |
iolet | -0.028 |
Token | 'm |
Token ID | 1101 |
Feature activation | +4.82e-03 |
Pos logit contributions | |
anish | +0.029 |
iv | +0.028 |
od | +0.027 |
And | +0.026 |
iolet | +0.026 |
Neg logit contributions | |
older | -0.041 |
heavier | -0.035 |
swe | -0.035 |
stays | -0.034 |
bigger | -0.033 |
Token | so |
Token ID | 523 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
older | +0.009 |
stays | +0.008 |
swe | +0.008 |
goes | +0.008 |
heavier | +0.008 |
Neg logit contributions | |
od | -0.008 |
iv | -0.008 |
Sus | -0.007 |
Jane | -0.007 |
Jack | -0.007 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
older | +0.059 |
stays | +0.055 |
swe | +0.055 |
reins | +0.054 |
brakes | +0.052 |
Neg logit contributions | |
od | -0.057 |
Sus | -0.055 |
iv | -0.054 |
Jane | -0.052 |
Jack | -0.052 |
Token | to |
Token ID | 284 |
Feature activation | +8.92e-03 |
Pos logit contributions | |
anish | +0.029 |
iv | +0.028 |
od | +0.028 |
iolet | +0.027 |
And | +0.027 |
Neg logit contributions | |
older | -0.024 |
heavier | -0.019 |
swe | -0.019 |
stays | -0.019 |
bigger | -0.018 |
Token | see |
Token ID | 766 |
Feature activation | +5.76e-03 |
Pos logit contributions | |
older | +0.015 |
heavier | +0.012 |
swe | +0.012 |
stays | +0.012 |
bigger | +0.011 |
Neg logit contributions | |
anish | -0.017 |
iv | -0.017 |
od | -0.016 |
iolet | -0.016 |
And | -0.016 |
Token | you |
Token ID | 345 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
older | +0.011 |
heavier | +0.010 |
stays | +0.010 |
gs | +0.009 |
lower | +0.009 |
Neg logit contributions | |
iv | -0.009 |
od | -0.009 |
anish | -0.009 |
Jack | -0.008 |
Sus | -0.008 |
Token | all |
Token ID | 477 |
Feature activation | +2.83e-02 |
Pos logit contributions | |
od | +0.026 |
anish | +0.026 |
iv | +0.026 |
oon | +0.024 |
And | +0.024 |
Neg logit contributions | |
older | -0.033 |
heavier | -0.028 |
stays | -0.028 |
swe | -0.028 |
sle | -0.027 |
Token | to |
Token ID | 284 |
Feature activation | +1.47e-03 |
Pos logit contributions | |
older | +0.001 |
heavier | +0.001 |
swe | +0.001 |
stays | +0.001 |
asks | +0.001 |
Neg logit contributions | |
anish | -0.001 |
od | -0.001 |
iv | -0.001 |
And | -0.001 |
Jack | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
older | +0.002 |
swe | +0.002 |
heavier | +0.002 |
stays | +0.002 |
bigger | +0.002 |
Neg logit contributions | |
anish | -0.003 |
iv | -0.003 |
od | -0.003 |
Jack | -0.003 |
iolet | -0.003 |
Token | living |
Token ID | 2877 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
older | +0.019 |
stays | +0.018 |
gs | +0.017 |
heavier | +0.017 |
swe | +0.017 |
Neg logit contributions | |
iv | -0.018 |
anish | -0.018 |
And | -0.017 |
iolet | -0.016 |
Jack | -0.016 |
Token | room |
Token ID | 2119 |
Feature activation | +5.26e-03 |
Pos logit contributions | |
anish | +0.038 |
iv | +0.037 |
od | +0.036 |
iolet | +0.035 |
Jack | +0.035 |
Neg logit contributions | |
older | -0.036 |
heavier | -0.030 |
swe | -0.029 |
stays | -0.029 |
bigger | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
older | +0.009 |
heavier | +0.007 |
swe | +0.007 |
stays | +0.007 |
bigger | +0.007 |
Neg logit contributions | |
anish | -0.010 |
iv | -0.009 |
od | -0.009 |
iolet | -0.009 |
And | -0.009 |
Token | placed |
Token ID | 4624 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
older | +0.022 |
swe | +0.018 |
heavier | +0.018 |
stays | +0.018 |
sle | +0.018 |
Neg logit contributions | |
anish | -0.020 |
iv | -0.020 |
Jack | -0.019 |
od | -0.019 |
bed | -0.019 |
Token | it |
Token ID | 340 |
Feature activation | +2.83e-03 |
Pos logit contributions | |
older | +0.023 |
stays | +0.020 |
swe | +0.019 |
younger | +0.019 |
stable | +0.018 |
Neg logit contributions | |
anish | -0.013 |
Jack | -0.013 |
bed | -0.013 |
iv | -0.012 |
od | -0.012 |
Token | on |
Token ID | 319 |
Feature activation | -5.90e-04 |
Pos logit contributions | |
older | +0.006 |
asks | +0.005 |
sle | +0.005 |
stays | +0.005 |
younger | +0.005 |
Neg logit contributions | |
bed | -0.004 |
anish | -0.004 |
od | -0.003 |
iv | -0.003 |
oon | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | +2.27e-03 |
Pos logit contributions | |
anish | +0.001 |
iv | +0.001 |
od | +0.001 |
Jack | +0.001 |
iolet | +0.001 |
Neg logit contributions | |
older | -0.001 |
stays | -0.001 |
swe | -0.001 |
heavier | -0.001 |
goes | -0.001 |
Token | scale |
Token ID | 5046 |
Feature activation | +4.53e-03 |
Pos logit contributions | |
older | +0.004 |
stays | +0.004 |
heavier | +0.004 |
swe | +0.004 |
goes | +0.004 |
Neg logit contributions | |
anish | -0.003 |
iv | -0.003 |
od | -0.003 |
And | -0.003 |
oon | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
older | +0.009 |
stays | +0.008 |
swe | +0.008 |
heavier | +0.008 |
goes | +0.008 |
Neg logit contributions | |
iv | -0.006 |
anish | -0.006 |
od | -0.006 |
bed | -0.006 |
oon | -0.006 |
Token | practice |
Token ID | 3357 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
anish | +0.026 |
iv | +0.025 |
od | +0.024 |
And | +0.023 |
iolet | +0.023 |
Neg logit contributions | |
older | -0.034 |
heavier | -0.029 |
swe | -0.029 |
stays | -0.028 |
sle | -0.028 |
Token | her |
Token ID | 607 |
Feature activation | +4.38e-03 |
Pos logit contributions | |
anish | +0.025 |
iv | +0.024 |
od | +0.023 |
iolet | +0.022 |
Jack | +0.022 |
Neg logit contributions | |
older | -0.026 |
swe | -0.022 |
heavier | -0.022 |
stays | -0.021 |
bigger | -0.021 |
Token | dance |
Token ID | 9280 |
Feature activation | -3.24e-03 |
Pos logit contributions | |
older | +0.008 |
heavier | +0.007 |
swe | +0.006 |
sle | +0.006 |
stays | +0.006 |
Neg logit contributions | |
anish | -0.008 |
iv | -0.007 |
od | -0.007 |
iolet | -0.007 |
And | -0.007 |
Token | moves |
Token ID | 6100 |
Feature activation | +3.97e-03 |
Pos logit contributions | |
anish | +0.006 |
iv | +0.005 |
od | +0.005 |
iolet | +0.005 |
And | +0.005 |
Neg logit contributions | |
older | -0.006 |
heavier | -0.005 |
swe | -0.005 |
stays | -0.005 |
bigger | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +2.04e-03 |
Pos logit contributions | |
older | +0.008 |
heavier | +0.006 |
stays | +0.006 |
swe | +0.006 |
sle | +0.006 |
Neg logit contributions | |
anish | -0.006 |
iv | -0.006 |
od | -0.006 |
And | -0.006 |
iolet | -0.006 |
Token | listen |
Token ID | 6004 |
Feature activation | +4.30e-03 |
Pos logit contributions | |
older | +0.004 |
heavier | +0.003 |
swe | +0.003 |
stays | +0.003 |
sle | +0.003 |
Neg logit contributions | |
anish | -0.003 |
iv | -0.003 |
od | -0.003 |
And | -0.003 |
iolet | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | +2.02e-03 |
Pos logit contributions | |
older | +0.007 |
sle | +0.006 |
heavier | +0.006 |
stays | +0.006 |
cooler | +0.006 |
Neg logit contributions | |
od | -0.008 |
anish | -0.008 |
iv | -0.008 |
And | -0.007 |
Jack | -0.007 |
Token | her |
Token ID | 607 |
Feature activation | -6.25e-04 |
Pos logit contributions | |
older | +0.004 |
heavier | +0.003 |
swe | +0.003 |
stays | +0.003 |
cooler | +0.003 |
Neg logit contributions | |
anish | -0.003 |
od | -0.003 |
iv | -0.003 |
iolet | -0.003 |
And | -0.003 |
Token | favourite |
Token ID | 12507 |
Feature activation | -3.11e-03 |
Pos logit contributions | |
anish | +0.001 |
iv | +0.001 |
od | +0.001 |
iolet | +0.001 |
And | +0.001 |
Neg logit contributions | |
older | -0.001 |
heavier | -0.001 |
stays | -0.001 |
sle | -0.001 |
swe | -0.001 |
Token | songs |
Token ID | 7259 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
anish | +0.004 |
iv | +0.004 |
od | +0.004 |
iolet | +0.004 |
Jack | +0.004 |
Neg logit contributions | |
older | -0.007 |
swe | -0.006 |
heavier | -0.006 |
stays | -0.006 |
bigger | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -4.17e-03 |
Pos logit contributions | |
anish | +0.034 |
iv | +0.032 |
od | +0.032 |
Jack | +0.031 |
iolet | +0.030 |
Neg logit contributions | |
older | -0.033 |
heavier | -0.027 |
swe | -0.027 |
stays | -0.027 |
bigger | -0.026 |
Token | for |
Token ID | 329 |
Feature activation | -2.92e-03 |
Pos logit contributions | |
older | +0.040 |
heavier | +0.032 |
stays | +0.031 |
swe | +0.031 |
sle | +0.029 |
Neg logit contributions | |
anish | -0.045 |
iv | -0.043 |
od | -0.041 |
And | -0.041 |
bed | -0.040 |
Token | help |
Token ID | 1037 |
Feature activation | +7.58e-03 |
Pos logit contributions | |
anish | +0.005 |
iv | +0.005 |
od | +0.005 |
iolet | +0.005 |
Jack | +0.005 |
Neg logit contributions | |
older | -0.005 |
heavier | -0.004 |
stays | -0.004 |
swe | -0.004 |
goes | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
older | +0.014 |
heavier | +0.012 |
swe | +0.012 |
stays | +0.012 |
bigger | +0.011 |
Neg logit contributions | |
anish | -0.013 |
iv | -0.012 |
od | -0.012 |
iolet | -0.012 |
And | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
older | +0.035 |
heavier | +0.028 |
old | +0.027 |
swe | +0.027 |
younger | +0.027 |
Neg logit contributions | |
anish | -0.032 |
iv | -0.031 |
od | -0.031 |
oon | -0.030 |
iolet | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | +2.55e-03 |
Pos logit contributions | |
older | +0.040 |
old | +0.032 |
bigger | +0.032 |
heavier | +0.032 |
swe | +0.031 |
Neg logit contributions | |
anish | -0.037 |
iv | -0.035 |
od | -0.034 |
iolet | -0.033 |
oon | -0.032 |
Token | Their |
Token ID | 14574 |
Feature activation | +7.21e-03 |
Pos logit contributions | |
older | +0.005 |
heavier | +0.005 |
bigger | +0.004 |
swe | +0.004 |
stays | +0.004 |
Neg logit contributions | |
anish | -0.004 |
od | -0.004 |
iv | -0.004 |
iolet | -0.003 |
oon | -0.003 |
Token | mom |
Token ID | 1995 |
Feature activation | +2.75e-03 |
Pos logit contributions | |
older | +0.011 |
heavier | +0.009 |
swe | +0.009 |
stays | +0.009 |
lower | +0.008 |
Neg logit contributions | |
anish | -0.014 |
iv | -0.014 |
od | -0.014 |
And | -0.013 |
iolet | -0.013 |
Token | heard |
Token ID | 2982 |
Feature activation | +7.28e-03 |
Pos logit contributions | |
older | +0.006 |
heavier | +0.005 |
swe | +0.005 |
stays | +0.005 |
sle | +0.005 |
Neg logit contributions | |
anish | -0.004 |
iv | -0.004 |
Jack | -0.004 |
od | -0.004 |
J | -0.004 |
Token | them |
Token ID | 606 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
older | +0.011 |
swe | +0.009 |
heavier | +0.009 |
stays | +0.008 |
bigger | +0.008 |
Neg logit contributions | |
anish | -0.015 |
iv | -0.015 |
od | -0.014 |
iolet | -0.014 |
And | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | +4.20e-02 |
Pos logit contributions | |
older | +0.021 |
heavier | +0.016 |
swe | +0.016 |
stays | +0.016 |
goes | +0.016 |
Neg logit contributions | |
anish | -0.025 |
iv | -0.023 |
od | -0.023 |
iolet | -0.023 |
Jack | -0.023 |
Token | ran |
Token ID | 4966 |
Feature activation | +3.73e-02 |
Pos logit contributions | |
older | +0.078 |
heavier | +0.067 |
swe | +0.066 |
sle | +0.065 |
stable | +0.064 |
Neg logit contributions | |
J | -0.065 |
od | -0.065 |
anish | -0.064 |
Jack | -0.064 |
Jane | -0.064 |
Token | sad |
Token ID | 6507 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
older | +0.056 |
sle | +0.056 |
swe | +0.053 |
heavier | +0.052 |
reins | +0.051 |
Neg logit contributions | |
bed | -0.028 |
iv | -0.028 |
anish | -0.027 |
Jack | -0.025 |
od | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
anish | +0.038 |
od | +0.036 |
iv | +0.035 |
Jack | +0.035 |
iolet | +0.035 |
Neg logit contributions | |
older | -0.032 |
swe | -0.025 |
heavier | -0.025 |
stays | -0.025 |
old | -0.024 |
Token | so |
Token ID | 523 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
anish | +0.033 |
iv | +0.031 |
od | +0.031 |
iolet | +0.029 |
And | +0.029 |
Neg logit contributions | |
older | -0.040 |
heavier | -0.034 |
swe | -0.034 |
stays | -0.033 |
bigger | -0.032 |
Token | she |
Token ID | 673 |
Feature activation | +2.27e-02 |
Pos logit contributions | |
older | +0.029 |
heavier | +0.025 |
swe | +0.024 |
stays | +0.024 |
sle | +0.023 |
Neg logit contributions | |
anish | -0.021 |
iv | -0.020 |
od | -0.020 |
Jack | -0.019 |
And | -0.018 |
Token | shut |
Token ID | 4423 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
older | +0.045 |
heavier | +0.040 |
sle | +0.039 |
swe | +0.037 |
bigger | +0.036 |
Neg logit contributions | |
iv | -0.034 |
Jack | -0.032 |
bed | -0.031 |
anish | -0.031 |
All | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
older | +0.066 |
heavier | +0.057 |
stays | +0.056 |
swe | +0.055 |
sle | +0.055 |
Neg logit contributions | |
iv | -0.051 |
anish | -0.049 |
od | -0.048 |
bed | -0.048 |
Sus | -0.047 |
Token | door |
Token ID | 3420 |
Feature activation | +3.13e-02 |
Pos logit contributions | |
older | +0.037 |
heavier | +0.035 |
swe | +0.034 |
stays | +0.032 |
sle | +0.032 |
Neg logit contributions | |
anish | -0.058 |
iv | -0.056 |
J | -0.054 |
And | -0.052 |
od | -0.052 |
Token | to |
Token ID | 284 |
Feature activation | -1.69e-03 |
Pos logit contributions | |
older | +0.062 |
stays | +0.055 |
sle | +0.054 |
swe | +0.053 |
heavier | +0.052 |
Neg logit contributions | |
anish | -0.044 |
iv | -0.043 |
od | -0.042 |
And | -0.041 |
Sus | -0.041 |
Token | her |
Token ID | 607 |
Feature activation | -7.34e-03 |
Pos logit contributions | |
anish | +0.003 |
iv | +0.003 |
od | +0.003 |
And | +0.003 |
iolet | +0.003 |
Neg logit contributions | |
older | -0.003 |
heavier | -0.002 |
swe | -0.002 |
stays | -0.002 |
sle | -0.002 |
Token | bedroom |
Token ID | 14043 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
iv | +0.013 |
anish | +0.013 |
od | +0.013 |
iolet | +0.013 |
And | +0.013 |
Neg logit contributions | |
older | -0.012 |
heavier | -0.010 |
stays | -0.010 |
swe | -0.010 |
gs | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
older | +0.052 |
heavier | +0.044 |
swe | +0.044 |
stays | +0.044 |
sle | +0.043 |
Neg logit contributions | |
anish | -0.042 |
iv | -0.041 |
od | -0.039 |
iolet | -0.037 |
And | -0.037 |
Token | their |
Token ID | 511 |
Feature activation | +8.36e-03 |
Pos logit contributions | |
older | +0.043 |
sle | +0.039 |
heavier | +0.038 |
stays | +0.038 |
swe | +0.036 |
Neg logit contributions | |
anish | -0.036 |
iv | -0.036 |
od | -0.035 |
J | -0.035 |
Jack | -0.034 |
Token | folders |
Token ID | 24512 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
older | +0.016 |
heavier | +0.015 |
sle | +0.014 |
stays | +0.013 |
lower | +0.013 |
Neg logit contributions | |
anish | -0.013 |
od | -0.012 |
iv | -0.012 |
iolet | -0.011 |
J | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | -3.97e-03 |
Pos logit contributions | |
older | +0.028 |
sle | +0.025 |
heavier | +0.024 |
swe | +0.024 |
stays | +0.024 |
Neg logit contributions | |
anish | -0.026 |
iv | -0.025 |
od | -0.024 |
oon | -0.023 |
iolet | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
anish | +0.007 |
iv | +0.007 |
od | +0.007 |
iolet | +0.006 |
Jack | +0.006 |
Neg logit contributions | |
older | -0.007 |
heavier | -0.006 |
stays | -0.006 |
swe | -0.006 |
sle | -0.006 |
Token | front |
Token ID | 2166 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
older | +0.021 |
heavier | +0.018 |
swe | +0.018 |
stays | +0.018 |
sle | +0.017 |
Neg logit contributions | |
anish | -0.017 |
iv | -0.016 |
od | -0.016 |
iolet | -0.015 |
And | -0.015 |
Token | of |
Token ID | 286 |
Feature activation | -8.53e-03 |
Pos logit contributions | |
older | +0.030 |
heavier | +0.027 |
swe | +0.026 |
stays | +0.026 |
sle | +0.025 |
Neg logit contributions | |
anish | -0.025 |
iv | -0.023 |
od | -0.023 |
iolet | -0.022 |
And | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | +9.22e-03 |
Pos logit contributions | |
anish | +0.015 |
iv | +0.015 |
od | +0.015 |
J | +0.014 |
Jack | +0.014 |
Neg logit contributions | |
older | -0.015 |
stays | -0.013 |
heavier | -0.013 |
swe | -0.012 |
sle | -0.012 |
Token | class |
Token ID | 1398 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
older | +0.018 |
heavier | +0.016 |
stays | +0.016 |
swe | +0.016 |
sle | +0.016 |
Neg logit contributions | |
anish | -0.014 |
iv | -0.013 |
od | -0.013 |
iolet | -0.012 |
J | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +2.59e-03 |
Pos logit contributions | |
anish | +0.029 |
od | +0.029 |
iv | +0.029 |
iolet | +0.027 |
J | +0.026 |
Neg logit contributions | |
older | -0.040 |
heavier | -0.035 |
swe | -0.035 |
sle | -0.035 |
stays | -0.035 |
Token | She |
Token ID | 1375 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
older | +0.005 |
heavier | +0.005 |
sle | +0.005 |
stays | +0.005 |
swe | +0.005 |
Neg logit contributions | |
anish | -0.003 |
iv | -0.003 |
Jack | -0.003 |
od | -0.003 |
iolet | -0.003 |
Token | wants |
Token ID | 3382 |
Feature activation | +8.50e-03 |
Pos logit contributions | |
heavier | +0.034 |
older | +0.033 |
sle | +0.033 |
reins | +0.030 |
forward | +0.030 |
Neg logit contributions | |
iv | -0.050 |
od | -0.047 |
anish | -0.045 |
iolet | -0.044 |
Jack | -0.044 |
Token | , |
Token ID | 11 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
older | +0.005 |
heavier | +0.004 |
swe | +0.004 |
stays | +0.004 |
bigger | +0.004 |
Neg logit contributions | |
anish | -0.008 |
iv | -0.008 |
od | -0.007 |
And | -0.007 |
iolet | -0.007 |
Token | there |
Token ID | 612 |
Feature activation | +2.50e-03 |
Pos logit contributions | |
anish | +0.031 |
iv | +0.031 |
od | +0.030 |
iolet | +0.029 |
And | +0.028 |
Neg logit contributions | |
older | -0.020 |
heavier | -0.015 |
bigger | -0.015 |
swe | -0.015 |
stays | -0.014 |
Token | was |
Token ID | 373 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
older | +0.004 |
goes | +0.003 |
stays | +0.003 |
heavier | +0.003 |
old | +0.003 |
Neg logit contributions | |
anish | -0.005 |
iv | -0.005 |
od | -0.005 |
iolet | -0.004 |
And | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | +2.09e-03 |
Pos logit contributions | |
anish | +0.059 |
iv | +0.055 |
od | +0.053 |
oon | +0.052 |
iolet | +0.052 |
Neg logit contributions | |
older | -0.024 |
old | -0.018 |
bigger | -0.018 |
cooler | -0.017 |
heavy | -0.017 |
Token | little |
Token ID | 1310 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
older | +0.003 |
stays | +0.003 |
heavier | +0.003 |
swe | +0.002 |
goes | +0.002 |
Neg logit contributions | |
anish | -0.004 |
iv | -0.004 |
od | -0.004 |
And | -0.004 |
Jack | -0.004 |
Token | girl |
Token ID | 2576 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
anish | +0.025 |
iv | +0.025 |
od | +0.024 |
Jack | +0.024 |
iolet | +0.023 |
Neg logit contributions | |
older | -0.023 |
stays | -0.020 |
heavier | -0.019 |
swe | -0.019 |
opens | -0.018 |
Token | named |
Token ID | 3706 |
Feature activation | +5.38e-03 |
Pos logit contributions | |
anish | +0.051 |
iolet | +0.046 |
od | +0.045 |
iv | +0.044 |
Jack | +0.043 |
Neg logit contributions | |
older | -0.053 |
old | -0.045 |
goes | -0.044 |
swe | -0.044 |
stable | -0.043 |
Token | Lily |
Token ID | 20037 |
Feature activation | +4.73e-03 |
Pos logit contributions | |
older | +0.009 |
swe | +0.007 |
bigger | +0.007 |
old | +0.007 |
heavier | +0.007 |
Neg logit contributions | |
anish | -0.011 |
iv | -0.010 |
od | -0.010 |
And | -0.010 |
iolet | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
older | +0.005 |
goes | +0.004 |
old | +0.004 |
down | +0.004 |
stable | +0.004 |
Neg logit contributions | |
anish | -0.012 |
iolet | -0.011 |
wards | -0.011 |
iv | -0.011 |
ocolate | -0.011 |
Token | One |
Token ID | 1881 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
anish | +0.039 |
iv | +0.039 |
od | +0.039 |
oon | +0.036 |
And | +0.036 |
Neg logit contributions | |
older | -0.027 |
swe | -0.022 |
old | -0.020 |
goes | -0.020 |
bigger | -0.019 |
Token | day |
Token ID | 1110 |
Feature activation | -2.55e-02 |
Pos logit contributions | |
older | +0.049 |
old | +0.040 |
goes | +0.037 |
bigger | +0.037 |
swe | +0.037 |
Neg logit contributions | |
anish | -0.064 |
iv | -0.062 |
od | -0.060 |
iolet | -0.059 |
bed | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
od | +0.005 |
iv | +0.005 |
anish | +0.005 |
And | +0.005 |
Jack | +0.005 |
Neg logit contributions | |
older | -0.007 |
heavier | -0.006 |
swe | -0.006 |
sle | -0.006 |
stays | -0.006 |
Token | Then |
Token ID | 3244 |
Feature activation | -8.85e-03 |
Pos logit contributions | |
anish | +0.022 |
iv | +0.021 |
od | +0.021 |
iolet | +0.020 |
And | +0.020 |
Neg logit contributions | |
older | -0.026 |
heavier | -0.021 |
swe | -0.021 |
stays | -0.021 |
bigger | -0.020 |
Token | a |
Token ID | 257 |
Feature activation | -4.98e-03 |
Pos logit contributions | |
anish | +0.015 |
iv | +0.015 |
od | +0.015 |
iolet | +0.014 |
And | +0.014 |
Neg logit contributions | |
older | -0.016 |
heavier | -0.013 |
swe | -0.013 |
stays | -0.013 |
sle | -0.013 |
Token | big |
Token ID | 1263 |
Feature activation | -9.69e-03 |
Pos logit contributions | |
anish | +0.010 |
And | +0.009 |
iv | +0.009 |
Jack | +0.009 |
od | +0.009 |
Neg logit contributions | |
older | -0.007 |
stays | -0.007 |
swe | -0.006 |
heavier | -0.006 |
goes | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -6.65e-03 |
Pos logit contributions | |
anish | +0.016 |
Jack | +0.015 |
od | +0.015 |
iv | +0.015 |
oon | +0.015 |
Neg logit contributions | |
older | -0.016 |
stays | -0.015 |
swe | -0.014 |
bars | -0.013 |
heavier | -0.013 |
Token | grey |
Token ID | 13791 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
Jack | +0.013 |
J | +0.013 |
od | +0.013 |
Sus | +0.013 |
And | +0.013 |
Neg logit contributions | |
older | -0.009 |
stays | -0.008 |
opens | -0.007 |
swe | -0.007 |
gs | -0.007 |
Token | cloud |
Token ID | 6279 |
Feature activation | +3.86e-04 |
Pos logit contributions | |
anish | +0.033 |
iv | +0.032 |
od | +0.032 |
iolet | +0.030 |
Jack | +0.030 |
Neg logit contributions | |
older | -0.033 |
heavier | -0.027 |
swe | -0.027 |
stays | -0.026 |
bigger | -0.025 |
Token | appeared |
Token ID | 4120 |
Feature activation | -8.43e-03 |
Pos logit contributions | |
older | +0.001 |
heavier | +0.001 |
swe | +0.001 |
stays | +0.001 |
bigger | +0.001 |
Neg logit contributions | |
anish | -0.001 |
iv | -0.001 |
od | -0.001 |
Jack | -0.001 |
iolet | -0.001 |
Token | on |
Token ID | 319 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
anish | +0.015 |
iv | +0.014 |
od | +0.014 |
And | +0.013 |
iolet | +0.013 |
Neg logit contributions | |
older | -0.015 |
heavier | -0.013 |
swe | -0.013 |
stays | -0.013 |
sle | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
anish | +0.042 |
od | +0.039 |
iv | +0.038 |
Jack | +0.038 |
iolet | +0.038 |
Neg logit contributions | |
older | -0.026 |
swe | -0.020 |
bigger | -0.020 |
heavier | -0.020 |
old | -0.019 |
Token | horizon |
Token ID | 17810 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
iv | +0.027 |
od | +0.026 |
anish | +0.026 |
And | +0.025 |
Jack | +0.024 |
Neg logit contributions | |
older | -0.039 |
stays | -0.034 |
swe | -0.034 |
heavier | -0.033 |
asks | -0.032 |
Token | in |
Token ID | 287 |
Feature activation | +8.35e-03 |
Pos logit contributions | |
bed | +0.047 |
od | +0.045 |
anish | +0.044 |
And | +0.042 |
iv | +0.042 |
Neg logit contributions | |
older | -0.058 |
stays | -0.049 |
swe | -0.049 |
heavier | -0.048 |
sle | -0.048 |
Token | the |
Token ID | 262 |
Feature activation | -5.17e-03 |
Pos logit contributions | |
older | +0.014 |
stays | +0.012 |
heavier | +0.012 |
swe | +0.012 |
gs | +0.011 |
Neg logit contributions | |
anish | -0.015 |
iv | -0.014 |
Jack | -0.014 |
J | -0.013 |
iolet | -0.013 |
Token | eyes |
Token ID | 2951 |
Feature activation | -1.69e-03 |
Pos logit contributions | |
anish | +0.009 |
Jack | +0.009 |
And | +0.009 |
All | +0.008 |
J | +0.008 |
Neg logit contributions | |
older | -0.009 |
heavier | -0.008 |
swe | -0.007 |
gs | -0.007 |
stays | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
od | +0.002 |
anish | +0.002 |
iv | +0.002 |
And | +0.002 |
bed | +0.002 |
Neg logit contributions | |
older | -0.003 |
stays | -0.003 |
swe | -0.003 |
sle | -0.003 |
asks | -0.003 |
Token | said |
Token ID | 531 |
Feature activation | +6.38e-03 |
Pos logit contributions | |
older | +0.030 |
swe | +0.025 |
heavier | +0.025 |
stays | +0.024 |
sle | +0.024 |
Neg logit contributions | |
anish | -0.027 |
od | -0.026 |
bed | -0.026 |
iv | -0.026 |
Jack | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | +2.55e-03 |
Pos logit contributions | |
older | +0.012 |
stays | +0.010 |
sle | +0.010 |
heavier | +0.010 |
swe | +0.010 |
Neg logit contributions | |
anish | -0.011 |
od | -0.011 |
iv | -0.010 |
And | -0.010 |
iolet | -0.010 |
Token | � |
Token ID | 6184 |
Feature activation | +1.45e-02 |
Pos logit contributions | |
older | +0.005 |
stays | +0.004 |
swe | +0.004 |
heavier | +0.004 |
goes | +0.004 |
Neg logit contributions | |
anish | -0.004 |
od | -0.004 |
iv | -0.004 |
And | -0.004 |
Jack | -0.004 |
Token | � |
Token ID | 95 |
Feature activation | +2.28e-03 |
Pos logit contributions | |
older | +0.028 |
heavier | +0.023 |
stays | +0.023 |
swe | +0.023 |
lower | +0.022 |
Neg logit contributions | |
anish | -0.024 |
od | -0.022 |
iv | -0.022 |
iolet | -0.021 |
Jack | -0.021 |
Token | € |
Token ID | 26391 |
Feature activation | -4.73e-02 |
Pos logit contributions | |
older | +0.004 |
stays | +0.004 |
heavier | +0.004 |
swe | +0.004 |
goes | +0.004 |
Neg logit contributions | |
anish | -0.004 |
od | -0.003 |
iv | -0.003 |
iolet | -0.003 |
Sus | -0.003 |
Token | � |
Token ID | 129 |
Feature activation | +2.19e-02 |
Pos logit contributions | |
anish | +0.080 |
iv | +0.077 |
od | +0.075 |
iolet | +0.072 |
And | +0.071 |
Neg logit contributions | |
older | -0.089 |
heavier | -0.074 |
swe | -0.073 |
stays | -0.073 |
bigger | -0.070 |
Token | � |
Token ID | 241 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
older | +0.045 |
swe | +0.039 |
heavier | +0.038 |
sle | +0.037 |
stays | +0.037 |
Neg logit contributions | |
iv | -0.032 |
od | -0.032 |
anish | -0.031 |
And | -0.031 |
oon | -0.029 |
Token | ending |
Token ID | 7464 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
anish | +0.045 |
iv | +0.040 |
od | +0.039 |
Jack | +0.036 |
oon | +0.036 |
Neg logit contributions | |
older | -0.108 |
heavier | -0.097 |
bars | -0.093 |
sle | -0.093 |
stays | -0.093 |
Token | was |
Token ID | 373 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
anish | +0.016 |
iv | +0.014 |
od | +0.014 |
Jack | +0.014 |
oon | +0.013 |
Neg logit contributions | |
older | -0.030 |
sle | -0.027 |
swe | -0.026 |
reins | -0.026 |
stable | -0.026 |
Token | finally |
Token ID | 3443 |
Feature activation | +1.88e-03 |
Pos logit contributions | |
anish | +0.020 |
Jack | +0.020 |
od | +0.020 |
wards | +0.020 |
And | +0.019 |
Neg logit contributions | |
older | -0.036 |
sle | -0.034 |
opens | -0.032 |
stays | -0.032 |
reins | -0.032 |
Token | there |
Token ID | 612 |
Feature activation | -9.99e-03 |
Pos logit contributions | |
older | +0.004 |
sle | +0.003 |
stays | +0.003 |
swe | +0.003 |
gs | +0.003 |
Neg logit contributions | |
anish | -0.003 |
od | -0.003 |
iv | -0.003 |
Jack | -0.003 |
And | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
anish | +0.015 |
iv | +0.014 |
od | +0.014 |
And | +0.014 |
Jack | +0.014 |
Neg logit contributions | |
older | -0.020 |
swe | -0.017 |
heavier | -0.017 |
stays | -0.017 |
sle | -0.017 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
anish | +0.026 |
iv | +0.025 |
od | +0.024 |
oon | +0.024 |
bed | +0.023 |
Neg logit contributions | |
older | -0.024 |
old | -0.018 |
younger | -0.018 |
heavier | -0.018 |
cooler | -0.018 |
Token | Sam |
Token ID | 16305 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
anish | +0.023 |
od | +0.022 |
iv | +0.022 |
And | +0.020 |
Sus | +0.020 |
Neg logit contributions | |
older | -0.021 |
goes | -0.017 |
cooler | -0.017 |
bigger | -0.017 |
heavier | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
older | +0.022 |
heavier | +0.020 |
swe | +0.019 |
sle | +0.019 |
bigger | +0.018 |
Neg logit contributions | |
anish | -0.019 |
iv | -0.019 |
Jack | -0.019 |
od | -0.019 |
oon | -0.018 |
Token | three |
Token ID | 1115 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
anish | +0.034 |
iv | +0.034 |
od | +0.034 |
bed | +0.033 |
Jack | +0.033 |
Neg logit contributions | |
older | -0.032 |
swe | -0.029 |
stays | -0.029 |
heavier | -0.027 |
sle | -0.027 |
Token | years |
Token ID | 812 |
Feature activation | +5.50e-02 |
Pos logit contributions | |
iv | +0.018 |
anish | +0.017 |
od | +0.017 |
And | +0.016 |
iolet | +0.015 |
Neg logit contributions | |
older | -0.039 |
heavier | -0.034 |
stays | -0.034 |
swe | -0.034 |
lower | -0.033 |
Token | old |
Token ID | 1468 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
stays | +0.053 |
lower | +0.043 |
swe | +0.042 |
heavier | +0.039 |
sits | +0.038 |
Neg logit contributions | |
od | -0.125 |
Sus | -0.125 |
Jane | -0.120 |
Jack | -0.120 |
iv | -0.120 |
Token | smile |
Token ID | 8212 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
anish | +0.046 |
iv | +0.045 |
od | +0.044 |
And | +0.043 |
iolet | +0.042 |
Neg logit contributions | |
older | -0.050 |
swe | -0.040 |
heavier | -0.040 |
stays | -0.040 |
younger | -0.038 |
Token | even |
Token ID | 772 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
od | +0.044 |
anish | +0.043 |
iv | +0.041 |
And | +0.040 |
iolet | +0.040 |
Neg logit contributions | |
older | -0.053 |
heavier | -0.045 |
stays | -0.044 |
swe | -0.043 |
sle | -0.043 |
Token | wider |
Token ID | 10595 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
sle | +0.046 |
older | +0.045 |
reins | +0.042 |
gs | +0.040 |
brakes | +0.039 |
Neg logit contributions | |
And | -0.041 |
od | -0.038 |
anish | -0.038 |
J | -0.038 |
Jane | -0.037 |
Token | when |
Token ID | 618 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
anish | +0.023 |
iv | +0.023 |
od | +0.023 |
And | +0.022 |
iolet | +0.021 |
Neg logit contributions | |
older | -0.036 |
heavier | -0.031 |
stays | -0.030 |
sle | -0.030 |
swe | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
older | +0.025 |
stays | +0.022 |
heavier | +0.022 |
swe | +0.022 |
sle | +0.022 |
Neg logit contributions | |
od | -0.015 |
anish | -0.015 |
iv | -0.014 |
Jack | -0.014 |
wards | -0.013 |
Token | lonely |
Token ID | 21757 |
Feature activation | -4.51e-02 |
Pos logit contributions | |
older | +0.034 |
stays | +0.031 |
gs | +0.030 |
heavier | +0.029 |
ches | +0.029 |
Neg logit contributions | |
anish | -0.035 |
iv | -0.035 |
Sus | -0.033 |
Jack | -0.033 |
And | -0.033 |
Token | lip |
Token ID | 10645 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
anish | +0.077 |
iv | +0.074 |
od | +0.074 |
J | +0.070 |
Jack | +0.070 |
Neg logit contributions | |
older | -0.076 |
heavier | -0.065 |
stays | -0.064 |
lower | -0.063 |
swe | -0.062 |
Token | was |
Token ID | 373 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
od | +0.044 |
anish | +0.044 |
iv | +0.043 |
bed | +0.040 |
iolet | +0.040 |
Neg logit contributions | |
older | -0.053 |
swe | -0.044 |
heavier | -0.044 |
sle | -0.044 |
stays | -0.043 |
Token | around |
Token ID | 1088 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
od | +0.026 |
iv | +0.025 |
Jane | +0.025 |
anish | +0.025 |
bed | +0.025 |
Neg logit contributions | |
older | -0.030 |
stays | -0.028 |
gs | -0.028 |
sle | -0.027 |
opens | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
anish | +0.046 |
od | +0.044 |
iv | +0.044 |
iolet | +0.041 |
And | +0.041 |
Neg logit contributions | |
older | -0.060 |
stays | -0.051 |
heavier | -0.051 |
swe | -0.050 |
sle | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | +3.49e-03 |
Pos logit contributions | |
anish | +0.026 |
iv | +0.025 |
od | +0.024 |
iolet | +0.023 |
oon | +0.022 |
Neg logit contributions | |
older | -0.030 |
swe | -0.025 |
heavier | -0.024 |
stays | -0.024 |
goes | -0.024 |
Token | missed |
Token ID | 6825 |
Feature activation | -3.64e-02 |
Pos logit contributions | |
anish | +0.019 |
iv | +0.018 |
od | +0.018 |
And | +0.017 |
iolet | +0.017 |
Neg logit contributions | |
older | -0.027 |
heavier | -0.023 |
swe | -0.022 |
sle | -0.022 |
stays | -0.022 |
Token | out |
Token ID | 503 |
Feature activation | -8.21e-03 |
Pos logit contributions | |
anish | +0.068 |
iv | +0.065 |
od | +0.064 |
iolet | +0.062 |
And | +0.061 |
Neg logit contributions | |
older | -0.063 |
heavier | -0.051 |
swe | -0.051 |
stays | -0.050 |
bigger | -0.048 |
Token | on |
Token ID | 319 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
od | +0.011 |
anish | +0.011 |
iv | +0.011 |
And | +0.011 |
J | +0.011 |
Neg logit contributions | |
older | -0.016 |
stays | -0.014 |
swe | -0.014 |
sle | -0.014 |
opens | -0.014 |
Token | all |
Token ID | 477 |
Feature activation | -2.64e-03 |
Pos logit contributions | |
anish | +0.032 |
iv | +0.031 |
od | +0.031 |
Jack | +0.029 |
iolet | +0.029 |
Neg logit contributions | |
older | -0.027 |
stays | -0.023 |
heavier | -0.023 |
swe | -0.022 |
sle | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | -4.07e-03 |
Pos logit contributions | |
anish | +0.005 |
od | +0.005 |
iv | +0.005 |
Jane | +0.005 |
Jack | +0.005 |
Neg logit contributions | |
older | -0.004 |
stays | -0.004 |
heavier | -0.004 |
swe | -0.004 |
sle | -0.003 |
Token | fun |
Token ID | 1257 |
Feature activation | -4.71e-02 |
Pos logit contributions | |
anish | +0.007 |
iv | +0.007 |
od | +0.007 |
Jack | +0.007 |
And | +0.007 |
Neg logit contributions | |
older | -0.007 |
heavier | -0.006 |
swe | -0.006 |
stays | -0.006 |
sle | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
anish | +0.083 |
iv | +0.080 |
od | +0.078 |
iolet | +0.075 |
And | +0.075 |
Neg logit contributions | |
older | -0.086 |
heavier | -0.071 |
swe | -0.071 |
stays | -0.070 |
bigger | -0.067 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
anish | +0.026 |
iv | +0.026 |
iolet | +0.024 |
oon | +0.024 |
od | +0.024 |
Neg logit contributions | |
older | -0.020 |
old | -0.017 |
goes | -0.016 |
swe | -0.016 |
bigger | -0.016 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
anish | +0.026 |
od | +0.025 |
iv | +0.024 |
itive | +0.023 |
edly | +0.023 |
Neg logit contributions | |
older | -0.021 |
old | -0.017 |
goes | -0.017 |
bigger | -0.017 |
cooler | -0.016 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
older | +0.022 |
heavier | +0.018 |
swe | +0.018 |
stays | +0.018 |
sle | +0.017 |
Neg logit contributions | |
anish | -0.032 |
od | -0.031 |
iv | -0.031 |
Jack | -0.030 |
iolet | -0.030 |
Token | a |
Token ID | 257 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
anish | +0.041 |
iv | +0.040 |
od | +0.039 |
iolet | +0.037 |
oon | +0.036 |
Neg logit contributions | |
older | -0.016 |
bigger | -0.013 |
heavy | -0.013 |
old | -0.012 |
cooler | -0.012 |
Token | frame |
Token ID | 5739 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
older | +0.005 |
stays | +0.004 |
heavier | +0.004 |
swe | +0.004 |
opens | +0.004 |
Neg logit contributions | |
anish | -0.006 |
And | -0.006 |
od | -0.005 |
iv | -0.005 |
Jack | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
anish | +0.015 |
iv | +0.015 |
od | +0.015 |
oon | +0.014 |
iolet | +0.014 |
Neg logit contributions | |
older | -0.021 |
stays | -0.018 |
heavier | -0.018 |
swe | -0.018 |
sle | -0.017 |
Token | They |
Token ID | 1119 |
Feature activation | +3.36e-03 |
Pos logit contributions | |
older | +0.022 |
swe | +0.018 |
bigger | +0.018 |
cooler | +0.018 |
heavier | +0.018 |
Neg logit contributions | |
anish | -0.022 |
iv | -0.022 |
od | -0.022 |
oon | -0.020 |
iolet | -0.020 |
Token | love |
Token ID | 1842 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
older | +0.006 |
heavier | +0.005 |
swe | +0.005 |
stays | +0.005 |
sle | +0.005 |
Neg logit contributions | |
iv | -0.006 |
anish | -0.006 |
od | -0.005 |
And | -0.005 |
iolet | -0.005 |
Token | their |
Token ID | 511 |
Feature activation | -6.63e-03 |
Pos logit contributions | |
anish | +0.023 |
od | +0.022 |
iv | +0.022 |
Jack | +0.021 |
iolet | +0.021 |
Neg logit contributions | |
older | -0.023 |
heavier | -0.019 |
stays | -0.019 |
swe | -0.019 |
sle | -0.018 |
Token | children |
Token ID | 1751 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
anish | +0.013 |
iv | +0.013 |
od | +0.012 |
Jack | +0.012 |
iolet | +0.012 |
Neg logit contributions | |
older | -0.011 |
swe | -0.009 |
heavier | -0.009 |
stays | -0.009 |
old | -0.008 |
Token | very |
Token ID | 845 |
Feature activation | +9.42e-03 |
Pos logit contributions | |
iv | +0.064 |
anish | +0.063 |
od | +0.063 |
And | +0.059 |
iolet | +0.059 |
Neg logit contributions | |
older | -0.085 |
heavier | -0.072 |
stays | -0.071 |
swe | -0.071 |
bigger | -0.067 |
Token | much |
Token ID | 881 |
Feature activation | +1.81e-04 |
Pos logit contributions | |
older | +0.016 |
heavier | +0.013 |
swe | +0.013 |
stays | +0.013 |
bigger | +0.013 |
Neg logit contributions | |
anish | -0.018 |
iv | -0.017 |
od | -0.017 |
iolet | -0.016 |
Jack | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +7.15e-03 |
Pos logit contributions | |
older | +0.000 |
heavier | +0.000 |
stays | +0.000 |
swe | +0.000 |
sle | +0.000 |
Neg logit contributions | |
anish | -0.000 |
iv | -0.000 |
od | -0.000 |
And | -0.000 |
iolet | -0.000 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.97e-03 |
Pos logit contributions | |
older | +0.013 |
swe | +0.010 |
old | +0.010 |
bigger | +0.010 |
goes | +0.010 |
Neg logit contributions | |
anish | -0.013 |
od | -0.013 |
iv | -0.013 |
oon | -0.012 |
bed | -0.012 |
Token | Tom |
Token ID | 13787 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
anish | +0.018 |
iv | +0.017 |
od | +0.017 |
And | +0.017 |
iolet | +0.017 |
Neg logit contributions | |
older | -0.014 |
heavier | -0.011 |
swe | -0.011 |
stays | -0.011 |
bigger | -0.010 |
Token | if |
Token ID | 611 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
older | +0.074 |
sle | +0.065 |
stays | +0.062 |
heavier | +0.062 |
swe | +0.061 |
Neg logit contributions | |
anish | -0.055 |
iv | -0.055 |
od | -0.053 |
Jack | -0.049 |
J | -0.049 |
Token | they |
Token ID | 484 |
Feature activation | -3.42e-03 |
Pos logit contributions | |
older | +0.029 |
sle | +0.026 |
lower | +0.026 |
heavier | +0.026 |
swe | +0.025 |
Neg logit contributions | |
anish | -0.018 |
iv | -0.017 |
od | -0.017 |
Jack | -0.017 |
J | -0.015 |
Token | have |
Token ID | 423 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
anish | +0.005 |
iv | +0.005 |
od | +0.005 |
Jack | +0.004 |
iolet | +0.004 |
Neg logit contributions | |
older | -0.007 |
swe | -0.006 |
heavier | -0.006 |
sle | -0.006 |
stable | -0.006 |
Token | less |
Token ID | 1342 |
Feature activation | -2.01e-03 |
Pos logit contributions | |
anish | +0.022 |
od | +0.021 |
iv | +0.021 |
Jack | +0.020 |
iolet | +0.020 |
Neg logit contributions | |
older | -0.025 |
sle | -0.022 |
heavier | -0.021 |
gs | -0.021 |
stable | -0.020 |
Token | than |
Token ID | 621 |
Feature activation | -2.96e-03 |
Pos logit contributions | |
anish | +0.003 |
iv | +0.003 |
od | +0.003 |
iolet | +0.003 |
bed | +0.003 |
Neg logit contributions | |
older | -0.004 |
heavier | -0.003 |
sle | -0.003 |
lower | -0.003 |
stays | -0.003 |
Token | you |
Token ID | 345 |
Feature activation | -4.07e-02 |
Pos logit contributions | |
anish | +0.005 |
od | +0.005 |
iv | +0.005 |
Jack | +0.004 |
bed | +0.004 |
Neg logit contributions | |
older | -0.005 |
stays | -0.005 |
sle | -0.005 |
swe | -0.004 |
heavier | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
anish | +0.062 |
iv | +0.060 |
od | +0.058 |
iolet | +0.054 |
Jack | +0.054 |
Neg logit contributions | |
older | -0.082 |
heavier | -0.072 |
swe | -0.071 |
sle | -0.071 |
stays | -0.068 |
Token | How |
Token ID | 1374 |
Feature activation | +3.71e-03 |
Pos logit contributions | |
older | +0.036 |
swe | +0.029 |
heavier | +0.029 |
stays | +0.029 |
goes | +0.028 |
Neg logit contributions | |
anish | -0.029 |
od | -0.028 |
iv | -0.028 |
iolet | -0.026 |
Jack | -0.026 |
Token | would |
Token ID | 561 |
Feature activation | +4.42e-03 |
Pos logit contributions | |
older | +0.006 |
heavier | +0.005 |
swe | +0.005 |
stays | +0.005 |
bigger | +0.005 |
Neg logit contributions | |
anish | -0.007 |
iv | -0.007 |
od | -0.006 |
iolet | -0.006 |
oon | -0.006 |
Token | you |
Token ID | 345 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
older | +0.008 |
heavier | +0.006 |
swe | +0.006 |
stays | +0.006 |
bigger | +0.006 |
Neg logit contributions | |
anish | -0.008 |
iv | -0.008 |
od | -0.008 |
iolet | -0.008 |
And | -0.008 |
Token | feel |
Token ID | 1254 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
anish | +0.045 |
iv | +0.043 |
od | +0.043 |
iolet | +0.041 |
And | +0.041 |
Neg logit contributions | |
older | -0.047 |
heavier | -0.039 |
swe | -0.039 |
stays | -0.038 |
sle | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | -9.70e-04 |
Pos logit contributions | |
older | +0.023 |
swe | +0.021 |
stays | +0.020 |
sle | +0.020 |
stable | +0.019 |
Neg logit contributions | |
anish | -0.016 |
iv | -0.016 |
Jack | -0.015 |
room | -0.015 |
od | -0.015 |
Token | He |
Token ID | 679 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
anish | +0.002 |
iv | +0.002 |
od | +0.002 |
iolet | +0.002 |
oon | +0.002 |
Neg logit contributions | |
older | -0.002 |
goes | -0.001 |
swe | -0.001 |
heavier | -0.001 |
stays | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
anish | +0.022 |
iv | +0.022 |
od | +0.021 |
And | +0.021 |
Jack | +0.021 |
Neg logit contributions | |
older | -0.029 |
heavier | -0.025 |
swe | -0.024 |
stays | -0.024 |
sle | -0.024 |
Token | a |
Token ID | 257 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
anish | +0.046 |
iv | +0.045 |
od | +0.045 |
And | +0.044 |
Sus | +0.042 |
Neg logit contributions | |
older | -0.054 |
swe | -0.046 |
sle | -0.046 |
stays | -0.046 |
heavier | -0.044 |
Token | very |
Token ID | 845 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
anish | +0.022 |
iv | +0.021 |
od | +0.021 |
iolet | +0.020 |
Jack | +0.020 |
Neg logit contributions | |
older | -0.015 |
heavier | -0.012 |
swe | -0.012 |
stays | -0.011 |
bigger | -0.011 |
Token | happy |
Token ID | 3772 |
Feature activation | -4.72e-02 |
Pos logit contributions | |
sle | +0.036 |
bars | +0.035 |
reins | +0.034 |
older | +0.034 |
stays | +0.033 |
Neg logit contributions | |
And | -0.029 |
J | -0.027 |
Jane | -0.026 |
Jack | -0.026 |
All | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
anish | +0.109 |
iv | +0.106 |
od | +0.106 |
oon | +0.102 |
iolet | +0.102 |
Neg logit contributions | |
older | -0.065 |
old | -0.052 |
cooler | -0.048 |
stable | -0.047 |
swe | -0.047 |
Token | proud |
Token ID | 6613 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
older | +0.039 |
sle | +0.033 |
swe | +0.033 |
stays | +0.032 |
heavier | +0.032 |
Neg logit contributions | |
od | -0.030 |
Jack | -0.030 |
iv | -0.030 |
anish | -0.030 |
And | -0.030 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
anish | +0.054 |
iv | +0.052 |
od | +0.051 |
iolet | +0.049 |
And | +0.049 |
Neg logit contributions | |
older | -0.050 |
heavier | -0.040 |
swe | -0.040 |
stays | -0.039 |
bigger | -0.038 |
Token | when |
Token ID | 618 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
anish | +0.032 |
iv | +0.031 |
od | +0.030 |
iolet | +0.029 |
And | +0.029 |
Neg logit contributions | |
older | -0.034 |
heavier | -0.028 |
swe | -0.028 |
stays | -0.028 |
bigger | -0.027 |
Token | he |
Token ID | 339 |
Feature activation | +6.35e-03 |
Pos logit contributions | |
older | +0.027 |
stays | +0.023 |
swe | +0.023 |
heavier | +0.022 |
sle | +0.022 |
Neg logit contributions | |
anish | -0.026 |
od | -0.026 |
Jack | -0.025 |
iv | -0.025 |
Sus | -0.024 |