Token | each |
Token ID | 1123 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
Sorry | +0.014 |
bye | +0.013 |
good | +0.012 |
day | +0.011 |
ho | +0.011 |
Neg logit contributions | |
Sad | -0.022 |
ice | -0.021 |
uster | -0.021 |
hole | -0.021 |
ist | -0.020 |
Token | other |
Token ID | 584 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
uster | +0.045 |
Sad | +0.044 |
hole | +0.044 |
ice | +0.043 |
ist | +0.042 |
Neg logit contributions | |
Sorry | -0.010 |
good | -0.010 |
bye | -0.010 |
they | -0.009 |
day | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +8.71e-03 |
Pos logit contributions | |
Sorry | +0.036 |
bye | +0.031 |
ho | +0.028 |
ge | +0.027 |
bye | +0.027 |
Neg logit contributions | |
hole | -0.033 |
ice | -0.031 |
ist | -0.030 |
Sad | -0.030 |
uster | -0.030 |
Token | Max |
Token ID | 5436 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
bye | +0.012 |
Sorry | +0.012 |
ho | +0.011 |
good | +0.011 |
day | +0.011 |
Neg logit contributions | |
Sad | -0.019 |
uster | -0.018 |
hole | -0.018 |
ice | -0.017 |
aron | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
Sorry | +0.090 |
bye | +0.089 |
day | +0.080 |
ho | +0.079 |
good | +0.072 |
Neg logit contributions | |
ice | -0.059 |
hole | -0.056 |
pled | -0.055 |
ist | -0.053 |
aron | -0.052 |
Token | Luna |
Token ID | 23694 |
Feature activation | +7.12e-02 |
Pos logit contributions | |
bye | +0.030 |
Sorry | +0.029 |
ho | +0.025 |
marriage | +0.025 |
good | +0.025 |
Neg logit contributions | |
ice | -0.025 |
father | -0.024 |
uster | -0.023 |
Sad | -0.023 |
run | -0.022 |
Token | hug |
Token ID | 16225 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
Sorry | +0.121 |
bye | +0.112 |
ho | +0.102 |
ww | +0.100 |
thing | +0.089 |
Neg logit contributions | |
hole | -0.111 |
ist | -0.108 |
pping | -0.106 |
ches | -0.104 |
ice | -0.103 |
Token | each |
Token ID | 1123 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
Sorry | +0.017 |
bye | +0.016 |
day | +0.013 |
good | +0.013 |
Prize | +0.013 |
Neg logit contributions | |
Sad | -0.023 |
ice | -0.022 |
uster | -0.022 |
ist | -0.021 |
aron | -0.021 |
Token | other |
Token ID | 584 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
uster | +0.037 |
Sad | +0.037 |
hole | +0.037 |
ice | +0.036 |
father | +0.035 |
Neg logit contributions | |
Sorry | -0.010 |
bye | -0.009 |
good | -0.009 |
they | -0.008 |
day | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +9.88e-03 |
Pos logit contributions | |
Sorry | +0.047 |
bye | +0.039 |
ho | +0.037 |
ge | +0.036 |
Prize | +0.035 |
Neg logit contributions | |
hole | -0.038 |
ice | -0.035 |
ist | -0.034 |
pled | -0.033 |
aron | -0.033 |
Token | They |
Token ID | 1119 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
bye | +0.014 |
Sorry | +0.013 |
ho | +0.012 |
day | +0.012 |
good | +0.011 |
Neg logit contributions | |
uster | -0.022 |
Sad | -0.021 |
hole | -0.020 |
ice | -0.020 |
aron | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | +1.62e-03 |
Pos logit contributions | |
uster | +0.001 |
hole | +0.001 |
Sad | +0.001 |
ice | +0.001 |
ist | +0.001 |
Neg logit contributions | |
Sorry | -0.000 |
bye | -0.000 |
good | -0.000 |
day | -0.000 |
they | -0.000 |
Token | please |
Token ID | 3387 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
Sorry | +0.002 |
bye | +0.002 |
good | +0.002 |
day | +0.002 |
ho | +0.002 |
Neg logit contributions | |
uster | -0.003 |
Sad | -0.003 |
hole | -0.003 |
ice | -0.003 |
pled | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
Sorry | +0.023 |
bye | +0.020 |
ho | +0.019 |
ge | +0.019 |
sunset | +0.019 |
Neg logit contributions | |
hole | -0.023 |
ist | -0.022 |
Sad | -0.021 |
ice | -0.020 |
aron | -0.020 |
Token | Max |
Token ID | 5436 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
good | +0.040 |
day | +0.038 |
bye | +0.037 |
Sorry | +0.037 |
well | +0.035 |
Neg logit contributions | |
uster | -0.045 |
Sad | -0.044 |
hole | -0.044 |
James | -0.041 |
innie | -0.040 |
Token | and |
Token ID | 290 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
Sorry | +0.059 |
bye | +0.053 |
ho | +0.048 |
sunset | +0.047 |
bye | +0.047 |
Neg logit contributions | |
hole | -0.066 |
ice | -0.064 |
uster | -0.062 |
ist | -0.062 |
pled | -0.062 |
Token | Luna |
Token ID | 23694 |
Feature activation | +6.74e-02 |
Pos logit contributions | |
Sorry | +0.022 |
bye | +0.022 |
ho | +0.021 |
good | +0.019 |
marriage | +0.018 |
Neg logit contributions | |
ice | -0.021 |
Sad | -0.021 |
uster | -0.020 |
father | -0.020 |
hole | -0.020 |
Token | can |
Token ID | 460 |
Feature activation | -2.36e-03 |
Pos logit contributions | |
Sorry | +0.115 |
bye | +0.104 |
ho | +0.101 |
bye | +0.088 |
sunset | +0.085 |
Neg logit contributions | |
hole | -0.119 |
ist | -0.112 |
ice | -0.110 |
uster | -0.107 |
pled | -0.105 |
Token | be |
Token ID | 307 |
Feature activation | +4.05e-03 |
Pos logit contributions | |
Sad | +0.005 |
hole | +0.005 |
uster | +0.005 |
ist | +0.005 |
ice | +0.005 |
Neg logit contributions | |
Sorry | -0.003 |
bye | -0.002 |
good | -0.002 |
day | -0.002 |
bye | -0.002 |
Token | friends |
Token ID | 2460 |
Feature activation | -6.92e-03 |
Pos logit contributions | |
Sorry | +0.006 |
bye | +0.005 |
ho | +0.005 |
day | +0.004 |
bye | +0.004 |
Neg logit contributions | |
Sad | -0.009 |
eping | -0.008 |
uster | -0.008 |
ice | -0.008 |
ady | -0.008 |
Token | too |
Token ID | 1165 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
hole | +0.014 |
Sad | +0.014 |
ice | +0.013 |
ist | +0.013 |
uster | +0.013 |
Neg logit contributions | |
Sorry | -0.011 |
bye | -0.010 |
good | -0.009 |
bye | -0.009 |
ho | -0.009 |
Token | ," |
Token ID | 553 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
Sorry | +0.028 |
bye | +0.026 |
ho | +0.024 |
bye | +0.024 |
good | +0.022 |
Neg logit contributions | |
hole | -0.028 |
Sad | -0.028 |
ice | -0.027 |
uster | -0.026 |
jo | -0.025 |
Token | so |
Token ID | 523 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
Sorry | +0.021 |
bye | +0.021 |
ho | +0.020 |
they | +0.018 |
bye | +0.018 |
Neg logit contributions | |
Sad | -0.035 |
hole | -0.034 |
father | -0.033 |
uster | -0.033 |
pled | -0.032 |
Token | kind |
Token ID | 1611 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
bye | +0.028 |
Sorry | +0.027 |
they | +0.027 |
ho | +0.027 |
day | +0.025 |
Neg logit contributions | |
Sad | -0.042 |
uster | -0.040 |
James | -0.038 |
father | -0.038 |
innie | -0.038 |
Token | !" |
Token ID | 2474 |
Feature activation | +5.47e-03 |
Pos logit contributions | |
Sorry | +0.020 |
bye | +0.017 |
good | +0.017 |
they | +0.017 |
ho | +0.016 |
Neg logit contributions | |
uster | -0.023 |
hole | -0.023 |
Sad | -0.023 |
ice | -0.022 |
pled | -0.022 |
Token | Spot |
Token ID | 15899 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
Sorry | +0.007 |
bye | +0.007 |
good | +0.007 |
ho | +0.007 |
day | +0.006 |
Neg logit contributions | |
Sad | -0.011 |
uster | -0.011 |
ice | -0.011 |
hole | -0.011 |
innie | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
Sorry | +0.079 |
bye | +0.066 |
day | +0.065 |
good | +0.064 |
they | +0.063 |
Neg logit contributions | |
ice | -0.069 |
Sad | -0.069 |
aron | -0.067 |
ist | -0.067 |
ilda | -0.065 |
Token | Blue |
Token ID | 4518 |
Feature activation | +6.49e-02 |
Pos logit contributions | |
Sorry | +0.023 |
bye | +0.023 |
ho | +0.022 |
ashes | +0.022 |
birthday | +0.021 |
Neg logit contributions | |
Sad | -0.023 |
father | -0.022 |
James | -0.021 |
Steven | -0.021 |
innie | -0.020 |
Token | went |
Token ID | 1816 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
Sorry | +0.103 |
bye | +0.079 |
birthday | +0.077 |
marriage | +0.077 |
good | +0.076 |
Neg logit contributions | |
ice | -0.124 |
ist | -0.123 |
Sad | -0.121 |
hole | -0.119 |
father | -0.118 |
Token | to |
Token ID | 284 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
Sorry | +0.029 |
bye | +0.026 |
bye | +0.022 |
good | +0.022 |
ho | +0.022 |
Neg logit contributions | |
uster | -0.045 |
hole | -0.045 |
Sad | -0.044 |
ilda | -0.042 |
ist | -0.041 |
Token | the |
Token ID | 262 |
Feature activation | -8.99e-03 |
Pos logit contributions | |
Sorry | +0.024 |
bye | +0.023 |
good | +0.020 |
they | +0.019 |
day | +0.019 |
Neg logit contributions | |
Sad | -0.061 |
uster | -0.060 |
hole | -0.060 |
father | -0.058 |
ice | -0.057 |
Token | small |
Token ID | 1402 |
Feature activation | +5.58e-03 |
Pos logit contributions | |
Sad | +0.013 |
hole | +0.013 |
x | +0.012 |
father | +0.012 |
uster | +0.011 |
Neg logit contributions | |
bye | -0.018 |
Sorry | -0.018 |
they | -0.017 |
ho | -0.017 |
bye | -0.016 |
Token | jail |
Token ID | 7356 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
Sorry | +0.010 |
bye | +0.009 |
they | +0.009 |
good | +0.009 |
bye | +0.008 |
Neg logit contributions | |
hole | -0.010 |
uster | -0.010 |
Sad | -0.010 |
ice | -0.009 |
father | -0.009 |
Token | They |
Token ID | 1119 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
bye | +0.020 |
ho | +0.019 |
Sorry | +0.019 |
day | +0.018 |
good | +0.018 |
Neg logit contributions | |
uster | -0.031 |
Sad | -0.030 |
hole | -0.029 |
ice | -0.028 |
aron | -0.028 |
Token | qu |
Token ID | 627 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
bye | +0.041 |
Sorry | +0.041 |
day | +0.034 |
bye | +0.032 |
good | +0.031 |
Neg logit contributions | |
ist | -0.046 |
ice | -0.046 |
pled | -0.045 |
hole | -0.044 |
father | -0.043 |
Token | ack |
Token ID | 441 |
Feature activation | +3.79e-02 |
Pos logit contributions | |
Sorry | +0.061 |
bye | +0.059 |
good | +0.058 |
they | +0.057 |
day | +0.056 |
Neg logit contributions | |
uster | -0.048 |
Sad | -0.048 |
hole | -0.047 |
ice | -0.045 |
father | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
Sorry | +0.096 |
bye | +0.086 |
ho | +0.077 |
day | +0.076 |
Prize | +0.076 |
Neg logit contributions | |
ice | -0.047 |
pled | -0.044 |
ist | -0.043 |
aron | -0.040 |
eping | -0.040 |
Token | They |
Token ID | 1119 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
bye | +0.028 |
ho | +0.028 |
Sorry | +0.027 |
day | +0.025 |
good | +0.025 |
Neg logit contributions | |
Sad | -0.039 |
uster | -0.037 |
James | -0.034 |
hole | -0.034 |
aron | -0.034 |
Token | swim |
Token ID | 9422 |
Feature activation | +6.48e-02 |
Pos logit contributions | |
Sorry | +0.043 |
bye | +0.043 |
day | +0.036 |
bye | +0.035 |
ho | +0.033 |
Neg logit contributions | |
ist | -0.047 |
ice | -0.045 |
pled | -0.045 |
hole | -0.045 |
Sad | -0.043 |
Token | away |
Token ID | 1497 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
Sorry | +0.152 |
bye | +0.124 |
Mall | +0.121 |
Hipp | +0.121 |
marriage | +0.121 |
Neg logit contributions | |
ist | -0.067 |
hole | -0.067 |
ak | -0.065 |
ches | -0.064 |
aron | -0.062 |
Token | . |
Token ID | 13 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
Sorry | +0.057 |
bye | +0.053 |
bye | +0.046 |
ww | +0.046 |
floats | +0.045 |
Neg logit contributions | |
ist | -0.036 |
hole | -0.033 |
aron | -0.032 |
Sad | -0.030 |
ge | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
bye | +0.028 |
ho | +0.028 |
Sorry | +0.027 |
good | +0.025 |
day | +0.025 |
Neg logit contributions | |
Sad | -0.037 |
uster | -0.035 |
James | -0.033 |
hole | -0.032 |
ice | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
Sorry | +0.016 |
bye | +0.015 |
good | +0.015 |
they | +0.014 |
day | +0.014 |
Neg logit contributions | |
uster | -0.027 |
hole | -0.027 |
Sad | -0.027 |
ice | -0.026 |
father | -0.025 |
Token | Tim |
Token ID | 14967 |
Feature activation | +5.88e-04 |
Pos logit contributions | |
Sad | +0.018 |
uster | +0.018 |
hole | +0.017 |
ice | +0.017 |
James | +0.016 |
Neg logit contributions | |
Sorry | -0.018 |
bye | -0.018 |
good | -0.017 |
bye | -0.016 |
day | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
Sorry | +0.063 |
bye | +0.059 |
ho | +0.058 |
moms | +0.054 |
Hipp | +0.054 |
Neg logit contributions | |
hole | -0.031 |
pled | -0.028 |
aron | -0.027 |
ist | -0.026 |
uster | -0.026 |
Token | They |
Token ID | 1119 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
bye | +0.017 |
ho | +0.017 |
Sorry | +0.016 |
day | +0.015 |
good | +0.015 |
Neg logit contributions | |
Sad | -0.023 |
uster | -0.022 |
hole | -0.021 |
aron | -0.020 |
James | -0.020 |
Token | hug |
Token ID | 16225 |
Feature activation | +5.43e-03 |
Pos logit contributions | |
Sorry | +0.025 |
bye | +0.023 |
ho | +0.020 |
day | +0.019 |
they | +0.019 |
Neg logit contributions | |
uster | -0.031 |
ist | -0.031 |
hole | -0.030 |
Sad | -0.029 |
ice | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
Sorry | +0.009 |
bye | +0.008 |
ho | +0.007 |
good | +0.007 |
Prize | +0.007 |
Neg logit contributions | |
Sad | -0.011 |
ice | -0.010 |
ist | -0.010 |
uster | -0.010 |
aron | -0.010 |
Token | They |
Token ID | 1119 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
ho | +0.032 |
bye | +0.032 |
Sorry | +0.031 |
day | +0.028 |
good | +0.027 |
Neg logit contributions | |
Sad | -0.041 |
uster | -0.038 |
aron | -0.036 |
James | -0.035 |
hole | -0.035 |
Token | swim |
Token ID | 9422 |
Feature activation | +6.48e-02 |
Pos logit contributions | |
Sorry | +0.028 |
bye | +0.025 |
ho | +0.023 |
day | +0.022 |
they | +0.022 |
Neg logit contributions | |
ist | -0.033 |
uster | -0.032 |
hole | -0.032 |
ice | -0.031 |
Sad | -0.031 |
Token | back |
Token ID | 736 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
Sorry | +0.148 |
bye | +0.121 |
ww | +0.119 |
Gig | +0.119 |
sunset | +0.117 |
Neg logit contributions | |
ist | -0.072 |
hole | -0.070 |
run | -0.068 |
uster | -0.065 |
aron | -0.061 |
Token | to |
Token ID | 284 |
Feature activation | +8.55e-03 |
Pos logit contributions | |
Sorry | +0.043 |
bye | +0.035 |
bye | +0.032 |
belong | +0.032 |
Prize | +0.032 |
Neg logit contributions | |
hole | -0.031 |
ist | -0.028 |
run | -0.027 |
Sad | -0.025 |
uster | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | -5.87e-03 |
Pos logit contributions | |
bye | +0.011 |
Sorry | +0.011 |
ho | +0.010 |
ww | +0.009 |
good | +0.008 |
Neg logit contributions | |
Sad | -0.018 |
father | -0.017 |
uster | -0.017 |
ice | -0.017 |
injuries | -0.016 |
Token | pool |
Token ID | 5933 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
uster | +0.012 |
hole | +0.012 |
Sad | +0.011 |
eping | +0.011 |
father | +0.011 |
Neg logit contributions | |
Sorry | -0.008 |
bye | -0.008 |
they | -0.007 |
ho | -0.006 |
bye | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
Sorry | +0.057 |
bye | +0.047 |
Fif | +0.045 |
good | +0.044 |
ww | +0.044 |
Neg logit contributions | |
hole | -0.044 |
ist | -0.042 |
aron | -0.037 |
pled | -0.037 |
box | -0.036 |
Token | pasta |
Token ID | 26296 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
ice | +0.032 |
Sad | +0.031 |
father | +0.030 |
pled | +0.030 |
ilda | +0.030 |
Neg logit contributions | |
bye | -0.024 |
Sorry | -0.024 |
ho | -0.022 |
bye | -0.019 |
day | -0.019 |
Token | soon |
Token ID | 2582 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
Sorry | +0.037 |
bye | +0.036 |
ho | +0.032 |
Hipp | +0.030 |
ge | +0.030 |
Neg logit contributions | |
ist | -0.037 |
ice | -0.033 |
hole | -0.033 |
box | -0.032 |
aker | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | +8.40e-03 |
Pos logit contributions | |
hole | +0.028 |
ice | +0.028 |
ist | +0.027 |
pled | +0.027 |
uster | +0.026 |
Neg logit contributions | |
Sorry | -0.029 |
bye | -0.028 |
ho | -0.024 |
Hipp | -0.023 |
bye | -0.023 |
Token | Mia |
Token ID | 32189 |
Feature activation | +2.44e-02 |
Pos logit contributions | |
bye | +0.013 |
Sorry | +0.013 |
good | +0.012 |
day | +0.012 |
ho | +0.011 |
Neg logit contributions | |
Sad | -0.016 |
uster | -0.015 |
hole | -0.015 |
ice | -0.015 |
innie | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | -4.74e-03 |
Pos logit contributions | |
Sorry | +0.043 |
bye | +0.041 |
good | +0.037 |
day | +0.036 |
they | +0.036 |
Neg logit contributions | |
ice | -0.045 |
hole | -0.044 |
uster | -0.044 |
Sad | -0.043 |
ist | -0.042 |
Token | Leo |
Token ID | 19632 |
Feature activation | +6.30e-02 |
Pos logit contributions | |
father | +0.008 |
ice | +0.008 |
Steven | +0.008 |
Sad | +0.007 |
hole | +0.007 |
Neg logit contributions | |
Sorry | -0.009 |
bye | -0.008 |
good | -0.008 |
well | -0.007 |
day | -0.007 |
Token | are |
Token ID | 389 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
Sorry | +0.098 |
bye | +0.093 |
ho | +0.078 |
bye | +0.076 |
ww | +0.072 |
Neg logit contributions | |
ist | -0.118 |
ice | -0.116 |
hole | -0.115 |
uster | -0.113 |
father | -0.110 |
Token | hungry |
Token ID | 14720 |
Feature activation | -9.56e-03 |
Pos logit contributions | |
bye | +0.016 |
Sorry | +0.016 |
ho | +0.014 |
bye | +0.012 |
seal | +0.011 |
Neg logit contributions | |
Sad | -0.021 |
eping | -0.021 |
ice | -0.020 |
pled | -0.020 |
hole | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
hole | +0.017 |
ice | +0.017 |
pled | +0.016 |
uster | +0.016 |
Sad | +0.016 |
Neg logit contributions | |
Sorry | -0.018 |
bye | -0.017 |
ho | -0.014 |
bye | -0.014 |
good | -0.014 |
Token | bored |
Token ID | 17533 |
Feature activation | +5.33e-03 |
Pos logit contributions | |
bye | +0.028 |
Sorry | +0.025 |
ho | +0.023 |
day | +0.022 |
Hipp | +0.021 |
Neg logit contributions | |
ice | -0.032 |
hole | -0.032 |
pled | -0.031 |
Sad | -0.031 |
uster | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
Sorry | +0.010 |
bye | +0.010 |
ho | +0.009 |
bye | +0.008 |
sw | +0.008 |
Neg logit contributions | |
hole | -0.009 |
ice | -0.009 |
pled | -0.009 |
Sad | -0.009 |
uster | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
Sorry | +0.015 |
bye | +0.015 |
ho | +0.012 |
bye | +0.011 |
day | +0.010 |
Neg logit contributions | |
Sad | -0.029 |
uster | -0.029 |
hole | -0.028 |
ice | -0.028 |
James | -0.027 |
Token | cheek |
Token ID | 19353 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
hole | +0.031 |
uster | +0.029 |
x | +0.028 |
ilda | +0.028 |
ist | +0.028 |
Neg logit contributions | |
Sorry | -0.027 |
they | -0.025 |
bye | -0.023 |
ho | -0.021 |
good | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
Sorry | +0.019 |
good | +0.017 |
bye | +0.017 |
ge | +0.015 |
seals | +0.015 |
Neg logit contributions | |
hole | -0.020 |
ist | -0.019 |
uster | -0.019 |
Sad | -0.018 |
ady | -0.018 |
Token | Max |
Token ID | 5436 |
Feature activation | +3.31e-02 |
Pos logit contributions | |
bye | +0.026 |
Sorry | +0.024 |
ho | +0.024 |
day | +0.023 |
game | +0.023 |
Neg logit contributions | |
uster | -0.030 |
aron | -0.028 |
Sad | -0.028 |
hole | -0.027 |
ice | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | +1.00e-02 |
Pos logit contributions | |
bye | +0.067 |
Sorry | +0.067 |
day | +0.060 |
ho | +0.057 |
good | +0.054 |
Neg logit contributions | |
ice | -0.051 |
hole | -0.046 |
pled | -0.046 |
aron | -0.046 |
ist | -0.046 |
Token | Luna |
Token ID | 23694 |
Feature activation | +6.29e-02 |
Pos logit contributions | |
bye | +0.021 |
Sorry | +0.020 |
ho | +0.017 |
day | +0.016 |
good | +0.016 |
Neg logit contributions | |
ice | -0.015 |
uster | -0.014 |
father | -0.013 |
ilda | -0.013 |
Sad | -0.013 |
Token | kiss |
Token ID | 9245 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
Sorry | +0.104 |
bye | +0.100 |
ho | +0.087 |
ww | +0.083 |
day | +0.082 |
Neg logit contributions | |
ist | -0.105 |
hole | -0.100 |
ice | -0.099 |
pping | -0.097 |
ches | -0.097 |
Token | each |
Token ID | 1123 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
Sorry | +0.043 |
bye | +0.039 |
nation | +0.034 |
Prize | +0.034 |
ww | +0.034 |
Neg logit contributions | |
aron | -0.035 |
Sad | -0.035 |
uster | -0.034 |
ice | -0.033 |
ist | -0.033 |
Token | other |
Token ID | 584 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
uster | +0.029 |
hole | +0.028 |
Sad | +0.028 |
ice | +0.028 |
father | +0.027 |
Neg logit contributions | |
Sorry | -0.014 |
bye | -0.011 |
they | -0.010 |
ho | -0.010 |
good | -0.010 |
Token | on |
Token ID | 319 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
Sorry | +0.047 |
bye | +0.038 |
ge | +0.036 |
ho | +0.035 |
moms | +0.035 |
Neg logit contributions | |
hole | -0.034 |
ist | -0.031 |
ice | -0.030 |
pled | -0.029 |
uster | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
Sorry | +0.014 |
bye | +0.014 |
ho | +0.011 |
bye | +0.010 |
day | +0.009 |
Neg logit contributions | |
Sad | -0.024 |
uster | -0.024 |
ice | -0.024 |
hole | -0.023 |
James | -0.023 |
Token | of |
Token ID | 286 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
Sorry | +0.013 |
bye | +0.011 |
bye | +0.011 |
ho | +0.011 |
good | +0.011 |
Neg logit contributions | |
uster | -0.009 |
Sad | -0.009 |
hole | -0.009 |
ist | -0.009 |
ice | -0.009 |
Token | fun |
Token ID | 1257 |
Feature activation | +9.25e-03 |
Pos logit contributions | |
good | +0.011 |
Sorry | +0.010 |
bye | +0.009 |
day | +0.009 |
they | +0.009 |
Neg logit contributions | |
uster | -0.058 |
hole | -0.057 |
Sad | -0.057 |
ice | -0.055 |
father | -0.055 |
Token | . |
Token ID | 13 |
Feature activation | +4.61e-03 |
Pos logit contributions | |
Sorry | +0.014 |
bye | +0.013 |
good | +0.012 |
bye | +0.011 |
ho | +0.011 |
Neg logit contributions | |
hole | -0.020 |
Sad | -0.020 |
uster | -0.019 |
ist | -0.019 |
ice | -0.019 |
Token | Max |
Token ID | 5436 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
bye | +0.007 |
Sorry | +0.007 |
ho | +0.007 |
day | +0.006 |
good | +0.006 |
Neg logit contributions | |
Sad | -0.009 |
uster | -0.009 |
hole | -0.008 |
ice | -0.008 |
aron | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | +5.25e-03 |
Pos logit contributions | |
Sorry | +0.068 |
bye | +0.064 |
day | +0.054 |
ho | +0.054 |
birthday | +0.052 |
Neg logit contributions | |
ice | -0.061 |
hole | -0.057 |
pled | -0.056 |
aron | -0.054 |
ist | -0.054 |
Token | Luna |
Token ID | 23694 |
Feature activation | +6.25e-02 |
Pos logit contributions | |
bye | +0.012 |
Sorry | +0.011 |
ho | +0.010 |
yacht | +0.010 |
marriage | +0.009 |
Neg logit contributions | |
ice | -0.007 |
Sad | -0.006 |
father | -0.006 |
uster | -0.006 |
ches | -0.006 |
Token | like |
Token ID | 588 |
Feature activation | +3.99e-03 |
Pos logit contributions | |
Sorry | +0.110 |
bye | +0.103 |
ho | +0.085 |
ww | +0.082 |
thing | +0.077 |
Neg logit contributions | |
pping | -0.102 |
ice | -0.097 |
ist | -0.094 |
ches | -0.094 |
cel | -0.094 |
Token | each |
Token ID | 1123 |
Feature activation | -2.33e-02 |
Pos logit contributions | |
Sorry | +0.008 |
bye | +0.007 |
ho | +0.005 |
bye | +0.005 |
ww | +0.005 |
Neg logit contributions | |
Sad | -0.007 |
pled | -0.007 |
ice | -0.007 |
uster | -0.007 |
James | -0.007 |
Token | other |
Token ID | 584 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
uster | +0.062 |
Sad | +0.061 |
hole | +0.061 |
ice | +0.060 |
father | +0.058 |
Neg logit contributions | |
Sorry | -0.023 |
bye | -0.020 |
good | -0.017 |
they | -0.017 |
bye | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
Sorry | +0.046 |
bye | +0.042 |
ho | +0.041 |
ge | +0.036 |
bye | +0.036 |
Neg logit contributions | |
pled | -0.028 |
ice | -0.028 |
hole | -0.027 |
aron | -0.026 |
ist | -0.025 |
Token | They |
Token ID | 1119 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
bye | +0.015 |
Sorry | +0.014 |
ho | +0.013 |
good | +0.013 |
day | +0.013 |
Neg logit contributions | |
Sad | -0.021 |
uster | -0.021 |
hole | -0.020 |
ice | -0.020 |
James | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +2.57e-02 |
Pos logit contributions | |
Sorry | +0.071 |
bye | +0.064 |
bye | +0.058 |
Prize | +0.057 |
festival | +0.054 |
Neg logit contributions | |
hole | -0.038 |
ist | -0.037 |
pled | -0.035 |
ice | -0.033 |
aron | -0.032 |
Token | They |
Token ID | 1119 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
bye | +0.037 |
Sorry | +0.034 |
ho | +0.033 |
day | +0.033 |
good | +0.032 |
Neg logit contributions | |
Sad | -0.053 |
uster | -0.051 |
hole | -0.049 |
ice | -0.048 |
pled | -0.046 |
Token | qu |
Token ID | 627 |
Feature activation | +2.91e-02 |
Pos logit contributions | |
bye | +0.038 |
Sorry | +0.036 |
day | +0.031 |
bye | +0.031 |
ww | +0.029 |
Neg logit contributions | |
hole | -0.041 |
ist | -0.039 |
ice | -0.039 |
pled | -0.039 |
Sad | -0.038 |
Token | ack |
Token ID | 441 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
Sorry | +0.054 |
good | +0.052 |
bye | +0.052 |
they | +0.050 |
day | +0.049 |
Neg logit contributions | |
uster | -0.050 |
Sad | -0.049 |
hole | -0.049 |
father | -0.046 |
ice | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
Sorry | +0.100 |
bye | +0.093 |
day | +0.083 |
ho | +0.081 |
Prize | +0.080 |
Neg logit contributions | |
ice | -0.045 |
pled | -0.043 |
ist | -0.041 |
eping | -0.040 |
aron | -0.038 |
Token | swim |
Token ID | 9422 |
Feature activation | +6.17e-02 |
Pos logit contributions | |
bye | +0.045 |
Sorry | +0.038 |
day | +0.035 |
ho | +0.035 |
ww | +0.031 |
Neg logit contributions | |
hole | -0.043 |
pled | -0.041 |
Sad | -0.041 |
ist | -0.041 |
aron | -0.041 |
Token | and |
Token ID | 290 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
Sorry | +0.151 |
bye | +0.126 |
marriage | +0.122 |
sunset | +0.122 |
Prize | +0.120 |
Neg logit contributions | |
ist | -0.064 |
aron | -0.061 |
hole | -0.059 |
ches | -0.058 |
run | -0.056 |
Token | flap |
Token ID | 37699 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
bye | +0.057 |
Sorry | +0.053 |
ww | +0.046 |
Prize | +0.043 |
bye | +0.043 |
Neg logit contributions | |
hole | -0.056 |
oscope | -0.053 |
ist | -0.053 |
injuries | -0.052 |
Sad | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | +2.58e-02 |
Pos logit contributions | |
Sorry | +0.048 |
bye | +0.043 |
festival | +0.038 |
Prize | +0.037 |
day | +0.035 |
Neg logit contributions | |
Sad | -0.059 |
hole | -0.058 |
ist | -0.057 |
pled | -0.057 |
uster | -0.055 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
bye | +0.038 |
Sorry | +0.036 |
ho | +0.035 |
day | +0.032 |
good | +0.032 |
Neg logit contributions | |
Sad | -0.054 |
uster | -0.050 |
hole | -0.047 |
ice | -0.046 |
James | -0.046 |
Token | laughs |
Token ID | 22051 |
Feature activation | +2.81e-02 |
Pos logit contributions | |
Sorry | +0.042 |
bye | +0.041 |
day | +0.036 |
bye | +0.034 |
good | +0.034 |
Neg logit contributions | |
ice | -0.037 |
pled | -0.036 |
hole | -0.036 |
ist | -0.033 |
Sad | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | +5.00e-04 |
Pos logit contributions | |
Sorry | +0.014 |
bye | +0.013 |
bye | +0.011 |
good | +0.011 |
day | +0.011 |
Neg logit contributions | |
hole | -0.018 |
Sad | -0.018 |
uster | -0.017 |
eping | -0.016 |
ist | -0.016 |
Token | They |
Token ID | 1119 |
Feature activation | +8.81e-03 |
Pos logit contributions | |
Sorry | +0.001 |
bye | +0.001 |
good | +0.001 |
they | +0.001 |
day | +0.001 |
Neg logit contributions | |
uster | -0.001 |
Sad | -0.001 |
hole | -0.001 |
ice | -0.001 |
father | -0.001 |
Token | qu |
Token ID | 627 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
Sorry | +0.013 |
bye | +0.012 |
good | +0.011 |
day | +0.011 |
they | +0.011 |
Neg logit contributions | |
hole | -0.018 |
uster | -0.018 |
Sad | -0.018 |
ist | -0.017 |
ice | -0.017 |
Token | ack |
Token ID | 441 |
Feature activation | +3.14e-02 |
Pos logit contributions | |
Sorry | +0.049 |
bye | +0.048 |
good | +0.047 |
they | +0.046 |
day | +0.045 |
Neg logit contributions | |
uster | -0.033 |
Sad | -0.033 |
hole | -0.033 |
ice | -0.030 |
father | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
Sorry | +0.069 |
bye | +0.065 |
day | +0.060 |
birthday | +0.058 |
sunset | +0.058 |
Neg logit contributions | |
ice | -0.045 |
ist | -0.044 |
pled | -0.041 |
eping | -0.041 |
ivering | -0.039 |
Token | swim |
Token ID | 9422 |
Feature activation | +6.15e-02 |
Pos logit contributions | |
bye | +0.036 |
Sorry | +0.031 |
day | +0.029 |
birthday | +0.028 |
good | +0.027 |
Neg logit contributions | |
ist | -0.038 |
hole | -0.038 |
ice | -0.037 |
run | -0.036 |
eping | -0.036 |
Token | away |
Token ID | 1497 |
Feature activation | +2.81e-02 |
Pos logit contributions | |
Sorry | +0.119 |
marriage | +0.101 |
sunset | +0.101 |
bye | +0.101 |
birthday | +0.101 |
Neg logit contributions | |
ist | -0.091 |
hole | -0.088 |
run | -0.084 |
ches | -0.081 |
uster | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
Sorry | +0.062 |
bye | +0.057 |
floats | +0.049 |
bye | +0.049 |
moms | +0.049 |
Neg logit contributions | |
ist | -0.040 |
hole | -0.036 |
aron | -0.034 |
run | -0.033 |
ge | -0.032 |
Token | Tom |
Token ID | 4186 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
bye | +0.017 |
Sorry | +0.016 |
good | +0.016 |
day | +0.015 |
ho | +0.014 |
Neg logit contributions | |
Sad | -0.025 |
uster | -0.025 |
hole | -0.024 |
ice | -0.023 |
James | -0.023 |
Token | is |
Token ID | 318 |
Feature activation | -8.91e-03 |
Pos logit contributions | |
Sorry | +0.026 |
bye | +0.024 |
good | +0.022 |
day | +0.022 |
they | +0.021 |
Neg logit contributions | |
ice | -0.024 |
hole | -0.023 |
uster | -0.023 |
pled | -0.022 |
ist | -0.022 |
Token | scared |
Token ID | 12008 |
Feature activation | -6.90e-03 |
Pos logit contributions | |
Sad | +0.016 |
eping | +0.016 |
uster | +0.016 |
ivering | +0.015 |
hole | +0.015 |
Neg logit contributions | |
Sorry | -0.016 |
bye | -0.015 |
ho | -0.014 |
day | -0.013 |
they | -0.012 |
Token | gifted |
Token ID | 22527 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
Sad | +0.001 |
uster | +0.001 |
ice | +0.001 |
hole | +0.001 |
father | +0.001 |
Neg logit contributions | |
bye | -0.001 |
Sorry | -0.001 |
good | -0.000 |
day | -0.000 |
they | -0.000 |
Token | book |
Token ID | 1492 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
Sorry | +0.031 |
bye | +0.028 |
good | +0.027 |
they | +0.026 |
day | +0.025 |
Neg logit contributions | |
uster | -0.041 |
ice | -0.040 |
father | -0.040 |
hole | -0.040 |
ist | -0.040 |
Token | ." |
Token ID | 526 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
Sorry | +0.039 |
good | +0.033 |
bye | +0.032 |
ge | +0.032 |
sunset | +0.032 |
Neg logit contributions | |
ist | -0.040 |
hole | -0.040 |
Sad | -0.037 |
aron | -0.036 |
ice | -0.036 |
Token | Mia |
Token ID | 32189 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
bye | +0.038 |
Sorry | +0.038 |
good | +0.036 |
day | +0.033 |
ho | +0.033 |
Neg logit contributions | |
Sad | -0.043 |
father | -0.039 |
uster | -0.039 |
hole | -0.039 |
ice | -0.039 |
Token | and |
Token ID | 290 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
Sorry | +0.042 |
bye | +0.040 |
good | +0.038 |
they | +0.036 |
day | +0.036 |
Neg logit contributions | |
hole | -0.051 |
uster | -0.050 |
Sad | -0.050 |
ice | -0.049 |
father | -0.048 |
Token | Leo |
Token ID | 19632 |
Feature activation | +6.06e-02 |
Pos logit contributions | |
bye | +0.039 |
Sorry | +0.038 |
good | +0.035 |
day | +0.033 |
ww | +0.032 |
Neg logit contributions | |
ice | -0.028 |
father | -0.027 |
Sad | -0.026 |
pled | -0.026 |
hole | -0.026 |
Token | say |
Token ID | 910 |
Feature activation | +9.43e-03 |
Pos logit contributions | |
Sorry | +0.109 |
bye | +0.104 |
good | +0.094 |
ho | +0.092 |
ww | +0.089 |
Neg logit contributions | |
hole | -0.098 |
ist | -0.096 |
father | -0.095 |
uster | -0.092 |
ice | -0.091 |
Token | . |
Token ID | 13 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
Sorry | +0.016 |
bye | +0.015 |
ge | +0.014 |
ho | +0.014 |
seals | +0.014 |
Neg logit contributions | |
uster | -0.017 |
Sad | -0.017 |
ist | -0.016 |
hole | -0.016 |
pled | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
bye | +0.017 |
Sorry | +0.017 |
good | +0.016 |
day | +0.015 |
well | +0.015 |
Neg logit contributions | |
Sad | -0.023 |
uster | -0.021 |
hole | -0.021 |
ice | -0.020 |
Steven | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -9.08e-03 |
Pos logit contributions | |
Sorry | +0.014 |
they | +0.013 |
good | +0.013 |
bye | +0.013 |
day | +0.012 |
Neg logit contributions | |
uster | -0.030 |
hole | -0.030 |
Sad | -0.029 |
ice | -0.029 |
ist | -0.028 |
Token | " |
Token ID | 1 |
Feature activation | +2.48e-02 |
Pos logit contributions | |
Sad | +0.017 |
uster | +0.015 |
hole | +0.015 |
James | +0.015 |
ice | +0.015 |
Neg logit contributions | |
Sorry | -0.016 |
bye | -0.015 |
good | -0.015 |
day | -0.015 |
bye | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
bye | +0.022 |
Sorry | +0.021 |
day | +0.020 |
good | +0.020 |
ho | +0.019 |
Neg logit contributions | |
uster | -0.019 |
Sad | -0.019 |
aron | -0.017 |
ice | -0.017 |
hole | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | -8.00e-03 |
Pos logit contributions | |
Sorry | +0.027 |
good | +0.026 |
bye | +0.026 |
they | +0.025 |
day | +0.024 |
Neg logit contributions | |
uster | -0.053 |
hole | -0.052 |
Sad | -0.051 |
ice | -0.050 |
ist | -0.049 |
Token | S |
Token ID | 50 |
Feature activation | +5.17e-03 |
Pos logit contributions | |
Sad | +0.015 |
uster | +0.015 |
ice | +0.014 |
hole | +0.014 |
ivering | +0.014 |
Neg logit contributions | |
Sorry | -0.013 |
bye | -0.012 |
day | -0.011 |
good | -0.011 |
bye | -0.011 |
Token | ara |
Token ID | 3301 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
Sorry | +0.011 |
bye | +0.011 |
good | +0.011 |
they | +0.010 |
day | +0.010 |
Neg logit contributions | |
uster | -0.007 |
hole | -0.007 |
Sad | -0.007 |
ice | -0.007 |
father | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | +2.01e-03 |
Pos logit contributions | |
Sorry | +0.044 |
bye | +0.041 |
good | +0.038 |
day | +0.038 |
they | +0.038 |
Neg logit contributions | |
hole | -0.035 |
ice | -0.034 |
uster | -0.034 |
Sad | -0.033 |
ist | -0.033 |
Token | Spot |
Token ID | 15899 |
Feature activation | +6.04e-02 |
Pos logit contributions | |
Sorry | +0.004 |
bye | +0.003 |
vows | +0.003 |
ho | +0.003 |
birthday | +0.003 |
Neg logit contributions | |
ice | -0.003 |
uster | -0.003 |
father | -0.003 |
x | -0.003 |
hole | -0.003 |
Token | go |
Token ID | 467 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
Sorry | +0.096 |
bye | +0.087 |
ho | +0.083 |
ww | +0.075 |
day | +0.074 |
Neg logit contributions | |
hole | -0.099 |
pping | -0.096 |
ist | -0.095 |
uster | -0.093 |
aron | -0.092 |
Token | back |
Token ID | 736 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
Sorry | +0.037 |
bye | +0.031 |
good | +0.028 |
bye | +0.027 |
ho | +0.027 |
Neg logit contributions | |
hole | -0.054 |
uster | -0.053 |
Sad | -0.052 |
ist | -0.049 |
father | -0.048 |
Token | to |
Token ID | 284 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
Sorry | +0.029 |
bye | +0.024 |
bye | +0.022 |
good | +0.022 |
ho | +0.021 |
Neg logit contributions | |
hole | -0.037 |
Sad | -0.035 |
uster | -0.034 |
father | -0.032 |
ist | -0.032 |
Token | playing |
Token ID | 2712 |
Feature activation | +6.47e-03 |
Pos logit contributions | |
bye | +0.016 |
Sorry | +0.016 |
ho | +0.013 |
good | +0.011 |
bye | +0.011 |
Neg logit contributions | |
Sad | -0.032 |
father | -0.030 |
hole | -0.030 |
uster | -0.030 |
innie | -0.029 |
Token | with |
Token ID | 351 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
Sorry | +0.010 |
bye | +0.009 |
ho | +0.008 |
bye | +0.008 |
good | +0.007 |
Neg logit contributions | |
hole | -0.013 |
Sad | -0.013 |
ist | -0.013 |
eping | -0.013 |
uster | -0.012 |
Token | love |
Token ID | 1842 |
Feature activation | +5.30e-03 |
Pos logit contributions | |
Sorry | +0.024 |
bye | +0.022 |
bye | +0.021 |
moms | +0.020 |
good | +0.020 |
Neg logit contributions | |
ist | -0.026 |
ice | -0.024 |
uster | -0.024 |
ady | -0.024 |
Sad | -0.024 |
Token | you |
Token ID | 345 |
Feature activation | -8.28e-03 |
Pos logit contributions | |
Sorry | +0.009 |
bye | +0.008 |
ho | +0.007 |
sw | +0.007 |
good | +0.007 |
Neg logit contributions | |
ice | -0.009 |
ilda | -0.008 |
uster | -0.008 |
father | -0.008 |
innie | -0.008 |
Token | ". |
Token ID | 1911 |
Feature activation | +5.50e-03 |
Pos logit contributions | |
ice | +0.015 |
uster | +0.014 |
Sad | +0.014 |
hole | +0.014 |
father | +0.014 |
Neg logit contributions | |
Sorry | -0.014 |
bye | -0.013 |
ho | -0.013 |
moms | -0.012 |
they | -0.012 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
Sorry | +0.008 |
bye | +0.008 |
good | +0.007 |
they | +0.007 |
day | +0.007 |
Neg logit contributions | |
uster | -0.012 |
hole | -0.011 |
Sad | -0.011 |
ice | -0.011 |
father | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +6.07e-03 |
Pos logit contributions | |
Sorry | +0.032 |
bye | +0.028 |
good | +0.027 |
day | +0.025 |
bye | +0.025 |
Neg logit contributions | |
ice | -0.032 |
Sad | -0.031 |
ist | -0.030 |
hole | -0.030 |
ilda | -0.029 |
Token | Coco |
Token ID | 48222 |
Feature activation | +6.03e-02 |
Pos logit contributions | |
Sorry | +0.009 |
bye | +0.009 |
ho | +0.008 |
day | +0.008 |
floats | +0.008 |
Neg logit contributions | |
ilda | -0.010 |
ice | -0.010 |
Steven | -0.010 |
Jason | -0.010 |
uster | -0.010 |
Token | were |
Token ID | 547 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
Sorry | +0.126 |
bye | +0.106 |
summer | +0.095 |
job | +0.095 |
birthday | +0.094 |
Neg logit contributions | |
ist | -0.089 |
ice | -0.081 |
pled | -0.078 |
ches | -0.076 |
cel | -0.076 |
Token | best |
Token ID | 1266 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
bye | +0.020 |
Sorry | +0.020 |
ho | +0.018 |
wait | +0.014 |
bye | +0.014 |
Neg logit contributions | |
Sad | -0.022 |
ice | -0.022 |
ady | -0.022 |
pled | -0.021 |
eping | -0.021 |
Token | friends |
Token ID | 2460 |
Feature activation | +3.50e-04 |
Pos logit contributions | |
Sorry | +0.017 |
bye | +0.016 |
ho | +0.016 |
melon | +0.014 |
day | +0.014 |
Neg logit contributions | |
uster | -0.020 |
Sad | -0.020 |
ice | -0.020 |
father | -0.019 |
pled | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +3.15e-03 |
Pos logit contributions | |
Sorry | +0.001 |
bye | +0.001 |
good | +0.000 |
ho | +0.000 |
bye | +0.000 |
Neg logit contributions | |
hole | -0.001 |
Sad | -0.001 |
pled | -0.001 |
ice | -0.001 |
ist | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
Sorry | +0.004 |
bye | +0.004 |
good | +0.004 |
bye | +0.004 |
ho | +0.004 |
Neg logit contributions | |
uster | -0.007 |
Sad | -0.007 |
hole | -0.007 |
ice | -0.007 |
father | -0.006 |
Token | They |
Token ID | 1119 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
good | +0.037 |
day | +0.037 |
bye | +0.037 |
Sorry | +0.036 |
ho | +0.034 |
Neg logit contributions | |
uster | -0.051 |
hole | -0.049 |
Sad | -0.049 |
ist | -0.045 |
ice | -0.045 |
Token | are |
Token ID | 389 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
Sorry | +0.036 |
bye | +0.035 |
bye | +0.031 |
ho | +0.030 |
moms | +0.028 |
Neg logit contributions | |
hole | -0.046 |
ist | -0.044 |
uster | -0.041 |
eping | -0.041 |
ice | -0.041 |
Token | pretty |
Token ID | 2495 |
Feature activation | +3.07e-03 |
Pos logit contributions | |
bye | +0.024 |
Sorry | +0.023 |
ho | +0.020 |
bye | +0.018 |
day | +0.015 |
Neg logit contributions | |
eping | -0.034 |
hole | -0.033 |
Sad | -0.033 |
ady | -0.033 |
uster | -0.032 |
Token | and |
Token ID | 290 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
Sorry | +0.006 |
bye | +0.006 |
bye | +0.005 |
day | +0.005 |
good | +0.005 |
Neg logit contributions | |
hole | -0.005 |
uster | -0.005 |
Sad | -0.005 |
ice | -0.005 |
ist | -0.005 |
Token | they |
Token ID | 484 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
bye | +0.027 |
Sorry | +0.026 |
ho | +0.024 |
bye | +0.022 |
day | +0.020 |
Neg logit contributions | |
Sad | -0.036 |
eping | -0.034 |
hole | -0.034 |
uster | -0.033 |
ist | -0.033 |
Token | swim |
Token ID | 9422 |
Feature activation | +6.00e-02 |
Pos logit contributions | |
Sorry | +0.041 |
bye | +0.040 |
bye | +0.037 |
ho | +0.036 |
moms | +0.034 |
Neg logit contributions | |
hole | -0.053 |
ist | -0.052 |
uster | -0.049 |
eping | -0.049 |
ice | -0.049 |
Token | in |
Token ID | 287 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
Sorry | +0.113 |
bye | +0.094 |
bye | +0.091 |
ending | +0.086 |
Mall | +0.083 |
Neg logit contributions | |
ist | -0.102 |
hole | -0.099 |
uster | -0.094 |
ak | -0.089 |
ches | -0.086 |
Token | the |
Token ID | 262 |
Feature activation | +2.17e-03 |
Pos logit contributions | |
Sorry | +0.016 |
bye | +0.014 |
ho | +0.012 |
bye | +0.012 |
they | +0.012 |
Neg logit contributions | |
hole | -0.034 |
Sad | -0.034 |
uster | -0.033 |
ice | -0.032 |
eping | -0.032 |
Token | water |
Token ID | 1660 |
Feature activation | +2.63e-02 |
Pos logit contributions | |
Sorry | +0.003 |
bye | +0.003 |
they | +0.003 |
ho | +0.003 |
good | +0.002 |
Neg logit contributions | |
uster | -0.005 |
hole | -0.005 |
Sad | -0.005 |
eping | -0.005 |
father | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
Sorry | +0.050 |
bye | +0.043 |
bye | +0.041 |
good | +0.040 |
moms | +0.038 |
Neg logit contributions | |
hole | -0.049 |
ist | -0.046 |
uster | -0.043 |
ice | -0.043 |
pled | -0.041 |
Token | Can |
Token ID | 1680 |
Feature activation | -8.80e-03 |
Pos logit contributions | |
bye | +0.033 |
Sorry | +0.033 |
good | +0.032 |
day | +0.032 |
they | +0.030 |
Neg logit contributions | |
uster | -0.042 |
Sad | -0.042 |
hole | -0.041 |
ice | -0.039 |
ist | -0.037 |
Token |
|
Token ID | 198 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
they | +0.016 |
good | +0.016 |
Sorry | +0.014 |
day | +0.014 |
bye | +0.014 |
Neg logit contributions | |
uster | -0.050 |
hole | -0.048 |
ist | -0.048 |
ice | -0.047 |
Sad | -0.046 |
Token | S |
Token ID | 50 |
Feature activation | -8.32e-03 |
Pos logit contributions | |
uster | +0.029 |
Sad | +0.029 |
hole | +0.028 |
ice | +0.027 |
father | +0.026 |
Neg logit contributions | |
Sorry | -0.033 |
bye | -0.032 |
good | -0.031 |
they | -0.030 |
day | -0.030 |
Token | ara |
Token ID | 3301 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
Sad | +0.016 |
uster | +0.015 |
hole | +0.014 |
ist | +0.013 |
father | +0.013 |
Neg logit contributions | |
bye | -0.015 |
Sorry | -0.015 |
good | -0.015 |
bye | -0.014 |
they | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | -9.01e-04 |
Pos logit contributions | |
Sorry | +0.043 |
bye | +0.041 |
good | +0.038 |
day | +0.037 |
they | +0.036 |
Neg logit contributions | |
hole | -0.044 |
ice | -0.043 |
uster | -0.042 |
Sad | -0.042 |
ist | -0.041 |
Token | Tom |
Token ID | 4186 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
hole | +0.001 |
ice | +0.001 |
uster | +0.001 |
father | +0.001 |
pled | +0.001 |
Neg logit contributions | |
Sorry | -0.002 |
bye | -0.002 |
good | -0.001 |
day | -0.001 |
ho | -0.001 |
Token | swim |
Token ID | 9422 |
Feature activation | +6.00e-02 |
Pos logit contributions | |
Sorry | +0.067 |
bye | +0.062 |
day | +0.054 |
they | +0.052 |
ho | +0.052 |
Neg logit contributions | |
hole | -0.079 |
ist | -0.075 |
uster | -0.075 |
eping | -0.074 |
pping | -0.073 |
Token | together |
Token ID | 1978 |
Feature activation | +3.59e-02 |
Pos logit contributions | |
Sorry | +0.128 |
ww | +0.106 |
marriage | +0.103 |
bye | +0.103 |
summer | +0.102 |
Neg logit contributions | |
run | -0.072 |
hole | -0.070 |
ist | -0.070 |
uster | -0.068 |
ches | -0.064 |
Token | in |
Token ID | 287 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
Sorry | +0.082 |
bye | +0.075 |
ww | +0.069 |
ge | +0.061 |
moms | +0.061 |
Neg logit contributions | |
hole | -0.048 |
ist | -0.042 |
frantically | -0.041 |
pping | -0.039 |
aron | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | +7.10e-03 |
Pos logit contributions | |
Sorry | +0.014 |
bye | +0.013 |
ho | +0.010 |
bye | +0.010 |
they | +0.010 |
Neg logit contributions | |
hole | -0.028 |
Sad | -0.028 |
uster | -0.027 |
father | -0.026 |
pled | -0.026 |
Token | lake |
Token ID | 13546 |
Feature activation | +3.14e-02 |
Pos logit contributions | |
Sorry | +0.009 |
ho | +0.008 |
bye | +0.008 |
they | +0.007 |
Fif | +0.007 |
Neg logit contributions | |
uster | -0.015 |
eping | -0.015 |
ady | -0.015 |
hole | -0.015 |
icer | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | +6.61e-03 |
Pos logit contributions | |
Sorry | +0.069 |
bye | +0.059 |
ww | +0.059 |
Prize | +0.054 |
bye | +0.052 |
Neg logit contributions | |
hole | -0.046 |
ist | -0.045 |
ently | -0.039 |
aron | -0.038 |
ice | -0.038 |
Token |
|
Token ID | 198 |
Feature activation | +6.14e-03 |
Pos logit contributions | |
Sorry | +0.001 |
bye | +0.001 |
good | +0.001 |
they | +0.001 |
day | +0.001 |
Neg logit contributions | |
uster | -0.002 |
hole | -0.002 |
Sad | -0.002 |
ice | -0.002 |
father | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
good | +0.007 |
Sorry | +0.007 |
they | +0.006 |
bye | +0.006 |
day | +0.006 |
Neg logit contributions | |
uster | -0.015 |
hole | -0.015 |
Sad | -0.015 |
ist | -0.015 |
ice | -0.015 |
Token | L |
Token ID | 43 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
uster | +0.030 |
Sad | +0.030 |
hole | +0.030 |
ice | +0.028 |
father | +0.028 |
Neg logit contributions | |
Sorry | -0.022 |
bye | -0.021 |
good | -0.021 |
they | -0.020 |
day | -0.019 |
Token | ily |
Token ID | 813 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
hole | +0.034 |
Sad | +0.034 |
eping | +0.033 |
uster | +0.032 |
Steven | +0.032 |
Neg logit contributions | |
good | -0.019 |
bye | -0.018 |
well | -0.017 |
day | -0.017 |
bye | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | +4.58e-03 |
Pos logit contributions | |
Sorry | +0.022 |
bye | +0.020 |
good | +0.019 |
day | +0.019 |
they | +0.018 |
Neg logit contributions | |
Sad | -0.019 |
ice | -0.019 |
hole | -0.019 |
uster | -0.019 |
pled | -0.018 |
Token | Sunny |
Token ID | 32241 |
Feature activation | +5.98e-02 |
Pos logit contributions | |
Sorry | +0.007 |
bye | +0.007 |
floats | +0.006 |
good | +0.006 |
day | +0.006 |
Neg logit contributions | |
ice | -0.008 |
father | -0.008 |
innie | -0.008 |
Sad | -0.007 |
uster | -0.007 |
Token | hugged |
Token ID | 41130 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
Sorry | +0.117 |
bye | +0.100 |
good | +0.094 |
thing | +0.091 |
day | +0.091 |
Neg logit contributions | |
ist | -0.095 |
pled | -0.094 |
ice | -0.093 |
hole | -0.089 |
Sad | -0.084 |
Token | and |
Token ID | 290 |
Feature activation | +9.23e-03 |
Pos logit contributions | |
Sorry | +0.022 |
bye | +0.019 |
ho | +0.017 |
day | +0.016 |
Prize | +0.016 |
Neg logit contributions | |
Sad | -0.032 |
ice | -0.031 |
uster | -0.031 |
aron | -0.030 |
ist | -0.029 |
Token | kissed |
Token ID | 27946 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
Sorry | +0.018 |
bye | +0.016 |
ho | +0.016 |
Prize | +0.015 |
birthday | +0.015 |
Neg logit contributions | |
Sad | -0.013 |
pled | -0.013 |
ice | -0.012 |
Steven | -0.011 |
ist | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
Sorry | +0.026 |
bye | +0.023 |
Prize | +0.020 |
day | +0.019 |
dive | +0.019 |
Neg logit contributions | |
Sad | -0.036 |
ice | -0.036 |
uster | -0.035 |
aron | -0.034 |
pled | -0.033 |
Token | They |
Token ID | 1119 |
Feature activation | +1.75e-04 |
Pos logit contributions | |
bye | +0.018 |
Sorry | +0.017 |
ho | +0.016 |
day | +0.015 |
race | +0.015 |
Neg logit contributions | |
Sad | -0.020 |
uster | -0.020 |
aron | -0.020 |
ice | -0.019 |
hole | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -3.65e-03 |
Pos logit contributions | |
uster | +0.012 |
Sad | +0.012 |
hole | +0.012 |
ice | +0.012 |
ist | +0.011 |
Neg logit contributions | |
Sorry | -0.007 |
bye | -0.006 |
good | -0.006 |
they | -0.006 |
day | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +1.58e-04 |
Pos logit contributions | |
uster | +0.008 |
hole | +0.008 |
Sad | +0.008 |
ice | +0.008 |
father | +0.008 |
Neg logit contributions | |
Sorry | -0.005 |
bye | -0.005 |
good | -0.005 |
they | -0.004 |
day | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
they | +0.000 |
good | +0.000 |
Sorry | +0.000 |
bye | +0.000 |
day | +0.000 |
Neg logit contributions | |
uster | -0.000 |
hole | -0.000 |
ice | -0.000 |
ist | -0.000 |
father | -0.000 |
Token | Tom |
Token ID | 13787 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
uster | +0.049 |
hole | +0.047 |
Sad | +0.047 |
ice | +0.045 |
father | +0.045 |
Neg logit contributions | |
Sorry | -0.043 |
bye | -0.042 |
good | -0.041 |
they | -0.040 |
day | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | +3.38e-03 |
Pos logit contributions | |
Sorry | +0.039 |
bye | +0.038 |
good | +0.035 |
they | +0.034 |
day | +0.034 |
Neg logit contributions | |
uster | -0.040 |
hole | -0.040 |
ice | -0.039 |
Sad | -0.039 |
father | -0.038 |
Token | Anna |
Token ID | 11735 |
Feature activation | +5.96e-02 |
Pos logit contributions | |
Sorry | +0.006 |
bye | +0.006 |
ho | +0.006 |
good | +0.006 |
they | +0.005 |
Neg logit contributions | |
Sad | -0.006 |
hole | -0.005 |
ice | -0.005 |
uster | -0.005 |
James | -0.005 |
Token | play |
Token ID | 711 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
Sorry | +0.089 |
bye | +0.085 |
ho | +0.071 |
they | +0.070 |
day | +0.069 |
Neg logit contributions | |
hole | -0.117 |
uster | -0.116 |
ist | -0.116 |
ice | -0.113 |
pled | -0.108 |
Token | with |
Token ID | 351 |
Feature activation | +5.30e-04 |
Pos logit contributions | |
Sorry | +0.019 |
bye | +0.018 |
ho | +0.016 |
bye | +0.015 |
they | +0.015 |
Neg logit contributions | |
Sad | -0.035 |
hole | -0.034 |
uster | -0.033 |
ist | -0.033 |
ice | -0.033 |
Token | Lily |
Token ID | 20037 |
Feature activation | +6.39e-03 |
Pos logit contributions | |
Sorry | +0.001 |
bye | +0.001 |
bye | +0.000 |
good | +0.000 |
day | +0.000 |
Neg logit contributions | |
Sad | -0.001 |
ice | -0.001 |
uster | -0.001 |
hole | -0.001 |
pled | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
Sorry | +0.013 |
bye | +0.012 |
bye | +0.010 |
ge | +0.010 |
ho | +0.010 |
Neg logit contributions | |
hole | -0.012 |
ice | -0.011 |
pled | -0.011 |
ist | -0.010 |
Sad | -0.010 |
Token | other |
Token ID | 584 |
Feature activation | +6.77e-03 |
Pos logit contributions | |
bye | +0.043 |
Sorry | +0.042 |
ho | +0.035 |
good | +0.035 |
Hipp | +0.034 |
Neg logit contributions | |
Sad | -0.051 |
James | -0.051 |
hole | -0.050 |
father | -0.050 |
ilda | -0.050 |
Token | hug |
Token ID | 16225 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
Sorry | +0.045 |
bye | +0.043 |
ho | +0.042 |
day | +0.037 |
Hipp | +0.037 |
Neg logit contributions | |
Sad | -0.038 |
ist | -0.035 |
pled | -0.035 |
ice | -0.034 |
hole | -0.034 |
Token | each |
Token ID | 1123 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
Sorry | +0.025 |
bye | +0.024 |
ho | +0.022 |
good | +0.020 |
Prize | +0.020 |
Neg logit contributions | |
Sad | -0.029 |
ice | -0.028 |
ist | -0.027 |
uster | -0.027 |
father | -0.026 |
Token | other |
Token ID | 584 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
uster | +0.053 |
hole | +0.053 |
Sad | +0.052 |
ice | +0.050 |
ist | +0.050 |
Neg logit contributions | |
Sorry | -0.011 |
good | -0.011 |
bye | -0.010 |
they | -0.010 |
day | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
Sorry | +0.033 |
bye | +0.029 |
ho | +0.027 |
ge | +0.025 |
Hipp | +0.025 |
Neg logit contributions | |
hole | -0.027 |
ice | -0.027 |
pled | -0.026 |
ist | -0.026 |
Sad | -0.025 |
Token | They |
Token ID | 1119 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
bye | +0.020 |
Sorry | +0.020 |
ho | +0.019 |
good | +0.017 |
day | +0.016 |
Neg logit contributions | |
Sad | -0.034 |
uster | -0.033 |
hole | -0.031 |
ice | -0.030 |
James | -0.030 |
Token | swim |
Token ID | 9422 |
Feature activation | +5.95e-02 |
Pos logit contributions | |
Sorry | +0.023 |
bye | +0.021 |
they | +0.019 |
day | +0.019 |
ho | +0.018 |
Neg logit contributions | |
ist | -0.031 |
Sad | -0.030 |
hole | -0.030 |
ice | -0.030 |
uster | -0.030 |
Token | back |
Token ID | 736 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
Sorry | +0.139 |
Hipp | +0.112 |
bye | +0.112 |
sunset | +0.112 |
Mall | +0.111 |
Neg logit contributions | |
ist | -0.064 |
hole | -0.060 |
uster | -0.056 |
run | -0.055 |
aron | -0.054 |
Token | to |
Token ID | 284 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
Sorry | +0.045 |
bye | +0.037 |
Prize | +0.034 |
bye | +0.033 |
belong | +0.033 |
Neg logit contributions | |
hole | -0.031 |
ist | -0.029 |
run | -0.026 |
Sad | -0.026 |
ently | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | -4.04e-03 |
Pos logit contributions | |
Sorry | +0.017 |
bye | +0.017 |
ho | +0.015 |
bye | +0.012 |
good | +0.012 |
Neg logit contributions | |
Sad | -0.032 |
father | -0.030 |
ice | -0.030 |
uster | -0.029 |
innie | -0.028 |
Token | shallow |
Token ID | 19337 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
hole | +0.007 |
uster | +0.007 |
Sad | +0.007 |
father | +0.007 |
ice | +0.007 |
Neg logit contributions | |
Sorry | -0.007 |
bye | -0.006 |
they | -0.006 |
bye | -0.005 |
ho | -0.005 |
Token | end |
Token ID | 886 |
Feature activation | -3.33e-03 |
Pos logit contributions | |
Sorry | +0.042 |
they | +0.033 |
good | +0.032 |
bye | +0.031 |
bye | +0.031 |
Neg logit contributions | |
hole | -0.048 |
uster | -0.047 |
ist | -0.046 |
ice | -0.045 |
ivering | -0.044 |
Token | butterfly |
Token ID | 35113 |
Feature activation | -3.46e-03 |
Pos logit contributions | |
Sorry | +0.008 |
they | +0.007 |
bye | +0.007 |
good | +0.006 |
ho | +0.006 |
Neg logit contributions | |
hole | -0.011 |
Sad | -0.011 |
uster | -0.011 |
ice | -0.011 |
ist | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -9.26e-03 |
Pos logit contributions | |
hole | +0.007 |
uster | +0.007 |
ice | +0.007 |
Sad | +0.007 |
ist | +0.006 |
Neg logit contributions | |
Sorry | -0.006 |
bye | -0.005 |
good | -0.005 |
bye | -0.005 |
moms | -0.005 |
Token | Bob |
Token ID | 5811 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
innie | +0.013 |
ice | +0.013 |
uster | +0.013 |
aron | +0.012 |
ilda | +0.012 |
Neg logit contributions | |
Sorry | -0.018 |
good | -0.018 |
bye | -0.017 |
day | -0.017 |
they | -0.016 |
Token | o |
Token ID | 78 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
Sorry | +0.024 |
bye | +0.022 |
good | +0.022 |
they | +0.021 |
bye | +0.021 |
Neg logit contributions | |
ice | -0.018 |
Sad | -0.018 |
hole | -0.018 |
ist | -0.017 |
uster | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | +6.15e-03 |
Pos logit contributions | |
Sorry | +0.043 |
bye | +0.037 |
good | +0.036 |
bye | +0.033 |
birthday | +0.033 |
Neg logit contributions | |
ice | -0.041 |
pled | -0.038 |
Sad | -0.037 |
hole | -0.036 |
ist | -0.036 |
Token | Coco |
Token ID | 48222 |
Feature activation | +5.94e-02 |
Pos logit contributions | |
bye | +0.009 |
Sorry | +0.009 |
floats | +0.009 |
ho | +0.008 |
birthday | +0.008 |
Neg logit contributions | |
father | -0.010 |
ilda | -0.009 |
ice | -0.009 |
innie | -0.009 |
Sad | -0.009 |
Token | were |
Token ID | 547 |
Feature activation | +5.15e-03 |
Pos logit contributions | |
Sorry | +0.113 |
bye | +0.092 |
bye | +0.086 |
birthday | +0.082 |
marriage | +0.080 |
Neg logit contributions | |
ist | -0.098 |
ice | -0.093 |
pled | -0.093 |
hole | -0.092 |
Sad | -0.090 |
Token | very |
Token ID | 845 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
Sorry | +0.009 |
bye | +0.009 |
ho | +0.008 |
bye | +0.007 |
seafood | +0.006 |
Neg logit contributions | |
Sad | -0.009 |
ady | -0.009 |
eping | -0.008 |
pled | -0.008 |
ivering | -0.008 |
Token | happy |
Token ID | 3772 |
Feature activation | +2.19e-02 |
Pos logit contributions | |
bye | +0.039 |
fl | +0.037 |
Sorry | +0.037 |
als | +0.037 |
names | +0.035 |
Neg logit contributions | |
ivering | -0.024 |
Sad | -0.022 |
ady | -0.021 |
aky | -0.021 |
ak | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -8.93e-03 |
Pos logit contributions | |
Sorry | +0.041 |
bye | +0.038 |
good | +0.033 |
ho | +0.033 |
bye | +0.033 |
Neg logit contributions | |
pled | -0.035 |
hole | -0.035 |
Sad | -0.035 |
father | -0.034 |
ist | -0.033 |
Token | They |
Token ID | 1119 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
aron | +0.016 |
uster | +0.014 |
hole | +0.014 |
ice | +0.014 |
innie | +0.014 |
Neg logit contributions | |
Sorry | -0.013 |
good | -0.013 |
bye | -0.013 |
dance | -0.013 |
go | -0.012 |
Token | for |
Token ID | 329 |
Feature activation | -7.24e-03 |
Pos logit contributions | |
Sorry | +0.036 |
bye | +0.034 |
bye | +0.031 |
Hipp | +0.029 |
ge | +0.029 |
Neg logit contributions | |
box | -0.028 |
ist | -0.028 |
hole | -0.027 |
aker | -0.027 |
pled | -0.026 |
Token | dinner |
Token ID | 8073 |
Feature activation | +1.38e-03 |
Pos logit contributions | |
Sad | +0.015 |
ice | +0.015 |
uster | +0.015 |
innie | +0.015 |
father | +0.014 |
Neg logit contributions | |
bye | -0.011 |
Sorry | -0.010 |
ho | -0.009 |
day | -0.009 |
bye | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +3.63e-03 |
Pos logit contributions | |
Sorry | +0.003 |
bye | +0.002 |
bye | +0.002 |
ho | +0.002 |
ge | +0.002 |
Neg logit contributions | |
hole | -0.002 |
ist | -0.002 |
ice | -0.002 |
uster | -0.002 |
ilda | -0.002 |
Token | Tim |
Token ID | 5045 |
Feature activation | +7.88e-03 |
Pos logit contributions | |
Sorry | +0.005 |
bye | +0.005 |
good | +0.005 |
they | +0.004 |
day | +0.004 |
Neg logit contributions | |
uster | -0.008 |
Sad | -0.008 |
hole | -0.008 |
ice | -0.008 |
father | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | +9.32e-03 |
Pos logit contributions | |
Sorry | +0.010 |
bye | +0.010 |
good | +0.009 |
they | +0.009 |
day | +0.008 |
Neg logit contributions | |
uster | -0.018 |
Sad | -0.018 |
hole | -0.018 |
ice | -0.017 |
father | -0.017 |
Token | Luna |
Token ID | 23694 |
Feature activation | +5.94e-02 |
Pos logit contributions | |
Sorry | +0.017 |
bye | +0.016 |
good | +0.015 |
they | +0.015 |
day | +0.014 |
Neg logit contributions | |
Sad | -0.016 |
uster | -0.016 |
hole | -0.016 |
ice | -0.015 |
father | -0.015 |
Token | were |
Token ID | 547 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
Sorry | +0.096 |
bye | +0.091 |
bye | +0.082 |
seals | +0.075 |
sunset | +0.074 |
Neg logit contributions | |
ice | -0.109 |
pled | -0.104 |
Sad | -0.096 |
ist | -0.096 |
innie | -0.094 |
Token | very |
Token ID | 845 |
Feature activation | +2.83e-02 |
Pos logit contributions | |
bye | +0.019 |
Sorry | +0.017 |
ho | +0.016 |
bye | +0.015 |
seal | +0.013 |
Neg logit contributions | |
Sad | -0.026 |
pled | -0.025 |
ice | -0.025 |
eping | -0.025 |
hole | -0.024 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
bye | +0.045 |
Sorry | +0.038 |
fl | +0.038 |
als | +0.036 |
seals | +0.035 |
Neg logit contributions | |
ivering | -0.047 |
eping | -0.047 |
pled | -0.047 |
ady | -0.046 |
ice | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
Sorry | +0.034 |
bye | +0.033 |
ho | +0.030 |
bye | +0.029 |
good | +0.028 |
Neg logit contributions | |
pled | -0.030 |
ice | -0.029 |
hole | -0.029 |
Sad | -0.029 |
father | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
Sorry | +0.013 |
bye | +0.013 |
good | +0.012 |
they | +0.012 |
day | +0.011 |
Neg logit contributions | |
uster | -0.023 |
Sad | -0.023 |
hole | -0.023 |
ice | -0.022 |
father | -0.021 |
Token | corner |
Token ID | 5228 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
uster | +0.034 |
hole | +0.033 |
Sad | +0.033 |
ice | +0.032 |
ist | +0.031 |
Neg logit contributions | |
good | -0.035 |
Sorry | -0.034 |
bye | -0.034 |
day | -0.033 |
they | -0.032 |
Token | of |
Token ID | 286 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
hole | +0.028 |
Sad | +0.027 |
uster | +0.027 |
ice | +0.026 |
ist | +0.026 |
Neg logit contributions | |
Sorry | -0.025 |
bye | -0.023 |
good | -0.022 |
bye | -0.021 |
they | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
uster | +0.035 |
hole | +0.034 |
Sad | +0.034 |
ice | +0.033 |
ist | +0.033 |
Neg logit contributions | |
Sorry | -0.009 |
good | -0.009 |
bye | -0.008 |
they | -0.008 |
day | -0.007 |
Token | room |
Token ID | 2119 |
Feature activation | -6.88e-06 |
Pos logit contributions | |
uster | +0.028 |
hole | +0.026 |
Sad | +0.026 |
ice | +0.026 |
father | +0.026 |
Neg logit contributions | |
good | -0.026 |
day | -0.026 |
Sorry | -0.024 |
they | -0.023 |
bye | -0.023 |
Token | she |
Token ID | 673 |
Feature activation | -3.32e-02 |
Token | hadn |
Token ID | 8020 |
Feature activation | -7.82e-02 |
Pos logit contributions | |
uster | +0.066 |
hole | +0.066 |
Sad | +0.066 |
ice | +0.063 |
father | +0.062 |
Neg logit contributions | |
Sorry | -0.053 |
bye | -0.050 |
good | -0.049 |
they | -0.048 |
day | -0.046 |
Token | 't |
Token ID | 470 |
Feature activation | -3.66e-02 |
Pos logit contributions | |
uster | +0.175 |
Sad | +0.172 |
ist | +0.171 |
ice | +0.171 |
eping | +0.169 |
Neg logit contributions | |
go | -0.098 |
they | -0.097 |
day | -0.086 |
bye | -0.086 |
good | -0.085 |
Token | seen |
Token ID | 1775 |
Feature activation | -6.54e-02 |
Pos logit contributions | |
uster | +0.039 |
Sad | +0.038 |
hole | +0.037 |
ice | +0.034 |
father | +0.033 |
Neg logit contributions | |
Sorry | -0.093 |
bye | -0.090 |
good | -0.090 |
they | -0.088 |
day | -0.086 |
Token | before |
Token ID | 878 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
uster | +0.161 |
Sad | +0.160 |
hole | +0.152 |
father | +0.148 |
innie | +0.147 |
Neg logit contributions | |
good | -0.071 |
bye | -0.070 |
day | -0.068 |
Sorry | -0.067 |
they | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
hole | +0.039 |
Sad | +0.037 |
uster | +0.037 |
ice | +0.036 |
ilda | +0.036 |
Neg logit contributions | |
Sorry | -0.030 |
bye | -0.028 |
good | -0.026 |
bye | -0.026 |
moms | -0.024 |
Token | It |
Token ID | 632 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
hole | +0.026 |
uster | +0.026 |
Sad | +0.025 |
ice | +0.024 |
ilda | +0.024 |
Neg logit contributions | |
bye | -0.030 |
Sorry | -0.029 |
good | -0.029 |
day | -0.028 |
they | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | -1.68e-03 |
Pos logit contributions | |
Sorry | +0.012 |
bye | +0.012 |
good | +0.011 |
they | +0.011 |
day | +0.011 |
Neg logit contributions | |
uster | -0.026 |
Sad | -0.025 |
hole | -0.025 |
ice | -0.024 |
father | -0.024 |
Token | she |
Token ID | 673 |
Feature activation | +1.32e-03 |
Pos logit contributions | |
Sad | +0.004 |
ilda | +0.003 |
hole | +0.003 |
James | +0.003 |
uster | +0.003 |
Neg logit contributions | |
bye | -0.002 |
Sorry | -0.002 |
ho | -0.002 |
good | -0.002 |
ge | -0.002 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
Sorry | +0.002 |
bye | +0.002 |
good | +0.002 |
they | +0.002 |
bye | +0.002 |
Neg logit contributions | |
hole | -0.002 |
uster | -0.002 |
Sad | -0.002 |
ice | -0.002 |
father | -0.002 |
Token | something |
Token ID | 1223 |
Feature activation | -4.42e-03 |
Pos logit contributions | |
Sorry | +0.017 |
bye | +0.015 |
ho | +0.013 |
good | +0.013 |
bye | +0.012 |
Neg logit contributions | |
Sad | -0.029 |
hole | -0.028 |
uster | -0.028 |
father | -0.028 |
pled | -0.027 |
Token | she |
Token ID | 673 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
hole | +0.009 |
Sad | +0.009 |
eping | +0.008 |
pled | +0.008 |
uster | +0.008 |
Neg logit contributions | |
Sorry | -0.007 |
bye | -0.007 |
bye | -0.006 |
ho | -0.005 |
moms | -0.005 |
Token | hadn |
Token ID | 8020 |
Feature activation | -7.58e-02 |
Pos logit contributions | |
uster | +0.042 |
hole | +0.040 |
Sad | +0.040 |
ice | +0.038 |
ist | +0.038 |
Neg logit contributions | |
Sorry | -0.020 |
good | -0.020 |
bye | -0.019 |
they | -0.019 |
day | -0.018 |
Token | 't |
Token ID | 470 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
ist | +0.180 |
uster | +0.180 |
ice | +0.178 |
aron | +0.177 |
ivering | +0.174 |
Neg logit contributions | |
go | -0.088 |
they | -0.085 |
day | -0.076 |
good | -0.072 |
bye | -0.070 |
Token | seen |
Token ID | 1775 |
Feature activation | -6.43e-02 |
Pos logit contributions | |
uster | +0.058 |
Sad | +0.055 |
ice | +0.054 |
hole | +0.054 |
innie | +0.052 |
Neg logit contributions | |
good | -0.076 |
day | -0.074 |
they | -0.074 |
bye | -0.073 |
go | -0.071 |
Token | before |
Token ID | 878 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
uster | +0.164 |
Sad | +0.164 |
hole | +0.151 |
Steven | +0.150 |
innie | +0.149 |
Neg logit contributions | |
day | -0.066 |
good | -0.065 |
bye | -0.064 |
they | -0.059 |
Sorry | -0.057 |
Token | - |
Token ID | 532 |
Feature activation | -6.98e-03 |
Pos logit contributions | |
hole | +0.034 |
Sad | +0.032 |
uster | +0.032 |
ice | +0.031 |
ilda | +0.031 |
Neg logit contributions | |
Sorry | -0.026 |
bye | -0.024 |
good | -0.022 |
bye | -0.022 |
ho | -0.021 |
Token | a |
Token ID | 257 |
Feature activation | -7.93e-03 |
Pos logit contributions | |
Sad | +0.015 |
hole | +0.015 |
uster | +0.014 |
pled | +0.014 |
ice | +0.014 |
Neg logit contributions | |
Sorry | -0.010 |
bye | -0.010 |
ho | -0.008 |
bye | -0.008 |
good | -0.008 |
Token | when |
Token ID | 618 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
Sorry | +0.008 |
bye | +0.007 |
good | +0.006 |
bye | +0.006 |
they | +0.006 |
Neg logit contributions | |
Sad | -0.014 |
hole | -0.014 |
uster | -0.014 |
ice | -0.013 |
father | -0.013 |
Token | she |
Token ID | 673 |
Feature activation | -3.93e-03 |
Pos logit contributions | |
Sorry | +0.018 |
bye | +0.018 |
good | +0.016 |
bye | +0.016 |
vows | +0.015 |
Neg logit contributions | |
Sad | -0.025 |
hole | -0.023 |
father | -0.023 |
James | -0.023 |
innie | -0.023 |
Token | spotted |
Token ID | 13489 |
Feature activation | -4.24e-04 |
Pos logit contributions | |
uster | +0.007 |
Sad | +0.007 |
hole | +0.007 |
ice | +0.007 |
father | +0.006 |
Neg logit contributions | |
Sorry | -0.007 |
bye | -0.007 |
good | -0.007 |
they | -0.006 |
bye | -0.006 |
Token | something |
Token ID | 1223 |
Feature activation | -8.11e-03 |
Pos logit contributions | |
Sad | +0.001 |
uster | +0.001 |
hole | +0.001 |
ice | +0.001 |
father | +0.001 |
Neg logit contributions | |
Sorry | -0.001 |
bye | -0.001 |
good | -0.000 |
they | -0.000 |
bye | -0.000 |
Token | she |
Token ID | 673 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
hole | +0.017 |
Sad | +0.016 |
uster | +0.016 |
pled | +0.016 |
eping | +0.016 |
Neg logit contributions | |
Sorry | -0.013 |
bye | -0.013 |
bye | -0.011 |
good | -0.009 |
moms | -0.009 |
Token | hadn |
Token ID | 8020 |
Feature activation | -7.41e-02 |
Pos logit contributions | |
uster | +0.055 |
Sad | +0.053 |
hole | +0.052 |
ist | +0.050 |
father | +0.049 |
Neg logit contributions | |
good | -0.025 |
Sorry | -0.025 |
bye | -0.024 |
they | -0.024 |
day | -0.024 |
Token | 't |
Token ID | 470 |
Feature activation | -3.63e-02 |
Pos logit contributions | |
uster | +0.175 |
ist | +0.174 |
aron | +0.169 |
ice | +0.169 |
Sad | +0.168 |
Neg logit contributions | |
go | -0.087 |
they | -0.084 |
day | -0.076 |
good | -0.073 |
bye | -0.070 |
Token | seen |
Token ID | 1775 |
Feature activation | -6.30e-02 |
Pos logit contributions | |
uster | +0.052 |
Sad | +0.049 |
ice | +0.048 |
hole | +0.047 |
innie | +0.046 |
Neg logit contributions | |
good | -0.081 |
day | -0.078 |
they | -0.078 |
go | -0.077 |
bye | -0.076 |
Token | before |
Token ID | 878 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
uster | +0.161 |
Sad | +0.161 |
hole | +0.150 |
Steven | +0.148 |
father | +0.147 |
Neg logit contributions | |
day | -0.063 |
good | -0.063 |
bye | -0.060 |
they | -0.056 |
well | -0.055 |
Token | . |
Token ID | 13 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
hole | +0.032 |
Sad | +0.031 |
uster | +0.030 |
ice | +0.030 |
ilda | +0.030 |
Neg logit contributions | |
Sorry | -0.027 |
bye | -0.025 |
good | -0.022 |
bye | -0.022 |
moms | -0.021 |
Token | It |
Token ID | 632 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
uster | +0.028 |
Sad | +0.028 |
hole | +0.028 |
ice | +0.026 |
father | +0.026 |
Neg logit contributions | |
Sorry | -0.027 |
bye | -0.026 |
good | -0.026 |
they | -0.025 |
day | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | -7.41e-04 |
Pos logit contributions | |
hole | +0.001 |
ist | +0.001 |
Sad | +0.001 |
uster | +0.001 |
ption | +0.001 |
Neg logit contributions | |
Sorry | -0.001 |
good | -0.001 |
bye | -0.001 |
vows | -0.001 |
bye | -0.001 |
Token | he |
Token ID | 339 |
Feature activation | +2.91e-04 |
Pos logit contributions | |
hole | +0.001 |
ilda | +0.001 |
Sad | +0.001 |
James | +0.001 |
uster | +0.001 |
Neg logit contributions | |
Sorry | -0.002 |
good | -0.001 |
bye | -0.001 |
birthday | -0.001 |
ho | -0.001 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
Sorry | +0.001 |
bye | +0.001 |
good | +0.000 |
bye | +0.000 |
they | +0.000 |
Neg logit contributions | |
hole | -0.000 |
Sad | -0.000 |
ice | -0.000 |
uster | -0.000 |
ilda | -0.000 |
Token | something |
Token ID | 1223 |
Feature activation | -6.57e-03 |
Pos logit contributions | |
Sorry | +0.019 |
bye | +0.017 |
ho | +0.015 |
good | +0.014 |
bye | +0.014 |
Neg logit contributions | |
Sad | -0.026 |
pled | -0.025 |
eping | -0.025 |
hole | -0.025 |
father | -0.025 |
Token | he |
Token ID | 339 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
hole | +0.013 |
eping | +0.013 |
pled | +0.013 |
Sad | +0.013 |
uster | +0.012 |
Neg logit contributions | |
Sorry | -0.011 |
bye | -0.011 |
bye | -0.009 |
moms | -0.008 |
ho | -0.008 |
Token | hadn |
Token ID | 8020 |
Feature activation | -7.26e-02 |
Pos logit contributions | |
uster | +0.070 |
Sad | +0.067 |
hole | +0.067 |
ice | +0.063 |
father | +0.063 |
Neg logit contributions | |
Sorry | -0.035 |
good | -0.035 |
bye | -0.034 |
they | -0.033 |
day | -0.033 |
Token | 't |
Token ID | 470 |
Feature activation | -3.45e-02 |
Pos logit contributions | |
uster | +0.179 |
ist | +0.171 |
ice | +0.171 |
aron | +0.169 |
Sad | +0.166 |
Neg logit contributions | |
go | -0.082 |
they | -0.080 |
day | -0.075 |
bye | -0.071 |
good | -0.067 |
Token | seen |
Token ID | 1775 |
Feature activation | -6.52e-02 |
Pos logit contributions | |
uster | +0.043 |
Sad | +0.041 |
hole | +0.039 |
ice | +0.038 |
innie | +0.036 |
Neg logit contributions | |
good | -0.081 |
bye | -0.080 |
Sorry | -0.079 |
day | -0.079 |
they | -0.079 |
Token | before |
Token ID | 878 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
Sad | +0.161 |
uster | +0.161 |
hole | +0.151 |
father | +0.147 |
innie | +0.147 |
Neg logit contributions | |
bye | -0.070 |
good | -0.070 |
day | -0.069 |
Sorry | -0.067 |
they | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | -5.35e-03 |
Pos logit contributions | |
hole | +0.024 |
uster | +0.023 |
Sad | +0.022 |
ice | +0.022 |
father | +0.022 |
Neg logit contributions | |
Sorry | -0.020 |
bye | -0.019 |
good | -0.017 |
bye | -0.017 |
ho | -0.016 |
Token | He |
Token ID | 679 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
uster | +0.010 |
Sad | +0.010 |
hole | +0.010 |
ice | +0.009 |
father | +0.009 |
Neg logit contributions | |
Sorry | -0.010 |
bye | -0.009 |
good | -0.009 |
they | -0.009 |
day | -0.009 |
Token | started |
Token ID | 2067 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
Sorry | +0.006 |
bye | +0.005 |
good | +0.005 |
bye | +0.005 |
day | +0.005 |
Neg logit contributions | |
hole | -0.004 |
Sad | -0.004 |
ice | -0.004 |
uster | -0.004 |
eping | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
Sorry | +0.043 |
bye | +0.040 |
bye | +0.033 |
good | +0.032 |
moms | +0.030 |
Neg logit contributions | |
Sad | -0.049 |
hole | -0.049 |
eping | -0.048 |
uster | -0.047 |
run | -0.047 |
Token | miss |
Token ID | 2051 |
Feature activation | +5.55e-03 |
Pos logit contributions | |
Sorry | +0.038 |
bye | +0.038 |
bye | +0.033 |
good | +0.032 |
they | +0.032 |
Neg logit contributions | |
hole | -0.038 |
Sad | -0.037 |
eping | -0.035 |
ice | -0.035 |
uster | -0.035 |
Token | something |
Token ID | 1223 |
Feature activation | +5.71e-03 |
Pos logit contributions | |
Sorry | +0.010 |
bye | +0.009 |
ho | +0.008 |
bye | +0.007 |
moms | +0.007 |
Neg logit contributions | |
hole | -0.010 |
uster | -0.010 |
Sad | -0.010 |
father | -0.010 |
ice | -0.010 |
Token | he |
Token ID | 339 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
Sorry | +0.011 |
bye | +0.010 |
bye | +0.009 |
moms | +0.008 |
ho | +0.008 |
Neg logit contributions | |
hole | -0.010 |
pled | -0.009 |
run | -0.009 |
uster | -0.009 |
ice | -0.009 |
Token | hadn |
Token ID | 8020 |
Feature activation | -7.23e-02 |
Pos logit contributions | |
uster | +0.063 |
Sad | +0.062 |
hole | +0.062 |
ice | +0.059 |
father | +0.058 |
Neg logit contributions | |
Sorry | -0.039 |
bye | -0.037 |
good | -0.036 |
they | -0.035 |
day | -0.034 |
Token | 't |
Token ID | 470 |
Feature activation | -3.07e-02 |
Pos logit contributions | |
uster | +0.165 |
ist | +0.164 |
Sad | +0.163 |
ice | +0.160 |
eping | +0.159 |
Neg logit contributions | |
go | -0.089 |
they | -0.088 |
day | -0.082 |
good | -0.078 |
bye | -0.078 |
Token | seen |
Token ID | 1775 |
Feature activation | -6.41e-02 |
Pos logit contributions | |
uster | +0.029 |
hole | +0.029 |
Sad | +0.029 |
ice | +0.026 |
father | +0.025 |
Neg logit contributions | |
Sorry | -0.082 |
bye | -0.079 |
good | -0.077 |
they | -0.076 |
bye | -0.074 |
Token | for |
Token ID | 329 |
Feature activation | +5.17e-03 |
Pos logit contributions | |
uster | +0.154 |
Sad | +0.154 |
hole | +0.147 |
father | +0.141 |
ice | +0.141 |
Neg logit contributions | |
good | -0.073 |
bye | -0.072 |
Sorry | -0.072 |
day | -0.070 |
they | -0.068 |
Token | a |
Token ID | 257 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
Sorry | +0.008 |
bye | +0.007 |
good | +0.007 |
they | +0.007 |
bye | +0.007 |
Neg logit contributions | |
uster | -0.011 |
hole | -0.011 |
Sad | -0.011 |
ice | -0.010 |
father | -0.010 |
Token | while |
Token ID | 981 |
Feature activation | -4.64e-03 |
Pos logit contributions | |
Sorry | +0.019 |
bye | +0.017 |
moms | +0.015 |
floats | +0.015 |
they | +0.014 |
Neg logit contributions | |
uster | -0.026 |
Sad | -0.024 |
x | -0.024 |
run | -0.023 |
hole | -0.023 |
Token | of |
Token ID | 286 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
uster | +0.020 |
Sad | +0.020 |
hole | +0.020 |
ice | +0.019 |
father | +0.019 |
Neg logit contributions | |
Sorry | -0.017 |
bye | -0.017 |
good | -0.016 |
they | -0.016 |
day | -0.015 |
Token | her |
Token ID | 607 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
uster | +0.041 |
hole | +0.041 |
Sad | +0.040 |
ist | +0.040 |
ice | +0.039 |
Neg logit contributions | |
good | -0.010 |
they | -0.010 |
Sorry | -0.010 |
day | -0.009 |
bye | -0.008 |
Token | garden |
Token ID | 11376 |
Feature activation | -3.77e-03 |
Pos logit contributions | |
uster | +0.047 |
Sad | +0.047 |
hole | +0.046 |
ice | +0.044 |
father | +0.044 |
Neg logit contributions | |
Sorry | -0.033 |
good | -0.032 |
bye | -0.032 |
they | -0.031 |
day | -0.030 |
Token | that |
Token ID | 326 |
Feature activation | +2.81e-03 |
Pos logit contributions | |
hole | +0.008 |
Sad | +0.007 |
ice | +0.007 |
pled | +0.007 |
uster | +0.007 |
Neg logit contributions | |
Sorry | -0.006 |
bye | -0.006 |
good | -0.005 |
bye | -0.005 |
they | -0.005 |
Token | she |
Token ID | 673 |
Feature activation | -3.38e-03 |
Pos logit contributions | |
Sorry | +0.005 |
bye | +0.004 |
seals | +0.004 |
bye | +0.004 |
Prize | +0.004 |
Neg logit contributions | |
hole | -0.005 |
Sad | -0.005 |
pled | -0.005 |
father | -0.005 |
ilda | -0.005 |
Token | hadn |
Token ID | 8020 |
Feature activation | -7.16e-02 |
Pos logit contributions | |
Sad | +0.008 |
uster | +0.008 |
hole | +0.007 |
ice | +0.007 |
father | +0.007 |
Neg logit contributions | |
Sorry | -0.005 |
bye | -0.004 |
good | -0.004 |
they | -0.004 |
bye | -0.004 |
Token | 't |
Token ID | 470 |
Feature activation | -3.10e-02 |
Pos logit contributions | |
uster | +0.168 |
ist | +0.166 |
ivering | +0.162 |
aron | +0.161 |
eping | +0.161 |
Neg logit contributions | |
go | -0.085 |
they | -0.083 |
day | -0.077 |
good | -0.072 |
bye | -0.069 |
Token | seen |
Token ID | 1775 |
Feature activation | -5.61e-02 |
Pos logit contributions | |
uster | +0.040 |
Sad | +0.039 |
hole | +0.039 |
ice | +0.036 |
father | +0.035 |
Neg logit contributions | |
Sorry | -0.071 |
bye | -0.069 |
good | -0.069 |
they | -0.067 |
day | -0.066 |
Token | before |
Token ID | 878 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
uster | +0.140 |
Sad | +0.138 |
hole | +0.132 |
father | +0.129 |
Steven | +0.128 |
Neg logit contributions | |
good | -0.060 |
day | -0.057 |
bye | -0.055 |
they | -0.054 |
Sorry | -0.054 |
Token | - |
Token ID | 532 |
Feature activation | -9.00e-03 |
Pos logit contributions | |
hole | +0.029 |
Sad | +0.028 |
uster | +0.027 |
ice | +0.027 |
ilda | +0.027 |
Neg logit contributions | |
Sorry | -0.025 |
bye | -0.023 |
bye | -0.021 |
good | -0.021 |
moms | -0.019 |
Token | a |
Token ID | 257 |
Feature activation | -9.74e-03 |
Pos logit contributions | |
hole | +0.019 |
Sad | +0.019 |
uster | +0.018 |
pled | +0.018 |
ice | +0.018 |
Neg logit contributions | |
Sorry | -0.013 |
bye | -0.013 |
good | -0.010 |
bye | -0.010 |
ho | -0.010 |
Token | places |
Token ID | 4113 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
uster | +0.006 |
hole | +0.006 |
Sad | +0.006 |
ice | +0.006 |
father | +0.006 |
Neg logit contributions | |
Sorry | -0.003 |
bye | -0.003 |
good | -0.003 |
they | -0.003 |
day | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
uster | +0.058 |
Sad | +0.057 |
hole | +0.057 |
ice | +0.055 |
father | +0.054 |
Neg logit contributions | |
Sorry | -0.028 |
bye | -0.027 |
good | -0.027 |
they | -0.026 |
day | -0.024 |
Token | saw |
Token ID | 2497 |
Feature activation | -2.14e-03 |
Pos logit contributions | |
hole | +0.019 |
Sad | +0.019 |
ilda | +0.019 |
father | +0.018 |
uster | +0.018 |
Neg logit contributions | |
Sorry | -0.018 |
bye | -0.015 |
good | -0.015 |
moms | -0.014 |
bye | -0.014 |
Token | things |
Token ID | 1243 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
hole | +0.005 |
pled | +0.005 |
father | +0.005 |
Sad | +0.005 |
uster | +0.005 |
Neg logit contributions | |
Sorry | -0.003 |
bye | -0.003 |
bye | -0.002 |
ho | -0.002 |
they | -0.002 |
Token | she |
Token ID | 673 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
hole | +0.027 |
ist | +0.025 |
uster | +0.025 |
pled | +0.025 |
ilda | +0.024 |
Neg logit contributions | |
Sorry | -0.019 |
bye | -0.017 |
bye | -0.015 |
moms | -0.014 |
good | -0.014 |
Token | hadn |
Token ID | 8020 |
Feature activation | -7.09e-02 |
Pos logit contributions | |
uster | +0.074 |
hole | +0.074 |
Sad | +0.074 |
ice | +0.071 |
father | +0.071 |
Neg logit contributions | |
Sorry | -0.051 |
bye | -0.048 |
good | -0.046 |
they | -0.045 |
bye | -0.044 |
Token | 't |
Token ID | 470 |
Feature activation | -3.52e-02 |
Pos logit contributions | |
Sad | +0.148 |
uster | +0.147 |
ice | +0.142 |
ivering | +0.142 |
eping | +0.141 |
Neg logit contributions | |
they | -0.101 |
go | -0.095 |
bye | -0.094 |
good | -0.094 |
Sorry | -0.093 |
Token | seen |
Token ID | 1775 |
Feature activation | -6.39e-02 |
Pos logit contributions | |
uster | +0.054 |
hole | +0.053 |
Sad | +0.053 |
ice | +0.050 |
father | +0.049 |
Neg logit contributions | |
Sorry | -0.074 |
bye | -0.069 |
good | -0.067 |
they | -0.066 |
bye | -0.065 |
Token | before |
Token ID | 878 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
uster | +0.144 |
Sad | +0.143 |
hole | +0.142 |
ice | +0.136 |
father | +0.134 |
Neg logit contributions | |
Sorry | -0.085 |
bye | -0.080 |
good | -0.078 |
they | -0.075 |
day | -0.072 |
Token | . |
Token ID | 13 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
hole | +0.024 |
Sad | +0.023 |
uster | +0.023 |
ice | +0.023 |
ilda | +0.023 |
Neg logit contributions | |
Sorry | -0.020 |
bye | -0.019 |
bye | -0.017 |
good | -0.017 |
moms | -0.016 |
Token | Most |
Token ID | 4042 |
Feature activation | -4.56e-03 |
Pos logit contributions | |
uster | +0.022 |
Sad | +0.022 |
hole | +0.022 |
ice | +0.021 |
father | +0.020 |
Neg logit contributions | |
Sorry | -0.019 |
bye | -0.018 |
good | -0.017 |
they | -0.017 |
day | -0.016 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.21e-03 |
Pos logit contributions | |
Sorry | +0.022 |
bye | +0.019 |
ho | +0.019 |
moms | +0.018 |
bye | +0.018 |
Neg logit contributions | |
Sad | -0.024 |
hole | -0.023 |
eping | -0.023 |
father | -0.022 |
uster | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | -1.15e-03 |
Pos logit contributions | |
Sorry | +0.002 |
bye | +0.002 |
ho | +0.001 |
bye | +0.001 |
good | +0.001 |
Neg logit contributions | |
Sad | -0.003 |
hole | -0.003 |
uster | -0.003 |
father | -0.002 |
eping | -0.002 |
Token | explore |
Token ID | 7301 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
Sad | +0.002 |
hole | +0.002 |
father | +0.002 |
uster | +0.002 |
eping | +0.002 |
Neg logit contributions | |
Sorry | -0.002 |
bye | -0.002 |
ho | -0.002 |
good | -0.002 |
they | -0.002 |
Token | places |
Token ID | 4113 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
uster | +0.035 |
Sad | +0.034 |
hole | +0.034 |
ice | +0.033 |
innie | +0.033 |
Neg logit contributions | |
good | -0.009 |
they | -0.009 |
day | -0.008 |
Sorry | -0.008 |
bye | -0.008 |
Token | he |
Token ID | 339 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
uster | +0.055 |
Sad | +0.054 |
hole | +0.053 |
ice | +0.052 |
father | +0.051 |
Neg logit contributions | |
Sorry | -0.025 |
bye | -0.023 |
good | -0.023 |
they | -0.023 |
day | -0.021 |
Token | hadn |
Token ID | 8020 |
Feature activation | -6.96e-02 |
Pos logit contributions | |
uster | +0.053 |
hole | +0.053 |
Sad | +0.053 |
ice | +0.050 |
father | +0.050 |
Neg logit contributions | |
Sorry | -0.033 |
bye | -0.031 |
good | -0.030 |
they | -0.029 |
bye | -0.027 |
Token | 't |
Token ID | 470 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
uster | +0.153 |
Sad | +0.151 |
ice | +0.149 |
ist | +0.148 |
ivering | +0.148 |
Neg logit contributions | |
go | -0.090 |
they | -0.088 |
bye | -0.085 |
day | -0.085 |
good | -0.082 |
Token | been |
Token ID | 587 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
uster | +0.030 |
hole | +0.030 |
Sad | +0.030 |
ice | +0.027 |
father | +0.026 |
Neg logit contributions | |
Sorry | -0.081 |
bye | -0.077 |
good | -0.076 |
they | -0.075 |
bye | -0.073 |
Token | before |
Token ID | 878 |
Feature activation | -6.68e-03 |
Pos logit contributions | |
hole | +0.044 |
uster | +0.044 |
Sad | +0.042 |
ice | +0.041 |
father | +0.040 |
Neg logit contributions | |
Sorry | -0.048 |
bye | -0.045 |
good | -0.042 |
they | -0.042 |
bye | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
hole | +0.014 |
uster | +0.013 |
Sad | +0.013 |
father | +0.013 |
ice | +0.013 |
Neg logit contributions | |
Sorry | -0.011 |
bye | -0.010 |
good | -0.009 |
bye | -0.009 |
ho | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
Sorry | +0.014 |
bye | +0.013 |
day | +0.013 |
good | +0.013 |
they | +0.013 |
Neg logit contributions | |
uster | -0.023 |
Sad | -0.022 |
hole | -0.022 |
eping | -0.022 |
ist | -0.022 |
Token | all |
Token ID | 477 |
Feature activation | -5.62e-03 |
Pos logit contributions | |
Sorry | +0.028 |
bye | +0.027 |
day | +0.021 |
good | +0.021 |
bye | +0.020 |
Neg logit contributions | |
uster | -0.049 |
Sad | -0.049 |
hole | -0.049 |
pled | -0.047 |
ice | -0.047 |
Token | sorts |
Token ID | 10524 |
Feature activation | +3.09e-03 |
Pos logit contributions | |
uster | +0.014 |
hole | +0.014 |
Sad | +0.014 |
ice | +0.013 |
father | +0.013 |
Neg logit contributions | |
Sorry | -0.007 |
bye | -0.006 |
good | -0.006 |
they | -0.005 |
bye | -0.005 |
Token | of |
Token ID | 286 |
Feature activation | +7.40e-03 |
Pos logit contributions | |
Sorry | +0.005 |
bye | +0.005 |
good | +0.005 |
they | +0.005 |
day | +0.005 |
Neg logit contributions | |
uster | -0.006 |
hole | -0.006 |
Sad | -0.006 |
ice | -0.005 |
ist | -0.005 |
Token | things |
Token ID | 1243 |
Feature activation | -8.64e-03 |
Pos logit contributions | |
Sorry | +0.009 |
bye | +0.008 |
good | +0.008 |
they | +0.008 |
day | +0.007 |
Neg logit contributions | |
uster | -0.018 |
Sad | -0.018 |
hole | -0.018 |
ice | -0.017 |
father | -0.017 |
Token | they |
Token ID | 484 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
hole | +0.018 |
ist | +0.017 |
uster | +0.017 |
pled | +0.016 |
ilda | +0.016 |
Neg logit contributions | |
Sorry | -0.015 |
bye | -0.014 |
bye | -0.012 |
good | -0.011 |
moms | -0.011 |
Token | hadn |
Token ID | 8020 |
Feature activation | -6.95e-02 |
Pos logit contributions | |
hole | +0.046 |
uster | +0.045 |
Sad | +0.044 |
father | +0.044 |
ice | +0.043 |
Neg logit contributions | |
Sorry | -0.039 |
bye | -0.036 |
good | -0.035 |
bye | -0.034 |
day | -0.031 |
Token | 't |
Token ID | 470 |
Feature activation | -3.10e-02 |
Pos logit contributions | |
Sad | +0.138 |
uster | +0.136 |
ice | +0.131 |
hole | +0.131 |
ivering | +0.129 |
Neg logit contributions | |
they | -0.106 |
bye | -0.102 |
Sorry | -0.102 |
good | -0.100 |
go | -0.099 |
Token | seen |
Token ID | 1775 |
Feature activation | -5.71e-02 |
Pos logit contributions | |
hole | +0.032 |
uster | +0.031 |
Sad | +0.030 |
eping | +0.028 |
ice | +0.028 |
Neg logit contributions | |
Sorry | -0.083 |
bye | -0.076 |
good | -0.074 |
bye | -0.073 |
they | -0.072 |
Token | before |
Token ID | 878 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
hole | +0.124 |
uster | +0.122 |
ice | +0.119 |
Sad | +0.119 |
ist | +0.116 |
Neg logit contributions | |
Sorry | -0.085 |
bye | -0.077 |
good | -0.073 |
they | -0.071 |
bye | -0.069 |
Token | . |
Token ID | 13 |
Feature activation | +9.19e-03 |
Pos logit contributions | |
hole | +0.025 |
uster | +0.023 |
Sad | +0.023 |
ice | +0.023 |
ilda | +0.023 |
Neg logit contributions | |
Sorry | -0.021 |
bye | -0.019 |
good | -0.018 |
bye | -0.017 |
ho | -0.017 |
Token | Together |
Token ID | 17083 |
Feature activation | +2.58e-02 |
Pos logit contributions | |
Sorry | +0.013 |
bye | +0.013 |
good | +0.012 |
they | +0.012 |
day | +0.012 |
Neg logit contributions | |
uster | -0.020 |
Sad | -0.020 |
hole | -0.019 |
ice | -0.019 |
father | -0.018 |
Token | saw |
Token ID | 2497 |
Feature activation | -7.28e-04 |
Pos logit contributions | |
Sad | +0.026 |
hole | +0.026 |
ice | +0.025 |
father | +0.025 |
uster | +0.024 |
Neg logit contributions | |
Sorry | -0.025 |
bye | -0.023 |
good | -0.021 |
day | -0.020 |
bye | -0.020 |
Token | things |
Token ID | 1243 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
pled | +0.002 |
ice | +0.001 |
father | +0.001 |
eping | +0.001 |
hole | +0.001 |
Neg logit contributions | |
Sorry | -0.001 |
bye | -0.001 |
ho | -0.001 |
day | -0.001 |
good | -0.001 |
Token | he |
Token ID | 339 |
Feature activation | -3.95e-02 |
Pos logit contributions | |
hole | +0.028 |
pled | +0.026 |
uster | +0.025 |
ist | +0.025 |
eping | +0.025 |
Neg logit contributions | |
Sorry | -0.022 |
bye | -0.021 |
bye | -0.017 |
ho | -0.016 |
good | -0.016 |
Token | had |
Token ID | 550 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
hole | +0.082 |
Sad | +0.080 |
uster | +0.080 |
ice | +0.078 |
father | +0.078 |
Neg logit contributions | |
Sorry | -0.060 |
bye | -0.056 |
good | -0.053 |
bye | -0.051 |
they | -0.050 |
Token | never |
Token ID | 1239 |
Feature activation | -5.03e-02 |
Pos logit contributions | |
hole | +0.051 |
ist | +0.047 |
father | +0.047 |
eping | +0.047 |
Sad | +0.047 |
Neg logit contributions | |
Sorry | -0.054 |
bye | -0.047 |
bye | -0.045 |
good | -0.043 |
ho | -0.042 |
Token | seen |
Token ID | 1775 |
Feature activation | -6.92e-02 |
Pos logit contributions | |
uster | +0.052 |
Sad | +0.050 |
hole | +0.049 |
ice | +0.045 |
father | +0.043 |
Neg logit contributions | |
Sorry | -0.128 |
bye | -0.126 |
good | -0.125 |
they | -0.122 |
day | -0.121 |
Token | before |
Token ID | 878 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
uster | +0.151 |
hole | +0.150 |
Sad | +0.149 |
ice | +0.144 |
father | +0.141 |
Neg logit contributions | |
Sorry | -0.098 |
bye | -0.091 |
good | -0.089 |
they | -0.087 |
day | -0.082 |
Token | . |
Token ID | 13 |
Feature activation | -1.45e-03 |
Pos logit contributions | |
hole | +0.037 |
uster | +0.037 |
Sad | +0.036 |
ice | +0.035 |
father | +0.035 |
Neg logit contributions | |
Sorry | -0.026 |
bye | -0.025 |
good | -0.023 |
bye | -0.022 |
ho | -0.021 |
Token | He |
Token ID | 679 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
hole | +0.003 |
uster | +0.003 |
Sad | +0.003 |
ist | +0.003 |
ice | +0.003 |
Neg logit contributions | |
Sorry | -0.002 |
they | -0.002 |
good | -0.002 |
bye | -0.002 |
day | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
Sad | +0.023 |
hole | +0.023 |
ice | +0.023 |
father | +0.022 |
uster | +0.022 |
Neg logit contributions | |
Sorry | -0.023 |
bye | -0.021 |
good | -0.020 |
day | -0.019 |
bye | -0.018 |
Token | amazed |
Token ID | 24562 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
hole | +0.021 |
Sad | +0.021 |
uster | +0.021 |
eping | +0.020 |
ady | +0.020 |
Neg logit contributions | |
Sorry | -0.018 |
bye | -0.017 |
ho | -0.015 |
bye | -0.014 |
they | -0.014 |
Token | able |
Token ID | 1498 |
Feature activation | -2.98e-03 |
Pos logit contributions | |
Sad | +0.029 |
uster | +0.028 |
hole | +0.028 |
ice | +0.027 |
eping | +0.027 |
Neg logit contributions | |
Sorry | -0.034 |
bye | -0.032 |
ho | -0.029 |
they | -0.029 |
bye | -0.028 |
Token | to |
Token ID | 284 |
Feature activation | -2.48e-03 |
Pos logit contributions | |
Sad | +0.007 |
hole | +0.006 |
uster | +0.006 |
pled | +0.006 |
ice | +0.006 |
Neg logit contributions | |
Sorry | -0.004 |
bye | -0.004 |
bye | -0.003 |
ho | -0.003 |
good | -0.003 |
Token | do |
Token ID | 466 |
Feature activation | -5.97e-03 |
Pos logit contributions | |
uster | +0.006 |
Sad | +0.006 |
hole | +0.006 |
ice | +0.006 |
father | +0.005 |
Neg logit contributions | |
Sorry | -0.003 |
bye | -0.003 |
good | -0.003 |
they | -0.003 |
day | -0.003 |
Token | things |
Token ID | 1243 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
Sad | +0.014 |
hole | +0.014 |
uster | +0.014 |
ice | +0.013 |
ilda | +0.013 |
Neg logit contributions | |
Sorry | -0.008 |
bye | -0.008 |
ho | -0.006 |
bye | -0.006 |
good | -0.006 |
Token | she |
Token ID | 673 |
Feature activation | -3.53e-02 |
Pos logit contributions | |
hole | +0.048 |
uster | +0.048 |
Sad | +0.047 |
ist | +0.047 |
ilda | +0.045 |
Neg logit contributions | |
Sorry | -0.032 |
bye | -0.029 |
bye | -0.026 |
good | -0.025 |
day | -0.024 |
Token | hadn |
Token ID | 8020 |
Feature activation | -6.87e-02 |
Pos logit contributions | |
uster | +0.079 |
Sad | +0.078 |
hole | +0.078 |
ice | +0.075 |
father | +0.074 |
Neg logit contributions | |
Sorry | -0.048 |
bye | -0.045 |
good | -0.044 |
they | -0.043 |
day | -0.041 |
Token | 't |
Token ID | 470 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
uster | +0.148 |
Sad | +0.147 |
hole | +0.143 |
ist | +0.142 |
eping | +0.142 |
Neg logit contributions | |
good | -0.090 |
they | -0.089 |
go | -0.089 |
day | -0.087 |
bye | -0.086 |
Token | done |
Token ID | 1760 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
uster | +0.048 |
Sad | +0.048 |
hole | +0.048 |
ice | +0.046 |
pled | +0.045 |
Neg logit contributions | |
Sorry | -0.050 |
bye | -0.045 |
good | -0.043 |
they | -0.043 |
ho | -0.042 |
Token | before |
Token ID | 878 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
uster | +0.030 |
Sad | +0.030 |
hole | +0.030 |
ice | +0.030 |
ilda | +0.029 |
Neg logit contributions | |
Sorry | -0.022 |
bye | -0.021 |
bye | -0.018 |
good | -0.018 |
they | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
hole | +0.029 |
Sad | +0.029 |
uster | +0.028 |
ice | +0.028 |
ilda | +0.028 |
Neg logit contributions | |
Sorry | -0.022 |
bye | -0.021 |
good | -0.019 |
bye | -0.019 |
ho | -0.018 |
Token | With |
Token ID | 2080 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
uster | +0.035 |
Sad | +0.035 |
hole | +0.035 |
ice | +0.033 |
father | +0.033 |
Neg logit contributions | |
Sorry | -0.028 |
bye | -0.027 |
good | -0.027 |
they | -0.026 |
day | -0.025 |
Token | her |
Token ID | 607 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
Sorry | +0.022 |
bye | +0.021 |
ho | +0.019 |
day | +0.017 |
ge | +0.017 |
Neg logit contributions | |
Sad | -0.026 |
ilda | -0.026 |
father | -0.026 |
uster | -0.024 |
ist | -0.024 |
Token | friends |
Token ID | 2460 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
bye | +0.035 |
Sorry | +0.034 |
ho | +0.031 |
Qu | +0.025 |
ww | +0.025 |
Neg logit contributions | |
Sad | -0.040 |
father | -0.038 |
ilda | -0.037 |
innie | -0.036 |
pled | -0.036 |
Token | what |
Token ID | 644 |
Feature activation | +9.00e-03 |
Pos logit contributions | |
Sorry | +0.033 |
bye | +0.031 |
ho | +0.028 |
moms | +0.026 |
good | +0.026 |
Neg logit contributions | |
hole | -0.020 |
ilda | -0.019 |
uster | -0.019 |
Sad | -0.019 |
ist | -0.019 |
Token | she |
Token ID | 673 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
Sorry | +0.013 |
bye | +0.013 |
bye | +0.011 |
ho | +0.010 |
good | +0.009 |
Neg logit contributions | |
Sad | -0.020 |
Sarah | -0.018 |
father | -0.018 |
Kelly | -0.018 |
ilda | -0.018 |
Token | had |
Token ID | 550 |
Feature activation | -3.18e-02 |
Pos logit contributions | |
uster | +0.099 |
Sad | +0.096 |
hole | +0.094 |
ice | +0.090 |
ist | +0.089 |
Neg logit contributions | |
Sorry | -0.047 |
bye | -0.047 |
good | -0.046 |
day | -0.044 |
they | -0.044 |
Token | seen |
Token ID | 1775 |
Feature activation | -6.82e-02 |
Pos logit contributions | |
uster | +0.061 |
Sad | +0.060 |
hole | +0.060 |
ice | +0.057 |
father | +0.056 |
Neg logit contributions | |
Sorry | -0.054 |
bye | -0.051 |
good | -0.051 |
they | -0.049 |
day | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | -7.57e-03 |
Pos logit contributions | |
hole | +0.144 |
uster | +0.140 |
Sad | +0.139 |
ice | +0.137 |
father | +0.135 |
Neg logit contributions | |
Sorry | -0.104 |
bye | -0.096 |
good | -0.092 |
bye | -0.087 |
they | -0.087 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.71e-03 |
Pos logit contributions | |
uster | +0.017 |
hole | +0.017 |
Sad | +0.017 |
ice | +0.017 |
ist | +0.016 |
Neg logit contributions | |
Sorry | -0.010 |
bye | -0.009 |
they | -0.009 |
good | -0.009 |
day | -0.009 |
Token | Once |
Token ID | 7454 |
Feature activation | -7.95e-03 |
Pos logit contributions | |
uster | +0.010 |
Sad | +0.010 |
hole | +0.010 |
ice | +0.009 |
father | +0.009 |
Neg logit contributions | |
Sorry | -0.007 |
bye | -0.007 |
good | -0.007 |
they | -0.006 |
day | -0.006 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.37e-03 |
Pos logit contributions | |
uster | +0.015 |
Sad | +0.015 |
ice | +0.014 |
eping | +0.014 |
ist | +0.014 |
Neg logit contributions | |
they | -0.015 |
day | -0.014 |
good | -0.014 |
Sorry | -0.013 |
go | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
good | +0.001 |
Sorry | +0.001 |
they | +0.001 |
bye | +0.001 |
day | +0.001 |
Neg logit contributions | |
uster | -0.004 |
hole | -0.004 |
Sad | -0.004 |
ice | -0.004 |
father | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -3.37e-03 |
Pos logit contributions | |
Sorry | +0.001 |
bye | +0.001 |
ho | +0.001 |
good | +0.001 |
day | +0.001 |
Neg logit contributions | |
Sad | -0.002 |
uster | -0.002 |
hole | -0.002 |
ice | -0.002 |
father | -0.002 |
Token | he |
Token ID | 339 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
Sad | +0.007 |
uster | +0.007 |
father | +0.007 |
hole | +0.007 |
ice | +0.007 |
Neg logit contributions | |
bye | -0.005 |
Sorry | -0.005 |
ho | -0.004 |
good | -0.004 |
bye | -0.004 |
Token | saw |
Token ID | 2497 |
Feature activation | +2.27e-03 |
Pos logit contributions | |
hole | +0.016 |
uster | +0.016 |
Sad | +0.016 |
ice | +0.016 |
father | +0.015 |
Neg logit contributions | |
Sorry | -0.020 |
bye | -0.020 |
good | -0.019 |
they | -0.018 |
day | -0.018 |
Token | something |
Token ID | 1223 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
Sorry | +0.003 |
bye | +0.003 |
good | +0.002 |
day | +0.002 |
bye | +0.002 |
Neg logit contributions | |
Sad | -0.005 |
uster | -0.005 |
hole | -0.005 |
ice | -0.005 |
father | -0.005 |
Token | he |
Token ID | 339 |
Feature activation | -3.20e-02 |
Pos logit contributions | |
hole | +0.030 |
Sad | +0.030 |
eping | +0.029 |
uster | +0.029 |
pled | +0.029 |
Neg logit contributions | |
Sorry | -0.023 |
bye | -0.022 |
bye | -0.018 |
good | -0.017 |
day | -0.016 |
Token | hadn |
Token ID | 8020 |
Feature activation | -6.80e-02 |
Pos logit contributions | |
uster | +0.073 |
Sad | +0.071 |
hole | +0.070 |
ice | +0.068 |
father | +0.067 |
Neg logit contributions | |
Sorry | -0.042 |
good | -0.041 |
bye | -0.041 |
they | -0.040 |
day | -0.038 |
Token | 't |
Token ID | 470 |
Feature activation | -3.29e-02 |
Pos logit contributions | |
ist | +0.158 |
uster | +0.158 |
ice | +0.157 |
Sad | +0.153 |
aron | +0.152 |
Neg logit contributions | |
they | -0.082 |
go | -0.082 |
good | -0.072 |
day | -0.072 |
bye | -0.071 |
Token | expected |
Token ID | 2938 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
uster | +0.048 |
Sad | +0.046 |
ice | +0.045 |
hole | +0.044 |
innie | +0.044 |
Neg logit contributions | |
good | -0.072 |
they | -0.069 |
day | -0.069 |
bye | -0.069 |
go | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
uster | +0.072 |
Sad | +0.072 |
hole | +0.071 |
ice | +0.068 |
father | +0.067 |
Neg logit contributions | |
Sorry | -0.037 |
bye | -0.035 |
good | -0.034 |
they | -0.033 |
day | -0.032 |
Token | A |
Token ID | 317 |
Feature activation | -2.38e-03 |
Pos logit contributions | |
uster | +0.029 |
hole | +0.029 |
Sad | +0.029 |
ist | +0.028 |
father | +0.028 |
Neg logit contributions | |
Sorry | -0.023 |
they | -0.021 |
bye | -0.020 |
good | -0.020 |
day | -0.019 |
Token | hole |
Token ID | 7604 |
Feature activation | -8.34e-03 |
Pos logit contributions | |
Sad | +0.005 |
uster | +0.005 |
hole | +0.005 |
Steven | +0.005 |
eping | +0.005 |
Neg logit contributions | |
Sorry | -0.003 |
bye | -0.003 |
they | -0.003 |
bye | -0.003 |
day | -0.003 |
Token | he |
Token ID | 339 |
Feature activation | +2.93e-03 |
Pos logit contributions | |
Sad | +0.022 |
hole | +0.022 |
uster | +0.022 |
father | +0.021 |
ice | +0.021 |
Neg logit contributions | |
Sorry | -0.014 |
bye | -0.013 |
good | -0.012 |
bye | -0.012 |
ho | -0.011 |
Token | found |
Token ID | 1043 |
Feature activation | -3.80e-03 |
Pos logit contributions | |
Sorry | +0.005 |
bye | +0.005 |
good | +0.005 |
they | +0.005 |
day | +0.005 |
Neg logit contributions | |
hole | -0.005 |
uster | -0.005 |
Sad | -0.005 |
ice | -0.005 |
father | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
uster | +0.010 |
Sad | +0.010 |
hole | +0.010 |
ice | +0.010 |
ist | +0.010 |
Neg logit contributions | |
Sorry | -0.003 |
good | -0.003 |
bye | -0.003 |
they | -0.003 |
day | -0.003 |
Token | door |
Token ID | 3420 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
hole | +0.031 |
uster | +0.031 |
Sad | +0.031 |
innie | +0.029 |
ilda | +0.029 |
Neg logit contributions | |
bye | -0.020 |
Sorry | -0.019 |
they | -0.017 |
bye | -0.016 |
good | -0.015 |
Token | he |
Token ID | 339 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
uster | +0.054 |
Sad | +0.054 |
ice | +0.052 |
innie | +0.051 |
father | +0.050 |
Neg logit contributions | |
they | -0.020 |
bye | -0.019 |
day | -0.019 |
go | -0.019 |
Sorry | -0.019 |
Token | hadn |
Token ID | 8020 |
Feature activation | -6.80e-02 |
Pos logit contributions | |
uster | +0.058 |
Sad | +0.056 |
hole | +0.054 |
ice | +0.054 |
ivering | +0.053 |
Neg logit contributions | |
good | -0.025 |
bye | -0.025 |
Sorry | -0.025 |
they | -0.024 |
day | -0.023 |
Token | 't |
Token ID | 470 |
Feature activation | -3.03e-02 |
Pos logit contributions | |
uster | +0.157 |
ist | +0.156 |
ivering | +0.155 |
Sad | +0.154 |
eping | +0.152 |
Neg logit contributions | |
go | -0.084 |
they | -0.080 |
day | -0.076 |
good | -0.074 |
bye | -0.073 |
Token | seen |
Token ID | 1775 |
Feature activation | -4.88e-02 |
Pos logit contributions | |
uster | +0.042 |
Sad | +0.040 |
ice | +0.038 |
hole | +0.037 |
ivering | +0.037 |
Neg logit contributions | |
good | -0.068 |
bye | -0.066 |
they | -0.066 |
day | -0.066 |
Sorry | -0.064 |
Token | before |
Token ID | 878 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
uster | +0.121 |
Sad | +0.121 |
hole | +0.114 |
ist | +0.111 |
father | +0.110 |
Neg logit contributions | |
good | -0.053 |
bye | -0.051 |
day | -0.050 |
Sorry | -0.049 |
they | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | +4.68e-03 |
Pos logit contributions | |
hole | +0.029 |
uster | +0.029 |
Sad | +0.028 |
ice | +0.027 |
father | +0.027 |
Neg logit contributions | |
Sorry | -0.018 |
bye | -0.017 |
good | -0.016 |
they | -0.015 |
bye | -0.015 |
Token | He |
Token ID | 679 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
Sorry | +0.006 |
they | +0.006 |
good | +0.006 |
day | +0.005 |
bye | +0.005 |
Neg logit contributions | |
ist | -0.011 |
uster | -0.011 |
hole | -0.011 |
eping | -0.011 |
ice | -0.011 |
Token | something |
Token ID | 1223 |
Feature activation | +8.93e-03 |
Pos logit contributions | |
Sad | +0.023 |
hole | +0.023 |
uster | +0.022 |
ice | +0.022 |
father | +0.022 |
Neg logit contributions | |
Sorry | -0.013 |
bye | -0.013 |
ho | -0.011 |
bye | -0.011 |
day | -0.011 |
Token | unknown |
Token ID | 6439 |
Feature activation | -1.15e-04 |
Pos logit contributions | |
bye | +0.015 |
Sorry | +0.013 |
bye | +0.012 |
ho | +0.012 |
ge | +0.011 |
Neg logit contributions | |
hole | -0.018 |
ist | -0.017 |
Sad | -0.016 |
ilda | -0.016 |
pled | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
hole | +0.000 |
ist | +0.000 |
Sad | +0.000 |
father | +0.000 |
uster | +0.000 |
Neg logit contributions | |
Sorry | -0.000 |
bye | -0.000 |
bye | -0.000 |
moms | -0.000 |
ge | -0.000 |
Token | something |
Token ID | 1223 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
uster | +0.059 |
hole | +0.059 |
Sad | +0.059 |
ice | +0.055 |
ist | +0.054 |
Neg logit contributions | |
Sorry | -0.068 |
bye | -0.065 |
good | -0.063 |
they | -0.061 |
day | -0.060 |
Token | she |
Token ID | 673 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
bye | +0.020 |
Sorry | +0.019 |
bye | +0.017 |
ho | +0.015 |
moms | +0.014 |
Neg logit contributions | |
Sad | -0.021 |
pled | -0.021 |
ist | -0.021 |
hole | -0.021 |
ice | -0.020 |
Token | hadn |
Token ID | 8020 |
Feature activation | -6.75e-02 |
Pos logit contributions | |
uster | +0.047 |
Sad | +0.046 |
hole | +0.045 |
ice | +0.043 |
father | +0.043 |
Neg logit contributions | |
Sorry | -0.022 |
good | -0.021 |
bye | -0.020 |
they | -0.020 |
day | -0.019 |
Token | 't |
Token ID | 470 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
uster | +0.159 |
ivering | +0.155 |
Sad | +0.153 |
aron | +0.152 |
ist | +0.152 |
Neg logit contributions | |
they | -0.080 |
go | -0.076 |
day | -0.071 |
good | -0.070 |
bye | -0.066 |
Token | seen |
Token ID | 1775 |
Feature activation | -6.10e-02 |
Pos logit contributions | |
uster | +0.052 |
Sad | +0.051 |
hole | +0.051 |
ice | +0.048 |
father | +0.047 |
Neg logit contributions | |
Sorry | -0.058 |
bye | -0.056 |
good | -0.056 |
they | -0.054 |
day | -0.053 |
Token | before |
Token ID | 878 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
uster | +0.148 |
Sad | +0.148 |
hole | +0.141 |
father | +0.135 |
innie | +0.135 |
Neg logit contributions | |
good | -0.067 |
bye | -0.067 |
Sorry | -0.066 |
day | -0.066 |
they | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
hole | +0.024 |
ice | +0.022 |
Sad | +0.022 |
uster | +0.022 |
ilda | +0.022 |
Neg logit contributions | |
Sorry | -0.020 |
bye | -0.019 |
bye | -0.017 |
good | -0.017 |
moms | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
Sorry | +0.017 |
bye | +0.015 |
good | +0.015 |
they | +0.015 |
day | +0.014 |
Neg logit contributions | |
uster | -0.028 |
hole | -0.028 |
Sad | -0.027 |
ice | -0.027 |
ist | -0.026 |
Token | of |
Token ID | 286 |
Feature activation | +2.02e-04 |
Pos logit contributions | |
Sorry | +0.006 |
bye | +0.006 |
good | +0.006 |
bye | +0.005 |
day | +0.005 |
Neg logit contributions | |
uster | -0.006 |
hole | -0.005 |
Sad | -0.005 |
pled | -0.005 |
ice | -0.005 |
Token | things |
Token ID | 1243 |
Feature activation | -8.53e-03 |
Pos logit contributions | |
Sorry | +0.000 |
bye | +0.000 |
bye | +0.000 |
they | +0.000 |
ho | +0.000 |
Neg logit contributions | |
hole | -0.000 |
uster | -0.000 |
Sad | -0.000 |
ice | -0.000 |
pled | -0.000 |
Token | she |
Token ID | 673 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
hole | +0.018 |
pled | +0.017 |
ist | +0.017 |
uster | +0.016 |
ilda | +0.016 |
Neg logit contributions | |
Sorry | -0.014 |
bye | -0.014 |
bye | -0.012 |
moms | -0.011 |
good | -0.011 |
Token | had |
Token ID | 550 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
uster | +0.090 |
Sad | +0.090 |
hole | +0.090 |
ice | +0.086 |
father | +0.085 |
Neg logit contributions | |
Sorry | -0.061 |
bye | -0.057 |
good | -0.056 |
they | -0.053 |
day | -0.052 |
Token | never |
Token ID | 1239 |
Feature activation | -5.17e-02 |
Pos logit contributions | |
hole | +0.057 |
uster | +0.057 |
Sad | +0.057 |
ice | +0.054 |
father | +0.053 |
Neg logit contributions | |
Sorry | -0.056 |
bye | -0.052 |
good | -0.050 |
they | -0.049 |
bye | -0.048 |
Token | seen |
Token ID | 1775 |
Feature activation | -6.74e-02 |
Pos logit contributions | |
uster | +0.057 |
Sad | +0.055 |
hole | +0.053 |
ice | +0.049 |
innie | +0.047 |
Neg logit contributions | |
Sorry | -0.128 |
bye | -0.127 |
good | -0.126 |
they | -0.124 |
day | -0.123 |
Token | before |
Token ID | 878 |
Feature activation | -9.39e-03 |
Pos logit contributions | |
uster | +0.150 |
Sad | +0.148 |
hole | +0.148 |
ice | +0.142 |
father | +0.140 |
Neg logit contributions | |
Sorry | -0.091 |
bye | -0.086 |
good | -0.084 |
they | -0.081 |
day | -0.078 |
Token | like |
Token ID | 588 |
Feature activation | -4.68e-04 |
Pos logit contributions | |
hole | +0.018 |
uster | +0.017 |
Sad | +0.017 |
ice | +0.017 |
ilda | +0.017 |
Neg logit contributions | |
Sorry | -0.016 |
bye | -0.015 |
bye | -0.014 |
good | -0.014 |
moms | -0.013 |
Token | birds |
Token ID | 10087 |
Feature activation | -9.53e-03 |
Pos logit contributions | |
ice | +0.001 |
hole | +0.001 |
pled | +0.001 |
uster | +0.001 |
ilda | +0.001 |
Neg logit contributions | |
Sorry | -0.001 |
bye | -0.001 |
bye | -0.001 |
good | -0.001 |
ho | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | -2.38e-03 |
Pos logit contributions | |
hole | +0.018 |
pled | +0.018 |
ice | +0.018 |
ist | +0.017 |
uster | +0.017 |
Neg logit contributions | |
Sorry | -0.018 |
bye | -0.017 |
moms | -0.014 |
bye | -0.014 |
good | -0.014 |
Token | plants |
Token ID | 6134 |
Feature activation | -3.59e-03 |
Pos logit contributions | |
ice | +0.005 |
Sad | +0.005 |
pled | +0.005 |
hole | +0.005 |
uster | +0.005 |
Neg logit contributions | |
Sorry | -0.003 |
bye | -0.003 |
bye | -0.003 |
good | -0.003 |
ho | -0.003 |
Token | noticed |
Token ID | 6810 |
Feature activation | +5.34e-03 |
Pos logit contributions | |
Sad | +0.006 |
uster | +0.006 |
hole | +0.006 |
ice | +0.006 |
father | +0.006 |
Neg logit contributions | |
Sorry | -0.006 |
bye | -0.006 |
good | -0.006 |
they | -0.005 |
bye | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
Sorry | +0.006 |
bye | +0.006 |
good | +0.005 |
they | +0.005 |
bye | +0.005 |
Neg logit contributions | |
Sad | -0.013 |
uster | -0.013 |
hole | -0.013 |
ice | -0.012 |
father | -0.012 |
Token | cup |
Token ID | 6508 |
Feature activation | -2.70e-03 |
Pos logit contributions | |
uster | +0.027 |
Sad | +0.027 |
hole | +0.027 |
innie | +0.026 |
eping | +0.026 |
Neg logit contributions | |
bye | -0.019 |
Sorry | -0.018 |
they | -0.016 |
bye | -0.015 |
moms | -0.014 |
Token | board |
Token ID | 3526 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
hole | +0.004 |
uster | +0.004 |
Sad | +0.004 |
ice | +0.004 |
pled | +0.004 |
Neg logit contributions | |
Sorry | -0.005 |
bye | -0.005 |
good | -0.005 |
they | -0.005 |
bye | -0.005 |
Token | he |
Token ID | 339 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
uster | +0.039 |
hole | +0.039 |
Sad | +0.039 |
ice | +0.037 |
father | +0.036 |
Neg logit contributions | |
Sorry | -0.022 |
bye | -0.021 |
good | -0.020 |
they | -0.019 |
day | -0.018 |
Token | hadn |
Token ID | 8020 |
Feature activation | -6.73e-02 |
Pos logit contributions | |
uster | +0.061 |
Sad | +0.059 |
hole | +0.058 |
ist | +0.056 |
father | +0.055 |
Neg logit contributions | |
good | -0.026 |
Sorry | -0.026 |
bye | -0.025 |
day | -0.024 |
they | -0.024 |
Token | 't |
Token ID | 470 |
Feature activation | -2.97e-02 |
Pos logit contributions | |
uster | +0.162 |
ist | +0.160 |
ivering | +0.157 |
eping | +0.155 |
Sad | +0.153 |
Neg logit contributions | |
go | -0.074 |
they | -0.071 |
day | -0.067 |
good | -0.064 |
bye | -0.060 |
Token | seen |
Token ID | 1775 |
Feature activation | -5.15e-02 |
Pos logit contributions | |
uster | +0.043 |
Sad | +0.041 |
hole | +0.039 |
ice | +0.038 |
ist | +0.038 |
Neg logit contributions | |
good | -0.065 |
day | -0.063 |
they | -0.062 |
bye | -0.062 |
Sorry | -0.060 |
Token | before |
Token ID | 878 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
uster | +0.134 |
Sad | +0.133 |
hole | +0.126 |
Steven | +0.124 |
ist | +0.123 |
Neg logit contributions | |
good | -0.051 |
day | -0.048 |
well | -0.044 |
bye | -0.044 |
they | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | +9.39e-05 |
Pos logit contributions | |
hole | +0.024 |
uster | +0.023 |
Sad | +0.023 |
ice | +0.023 |
father | +0.022 |
Neg logit contributions | |
Sorry | -0.018 |
bye | -0.017 |
good | -0.015 |
bye | -0.015 |
they | -0.014 |
Token | He |
Token ID | 679 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
Sorry | +0.000 |
bye | +0.000 |
good | +0.000 |
they | +0.000 |
day | +0.000 |
Neg logit contributions | |
uster | -0.000 |
Sad | -0.000 |
hole | -0.000 |
ice | -0.000 |
father | -0.000 |
Token | for |
Token ID | 329 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
hole | +0.024 |
uster | +0.023 |
Sad | +0.022 |
ist | +0.022 |
ice | +0.022 |
Neg logit contributions | |
Sorry | -0.017 |
bye | -0.015 |
good | -0.014 |
bye | -0.013 |
they | -0.013 |
Token | him |
Token ID | 683 |
Feature activation | -8.66e-03 |
Pos logit contributions | |
uster | +0.022 |
Sad | +0.021 |
hole | +0.021 |
ice | +0.020 |
father | +0.020 |
Neg logit contributions | |
Sorry | -0.021 |
bye | -0.020 |
good | -0.019 |
they | -0.019 |
bye | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
hole | +0.017 |
uster | +0.016 |
ist | +0.015 |
ilda | +0.015 |
Sad | +0.015 |
Neg logit contributions | |
Sorry | -0.016 |
bye | -0.014 |
good | -0.013 |
bye | -0.012 |
moms | -0.012 |
Token | something |
Token ID | 1223 |
Feature activation | -1.16e-03 |
Pos logit contributions | |
uster | +0.057 |
hole | +0.057 |
Sad | +0.056 |
ice | +0.053 |
ist | +0.052 |
Neg logit contributions | |
Sorry | -0.058 |
bye | -0.056 |
good | -0.053 |
they | -0.050 |
day | -0.050 |
Token | he |
Token ID | 339 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
hole | +0.002 |
Sad | +0.002 |
uster | +0.002 |
pled | +0.002 |
eping | +0.002 |
Neg logit contributions | |
Sorry | -0.002 |
bye | -0.002 |
bye | -0.002 |
moms | -0.001 |
ho | -0.001 |
Token | hadn |
Token ID | 8020 |
Feature activation | -6.73e-02 |
Pos logit contributions | |
uster | +0.062 |
hole | +0.061 |
Sad | +0.061 |
ice | +0.059 |
father | +0.058 |
Neg logit contributions | |
Sorry | -0.045 |
bye | -0.043 |
good | -0.041 |
they | -0.040 |
day | -0.039 |
Token | 't |
Token ID | 470 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
uster | +0.155 |
ist | +0.153 |
ice | +0.152 |
Sad | +0.152 |
ivering | +0.149 |
Neg logit contributions | |
go | -0.081 |
they | -0.081 |
good | -0.074 |
day | -0.073 |
bye | -0.071 |
Token | expected |
Token ID | 2938 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
uster | +0.044 |
Sad | +0.043 |
hole | +0.043 |
ice | +0.040 |
father | +0.039 |
Neg logit contributions | |
Sorry | -0.075 |
bye | -0.071 |
good | -0.070 |
they | -0.068 |
bye | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
uster | +0.074 |
hole | +0.074 |
Sad | +0.073 |
ice | +0.071 |
father | +0.070 |
Neg logit contributions | |
Sorry | -0.047 |
bye | -0.043 |
good | -0.041 |
they | -0.040 |
bye | -0.038 |
Token | |
Token ID | 220 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
uster | +0.037 |
Sad | +0.037 |
hole | +0.037 |
ist | +0.035 |
ice | +0.035 |
Neg logit contributions | |
Sorry | -0.030 |
bye | -0.028 |
good | -0.028 |
they | -0.028 |
day | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | -2.49e-03 |
Pos logit contributions | |
Sad | +0.035 |
uster | +0.035 |
hole | +0.034 |
ice | +0.034 |
ist | +0.034 |
Neg logit contributions | |
Sorry | -0.014 |
they | -0.014 |
good | -0.014 |
bye | -0.012 |
day | -0.012 |
Token | all |
Token ID | 477 |
Feature activation | +7.20e-03 |
Pos logit contributions | |
Sorry | +0.034 |
bye | +0.033 |
ho | +0.027 |
day | +0.026 |
good | +0.025 |
Neg logit contributions | |
Sad | -0.039 |
ilda | -0.038 |
hole | -0.038 |
uster | -0.038 |
pled | -0.037 |
Token | about |
Token ID | 546 |
Feature activation | -5.19e-04 |
Pos logit contributions | |
Sorry | +0.015 |
bye | +0.015 |
good | +0.014 |
they | +0.014 |
ho | +0.013 |
Neg logit contributions | |
hole | -0.011 |
Sad | -0.011 |
uster | -0.010 |
ice | -0.010 |
father | -0.010 |
Token | what |
Token ID | 644 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
father | +0.001 |
Sad | +0.001 |
ice | +0.001 |
pled | +0.001 |
hole | +0.001 |
Neg logit contributions | |
Sorry | -0.001 |
bye | -0.001 |
ho | -0.001 |
bye | -0.001 |
good | -0.001 |
Token | he |
Token ID | 339 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
bye | +0.019 |
Sorry | +0.019 |
ho | +0.017 |
bye | +0.017 |
ashes | +0.015 |
Neg logit contributions | |
father | -0.019 |
James | -0.019 |
Sad | -0.019 |
pled | -0.019 |
Steven | -0.018 |
Token | 'd |
Token ID | 1549 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
uster | +0.111 |
Sad | +0.105 |
hole | +0.101 |
ist | +0.099 |
ivering | +0.096 |
Neg logit contributions | |
bye | -0.047 |
good | -0.047 |
day | -0.046 |
Sorry | -0.043 |
they | -0.042 |
Token | seen |
Token ID | 1775 |
Feature activation | -6.66e-02 |
Pos logit contributions | |
uster | +0.050 |
Sad | +0.048 |
hole | +0.047 |
ice | +0.046 |
innie | +0.045 |
Neg logit contributions | |
bye | -0.041 |
good | -0.040 |
day | -0.040 |
Sorry | -0.039 |
they | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
uster | +0.150 |
Sad | +0.148 |
hole | +0.147 |
ice | +0.141 |
father | +0.139 |
Neg logit contributions | |
Sorry | -0.089 |
bye | -0.085 |
good | -0.083 |
they | -0.080 |
day | -0.077 |
Token | His |
Token ID | 2399 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
Sorry | +0.026 |
good | +0.026 |
bye | +0.026 |
ho | +0.025 |
ge | +0.024 |
Neg logit contributions | |
uster | -0.019 |
Sad | -0.019 |
hole | -0.017 |
innie | -0.017 |
James | -0.017 |
Token | friends |
Token ID | 2460 |
Feature activation | +8.69e-03 |
Pos logit contributions | |
Sorry | +0.021 |
bye | +0.020 |
they | +0.019 |
ho | +0.017 |
bye | +0.015 |
Neg logit contributions | |
father | -0.018 |
iz | -0.018 |
innie | -0.018 |
x | -0.017 |
Sad | -0.016 |
Token | thought |
Token ID | 1807 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
Sorry | +0.012 |
bye | +0.011 |
good | +0.011 |
day | +0.010 |
bye | +0.010 |
Neg logit contributions | |
hole | -0.019 |
uster | -0.019 |
Sad | -0.019 |
ist | -0.018 |
ice | -0.018 |
Token | it |
Token ID | 340 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
Sad | +0.111 |
uster | +0.110 |
hole | +0.108 |
ivering | +0.106 |
ice | +0.104 |
Neg logit contributions | |
they | -0.033 |
good | -0.031 |
day | -0.028 |
go | -0.025 |
bye | -0.023 |
Token | baffled |
Token ID | 44432 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
hole | +0.011 |
Sad | +0.010 |
uster | +0.010 |
pled | +0.010 |
ice | +0.010 |
Neg logit contributions | |
Sorry | -0.010 |
bye | -0.010 |
bye | -0.008 |
they | -0.008 |
good | -0.008 |
Token | by |
Token ID | 416 |
Feature activation | -5.33e-03 |
Pos logit contributions | |
uster | +0.044 |
Sad | +0.043 |
hole | +0.043 |
ice | +0.041 |
father | +0.040 |
Neg logit contributions | |
good | -0.026 |
Sorry | -0.026 |
they | -0.026 |
bye | -0.025 |
day | -0.025 |
Token | what |
Token ID | 644 |
Feature activation | -1.24e-03 |
Pos logit contributions | |
Sad | +0.013 |
hole | +0.013 |
uster | +0.013 |
ice | +0.012 |
father | +0.012 |
Neg logit contributions | |
Sorry | -0.007 |
bye | -0.006 |
good | -0.005 |
they | -0.005 |
day | -0.005 |
Token | she |
Token ID | 673 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
Sad | +0.003 |
father | +0.003 |
ice | +0.002 |
pled | +0.002 |
hole | +0.002 |
Neg logit contributions | |
Sorry | -0.002 |
bye | -0.002 |
bye | -0.002 |
good | -0.001 |
ho | -0.001 |
Token | had |
Token ID | 550 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
uster | +0.107 |
Sad | +0.103 |
hole | +0.100 |
ice | +0.096 |
ist | +0.095 |
Neg logit contributions | |
bye | -0.047 |
Sorry | -0.047 |
good | -0.046 |
day | -0.045 |
they | -0.044 |
Token | seen |
Token ID | 1775 |
Feature activation | -6.66e-02 |
Pos logit contributions | |
uster | +0.062 |
Sad | +0.060 |
hole | +0.058 |
ice | +0.056 |
innie | +0.055 |
Neg logit contributions | |
Sorry | -0.056 |
good | -0.056 |
bye | -0.056 |
they | -0.055 |
day | -0.054 |
Token | , |
Token ID | 11 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
uster | +0.142 |
Sad | +0.140 |
hole | +0.139 |
ice | +0.133 |
father | +0.131 |
Neg logit contributions | |
Sorry | -0.097 |
bye | -0.092 |
good | -0.091 |
they | -0.088 |
day | -0.085 |
Token | but |
Token ID | 475 |
Feature activation | -9.40e-03 |
Pos logit contributions | |
hole | +0.049 |
uster | +0.049 |
Sad | +0.048 |
ice | +0.045 |
father | +0.044 |
Neg logit contributions | |
Sorry | -0.066 |
bye | -0.064 |
good | -0.061 |
they | -0.059 |
day | -0.058 |
Token | she |
Token ID | 673 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
Sad | +0.020 |
hole | +0.020 |
father | +0.019 |
ice | +0.018 |
ilda | +0.018 |
Neg logit contributions | |
Sorry | -0.014 |
bye | -0.013 |
good | -0.011 |
bye | -0.011 |
seals | -0.011 |
Token | slept |
Token ID | 21256 |
Feature activation | -7.44e-03 |
Pos logit contributions | |
Sad | +0.023 |
hole | +0.023 |
uster | +0.023 |
ice | +0.023 |
father | +0.022 |
Neg logit contributions | |
Sorry | -0.018 |
bye | -0.016 |
good | -0.016 |
they | -0.015 |
bye | -0.015 |
Token | peacefully |
Token ID | 30996 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
uster | +0.018 |
Sad | +0.017 |
hole | +0.017 |
ice | +0.017 |
father | +0.016 |
Neg logit contributions | |
Sorry | -0.009 |
bye | -0.009 |
good | -0.008 |
they | -0.008 |
day | -0.008 |
Token | smile |
Token ID | 8212 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
Sorry | +0.020 |
bye | +0.019 |
ho | +0.017 |
Hipp | +0.016 |
seals | +0.015 |
Neg logit contributions | |
ist | -0.024 |
Sad | -0.024 |
eping | -0.024 |
hole | -0.023 |
ivering | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
Sorry | +0.043 |
bye | +0.040 |
ge | +0.037 |
moms | +0.037 |
ho | +0.036 |
Neg logit contributions | |
ice | -0.026 |
pled | -0.026 |
hole | -0.025 |
aron | -0.025 |
Sad | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
bye | +0.029 |
Sorry | +0.029 |
good | +0.027 |
day | +0.026 |
ho | +0.026 |
Neg logit contributions | |
uster | -0.029 |
Sad | -0.029 |
hole | -0.027 |
ice | -0.027 |
innie | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
they | +0.017 |
good | +0.017 |
Sorry | +0.016 |
day | +0.015 |
bye | +0.015 |
Neg logit contributions | |
uster | -0.047 |
hole | -0.046 |
ist | -0.045 |
ice | -0.045 |
Sad | -0.044 |
Token | Anna |
Token ID | 31160 |
Feature activation | +3.13e-02 |
Pos logit contributions | |
uster | +0.029 |
hole | +0.028 |
Sad | +0.028 |
ice | +0.027 |
father | +0.027 |
Neg logit contributions | |
Sorry | -0.021 |
bye | -0.020 |
good | -0.020 |
they | -0.020 |
day | -0.019 |
Token | cl |
Token ID | 537 |
Feature activation | +3.78e-02 |
Pos logit contributions | |
Sorry | +0.063 |
bye | +0.060 |
day | +0.054 |
good | +0.053 |
they | +0.052 |
Neg logit contributions | |
ice | -0.051 |
hole | -0.049 |
pled | -0.048 |
ist | -0.047 |
uster | -0.046 |
Token | aps |
Token ID | 1686 |
Feature activation | +3.37e-02 |
Pos logit contributions | |
bye | +0.077 |
Sorry | +0.076 |
good | +0.076 |
they | +0.076 |
day | +0.075 |
Neg logit contributions | |
uster | -0.055 |
Sad | -0.053 |
hole | -0.053 |
ice | -0.050 |
ady | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
Sorry | +0.047 |
bye | +0.044 |
good | +0.040 |
they | +0.039 |
day | +0.038 |
Neg logit contributions | |
uster | -0.074 |
Sad | -0.074 |
hole | -0.073 |
ice | -0.070 |
father | -0.069 |
Token | She |
Token ID | 1375 |
Feature activation | -1.91e-04 |
Pos logit contributions | |
bye | +0.041 |
Sorry | +0.040 |
good | +0.038 |
day | +0.037 |
marriage | +0.036 |
Neg logit contributions | |
uster | -0.027 |
Sad | -0.025 |
hole | -0.024 |
ice | -0.023 |
innie | -0.023 |
Token | is |
Token ID | 318 |
Feature activation | -6.96e-03 |
Pos logit contributions | |
ice | +0.000 |
hole | +0.000 |
pled | +0.000 |
ist | +0.000 |
Sad | +0.000 |
Neg logit contributions | |
Sorry | -0.000 |
bye | -0.000 |
day | -0.000 |
ho | -0.000 |
good | -0.000 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
Sad | +0.014 |
eping | +0.014 |
pled | +0.013 |
ady | +0.013 |
ice | +0.013 |
Neg logit contributions | |
Sorry | -0.011 |
bye | -0.011 |
ho | -0.010 |
bye | -0.009 |
day | -0.009 |
Token | you |
Token ID | 345 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
hole | +0.023 |
Sad | +0.022 |
uster | +0.022 |
father | +0.022 |
ice | +0.022 |
Neg logit contributions | |
Sorry | -0.014 |
bye | -0.014 |
good | -0.012 |
ho | -0.012 |
day | -0.011 |
Token | can |
Token ID | 460 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
Sorry | +0.023 |
bye | +0.021 |
bye | +0.019 |
good | +0.019 |
they | +0.019 |
Neg logit contributions | |
hole | -0.030 |
uster | -0.029 |
ist | -0.028 |
ice | -0.028 |
father | -0.027 |
Token | try |
Token ID | 1949 |
Feature activation | -5.82e-03 |
Pos logit contributions | |
uster | +0.027 |
hole | +0.027 |
Sad | +0.027 |
father | +0.025 |
ice | +0.025 |
Neg logit contributions | |
Sorry | -0.011 |
bye | -0.010 |
good | -0.010 |
they | -0.009 |
day | -0.009 |
Token | it |
Token ID | 340 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
hole | +0.014 |
uster | +0.014 |
Sad | +0.013 |
ice | +0.013 |
father | +0.013 |
Neg logit contributions | |
Sorry | -0.007 |
bye | -0.007 |
good | -0.006 |
they | -0.006 |
day | -0.006 |
Token | ." |
Token ID | 526 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
Sorry | +0.022 |
ho | +0.019 |
bye | +0.019 |
ge | +0.018 |
bye | +0.017 |
Neg logit contributions | |
hole | -0.021 |
ist | -0.020 |
ice | -0.019 |
Sad | -0.019 |
uster | -0.019 |
Token | Anna |
Token ID | 11735 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
Sorry | +0.018 |
bye | +0.017 |
good | +0.017 |
they | +0.017 |
day | +0.016 |
Neg logit contributions | |
uster | -0.029 |
Sad | -0.028 |
hole | -0.028 |
ice | -0.027 |
father | -0.026 |
Token | comes |
Token ID | 2058 |
Feature activation | -1.41e-03 |
Pos logit contributions | |
Sorry | +0.061 |
bye | +0.059 |
good | +0.055 |
day | +0.054 |
they | +0.053 |
Neg logit contributions | |
hole | -0.066 |
ice | -0.065 |
uster | -0.065 |
Sad | -0.064 |
ist | -0.062 |
Token | to |
Token ID | 284 |
Feature activation | -1.45e-03 |
Pos logit contributions | |
uster | +0.004 |
Sad | +0.003 |
hole | +0.003 |
ice | +0.003 |
father | +0.003 |
Neg logit contributions | |
Sorry | -0.002 |
bye | -0.001 |
good | -0.001 |
they | -0.001 |
day | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
uster | +0.004 |
Sad | +0.004 |
hole | +0.004 |
ice | +0.004 |
father | +0.004 |
Neg logit contributions | |
Sorry | -0.001 |
bye | -0.001 |
good | -0.001 |
they | -0.001 |
day | -0.001 |
Token | door |
Token ID | 3420 |
Feature activation | -8.23e-03 |
Pos logit contributions | |
hole | +0.022 |
uster | +0.020 |
Sad | +0.019 |
box | +0.019 |
James | +0.019 |
Neg logit contributions | |
Sorry | -0.021 |
bye | -0.021 |
they | -0.019 |
good | -0.018 |
bye | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +4.72e-03 |
Pos logit contributions | |
uster | +0.021 |
Sad | +0.020 |
hole | +0.020 |
ice | +0.019 |
father | +0.019 |
Neg logit contributions | |
Sorry | -0.009 |
good | -0.009 |
bye | -0.009 |
they | -0.009 |
day | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
Sorry | +0.025 |
bye | +0.024 |
good | +0.023 |
they | +0.022 |
day | +0.022 |
Neg logit contributions | |
uster | -0.036 |
Sad | -0.036 |
hole | -0.036 |
ice | -0.034 |
father | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | -6.19e-03 |
Pos logit contributions | |
they | +0.015 |
good | +0.014 |
Sorry | +0.014 |
day | +0.012 |
bye | +0.012 |
Neg logit contributions | |
uster | -0.044 |
hole | -0.043 |
ist | -0.043 |
ice | -0.042 |
pled | -0.041 |
Token | " |
Token ID | 1 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
uster | +0.012 |
Sad | +0.012 |
hole | +0.012 |
ice | +0.012 |
father | +0.012 |
Neg logit contributions | |
Sorry | -0.010 |
bye | -0.009 |
good | -0.009 |
they | -0.009 |
day | -0.009 |
Token | Sorry |
Token ID | 14385 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
Sorry | +0.019 |
bye | +0.018 |
good | +0.018 |
they | +0.017 |
day | +0.017 |
Neg logit contributions | |
Sad | -0.018 |
uster | -0.018 |
hole | -0.018 |
ice | -0.017 |
father | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | +3.67e-03 |
Pos logit contributions | |
Sorry | +0.013 |
bye | +0.012 |
good | +0.011 |
day | +0.010 |
ho | +0.010 |
Neg logit contributions | |
uster | -0.025 |
Sad | -0.025 |
hole | -0.025 |
ice | -0.025 |
pled | -0.025 |
Token | sorry |
Token ID | 7926 |
Feature activation | +3.97e-02 |
Pos logit contributions | |
bye | +0.007 |
Sorry | +0.006 |
ho | +0.005 |
day | +0.005 |
west | +0.005 |
Neg logit contributions | |
ice | -0.006 |
uster | -0.006 |
Sad | -0.006 |
pled | -0.006 |
hole | -0.006 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
Sorry | +0.059 |
bye | +0.057 |
day | +0.055 |
ho | +0.053 |
marriage | +0.052 |
Neg logit contributions | |
uster | -0.065 |
ice | -0.064 |
hole | -0.064 |
Sad | -0.064 |
pled | -0.064 |
Token | Ben |
Token ID | 3932 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
bye | +0.033 |
Sorry | +0.031 |
good | +0.029 |
day | +0.028 |
ho | +0.028 |
Neg logit contributions | |
Sad | -0.040 |
uster | -0.039 |
ice | -0.038 |
hole | -0.037 |
ilda | -0.036 |
Token | says |
Token ID | 1139 |
Feature activation | +6.26e-03 |
Pos logit contributions | |
Sorry | +0.015 |
bye | +0.014 |
good | +0.014 |
they | +0.013 |
day | +0.013 |
Neg logit contributions | |
uster | -0.024 |
Sad | -0.024 |
hole | -0.024 |
ice | -0.023 |
father | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +4.52e-03 |
Pos logit contributions | |
Sorry | +0.008 |
bye | +0.007 |
good | +0.007 |
they | +0.007 |
day | +0.007 |
Neg logit contributions | |
uster | -0.015 |
Sad | -0.015 |
hole | -0.015 |
ice | -0.014 |
father | -0.014 |
Token | He |
Token ID | 679 |
Feature activation | -8.15e-04 |
Pos logit contributions | |
Sorry | +0.009 |
bye | +0.008 |
good | +0.008 |
they | +0.008 |
day | +0.008 |
Neg logit contributions | |
uster | -0.008 |
Sad | -0.008 |
hole | -0.007 |
ice | -0.007 |
father | -0.007 |
Token | turns |
Token ID | 4962 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
ho | +0.023 |
Sorry | +0.022 |
bye | +0.021 |
Hipp | +0.018 |
melon | +0.017 |
Neg logit contributions | |
ice | -0.027 |
pled | -0.027 |
uster | -0.026 |
Sad | -0.025 |
eping | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
Sorry | +0.031 |
bye | +0.028 |
bye | +0.024 |
good | +0.023 |
ho | +0.022 |
Neg logit contributions | |
hole | -0.031 |
pled | -0.030 |
eping | -0.029 |
run | -0.029 |
ist | -0.028 |
Token | They |
Token ID | 1119 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
bye | +0.030 |
Sorry | +0.028 |
ho | +0.027 |
ge | +0.024 |
seal | +0.024 |
Neg logit contributions | |
Sad | -0.041 |
uster | -0.040 |
ice | -0.038 |
hole | -0.037 |
innie | -0.036 |
Token | laughed |
Token ID | 13818 |
Feature activation | +2.83e-02 |
Pos logit contributions | |
Sorry | +0.019 |
bye | +0.018 |
bye | +0.016 |
good | +0.015 |
day | +0.015 |
Neg logit contributions | |
ice | -0.021 |
Sad | -0.021 |
pled | -0.021 |
hole | -0.020 |
ist | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
Sorry | +0.068 |
bye | +0.062 |
ho | +0.061 |
ge | +0.057 |
moms | +0.055 |
Neg logit contributions | |
hole | -0.034 |
pled | -0.034 |
Sad | -0.034 |
uster | -0.034 |
ice | -0.032 |
Token | talked |
Token ID | 6619 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
Sorry | +0.033 |
bye | +0.032 |
ho | +0.029 |
birthday | +0.029 |
marriage | +0.028 |
Neg logit contributions | |
pled | -0.030 |
Sad | -0.028 |
eping | -0.027 |
ice | -0.026 |
hole | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
Sorry | +0.089 |
bye | +0.077 |
ho | +0.074 |
ge | +0.074 |
Hipp | +0.073 |
Neg logit contributions | |
hole | -0.056 |
ist | -0.054 |
pled | -0.053 |
Sad | -0.053 |
aron | -0.052 |
Token | Mom |
Token ID | 11254 |
Feature activation | -6.21e-03 |
Pos logit contributions | |
bye | +0.030 |
Sorry | +0.028 |
ho | +0.028 |
seal | +0.025 |
ge | +0.025 |
Neg logit contributions | |
Sad | -0.043 |
uster | -0.041 |
ice | -0.039 |
innie | -0.038 |
hole | -0.038 |
Token | watched |
Token ID | 7342 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
uster | +0.014 |
Sad | +0.013 |
hole | +0.013 |
ice | +0.013 |
father | +0.013 |
Neg logit contributions | |
Sorry | -0.009 |
bye | -0.008 |
good | -0.008 |
they | -0.008 |
day | -0.008 |
Token | them |
Token ID | 606 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
uster | +0.028 |
Sad | +0.028 |
hole | +0.028 |
ice | +0.027 |
father | +0.026 |
Neg logit contributions | |
Sorry | -0.020 |
bye | -0.019 |
good | -0.018 |
they | -0.017 |
day | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | +5.27e-03 |
Pos logit contributions | |
Sorry | +0.035 |
bye | +0.033 |
ho | +0.031 |
ge | +0.029 |
ww | +0.028 |
Neg logit contributions | |
hole | -0.033 |
ist | -0.032 |
Sad | -0.030 |
uster | -0.030 |
ice | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | +9.49e-04 |
Pos logit contributions | |
uster | +0.030 |
hole | +0.030 |
ist | +0.029 |
eping | +0.029 |
ice | +0.029 |
Neg logit contributions | |
they | -0.011 |
Sorry | -0.011 |
good | -0.010 |
bye | -0.009 |
day | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
they | +0.001 |
good | +0.001 |
Sorry | +0.001 |
go | +0.000 |
bye | +0.000 |
Neg logit contributions | |
uster | -0.003 |
ist | -0.003 |
ice | -0.003 |
hole | -0.003 |
father | -0.003 |
Token | Tom |
Token ID | 13787 |
Feature activation | -4.03e-03 |
Pos logit contributions | |
uster | +0.042 |
hole | +0.041 |
Sad | +0.039 |
ice | +0.039 |
ist | +0.039 |
Neg logit contributions | |
Sorry | -0.034 |
good | -0.033 |
they | -0.033 |
bye | -0.033 |
bye | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | +3.43e-03 |
Pos logit contributions | |
uster | +0.010 |
Sad | +0.010 |
hole | +0.010 |
father | +0.009 |
innie | +0.009 |
Neg logit contributions | |
Sorry | -0.005 |
good | -0.005 |
they | -0.004 |
bye | -0.004 |
go | -0.004 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
Sorry | +0.005 |
bye | +0.004 |
good | +0.004 |
they | +0.004 |
day | +0.004 |
Neg logit contributions | |
Sad | -0.008 |
uster | -0.008 |
hole | -0.008 |
ice | -0.007 |
father | -0.007 |
Token | started |
Token ID | 2067 |
Feature activation | +3.52e-02 |
Pos logit contributions | |
Sorry | +0.026 |
bye | +0.024 |
good | +0.023 |
they | +0.022 |
day | +0.022 |
Neg logit contributions | |
uster | -0.049 |
hole | -0.049 |
Sad | -0.049 |
ice | -0.047 |
father | -0.046 |
Token | to |
Token ID | 284 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
Sorry | +0.047 |
bye | +0.046 |
bye | +0.039 |
good | +0.038 |
day | +0.036 |
Neg logit contributions | |
Sad | -0.078 |
hole | -0.076 |
uster | -0.075 |
pled | -0.073 |
ice | -0.072 |
Token | fight |
Token ID | 1907 |
Feature activation | +7.96e-03 |
Pos logit contributions | |
Sorry | +0.031 |
bye | +0.031 |
good | +0.028 |
they | +0.027 |
bye | +0.027 |
Neg logit contributions | |
Sad | -0.043 |
hole | -0.042 |
uster | -0.042 |
ice | -0.040 |
ist | -0.039 |
Token | over |
Token ID | 625 |
Feature activation | +8.80e-03 |
Pos logit contributions | |
Sorry | +0.012 |
bye | +0.011 |
good | +0.010 |
ho | +0.010 |
day | +0.010 |
Neg logit contributions | |
Sad | -0.017 |
hole | -0.017 |
uster | -0.017 |
ice | -0.016 |
ist | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
Sorry | +0.009 |
bye | +0.009 |
good | +0.007 |
bye | +0.007 |
day | +0.007 |
Neg logit contributions | |
hole | -0.023 |
Sad | -0.023 |
uster | -0.022 |
ice | -0.022 |
pled | -0.022 |
Token | shiny |
Token ID | 22441 |
Feature activation | +4.01e-03 |
Pos logit contributions | |
good | +0.027 |
Sorry | +0.026 |
bye | +0.025 |
day | +0.024 |
they | +0.024 |
Neg logit contributions | |
uster | -0.046 |
Sad | -0.046 |
hole | -0.045 |
ice | -0.044 |
ist | -0.044 |
Token | " |
Token ID | 366 |
Feature activation | +8.28e-03 |
Pos logit contributions | |
Sad | +0.007 |
uster | +0.007 |
hole | +0.007 |
ice | +0.007 |
pled | +0.007 |
Neg logit contributions | |
Sorry | -0.005 |
bye | -0.005 |
good | -0.004 |
day | -0.004 |
bye | -0.004 |
Token | L |
Token ID | 43 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
Sorry | +0.014 |
bye | +0.013 |
good | +0.013 |
they | +0.012 |
day | +0.012 |
Neg logit contributions | |
uster | -0.016 |
hole | -0.016 |
Sad | -0.016 |
ice | -0.015 |
father | -0.015 |
Token | ily |
Token ID | 813 |
Feature activation | -6.53e-05 |
Pos logit contributions | |
Sad | +0.035 |
uster | +0.033 |
hole | +0.033 |
Steven | +0.032 |
ilda | +0.032 |
Neg logit contributions | |
bye | -0.021 |
good | -0.021 |
day | -0.019 |
bye | -0.019 |
go | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
hole | +0.000 |
uster | +0.000 |
ice | +0.000 |
pled | +0.000 |
Sad | +0.000 |
Neg logit contributions | |
Sorry | -0.000 |
bye | -0.000 |
good | -0.000 |
they | -0.000 |
day | -0.000 |
Token | wait |
Token ID | 4043 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
bye | +0.019 |
Sorry | +0.018 |
day | +0.015 |
good | +0.015 |
ho | +0.015 |
Neg logit contributions | |
Sad | -0.026 |
uster | -0.026 |
hole | -0.025 |
pled | -0.025 |
ice | -0.024 |
Token | ! |
Token ID | 0 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
Sorry | +0.035 |
bye | +0.033 |
ho | +0.029 |
they | +0.028 |
good | +0.028 |
Neg logit contributions | |
hole | -0.047 |
uster | -0.046 |
ist | -0.045 |
Sad | -0.045 |
ice | -0.043 |
Token | I |
Token ID | 314 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
good | +0.021 |
bye | +0.021 |
Sorry | +0.021 |
they | +0.020 |
day | +0.020 |
Neg logit contributions | |
Sad | -0.025 |
uster | -0.025 |
hole | -0.024 |
ice | -0.023 |
father | -0.023 |
Token | 'm |
Token ID | 1101 |
Feature activation | -3.38e-03 |
Pos logit contributions | |
Sorry | +0.016 |
bye | +0.015 |
bye | +0.014 |
they | +0.014 |
good | +0.014 |
Neg logit contributions | |
Sad | -0.020 |
uster | -0.019 |
hole | -0.019 |
ist | -0.019 |
ice | -0.018 |
Token | sorry |
Token ID | 7926 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
uster | +0.007 |
Sad | +0.007 |
hole | +0.006 |
ice | +0.006 |
eping | +0.006 |
Neg logit contributions | |
Sorry | -0.006 |
bye | -0.006 |
day | -0.005 |
ho | -0.005 |
good | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
Sorry | +0.014 |
bye | +0.014 |
day | +0.012 |
ho | +0.012 |
good | +0.012 |
Neg logit contributions | |
uster | -0.023 |
Sad | -0.023 |
hole | -0.023 |
ice | -0.022 |
ist | -0.022 |
Token | I |
Token ID | 314 |
Feature activation | +9.54e-03 |
Pos logit contributions | |
good | +0.033 |
bye | +0.032 |
Sorry | +0.032 |
they | +0.032 |
day | +0.031 |
Neg logit contributions | |
Sad | -0.047 |
uster | -0.047 |
hole | -0.046 |
ice | -0.044 |
father | -0.044 |
Token | ink |
Token ID | 16882 |
Feature activation | +9.07e-04 |
Pos logit contributions | |
Sad | +0.002 |
uster | +0.002 |
hole | +0.002 |
ice | +0.001 |
eping | +0.001 |
Neg logit contributions | |
Sorry | -0.001 |
bye | -0.001 |
they | -0.001 |
ho | -0.001 |
good | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +1.67e-03 |
Pos logit contributions | |
Sorry | +0.002 |
bye | +0.001 |
good | +0.001 |
bye | +0.001 |
day | +0.001 |
Neg logit contributions | |
hole | -0.002 |
uster | -0.002 |
ist | -0.002 |
ice | -0.002 |
Sad | -0.002 |
Token | The |
Token ID | 383 |
Feature activation | +5.66e-03 |
Pos logit contributions | |
Sorry | +0.003 |
bye | +0.003 |
good | +0.003 |
day | +0.003 |
they | +0.003 |
Neg logit contributions | |
uster | -0.003 |
Sad | -0.003 |
hole | -0.003 |
innie | -0.002 |
ilda | -0.002 |
Token | ink |
Token ID | 16882 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
Sorry | +0.011 |
bye | +0.010 |
they | +0.010 |
bye | +0.009 |
good | +0.009 |
Neg logit contributions | |
Sad | -0.010 |
uster | -0.009 |
ice | -0.009 |
hole | -0.009 |
father | -0.009 |
Token | spl |
Token ID | 4328 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
Sorry | +0.021 |
bye | +0.019 |
bye | +0.017 |
good | +0.017 |
day | +0.015 |
Neg logit contributions | |
hole | -0.023 |
pled | -0.022 |
Sad | -0.022 |
uster | -0.022 |
ice | -0.022 |
Token | ashed |
Token ID | 5263 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
Sorry | +0.023 |
good | +0.021 |
bye | +0.021 |
they | +0.020 |
day | +0.019 |
Neg logit contributions | |
uster | -0.044 |
hole | -0.042 |
Sad | -0.042 |
ist | -0.041 |
ice | -0.041 |
Token | everywhere |
Token ID | 8347 |
Feature activation | +3.52e-03 |
Pos logit contributions | |
Sorry | +0.030 |
bye | +0.027 |
bye | +0.024 |
moms | +0.023 |
marriage | +0.022 |
Neg logit contributions | |
uster | -0.032 |
Sad | -0.032 |
hole | -0.031 |
pled | -0.031 |
ice | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | -2.55e-02 |
Pos logit contributions | |
Sorry | +0.005 |
bye | +0.005 |
good | +0.004 |
bye | +0.004 |
day | +0.004 |
Neg logit contributions | |
Sad | -0.008 |
uster | -0.007 |
hole | -0.007 |
ist | -0.007 |
ice | -0.007 |
Token | making |
Token ID | 1642 |
Feature activation | +3.17e-03 |
Pos logit contributions | |
uster | +0.044 |
hole | +0.044 |
Sad | +0.043 |
ice | +0.041 |
ist | +0.041 |
Neg logit contributions | |
Sorry | -0.048 |
bye | -0.046 |
good | -0.043 |
bye | -0.041 |
day | -0.041 |
Token | a |
Token ID | 257 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
Sorry | +0.005 |
bye | +0.005 |
ho | +0.004 |
bye | +0.004 |
day | +0.003 |
Neg logit contributions | |
Sad | -0.007 |
pled | -0.007 |
ice | -0.007 |
uster | -0.007 |
eping | -0.007 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
bye | +0.021 |
Sorry | +0.020 |
they | +0.017 |
ho | +0.016 |
day | +0.016 |
Neg logit contributions | |
Sad | -0.032 |
uster | -0.030 |
hole | -0.030 |
ivering | -0.030 |
eping | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | -7.26e-03 |
Pos logit contributions | |
bye | +0.019 |
Sorry | +0.018 |
ho | +0.018 |
moms | +0.016 |
vows | +0.016 |
Neg logit contributions | |
ice | -0.014 |
pled | -0.014 |
Sad | -0.013 |
uster | -0.012 |
innie | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | -5.22e-04 |
Pos logit contributions | |
Sad | +0.013 |
ice | +0.012 |
James | +0.011 |
nowhere | +0.011 |
pled | +0.011 |
Neg logit contributions | |
bye | -0.012 |
Sorry | -0.011 |
ho | -0.011 |
vows | -0.010 |
bye | -0.010 |
Token | human |
Token ID | 1692 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
uster | +0.001 |
Sad | +0.001 |
hole | +0.001 |
ice | +0.001 |
father | +0.001 |
Neg logit contributions | |
Sorry | -0.001 |
bye | -0.001 |
ho | -0.001 |
they | -0.001 |
good | -0.001 |
Token | saw |
Token ID | 2497 |
Feature activation | +2.49e-03 |
Pos logit contributions | |
Sorry | +0.023 |
bye | +0.021 |
good | +0.020 |
they | +0.019 |
bye | +0.019 |
Neg logit contributions | |
ice | -0.030 |
hole | -0.030 |
Sad | -0.029 |
uster | -0.029 |
father | -0.029 |
Token | a |
Token ID | 257 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
Sorry | +0.004 |
bye | +0.004 |
ho | +0.003 |
bye | +0.003 |
good | +0.003 |
Neg logit contributions | |
Sad | -0.005 |
ice | -0.005 |
pled | -0.005 |
father | -0.005 |
eping | -0.005 |
Token | little |
Token ID | 1310 |
Feature activation | +3.86e-03 |
Pos logit contributions | |
Sad | +0.031 |
uster | +0.031 |
hole | +0.030 |
eping | +0.029 |
innie | +0.029 |
Neg logit contributions | |
bye | -0.019 |
Sorry | -0.018 |
they | -0.015 |
bye | -0.014 |
ho | -0.014 |
Token | cater |
Token ID | 20825 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
Sorry | +0.006 |
bye | +0.006 |
good | +0.005 |
they | +0.005 |
bye | +0.005 |
Neg logit contributions | |
hole | -0.008 |
uster | -0.008 |
pled | -0.008 |
Sad | -0.008 |
ice | -0.008 |
Token | pillar |
Token ID | 41643 |
Feature activation | -6.22e-03 |
Pos logit contributions | |
uster | +0.026 |
hole | +0.026 |
Sad | +0.026 |
father | +0.024 |
innie | +0.024 |
Neg logit contributions | |
bye | -0.028 |
good | -0.027 |
they | -0.027 |
Sorry | -0.027 |
day | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | +3.04e-03 |
Pos logit contributions | |
hole | +0.013 |
Sad | +0.013 |
uster | +0.013 |
ice | +0.013 |
father | +0.012 |
Neg logit contributions | |
Sorry | -0.009 |
bye | -0.009 |
good | -0.008 |
they | -0.008 |
bye | -0.008 |
Token | decided |
Token ID | 3066 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
bye | +0.005 |
Sorry | +0.005 |
day | +0.004 |
good | +0.004 |
ho | +0.004 |
Neg logit contributions | |
Sad | -0.005 |
ice | -0.005 |
pled | -0.005 |
hole | -0.005 |
eping | -0.005 |
Token | to |
Token ID | 284 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
Sorry | +0.031 |
bye | +0.029 |
day | +0.026 |
good | +0.024 |
ho | +0.024 |
Neg logit contributions | |
Sad | -0.036 |
ilda | -0.035 |
hole | -0.034 |
uster | -0.033 |
pled | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | -1.79e-04 |
Pos logit contributions | |
Sorry | +0.023 |
bye | +0.021 |
good | +0.020 |
they | +0.019 |
day | +0.019 |
Neg logit contributions | |
uster | -0.025 |
hole | -0.025 |
Sad | -0.024 |
ice | -0.023 |
ist | -0.023 |
Token | then |
Token ID | 788 |
Feature activation | +2.84e-03 |
Pos logit contributions | |
Sad | +0.000 |
pled | +0.000 |
ilda | +0.000 |
Sarah | +0.000 |
ist | +0.000 |
Neg logit contributions | |
Sorry | -0.000 |
bye | -0.000 |
day | -0.000 |
birthday | -0.000 |
seals | -0.000 |
Token | she |
Token ID | 673 |
Feature activation | +3.69e-03 |
Pos logit contributions | |
Sorry | +0.006 |
bye | +0.005 |
birthday | +0.005 |
day | +0.004 |
seals | +0.004 |
Neg logit contributions | |
pled | -0.004 |
hole | -0.004 |
ice | -0.004 |
ilda | -0.004 |
Sad | -0.004 |
Token | said |
Token ID | 531 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
Sorry | +0.007 |
bye | +0.006 |
good | +0.006 |
day | +0.005 |
bye | +0.005 |
Neg logit contributions | |
hole | -0.007 |
ice | -0.007 |
pled | -0.007 |
uster | -0.007 |
Sad | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | -9.08e-03 |
Pos logit contributions | |
Sorry | +0.029 |
bye | +0.026 |
seals | +0.025 |
day | +0.025 |
floats | +0.024 |
Neg logit contributions | |
ently | -0.033 |
hole | -0.032 |
run | -0.032 |
ist | -0.032 |
pled | -0.032 |
Token | " |
Token ID | 366 |
Feature activation | +4.29e-03 |
Pos logit contributions | |
Sad | +0.018 |
hole | +0.018 |
uster | +0.018 |
ice | +0.017 |
ilda | +0.017 |
Neg logit contributions | |
bye | -0.014 |
Sorry | -0.014 |
good | -0.012 |
day | -0.012 |
moms | -0.011 |
Token | Let |
Token ID | 5756 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
Sorry | +0.007 |
good | +0.006 |
bye | +0.006 |
they | +0.006 |
bye | +0.006 |
Neg logit contributions | |
Sad | -0.009 |
uster | -0.009 |
hole | -0.009 |
ice | -0.008 |
ist | -0.008 |
Token | 's |
Token ID | 338 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
Sad | +0.043 |
ice | +0.042 |
hole | +0.042 |
uster | +0.041 |
ilda | +0.040 |
Neg logit contributions | |
Sorry | -0.042 |
bye | -0.039 |
ho | -0.037 |
good | -0.037 |
they | -0.036 |
Token | make |
Token ID | 787 |
Feature activation | -2.55e-03 |
Pos logit contributions | |
bye | +0.025 |
Sorry | +0.025 |
ho | +0.024 |
good | +0.023 |
they | +0.022 |
Neg logit contributions | |
Sad | -0.050 |
hole | -0.049 |
ist | -0.049 |
uster | -0.048 |
father | -0.047 |
Token | a |
Token ID | 257 |
Feature activation | +2.48e-02 |
Pos logit contributions | |
Sad | +0.006 |
ice | +0.006 |
hole | +0.006 |
eping | +0.005 |
pled | +0.005 |
Neg logit contributions | |
Sorry | -0.003 |
bye | -0.003 |
ho | -0.003 |
day | -0.003 |
they | -0.003 |
Token | game |
Token ID | 983 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
bye | +0.036 |
Sorry | +0.035 |
they | +0.032 |
ho | +0.030 |
Hipp | +0.029 |
Neg logit contributions | |
Sad | -0.045 |
hole | -0.044 |
x | -0.043 |
eping | -0.043 |
ist | -0.043 |
Token | pole |
Token ID | 16825 |
Feature activation | -5.58e-03 |
Pos logit contributions | |
uster | +0.030 |
Sad | +0.030 |
hole | +0.030 |
eping | +0.029 |
Steven | +0.028 |
Neg logit contributions | |
bye | -0.022 |
Sorry | -0.020 |
they | -0.017 |
bye | -0.016 |
every | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | -4.75e-03 |
Pos logit contributions | |
hole | +0.013 |
uster | +0.012 |
Sad | +0.012 |
ist | +0.012 |
ice | +0.012 |
Neg logit contributions | |
Sorry | -0.008 |
bye | -0.007 |
good | -0.007 |
bye | -0.006 |
they | -0.006 |
Token | thought |
Token ID | 1807 |
Feature activation | -7.89e-03 |
Pos logit contributions | |
Sad | +0.009 |
hole | +0.009 |
ice | +0.009 |
uster | +0.008 |
ilda | +0.008 |
Neg logit contributions | |
Sorry | -0.008 |
bye | -0.008 |
good | -0.007 |
day | -0.007 |
bye | -0.007 |
Token | it |
Token ID | 340 |
Feature activation | +1.57e-04 |
Pos logit contributions | |
uster | +0.021 |
Sad | +0.021 |
hole | +0.021 |
ice | +0.020 |
father | +0.020 |
Neg logit contributions | |
Sorry | -0.008 |
bye | -0.007 |
good | -0.007 |
they | -0.007 |
day | -0.006 |
Token | would |
Token ID | 561 |
Feature activation | +7.55e-03 |
Pos logit contributions | |
Sorry | +0.000 |
bye | +0.000 |
bye | +0.000 |
good | +0.000 |
ge | +0.000 |
Neg logit contributions | |
Sad | -0.000 |
hole | -0.000 |
ist | -0.000 |
uster | -0.000 |
ilda | -0.000 |
Token | be |
Token ID | 307 |
Feature activation | +1.38e-03 |
Pos logit contributions | |
Sorry | +0.012 |
bye | +0.012 |
bye | +0.011 |
seals | +0.010 |
ho | +0.010 |
Neg logit contributions | |
Sad | -0.014 |
hole | -0.013 |
ist | -0.012 |
x | -0.012 |
eping | -0.012 |
Token | fun |
Token ID | 1257 |
Feature activation | -8.10e-04 |
Pos logit contributions | |
Sorry | +0.002 |
bye | +0.002 |
ho | +0.002 |
bye | +0.002 |
moms | +0.001 |
Neg logit contributions | |
eping | -0.003 |
hole | -0.003 |
Sad | -0.003 |
pled | -0.003 |
ice | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
hole | +0.002 |
Sad | +0.002 |
uster | +0.002 |
ist | +0.002 |
ilda | +0.002 |
Neg logit contributions | |
Sorry | -0.001 |
bye | -0.001 |
bye | -0.001 |
ho | -0.001 |
good | -0.001 |
Token | swing |
Token ID | 9628 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
Sorry | +0.025 |
bye | +0.021 |
good | +0.021 |
they | +0.020 |
ho | +0.020 |
Neg logit contributions | |
ist | -0.040 |
hole | -0.039 |
Sad | -0.039 |
uster | -0.038 |
eping | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
Sorry | +0.017 |
bye | +0.016 |
bye | +0.014 |
ho | +0.013 |
good | +0.013 |
Neg logit contributions | |
hole | -0.020 |
Sad | -0.019 |
ist | -0.019 |
eping | -0.019 |
uster | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
Sorry | +0.014 |
bye | +0.013 |
they | +0.013 |
good | +0.013 |
day | +0.012 |
Neg logit contributions | |
uster | -0.025 |
hole | -0.025 |
Sad | -0.024 |
ice | -0.024 |
ist | -0.024 |
Token | They |
Token ID | 1119 |
Feature activation | -4.55e-04 |
Pos logit contributions | |
uster | +0.015 |
Sad | +0.015 |
hole | +0.015 |
ice | +0.015 |
father | +0.014 |
Neg logit contributions | |
Sorry | -0.012 |
bye | -0.011 |
good | -0.011 |
they | -0.010 |
day | -0.010 |
Token | listened |
Token ID | 16399 |
Feature activation | +2.07e-02 |
Pos logit contributions | |
hole | +0.001 |
Sad | +0.001 |
ist | +0.001 |
uster | +0.001 |
father | +0.001 |
Neg logit contributions | |
Sorry | -0.001 |
bye | -0.001 |
good | -0.001 |
day | -0.001 |
bye | -0.001 |
Token | carefully |
Token ID | 7773 |
Feature activation | -2.16e-05 |
Pos logit contributions | |
Sorry | +0.035 |
bye | +0.030 |
moms | +0.027 |
ge | +0.026 |
day | +0.026 |
Neg logit contributions | |
hole | -0.041 |
Sad | -0.041 |
uster | -0.041 |
ist | -0.040 |
ilda | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
hole | +0.000 |
uster | +0.000 |
ist | +0.000 |
Sad | +0.000 |
eping | +0.000 |
Neg logit contributions | |
Sorry | -0.000 |
bye | -0.000 |
good | -0.000 |
moms | -0.000 |
day | -0.000 |
Token | just |
Token ID | 655 |
Feature activation | -9.63e-03 |
Pos logit contributions | |
hole | +0.053 |
Sad | +0.053 |
uster | +0.052 |
ice | +0.049 |
ist | +0.049 |
Neg logit contributions | |
Sorry | -0.060 |
bye | -0.057 |
good | -0.053 |
day | -0.051 |
bye | -0.050 |
Token | in |
Token ID | 287 |
Feature activation | -5.02e-03 |
Pos logit contributions | |
hole | +0.019 |
ist | +0.018 |
pled | +0.017 |
Sad | +0.017 |
uster | +0.017 |
Neg logit contributions | |
Sorry | -0.017 |
bye | -0.014 |
moms | -0.014 |
good | -0.013 |
vows | -0.012 |
Token | case |
Token ID | 1339 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
hole | +0.011 |
Sad | +0.011 |
uster | +0.011 |
father | +0.011 |
ice | +0.011 |
Neg logit contributions | |
Sorry | -0.007 |
bye | -0.006 |
bye | -0.005 |
good | -0.005 |
they | -0.005 |
Token | they |
Token ID | 484 |
Feature activation | -4.99e-03 |
Pos logit contributions | |
hole | +0.033 |
uster | +0.033 |
Sad | +0.033 |
ice | +0.032 |
ist | +0.032 |
Neg logit contributions | |
Sorry | -0.014 |
bye | -0.012 |
good | -0.011 |
day | -0.010 |
bye | -0.010 |
Token | could |
Token ID | 714 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
hole | +0.010 |
ist | +0.009 |
father | +0.009 |
uster | +0.009 |
Sad | +0.009 |
Neg logit contributions | |
Sorry | -0.008 |
bye | -0.007 |
good | -0.007 |
bye | -0.007 |
day | -0.006 |
Token | hear |
Token ID | 3285 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
Sad | +0.043 |
hole | +0.043 |
uster | +0.042 |
ist | +0.041 |
father | +0.041 |
Neg logit contributions | |
Sorry | -0.026 |
bye | -0.024 |
good | -0.023 |
bye | -0.022 |
they | -0.021 |
Token | it |
Token ID | 340 |
Feature activation | -2.45e-04 |
Pos logit contributions | |
Sorry | +0.022 |
bye | +0.019 |
ho | +0.018 |
moms | +0.017 |
melon | +0.016 |
Neg logit contributions | |
hole | -0.029 |
ist | -0.028 |
pled | -0.027 |
eping | -0.027 |
Sad | -0.027 |
Token | when |
Token ID | 618 |
Feature activation | +4.76e-03 |
Pos logit contributions | |
Sorry | +0.016 |
bye | +0.015 |
ho | +0.014 |
ge | +0.014 |
seals | +0.013 |
Neg logit contributions | |
ist | -0.019 |
Sad | -0.019 |
ice | -0.019 |
uster | -0.019 |
ilda | -0.019 |
Token | his |
Token ID | 465 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
bye | +0.009 |
Sorry | +0.009 |
ho | +0.008 |
bye | +0.008 |
good | +0.008 |
Neg logit contributions | |
Sad | -0.008 |
father | -0.007 |
James | -0.007 |
uster | -0.007 |
ice | -0.007 |
Token | Dad |
Token ID | 17415 |
Feature activation | -9.73e-03 |
Pos logit contributions | |
Sorry | +0.031 |
bye | +0.029 |
they | +0.028 |
good | +0.027 |
ho | +0.027 |
Neg logit contributions | |
uster | -0.025 |
hole | -0.024 |
father | -0.024 |
Sad | -0.024 |
ice | -0.023 |
Token | came |
Token ID | 1625 |
Feature activation | -9.17e-03 |
Pos logit contributions | |
uster | +0.021 |
Sad | +0.021 |
hole | +0.021 |
ice | +0.020 |
father | +0.020 |
Neg logit contributions | |
Sorry | -0.014 |
bye | -0.013 |
good | -0.013 |
they | -0.012 |
day | -0.012 |
Token | in |
Token ID | 287 |
Feature activation | -8.17e-03 |
Pos logit contributions | |
uster | +0.022 |
Sad | +0.021 |
hole | +0.021 |
ice | +0.020 |
father | +0.020 |
Neg logit contributions | |
Sorry | -0.012 |
bye | -0.011 |
good | -0.010 |
they | -0.010 |
day | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | -8.37e-03 |
Pos logit contributions | |
uster | +0.020 |
Sad | +0.020 |
hole | +0.020 |
ice | +0.019 |
ist | +0.019 |
Neg logit contributions | |
Sorry | -0.009 |
good | -0.009 |
they | -0.009 |
bye | -0.008 |
day | -0.008 |
Token | said |
Token ID | 531 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
Sad | +0.014 |
uster | +0.014 |
hole | +0.014 |
pled | +0.014 |
ilda | +0.014 |
Neg logit contributions | |
bye | -0.015 |
Sorry | -0.014 |
good | -0.013 |
bye | -0.013 |
ho | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | -3.95e-03 |
Pos logit contributions | |
bye | +0.033 |
Sorry | +0.030 |
day | +0.028 |
moms | +0.026 |
sunset | +0.026 |
Neg logit contributions | |
uster | -0.039 |
pled | -0.039 |
Sad | -0.038 |
James | -0.038 |
run | -0.038 |
Token | � |
Token ID | 6184 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
uster | +0.008 |
Sad | +0.008 |
hole | +0.008 |
ice | +0.008 |
ilda | +0.008 |
Neg logit contributions | |
bye | -0.006 |
Sorry | -0.006 |
good | -0.005 |
day | -0.005 |
bye | -0.005 |
Token | � |
Token ID | 95 |
Feature activation | -8.39e-03 |
Pos logit contributions | |
Sorry | +0.025 |
bye | +0.024 |
good | +0.024 |
they | +0.023 |
day | +0.022 |
Neg logit contributions | |
uster | -0.023 |
Sad | -0.023 |
hole | -0.023 |
ice | -0.022 |
father | -0.021 |
Token | € |
Token ID | 26391 |
Feature activation | +2.31e-03 |
Pos logit contributions | |
uster | +0.015 |
hole | +0.015 |
Sad | +0.015 |
pled | +0.014 |
father | +0.014 |
Neg logit contributions | |
bye | -0.015 |
Sorry | -0.014 |
good | -0.014 |
day | -0.013 |
go | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
hole | +0.001 |
uster | +0.001 |
Sad | +0.001 |
ist | +0.001 |
father | +0.001 |
Neg logit contributions | |
Sorry | -0.001 |
bye | -0.001 |
good | -0.001 |
bye | -0.001 |
they | -0.001 |
Token | She |
Token ID | 1375 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
uster | +0.021 |
Sad | +0.021 |
hole | +0.021 |
ice | +0.020 |
father | +0.020 |
Neg logit contributions | |
Sorry | -0.019 |
bye | -0.018 |
good | -0.017 |
they | -0.017 |
day | -0.016 |
Token | reached |
Token ID | 4251 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
uster | +0.022 |
Sad | +0.021 |
hole | +0.021 |
ice | +0.021 |
ist | +0.020 |
Neg logit contributions | |
Sorry | -0.017 |
bye | -0.016 |
good | -0.015 |
they | -0.014 |
day | -0.014 |
Token | out |
Token ID | 503 |
Feature activation | -6.02e-03 |
Pos logit contributions | |
uster | +0.047 |
Sad | +0.047 |
hole | +0.046 |
ice | +0.045 |
father | +0.045 |
Neg logit contributions | |
Sorry | -0.020 |
good | -0.019 |
bye | -0.018 |
they | -0.018 |
day | -0.017 |
Token | with |
Token ID | 351 |
Feature activation | -7.99e-03 |
Pos logit contributions | |
uster | +0.015 |
hole | +0.015 |
Sad | +0.014 |
ice | +0.014 |
father | +0.014 |
Neg logit contributions | |
good | -0.007 |
they | -0.006 |
Sorry | -0.006 |
bye | -0.006 |
day | -0.006 |
Token | her |
Token ID | 607 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
uster | +0.021 |
Sad | +0.020 |
hole | +0.020 |
ice | +0.020 |
father | +0.020 |
Neg logit contributions | |
Sorry | -0.008 |
good | -0.008 |
bye | -0.007 |
they | -0.007 |
day | -0.007 |
Token | tiny |
Token ID | 7009 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
uster | +0.057 |
Sad | +0.057 |
hole | +0.055 |
ice | +0.054 |
father | +0.053 |
Neg logit contributions | |
Sorry | -0.039 |
good | -0.039 |
they | -0.038 |
bye | -0.037 |
day | -0.036 |
Token | hands |
Token ID | 2832 |
Feature activation | -2.82e-03 |
Pos logit contributions | |
Sad | +0.031 |
uster | +0.030 |
hole | +0.029 |
father | +0.029 |
ice | +0.028 |
Neg logit contributions | |
good | -0.024 |
Sorry | -0.024 |
bye | -0.023 |
they | -0.022 |
day | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
hole | +0.006 |
uster | +0.006 |
Sad | +0.006 |
ice | +0.006 |
ist | +0.006 |
Neg logit contributions | |
Sorry | -0.004 |
bye | -0.004 |
good | -0.004 |
day | -0.004 |
they | -0.004 |
Token | wrapped |
Token ID | 12908 |
Feature activation | -5.79e-03 |
Pos logit contributions | |
Sad | +0.018 |
hole | +0.018 |
uster | +0.017 |
ice | +0.017 |
pled | +0.016 |
Neg logit contributions | |
Sorry | -0.020 |
bye | -0.019 |
good | -0.017 |
day | -0.017 |
bye | -0.016 |
Token | it |
Token ID | 340 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
uster | +0.014 |
Sad | +0.014 |
hole | +0.014 |
ice | +0.013 |
father | +0.013 |
Neg logit contributions | |
Sorry | -0.007 |
bye | -0.006 |
good | -0.006 |
they | -0.006 |
day | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | +2.23e-03 |
Pos logit contributions | |
Sorry | +0.025 |
bye | +0.023 |
bye | +0.020 |
ho | +0.019 |
good | +0.017 |
Neg logit contributions | |
Sad | -0.040 |
hole | -0.040 |
ice | -0.039 |
uster | -0.038 |
father | -0.038 |
Token | small |
Token ID | 1402 |
Feature activation | -2.37e-03 |
Pos logit contributions | |
Sorry | +0.003 |
bye | +0.003 |
good | +0.002 |
they | +0.002 |
bye | +0.002 |
Neg logit contributions | |
uster | -0.005 |
Sad | -0.005 |
hole | -0.005 |
ice | -0.005 |
father | -0.005 |
Token | one |
Token ID | 530 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
uster | +0.005 |
Sad | +0.005 |
hole | +0.005 |
ice | +0.004 |
father | +0.004 |
Neg logit contributions | |
Sorry | -0.004 |
bye | -0.004 |
good | -0.004 |
they | -0.003 |
day | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +3.84e-03 |
Pos logit contributions | |
hole | +0.028 |
uster | +0.025 |
ist | +0.025 |
ice | +0.025 |
father | +0.024 |
Neg logit contributions | |
Sorry | -0.025 |
bye | -0.023 |
bye | -0.021 |
ge | -0.020 |
good | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | +7.52e-03 |
Pos logit contributions | |
Sorry | +0.004 |
good | +0.004 |
bye | +0.004 |
they | +0.004 |
day | +0.004 |
Neg logit contributions | |
uster | -0.009 |
hole | -0.009 |
Sad | -0.009 |
ice | -0.009 |
father | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
they | +0.007 |
good | +0.006 |
day | +0.006 |
Sorry | +0.005 |
bye | +0.005 |
Neg logit contributions | |
uster | -0.021 |
father | -0.021 |
hole | -0.021 |
ice | -0.021 |
ist | -0.021 |
Token | " |
Token ID | 1 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
uster | +0.031 |
hole | +0.031 |
Sad | +0.031 |
father | +0.030 |
ice | +0.030 |
Neg logit contributions | |
Sorry | -0.022 |
good | -0.021 |
they | -0.021 |
bye | -0.021 |
day | -0.020 |
Token | Look |
Token ID | 8567 |
Feature activation | +4.51e-03 |
Pos logit contributions | |
Sorry | +0.024 |
bye | +0.022 |
good | +0.022 |
they | +0.021 |
bye | +0.021 |
Neg logit contributions | |
Sad | -0.031 |
uster | -0.030 |
hole | -0.029 |
ice | -0.029 |
James | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | +2.42e-02 |
Pos logit contributions | |
bye | +0.005 |
Sorry | +0.005 |
day | +0.005 |
good | +0.005 |
ho | +0.004 |
Neg logit contributions | |
uster | -0.011 |
hole | -0.010 |
ice | -0.010 |
Sad | -0.010 |
ist | -0.010 |
Token | Lily |
Token ID | 20037 |
Feature activation | +9.00e-04 |
Pos logit contributions | |
bye | +0.047 |
ho | +0.042 |
Sorry | +0.038 |
day | +0.038 |
became | +0.034 |
Neg logit contributions | |
pled | -0.037 |
uster | -0.037 |
ice | -0.037 |
Sad | -0.036 |
hole | -0.035 |
Token | , |
Token ID | 11 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
Sorry | +0.001 |
bye | +0.001 |
good | +0.001 |
they | +0.001 |
day | +0.001 |
Neg logit contributions | |
uster | -0.002 |
Sad | -0.002 |
hole | -0.002 |
ice | -0.002 |
father | -0.002 |
Token | had |
Token ID | 550 |
Feature activation | -1.28e-03 |
Pos logit contributions | |
uster | +0.012 |
Sad | +0.012 |
hole | +0.012 |
ice | +0.012 |
father | +0.011 |
Neg logit contributions | |
Sorry | -0.008 |
bye | -0.007 |
good | -0.007 |
they | -0.007 |
bye | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
uster | +0.003 |
Sad | +0.003 |
hole | +0.003 |
ice | +0.003 |
father | +0.003 |
Neg logit contributions | |
Sorry | -0.002 |
bye | -0.002 |
good | -0.002 |
they | -0.001 |
day | -0.001 |
Token | nasty |
Token ID | 17166 |
Feature activation | -4.61e-03 |
Pos logit contributions | |
uster | +0.022 |
Sad | +0.022 |
hole | +0.022 |
ice | +0.021 |
father | +0.021 |
Neg logit contributions | |
Sorry | -0.016 |
bye | -0.016 |
they | -0.014 |
good | -0.014 |
bye | -0.013 |
Token | cold |
Token ID | 4692 |
Feature activation | -1.33e-03 |
Pos logit contributions | |
uster | +0.009 |
hole | +0.009 |
Sad | +0.009 |
ice | +0.009 |
father | +0.009 |
Neg logit contributions | |
Sorry | -0.007 |
bye | -0.007 |
they | -0.007 |
good | -0.007 |
day | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
uster | +0.003 |
hole | +0.003 |
Sad | +0.003 |
ice | +0.003 |
ist | +0.003 |
Neg logit contributions | |
Sorry | -0.002 |
bye | -0.002 |
good | -0.002 |
they | -0.002 |
bye | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
Sad | +0.023 |
ice | +0.023 |
uster | +0.022 |
hole | +0.022 |
pled | +0.022 |
Neg logit contributions | |
Sorry | -0.016 |
bye | -0.015 |
bye | -0.013 |
ho | -0.013 |
good | -0.013 |
Token | feeling |
Token ID | 4203 |
Feature activation | +3.72e-03 |
Pos logit contributions | |
Sad | +0.028 |
uster | +0.028 |
hole | +0.027 |
ice | +0.027 |
eping | +0.026 |
Neg logit contributions | |
Sorry | -0.024 |
bye | -0.023 |
good | -0.021 |
bye | -0.020 |
they | -0.020 |
Token | very |
Token ID | 845 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
Sorry | +0.005 |
bye | +0.005 |
bye | +0.004 |
ho | +0.004 |
moms | +0.004 |
Neg logit contributions | |
uster | -0.008 |
ice | -0.008 |
Sad | -0.008 |
pled | -0.008 |
hole | -0.008 |
Token | un |
Token ID | 555 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
bye | +0.043 |
Sorry | +0.042 |
ho | +0.036 |
moms | +0.035 |
they | +0.035 |
Neg logit contributions | |
ivering | -0.045 |
ice | -0.044 |
uster | -0.044 |
Sad | -0.042 |
eping | -0.042 |
Token | well |
Token ID | 4053 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
hole | +0.015 |
uster | +0.014 |
Sad | +0.014 |
ice | +0.013 |
father | +0.013 |
Neg logit contributions | |
Sorry | -0.021 |
bye | -0.020 |
good | -0.020 |
they | -0.019 |
ho | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
Sad | +0.094 |
uster | +0.092 |
hole | +0.089 |
ice | +0.087 |
father | +0.087 |
Neg logit contributions | |
good | -0.033 |
they | -0.031 |
bye | -0.031 |
day | -0.030 |
Sorry | -0.030 |
Token | a |
Token ID | 257 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
hole | +0.003 |
eping | +0.003 |
Sad | +0.003 |
pled | +0.002 |
uster | +0.002 |
Neg logit contributions | |
Sorry | -0.002 |
bye | -0.002 |
day | -0.002 |
they | -0.001 |
ho | -0.001 |
Token | shiny |
Token ID | 22441 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
hole | +0.044 |
uster | +0.044 |
Sad | +0.044 |
eping | +0.042 |
ist | +0.041 |
Neg logit contributions | |
Sorry | -0.025 |
bye | -0.025 |
they | -0.022 |
good | -0.021 |
day | -0.021 |
Token | thing |
Token ID | 1517 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
uster | +0.064 |
Sad | +0.064 |
hole | +0.063 |
ice | +0.061 |
father | +0.060 |
Neg logit contributions | |
Sorry | -0.038 |
bye | -0.036 |
good | -0.036 |
they | -0.034 |
day | -0.033 |
Token | in |
Token ID | 287 |
Feature activation | -3.04e-03 |
Pos logit contributions | |
hole | +0.023 |
pled | +0.020 |
ist | +0.020 |
uster | +0.020 |
eping | +0.019 |
Neg logit contributions | |
Sorry | -0.018 |
bye | -0.017 |
good | -0.015 |
bye | -0.014 |
day | -0.014 |
Token | its |
Token ID | 663 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
hole | +0.007 |
Sad | +0.007 |
uster | +0.006 |
ice | +0.006 |
eping | +0.006 |
Neg logit contributions | |
Sorry | -0.004 |
bye | -0.004 |
good | -0.004 |
bye | -0.004 |
day | -0.004 |
Token | be |
Token ID | 307 |
Feature activation | -4.16e-02 |
Pos logit contributions | |
hole | +0.053 |
uster | +0.049 |
Sad | +0.049 |
ady | +0.047 |
father | +0.047 |
Neg logit contributions | |
Sorry | -0.045 |
they | -0.040 |
bye | -0.040 |
good | -0.039 |
day | -0.038 |
Token | ak |
Token ID | 461 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
uster | +0.028 |
Sad | +0.027 |
hole | +0.025 |
ice | +0.022 |
father | +0.021 |
Neg logit contributions | |
Sorry | -0.120 |
bye | -0.118 |
good | -0.118 |
they | -0.116 |
day | -0.115 |
Token | . |
Token ID | 13 |
Feature activation | -4.50e-03 |
Pos logit contributions | |
hole | +0.021 |
uster | +0.019 |
ist | +0.019 |
ice | +0.019 |
pled | +0.019 |
Neg logit contributions | |
Sorry | -0.018 |
bye | -0.015 |
good | -0.015 |
bye | -0.014 |
they | -0.014 |
Token | It |
Token ID | 632 |
Feature activation | -1.11e-03 |
Pos logit contributions | |
uster | +0.009 |
Sad | +0.009 |
hole | +0.009 |
ice | +0.009 |
father | +0.009 |
Neg logit contributions | |
Sorry | -0.007 |
bye | -0.007 |
good | -0.007 |
they | -0.007 |
day | -0.006 |
Token | looked |
Token ID | 3114 |
Feature activation | +9.80e-03 |
Pos logit contributions | |
hole | +0.002 |
Sad | +0.002 |
ice | +0.002 |
pled | +0.002 |
eping | +0.002 |
Neg logit contributions | |
Sorry | -0.002 |
bye | -0.002 |
good | -0.002 |
bye | -0.002 |
day | -0.002 |
Token | like |
Token ID | 588 |
Feature activation | +3.58e-03 |
Pos logit contributions | |
Sorry | +0.015 |
bye | +0.015 |
day | +0.012 |
ho | +0.012 |
moms | +0.012 |
Neg logit contributions | |
hole | -0.020 |
eping | -0.019 |
uster | -0.019 |
pled | -0.019 |
Sad | -0.018 |
Token | of |
Token ID | 286 |
Feature activation | -5.26e-03 |
Pos logit contributions | |
uster | +0.006 |
Sad | +0.006 |
hole | +0.006 |
ice | +0.005 |
father | +0.005 |
Neg logit contributions | |
Sorry | -0.003 |
bye | -0.003 |
good | -0.003 |
they | -0.003 |
day | -0.003 |
Token | something |
Token ID | 1223 |
Feature activation | -7.76e-03 |
Pos logit contributions | |
uster | +0.011 |
Sad | +0.011 |
ice | +0.011 |
hole | +0.011 |
pled | +0.011 |
Neg logit contributions | |
Sorry | -0.008 |
bye | -0.007 |
they | -0.006 |
bye | -0.006 |
ho | -0.006 |
Token | strange |
Token ID | 6283 |
Feature activation | -6.46e-03 |
Pos logit contributions | |
hole | +0.015 |
pled | +0.015 |
ice | +0.014 |
eping | +0.014 |
Sad | +0.013 |
Neg logit contributions | |
bye | -0.014 |
Sorry | -0.013 |
bye | -0.011 |
moms | -0.010 |
ho | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +6.40e-03 |
Pos logit contributions | |
hole | +0.013 |
pled | +0.012 |
ist | +0.012 |
eping | +0.012 |
ice | +0.012 |
Neg logit contributions | |
Sorry | -0.011 |
bye | -0.010 |
good | -0.009 |
bye | -0.009 |
moms | -0.009 |
Token | She |
Token ID | 1375 |
Feature activation | -9.38e-03 |
Pos logit contributions | |
Sorry | +0.009 |
bye | +0.009 |
good | +0.009 |
they | +0.009 |
day | +0.008 |
Neg logit contributions | |
uster | -0.014 |
hole | -0.014 |
Sad | -0.014 |
ist | -0.013 |
ice | -0.013 |
Token | couldn |
Token ID | 3521 |
Feature activation | -4.14e-02 |
Pos logit contributions | |
uster | +0.020 |
Sad | +0.020 |
hole | +0.020 |
ice | +0.019 |
father | +0.019 |
Neg logit contributions | |
Sorry | -0.013 |
bye | -0.013 |
good | -0.012 |
they | -0.012 |
day | -0.011 |
Token | 't |
Token ID | 470 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
uster | +0.082 |
hole | +0.080 |
Sad | +0.078 |
ist | +0.076 |
ice | +0.075 |
Neg logit contributions | |
Sorry | -0.068 |
they | -0.066 |
good | -0.065 |
bye | -0.064 |
go | -0.062 |
Token | identify |
Token ID | 5911 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
uster | +0.043 |
hole | +0.042 |
Sad | +0.042 |
ice | +0.042 |
pled | +0.040 |
Neg logit contributions | |
good | -0.021 |
bye | -0.020 |
they | -0.020 |
Sorry | -0.020 |
go | -0.020 |
Token | what |
Token ID | 644 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
uster | +0.059 |
Sad | +0.058 |
hole | +0.056 |
father | +0.054 |
ist | +0.054 |
Neg logit contributions | |
good | -0.023 |
Sorry | -0.022 |
they | -0.022 |
bye | -0.020 |
day | -0.020 |
Token | was |
Token ID | 373 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
bye | +0.021 |
Sorry | +0.019 |
bye | +0.018 |
ho | +0.016 |
seals | +0.015 |
Neg logit contributions | |
pled | -0.023 |
Sad | -0.022 |
ice | -0.022 |
hole | -0.021 |
father | -0.021 |
Token | inside |
Token ID | 2641 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
hole | +0.069 |
uster | +0.068 |
Sad | +0.067 |
ice | +0.065 |
ist | +0.064 |
Neg logit contributions | |
Sorry | -0.053 |
bye | -0.050 |
good | -0.048 |
they | -0.046 |
bye | -0.045 |
Token | h |
Token ID | 289 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
uster | +0.042 |
hole | +0.042 |
Sad | +0.042 |
ice | +0.039 |
eping | +0.039 |
Neg logit contributions | |
Sorry | -0.035 |
bye | -0.035 |
they | -0.032 |
good | -0.031 |
day | -0.031 |
Token | alo |
Token ID | 7335 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
Sad | +0.033 |
ice | +0.031 |
father | +0.030 |
James | +0.029 |
Steven | +0.029 |
Neg logit contributions | |
bye | -0.028 |
bye | -0.027 |
Sorry | -0.026 |
ho | -0.026 |
they | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +7.19e-03 |
Pos logit contributions | |
hole | +0.032 |
Sad | +0.030 |
uster | +0.030 |
pled | +0.030 |
ist | +0.030 |
Neg logit contributions | |
Sorry | -0.021 |
bye | -0.018 |
good | -0.018 |
bye | -0.016 |
day | -0.016 |
Token | He |
Token ID | 679 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
Sorry | +0.011 |
they | +0.011 |
good | +0.011 |
bye | +0.010 |
day | +0.010 |
Neg logit contributions | |
hole | -0.015 |
uster | -0.015 |
Sad | -0.015 |
ice | -0.014 |
ist | -0.014 |
Token | was |
Token ID | 373 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
Sad | +0.029 |
hole | +0.029 |
uster | +0.029 |
ice | +0.028 |
ist | +0.027 |
Neg logit contributions | |
Sorry | -0.020 |
bye | -0.019 |
good | -0.018 |
day | -0.017 |
they | -0.017 |
Token | an |
Token ID | 281 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
uster | +0.022 |
Sad | +0.022 |
hole | +0.022 |
eping | +0.021 |
ice | +0.021 |
Neg logit contributions | |
Sorry | -0.019 |
bye | -0.019 |
good | -0.016 |
they | -0.016 |
day | -0.016 |
Token | angel |
Token ID | 18304 |
Feature activation | -4.73e-03 |
Pos logit contributions | |
Sad | +0.101 |
uster | +0.096 |
hole | +0.094 |
ist | +0.091 |
James | +0.090 |
Neg logit contributions | |
good | -0.054 |
bye | -0.054 |
day | -0.051 |
bye | -0.049 |
go | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | +6.31e-04 |
Pos logit contributions | |
hole | +0.010 |
Sad | +0.009 |
uster | +0.009 |
ist | +0.009 |
ice | +0.009 |
Neg logit contributions | |
Sorry | -0.007 |
good | -0.007 |
bye | -0.007 |
bye | -0.006 |
they | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
Sorry | +0.001 |
they | +0.001 |
good | +0.001 |
bye | +0.001 |
day | +0.001 |
Neg logit contributions | |
hole | -0.002 |
uster | -0.001 |
Sad | -0.001 |
ist | -0.001 |
ice | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
they | +0.011 |
good | +0.010 |
Sorry | +0.009 |
bye | +0.009 |
day | +0.008 |
Neg logit contributions | |
uster | -0.035 |
hole | -0.034 |
ice | -0.034 |
ist | -0.034 |
ivering | -0.033 |
Token | " |
Token ID | 1 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
uster | +0.028 |
hole | +0.028 |
Sad | +0.028 |
ice | +0.027 |
father | +0.026 |
Neg logit contributions | |
Sorry | -0.018 |
bye | -0.018 |
good | -0.017 |
they | -0.017 |
day | -0.016 |
Token | what |
Token ID | 644 |
Feature activation | +6.70e-03 |
Pos logit contributions | |
hole | +0.052 |
uster | +0.052 |
Sad | +0.052 |
ice | +0.049 |
father | +0.049 |
Neg logit contributions | |
Sorry | -0.038 |
bye | -0.037 |
good | -0.034 |
they | -0.032 |
day | -0.032 |
Token | an |
Token ID | 281 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
Sorry | +0.008 |
bye | +0.008 |
good | +0.007 |
bye | +0.007 |
ho | +0.006 |
Neg logit contributions | |
Sad | -0.016 |
hole | -0.016 |
uster | -0.016 |
father | -0.015 |
ice | -0.015 |
Token | X |
Token ID | 1395 |
Feature activation | -4.02e-02 |
Pos logit contributions | |
Sad | +0.038 |
uster | +0.038 |
hole | +0.037 |
ice | +0.035 |
pled | +0.035 |
Neg logit contributions | |
bye | -0.021 |
good | -0.021 |
day | -0.020 |
Sorry | -0.020 |
go | -0.019 |
Token | - |
Token ID | 12 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
uster | +0.093 |
Sad | +0.092 |
hole | +0.091 |
ice | +0.088 |
father | +0.086 |
Neg logit contributions | |
Sorry | -0.051 |
bye | -0.048 |
good | -0.047 |
they | -0.046 |
day | -0.044 |
Token | ray |
Token ID | 2433 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
uster | +0.019 |
Sad | +0.018 |
hole | +0.018 |
ice | +0.017 |
father | +0.016 |
Neg logit contributions | |
Sorry | -0.026 |
bye | -0.026 |
good | -0.026 |
they | -0.025 |
day | -0.025 |
Token | was |
Token ID | 373 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
hole | +0.050 |
uster | +0.048 |
Sad | +0.048 |
ice | +0.047 |
father | +0.047 |
Neg logit contributions | |
Sorry | -0.034 |
bye | -0.031 |
good | -0.030 |
bye | -0.028 |
they | -0.027 |
Token | so |
Token ID | 523 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
uster | +0.105 |
Sad | +0.104 |
hole | +0.099 |
ice | +0.097 |
innie | +0.096 |
Neg logit contributions | |
Sorry | -0.049 |
good | -0.049 |
bye | -0.048 |
they | -0.047 |
day | -0.046 |
Token | she |
Token ID | 673 |
Feature activation | -6.06e-03 |
Pos logit contributions | |
Sorry | +0.015 |
bye | +0.015 |
bye | +0.013 |
seals | +0.013 |
ho | +0.012 |
Neg logit contributions | |
Sad | -0.021 |
hole | -0.020 |
ilda | -0.020 |
uster | -0.020 |
ice | -0.020 |
Token | asked |
Token ID | 1965 |
Feature activation | -4.30e-03 |
Pos logit contributions | |
uster | +0.014 |
Sad | +0.014 |
hole | +0.014 |
ice | +0.013 |
father | +0.013 |
Neg logit contributions | |
Sorry | -0.008 |
bye | -0.007 |
good | -0.007 |
they | -0.007 |
day | -0.007 |
Token | her |
Token ID | 607 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
Sad | +0.009 |
hole | +0.009 |
uster | +0.008 |
ice | +0.008 |
father | +0.008 |
Neg logit contributions | |
Sorry | -0.007 |
bye | -0.006 |
good | -0.006 |
bye | -0.005 |
day | -0.005 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
Sorry | +0.023 |
bye | +0.023 |
ho | +0.019 |
floats | +0.018 |
bye | +0.018 |
Neg logit contributions | |
father | -0.030 |
Sad | -0.030 |
ilda | -0.029 |
innie | -0.028 |
hole | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -9.37e-03 |
Pos logit contributions | |
hole | +0.027 |
father | +0.026 |
ice | +0.025 |
Sad | +0.025 |
uster | +0.025 |
Neg logit contributions | |
Sorry | -0.024 |
bye | -0.022 |
good | -0.020 |
bye | -0.019 |
they | -0.019 |
Token | He |
Token ID | 679 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
uster | +0.018 |
Sad | +0.017 |
hole | +0.017 |
ice | +0.016 |
father | +0.016 |
Neg logit contributions | |
Sorry | -0.016 |
bye | -0.015 |
good | -0.015 |
they | -0.015 |
day | -0.014 |
Token | went |
Token ID | 1816 |
Feature activation | +5.02e-03 |
Pos logit contributions | |
Sad | +0.030 |
uster | +0.030 |
hole | +0.030 |
ice | +0.030 |
ist | +0.029 |
Neg logit contributions | |
Sorry | -0.025 |
bye | -0.023 |
good | -0.022 |
day | -0.021 |
they | -0.021 |
Token | to |
Token ID | 284 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
Sorry | +0.005 |
bye | +0.005 |
good | +0.005 |
they | +0.005 |
day | +0.004 |
Neg logit contributions | |
uster | -0.013 |
Sad | -0.013 |
hole | -0.013 |
ice | -0.012 |
father | -0.012 |
Token | his |
Token ID | 465 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
Sorry | +0.013 |
bye | +0.012 |
good | +0.012 |
they | +0.011 |
day | +0.011 |
Neg logit contributions | |
uster | -0.025 |
Sad | -0.025 |
hole | -0.025 |
ice | -0.024 |
father | -0.024 |
Token | mom |
Token ID | 1995 |
Feature activation | -4.51e-02 |
Pos logit contributions | |
hole | +0.027 |
father | +0.027 |
Sad | +0.026 |
uster | +0.026 |
ice | +0.025 |
Neg logit contributions | |
Sorry | -0.033 |
bye | -0.031 |
they | -0.029 |
ho | -0.028 |
good | -0.027 |
Token | 's |
Token ID | 338 |
Feature activation | -9.14e-03 |
Pos logit contributions | |
Sad | +0.102 |
uster | +0.101 |
hole | +0.094 |
innie | +0.093 |
eping | +0.092 |
Neg logit contributions | |
they | -0.060 |
day | -0.058 |
go | -0.058 |
good | -0.056 |
bye | -0.054 |
Token | needle |
Token ID | 17598 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
hole | +0.014 |
uster | +0.014 |
Sad | +0.014 |
father | +0.013 |
ice | +0.013 |
Neg logit contributions | |
Sorry | -0.019 |
bye | -0.018 |
they | -0.017 |
good | -0.017 |
bye | -0.016 |
Token | work |
Token ID | 1818 |
Feature activation | +4.41e-04 |
Pos logit contributions | |
Sad | +0.051 |
uster | +0.051 |
hole | +0.049 |
ice | +0.048 |
innie | +0.048 |
Neg logit contributions | |
they | -0.031 |
good | -0.031 |
bye | -0.031 |
Sorry | -0.030 |
day | -0.030 |
Token | box |
Token ID | 3091 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
Sorry | +0.001 |
bye | +0.001 |
good | +0.001 |
bye | +0.001 |
they | +0.001 |
Neg logit contributions | |
hole | -0.001 |
uster | -0.001 |
ist | -0.001 |
Sad | -0.001 |
ice | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | -6.58e-03 |
Pos logit contributions | |
Sad | +0.067 |
uster | +0.067 |
hole | +0.064 |
ice | +0.063 |
father | +0.063 |
Neg logit contributions | |
they | -0.037 |
bye | -0.036 |
good | -0.036 |
Sorry | -0.036 |
day | -0.034 |