Token | together |
Token ID | 1978 |
Feature activation | +3.31e-01 |
Pos logit contributions | |
Spot | +0.004 |
girls | +0.004 |
Zip | +0.004 |
Buddy | +0.004 |
Max | +0.004 |
Neg logit contributions | |
canvas | -0.011 |
line | -0.011 |
tide | -0.011 |
pulp | -0.011 |
product | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -3.81e-01 |
Pos logit contributions | |
Spot | +0.011 |
girls | +0.010 |
Buddy | +0.010 |
Zip | +0.010 |
Wh | +0.010 |
Neg logit contributions | |
canvas | -0.012 |
line | -0.011 |
product | -0.011 |
tide | -0.011 |
lee | -0.011 |
Token | They |
Token ID | 1119 |
Feature activation | -4.99e-01 |
Pos logit contributions | |
lee | +0.015 |
canvas | +0.015 |
line | +0.015 |
harbor | +0.015 |
product | +0.014 |
Neg logit contributions | |
girls | -0.010 |
oof | -0.010 |
uddy | -0.009 |
Zip | -0.009 |
ver | -0.009 |
Token | played |
Token ID | 2826 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
harbor | +0.017 |
canvas | +0.017 |
lee | +0.017 |
line | +0.016 |
shell | +0.016 |
Neg logit contributions | |
uddy | -0.016 |
oof | -0.016 |
girls | -0.015 |
Zip | -0.015 |
Johnson | -0.015 |
Token | hop |
Token ID | 1725 |
Feature activation | +8.07e-01 |
Pos logit contributions | |
Johnson | +0.042 |
uddy | +0.041 |
Understand | +0.038 |
girls | +0.037 |
oof | +0.036 |
Neg logit contributions | |
harbor | -0.049 |
shell | -0.047 |
line | -0.047 |
tide | -0.047 |
canvas | -0.046 |
Token | sc |
Token ID | 1416 |
Feature activation | +4.98e+00 |
Pos logit contributions | |
Okay | +0.030 |
OK | +0.029 |
Understand | +0.028 |
Deal | +0.028 |
Johnson | +0.028 |
Neg logit contributions | |
harbor | -0.024 |
shell | -0.023 |
line | -0.023 |
cliff | -0.022 |
tide | -0.021 |
Token | ot |
Token ID | 313 |
Feature activation | -8.27e-01 |
Pos logit contributions | |
Spot | +0.224 |
girls | +0.219 |
Buddy | +0.217 |
Zip | +0.216 |
Max | +0.213 |
Neg logit contributions | |
canvas | -0.111 |
line | -0.107 |
product | -0.106 |
tide | -0.103 |
lee | -0.097 |
Token | ch |
Token ID | 354 |
Feature activation | +2.42e+00 |
Pos logit contributions | |
canvas | +0.035 |
line | +0.034 |
product | +0.033 |
tide | +0.033 |
lee | +0.033 |
Neg logit contributions | |
girls | -0.019 |
Spot | -0.019 |
Johnson | -0.019 |
Buddy | -0.018 |
Monkey | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | +6.28e-01 |
Pos logit contributions | |
girls | +0.066 |
Spot | +0.066 |
Johnson | +0.065 |
Monkey | +0.065 |
Buddy | +0.062 |
Neg logit contributions | |
line | -0.089 |
canvas | -0.088 |
product | -0.086 |
tide | -0.085 |
lee | -0.084 |
Token | picked |
Token ID | 6497 |
Feature activation | -5.53e-01 |
Pos logit contributions | |
Deal | +0.023 |
oof | +0.023 |
Understand | +0.023 |
OK | +0.022 |
ver | +0.022 |
Neg logit contributions | |
line | -0.018 |
canvas | -0.018 |
harbor | -0.017 |
shell | -0.017 |
tide | -0.016 |
Token | more |
Token ID | 517 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
canvas | +0.018 |
tide | +0.018 |
lee | +0.018 |
shell | +0.017 |
product | +0.017 |
Neg logit contributions | |
uddy | -0.018 |
girls | -0.018 |
Spot | -0.017 |
Buddy | -0.017 |
Johnson | -0.017 |
Token | love |
Token ID | 1842 |
Feature activation | +4.85e-01 |
Pos logit contributions | |
uddy | +0.027 |
girls | +0.025 |
ver | +0.025 |
oof | +0.025 |
isk | +0.025 |
Neg logit contributions | |
tide | -0.021 |
canvas | -0.021 |
harbor | -0.020 |
shell | -0.020 |
line | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -3.23e-01 |
Pos logit contributions | |
girls | +0.014 |
uddy | +0.014 |
oof | +0.014 |
Zip | +0.014 |
Monkey | +0.014 |
Neg logit contributions | |
canvas | -0.018 |
line | -0.018 |
product | -0.018 |
tide | -0.017 |
shell | -0.017 |
Token | They |
Token ID | 1119 |
Feature activation | -6.59e-01 |
Pos logit contributions | |
line | +0.013 |
canvas | +0.013 |
lee | +0.012 |
tide | +0.012 |
product | +0.012 |
Neg logit contributions | |
girls | -0.009 |
Spot | -0.009 |
oof | -0.008 |
Zip | -0.008 |
uddy | -0.008 |
Token | played |
Token ID | 2826 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
line | +0.021 |
harbor | +0.021 |
canvas | +0.021 |
lee | +0.020 |
tide | +0.020 |
Neg logit contributions | |
oof | -0.023 |
girls | -0.022 |
uddy | -0.022 |
Zip | -0.022 |
Buddy | -0.022 |
Token | hop |
Token ID | 1725 |
Feature activation | +5.77e-01 |
Pos logit contributions | |
uddy | +0.040 |
Johnson | +0.038 |
girls | +0.037 |
Monkey | +0.036 |
oof | +0.036 |
Neg logit contributions | |
line | -0.041 |
tide | -0.041 |
harbor | -0.040 |
canvas | -0.038 |
shell | -0.037 |
Token | sc |
Token ID | 1416 |
Feature activation | +4.69e+00 |
Pos logit contributions | |
Okay | +0.020 |
oof | +0.020 |
girls | +0.020 |
Understand | +0.019 |
Johnson | +0.019 |
Neg logit contributions | |
line | -0.018 |
harbor | -0.018 |
tide | -0.017 |
cliff | -0.017 |
shell | -0.017 |
Token | ot |
Token ID | 313 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
Spot | +0.224 |
girls | +0.224 |
Buddy | +0.218 |
Zip | +0.218 |
Monkey | +0.218 |
Neg logit contributions | |
canvas | -0.088 |
product | -0.084 |
line | -0.083 |
lee | -0.078 |
tide | -0.077 |
Token | ch |
Token ID | 354 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
line | +0.053 |
canvas | +0.052 |
product | +0.051 |
tide | +0.050 |
lee | +0.050 |
Neg logit contributions | |
Johnson | -0.031 |
girls | -0.030 |
Spot | -0.029 |
Deal | -0.029 |
Kitty | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | +2.76e-01 |
Pos logit contributions | |
Johnson | +0.061 |
Monkey | +0.060 |
girls | +0.060 |
Spot | +0.059 |
Deal | +0.057 |
Neg logit contributions | |
line | -0.075 |
canvas | -0.071 |
tide | -0.070 |
product | -0.070 |
lee | -0.068 |
Token | jumped |
Token ID | 11687 |
Feature activation | -3.13e-01 |
Pos logit contributions | |
oof | +0.009 |
girls | +0.008 |
uddy | +0.008 |
Johnson | +0.008 |
Deal | +0.008 |
Neg logit contributions | |
line | -0.010 |
canvas | -0.010 |
tide | -0.010 |
harbor | -0.009 |
shell | -0.009 |
Token | rope |
Token ID | 17182 |
Feature activation | +7.22e-01 |
Pos logit contributions | |
tide | +0.012 |
line | +0.011 |
canvas | +0.011 |
harbor | +0.011 |
shell | +0.011 |
Neg logit contributions | |
girls | -0.009 |
oof | -0.009 |
Deal | -0.009 |
uddy | -0.009 |
Monkey | -0.009 |
Token | day |
Token ID | 1110 |
Feature activation | -3.76e-01 |
Pos logit contributions | |
Spot | +0.010 |
Buddy | +0.009 |
Zip | +0.009 |
girls | +0.009 |
Max | +0.009 |
Neg logit contributions | |
canvas | -0.018 |
line | -0.018 |
product | -0.018 |
tide | -0.018 |
lee | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | -6.32e-01 |
Pos logit contributions | |
canvas | +0.016 |
line | +0.016 |
product | +0.016 |
tide | +0.015 |
lee | +0.015 |
Neg logit contributions | |
Spot | -0.009 |
girls | -0.009 |
Buddy | -0.009 |
Zip | -0.009 |
Wh | -0.008 |
Token | they |
Token ID | 484 |
Feature activation | -6.16e-01 |
Pos logit contributions | |
tide | +0.024 |
canvas | +0.023 |
line | +0.023 |
product | +0.022 |
pumpkin | +0.021 |
Neg logit contributions | |
girls | -0.020 |
isk | -0.018 |
Monkey | -0.018 |
uddy | -0.018 |
oof | -0.018 |
Token | play |
Token ID | 711 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
product | +0.024 |
canvas | +0.023 |
line | +0.023 |
lee | +0.022 |
tide | +0.022 |
Neg logit contributions | |
Spot | -0.018 |
Max | -0.017 |
girls | -0.017 |
Buddy | -0.017 |
Zip | -0.016 |
Token | hop |
Token ID | 1725 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
girls | +0.017 |
Johnson | +0.017 |
uddy | +0.016 |
Kitty | +0.016 |
oof | +0.015 |
Neg logit contributions | |
canvas | -0.021 |
tide | -0.021 |
harbor | -0.020 |
product | -0.020 |
shell | -0.020 |
Token | sc |
Token ID | 1416 |
Feature activation | +4.66e+00 |
Pos logit contributions | |
Johnson | +0.000 |
girls | +0.000 |
oof | +0.000 |
Kitty | +0.000 |
uddy | +0.000 |
Neg logit contributions | |
tide | -0.000 |
line | -0.000 |
harbor | -0.000 |
shell | -0.000 |
canvas | -0.000 |
Token | ot |
Token ID | 313 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
Spot | +0.204 |
girls | +0.198 |
Buddy | +0.197 |
Zip | +0.196 |
Max | +0.193 |
Neg logit contributions | |
canvas | -0.111 |
line | -0.107 |
product | -0.106 |
tide | -0.103 |
lee | -0.097 |
Token | ch |
Token ID | 354 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
canvas | +0.059 |
line | +0.059 |
product | +0.058 |
tide | +0.057 |
lee | +0.055 |
Neg logit contributions | |
Johnson | -0.036 |
girls | -0.036 |
Spot | -0.035 |
Kitty | -0.034 |
oof | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -8.05e-02 |
Pos logit contributions | |
girls | +0.054 |
Spot | +0.054 |
Johnson | +0.052 |
Monkey | +0.052 |
Buddy | +0.051 |
Neg logit contributions | |
line | -0.068 |
canvas | -0.066 |
product | -0.066 |
tide | -0.065 |
lee | -0.062 |
Token | They |
Token ID | 1119 |
Feature activation | -4.71e-01 |
Pos logit contributions | |
canvas | +0.004 |
line | +0.003 |
tide | +0.003 |
product | +0.003 |
lee | +0.003 |
Neg logit contributions | |
girls | -0.002 |
Buddy | -0.002 |
Spot | -0.002 |
Monkey | -0.002 |
uddy | -0.002 |
Token | draw |
Token ID | 3197 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
canvas | +0.018 |
line | +0.018 |
product | +0.017 |
tide | +0.017 |
harbor | +0.017 |
Neg logit contributions | |
Spot | -0.014 |
girls | -0.013 |
Buddy | -0.013 |
Zip | -0.013 |
uddy | -0.013 |
Token | She |
Token ID | 1375 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
canvas | +0.006 |
product | +0.005 |
line | +0.005 |
tide | +0.005 |
lee | +0.005 |
Neg logit contributions | |
Spot | -0.003 |
girls | -0.003 |
Buddy | -0.003 |
Zip | -0.003 |
Fl | -0.003 |
Token | played |
Token ID | 2826 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
canvas | +0.015 |
lee | +0.014 |
line | +0.014 |
shell | +0.014 |
tide | +0.014 |
Neg logit contributions | |
girls | -0.016 |
Zip | -0.016 |
uddy | -0.015 |
oof | -0.015 |
Spot | -0.015 |
Token | tag |
Token ID | 7621 |
Feature activation | +2.26e+00 |
Pos logit contributions | |
uddy | +0.040 |
Johnson | +0.039 |
girls | +0.038 |
oof | +0.037 |
Monkey | +0.035 |
Neg logit contributions | |
line | -0.047 |
tide | -0.046 |
canvas | -0.045 |
harbor | -0.044 |
product | -0.044 |
Token | and |
Token ID | 290 |
Feature activation | +5.92e-01 |
Pos logit contributions | |
Okay | +0.083 |
oof | +0.082 |
girls | +0.082 |
Deal | +0.081 |
Understand | +0.081 |
Neg logit contributions | |
line | -0.065 |
shell | -0.064 |
canvas | -0.063 |
product | -0.061 |
harbor | -0.061 |
Token | hop |
Token ID | 1725 |
Feature activation | +4.78e-01 |
Pos logit contributions | |
uddy | +0.021 |
Understand | +0.021 |
ver | +0.021 |
Deal | +0.021 |
oof | +0.021 |
Neg logit contributions | |
shell | -0.017 |
line | -0.017 |
canvas | -0.016 |
tide | -0.016 |
shells | -0.015 |
Token | sc |
Token ID | 1416 |
Feature activation | +4.61e+00 |
Pos logit contributions | |
oof | +0.019 |
Okay | +0.019 |
girls | +0.018 |
Understand | +0.018 |
Johnson | +0.018 |
Neg logit contributions | |
line | -0.013 |
harbor | -0.011 |
tide | -0.011 |
shell | -0.011 |
canvas | -0.011 |
Token | ot |
Token ID | 313 |
Feature activation | -8.61e-01 |
Pos logit contributions | |
Spot | +0.209 |
girls | +0.206 |
Buddy | +0.203 |
Zip | +0.202 |
Max | +0.198 |
Neg logit contributions | |
canvas | -0.101 |
line | -0.096 |
product | -0.096 |
tide | -0.092 |
lee | -0.088 |
Token | ch |
Token ID | 354 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
line | +0.030 |
canvas | +0.030 |
lee | +0.029 |
tide | +0.028 |
product | +0.027 |
Neg logit contributions | |
Johnson | -0.025 |
Deal | -0.024 |
girls | -0.024 |
oof | -0.023 |
Monkey | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
Enjoy | +0.065 |
Monkey | +0.065 |
Johnson | +0.065 |
girls | +0.063 |
Deal | +0.062 |
Neg logit contributions | |
line | -0.078 |
canvas | -0.072 |
lee | -0.072 |
tide | -0.070 |
product | -0.069 |
Token | pretended |
Token ID | 36766 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
oof | +0.015 |
uddy | +0.014 |
ver | +0.014 |
Understand | +0.014 |
girls | +0.014 |
Neg logit contributions | |
line | -0.013 |
canvas | -0.012 |
tide | -0.012 |
shell | -0.012 |
harbor | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
girls | +0.043 |
oof | +0.042 |
uddy | +0.040 |
Spot | +0.039 |
Buddy | +0.039 |
Neg logit contributions | |
canvas | -0.042 |
line | -0.042 |
product | -0.041 |
tide | -0.040 |
lee | -0.039 |
Token | They |
Token ID | 1119 |
Feature activation | -8.01e-01 |
Pos logit contributions | |
canvas | +0.005 |
line | +0.005 |
lee | +0.005 |
product | +0.004 |
tide | +0.004 |
Neg logit contributions | |
girls | -0.003 |
Zip | -0.003 |
Spot | -0.003 |
oof | -0.003 |
Buddy | -0.003 |
Token | started |
Token ID | 2067 |
Feature activation | -7.08e-01 |
Pos logit contributions | |
canvas | +0.030 |
line | +0.029 |
product | +0.028 |
tide | +0.028 |
lee | +0.028 |
Neg logit contributions | |
Spot | -0.024 |
girls | -0.024 |
Zip | -0.023 |
Buddy | -0.023 |
oof | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | -6.72e-01 |
Pos logit contributions | |
canvas | +0.034 |
line | +0.033 |
product | +0.033 |
tide | +0.033 |
lee | +0.032 |
Neg logit contributions | |
Spot | -0.014 |
girls | -0.013 |
Buddy | -0.013 |
Zip | -0.013 |
Max | -0.012 |
Token | play |
Token ID | 711 |
Feature activation | +7.48e-01 |
Pos logit contributions | |
canvas | +0.023 |
line | +0.020 |
tide | +0.020 |
product | +0.020 |
harbor | +0.020 |
Neg logit contributions | |
uddy | -0.023 |
oof | -0.022 |
Spot | -0.022 |
girls | -0.022 |
Zip | -0.021 |
Token | hop |
Token ID | 1725 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
uddy | +0.021 |
Johnson | +0.020 |
oof | +0.020 |
girls | +0.020 |
Monkey | +0.019 |
Neg logit contributions | |
canvas | -0.029 |
line | -0.029 |
tide | -0.028 |
harbor | -0.028 |
product | -0.028 |
Token | sc |
Token ID | 1416 |
Feature activation | +4.47e+00 |
Pos logit contributions | |
oof | +0.004 |
Johnson | +0.004 |
Enjoy | +0.004 |
girls | +0.004 |
Okay | +0.004 |
Neg logit contributions | |
line | -0.003 |
harbor | -0.003 |
tide | -0.003 |
cliff | -0.003 |
shell | -0.003 |
Token | ot |
Token ID | 313 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
Spot | +0.201 |
girls | +0.197 |
Buddy | +0.195 |
Zip | +0.194 |
Max | +0.191 |
Neg logit contributions | |
canvas | -0.099 |
line | -0.095 |
product | -0.095 |
tide | -0.091 |
lee | -0.087 |
Token | ch |
Token ID | 354 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
canvas | +0.056 |
line | +0.055 |
product | +0.054 |
tide | +0.054 |
lee | +0.052 |
Neg logit contributions | |
Johnson | -0.032 |
girls | -0.032 |
Spot | -0.031 |
oof | -0.030 |
Monkey | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | -7.43e-02 |
Pos logit contributions | |
girls | +0.049 |
Spot | +0.049 |
Monkey | +0.049 |
Johnson | +0.048 |
Enjoy | +0.047 |
Neg logit contributions | |
line | -0.063 |
canvas | -0.061 |
product | -0.059 |
tide | -0.059 |
lee | -0.057 |
Token | had |
Token ID | 550 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
line | +0.002 |
canvas | +0.002 |
tide | +0.002 |
shell | +0.002 |
harbor | +0.002 |
Neg logit contributions | |
oof | -0.003 |
uddy | -0.003 |
Deal | -0.003 |
Johnson | -0.003 |
Understand | -0.002 |
Token | lots |
Token ID | 6041 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
canvas | +0.047 |
tide | +0.046 |
lee | +0.045 |
product | +0.045 |
harbor | +0.045 |
Neg logit contributions | |
uddy | -0.036 |
girls | -0.036 |
oof | -0.034 |
Monkey | -0.034 |
Johnson | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
line | +0.062 |
canvas | +0.060 |
product | +0.058 |
harbor | +0.056 |
lee | +0.055 |
Neg logit contributions | |
oof | -0.053 |
Zip | -0.052 |
Spot | -0.052 |
Wh | -0.051 |
girls | -0.051 |
Token | Barry |
Token ID | 14488 |
Feature activation | +7.63e-01 |
Pos logit contributions | |
girls | +0.001 |
oof | +0.001 |
losers | +0.001 |
Deal | +0.001 |
isk | +0.001 |
Neg logit contributions | |
lee | -0.001 |
harbor | -0.001 |
line | -0.001 |
canvas | -0.001 |
pumpkin | -0.001 |
Token | enjoyed |
Token ID | 8359 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
oof | +0.024 |
Spot | +0.024 |
Enjoy | +0.024 |
Zip | +0.024 |
girls | +0.024 |
Neg logit contributions | |
lee | -0.026 |
line | -0.023 |
shell | -0.023 |
product | -0.023 |
harbor | -0.023 |
Token | playing |
Token ID | 2712 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
uddy | +0.010 |
girls | +0.009 |
oof | +0.009 |
ver | +0.009 |
Deal | +0.008 |
Neg logit contributions | |
line | -0.009 |
product | -0.009 |
tide | -0.009 |
lee | -0.009 |
canvas | -0.009 |
Token | hop |
Token ID | 1725 |
Feature activation | +3.57e-01 |
Pos logit contributions | |
uddy | +0.035 |
Johnson | +0.033 |
girls | +0.032 |
oof | +0.032 |
ver | +0.031 |
Neg logit contributions | |
line | -0.033 |
tide | -0.033 |
canvas | -0.033 |
harbor | -0.032 |
product | -0.032 |
Token | sc |
Token ID | 1416 |
Feature activation | +4.42e+00 |
Pos logit contributions | |
Okay | +0.012 |
Johnson | +0.012 |
girls | +0.012 |
oof | +0.012 |
OK | +0.012 |
Neg logit contributions | |
line | -0.012 |
harbor | -0.011 |
tide | -0.011 |
shell | -0.011 |
canvas | -0.010 |
Token | ot |
Token ID | 313 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
girls | +0.209 |
Spot | +0.208 |
Buddy | +0.203 |
Zip | +0.202 |
Monkey | +0.202 |
Neg logit contributions | |
canvas | -0.085 |
line | -0.083 |
product | -0.083 |
lee | -0.078 |
tide | -0.075 |
Token | ch |
Token ID | 354 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
line | +0.058 |
canvas | +0.057 |
product | +0.055 |
lee | +0.055 |
tide | +0.054 |
Neg logit contributions | |
Johnson | -0.041 |
girls | -0.040 |
Deal | -0.039 |
Kitty | -0.038 |
Spot | -0.038 |
Token | with |
Token ID | 351 |
Feature activation | +4.36e-01 |
Pos logit contributions | |
Johnson | +0.054 |
Monkey | +0.051 |
girls | +0.050 |
Enjoy | +0.050 |
Spot | +0.050 |
Neg logit contributions | |
line | -0.061 |
tide | -0.055 |
canvas | -0.055 |
product | -0.055 |
lee | -0.054 |
Token | them |
Token ID | 606 |
Feature activation | -3.53e-01 |
Pos logit contributions | |
Deal | +0.016 |
ver | +0.016 |
uddy | +0.015 |
oof | +0.015 |
ll | +0.015 |
Neg logit contributions | |
lee | -0.014 |
tide | -0.014 |
shell | -0.014 |
shells | -0.013 |
harbor | -0.013 |
Token | so |
Token ID | 523 |
Feature activation | -3.89e-01 |
Pos logit contributions | |
line | +0.011 |
shell | +0.010 |
canvas | +0.010 |
harbor | +0.010 |
tide | +0.009 |
Neg logit contributions | |
oof | -0.013 |
Enjoy | -0.012 |
Deal | -0.012 |
Understand | -0.012 |
Johnson | -0.012 |
Token | would |
Token ID | 561 |
Feature activation | +7.32e-01 |
Pos logit contributions | |
canvas | +0.010 |
product | +0.009 |
tide | +0.009 |
line | +0.009 |
harbor | +0.009 |
Neg logit contributions | |
girls | -0.008 |
Zip | -0.007 |
Spot | -0.007 |
oof | -0.007 |
Buddy | -0.007 |
Token | play |
Token ID | 711 |
Feature activation | +9.61e-01 |
Pos logit contributions | |
girls | +0.023 |
Spot | +0.023 |
Buddy | +0.022 |
uddy | +0.022 |
Zip | +0.022 |
Neg logit contributions | |
canvas | -0.027 |
line | -0.026 |
product | -0.025 |
tide | -0.025 |
harbor | -0.025 |
Token | tag |
Token ID | 7621 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
girls | +0.033 |
uddy | +0.032 |
Johnson | +0.032 |
Kitty | +0.029 |
oof | +0.029 |
Neg logit contributions | |
canvas | -0.032 |
line | -0.032 |
product | -0.031 |
tide | -0.031 |
harbor | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | +5.78e-01 |
Pos logit contributions | |
Okay | +0.067 |
Johnson | +0.067 |
girls | +0.066 |
uddy | +0.066 |
Deal | +0.065 |
Neg logit contributions | |
line | -0.049 |
tide | -0.046 |
shell | -0.045 |
canvas | -0.045 |
product | -0.044 |
Token | hop |
Token ID | 1725 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
uddy | +0.021 |
oof | +0.021 |
girls | +0.021 |
ver | +0.021 |
Johnson | +0.020 |
Neg logit contributions | |
canvas | -0.018 |
line | -0.018 |
tide | -0.017 |
shell | -0.016 |
harbor | -0.016 |
Token | sc |
Token ID | 1416 |
Feature activation | +4.35e+00 |
Pos logit contributions | |
Johnson | +0.005 |
Okay | +0.005 |
OK | +0.005 |
oof | +0.005 |
girls | +0.005 |
Neg logit contributions | |
line | -0.004 |
harbor | -0.004 |
tide | -0.004 |
shell | -0.004 |
canvas | -0.004 |
Token | ot |
Token ID | 313 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
Spot | +0.191 |
girls | +0.186 |
Buddy | +0.185 |
Zip | +0.184 |
Max | +0.181 |
Neg logit contributions | |
canvas | -0.102 |
line | -0.098 |
product | -0.097 |
tide | -0.095 |
lee | -0.089 |
Token | ch |
Token ID | 354 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
canvas | +0.061 |
line | +0.060 |
product | +0.058 |
tide | +0.057 |
lee | +0.057 |
Neg logit contributions | |
Johnson | -0.035 |
girls | -0.035 |
Spot | -0.034 |
Deal | -0.033 |
Zip | -0.033 |
Token | , |
Token ID | 11 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
girls | +0.044 |
Johnson | +0.043 |
Spot | +0.042 |
Monkey | +0.042 |
Buddy | +0.040 |
Neg logit contributions | |
line | -0.067 |
canvas | -0.066 |
product | -0.064 |
tide | -0.064 |
lee | -0.062 |
Token | and |
Token ID | 290 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
girls | +0.007 |
oof | +0.007 |
uddy | +0.007 |
Spot | +0.007 |
Buddy | +0.006 |
Neg logit contributions | |
line | -0.011 |
canvas | -0.011 |
tide | -0.010 |
product | -0.010 |
shell | -0.010 |
Token | many |
Token ID | 867 |
Feature activation | -3.23e-01 |
Pos logit contributions | |
oof | +0.010 |
uddy | +0.009 |
girls | +0.009 |
Johnson | +0.009 |
Deal | +0.009 |
Neg logit contributions | |
canvas | -0.009 |
line | -0.009 |
tide | -0.008 |
shell | -0.008 |
product | -0.008 |
Token | and |
Token ID | 392 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
Spot | +0.036 |
girls | +0.035 |
Buddy | +0.034 |
Zip | +0.033 |
Max | +0.033 |
Neg logit contributions | |
canvas | -0.052 |
line | -0.052 |
product | -0.050 |
tide | -0.049 |
lee | -0.048 |
Token | - |
Token ID | 12 |
Feature activation | +2.26e+00 |
Pos logit contributions | |
Spot | +0.073 |
Buddy | +0.071 |
girls | +0.069 |
Zip | +0.069 |
Fl | +0.068 |
Neg logit contributions | |
canvas | -0.096 |
tide | -0.093 |
line | -0.093 |
product | -0.093 |
harbor | -0.089 |
Token | seek |
Token ID | 36163 |
Feature activation | +9.96e-01 |
Pos logit contributions | |
Spot | +0.067 |
girls | +0.065 |
Buddy | +0.064 |
Max | +0.064 |
Monkey | +0.063 |
Neg logit contributions | |
line | -0.085 |
canvas | -0.081 |
product | -0.077 |
lee | -0.074 |
tide | -0.072 |
Token | and |
Token ID | 290 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
girls | +0.037 |
oof | +0.037 |
uddy | +0.037 |
Monkey | +0.036 |
Johnson | +0.035 |
Neg logit contributions | |
line | -0.027 |
canvas | -0.027 |
tide | -0.026 |
harbor | -0.026 |
product | -0.025 |
Token | hop |
Token ID | 1725 |
Feature activation | +7.84e-02 |
Pos logit contributions | |
ver | +0.005 |
Deal | +0.005 |
oof | +0.005 |
uddy | +0.005 |
Understand | +0.005 |
Neg logit contributions | |
shell | -0.004 |
line | -0.003 |
canvas | -0.003 |
harbor | -0.003 |
tide | -0.003 |
Token | sc |
Token ID | 1416 |
Feature activation | +4.33e+00 |
Pos logit contributions | |
Johnson | +0.003 |
Okay | +0.003 |
Deal | +0.003 |
oof | +0.003 |
Understand | +0.003 |
Neg logit contributions | |
line | -0.002 |
harbor | -0.002 |
shell | -0.002 |
tide | -0.002 |
cliff | -0.002 |
Token | ot |
Token ID | 313 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
Spot | +0.190 |
girls | +0.186 |
Buddy | +0.184 |
Zip | +0.183 |
Max | +0.180 |
Neg logit contributions | |
canvas | -0.101 |
line | -0.098 |
product | -0.097 |
tide | -0.094 |
lee | -0.089 |
Token | ch |
Token ID | 354 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
canvas | +0.036 |
line | +0.035 |
lee | +0.035 |
product | +0.034 |
tide | +0.033 |
Neg logit contributions | |
Johnson | -0.031 |
girls | -0.029 |
Deal | -0.029 |
oof | -0.029 |
Spot | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
Monkey | +0.053 |
Johnson | +0.053 |
Enjoy | +0.052 |
girls | +0.052 |
Spot | +0.052 |
Neg logit contributions | |
line | -0.060 |
canvas | -0.058 |
lee | -0.056 |
product | -0.056 |
tide | -0.055 |
Token | Then |
Token ID | 3244 |
Feature activation | -9.68e-03 |
Pos logit contributions | |
lee | +0.001 |
harbor | +0.001 |
canvas | +0.001 |
line | +0.001 |
product | +0.001 |
Neg logit contributions | |
girls | -0.001 |
oof | -0.001 |
ver | -0.001 |
ever | -0.001 |
Deal | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -4.13e-01 |
Pos logit contributions | |
canvas | +0.000 |
product | +0.000 |
line | +0.000 |
tide | +0.000 |
harbor | +0.000 |
Neg logit contributions | |
girls | -0.000 |
oof | -0.000 |
uddy | -0.000 |
Spot | -0.000 |
Zip | -0.000 |
Token | also |
Token ID | 635 |
Feature activation | +4.66e-02 |
Pos logit contributions | |
Spot | +0.000 |
girls | +0.000 |
Buddy | +0.000 |
Zip | +0.000 |
Max | +0.000 |
Neg logit contributions | |
canvas | -0.000 |
line | -0.000 |
product | -0.000 |
tide | -0.000 |
lee | -0.000 |
Token | played |
Token ID | 2826 |
Feature activation | +6.19e-01 |
Pos logit contributions | |
Spot | +0.001 |
girls | +0.001 |
Buddy | +0.001 |
Zip | +0.001 |
Max | +0.001 |
Neg logit contributions | |
canvas | -0.002 |
line | -0.002 |
product | -0.002 |
tide | -0.002 |
lee | -0.002 |
Token | tag |
Token ID | 7621 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
uddy | +0.018 |
Johnson | +0.017 |
girls | +0.017 |
Spot | +0.017 |
oof | +0.016 |
Neg logit contributions | |
canvas | -0.024 |
tide | -0.024 |
harbor | -0.023 |
shell | -0.023 |
product | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
Understand | +0.055 |
oof | +0.055 |
Deal | +0.054 |
Okay | +0.053 |
OK | +0.053 |
Neg logit contributions | |
canvas | -0.054 |
shell | -0.054 |
harbor | -0.054 |
line | -0.053 |
pumpkin | -0.052 |
Token | hop |
Token ID | 1725 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
canvas | +0.005 |
shell | +0.004 |
lee | +0.004 |
pumpkin | +0.004 |
pulp | +0.004 |
Neg logit contributions | |
Understand | -0.006 |
ver | -0.006 |
Deal | -0.006 |
uddy | -0.005 |
If | -0.005 |
Token | sc |
Token ID | 1416 |
Feature activation | +4.24e+00 |
Pos logit contributions | |
oof | +0.000 |
Understand | +0.000 |
Johnson | +0.000 |
Okay | +0.000 |
OK | +0.000 |
Neg logit contributions | |
line | -0.000 |
harbor | -0.000 |
tide | -0.000 |
shell | -0.000 |
canvas | -0.000 |
Token | ot |
Token ID | 313 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
Spot | +0.195 |
girls | +0.193 |
Buddy | +0.189 |
Zip | +0.188 |
Monkey | +0.186 |
Neg logit contributions | |
canvas | -0.090 |
product | -0.086 |
line | -0.085 |
lee | -0.081 |
tide | -0.080 |
Token | ch |
Token ID | 354 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
canvas | +0.058 |
line | +0.056 |
lee | +0.055 |
product | +0.055 |
tide | +0.054 |
Neg logit contributions | |
Johnson | -0.038 |
girls | -0.037 |
Spot | -0.037 |
Deal | -0.036 |
oof | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | -4.59e-01 |
Pos logit contributions | |
Spot | +0.050 |
Monkey | +0.049 |
girls | +0.049 |
Johnson | +0.048 |
Enjoy | +0.047 |
Neg logit contributions | |
line | -0.063 |
canvas | -0.063 |
product | -0.061 |
lee | -0.060 |
tide | -0.060 |
Token | They |
Token ID | 1119 |
Feature activation | -3.20e-01 |
Pos logit contributions | |
lee | +0.017 |
canvas | +0.017 |
line | +0.017 |
tide | +0.016 |
harbor | +0.016 |
Neg logit contributions | |
girls | -0.012 |
Deal | -0.012 |
losers | -0.012 |
oof | -0.012 |
ever | -0.012 |
Token | laughed |
Token ID | 13818 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
canvas | +0.011 |
line | +0.010 |
shell | +0.010 |
harbor | +0.010 |
lee | +0.010 |
Neg logit contributions | |
oof | -0.011 |
uddy | -0.010 |
girls | -0.010 |
Johnson | -0.010 |
Spot | -0.010 |
Token | play |
Token ID | 711 |
Feature activation | +9.12e-01 |
Pos logit contributions | |
canvas | +0.007 |
line | +0.007 |
tide | +0.007 |
product | +0.007 |
harbor | +0.006 |
Neg logit contributions | |
girls | -0.007 |
Spot | -0.007 |
Zip | -0.006 |
Buddy | -0.006 |
uddy | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
girls | +0.024 |
Spot | +0.024 |
uddy | +0.023 |
Zip | +0.023 |
Buddy | +0.022 |
Neg logit contributions | |
canvas | -0.037 |
line | -0.036 |
product | -0.036 |
tide | -0.036 |
harbor | -0.034 |
Token | game |
Token ID | 983 |
Feature activation | +6.26e-01 |
Pos logit contributions | |
Spot | +0.027 |
girls | +0.026 |
Buddy | +0.026 |
Zip | +0.025 |
Fl | +0.025 |
Neg logit contributions | |
canvas | -0.045 |
line | -0.044 |
product | -0.044 |
tide | -0.043 |
lee | -0.042 |
Token | of |
Token ID | 286 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
girls | +0.017 |
Spot | +0.017 |
Buddy | +0.016 |
Zip | +0.016 |
oof | +0.015 |
Neg logit contributions | |
canvas | -0.025 |
line | -0.025 |
product | -0.024 |
tide | -0.024 |
harbor | -0.023 |
Token | hop |
Token ID | 1725 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
uddy | +0.035 |
Johnson | +0.031 |
ver | +0.031 |
girls | +0.031 |
oof | +0.030 |
Neg logit contributions | |
canvas | -0.053 |
tide | -0.053 |
shell | -0.050 |
cliff | -0.050 |
line | -0.050 |
Token | sc |
Token ID | 1416 |
Feature activation | +4.24e+00 |
Pos logit contributions | |
line | +0.006 |
tide | +0.006 |
harbor | +0.005 |
shell | +0.005 |
canvas | +0.005 |
Neg logit contributions | |
girls | -0.005 |
Johnson | -0.005 |
oof | -0.005 |
uddy | -0.005 |
Monkey | -0.005 |
Token | ot |
Token ID | 313 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
Spot | +0.178 |
Buddy | +0.173 |
oof | +0.172 |
Zip | +0.170 |
Fl | +0.169 |
Neg logit contributions | |
canvas | -0.109 |
line | -0.105 |
product | -0.104 |
tide | -0.103 |
harbor | -0.097 |
Token | ch |
Token ID | 354 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
canvas | +0.090 |
line | +0.089 |
product | +0.088 |
tide | +0.087 |
lee | +0.085 |
Neg logit contributions | |
Spot | -0.040 |
girls | -0.040 |
Buddy | -0.038 |
Zip | -0.037 |
Johnson | -0.037 |
Token | , |
Token ID | 11 |
Feature activation | +6.22e-03 |
Pos logit contributions | |
Spot | +0.038 |
girls | +0.037 |
Buddy | +0.036 |
Zip | +0.036 |
Monkey | +0.035 |
Neg logit contributions | |
canvas | -0.056 |
line | -0.056 |
product | -0.055 |
tide | -0.055 |
lee | -0.053 |
Token | but |
Token ID | 475 |
Feature activation | -7.54e-01 |
Pos logit contributions | |
girls | +0.000 |
oof | +0.000 |
uddy | +0.000 |
Spot | +0.000 |
Zip | +0.000 |
Neg logit contributions | |
line | -0.000 |
canvas | -0.000 |
tide | -0.000 |
harbor | -0.000 |
shell | -0.000 |
Token | it |
Token ID | 340 |
Feature activation | +8.51e-01 |
Pos logit contributions | |
canvas | +0.018 |
line | +0.018 |
shell | +0.017 |
tide | +0.016 |
harbor | +0.016 |
Neg logit contributions | |
girls | -0.035 |
uddy | -0.033 |
oof | -0.033 |
Enjoy | -0.032 |
Hero | -0.030 |
Token | group |
Token ID | 1448 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
girls | +0.017 |
oof | +0.016 |
Zip | +0.015 |
uddy | +0.015 |
Spot | +0.015 |
Neg logit contributions | |
canvas | -0.013 |
product | -0.012 |
shell | -0.012 |
tide | -0.012 |
harbor | -0.012 |
Token | of |
Token ID | 286 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
canvas | +0.047 |
line | +0.046 |
product | +0.045 |
tide | +0.045 |
lee | +0.044 |
Neg logit contributions | |
Spot | -0.024 |
girls | -0.023 |
Buddy | -0.023 |
Zip | -0.023 |
Max | -0.022 |
Token | girls |
Token ID | 4813 |
Feature activation | -8.15e-01 |
Pos logit contributions | |
uddy | +0.083 |
ver | +0.077 |
oof | +0.077 |
isk | +0.077 |
Johnson | +0.077 |
Neg logit contributions | |
tide | -0.059 |
canvas | -0.058 |
harbor | -0.056 |
shell | -0.054 |
waves | -0.053 |
Token | playing |
Token ID | 2712 |
Feature activation | +7.90e-01 |
Pos logit contributions | |
canvas | +0.032 |
line | +0.031 |
product | +0.031 |
tide | +0.031 |
lee | +0.030 |
Neg logit contributions | |
Spot | -0.023 |
girls | -0.022 |
Buddy | -0.022 |
Zip | -0.021 |
Max | -0.021 |
Token | hop |
Token ID | 1725 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
uddy | +0.027 |
Johnson | +0.025 |
Kitty | +0.024 |
oof | +0.024 |
girls | +0.024 |
Neg logit contributions | |
line | -0.024 |
canvas | -0.024 |
product | -0.024 |
tide | -0.023 |
shell | -0.023 |
Token | sc |
Token ID | 1416 |
Feature activation | +4.23e+00 |
Pos logit contributions | |
line | +0.007 |
harbor | +0.007 |
tide | +0.007 |
shell | +0.006 |
canvas | +0.006 |
Neg logit contributions | |
Johnson | -0.005 |
Kitty | -0.005 |
oof | -0.005 |
girls | -0.005 |
Monkey | -0.005 |
Token | ot |
Token ID | 313 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
Spot | +0.178 |
girls | +0.174 |
Buddy | +0.172 |
Zip | +0.171 |
Max | +0.168 |
Neg logit contributions | |
canvas | -0.107 |
line | -0.103 |
product | -0.102 |
tide | -0.100 |
lee | -0.094 |
Token | ch |
Token ID | 354 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
canvas | +0.072 |
line | +0.071 |
product | +0.070 |
tide | +0.069 |
lee | +0.068 |
Neg logit contributions | |
Johnson | -0.042 |
girls | -0.041 |
Spot | -0.040 |
Kitty | -0.040 |
oof | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
Monkey | +0.055 |
Spot | +0.055 |
girls | +0.054 |
Johnson | +0.054 |
Buddy | +0.053 |
Neg logit contributions | |
line | -0.056 |
canvas | -0.054 |
tide | -0.053 |
product | -0.053 |
lee | -0.051 |
Token | They |
Token ID | 1119 |
Feature activation | -3.21e-01 |
Pos logit contributions | |
canvas | +0.006 |
lee | +0.006 |
line | +0.006 |
product | +0.006 |
harbor | +0.006 |
Neg logit contributions | |
girls | -0.004 |
oof | -0.004 |
Zip | -0.004 |
uddy | -0.004 |
Spot | -0.003 |
Token | joined |
Token ID | 5399 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
harbor | +0.010 |
line | +0.010 |
lee | +0.010 |
shell | +0.010 |
canvas | +0.010 |
Neg logit contributions | |
oof | -0.011 |
uddy | -0.011 |
Johnson | -0.010 |
girls | -0.010 |
Zip | -0.010 |
Token | a |
Token ID | 257 |
Feature activation | +7.42e-01 |
Pos logit contributions | |
girls | +0.040 |
uddy | +0.040 |
oof | +0.037 |
Deal | +0.037 |
Buddy | +0.036 |
Neg logit contributions | |
canvas | -0.056 |
tide | -0.056 |
product | -0.056 |
shell | -0.054 |
line | -0.053 |
Token | big |
Token ID | 1263 |
Feature activation | -7.50e-01 |
Pos logit contributions | |
girls | +0.029 |
Spot | +0.028 |
Zip | +0.028 |
Buddy | +0.028 |
oof | +0.027 |
Neg logit contributions | |
canvas | -0.022 |
product | -0.021 |
tide | -0.021 |
line | -0.020 |
harbor | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
canvas | +0.026 |
product | +0.025 |
line | +0.025 |
tide | +0.024 |
harbor | +0.023 |
Neg logit contributions | |
Spot | -0.025 |
girls | -0.024 |
Buddy | -0.024 |
Zip | -0.024 |
uddy | -0.023 |
Token | pink |
Token ID | 11398 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
girls | +0.007 |
Spot | +0.007 |
uddy | +0.007 |
Buddy | +0.007 |
Zip | +0.007 |
Neg logit contributions | |
canvas | -0.005 |
product | -0.004 |
tide | -0.004 |
line | -0.004 |
shell | -0.004 |
Token | elephant |
Token ID | 20950 |
Feature activation | +9.68e-01 |
Pos logit contributions | |
Spot | +0.004 |
girls | +0.004 |
Buddy | +0.004 |
Zip | +0.004 |
Max | +0.004 |
Neg logit contributions | |
canvas | -0.004 |
product | -0.004 |
line | -0.004 |
tide | -0.004 |
harbor | -0.004 |
Token | stuffed |
Token ID | 22259 |
Feature activation | +4.22e+00 |
Pos logit contributions | |
Spot | +0.026 |
Buddy | +0.025 |
girls | +0.025 |
Zip | +0.024 |
Max | +0.024 |
Neg logit contributions | |
canvas | -0.040 |
line | -0.039 |
product | -0.039 |
tide | -0.038 |
lee | -0.037 |
Token | animal |
Token ID | 5044 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
ver | +0.149 |
OK | +0.144 |
uddy | +0.137 |
Deal | +0.137 |
Deal | +0.136 |
Neg logit contributions | |
pumpkin | -0.135 |
lee | -0.131 |
shell | -0.129 |
canvas | -0.127 |
line | -0.121 |
Token | . |
Token ID | 13 |
Feature activation | -4.14e-01 |
Pos logit contributions | |
Spot | +0.054 |
girls | +0.054 |
Buddy | +0.052 |
Zip | +0.052 |
oof | +0.051 |
Neg logit contributions | |
canvas | -0.072 |
product | -0.070 |
line | -0.070 |
tide | -0.068 |
lee | -0.066 |
Token | He |
Token ID | 679 |
Feature activation | -5.39e-01 |
Pos logit contributions | |
canvas | +0.017 |
line | +0.017 |
product | +0.016 |
lee | +0.016 |
tide | +0.016 |
Neg logit contributions | |
girls | -0.011 |
oof | -0.010 |
Zip | -0.010 |
Spot | -0.010 |
uddy | -0.010 |
Token | had |
Token ID | 550 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
canvas | +0.016 |
lee | +0.016 |
product | +0.015 |
shell | +0.015 |
line | +0.015 |
Neg logit contributions | |
girls | -0.020 |
Spot | -0.019 |
oof | -0.019 |
uddy | -0.018 |
Zip | -0.018 |
Token | never |
Token ID | 1239 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
canvas | +0.008 |
product | +0.008 |
lee | +0.008 |
shell | +0.008 |
line | +0.008 |
Neg logit contributions | |
girls | -0.005 |
uddy | -0.004 |
Deal | -0.004 |
oof | -0.004 |
Buddy | -0.004 |
Token | hide |
Token ID | 7808 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
uddy | +0.023 |
oof | +0.022 |
Deal | +0.021 |
Johnson | +0.021 |
Finally | +0.021 |
Neg logit contributions | |
line | -0.024 |
canvas | -0.023 |
tide | -0.023 |
harbor | -0.023 |
shell | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | +5.26e-01 |
Pos logit contributions | |
uddy | +0.054 |
Monkey | +0.053 |
oof | +0.053 |
girls | +0.052 |
Zip | +0.052 |
Neg logit contributions | |
line | -0.066 |
canvas | -0.066 |
lee | -0.064 |
shell | -0.063 |
tide | -0.062 |
Token | seek |
Token ID | 5380 |
Feature activation | +8.89e-02 |
Pos logit contributions | |
uddy | +0.019 |
ver | +0.018 |
oof | +0.018 |
isk | +0.017 |
girls | +0.017 |
Neg logit contributions | |
canvas | -0.017 |
line | -0.017 |
tide | -0.017 |
shell | -0.016 |
harbor | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | -3.40e-03 |
Pos logit contributions | |
uddy | +0.003 |
oof | +0.003 |
Monkey | +0.003 |
Wh | +0.003 |
girls | +0.003 |
Neg logit contributions | |
canvas | -0.003 |
line | -0.003 |
product | -0.003 |
harbor | -0.003 |
tide | -0.003 |
Token | hop |
Token ID | 1725 |
Feature activation | -3.39e-02 |
Pos logit contributions | |
line | +0.000 |
canvas | +0.000 |
tide | +0.000 |
harbor | +0.000 |
shell | +0.000 |
Neg logit contributions | |
uddy | -0.000 |
oof | -0.000 |
girls | -0.000 |
Monkey | -0.000 |
Johnson | -0.000 |
Token | sc |
Token ID | 1416 |
Feature activation | +4.17e+00 |
Pos logit contributions | |
line | +0.001 |
harbor | +0.001 |
tide | +0.001 |
canvas | +0.001 |
shell | +0.001 |
Neg logit contributions | |
oof | -0.001 |
girls | -0.001 |
Deal | -0.001 |
Johnson | -0.001 |
Monkey | -0.001 |
Token | ot |
Token ID | 313 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
Spot | +0.188 |
girls | +0.186 |
Buddy | +0.183 |
Zip | +0.182 |
Max | +0.179 |
Neg logit contributions | |
canvas | -0.092 |
line | -0.088 |
product | -0.087 |
tide | -0.084 |
lee | -0.081 |
Token | ch |
Token ID | 354 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
canvas | +0.055 |
line | +0.055 |
lee | +0.053 |
tide | +0.052 |
product | +0.052 |
Neg logit contributions | |
Johnson | -0.041 |
Deal | -0.040 |
girls | -0.039 |
oof | -0.038 |
Spot | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
Johnson | +0.051 |
girls | +0.051 |
Monkey | +0.051 |
Spot | +0.050 |
Enjoy | +0.049 |
Neg logit contributions | |
line | -0.066 |
canvas | -0.064 |
tide | -0.062 |
lee | -0.061 |
product | -0.061 |
Token | much |
Token ID | 881 |
Feature activation | -3.99e-01 |
Pos logit contributions | |
oof | +0.010 |
uddy | +0.010 |
ver | +0.010 |
girls | +0.010 |
Johnson | +0.010 |
Neg logit contributions | |
canvas | -0.010 |
tide | -0.010 |
line | -0.010 |
shell | -0.009 |
harbor | -0.009 |
Token | more |
Token ID | 517 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
canvas | +0.014 |
tide | +0.013 |
line | +0.013 |
product | +0.013 |
harbor | +0.012 |
Neg logit contributions | |
girls | -0.013 |
Spot | -0.013 |
oof | -0.013 |
uddy | -0.013 |
Zip | -0.012 |
Token | a |
Token ID | 257 |
Feature activation | +7.36e-01 |
Pos logit contributions | |
girls | +0.041 |
uddy | +0.040 |
oof | +0.038 |
Deal | +0.037 |
Buddy | +0.036 |
Neg logit contributions | |
tide | -0.055 |
canvas | -0.055 |
product | -0.055 |
shell | -0.053 |
harbor | -0.052 |
Token | big |
Token ID | 1263 |
Feature activation | -7.53e-01 |
Pos logit contributions | |
girls | +0.029 |
Spot | +0.028 |
Zip | +0.027 |
Buddy | +0.027 |
oof | +0.027 |
Neg logit contributions | |
canvas | -0.022 |
product | -0.021 |
tide | -0.020 |
line | -0.020 |
harbor | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
canvas | +0.026 |
product | +0.025 |
line | +0.025 |
tide | +0.024 |
harbor | +0.023 |
Neg logit contributions | |
Spot | -0.025 |
girls | -0.025 |
Buddy | -0.024 |
Zip | -0.024 |
uddy | -0.023 |
Token | pink |
Token ID | 11398 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
girls | +0.006 |
uddy | +0.006 |
Spot | +0.006 |
Buddy | +0.006 |
Zip | +0.006 |
Neg logit contributions | |
canvas | -0.004 |
product | -0.004 |
tide | -0.004 |
line | -0.004 |
shell | -0.003 |
Token | elephant |
Token ID | 20950 |
Feature activation | +9.46e-01 |
Pos logit contributions | |
Spot | +0.003 |
girls | +0.003 |
Buddy | +0.003 |
Zip | +0.003 |
Max | +0.003 |
Neg logit contributions | |
canvas | -0.004 |
product | -0.003 |
line | -0.003 |
tide | -0.003 |
harbor | -0.003 |
Token | stuffed |
Token ID | 22259 |
Feature activation | +4.17e+00 |
Pos logit contributions | |
Spot | +0.025 |
Buddy | +0.024 |
girls | +0.024 |
Zip | +0.024 |
Max | +0.023 |
Neg logit contributions | |
canvas | -0.039 |
line | -0.038 |
product | -0.038 |
tide | -0.038 |
lee | -0.036 |
Token | animal |
Token ID | 5044 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
ver | +0.147 |
OK | +0.143 |
uddy | +0.136 |
Deal | +0.135 |
Deal | +0.135 |
Neg logit contributions | |
pumpkin | -0.133 |
lee | -0.130 |
shell | -0.128 |
canvas | -0.126 |
line | -0.120 |
Token | . |
Token ID | 13 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
Spot | +0.053 |
girls | +0.053 |
Buddy | +0.051 |
Zip | +0.051 |
oof | +0.050 |
Neg logit contributions | |
canvas | -0.070 |
product | -0.068 |
line | -0.068 |
tide | -0.066 |
lee | -0.065 |
Token | He |
Token ID | 679 |
Feature activation | -5.38e-01 |
Pos logit contributions | |
canvas | +0.017 |
line | +0.016 |
product | +0.016 |
tide | +0.016 |
lee | +0.016 |
Neg logit contributions | |
girls | -0.010 |
oof | -0.010 |
Zip | -0.009 |
Spot | -0.009 |
uddy | -0.009 |
Token | had |
Token ID | 550 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
canvas | +0.016 |
lee | +0.016 |
product | +0.015 |
shell | +0.015 |
line | +0.015 |
Neg logit contributions | |
girls | -0.020 |
Spot | -0.019 |
oof | -0.019 |
Zip | -0.018 |
uddy | -0.018 |
Token | never |
Token ID | 1239 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
canvas | +0.009 |
product | +0.009 |
lee | +0.009 |
shell | +0.009 |
line | +0.009 |
Neg logit contributions | |
girls | -0.005 |
uddy | -0.005 |
Deal | -0.005 |
Buddy | -0.005 |
oof | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +3.39e-01 |
Pos logit contributions | |
canvas | +0.005 |
line | +0.005 |
tide | +0.004 |
product | +0.004 |
pumpkin | +0.004 |
Neg logit contributions | |
uddy | -0.007 |
Kenny | -0.007 |
Zip | -0.007 |
Monkey | -0.007 |
Buddy | -0.007 |
Token | You |
Token ID | 921 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
girls | +0.010 |
uddy | +0.009 |
ll | +0.009 |
isk | +0.009 |
ver | +0.009 |
Neg logit contributions | |
canvas | -0.012 |
line | -0.011 |
tide | -0.011 |
rys | -0.011 |
product | -0.011 |
Token | can |
Token ID | 460 |
Feature activation | +4.33e-01 |
Pos logit contributions | |
canvas | +0.010 |
tide | +0.009 |
shell | +0.009 |
line | +0.009 |
harbor | +0.009 |
Neg logit contributions | |
uddy | -0.007 |
Zip | -0.007 |
girls | -0.007 |
Spot | -0.007 |
Johnson | -0.007 |
Token | play |
Token ID | 711 |
Feature activation | +7.47e-01 |
Pos logit contributions | |
girls | +0.013 |
Zip | +0.012 |
Spot | +0.012 |
uddy | +0.012 |
Buddy | +0.012 |
Neg logit contributions | |
canvas | -0.016 |
line | -0.015 |
tide | -0.015 |
harbor | -0.015 |
product | -0.015 |
Token | hop |
Token ID | 1725 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
uddy | +0.027 |
Johnson | +0.025 |
girls | +0.024 |
ver | +0.023 |
Kenny | +0.023 |
Neg logit contributions | |
canvas | -0.024 |
shell | -0.023 |
product | -0.023 |
line | -0.023 |
tide | -0.023 |
Token | sc |
Token ID | 1416 |
Feature activation | +4.11e+00 |
Pos logit contributions | |
line | +0.008 |
harbor | +0.008 |
shell | +0.008 |
tide | +0.008 |
canvas | +0.007 |
Neg logit contributions | |
Johnson | -0.009 |
uddy | -0.008 |
Kenny | -0.008 |
girls | -0.008 |
Monkey | -0.008 |
Token | ot |
Token ID | 313 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
Spot | +0.191 |
girls | +0.190 |
Buddy | +0.187 |
Monkey | +0.186 |
Johnson | +0.185 |
Neg logit contributions | |
canvas | -0.083 |
product | -0.078 |
line | -0.077 |
lee | -0.072 |
tide | -0.072 |
Token | ch |
Token ID | 354 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
canvas | +0.082 |
line | +0.081 |
tide | +0.079 |
product | +0.079 |
lee | +0.076 |
Neg logit contributions | |
Spot | -0.042 |
Johnson | -0.042 |
girls | -0.041 |
Buddy | -0.040 |
Kitty | -0.040 |
Token | again |
Token ID | 757 |
Feature activation | +8.98e-01 |
Pos logit contributions | |
Spot | +0.043 |
girls | +0.042 |
Buddy | +0.041 |
Johnson | +0.041 |
Monkey | +0.041 |
Neg logit contributions | |
canvas | -0.058 |
line | -0.058 |
tide | -0.057 |
product | -0.057 |
lee | -0.054 |
Token | . |
Token ID | 13 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
uddy | +0.029 |
girls | +0.028 |
Spot | +0.028 |
Zip | +0.028 |
Buddy | +0.028 |
Neg logit contributions | |
canvas | -0.032 |
line | -0.031 |
product | -0.030 |
tide | -0.029 |
shell | -0.029 |
Token | But |
Token ID | 887 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
ll | +0.008 |
girls | +0.008 |
Johnson | +0.007 |
isk | +0.007 |
uddy | +0.007 |
Neg logit contributions | |
canvas | -0.010 |
line | -0.010 |
rys | -0.010 |
tide | -0.009 |
harbor | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
Spot | +0.035 |
girls | +0.034 |
Zip | +0.034 |
Buddy | +0.034 |
oof | +0.034 |
Neg logit contributions | |
line | -0.045 |
canvas | -0.044 |
product | -0.043 |
tide | -0.042 |
lee | -0.041 |
Token | He |
Token ID | 679 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
canvas | +0.016 |
product | +0.015 |
line | +0.015 |
tide | +0.015 |
lee | +0.014 |
Neg logit contributions | |
Spot | -0.007 |
girls | -0.007 |
Buddy | -0.006 |
Zip | -0.006 |
Max | -0.006 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
canvas | +0.004 |
product | +0.004 |
tide | +0.004 |
line | +0.004 |
lee | +0.004 |
Neg logit contributions | |
girls | -0.004 |
Zip | -0.004 |
Spot | -0.004 |
Buddy | -0.004 |
uddy | -0.004 |
Token | as |
Token ID | 355 |
Feature activation | -7.91e-02 |
Pos logit contributions | |
Spot | +0.004 |
girls | +0.004 |
Buddy | +0.004 |
Zip | +0.004 |
Max | +0.004 |
Neg logit contributions | |
canvas | -0.009 |
line | -0.008 |
product | -0.008 |
tide | -0.008 |
lee | -0.008 |
Token | he |
Token ID | 339 |
Feature activation | +6.78e-01 |
Pos logit contributions | |
canvas | +0.003 |
product | +0.003 |
line | +0.003 |
tide | +0.003 |
lee | +0.003 |
Neg logit contributions | |
Spot | -0.002 |
Buddy | -0.002 |
Zip | -0.002 |
Max | -0.002 |
girls | -0.002 |
Token | stuffed |
Token ID | 22259 |
Feature activation | +4.08e+00 |
Pos logit contributions | |
Spot | +0.024 |
Zip | +0.024 |
Buddy | +0.024 |
girls | +0.024 |
uddy | +0.023 |
Neg logit contributions | |
canvas | -0.020 |
product | -0.020 |
lee | -0.020 |
line | -0.020 |
tide | -0.019 |
Token | them |
Token ID | 606 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
OK | +0.150 |
Deal | +0.146 |
uddy | +0.141 |
Okay | +0.140 |
ver | +0.138 |
Neg logit contributions | |
lee | -0.127 |
line | -0.123 |
shells | -0.122 |
shell | -0.121 |
tides | -0.118 |
Token | into |
Token ID | 656 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
oof | +0.033 |
Zip | +0.032 |
girls | +0.032 |
uddy | +0.031 |
Buddy | +0.031 |
Neg logit contributions | |
line | -0.039 |
canvas | -0.037 |
product | -0.035 |
harbor | -0.035 |
lee | -0.035 |
Token | his |
Token ID | 465 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
girls | +0.009 |
Spot | +0.008 |
uddy | +0.008 |
Buddy | +0.008 |
Zip | +0.008 |
Neg logit contributions | |
canvas | -0.011 |
product | -0.011 |
line | -0.011 |
tide | -0.011 |
lee | -0.010 |
Token | backpack |
Token ID | 22526 |
Feature activation | +8.41e-01 |
Pos logit contributions | |
Spot | +0.005 |
girls | +0.005 |
Buddy | +0.005 |
Zip | +0.005 |
Wh | +0.005 |
Neg logit contributions | |
canvas | -0.004 |
product | -0.004 |
line | -0.004 |
tide | -0.004 |
lee | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
girls | +0.029 |
oof | +0.028 |
Zip | +0.028 |
Buddy | +0.027 |
Spot | +0.027 |
Neg logit contributions | |
line | -0.031 |
canvas | -0.029 |
product | -0.027 |
shell | -0.026 |
harbor | -0.026 |
Token | pieces |
Token ID | 5207 |
Feature activation | -7.62e-01 |
Pos logit contributions | |
line | +0.003 |
canvas | +0.003 |
harbor | +0.003 |
lee | +0.003 |
tide | +0.003 |
Neg logit contributions | |
Spot | -0.002 |
Fl | -0.001 |
Max | -0.001 |
girls | -0.001 |
Buddy | -0.001 |
Token | of |
Token ID | 286 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
canvas | +0.032 |
product | +0.031 |
tide | +0.030 |
line | +0.030 |
lee | +0.029 |
Neg logit contributions | |
Spot | -0.020 |
girls | -0.019 |
Buddy | -0.018 |
Zip | -0.018 |
Max | -0.018 |
Token | chocolate |
Token ID | 11311 |
Feature activation | +5.89e-01 |
Pos logit contributions | |
girls | +0.066 |
uddy | +0.065 |
Zip | +0.061 |
Buddy | +0.061 |
Monkey | +0.061 |
Neg logit contributions | |
canvas | -0.066 |
product | -0.065 |
tide | -0.062 |
pumpkin | -0.061 |
line | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
Spot | +0.017 |
girls | +0.017 |
Buddy | +0.017 |
Zip | +0.016 |
Fl | +0.016 |
Neg logit contributions | |
canvas | -0.022 |
line | -0.022 |
product | -0.022 |
tide | -0.021 |
lee | -0.021 |
Token | They |
Token ID | 1119 |
Feature activation | -3.73e-01 |
Pos logit contributions | |
girls | +0.005 |
Spot | +0.005 |
Buddy | +0.005 |
Zip | +0.005 |
oof | +0.005 |
Neg logit contributions | |
canvas | -0.008 |
line | -0.008 |
product | -0.008 |
tide | -0.008 |
lee | -0.008 |
Token | stuffed |
Token ID | 22259 |
Feature activation | +4.07e+00 |
Pos logit contributions | |
canvas | +0.012 |
line | +0.012 |
product | +0.011 |
tide | +0.011 |
shell | +0.011 |
Neg logit contributions | |
Zip | -0.013 |
girls | -0.013 |
oof | -0.013 |
Buddy | -0.013 |
uddy | -0.013 |
Token | them |
Token ID | 606 |
Feature activation | +5.63e-01 |
Pos logit contributions | |
uddy | +0.156 |
ver | +0.151 |
OK | +0.150 |
Deal | +0.143 |
girls | +0.142 |
Neg logit contributions | |
pumpkin | -0.119 |
pulp | -0.112 |
canvas | -0.110 |
lee | -0.108 |
product | -0.107 |
Token | in |
Token ID | 287 |
Feature activation | +9.41e-02 |
Pos logit contributions | |
girls | +0.017 |
oof | +0.017 |
uddy | +0.017 |
Zip | +0.017 |
Buddy | +0.016 |
Neg logit contributions | |
canvas | -0.021 |
product | -0.020 |
line | -0.020 |
pumpkin | -0.020 |
shell | -0.019 |
Token | their |
Token ID | 511 |
Feature activation | +9.73e-02 |
Pos logit contributions | |
girls | +0.003 |
uddy | +0.003 |
Deal | +0.003 |
oof | +0.003 |
Buddy | +0.003 |
Neg logit contributions | |
product | -0.003 |
canvas | -0.003 |
shell | -0.003 |
pumpkin | -0.003 |
line | -0.003 |
Token | mouths |
Token ID | 28552 |
Feature activation | +5.18e-01 |
Pos logit contributions | |
girls | +0.004 |
Spot | +0.004 |
Zip | +0.004 |
Buddy | +0.004 |
uddy | +0.004 |
Neg logit contributions | |
canvas | -0.003 |
product | -0.003 |
line | -0.003 |
tide | -0.003 |
lee | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
girls | +0.015 |
Buddy | +0.015 |
Zip | +0.015 |
uddy | +0.014 |
oof | +0.014 |
Neg logit contributions | |
line | -0.021 |
canvas | -0.020 |
product | -0.020 |
tide | -0.019 |
shell | -0.019 |
Token | shelves |
Token ID | 20726 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
Spot | +0.015 |
girls | +0.014 |
Buddy | +0.014 |
Zip | +0.014 |
Fl | +0.014 |
Neg logit contributions | |
canvas | -0.017 |
line | -0.016 |
product | -0.016 |
tide | -0.016 |
lee | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | +4.72e-01 |
Pos logit contributions | |
line | +0.016 |
canvas | +0.015 |
product | +0.015 |
tide | +0.014 |
lee | +0.014 |
Neg logit contributions | |
oof | -0.012 |
girls | -0.012 |
Spot | -0.011 |
Buddy | -0.011 |
Zip | -0.011 |
Token | old |
Token ID | 1468 |
Feature activation | -8.75e-01 |
Pos logit contributions | |
Spot | +0.011 |
girls | +0.011 |
Zip | +0.011 |
Buddy | +0.011 |
oof | +0.010 |
Neg logit contributions | |
canvas | -0.021 |
line | -0.020 |
product | -0.020 |
tide | -0.020 |
lee | -0.019 |
Token | blankets |
Token ID | 34794 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
canvas | +0.029 |
product | +0.028 |
line | +0.027 |
lee | +0.026 |
tide | +0.026 |
Neg logit contributions | |
Spot | -0.031 |
Zip | -0.030 |
oof | -0.030 |
girls | -0.030 |
Buddy | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | +6.77e-02 |
Pos logit contributions | |
oof | +0.041 |
girls | +0.040 |
Zip | +0.038 |
Spot | +0.038 |
Buddy | +0.037 |
Neg logit contributions | |
line | -0.053 |
canvas | -0.051 |
product | -0.048 |
harbor | -0.047 |
tide | -0.047 |
Token | stuffed |
Token ID | 22259 |
Feature activation | +4.07e+00 |
Pos logit contributions | |
Spot | +0.002 |
girls | +0.002 |
Zip | +0.002 |
uddy | +0.002 |
oof | +0.002 |
Neg logit contributions | |
canvas | -0.002 |
product | -0.002 |
line | -0.002 |
tide | -0.002 |
lee | -0.002 |
Token | animals |
Token ID | 4695 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
ver | +0.161 |
oof | +0.159 |
OK | +0.156 |
uddy | +0.156 |
Deal | +0.155 |
Neg logit contributions | |
lee | -0.109 |
canvas | -0.109 |
product | -0.105 |
pans | -0.104 |
floors | -0.102 |
Token | . |
Token ID | 13 |
Feature activation | -4.89e-01 |
Pos logit contributions | |
oof | +0.004 |
Spot | +0.004 |
Zip | +0.004 |
Buddy | +0.004 |
girls | +0.004 |
Neg logit contributions | |
line | -0.004 |
canvas | -0.004 |
product | -0.004 |
harbor | -0.003 |
shell | -0.003 |
Token | She |
Token ID | 1375 |
Feature activation | -7.38e-01 |
Pos logit contributions | |
canvas | +0.016 |
line | +0.015 |
lee | +0.015 |
product | +0.014 |
harbor | +0.013 |
Neg logit contributions | |
oof | -0.018 |
girls | -0.018 |
losers | -0.016 |
uddy | -0.016 |
ever | -0.015 |
Token | worked |
Token ID | 3111 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
canvas | +0.025 |
product | +0.025 |
line | +0.024 |
lee | +0.023 |
tide | +0.023 |
Neg logit contributions | |
Spot | -0.025 |
girls | -0.025 |
Zip | -0.024 |
Buddy | -0.024 |
oof | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | +4.36e-01 |
Pos logit contributions | |
girls | +0.036 |
oof | +0.034 |
Spot | +0.033 |
uddy | +0.032 |
Enjoy | +0.031 |
Neg logit contributions | |
product | -0.043 |
canvas | -0.040 |
lee | -0.039 |
line | -0.039 |
tide | -0.036 |
Token | Inside |
Token ID | 14384 |
Feature activation | +4.14e-01 |
Pos logit contributions | |
canvas | +0.014 |
line | +0.014 |
product | +0.014 |
lee | +0.014 |
tide | +0.013 |
Neg logit contributions | |
girls | -0.009 |
oof | -0.008 |
Spot | -0.008 |
Zip | -0.008 |
Wh | -0.008 |
Token | were |
Token ID | 547 |
Feature activation | +7.20e-01 |
Pos logit contributions | |
girls | +0.013 |
oof | +0.012 |
Zip | +0.012 |
Spot | +0.012 |
Wh | +0.012 |
Neg logit contributions | |
product | -0.015 |
line | -0.015 |
canvas | -0.015 |
tide | -0.015 |
harbor | -0.015 |
Token | all |
Token ID | 477 |
Feature activation | +6.06e-01 |
Pos logit contributions | |
uddy | +0.022 |
girls | +0.022 |
oof | +0.022 |
isk | +0.021 |
Deal | +0.021 |
Neg logit contributions | |
tide | -0.026 |
canvas | -0.025 |
lee | -0.025 |
product | -0.025 |
harbor | -0.025 |
Token | kinds |
Token ID | 6982 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
Spot | +0.017 |
Buddy | +0.016 |
girls | +0.016 |
Zip | +0.016 |
Max | +0.015 |
Neg logit contributions | |
canvas | -0.024 |
line | -0.024 |
product | -0.023 |
tide | -0.023 |
lee | -0.022 |
Token | of |
Token ID | 286 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
Spot | +0.009 |
girls | +0.009 |
Buddy | +0.008 |
Zip | +0.008 |
Wh | +0.008 |
Neg logit contributions | |
canvas | -0.015 |
line | -0.015 |
product | -0.015 |
tide | -0.015 |
lee | -0.014 |
Token | stuffed |
Token ID | 22259 |
Feature activation | +4.03e+00 |
Pos logit contributions | |
uddy | +0.099 |
girls | +0.098 |
isk | +0.095 |
ver | +0.095 |
Spot | +0.094 |
Neg logit contributions | |
canvas | -0.080 |
tide | -0.079 |
product | -0.079 |
lee | -0.077 |
shell | -0.073 |
Token | animals |
Token ID | 4695 |
Feature activation | +5.56e-01 |
Pos logit contributions | |
ver | +0.185 |
OK | +0.181 |
Okay | +0.170 |
Deal | +0.170 |
uddy | +0.168 |
Neg logit contributions | |
lee | -0.111 |
shells | -0.098 |
pumpkin | -0.097 |
waves | -0.097 |
shell | -0.092 |
Token | , |
Token ID | 11 |
Feature activation | -8.37e-02 |
Pos logit contributions | |
oof | +0.015 |
Spot | +0.015 |
girls | +0.015 |
Wh | +0.015 |
Zip | +0.015 |
Neg logit contributions | |
line | -0.021 |
canvas | -0.021 |
product | -0.021 |
tide | -0.020 |
harbor | -0.020 |
Token | games |
Token ID | 1830 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
canvas | +0.003 |
tide | +0.003 |
pumpkin | +0.003 |
line | +0.003 |
product | +0.003 |
Neg logit contributions | |
uddy | -0.003 |
ver | -0.003 |
girls | -0.003 |
isk | -0.003 |
Deal | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
girls | +0.013 |
oof | +0.013 |
Spot | +0.013 |
Zip | +0.013 |
Wh | +0.013 |
Neg logit contributions | |
line | -0.019 |
canvas | -0.018 |
product | -0.018 |
tide | -0.017 |
harbor | -0.017 |
Token | humor |
Token ID | 14733 |
Feature activation | +6.58e-01 |
Pos logit contributions | |
uddy | +0.010 |
girls | +0.010 |
ver | +0.010 |
Deal | +0.010 |
oof | +0.009 |
Neg logit contributions | |
canvas | -0.010 |
tide | -0.010 |
line | -0.010 |
pumpkin | -0.010 |
harbor | -0.010 |
Token | know |
Token ID | 760 |
Feature activation | -6.44e-01 |
Pos logit contributions | |
girls | +0.000 |
Buddy | +0.000 |
Spot | +0.000 |
Zip | +0.000 |
Kitty | +0.000 |
Neg logit contributions | |
product | -0.000 |
canvas | -0.000 |
line | -0.000 |
tide | -0.000 |
shell | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | -6.14e-01 |
Pos logit contributions | |
canvas | +0.025 |
tide | +0.024 |
line | +0.024 |
product | +0.023 |
lee | +0.023 |
Neg logit contributions | |
girls | -0.018 |
Spot | -0.018 |
uddy | -0.018 |
Buddy | -0.017 |
Zip | -0.017 |
Token | right |
Token ID | 826 |
Feature activation | -5.08e-01 |
Pos logit contributions | |
canvas | +0.024 |
line | +0.024 |
tide | +0.023 |
shell | +0.022 |
product | +0.022 |
Neg logit contributions | |
girls | -0.018 |
uddy | -0.017 |
oof | -0.017 |
Spot | -0.017 |
Buddy | -0.017 |
Token | ? |
Token ID | 30 |
Feature activation | -5.46e-01 |
Pos logit contributions | |
line | +0.024 |
canvas | +0.023 |
product | +0.022 |
lee | +0.022 |
pulp | +0.022 |
Neg logit contributions | |
Spot | -0.011 |
Max | -0.010 |
Zip | -0.010 |
Fl | -0.010 |
Buddy | -0.010 |
Token | Micro |
Token ID | 4527 |
Feature activation | -5.33e-01 |
Pos logit contributions | |
canvas | +0.020 |
line | +0.020 |
tide | +0.019 |
product | +0.019 |
lee | +0.018 |
Neg logit contributions | |
girls | -0.016 |
Spot | -0.015 |
Buddy | -0.015 |
oof | -0.015 |
uddy | -0.015 |
Token | sc |
Token ID | 1416 |
Feature activation | +3.97e+00 |
Pos logit contributions | |
canvas | +0.020 |
line | +0.019 |
tide | +0.018 |
product | +0.018 |
lee | +0.018 |
Neg logit contributions | |
girls | -0.016 |
Spot | -0.016 |
Buddy | -0.015 |
uddy | -0.015 |
Johnson | -0.015 |
Token | opes |
Token ID | 13920 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
Spot | +0.167 |
Buddy | +0.161 |
Zip | +0.161 |
girls | +0.161 |
oof | +0.159 |
Neg logit contributions | |
canvas | -0.101 |
line | -0.099 |
product | -0.097 |
tide | -0.095 |
harbor | -0.090 |
Token | are |
Token ID | 389 |
Feature activation | +3.96e-01 |
Pos logit contributions | |
Spot | +0.007 |
girls | +0.007 |
Buddy | +0.007 |
Zip | +0.006 |
oof | +0.006 |
Neg logit contributions | |
canvas | -0.011 |
line | -0.011 |
product | -0.010 |
tide | -0.010 |
harbor | -0.010 |
Token | amazing |
Token ID | 4998 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
uddy | +0.012 |
girls | +0.011 |
oof | +0.011 |
Spot | +0.011 |
Johnson | +0.011 |
Neg logit contributions | |
canvas | -0.014 |
product | -0.014 |
tide | -0.014 |
line | -0.014 |
shell | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
Spot | +0.008 |
uddy | +0.008 |
Zip | +0.008 |
oof | +0.008 |
Buddy | +0.008 |
Neg logit contributions | |
canvas | -0.008 |
product | -0.008 |
tide | -0.008 |
lee | -0.008 |
line | -0.007 |
Token | They |
Token ID | 1119 |
Feature activation | -6.97e-01 |
Pos logit contributions | |
girls | +0.003 |
Johnson | +0.003 |
isk | +0.003 |
oof | +0.003 |
ll | +0.003 |
Neg logit contributions | |
line | -0.004 |
canvas | -0.004 |
tide | -0.004 |
harbor | -0.004 |
lee | -0.004 |
Token | place |
Token ID | 1295 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
canvas | +0.008 |
tide | +0.007 |
product | +0.007 |
pumpkin | +0.007 |
harbor | +0.007 |
Neg logit contributions | |
girls | -0.009 |
Zip | -0.009 |
Spot | -0.008 |
oof | -0.008 |
Buddy | -0.008 |
Token | of |
Token ID | 286 |
Feature activation | +7.44e-01 |
Pos logit contributions | |
canvas | +0.012 |
line | +0.012 |
product | +0.012 |
tide | +0.012 |
shell | +0.011 |
Neg logit contributions | |
girls | -0.007 |
Spot | -0.007 |
Zip | -0.007 |
oof | -0.007 |
Buddy | -0.007 |
Token | peace |
Token ID | 4167 |
Feature activation | -6.15e-01 |
Pos logit contributions | |
uddy | +0.030 |
ver | +0.029 |
girls | +0.028 |
Zip | +0.026 |
oof | +0.026 |
Neg logit contributions | |
tide | -0.023 |
canvas | -0.022 |
product | -0.022 |
shell | -0.022 |
pumpkin | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | -5.91e-01 |
Pos logit contributions | |
canvas | +0.022 |
line | +0.022 |
product | +0.021 |
tide | +0.021 |
harbor | +0.021 |
Neg logit contributions | |
Spot | -0.019 |
girls | -0.019 |
Zip | -0.018 |
Buddy | -0.018 |
oof | -0.018 |
Token | content |
Token ID | 2695 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
canvas | +0.019 |
tide | +0.019 |
line | +0.018 |
product | +0.018 |
shell | +0.018 |
Neg logit contributions | |
uddy | -0.022 |
girls | -0.020 |
oof | -0.020 |
Zip | -0.020 |
Johnson | -0.020 |
Token | ment |
Token ID | 434 |
Feature activation | -3.87e+00 |
Pos logit contributions | |
canvas | +0.049 |
product | +0.048 |
line | +0.048 |
tide | +0.046 |
lee | +0.045 |
Neg logit contributions | |
Spot | -0.045 |
girls | -0.043 |
Buddy | -0.043 |
Zip | -0.042 |
Fl | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
line | +0.093 |
product | +0.081 |
canvas | +0.079 |
tide | +0.075 |
shell | +0.073 |
Neg logit contributions | |
uddy | -0.167 |
oof | -0.166 |
Zip | -0.162 |
Wh | -0.161 |
Monkey | -0.159 |
Token | |
Token ID | 220 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
line | +0.012 |
lee | +0.011 |
canvas | +0.011 |
tide | +0.011 |
harbor | +0.011 |
Neg logit contributions | |
girls | -0.011 |
losers | -0.010 |
oof | -0.010 |
ever | -0.010 |
isk | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -6.14e-01 |
Pos logit contributions | |
line | +0.012 |
product | +0.011 |
canvas | +0.011 |
lee | +0.011 |
tide | +0.010 |
Neg logit contributions | |
girls | -0.009 |
oof | -0.008 |
Monkey | -0.008 |
Spot | -0.008 |
Buddy | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -7.23e-01 |
Pos logit contributions | |
line | +0.021 |
product | +0.019 |
canvas | +0.019 |
tide | +0.018 |
harbor | +0.018 |
Neg logit contributions | |
girls | -0.020 |
Spot | -0.018 |
losers | -0.018 |
Wh | -0.018 |
oof | -0.018 |
Token | One |
Token ID | 3198 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
line | +0.027 |
canvas | +0.026 |
product | +0.026 |
tide | +0.026 |
harbor | +0.025 |
Neg logit contributions | |
girls | -0.021 |
Spot | -0.020 |
Buddy | -0.019 |
Max | -0.019 |
Fl | -0.019 |
Token | stopped |
Token ID | 5025 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
Spot | +0.018 |
Fl | +0.017 |
Will | +0.017 |
Buddy | +0.017 |
Max | +0.017 |
Neg logit contributions | |
product | -0.027 |
canvas | -0.026 |
line | -0.025 |
tide | -0.024 |
harbor | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
canvas | +0.011 |
line | +0.010 |
product | +0.010 |
tide | +0.010 |
lee | +0.010 |
Neg logit contributions | |
Spot | -0.006 |
girls | -0.006 |
Buddy | -0.005 |
Zip | -0.005 |
Fl | -0.005 |
Token | froze |
Token ID | 37785 |
Feature activation | -9.48e-01 |
Pos logit contributions | |
canvas | +0.009 |
product | +0.009 |
line | +0.009 |
tide | +0.008 |
lee | +0.008 |
Neg logit contributions | |
Spot | -0.010 |
Buddy | -0.009 |
girls | -0.009 |
Zip | -0.009 |
Max | -0.009 |
Token | in |
Token ID | 287 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
canvas | +0.041 |
product | +0.040 |
line | +0.039 |
tide | +0.037 |
rys | +0.037 |
Neg logit contributions | |
Spot | -0.023 |
Buddy | -0.021 |
Max | -0.021 |
Fl | -0.021 |
Zip | -0.021 |
Token | amaz |
Token ID | 40642 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
line | +0.019 |
canvas | +0.019 |
product | +0.018 |
tide | +0.018 |
lee | +0.017 |
Neg logit contributions | |
Spot | -0.019 |
Buddy | -0.018 |
Max | -0.018 |
Zip | -0.018 |
girls | -0.018 |
Token | ment |
Token ID | 434 |
Feature activation | -3.78e+00 |
Pos logit contributions | |
line | +0.071 |
canvas | +0.070 |
product | +0.066 |
harbor | +0.066 |
pulp | +0.065 |
Neg logit contributions | |
Spot | -0.052 |
Max | -0.049 |
just | -0.049 |
Zip | -0.048 |
Buddy | -0.048 |
Token | , |
Token ID | 11 |
Feature activation | -6.37e-01 |
Pos logit contributions | |
canvas | +0.103 |
line | +0.101 |
product | +0.097 |
tide | +0.097 |
harbor | +0.089 |
Neg logit contributions | |
uddy | -0.156 |
girls | -0.153 |
oof | -0.151 |
Buddy | -0.150 |
Monkey | -0.148 |
Token | watching |
Token ID | 4964 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
line | +0.025 |
canvas | +0.024 |
tide | +0.023 |
product | +0.023 |
lee | +0.022 |
Neg logit contributions | |
girls | -0.019 |
uddy | -0.018 |
oof | -0.018 |
Buddy | -0.018 |
Zip | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
canvas | +0.037 |
line | +0.035 |
tide | +0.035 |
product | +0.035 |
lee | +0.034 |
Neg logit contributions | |
girls | -0.031 |
uddy | -0.030 |
Spot | -0.030 |
oof | -0.030 |
Zip | -0.029 |
Token | bright |
Token ID | 6016 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
canvas | +0.010 |
line | +0.010 |
product | +0.010 |
tide | +0.009 |
lee | +0.009 |
Neg logit contributions | |
Spot | -0.012 |
girls | -0.011 |
Buddy | -0.011 |
Zip | -0.011 |
Max | -0.011 |
Token | colours |
Token ID | 18915 |
Feature activation | -8.67e-01 |
Pos logit contributions | |
line | +0.045 |
canvas | +0.043 |
product | +0.042 |
lee | +0.042 |
rys | +0.042 |
Neg logit contributions | |
Spot | -0.041 |
Buddy | -0.039 |
George | -0.039 |
Max | -0.039 |
Bob | -0.038 |
Token | : |
Token ID | 25 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
line | +0.009 |
canvas | +0.009 |
tide | +0.009 |
product | +0.009 |
harbor | +0.008 |
Neg logit contributions | |
girls | -0.009 |
uddy | -0.009 |
oof | -0.009 |
Buddy | -0.008 |
Zip | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -3.37e-01 |
Pos logit contributions | |
line | +0.054 |
tide | +0.052 |
canvas | +0.052 |
product | +0.050 |
pumpkin | +0.050 |
Neg logit contributions | |
girls | -0.052 |
uddy | -0.048 |
oof | -0.046 |
isk | -0.045 |
Buddy | -0.045 |
Token |
|
Token ID | 198 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
canvas | +0.014 |
line | +0.014 |
product | +0.014 |
tide | +0.014 |
lee | +0.014 |
Neg logit contributions | |
girls | -0.008 |
Spot | -0.008 |
Buddy | -0.007 |
Zip | -0.007 |
Fl | -0.007 |
Token | " |
Token ID | 1 |
Feature activation | -7.43e-01 |
Pos logit contributions | |
canvas | +0.010 |
line | +0.010 |
product | +0.010 |
tide | +0.010 |
lee | +0.009 |
Neg logit contributions | |
Spot | -0.009 |
girls | -0.009 |
Buddy | -0.009 |
Zip | -0.009 |
Fl | -0.008 |
Token | Measure |
Token ID | 47384 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
line | +0.026 |
canvas | +0.025 |
tide | +0.025 |
product | +0.024 |
harbor | +0.023 |
Neg logit contributions | |
girls | -0.024 |
Spot | -0.023 |
Buddy | -0.023 |
uddy | -0.022 |
Max | -0.022 |
Token | ment |
Token ID | 434 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
girls | +0.008 |
Spot | +0.008 |
Buddy | +0.008 |
Zip | +0.008 |
uddy | +0.008 |
Neg logit contributions | |
canvas | -0.009 |
product | -0.009 |
line | -0.008 |
tide | -0.008 |
lee | -0.008 |
Token | complete |
Token ID | 1844 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
canvas | +0.150 |
line | +0.141 |
product | +0.141 |
tide | +0.136 |
pumpkin | +0.131 |
Neg logit contributions | |
girls | -0.102 |
Spot | -0.100 |
uddy | -0.099 |
Buddy | -0.098 |
Wh | -0.098 |
Token | . |
Token ID | 13 |
Feature activation | +6.60e-01 |
Pos logit contributions | |
girls | +0.008 |
uddy | +0.007 |
Spot | +0.007 |
Buddy | +0.007 |
Zip | +0.007 |
Neg logit contributions | |
canvas | -0.009 |
product | -0.009 |
line | -0.008 |
tide | -0.008 |
lee | -0.008 |
Token | Height |
Token ID | 27280 |
Feature activation | +8.66e-01 |
Pos logit contributions | |
girls | +0.019 |
Spot | +0.018 |
Buddy | +0.018 |
uddy | +0.017 |
Zip | +0.017 |
Neg logit contributions | |
canvas | -0.026 |
line | -0.025 |
tide | -0.024 |
product | -0.024 |
lee | -0.023 |
Token | : |
Token ID | 25 |
Feature activation | -4.59e-01 |
Pos logit contributions | |
Spot | +0.031 |
girls | +0.031 |
Buddy | +0.030 |
Zip | +0.030 |
Fl | +0.030 |
Neg logit contributions | |
canvas | -0.027 |
line | -0.026 |
product | -0.026 |
tide | -0.025 |
lee | -0.024 |
Token | 90 |
Token ID | 4101 |
Feature activation | -7.10e-01 |
Pos logit contributions | |
canvas | +0.017 |
line | +0.016 |
product | +0.015 |
tide | +0.015 |
pumpkin | +0.015 |
Neg logit contributions | |
girls | -0.015 |
Spot | -0.015 |
Buddy | -0.014 |
uddy | -0.014 |
Zip | -0.014 |
Token | we |
Token ID | 356 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
girls | +0.005 |
oof | +0.005 |
uddy | +0.005 |
Zip | +0.005 |
Buddy | +0.005 |
Neg logit contributions | |
canvas | -0.007 |
line | -0.006 |
tide | -0.006 |
product | -0.006 |
pumpkin | -0.006 |
Token | might |
Token ID | 1244 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
canvas | +0.043 |
line | +0.042 |
tide | +0.041 |
pumpkin | +0.041 |
shell | +0.041 |
Neg logit contributions | |
Zip | -0.027 |
oof | -0.026 |
Buddy | -0.026 |
Spot | -0.026 |
Kenny | -0.025 |
Token | break |
Token ID | 2270 |
Feature activation | -8.36e-03 |
Pos logit contributions | |
canvas | +0.012 |
line | +0.012 |
product | +0.012 |
tide | +0.012 |
lee | +0.011 |
Neg logit contributions | |
Spot | -0.010 |
girls | -0.010 |
Buddy | -0.010 |
Zip | -0.010 |
Fl | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | -4.08e-01 |
Pos logit contributions | |
canvas | +0.000 |
line | +0.000 |
product | +0.000 |
tide | +0.000 |
pumpkin | +0.000 |
Neg logit contributions | |
uddy | -0.000 |
girls | -0.000 |
Zip | -0.000 |
Spot | -0.000 |
Buddy | -0.000 |
Token | st |
Token ID | 336 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
canvas | +0.011 |
line | +0.010 |
product | +0.010 |
tide | +0.010 |
harbor | +0.010 |
Neg logit contributions | |
Spot | -0.017 |
girls | -0.016 |
Buddy | -0.016 |
Zip | -0.016 |
Wh | -0.016 |
Token | alks |
Token ID | 23833 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
canvas | +0.055 |
tide | +0.053 |
product | +0.052 |
line | +0.052 |
lee | +0.050 |
Neg logit contributions | |
girls | -0.062 |
uddy | -0.060 |
Spot | -0.059 |
Fl | -0.057 |
oof | -0.057 |
Token | ." |
Token ID | 526 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
canvas | +0.145 |
line | +0.144 |
product | +0.140 |
tide | +0.138 |
lee | +0.132 |
Neg logit contributions | |
Spot | -0.100 |
girls | -0.099 |
Buddy | -0.097 |
Zip | -0.096 |
oof | -0.093 |
Token |
|
Token ID | 198 |
Feature activation | -5.02e-01 |
Pos logit contributions | |
line | +0.007 |
canvas | +0.007 |
product | +0.006 |
lee | +0.006 |
tide | +0.006 |
Neg logit contributions | |
girls | -0.004 |
Zip | -0.004 |
oof | -0.004 |
Buddy | -0.004 |
uddy | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -5.03e-01 |
Pos logit contributions | |
line | +0.018 |
product | +0.018 |
canvas | +0.017 |
lee | +0.017 |
tide | +0.016 |
Neg logit contributions | |
girls | -0.015 |
Monkey | -0.014 |
Zip | -0.014 |
oof | -0.014 |
Buddy | -0.013 |
Token | L |
Token ID | 43 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
canvas | +0.020 |
line | +0.020 |
product | +0.020 |
tide | +0.019 |
lee | +0.019 |
Neg logit contributions | |
Spot | -0.014 |
girls | -0.014 |
Buddy | -0.013 |
Zip | -0.013 |
Fl | -0.013 |
Token | ily |
Token ID | 813 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
canvas | +0.009 |
line | +0.009 |
product | +0.009 |
lee | +0.009 |
tide | +0.009 |
Neg logit contributions | |
girls | -0.012 |
Spot | -0.012 |
Buddy | -0.012 |
Monkey | -0.011 |
Zip | -0.011 |
Token | looked |
Token ID | 3114 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
Spot | +0.008 |
girls | +0.007 |
Buddy | +0.007 |
Zip | +0.007 |
Max | +0.007 |
Neg logit contributions | |
canvas | -0.008 |
line | -0.008 |
product | -0.008 |
tide | -0.008 |
lee | -0.008 |
Token | through |
Token ID | 832 |
Feature activation | -5.77e-01 |
Pos logit contributions | |
canvas | +0.060 |
product | +0.059 |
line | +0.058 |
tide | +0.057 |
harbor | +0.055 |
Neg logit contributions | |
Spot | -0.030 |
Max | -0.027 |
Buddy | -0.026 |
Fl | -0.026 |
girls | -0.026 |
Token | the |
Token ID | 262 |
Feature activation | -4.42e-01 |
Pos logit contributions | |
canvas | +0.022 |
product | +0.022 |
line | +0.021 |
tide | +0.021 |
lee | +0.021 |
Neg logit contributions | |
Spot | -0.017 |
girls | -0.017 |
Buddy | -0.016 |
Zip | -0.016 |
uddy | -0.016 |
Token | ey |
Token ID | 1926 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
line | +0.011 |
product | +0.011 |
canvas | +0.010 |
lee | +0.010 |
tide | +0.010 |
Neg logit contributions | |
Spot | -0.018 |
Max | -0.017 |
Fl | -0.017 |
girls | -0.017 |
Buddy | -0.017 |
Token | ep |
Token ID | 538 |
Feature activation | -8.23e-01 |
Pos logit contributions | |
canvas | +0.114 |
harbor | +0.110 |
line | +0.109 |
tide | +0.108 |
product | +0.105 |
Neg logit contributions | |
Zip | -0.107 |
Spot | -0.107 |
girls | -0.104 |
oof | -0.103 |
Buddy | -0.102 |
Token | iece |
Token ID | 8535 |
Feature activation | -3.54e+00 |
Pos logit contributions | |
line | +0.023 |
canvas | +0.020 |
product | +0.020 |
tide | +0.019 |
lee | +0.019 |
Neg logit contributions | |
Buddy | -0.035 |
Spot | -0.034 |
oof | -0.034 |
girls | -0.034 |
Kitty | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
canvas | +0.127 |
line | +0.125 |
product | +0.123 |
tide | +0.121 |
lee | +0.117 |
Neg logit contributions | |
Spot | -0.111 |
girls | -0.108 |
Buddy | -0.107 |
Zip | -0.105 |
Wh | -0.103 |
Token | He |
Token ID | 679 |
Feature activation | -6.44e-01 |
Pos logit contributions | |
girls | +0.001 |
oof | +0.001 |
Monkey | +0.001 |
uddy | +0.001 |
Buddy | +0.001 |
Neg logit contributions | |
line | -0.001 |
canvas | -0.001 |
tide | -0.001 |
lee | -0.001 |
product | -0.001 |
Token | saw |
Token ID | 2497 |
Feature activation | -4.27e-01 |
Pos logit contributions | |
canvas | +0.018 |
line | +0.018 |
product | +0.018 |
tide | +0.017 |
lee | +0.017 |
Neg logit contributions | |
girls | -0.024 |
Buddy | -0.024 |
Spot | -0.024 |
uddy | -0.024 |
oof | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
canvas | +0.017 |
line | +0.017 |
tide | +0.017 |
product | +0.016 |
lee | +0.016 |
Neg logit contributions | |
girls | -0.012 |
Spot | -0.011 |
uddy | -0.011 |
Buddy | -0.011 |
oof | -0.011 |
Token | tiny |
Token ID | 7009 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
canvas | +0.001 |
line | +0.001 |
product | +0.001 |
tide | +0.001 |
lee | +0.001 |
Neg logit contributions | |
Spot | -0.001 |
girls | -0.001 |
Buddy | -0.001 |
Zip | -0.001 |
Max | -0.001 |
Token | fingers |
Token ID | 9353 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
canvas | +0.033 |
product | +0.032 |
shell | +0.031 |
tide | +0.030 |
line | +0.030 |
Neg logit contributions | |
Spot | -0.041 |
Buddy | -0.041 |
Zip | -0.041 |
girls | -0.041 |
uddy | -0.040 |
Token | and |
Token ID | 290 |
Feature activation | -9.08e-01 |
Pos logit contributions | |
line | +0.077 |
canvas | +0.077 |
product | +0.075 |
tide | +0.073 |
lee | +0.071 |
Neg logit contributions | |
girls | -0.053 |
Spot | -0.053 |
Buddy | -0.051 |
Zip | -0.051 |
oof | -0.050 |
Token | feel |
Token ID | 1254 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
canvas | +0.031 |
product | +0.030 |
line | +0.029 |
tide | +0.029 |
shell | +0.028 |
Neg logit contributions | |
Spot | -0.030 |
uddy | -0.030 |
girls | -0.030 |
oof | -0.029 |
Zip | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
canvas | +0.088 |
product | +0.085 |
line | +0.084 |
tide | +0.082 |
lee | +0.081 |
Neg logit contributions | |
Spot | -0.038 |
Fl | -0.035 |
Buddy | -0.034 |
Max | -0.033 |
love | -0.032 |
Token | smooth |
Token ID | 7209 |
Feature activation | -2.36e+00 |
Pos logit contributions | |
canvas | +0.044 |
line | +0.044 |
product | +0.043 |
lee | +0.042 |
tide | +0.041 |
Neg logit contributions | |
Spot | -0.047 |
Fl | -0.044 |
girls | -0.044 |
Buddy | -0.044 |
Max | -0.043 |
Token | ness |
Token ID | 1108 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
canvas | +0.081 |
lee | +0.079 |
line | +0.077 |
product | +0.075 |
tide | +0.075 |
Neg logit contributions | |
Spot | -0.074 |
Fl | -0.069 |
girls | -0.068 |
Zip | -0.067 |
Wh | -0.066 |
Token | between |
Token ID | 1022 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
canvas | +0.148 |
product | +0.144 |
line | +0.143 |
tide | +0.140 |
lee | +0.140 |
Neg logit contributions | |
Spot | -0.090 |
Buddy | -0.083 |
Zip | -0.082 |
girls | -0.082 |
Fl | -0.081 |
Token | his |
Token ID | 465 |
Feature activation | -7.19e-01 |
Pos logit contributions | |
canvas | +0.058 |
line | +0.057 |
product | +0.056 |
tide | +0.055 |
harbor | +0.054 |
Neg logit contributions | |
Spot | -0.029 |
girls | -0.027 |
Buddy | -0.027 |
Zip | -0.027 |
Max | -0.026 |
Token | fingers |
Token ID | 9353 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
canvas | +0.019 |
product | +0.018 |
line | +0.018 |
tide | +0.018 |
lee | +0.017 |
Neg logit contributions | |
Spot | -0.029 |
girls | -0.029 |
Buddy | -0.029 |
Zip | -0.029 |
Wh | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | -9.14e-01 |
Pos logit contributions | |
line | +0.075 |
canvas | +0.074 |
product | +0.072 |
tide | +0.070 |
lee | +0.068 |
Neg logit contributions | |
Spot | -0.080 |
girls | -0.079 |
Zip | -0.078 |
Buddy | -0.078 |
oof | -0.077 |
Token | think |
Token ID | 892 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
canvas | +0.032 |
line | +0.031 |
product | +0.031 |
tide | +0.030 |
lee | +0.029 |
Neg logit contributions | |
Spot | -0.029 |
girls | -0.029 |
Buddy | -0.028 |
Zip | -0.028 |
uddy | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
oof | +0.047 |
girls | +0.046 |
Wh | +0.046 |
uddy | +0.046 |
Spot | +0.045 |
Neg logit contributions | |
canvas | -0.046 |
product | -0.046 |
line | -0.045 |
tide | -0.043 |
harbor | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
girls | +0.010 |
uddy | +0.010 |
oof | +0.010 |
Bones | +0.009 |
Monkey | +0.009 |
Neg logit contributions | |
line | -0.012 |
canvas | -0.012 |
harbor | -0.012 |
product | -0.012 |
tide | -0.011 |
Token | enjoyed |
Token ID | 8359 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
oof | +0.008 |
uddy | +0.008 |
girls | +0.008 |
Johnson | +0.007 |
Wh | +0.007 |
Neg logit contributions | |
line | -0.007 |
lee | -0.007 |
canvas | -0.007 |
product | -0.007 |
harbor | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
line | +0.006 |
product | +0.006 |
canvas | +0.006 |
tide | +0.006 |
harbor | +0.006 |
Neg logit contributions | |
uddy | -0.006 |
girls | -0.006 |
oof | -0.005 |
Buddy | -0.005 |
Spot | -0.005 |
Token | refresh |
Token ID | 14976 |
Feature activation | -7.01e-01 |
Pos logit contributions | |
girls | +0.010 |
Spot | +0.010 |
Wh | +0.009 |
uddy | +0.009 |
Zip | +0.009 |
Neg logit contributions | |
tide | -0.009 |
canvas | -0.008 |
harbor | -0.008 |
product | -0.008 |
line | -0.008 |
Token | ment |
Token ID | 434 |
Feature activation | -3.42e+00 |
Pos logit contributions | |
canvas | +0.015 |
line | +0.014 |
pulp | +0.014 |
product | +0.013 |
tide | +0.012 |
Neg logit contributions | |
Spot | -0.030 |
ever | -0.030 |
girls | -0.030 |
Zip | -0.029 |
Bob | -0.028 |
Token | of |
Token ID | 286 |
Feature activation | +9.05e-01 |
Pos logit contributions | |
line | +0.129 |
canvas | +0.127 |
product | +0.123 |
tide | +0.121 |
lee | +0.116 |
Neg logit contributions | |
girls | -0.101 |
oof | -0.100 |
Wh | -0.100 |
Spot | -0.099 |
Buddy | -0.098 |
Token | the |
Token ID | 262 |
Feature activation | -6.86e-02 |
Pos logit contributions | |
uddy | +0.034 |
ver | +0.032 |
girls | +0.031 |
oof | +0.030 |
Monkey | +0.029 |
Neg logit contributions | |
tide | -0.030 |
canvas | -0.029 |
product | -0.028 |
harbor | -0.028 |
pumpkin | -0.028 |
Token | cool |
Token ID | 3608 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
tide | +0.002 |
canvas | +0.002 |
harbor | +0.002 |
product | +0.002 |
shell | +0.002 |
Neg logit contributions | |
girls | -0.003 |
Spot | -0.003 |
Wh | -0.003 |
uddy | -0.003 |
Buddy | -0.003 |
Token | lake |
Token ID | 13546 |
Feature activation | +9.95e-02 |
Pos logit contributions | |
canvas | +0.031 |
line | +0.030 |
product | +0.030 |
lee | +0.028 |
tide | +0.028 |
Neg logit contributions | |
Spot | -0.037 |
Buddy | -0.036 |
Zip | -0.035 |
girls | -0.035 |
Fl | -0.035 |
Token | water |
Token ID | 1660 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
girls | +0.003 |
Spot | +0.003 |
oof | +0.003 |
Zip | +0.003 |
Wh | +0.003 |
Neg logit contributions | |
line | -0.004 |
product | -0.004 |
canvas | -0.004 |
tide | -0.003 |
harbor | -0.003 |
Token | run |
Token ID | 1057 |
Feature activation | -9.47e-01 |
Pos logit contributions | |
canvas | +0.019 |
line | +0.019 |
product | +0.018 |
tide | +0.018 |
harbor | +0.018 |
Neg logit contributions | |
Spot | -0.020 |
girls | -0.019 |
Buddy | -0.019 |
Zip | -0.019 |
uddy | -0.019 |
Token | through |
Token ID | 832 |
Feature activation | -8.81e-01 |
Pos logit contributions | |
canvas | +0.038 |
product | +0.037 |
line | +0.037 |
tide | +0.036 |
lee | +0.035 |
Neg logit contributions | |
Spot | -0.027 |
girls | -0.026 |
Buddy | -0.025 |
Zip | -0.025 |
Fl | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | -4.35e-01 |
Pos logit contributions | |
line | +0.043 |
canvas | +0.041 |
product | +0.041 |
tide | +0.041 |
lee | +0.040 |
Neg logit contributions | |
Spot | -0.018 |
Fl | -0.017 |
Kitty | -0.016 |
Buddy | -0.016 |
Max | -0.015 |
Token | tall |
Token ID | 7331 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
line | +0.013 |
canvas | +0.013 |
product | +0.013 |
complexion | +0.012 |
lee | +0.012 |
Neg logit contributions | |
Spot | -0.016 |
girls | -0.016 |
Buddy | -0.015 |
Max | -0.015 |
Fl | -0.015 |
Token | st |
Token ID | 336 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
canvas | +0.066 |
line | +0.066 |
product | +0.066 |
tide | +0.063 |
rys | +0.063 |
Neg logit contributions | |
Spot | -0.068 |
girls | -0.067 |
Buddy | -0.061 |
Fl | -0.061 |
Max | -0.061 |
Token | alks |
Token ID | 23833 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
canvas | +0.044 |
tide | +0.042 |
product | +0.041 |
pulp | +0.038 |
pumpkin | +0.037 |
Neg logit contributions | |
girls | -0.069 |
uddy | -0.066 |
Spot | -0.064 |
oof | -0.063 |
Wh | -0.062 |
Token | . |
Token ID | 13 |
Feature activation | -4.22e-01 |
Pos logit contributions | |
canvas | +0.133 |
product | +0.131 |
tide | +0.128 |
line | +0.126 |
lee | +0.124 |
Neg logit contributions | |
Spot | -0.093 |
girls | -0.091 |
Buddy | -0.088 |
Zip | -0.087 |
Max | -0.086 |
Token |
|
Token ID | 198 |
Feature activation | -5.77e-01 |
Pos logit contributions | |
canvas | +0.018 |
line | +0.018 |
product | +0.017 |
tide | +0.017 |
lee | +0.017 |
Neg logit contributions | |
girls | -0.011 |
Spot | -0.010 |
Zip | -0.010 |
Buddy | -0.010 |
oof | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -7.82e-01 |
Pos logit contributions | |
canvas | +0.024 |
line | +0.024 |
product | +0.024 |
tide | +0.023 |
lee | +0.023 |
Neg logit contributions | |
girls | -0.014 |
Spot | -0.013 |
Buddy | -0.013 |
Zip | -0.013 |
Wh | -0.012 |
Token | One |
Token ID | 3198 |
Feature activation | +5.14e-01 |
Pos logit contributions | |
canvas | +0.035 |
line | +0.034 |
product | +0.034 |
tide | +0.033 |
harbor | +0.033 |
Neg logit contributions | |
girls | -0.018 |
Spot | -0.018 |
Buddy | -0.017 |
Zip | -0.017 |
Fl | -0.016 |
Token | day |
Token ID | 1110 |
Feature activation | -5.00e-01 |
Pos logit contributions | |
Spot | +0.012 |
Buddy | +0.011 |
Zip | +0.010 |
Fl | +0.010 |
Kitty | +0.010 |
Neg logit contributions | |
canvas | -0.024 |
line | -0.023 |
product | -0.023 |
lee | -0.023 |
tide | -0.022 |
Token | slide |
Token ID | 10649 |
Feature activation | -3.67e-01 |
Pos logit contributions | |
canvas | +0.006 |
line | +0.005 |
tide | +0.005 |
product | +0.005 |
shell | +0.005 |
Neg logit contributions | |
Spot | -0.010 |
girls | -0.010 |
Zip | -0.010 |
Buddy | -0.010 |
uddy | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
line | +0.014 |
canvas | +0.012 |
pumpkin | +0.011 |
tide | +0.011 |
product | +0.011 |
Neg logit contributions | |
oof | -0.013 |
girls | -0.013 |
Buddy | -0.012 |
Enjoy | -0.012 |
Kitty | -0.012 |
Token | looked |
Token ID | 3114 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
Spot | +0.008 |
girls | +0.007 |
Buddy | +0.007 |
Zip | +0.007 |
Max | +0.007 |
Neg logit contributions | |
canvas | -0.008 |
line | -0.008 |
product | -0.008 |
tide | -0.008 |
lee | -0.008 |
Token | through |
Token ID | 832 |
Feature activation | -5.77e-01 |
Pos logit contributions | |
canvas | +0.060 |
product | +0.059 |
line | +0.058 |
tide | +0.057 |
harbor | +0.055 |
Neg logit contributions | |
Spot | -0.030 |
Max | -0.027 |
Buddy | -0.026 |
Fl | -0.026 |
girls | -0.026 |
Token | the |
Token ID | 262 |
Feature activation | -4.42e-01 |
Pos logit contributions | |
canvas | +0.022 |
product | +0.022 |
line | +0.021 |
tide | +0.021 |
lee | +0.021 |
Neg logit contributions | |
Spot | -0.017 |
girls | -0.017 |
Buddy | -0.016 |
Zip | -0.016 |
uddy | -0.016 |
Token | ey |
Token ID | 1926 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
line | +0.011 |
product | +0.011 |
canvas | +0.010 |
lee | +0.010 |
tide | +0.010 |
Neg logit contributions | |
Spot | -0.018 |
Max | -0.017 |
Fl | -0.017 |
girls | -0.017 |
Buddy | -0.017 |
Token | ep |
Token ID | 538 |
Feature activation | -8.23e-01 |
Pos logit contributions | |
canvas | +0.114 |
harbor | +0.110 |
line | +0.109 |
tide | +0.108 |
product | +0.105 |
Neg logit contributions | |
Zip | -0.107 |
Spot | -0.107 |
girls | -0.104 |
oof | -0.103 |
Buddy | -0.102 |
Token | iece |
Token ID | 8535 |
Feature activation | -3.54e+00 |
Pos logit contributions | |
line | +0.023 |
canvas | +0.020 |
product | +0.020 |
tide | +0.019 |
lee | +0.019 |
Neg logit contributions | |
Buddy | -0.035 |
Spot | -0.034 |
oof | -0.034 |
girls | -0.034 |
Kitty | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
canvas | +0.127 |
line | +0.125 |
product | +0.123 |
tide | +0.121 |
lee | +0.117 |
Neg logit contributions | |
Spot | -0.111 |
girls | -0.108 |
Buddy | -0.107 |
Zip | -0.105 |
Wh | -0.103 |
Token | He |
Token ID | 679 |
Feature activation | -6.44e-01 |
Pos logit contributions | |
girls | +0.001 |
oof | +0.001 |
Monkey | +0.001 |
uddy | +0.001 |
Buddy | +0.001 |
Neg logit contributions | |
line | -0.001 |
canvas | -0.001 |
tide | -0.001 |
lee | -0.001 |
product | -0.001 |
Token | saw |
Token ID | 2497 |
Feature activation | -4.27e-01 |
Pos logit contributions | |
canvas | +0.018 |
line | +0.018 |
product | +0.018 |
tide | +0.017 |
lee | +0.017 |
Neg logit contributions | |
girls | -0.024 |
Buddy | -0.024 |
Spot | -0.024 |
uddy | -0.024 |
oof | -0.024 |
Token | rub |
Token ID | 6437 |
Feature activation | +3.36e-01 |
Pos logit contributions | |
canvas | +0.015 |
line | +0.015 |
product | +0.015 |
tide | +0.015 |
shell | +0.014 |
Neg logit contributions | |
uddy | -0.016 |
girls | -0.016 |
Zip | -0.016 |
oof | -0.016 |
Spot | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | -7.55e-01 |
Pos logit contributions | |
girls | +0.011 |
uddy | +0.010 |
Zip | +0.010 |
oof | +0.010 |
Spot | +0.010 |
Neg logit contributions | |
line | -0.012 |
canvas | -0.011 |
product | -0.011 |
tide | -0.011 |
lee | -0.011 |
Token | magical |
Token ID | 10883 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
canvas | +0.020 |
tide | +0.019 |
line | +0.018 |
product | +0.018 |
harbor | +0.018 |
Neg logit contributions | |
Spot | -0.032 |
girls | -0.031 |
Zip | -0.031 |
Buddy | -0.031 |
uddy | -0.031 |
Token | o |
Token ID | 267 |
Feature activation | -6.23e-01 |
Pos logit contributions | |
canvas | +0.032 |
line | +0.031 |
product | +0.031 |
tide | +0.030 |
lee | +0.029 |
Neg logit contributions | |
Spot | -0.043 |
girls | -0.042 |
Buddy | -0.041 |
Zip | -0.041 |
Max | -0.040 |
Token | int |
Token ID | 600 |
Feature activation | +5.93e-03 |
Pos logit contributions | |
canvas | +0.014 |
product | +0.013 |
line | +0.013 |
tide | +0.013 |
lee | +0.012 |
Neg logit contributions | |
girls | -0.027 |
Spot | -0.027 |
Buddy | -0.026 |
Zip | -0.026 |
Fl | -0.026 |
Token | ment |
Token ID | 434 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
Spot | +0.000 |
girls | +0.000 |
Buddy | +0.000 |
Zip | +0.000 |
Max | +0.000 |
Neg logit contributions | |
line | -0.000 |
canvas | -0.000 |
product | -0.000 |
tide | -0.000 |
lee | -0.000 |
Token | on |
Token ID | 319 |
Feature activation | +4.14e-02 |
Pos logit contributions | |
canvas | +0.114 |
line | +0.112 |
tide | +0.106 |
product | +0.105 |
shell | +0.099 |
Neg logit contributions | |
Buddy | -0.117 |
girls | -0.117 |
oof | -0.117 |
Zip | -0.117 |
Spot | -0.117 |
Token | her |
Token ID | 607 |
Feature activation | -2.39e-01 |
Pos logit contributions | |
girls | +0.001 |
uddy | +0.001 |
oof | +0.001 |
ver | +0.001 |
Zip | +0.001 |
Neg logit contributions | |
canvas | -0.001 |
product | -0.001 |
tide | -0.001 |
line | -0.001 |
lee | -0.001 |
Token | forehead |
Token ID | 22645 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
canvas | +0.006 |
line | +0.006 |
product | +0.006 |
tide | +0.006 |
lee | +0.006 |
Neg logit contributions | |
Spot | -0.010 |
girls | -0.010 |
Buddy | -0.009 |
Zip | -0.009 |
Max | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
girls | +0.000 |
Spot | +0.000 |
Zip | +0.000 |
Buddy | +0.000 |
oof | +0.000 |
Neg logit contributions | |
line | -0.000 |
canvas | -0.000 |
product | -0.000 |
tide | -0.000 |
lee | -0.000 |
Token | She |
Token ID | 1375 |
Feature activation | -7.37e-01 |
Pos logit contributions | |
lee | +0.007 |
line | +0.007 |
canvas | +0.007 |
product | +0.007 |
harbor | +0.007 |
Neg logit contributions | |
girls | -0.007 |
losers | -0.007 |
oof | -0.006 |
ever | -0.006 |
Deal | -0.006 |
Token | by |
Token ID | 416 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
uddy | +0.004 |
girls | +0.004 |
Johnson | +0.003 |
Zip | +0.003 |
isk | +0.003 |
Neg logit contributions | |
canvas | -0.004 |
product | -0.004 |
shell | -0.004 |
line | -0.004 |
lee | -0.004 |
Token | singing |
Token ID | 13777 |
Feature activation | +4.13e-01 |
Pos logit contributions | |
canvas | +0.006 |
line | +0.006 |
product | +0.006 |
tide | +0.006 |
lee | +0.005 |
Neg logit contributions | |
girls | -0.006 |
uddy | -0.006 |
Spot | -0.006 |
Buddy | -0.006 |
Zip | -0.006 |
Token | with |
Token ID | 351 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
uddy | +0.014 |
Johnson | +0.014 |
girls | +0.013 |
Wh | +0.013 |
Kenny | +0.013 |
Neg logit contributions | |
canvas | -0.015 |
line | -0.014 |
tide | -0.013 |
product | -0.013 |
harbor | -0.013 |
Token | others |
Token ID | 1854 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
uddy | +0.008 |
girls | +0.008 |
ver | +0.007 |
Deal | +0.007 |
Johnson | +0.007 |
Neg logit contributions | |
canvas | -0.007 |
lee | -0.006 |
tide | -0.006 |
shell | -0.006 |
product | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -7.78e-01 |
Pos logit contributions | |
canvas | +0.041 |
line | +0.040 |
product | +0.038 |
tide | +0.036 |
shell | +0.036 |
Neg logit contributions | |
Wh | -0.035 |
uddy | -0.035 |
girls | -0.035 |
Zip | -0.035 |
oof | -0.034 |
Token | getting |
Token ID | 1972 |
Feature activation | -3.36e+00 |
Pos logit contributions | |
canvas | +0.029 |
product | +0.027 |
line | +0.027 |
tide | +0.026 |
harbor | +0.026 |
Neg logit contributions | |
uddy | -0.024 |
Spot | -0.024 |
girls | -0.023 |
Zip | -0.023 |
oof | -0.023 |
Token | feedback |
Token ID | 7538 |
Feature activation | -5.05e-01 |
Pos logit contributions | |
canvas | +0.141 |
line | +0.137 |
product | +0.137 |
tide | +0.136 |
lee | +0.133 |
Neg logit contributions | |
Spot | -0.085 |
uddy | -0.085 |
girls | -0.084 |
Buddy | -0.083 |
Zip | -0.083 |
Token | . |
Token ID | 13 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
canvas | +0.017 |
line | +0.017 |
product | +0.017 |
tide | +0.016 |
harbor | +0.015 |
Neg logit contributions | |
Spot | -0.017 |
Buddy | -0.017 |
girls | -0.016 |
Wh | -0.016 |
Zip | -0.016 |
Token | You |
Token ID | 921 |
Feature activation | -5.86e-01 |
Pos logit contributions | |
canvas | +0.009 |
line | +0.009 |
rys | +0.008 |
lee | +0.008 |
tide | +0.008 |
Neg logit contributions | |
girls | -0.006 |
uddy | -0.006 |
isk | -0.006 |
Buddy | -0.006 |
Johnson | -0.006 |
Token | can |
Token ID | 460 |
Feature activation | +3.13e-01 |
Pos logit contributions | |
canvas | +0.024 |
line | +0.022 |
harbor | +0.022 |
tide | +0.021 |
product | +0.021 |
Neg logit contributions | |
Johnson | -0.016 |
Spot | -0.016 |
uddy | -0.016 |
girls | -0.015 |
oof | -0.015 |
Token | improve |
Token ID | 2987 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
Spot | +0.009 |
girls | +0.009 |
Buddy | +0.009 |
Zip | +0.009 |
uddy | +0.009 |
Neg logit contributions | |
canvas | -0.012 |
line | -0.012 |
product | -0.011 |
tide | -0.011 |
harbor | -0.011 |
Token | you |
Token ID | 345 |
Feature activation | -7.13e-01 |
Pos logit contributions | |
canvas | +0.013 |
product | +0.012 |
line | +0.012 |
tide | +0.012 |
shell | +0.011 |
Neg logit contributions | |
uddy | -0.015 |
girls | -0.015 |
Zip | -0.014 |
oof | -0.014 |
Johnson | -0.014 |
Token | from |
Token ID | 422 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
line | +0.023 |
canvas | +0.023 |
product | +0.022 |
tide | +0.021 |
shell | +0.021 |
Neg logit contributions | |
uddy | -0.025 |
Zip | -0.024 |
Spot | -0.024 |
girls | -0.024 |
oof | -0.024 |
Token | losing |
Token ID | 6078 |
Feature activation | -3.45e-01 |
Pos logit contributions | |
canvas | +0.038 |
product | +0.037 |
line | +0.037 |
tide | +0.037 |
harbor | +0.036 |
Neg logit contributions | |
uddy | -0.040 |
girls | -0.037 |
Zip | -0.036 |
oof | -0.035 |
Buddy | -0.035 |
Token | out |
Token ID | 503 |
Feature activation | -6.17e-03 |
Pos logit contributions | |
canvas | +0.011 |
product | +0.011 |
line | +0.011 |
tide | +0.011 |
shell | +0.010 |
Neg logit contributions | |
uddy | -0.012 |
girls | -0.012 |
Zip | -0.011 |
Spot | -0.011 |
oof | -0.011 |
Token | or |
Token ID | 393 |
Feature activation | -4.82e-01 |
Pos logit contributions | |
canvas | +0.000 |
product | +0.000 |
line | +0.000 |
tide | +0.000 |
lee | +0.000 |
Neg logit contributions | |
girls | -0.000 |
uddy | -0.000 |
oof | -0.000 |
Zip | -0.000 |
Spot | -0.000 |
Token | getting |
Token ID | 1972 |
Feature activation | -3.36e+00 |
Pos logit contributions | |
canvas | +0.017 |
line | +0.017 |
product | +0.016 |
tide | +0.016 |
harbor | +0.016 |
Neg logit contributions | |
uddy | -0.015 |
girls | -0.015 |
Zip | -0.015 |
oof | -0.015 |
Spot | -0.015 |
Token | into |
Token ID | 656 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
canvas | +0.128 |
line | +0.125 |
tide | +0.125 |
product | +0.125 |
lee | +0.120 |
Neg logit contributions | |
uddy | -0.098 |
girls | -0.096 |
Zip | -0.096 |
Spot | -0.093 |
oof | -0.093 |
Token | trouble |
Token ID | 5876 |
Feature activation | -9.73e-01 |
Pos logit contributions | |
canvas | +0.038 |
product | +0.038 |
tide | +0.036 |
line | +0.036 |
harbor | +0.035 |
Neg logit contributions | |
uddy | -0.035 |
girls | -0.035 |
Zip | -0.034 |
Spot | -0.033 |
Buddy | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -2.84e-01 |
Pos logit contributions | |
line | +0.029 |
product | +0.028 |
canvas | +0.028 |
tide | +0.026 |
harbor | +0.026 |
Neg logit contributions | |
oof | -0.036 |
Zip | -0.036 |
uddy | -0.036 |
Spot | -0.036 |
girls | -0.035 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +8.40e-02 |
Pos logit contributions | |
line | +0.010 |
product | +0.010 |
canvas | +0.010 |
lee | +0.009 |
rys | +0.009 |
Neg logit contributions | |
girls | -0.008 |
Spot | -0.007 |
Buddy | -0.007 |
uddy | -0.007 |
Zip | -0.007 |
Token | Once |
Token ID | 7454 |
Feature activation | -7.94e-01 |
Pos logit contributions | |
oof | +0.003 |
girls | +0.003 |
losers | +0.003 |
Deal | +0.003 |
uddy | +0.003 |
Neg logit contributions | |
line | -0.002 |
tide | -0.002 |
product | -0.002 |
harbor | -0.002 |
cliff | -0.002 |
Token | flexible |
Token ID | 12846 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
canvas | +0.054 |
product | +0.053 |
line | +0.051 |
lee | +0.050 |
tide | +0.050 |
Neg logit contributions | |
Spot | -0.047 |
girls | -0.047 |
uddy | -0.046 |
Zip | -0.046 |
Buddy | -0.046 |
Token | to |
Token ID | 284 |
Feature activation | +7.81e-02 |
Pos logit contributions | |
canvas | +0.007 |
line | +0.007 |
product | +0.007 |
tide | +0.006 |
shell | +0.006 |
Neg logit contributions | |
Zip | -0.008 |
oof | -0.008 |
uddy | -0.008 |
girls | -0.008 |
Spot | -0.008 |
Token | hold |
Token ID | 1745 |
Feature activation | -8.76e-01 |
Pos logit contributions | |
girls | +0.003 |
uddy | +0.003 |
Spot | +0.003 |
oof | +0.002 |
Johnson | +0.002 |
Neg logit contributions | |
canvas | -0.003 |
line | -0.002 |
product | -0.002 |
tide | -0.002 |
shell | -0.002 |
Token | onto |
Token ID | 4291 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
canvas | +0.029 |
line | +0.028 |
shell | +0.028 |
product | +0.027 |
tide | +0.027 |
Neg logit contributions | |
girls | -0.029 |
isk | -0.029 |
uddy | -0.028 |
Monkey | -0.028 |
Deal | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | -7.83e-01 |
Pos logit contributions | |
canvas | +0.004 |
product | +0.004 |
tide | +0.004 |
line | +0.004 |
pumpkin | +0.004 |
Neg logit contributions | |
uddy | -0.005 |
girls | -0.005 |
oof | -0.004 |
Deal | -0.004 |
ver | -0.004 |
Token | palm |
Token ID | 18057 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
canvas | +0.023 |
tide | +0.021 |
pumpkin | +0.020 |
shell | +0.019 |
pulp | +0.019 |
Neg logit contributions | |
girls | -0.034 |
oof | -0.032 |
uddy | -0.032 |
Spot | -0.032 |
Buddy | -0.032 |
Token | branches |
Token ID | 13737 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
canvas | +0.125 |
product | +0.122 |
line | +0.122 |
tide | +0.119 |
lee | +0.116 |
Neg logit contributions | |
Spot | -0.099 |
girls | -0.095 |
Buddy | -0.095 |
Zip | -0.094 |
Fl | -0.091 |
Token | ." |
Token ID | 526 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
line | +0.040 |
canvas | +0.038 |
product | +0.035 |
tide | +0.035 |
harbor | +0.034 |
Neg logit contributions | |
girls | -0.039 |
oof | -0.038 |
Spot | -0.038 |
Zip | -0.038 |
Buddy | -0.037 |
Token |
|
Token ID | 198 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
line | +0.008 |
lee | +0.008 |
canvas | +0.008 |
product | +0.007 |
tide | +0.007 |
Neg logit contributions | |
girls | -0.006 |
oof | -0.006 |
uddy | -0.006 |
ver | -0.006 |
isk | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -5.36e-01 |
Pos logit contributions | |
line | +0.018 |
lee | +0.018 |
product | +0.018 |
canvas | +0.018 |
tide | +0.017 |
Neg logit contributions | |
girls | -0.017 |
Monkey | -0.015 |
losers | -0.015 |
Johnson | -0.015 |
Spot | -0.015 |
Token | Tim |
Token ID | 14967 |
Feature activation | +3.24e-01 |
Pos logit contributions | |
canvas | +0.016 |
product | +0.016 |
line | +0.016 |
tide | +0.016 |
harbor | +0.015 |
Neg logit contributions | |
girls | -0.018 |
Spot | -0.018 |
Johnson | -0.017 |
Buddy | -0.017 |
Fl | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | -4.08e-01 |
Pos logit contributions | |
canvas | +0.029 |
product | +0.029 |
tide | +0.029 |
pumpkin | +0.028 |
shell | +0.028 |
Neg logit contributions | |
uddy | -0.039 |
girls | -0.037 |
ver | -0.036 |
Deal | -0.036 |
ll | -0.036 |
Token | hook |
Token ID | 8011 |
Feature activation | -8.75e-01 |
Pos logit contributions | |
product | +0.011 |
canvas | +0.011 |
line | +0.011 |
lee | +0.011 |
rys | +0.010 |
Neg logit contributions | |
Spot | -0.016 |
girls | -0.016 |
Buddy | -0.015 |
Fl | -0.015 |
Zip | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
line | +0.029 |
canvas | +0.029 |
product | +0.027 |
tide | +0.027 |
shell | +0.026 |
Neg logit contributions | |
Spot | -0.030 |
girls | -0.030 |
Buddy | -0.029 |
Zip | -0.029 |
Wh | -0.029 |
Token | throw |
Token ID | 3714 |
Feature activation | -9.32e-01 |
Pos logit contributions | |
canvas | +0.007 |
line | +0.007 |
tide | +0.007 |
product | +0.007 |
shell | +0.007 |
Neg logit contributions | |
girls | -0.008 |
Spot | -0.008 |
uddy | -0.008 |
Buddy | -0.008 |
Zip | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
canvas | +0.031 |
tide | +0.031 |
shell | +0.030 |
line | +0.030 |
product | +0.030 |
Neg logit contributions | |
uddy | -0.031 |
girls | -0.030 |
Buddy | -0.029 |
Spot | -0.029 |
Johnson | -0.029 |
Token | line |
Token ID | 1627 |
Feature activation | -3.32e+00 |
Pos logit contributions | |
canvas | +0.030 |
line | +0.029 |
product | +0.029 |
tide | +0.027 |
lee | +0.027 |
Neg logit contributions | |
Spot | -0.044 |
girls | -0.043 |
Buddy | -0.042 |
Zip | -0.042 |
Fl | -0.041 |
Token | in |
Token ID | 287 |
Feature activation | -9.85e-01 |
Pos logit contributions | |
canvas | +0.132 |
line | +0.129 |
product | +0.129 |
tide | +0.126 |
lee | +0.123 |
Neg logit contributions | |
Spot | -0.091 |
girls | -0.088 |
Buddy | -0.086 |
Zip | -0.086 |
Max | -0.083 |
Token | the |
Token ID | 262 |
Feature activation | -7.58e-01 |
Pos logit contributions | |
canvas | +0.030 |
product | +0.030 |
tide | +0.029 |
shell | +0.029 |
line | +0.028 |
Neg logit contributions | |
uddy | -0.038 |
girls | -0.037 |
Deal | -0.036 |
oof | -0.036 |
ver | -0.035 |
Token | water |
Token ID | 1660 |
Feature activation | -5.41e-01 |
Pos logit contributions | |
line | +0.026 |
product | +0.025 |
canvas | +0.025 |
lee | +0.024 |
complexion | +0.023 |
Neg logit contributions | |
Spot | -0.026 |
Fl | -0.024 |
girls | -0.024 |
Max | -0.024 |
Buddy | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
canvas | +0.021 |
product | +0.020 |
line | +0.020 |
tide | +0.020 |
lee | +0.019 |
Neg logit contributions | |
Spot | -0.016 |
girls | -0.015 |
Buddy | -0.015 |
Fl | -0.014 |
Zip | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
Spot | +0.004 |
Max | +0.003 |
Fl | +0.003 |
Buddy | +0.003 |
Tim | +0.003 |
Neg logit contributions | |
canvas | -0.008 |
product | -0.008 |
line | -0.008 |
harbor | -0.007 |
rys | -0.007 |
Token | rough |
Token ID | 5210 |
Feature activation | -6.90e-01 |
Pos logit contributions | |
canvas | +0.029 |
product | +0.029 |
line | +0.027 |
harbor | +0.027 |
rys | +0.027 |
Neg logit contributions | |
Spot | -0.016 |
Tim | -0.015 |
ever | -0.014 |
Max | -0.014 |
Buddy | -0.014 |
Token | it |
Token ID | 340 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
canvas | +0.025 |
line | +0.025 |
tide | +0.024 |
product | +0.024 |
lee | +0.023 |
Neg logit contributions | |
girls | -0.021 |
Spot | -0.020 |
uddy | -0.020 |
Buddy | -0.020 |
oof | -0.020 |
Token | was |
Token ID | 373 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
Spot | +0.020 |
girls | +0.019 |
Buddy | +0.019 |
Zip | +0.019 |
Fl | +0.019 |
Neg logit contributions | |
canvas | -0.027 |
line | -0.026 |
product | -0.026 |
tide | -0.025 |
lee | -0.025 |
Token | with |
Token ID | 351 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
girls | +0.015 |
Spot | +0.015 |
Zip | +0.014 |
Buddy | +0.014 |
oof | +0.014 |
Neg logit contributions | |
canvas | -0.016 |
line | -0.016 |
product | -0.015 |
tide | -0.015 |
shell | -0.014 |
Token | his |
Token ID | 465 |
Feature activation | -7.27e-01 |
Pos logit contributions | |
girls | +0.005 |
uddy | +0.005 |
Deal | +0.004 |
Monkey | +0.004 |
Johnson | +0.004 |
Neg logit contributions | |
canvas | -0.005 |
tide | -0.004 |
lee | -0.004 |
shell | -0.004 |
product | -0.004 |
Token | palm |
Token ID | 18057 |
Feature activation | -3.32e+00 |
Pos logit contributions | |
canvas | +0.022 |
product | +0.021 |
line | +0.021 |
tide | +0.021 |
shell | +0.020 |
Neg logit contributions | |
girls | -0.027 |
Spot | -0.027 |
Zip | -0.026 |
Buddy | -0.026 |
Monkey | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -5.32e-02 |
Pos logit contributions | |
line | +0.131 |
product | +0.129 |
canvas | +0.129 |
harbor | +0.127 |
tide | +0.126 |
Neg logit contributions | |
Spot | -0.091 |
Buddy | -0.086 |
Zip | -0.085 |
Fl | -0.084 |
Max | -0.083 |
Token | |
Token ID | 220 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
line | +0.002 |
canvas | +0.002 |
product | +0.002 |
tide | +0.002 |
lee | +0.002 |
Neg logit contributions | |
girls | -0.001 |
Spot | -0.001 |
Zip | -0.001 |
Buddy | -0.001 |
oof | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -4.18e-01 |
Pos logit contributions | |
girls | +0.004 |
Spot | +0.004 |
Monkey | +0.004 |
Buddy | +0.004 |
Zip | +0.004 |
Neg logit contributions | |
line | -0.007 |
canvas | -0.006 |
product | -0.006 |
lee | -0.006 |
tide | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -5.44e-01 |
Pos logit contributions | |
line | +0.016 |
product | +0.015 |
canvas | +0.015 |
lee | +0.014 |
tide | +0.014 |
Neg logit contributions | |
girls | -0.012 |
Spot | -0.011 |
oof | -0.011 |
Buddy | -0.011 |
Monkey | -0.011 |
Token | Tim |
Token ID | 14967 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
canvas | +0.027 |
product | +0.026 |
line | +0.026 |
tide | +0.026 |
lee | +0.026 |
Neg logit contributions | |
Spot | -0.010 |
Buddy | -0.010 |
Zip | -0.009 |
Wh | -0.009 |
girls | -0.009 |
Token | water |
Token ID | 1660 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
canvas | +0.002 |
product | +0.002 |
line | +0.002 |
lee | +0.002 |
pulp | +0.002 |
Neg logit contributions | |
Spot | -0.002 |
Fl | -0.002 |
Buddy | -0.002 |
Max | -0.002 |
girls | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | -6.65e-01 |
Pos logit contributions | |
girls | +0.003 |
Spot | +0.003 |
Buddy | +0.003 |
Zip | +0.003 |
uddy | +0.003 |
Neg logit contributions | |
line | -0.004 |
canvas | -0.004 |
tide | -0.004 |
product | -0.004 |
harbor | -0.004 |
Token | feel |
Token ID | 1254 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
canvas | +0.022 |
line | +0.022 |
tide | +0.021 |
product | +0.021 |
harbor | +0.020 |
Neg logit contributions | |
Spot | -0.022 |
girls | -0.022 |
Buddy | -0.021 |
Zip | -0.021 |
uddy | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
canvas | +0.078 |
product | +0.076 |
pulp | +0.071 |
lee | +0.070 |
line | +0.069 |
Neg logit contributions | |
Spot | -0.035 |
Fl | -0.034 |
Buddy | -0.030 |
Max | -0.029 |
love | -0.029 |
Token | cool |
Token ID | 3608 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
canvas | +0.034 |
product | +0.033 |
line | +0.033 |
lee | +0.031 |
tide | +0.030 |
Neg logit contributions | |
Spot | -0.047 |
girls | -0.045 |
Buddy | -0.045 |
Fl | -0.044 |
Zip | -0.044 |
Token | ness |
Token ID | 1108 |
Feature activation | -3.32e+00 |
Pos logit contributions | |
canvas | +0.058 |
product | +0.055 |
line | +0.053 |
lee | +0.052 |
pulp | +0.050 |
Neg logit contributions | |
Spot | -0.074 |
Fl | -0.070 |
Buddy | -0.069 |
Max | -0.068 |
Zip | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
canvas | +0.151 |
product | +0.149 |
lee | +0.148 |
harbor | +0.145 |
pulp | +0.143 |
Neg logit contributions | |
Spot | -0.070 |
ever | -0.066 |
Fl | -0.065 |
Max | -0.061 |
Buddy | -0.060 |
Token |
|
Token ID | 198 |
Feature activation | -5.81e-01 |
Pos logit contributions | |
canvas | +0.016 |
line | +0.015 |
lee | +0.015 |
tide | +0.015 |
product | +0.014 |
Neg logit contributions | |
girls | -0.012 |
Spot | -0.010 |
losers | -0.010 |
Buddy | -0.010 |
Monkey | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -7.46e-01 |
Pos logit contributions | |
line | +0.023 |
canvas | +0.023 |
product | +0.023 |
tide | +0.022 |
lee | +0.022 |
Neg logit contributions | |
girls | -0.015 |
Spot | -0.014 |
Buddy | -0.014 |
Wh | -0.014 |
Monkey | -0.014 |
Token | One |
Token ID | 3198 |
Feature activation | +4.51e-01 |
Pos logit contributions | |
canvas | +0.034 |
line | +0.034 |
product | +0.034 |
tide | +0.033 |
lee | +0.032 |
Neg logit contributions | |
Spot | -0.016 |
girls | -0.015 |
Buddy | -0.015 |
Zip | -0.015 |
Max | -0.014 |
Token | day |
Token ID | 1110 |
Feature activation | -5.02e-01 |
Pos logit contributions | |
Spot | +0.010 |
Buddy | +0.010 |
Zip | +0.009 |
Fl | +0.009 |
Max | +0.009 |
Neg logit contributions | |
canvas | -0.021 |
line | -0.020 |
product | -0.020 |
lee | -0.020 |
pulp | -0.019 |
Token | oy |
Token ID | 35104 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
canvas | +0.035 |
product | +0.034 |
line | +0.033 |
tide | +0.033 |
harbor | +0.031 |
Neg logit contributions | |
uddy | -0.032 |
girls | -0.031 |
Zip | -0.030 |
Buddy | -0.029 |
oof | -0.029 |
Token | sters |
Token ID | 5937 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
canvas | +0.020 |
product | +0.017 |
tide | +0.016 |
lee | +0.016 |
rys | +0.016 |
Neg logit contributions | |
girls | -0.050 |
Spot | -0.047 |
Monkey | -0.046 |
Wh | -0.045 |
just | -0.045 |
Token | and |
Token ID | 290 |
Feature activation | -8.55e-01 |
Pos logit contributions | |
Spot | +0.001 |
Zip | +0.001 |
girls | +0.001 |
Buddy | +0.001 |
uddy | +0.000 |
Neg logit contributions | |
line | -0.001 |
canvas | -0.001 |
product | -0.001 |
tide | -0.001 |
harbor | -0.001 |
Token | ended |
Token ID | 4444 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
canvas | +0.030 |
product | +0.029 |
line | +0.029 |
tide | +0.028 |
shell | +0.027 |
Neg logit contributions | |
girls | -0.028 |
uddy | -0.028 |
Zip | -0.027 |
oof | -0.027 |
Spot | -0.027 |
Token | up |
Token ID | 510 |
Feature activation | -4.53e-01 |
Pos logit contributions | |
canvas | +0.044 |
product | +0.043 |
line | +0.043 |
tide | +0.042 |
lee | +0.041 |
Neg logit contributions | |
Spot | -0.029 |
girls | -0.028 |
Buddy | -0.028 |
Zip | -0.028 |
uddy | -0.027 |
Token | getting |
Token ID | 1972 |
Feature activation | -3.31e+00 |
Pos logit contributions | |
canvas | +0.017 |
line | +0.017 |
product | +0.017 |
tide | +0.017 |
lee | +0.016 |
Neg logit contributions | |
Spot | -0.013 |
girls | -0.013 |
Buddy | -0.013 |
Zip | -0.012 |
Max | -0.012 |
Token | sick |
Token ID | 6639 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
canvas | +0.127 |
line | +0.124 |
product | +0.124 |
tide | +0.121 |
lee | +0.118 |
Neg logit contributions | |
Spot | -0.095 |
girls | -0.092 |
Buddy | -0.090 |
Zip | -0.089 |
Fl | -0.087 |
Token | from |
Token ID | 422 |
Feature activation | -8.37e-01 |
Pos logit contributions | |
canvas | +0.042 |
line | +0.042 |
product | +0.041 |
tide | +0.040 |
lee | +0.038 |
Neg logit contributions | |
Spot | -0.040 |
girls | -0.040 |
Zip | -0.040 |
Buddy | -0.040 |
oof | -0.039 |
Token | eating |
Token ID | 6600 |
Feature activation | -8.32e-01 |
Pos logit contributions | |
canvas | +0.030 |
product | +0.030 |
tide | +0.029 |
line | +0.029 |
harbor | +0.028 |
Neg logit contributions | |
uddy | -0.027 |
girls | -0.026 |
Zip | -0.025 |
oof | -0.025 |
Spot | -0.025 |
Token | them |
Token ID | 606 |
Feature activation | -5.68e-01 |
Pos logit contributions | |
canvas | +0.030 |
line | +0.029 |
product | +0.029 |
tide | +0.029 |
harbor | +0.027 |
Neg logit contributions | |
Spot | -0.026 |
girls | -0.026 |
Zip | -0.025 |
Buddy | -0.025 |
uddy | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
product | +0.017 |
canvas | +0.017 |
line | +0.016 |
tide | +0.015 |
pumpkin | +0.015 |
Neg logit contributions | |
Zip | -0.022 |
girls | -0.021 |
uddy | -0.021 |
oof | -0.021 |
Spot | -0.021 |
Token | to |
Token ID | 284 |
Feature activation | -8.09e-01 |
Pos logit contributions | |
canvas | +0.009 |
product | +0.009 |
line | +0.009 |
lee | +0.009 |
tide | +0.009 |
Neg logit contributions | |
oof | -0.010 |
uddy | -0.010 |
girls | -0.009 |
Buddy | -0.009 |
Wh | -0.009 |
Token | 10 |
Token ID | 838 |
Feature activation | -9.34e-01 |
Pos logit contributions | |
canvas | +0.028 |
line | +0.026 |
product | +0.026 |
tide | +0.025 |
shell | +0.023 |
Neg logit contributions | |
uddy | -0.029 |
girls | -0.028 |
oof | -0.027 |
Spot | -0.027 |
Buddy | -0.027 |
Token | out |
Token ID | 503 |
Feature activation | +6.34e-02 |
Pos logit contributions | |
canvas | +0.032 |
product | +0.032 |
line | +0.031 |
tide | +0.029 |
lee | +0.029 |
Neg logit contributions | |
Spot | -0.030 |
Buddy | -0.030 |
girls | -0.030 |
oof | -0.030 |
uddy | -0.030 |
Token | loud |
Token ID | 7812 |
Feature activation | -8.22e-01 |
Pos logit contributions | |
Spot | +0.002 |
girls | +0.002 |
oof | +0.002 |
Buddy | +0.002 |
uddy | +0.002 |
Neg logit contributions | |
canvas | -0.002 |
product | -0.002 |
lee | -0.002 |
line | -0.002 |
tide | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | -5.05e-01 |
Pos logit contributions | |
canvas | +0.029 |
product | +0.028 |
line | +0.028 |
tide | +0.027 |
lee | +0.027 |
Neg logit contributions | |
Spot | -0.026 |
girls | -0.026 |
Buddy | -0.025 |
oof | -0.025 |
uddy | -0.025 |
Token | getting |
Token ID | 1972 |
Feature activation | -3.29e+00 |
Pos logit contributions | |
canvas | +0.018 |
line | +0.018 |
product | +0.017 |
tide | +0.017 |
lee | +0.016 |
Neg logit contributions | |
oof | -0.017 |
uddy | -0.016 |
girls | -0.016 |
Buddy | -0.015 |
Johnson | -0.015 |
Token | ready |
Token ID | 3492 |
Feature activation | -6.54e-01 |
Pos logit contributions | |
canvas | +0.116 |
tide | +0.114 |
line | +0.114 |
product | +0.114 |
lee | +0.108 |
Neg logit contributions | |
uddy | -0.106 |
oof | -0.105 |
girls | -0.100 |
Buddy | -0.100 |
Spot | -0.099 |
Token | to |
Token ID | 284 |
Feature activation | -3.72e-01 |
Pos logit contributions | |
canvas | +0.028 |
line | +0.028 |
product | +0.028 |
tide | +0.027 |
lee | +0.026 |
Neg logit contributions | |
girls | -0.016 |
oof | -0.016 |
Spot | -0.015 |
uddy | -0.015 |
Buddy | -0.015 |
Token | freeze |
Token ID | 16611 |
Feature activation | -5.48e-01 |
Pos logit contributions | |
canvas | +0.013 |
product | +0.013 |
tide | +0.013 |
line | +0.012 |
shell | +0.012 |
Neg logit contributions | |
girls | -0.012 |
uddy | -0.012 |
Spot | -0.011 |
oof | -0.011 |
Buddy | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
product | +0.018 |
canvas | +0.018 |
line | +0.018 |
tide | +0.017 |
lee | +0.017 |
Neg logit contributions | |
oof | -0.018 |
uddy | -0.018 |
girls | -0.018 |
Spot | -0.017 |
Buddy | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | -4.10e-01 |
Pos logit contributions | |
line | +0.011 |
canvas | +0.011 |
product | +0.010 |
lee | +0.010 |
harbor | +0.010 |
Neg logit contributions | |
girls | -0.010 |
oof | -0.009 |
Deal | -0.009 |
losers | -0.009 |
isk | -0.008 |
Token | sense |
Token ID | 2565 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
Spot | +0.022 |
Fl | +0.021 |
Buddy | +0.020 |
Max | +0.020 |
Zip | +0.020 |
Neg logit contributions | |
line | -0.031 |
canvas | -0.030 |
product | -0.030 |
lee | -0.030 |
tide | -0.029 |
Token | of |
Token ID | 286 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
canvas | +0.044 |
line | +0.043 |
product | +0.043 |
tide | +0.042 |
lee | +0.041 |
Neg logit contributions | |
Spot | -0.024 |
girls | -0.023 |
Buddy | -0.023 |
Zip | -0.023 |
Max | -0.022 |
Token | joy |
Token ID | 8716 |
Feature activation | -9.46e-01 |
Pos logit contributions | |
girls | +0.033 |
Spot | +0.032 |
Buddy | +0.031 |
uddy | +0.031 |
Zip | +0.031 |
Neg logit contributions | |
canvas | -0.047 |
product | -0.046 |
tide | -0.046 |
line | -0.046 |
lee | -0.044 |
Token | and |
Token ID | 290 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
canvas | +0.035 |
line | +0.035 |
product | +0.034 |
tide | +0.034 |
lee | +0.033 |
Neg logit contributions | |
Spot | -0.028 |
girls | -0.027 |
Buddy | -0.027 |
Zip | -0.027 |
Max | -0.026 |
Token | content |
Token ID | 2695 |
Feature activation | -8.98e-01 |
Pos logit contributions | |
canvas | +0.005 |
line | +0.005 |
product | +0.005 |
tide | +0.005 |
lee | +0.005 |
Neg logit contributions | |
girls | -0.005 |
Spot | -0.005 |
uddy | -0.005 |
Buddy | -0.005 |
Zip | -0.005 |
Token | ment |
Token ID | 434 |
Feature activation | -3.29e+00 |
Pos logit contributions | |
canvas | +0.031 |
product | +0.029 |
line | +0.029 |
tide | +0.029 |
pulp | +0.028 |
Neg logit contributions | |
Spot | -0.030 |
Fl | -0.028 |
Zip | -0.027 |
Buddy | -0.027 |
Max | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
line | +0.090 |
product | +0.082 |
canvas | +0.079 |
tide | +0.074 |
lee | +0.074 |
Neg logit contributions | |
oof | -0.137 |
uddy | -0.134 |
girls | -0.134 |
Buddy | -0.133 |
Zip | -0.132 |
Token | After |
Token ID | 2293 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
line | +0.008 |
lee | +0.008 |
canvas | +0.008 |
product | +0.008 |
tide | +0.008 |
Neg logit contributions | |
girls | -0.006 |
oof | -0.005 |
Spot | -0.005 |
Buddy | -0.005 |
Zip | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
tide | +0.005 |
line | +0.005 |
canvas | +0.005 |
product | +0.005 |
harbor | +0.005 |
Neg logit contributions | |
girls | -0.004 |
uddy | -0.004 |
oof | -0.004 |
Spot | -0.004 |
Deal | -0.004 |
Token | few |
Token ID | 1178 |
Feature activation | +3.35e-01 |
Pos logit contributions | |
Spot | +0.032 |
girls | +0.030 |
Buddy | +0.030 |
Zip | +0.030 |
Fl | +0.029 |
Neg logit contributions | |
canvas | -0.055 |
line | -0.054 |
product | -0.054 |
tide | -0.053 |
lee | -0.052 |
Token | moments |
Token ID | 7188 |
Feature activation | -6.92e-02 |
Pos logit contributions | |
Spot | +0.011 |
girls | +0.011 |
Buddy | +0.011 |
Zip | +0.011 |
Wh | +0.011 |
Neg logit contributions | |
canvas | -0.011 |
product | -0.011 |
line | -0.011 |
tide | -0.011 |
lee | -0.010 |
Token | clean |
Token ID | 3424 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
canvas | +0.007 |
line | +0.006 |
tide | +0.006 |
product | +0.006 |
harbor | +0.006 |
Neg logit contributions | |
girls | -0.007 |
Spot | -0.007 |
Zip | -0.007 |
Buddy | -0.006 |
uddy | -0.006 |
Token | up |
Token ID | 510 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
canvas | +0.014 |
line | +0.014 |
product | +0.014 |
tide | +0.013 |
pulp | +0.013 |
Neg logit contributions | |
Spot | -0.007 |
Max | -0.006 |
Buddy | -0.006 |
Fl | -0.006 |
Wh | -0.006 |
Token | his |
Token ID | 465 |
Feature activation | -3.81e-01 |
Pos logit contributions | |
girls | +0.007 |
Spot | +0.007 |
Zip | +0.007 |
Buddy | +0.006 |
uddy | +0.006 |
Neg logit contributions | |
canvas | -0.010 |
product | -0.010 |
line | -0.010 |
tide | -0.009 |
lee | -0.009 |
Token | toys |
Token ID | 14958 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
line | +0.016 |
canvas | +0.015 |
pulp | +0.014 |
rys | +0.014 |
lee | +0.014 |
Neg logit contributions | |
Spot | -0.010 |
Max | -0.009 |
first | -0.009 |
for | -0.009 |
Fl | -0.009 |
Token | before |
Token ID | 878 |
Feature activation | -7.78e-01 |
Pos logit contributions | |
girls | +0.014 |
Zip | +0.014 |
oof | +0.014 |
uddy | +0.014 |
Monkey | +0.013 |
Neg logit contributions | |
line | -0.020 |
canvas | -0.019 |
product | -0.018 |
tide | -0.018 |
harbor | -0.018 |
Token | getting |
Token ID | 1972 |
Feature activation | -3.29e+00 |
Pos logit contributions | |
canvas | +0.031 |
lee | +0.031 |
line | +0.031 |
product | +0.030 |
rys | +0.030 |
Neg logit contributions | |
Spot | -0.022 |
Max | -0.021 |
Fl | -0.020 |
Buddy | -0.020 |
Wh | -0.019 |
Token | out |
Token ID | 503 |
Feature activation | +5.66e-02 |
Pos logit contributions | |
canvas | +0.146 |
line | +0.141 |
product | +0.141 |
lee | +0.136 |
rys | +0.135 |
Neg logit contributions | |
Spot | -0.076 |
Max | -0.071 |
Fl | -0.068 |
Wh | -0.068 |
Will | -0.066 |
Token | of |
Token ID | 286 |
Feature activation | +4.24e-01 |
Pos logit contributions | |
Spot | +0.001 |
Buddy | +0.001 |
Max | +0.001 |
Fl | +0.001 |
Wh | +0.001 |
Neg logit contributions | |
canvas | -0.003 |
line | -0.003 |
product | -0.003 |
tide | -0.003 |
pulp | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | -8.87e-01 |
Pos logit contributions | |
Spot | +0.012 |
girls | +0.011 |
Buddy | +0.011 |
Zip | +0.011 |
Max | +0.011 |
Neg logit contributions | |
canvas | -0.017 |
line | -0.017 |
product | -0.017 |
tide | -0.016 |
lee | -0.016 |
Token | tub |
Token ID | 12202 |
Feature activation | -9.13e-01 |
Pos logit contributions | |
canvas | +0.031 |
line | +0.031 |
lee | +0.030 |
pulp | +0.029 |
product | +0.029 |
Neg logit contributions | |
Spot | -0.028 |
Fl | -0.026 |
Max | -0.026 |
Wh | -0.025 |
first | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
canvas | +0.034 |
line | +0.033 |
product | +0.033 |
tide | +0.033 |
lee | +0.032 |
Neg logit contributions | |
Spot | -0.027 |
girls | -0.026 |
Buddy | -0.026 |
Zip | -0.026 |
Max | -0.025 |
Token | serve |
Token ID | 4691 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
canvas | +0.024 |
product | +0.023 |
line | +0.023 |
tide | +0.023 |
harbor | +0.022 |
Neg logit contributions | |
Spot | -0.020 |
girls | -0.020 |
Zip | -0.019 |
Buddy | -0.019 |
uddy | -0.019 |
Token | a |
Token ID | 257 |
Feature activation | +8.10e-02 |
Pos logit contributions | |
uddy | +0.010 |
girls | +0.009 |
Zip | +0.009 |
Monkey | +0.009 |
ver | +0.009 |
Neg logit contributions | |
canvas | -0.009 |
product | -0.009 |
tide | -0.009 |
pumpkin | -0.009 |
shell | -0.009 |
Token | picnic |
Token ID | 35715 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
girls | +0.003 |
Zip | +0.003 |
Spot | +0.003 |
oof | +0.003 |
isk | +0.003 |
Neg logit contributions | |
canvas | -0.003 |
pumpkin | -0.002 |
tide | -0.002 |
product | -0.002 |
harbor | -0.002 |
Token | to |
Token ID | 284 |
Feature activation | -5.26e-01 |
Pos logit contributions | |
Spot | +0.011 |
girls | +0.011 |
Buddy | +0.011 |
Zip | +0.010 |
Max | +0.010 |
Neg logit contributions | |
canvas | -0.015 |
line | -0.015 |
product | -0.015 |
tide | -0.015 |
lee | -0.014 |
Token | their |
Token ID | 511 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
canvas | +0.019 |
product | +0.018 |
tide | +0.018 |
line | +0.018 |
harbor | +0.018 |
Neg logit contributions | |
uddy | -0.017 |
girls | -0.016 |
Zip | -0.015 |
oof | -0.015 |
Spot | -0.015 |
Token | stuffed |
Token ID | 22259 |
Feature activation | +3.26e+00 |
Pos logit contributions | |
harbor | +0.007 |
tide | +0.006 |
pumpkin | +0.006 |
canvas | +0.006 |
product | +0.006 |
Neg logit contributions | |
ver | -0.010 |
oof | -0.009 |
uddy | -0.009 |
Zip | -0.009 |
girls | -0.009 |
Token | animals |
Token ID | 4695 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
ver | +0.118 |
OK | +0.114 |
uddy | +0.114 |
Deal | +0.107 |
oof | +0.107 |
Neg logit contributions | |
pumpkin | -0.105 |
harbor | -0.104 |
shell | -0.100 |
canvas | -0.099 |
tide | -0.098 |
Token | . |
Token ID | 13 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
oof | +0.007 |
Zip | +0.007 |
Wh | +0.007 |
Spot | +0.007 |
uddy | +0.007 |
Neg logit contributions | |
line | -0.006 |
canvas | -0.006 |
harbor | -0.006 |
product | -0.006 |
pumpkin | -0.005 |
Token | They |
Token ID | 1119 |
Feature activation | -5.33e-01 |
Pos logit contributions | |
canvas | +0.018 |
product | +0.018 |
line | +0.018 |
tide | +0.018 |
lee | +0.017 |
Neg logit contributions | |
Spot | -0.009 |
girls | -0.008 |
Buddy | -0.008 |
Zip | -0.008 |
Max | -0.008 |
Token | put |
Token ID | 1234 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
canvas | +0.018 |
line | +0.018 |
product | +0.018 |
tide | +0.018 |
harbor | +0.017 |
Neg logit contributions | |
Spot | -0.017 |
Zip | -0.017 |
girls | -0.017 |
Buddy | -0.017 |
oof | -0.016 |
Token | some |
Token ID | 617 |
Feature activation | +8.56e-01 |
Pos logit contributions | |
pumpkin | +0.005 |
canvas | +0.005 |
tide | +0.005 |
product | +0.005 |
pulp | +0.005 |
Neg logit contributions | |
uddy | -0.005 |
ver | -0.005 |
girls | -0.005 |
ll | -0.005 |
oof | -0.005 |
Token | water |
Token ID | 1660 |
Feature activation | +5.14e-01 |
Pos logit contributions | |
Spot | +0.012 |
girls | +0.012 |
uddy | +0.012 |
Zip | +0.012 |
Buddy | +0.012 |
Neg logit contributions | |
canvas | -0.010 |
tide | -0.010 |
product | -0.010 |
line | -0.009 |
shell | -0.009 |
Token | every |
Token ID | 790 |
Feature activation | -7.57e-01 |
Pos logit contributions | |
Spot | +0.014 |
girls | +0.014 |
Buddy | +0.014 |
Zip | +0.013 |
uddy | +0.013 |
Neg logit contributions | |
line | -0.020 |
canvas | -0.020 |
product | -0.020 |
tide | -0.020 |
lee | -0.019 |
Token | day |
Token ID | 1110 |
Feature activation | -8.07e-01 |
Pos logit contributions | |
canvas | +0.029 |
tide | +0.029 |
product | +0.028 |
shell | +0.028 |
pumpkin | +0.027 |
Neg logit contributions | |
girls | -0.023 |
Spot | -0.022 |
Wh | -0.021 |
Buddy | -0.021 |
Zip | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
canvas | +0.027 |
line | +0.027 |
product | +0.027 |
tide | +0.026 |
lee | +0.025 |
Neg logit contributions | |
Spot | -0.026 |
girls | -0.026 |
Buddy | -0.026 |
Zip | -0.026 |
oof | -0.025 |
Token | He |
Token ID | 679 |
Feature activation | -7.30e-01 |
Pos logit contributions | |
line | +0.009 |
canvas | +0.009 |
tide | +0.008 |
product | +0.008 |
lee | +0.008 |
Neg logit contributions | |
girls | -0.005 |
Buddy | -0.005 |
Monkey | -0.005 |
uddy | -0.005 |
Spot | -0.005 |
Token | named |
Token ID | 3706 |
Feature activation | +2.89e+00 |
Pos logit contributions | |
canvas | +0.027 |
product | +0.026 |
tide | +0.026 |
line | +0.026 |
shell | +0.025 |
Neg logit contributions | |
girls | -0.021 |
Spot | -0.021 |
Zip | -0.021 |
Buddy | -0.021 |
uddy | -0.020 |
Token | it |
Token ID | 340 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
girls | +0.083 |
uddy | +0.079 |
ver | +0.077 |
Deal | +0.076 |
Johnson | +0.075 |
Neg logit contributions | |
canvas | -0.109 |
tide | -0.109 |
lee | -0.108 |
shell | -0.107 |
line | -0.105 |
Token | Buzz |
Token ID | 15753 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
girls | +0.034 |
uddy | +0.032 |
oof | +0.032 |
Johnson | +0.032 |
Spot | +0.032 |
Neg logit contributions | |
shell | -0.045 |
tide | -0.044 |
canvas | -0.044 |
line | -0.044 |
pumpkin | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
Zip | +0.006 |
Kenny | +0.006 |
Buddy | +0.006 |
Spot | +0.006 |
oof | +0.006 |
Neg logit contributions | |
line | -0.007 |
product | -0.007 |
canvas | -0.007 |
shell | -0.007 |
harbor | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +8.50e-02 |
Pos logit contributions | |
atum | +0.013 |
isk | +0.013 |
girls | +0.012 |
oof | +0.012 |
uddy | +0.012 |
Neg logit contributions | |
tide | -0.012 |
line | -0.012 |
pumpkin | -0.011 |
product | -0.011 |
harbor | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
girls | +0.003 |
Buddy | +0.003 |
Wh | +0.003 |
Monkey | +0.002 |
Spot | +0.002 |
Neg logit contributions | |
product | -0.003 |
line | -0.003 |
tide | -0.003 |
harbor | -0.003 |
canvas | -0.003 |
Token | same |
Token ID | 976 |
Feature activation | -3.35e-01 |
Pos logit contributions | |
ver | +0.011 |
girls | +0.011 |
uddy | +0.011 |
Kitty | +0.010 |
Zip | +0.010 |
Neg logit contributions | |
shell | -0.007 |
canvas | -0.007 |
tide | -0.007 |
harbor | -0.007 |
pumpkin | -0.006 |
Token | clearing |
Token ID | 17304 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
canvas | +0.010 |
shell | +0.010 |
product | +0.010 |
harbor | +0.009 |
tide | +0.009 |
Neg logit contributions | |
girls | -0.014 |
Zip | -0.013 |
Spot | -0.013 |
uddy | -0.013 |
Kitty | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -7.09e-03 |
Pos logit contributions | |
girls | +0.034 |
oof | +0.033 |
Spot | +0.032 |
Zip | +0.032 |
uddy | +0.031 |
Neg logit contributions | |
canvas | -0.042 |
line | -0.042 |
product | -0.041 |
tide | -0.039 |
lee | -0.039 |
Token | though |
Token ID | 996 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
canvas | +0.000 |
line | +0.000 |
product | +0.000 |
tide | +0.000 |
shell | +0.000 |
Neg logit contributions | |
girls | -0.000 |
uddy | -0.000 |
oof | -0.000 |
Zip | -0.000 |
Deal | -0.000 |
Token | it |
Token ID | 340 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
canvas | +0.010 |
lee | +0.010 |
product | +0.009 |
tide | +0.009 |
harbor | +0.009 |
Neg logit contributions | |
girls | -0.014 |
Deal | -0.013 |
oof | -0.013 |
Enjoy | -0.013 |
losers | -0.012 |
Token | hadn |
Token ID | 8020 |
Feature activation | +3.00e+00 |
Pos logit contributions | |
oof | +0.051 |
girls | +0.050 |
Enjoy | +0.050 |
Deal | +0.047 |
Spot | +0.045 |
Neg logit contributions | |
product | -0.044 |
lee | -0.044 |
shell | -0.043 |
canvas | -0.041 |
tide | -0.041 |
Token | 't |
Token ID | 470 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
Deal | +0.085 |
girls | +0.075 |
oof | +0.066 |
Enjoy | +0.061 |
Spot | +0.057 |
Neg logit contributions | |
line | -0.123 |
product | -0.122 |
lee | -0.121 |
canvas | -0.118 |
tide | -0.116 |
Token | looked |
Token ID | 3114 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
girls | +0.035 |
Spot | +0.034 |
oof | +0.033 |
uddy | +0.033 |
Zip | +0.033 |
Neg logit contributions | |
canvas | -0.038 |
product | -0.037 |
tide | -0.037 |
line | -0.036 |
lee | -0.035 |
Token | as |
Token ID | 355 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
girls | +0.008 |
uddy | +0.008 |
oof | +0.008 |
Zip | +0.008 |
Buddy | +0.008 |
Neg logit contributions | |
canvas | -0.010 |
line | -0.010 |
product | -0.010 |
tide | -0.010 |
lee | -0.010 |
Token | hurt |
Token ID | 5938 |
Feature activation | +4.80e-01 |
Pos logit contributions | |
tide | +0.004 |
canvas | +0.003 |
product | +0.003 |
line | +0.003 |
shell | +0.003 |
Neg logit contributions | |
girls | -0.005 |
oof | -0.005 |
uddy | -0.004 |
Monkey | -0.004 |
Spot | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -7.05e-01 |
Pos logit contributions | |
girls | +0.019 |
Deal | +0.019 |
oof | +0.019 |
uddy | +0.018 |
Enjoy | +0.018 |
Neg logit contributions | |
line | -0.013 |
canvas | -0.013 |
lee | -0.012 |
tide | -0.012 |
shell | -0.012 |
Token | It |
Token ID | 632 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
Spot | +0.002 |
girls | +0.002 |
Buddy | +0.002 |
Zip | +0.002 |
uddy | +0.002 |
Neg logit contributions | |
canvas | -0.003 |
line | -0.003 |
product | -0.003 |
tide | -0.003 |
lee | -0.003 |
Token | was |
Token ID | 373 |
Feature activation | +9.47e-01 |
Pos logit contributions | |
Spot | +0.023 |
Buddy | +0.021 |
Zip | +0.021 |
Fl | +0.021 |
Max | +0.021 |
Neg logit contributions | |
canvas | -0.056 |
line | -0.056 |
product | -0.054 |
tide | -0.053 |
lee | -0.052 |
Token | a |
Token ID | 257 |
Feature activation | +8.48e-01 |
Pos logit contributions | |
Spot | +0.025 |
girls | +0.025 |
Buddy | +0.024 |
Zip | +0.024 |
uddy | +0.024 |
Neg logit contributions | |
canvas | -0.038 |
product | -0.038 |
line | -0.037 |
tide | -0.037 |
harbor | -0.036 |
Token | big |
Token ID | 1263 |
Feature activation | -5.50e-01 |
Pos logit contributions | |
girls | +0.034 |
Spot | +0.033 |
Zip | +0.033 |
Buddy | +0.033 |
uddy | +0.032 |
Neg logit contributions | |
canvas | -0.024 |
product | -0.023 |
tide | -0.022 |
harbor | -0.022 |
line | -0.022 |
Token | arc |
Token ID | 10389 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
canvas | +0.017 |
tide | +0.017 |
product | +0.016 |
line | +0.016 |
harbor | +0.016 |
Neg logit contributions | |
girls | -0.020 |
Spot | -0.020 |
Zip | -0.019 |
Buddy | -0.019 |
uddy | -0.019 |
Token | of |
Token ID | 286 |
Feature activation | +2.99e+00 |
Pos logit contributions | |
canvas | +0.007 |
product | +0.007 |
lee | +0.007 |
line | +0.007 |
pulp | +0.007 |
Neg logit contributions | |
Spot | -0.002 |
ever | -0.002 |
Zip | -0.002 |
Max | -0.002 |
girls | -0.002 |
Token | many |
Token ID | 867 |
Feature activation | +8.85e-01 |
Pos logit contributions | |
girls | +0.124 |
Spot | +0.122 |
uddy | +0.122 |
Buddy | +0.121 |
Zip | +0.120 |
Neg logit contributions | |
canvas | -0.080 |
tide | -0.075 |
product | -0.073 |
line | -0.072 |
harbor | -0.070 |
Token | colors |
Token ID | 7577 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
Spot | +0.032 |
girls | +0.031 |
Buddy | +0.030 |
Zip | +0.030 |
Fl | +0.030 |
Neg logit contributions | |
canvas | -0.027 |
line | -0.027 |
product | -0.026 |
tide | -0.026 |
lee | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
canvas | +0.006 |
product | +0.006 |
line | +0.005 |
tide | +0.005 |
lee | +0.005 |
Neg logit contributions | |
Spot | -0.002 |
Max | -0.002 |
Zip | -0.002 |
girls | -0.002 |
Buddy | -0.002 |
Token | It |
Token ID | 632 |
Feature activation | +9.34e-01 |
Pos logit contributions | |
canvas | +0.005 |
line | +0.005 |
product | +0.005 |
tide | +0.005 |
harbor | +0.005 |
Neg logit contributions | |
girls | -0.004 |
uddy | -0.004 |
oof | -0.004 |
Buddy | -0.004 |
Johnson | -0.004 |
Token | was |
Token ID | 373 |
Feature activation | +9.02e-01 |
Pos logit contributions | |
Spot | +0.019 |
Fl | +0.018 |
Zip | +0.018 |
Max | +0.018 |
Buddy | +0.017 |
Neg logit contributions | |
line | -0.045 |
canvas | -0.045 |
product | -0.043 |
tide | -0.042 |
lee | -0.042 |
Token | it |
Token ID | 340 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
Spot | +0.001 |
girls | +0.001 |
Buddy | +0.001 |
Zip | +0.001 |
Max | +0.001 |
Neg logit contributions | |
canvas | -0.002 |
line | -0.002 |
product | -0.002 |
tide | -0.002 |
lee | -0.002 |
Token | out |
Token ID | 503 |
Feature activation | +9.32e-01 |
Pos logit contributions | |
Spot | +0.007 |
Buddy | +0.007 |
Zip | +0.007 |
girls | +0.007 |
Max | +0.006 |
Neg logit contributions | |
canvas | -0.010 |
line | -0.010 |
product | -0.010 |
tide | -0.010 |
lee | -0.009 |
Token | of |
Token ID | 286 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
Spot | +0.025 |
girls | +0.025 |
oof | +0.023 |
Buddy | +0.023 |
Zip | +0.023 |
Neg logit contributions | |
canvas | -0.038 |
product | -0.037 |
line | -0.037 |
tide | -0.036 |
lee | -0.036 |
Token | blocks |
Token ID | 7021 |
Feature activation | +4.38e-01 |
Pos logit contributions | |
uddy | +0.060 |
girls | +0.058 |
ver | +0.055 |
Deal | +0.054 |
ll | +0.054 |
Neg logit contributions | |
canvas | -0.054 |
product | -0.051 |
tide | -0.051 |
harbor | -0.049 |
shell | -0.048 |
Token | and |
Token ID | 290 |
Feature activation | -3.23e-01 |
Pos logit contributions | |
girls | +0.011 |
Spot | +0.011 |
Buddy | +0.010 |
oof | +0.010 |
Zip | +0.010 |
Neg logit contributions | |
line | -0.019 |
canvas | -0.019 |
product | -0.018 |
tide | -0.018 |
harbor | -0.017 |
Token | stuffed |
Token ID | 22259 |
Feature activation | +3.22e+00 |
Pos logit contributions | |
canvas | +0.011 |
line | +0.010 |
product | +0.010 |
tide | +0.010 |
harbor | +0.010 |
Neg logit contributions | |
oof | -0.011 |
uddy | -0.011 |
girls | -0.011 |
Spot | -0.011 |
Zip | -0.011 |
Token | animals |
Token ID | 4695 |
Feature activation | -8.10e-02 |
Pos logit contributions | |
uddy | +0.107 |
ver | +0.103 |
OK | +0.100 |
Deal | +0.100 |
oof | +0.099 |
Neg logit contributions | |
canvas | -0.104 |
lee | -0.101 |
product | -0.101 |
pumpkin | -0.100 |
line | -0.099 |
Token | , |
Token ID | 11 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
line | +0.003 |
canvas | +0.003 |
product | +0.002 |
pumpkin | +0.002 |
shell | +0.002 |
Neg logit contributions | |
oof | -0.003 |
Zip | -0.003 |
Spot | -0.003 |
Kitty | -0.003 |
Buddy | -0.003 |
Token | but |
Token ID | 475 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
canvas | +0.008 |
line | +0.008 |
tide | +0.007 |
harbor | +0.007 |
pumpkin | +0.007 |
Neg logit contributions | |
uddy | -0.007 |
oof | -0.007 |
girls | -0.007 |
Zip | -0.006 |
Buddy | -0.006 |
Token | it |
Token ID | 340 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
line | +0.011 |
tide | +0.011 |
canvas | +0.011 |
product | +0.010 |
shell | +0.010 |
Neg logit contributions | |
girls | -0.017 |
oof | -0.016 |
uddy | -0.016 |
Enjoy | -0.015 |
Zip | -0.014 |
Token | didn |
Token ID | 1422 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
girls | +0.030 |
oof | +0.030 |
Spot | +0.029 |
Enjoy | +0.029 |
Zip | +0.028 |
Neg logit contributions | |
product | -0.040 |
tide | -0.040 |
canvas | -0.040 |
line | -0.038 |
shell | -0.038 |
Token | His |
Token ID | 2399 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
canvas | +0.030 |
line | +0.030 |
product | +0.030 |
tide | +0.029 |
lee | +0.029 |
Neg logit contributions | |
Spot | -0.013 |
girls | -0.013 |
Buddy | -0.012 |
Zip | -0.012 |
Wh | -0.012 |
Token | mum |
Token ID | 25682 |
Feature activation | -9.31e-02 |
Pos logit contributions | |
girls | +0.011 |
Spot | +0.011 |
Zip | +0.011 |
Buddy | +0.010 |
Wh | +0.010 |
Neg logit contributions | |
canvas | -0.008 |
product | -0.008 |
tide | -0.008 |
shell | -0.007 |
harbor | -0.007 |
Token | said |
Token ID | 531 |
Feature activation | +7.34e-01 |
Pos logit contributions | |
canvas | +0.003 |
product | +0.003 |
line | +0.003 |
pumpkin | +0.003 |
lee | +0.003 |
Neg logit contributions | |
girls | -0.003 |
oof | -0.003 |
Zip | -0.003 |
uddy | -0.003 |
Buddy | -0.003 |
Token | it |
Token ID | 340 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
girls | +0.025 |
oof | +0.024 |
Spot | +0.024 |
Zip | +0.023 |
uddy | +0.023 |
Neg logit contributions | |
canvas | -0.025 |
line | -0.025 |
product | -0.024 |
tide | -0.024 |
pumpkin | -0.023 |
Token | was |
Token ID | 373 |
Feature activation | +5.44e-01 |
Pos logit contributions | |
oof | +0.047 |
girls | +0.047 |
Zip | +0.042 |
losers | +0.042 |
Deal | +0.041 |
Neg logit contributions | |
product | -0.039 |
canvas | -0.037 |
line | -0.035 |
pumpkin | -0.034 |
tide | -0.034 |
Token | time |
Token ID | 640 |
Feature activation | +9.19e-01 |
Pos logit contributions | |
girls | +0.017 |
uddy | +0.016 |
Zip | +0.016 |
Buddy | +0.015 |
oof | +0.015 |
Neg logit contributions | |
product | -0.021 |
canvas | -0.021 |
tide | -0.020 |
lee | -0.019 |
harbor | -0.019 |
Token | to |
Token ID | 284 |
Feature activation | +7.03e-02 |
Pos logit contributions | |
girls | +0.020 |
Zip | +0.018 |
Spot | +0.018 |
Buddy | +0.018 |
Monkey | +0.018 |
Neg logit contributions | |
line | -0.044 |
canvas | -0.043 |
product | -0.042 |
tide | -0.041 |
lee | -0.041 |
Token | buy |
Token ID | 2822 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
Spot | +0.002 |
girls | +0.002 |
Buddy | +0.002 |
Zip | +0.002 |
Max | +0.002 |
Neg logit contributions | |
canvas | -0.002 |
line | -0.002 |
product | -0.002 |
tide | -0.002 |
lee | -0.002 |
Token | some |
Token ID | 617 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
product | +0.004 |
canvas | +0.004 |
tide | +0.004 |
line | +0.004 |
pumpkin | +0.004 |
Neg logit contributions | |
uddy | -0.003 |
girls | -0.003 |
Zip | -0.003 |
Buddy | -0.003 |
Monkey | -0.003 |
Token | salt |
Token ID | 8268 |
Feature activation | -3.89e-01 |
Pos logit contributions | |
Spot | +0.007 |
girls | +0.007 |
Zip | +0.006 |
Buddy | +0.006 |
uddy | +0.006 |
Neg logit contributions | |
canvas | -0.007 |
product | -0.007 |
tide | -0.007 |
line | -0.007 |
shell | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -5.72e-01 |
Pos logit contributions | |
line | +0.015 |
product | +0.015 |
canvas | +0.015 |
tide | +0.015 |
shell | +0.014 |
Neg logit contributions | |
Zip | -0.011 |
girls | -0.011 |
Buddy | -0.011 |
Spot | -0.011 |
uddy | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | +8.16e-01 |
Pos logit contributions | |
Spot | +0.014 |
girls | +0.014 |
Buddy | +0.013 |
Zip | +0.013 |
oof | +0.013 |
Neg logit contributions | |
canvas | -0.023 |
line | -0.023 |
product | -0.022 |
tide | -0.022 |
lee | -0.021 |
Token | " |
Token ID | 366 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
girls | +0.023 |
oof | +0.023 |
uddy | +0.022 |
Spot | +0.022 |
Zip | +0.022 |
Neg logit contributions | |
line | -0.032 |
canvas | -0.031 |
tide | -0.030 |
product | -0.030 |
lee | -0.029 |
Token | OK |
Token ID | 11380 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
Spot | +0.006 |
girls | +0.006 |
Buddy | +0.006 |
Zip | +0.005 |
Max | +0.005 |
Neg logit contributions | |
canvas | -0.012 |
line | -0.012 |
product | -0.011 |
tide | -0.011 |
lee | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | +6.02e-01 |
Pos logit contributions | |
girls | +0.005 |
Spot | +0.005 |
Buddy | +0.005 |
Zip | +0.005 |
uddy | +0.005 |
Neg logit contributions | |
canvas | -0.007 |
line | -0.007 |
product | -0.007 |
tide | -0.007 |
lee | -0.006 |
Token | Mia |
Token ID | 32189 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
girls | +0.022 |
uddy | +0.020 |
isk | +0.020 |
oof | +0.019 |
ver | +0.019 |
Neg logit contributions | |
canvas | -0.020 |
tide | -0.019 |
line | -0.019 |
shell | -0.018 |
pumpkin | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +8.41e-01 |
Pos logit contributions | |
line | +0.009 |
canvas | +0.009 |
product | +0.009 |
lee | +0.009 |
pumpkin | +0.009 |
Neg logit contributions | |
oof | -0.017 |
Zip | -0.017 |
girls | -0.017 |
Monkey | -0.017 |
Kenny | -0.017 |
Token | No |
Token ID | 1400 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
girls | +0.027 |
isk | +0.026 |
oof | +0.026 |
Monkey | +0.025 |
Johnson | +0.025 |
Neg logit contributions | |
line | -0.025 |
canvas | -0.024 |
tide | -0.023 |
pumpkin | -0.021 |
shell | -0.021 |
Token | more |
Token ID | 517 |
Feature activation | -6.30e-02 |
Pos logit contributions | |
canvas | +0.010 |
tide | +0.010 |
product | +0.010 |
shell | +0.009 |
pumpkin | +0.009 |
Neg logit contributions | |
girls | -0.009 |
uddy | -0.008 |
isk | -0.008 |
Spot | -0.008 |
Buddy | -0.008 |
Token | balancing |
Token ID | 22486 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
canvas | +0.002 |
line | +0.002 |
product | +0.002 |
tide | +0.002 |
lee | +0.002 |
Neg logit contributions | |
Spot | -0.002 |
girls | -0.002 |
Buddy | -0.002 |
Zip | -0.002 |
Max | -0.002 |
Token | on |
Token ID | 319 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
line | +0.023 |
canvas | +0.023 |
tide | +0.022 |
product | +0.022 |
shell | +0.021 |
Neg logit contributions | |
uddy | -0.020 |
girls | -0.019 |
Johnson | -0.019 |
Buddy | -0.019 |
Monkey | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | -7.15e-02 |
Pos logit contributions | |
canvas | +0.006 |
product | +0.005 |
tide | +0.005 |
pumpkin | +0.005 |
shell | +0.005 |
Neg logit contributions | |
girls | -0.006 |
uddy | -0.006 |
Deal | -0.005 |
ver | -0.005 |
Johnson | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
girls | +0.053 |
oof | +0.052 |
Zip | +0.050 |
uddy | +0.049 |
Buddy | +0.049 |
Neg logit contributions | |
line | -0.070 |
product | -0.069 |
canvas | -0.068 |
lee | -0.065 |
harbor | -0.063 |
Token | Ab |
Token ID | 2275 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
girls | +0.004 |
oof | +0.003 |
uddy | +0.003 |
Zip | +0.003 |
Bones | +0.003 |
Neg logit contributions | |
canvas | -0.003 |
line | -0.003 |
lee | -0.003 |
product | -0.003 |
harbor | -0.003 |
Token | bie |
Token ID | 12590 |
Feature activation | -6.37e-01 |
Pos logit contributions | |
line | +0.006 |
canvas | +0.006 |
product | +0.006 |
tide | +0.006 |
harbor | +0.006 |
Neg logit contributions | |
Spot | -0.004 |
girls | -0.004 |
oof | -0.004 |
Fl | -0.004 |
uddy | -0.004 |
Token | told |
Token ID | 1297 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
canvas | +0.024 |
line | +0.023 |
product | +0.023 |
tide | +0.023 |
lee | +0.022 |
Neg logit contributions | |
Spot | -0.019 |
girls | -0.018 |
Buddy | -0.018 |
Zip | -0.018 |
Wh | -0.017 |
Token | everyone |
Token ID | 2506 |
Feature activation | +8.21e-02 |
Pos logit contributions | |
Spot | +0.027 |
girls | +0.025 |
Buddy | +0.025 |
Zip | +0.025 |
Fl | +0.024 |
Neg logit contributions | |
canvas | -0.043 |
line | -0.043 |
product | -0.042 |
tide | -0.041 |
lee | -0.040 |
Token | about |
Token ID | 546 |
Feature activation | +6.27e-01 |
Pos logit contributions | |
Spot | +0.002 |
Buddy | +0.002 |
girls | +0.002 |
Zip | +0.002 |
Max | +0.002 |
Neg logit contributions | |
canvas | -0.004 |
line | -0.004 |
product | -0.004 |
tide | -0.004 |
lee | -0.003 |
Token | her |
Token ID | 607 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
uddy | +0.023 |
girls | +0.023 |
oof | +0.022 |
Zip | +0.021 |
ver | +0.021 |
Neg logit contributions | |
canvas | -0.021 |
product | -0.020 |
pumpkin | -0.019 |
tide | -0.019 |
line | -0.019 |
Token | special |
Token ID | 2041 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
girls | +0.006 |
Spot | +0.006 |
Buddy | +0.005 |
Zip | +0.005 |
Fl | +0.005 |
Neg logit contributions | |
canvas | -0.004 |
product | -0.004 |
tide | -0.004 |
line | -0.004 |
lee | -0.004 |
Token | exam |
Token ID | 2814 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
canvas | +0.011 |
product | +0.010 |
line | +0.010 |
tide | +0.010 |
lee | +0.009 |
Neg logit contributions | |
Spot | -0.015 |
girls | -0.015 |
Buddy | -0.015 |
Zip | -0.015 |
Fl | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
canvas | +0.011 |
line | +0.011 |
product | +0.011 |
lee | +0.009 |
tide | +0.009 |
Neg logit contributions | |
oof | -0.011 |
girls | -0.011 |
Buddy | -0.010 |
Spot | -0.010 |
uddy | -0.010 |
Token | They |
Token ID | 1119 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
canvas | +0.001 |
line | +0.001 |
product | +0.001 |
tide | +0.001 |
harbor | +0.001 |
Neg logit contributions | |
Spot | -0.001 |
Fl | -0.001 |
Buddy | -0.000 |
Zip | -0.000 |
Kitty | -0.000 |
Token | upon |
Token ID | 2402 |
Feature activation | +4.81e-02 |
Pos logit contributions | |
canvas | +0.041 |
line | +0.040 |
product | +0.040 |
tide | +0.039 |
lee | +0.039 |
Neg logit contributions | |
Spot | -0.013 |
girls | -0.013 |
Buddy | -0.012 |
Zip | -0.012 |
Wh | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | +6.47e-01 |
Pos logit contributions | |
girls | +0.002 |
uddy | +0.002 |
Monkey | +0.002 |
Johnson | +0.001 |
isk | +0.001 |
Neg logit contributions | |
tide | -0.002 |
canvas | -0.002 |
line | -0.001 |
shell | -0.001 |
product | -0.001 |
Token | time |
Token ID | 640 |
Feature activation | -4.30e-01 |
Pos logit contributions | |
girls | +0.020 |
Monkey | +0.018 |
Wh | +0.018 |
Spot | +0.018 |
Zip | +0.017 |
Neg logit contributions | |
canvas | -0.024 |
product | -0.023 |
tide | -0.023 |
harbor | -0.022 |
shell | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | +5.88e-01 |
Pos logit contributions | |
line | +0.014 |
canvas | +0.014 |
tide | +0.014 |
product | +0.013 |
lee | +0.013 |
Neg logit contributions | |
girls | -0.015 |
oof | -0.013 |
Spot | -0.013 |
Deal | -0.013 |
Monkey | -0.012 |
Token | there |
Token ID | 612 |
Feature activation | +5.21e-01 |
Pos logit contributions | |
girls | +0.026 |
oof | +0.023 |
Monkey | +0.023 |
Spot | +0.023 |
uddy | +0.023 |
Neg logit contributions | |
canvas | -0.014 |
line | -0.014 |
tide | -0.013 |
product | -0.013 |
harbor | -0.013 |
Token | was |
Token ID | 373 |
Feature activation | +7.70e-01 |
Pos logit contributions | |
girls | +0.020 |
oof | +0.018 |
losers | +0.017 |
Monkey | +0.017 |
uddy | +0.017 |
Neg logit contributions | |
product | -0.016 |
tide | -0.015 |
line | -0.015 |
shell | -0.015 |
pumpkin | -0.014 |
Token | a |
Token ID | 257 |
Feature activation | +6.08e-01 |
Pos logit contributions | |
girls | +0.025 |
uddy | +0.024 |
isk | +0.023 |
Monkey | +0.022 |
Deal | +0.022 |
Neg logit contributions | |
tide | -0.028 |
canvas | -0.027 |
harbor | -0.027 |
pumpkin | -0.027 |
shell | -0.025 |
Token | little |
Token ID | 1310 |
Feature activation | -3.17e-02 |
Pos logit contributions | |
Spot | +0.019 |
Buddy | +0.018 |
girls | +0.018 |
Max | +0.018 |
Fl | +0.018 |
Neg logit contributions | |
canvas | -0.022 |
line | -0.022 |
product | -0.021 |
tide | -0.021 |
lee | -0.021 |
Token | girl |
Token ID | 2576 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
canvas | +0.001 |
line | +0.001 |
product | +0.001 |
tide | +0.001 |
lee | +0.001 |
Neg logit contributions | |
Spot | -0.001 |
Buddy | -0.001 |
girls | -0.001 |
Zip | -0.001 |
Fl | -0.001 |
Token | named |
Token ID | 3706 |
Feature activation | +9.00e-01 |
Pos logit contributions | |
line | +0.002 |
tide | +0.002 |
product | +0.002 |
pulp | +0.002 |
tides | +0.002 |
Neg logit contributions | |
Spot | -0.000 |
Fl | -0.000 |
Max | -0.000 |
Wh | -0.000 |
Kitty | -0.000 |
Token | Jane |
Token ID | 12091 |
Feature activation | -5.90e-01 |
Pos logit contributions | |
girls | +0.022 |
Spot | +0.019 |
Monkey | +0.018 |
Zip | +0.018 |
oof | +0.018 |
Neg logit contributions | |
canvas | -0.042 |
line | -0.039 |
lee | -0.039 |
product | -0.038 |
tide | -0.038 |
Token | from |
Token ID | 422 |
Feature activation | +3.85e-01 |
Pos logit contributions | |
girls | +0.042 |
oof | +0.041 |
uddy | +0.040 |
Deal | +0.039 |
Buddy | +0.038 |
Neg logit contributions | |
canvas | -0.047 |
product | -0.045 |
tide | -0.045 |
line | -0.044 |
lee | -0.043 |
Token | the |
Token ID | 262 |
Feature activation | +5.61e-02 |
Pos logit contributions | |
uddy | +0.016 |
girls | +0.015 |
ver | +0.015 |
ll | +0.015 |
Deal | +0.014 |
Neg logit contributions | |
canvas | -0.012 |
tide | -0.012 |
pumpkin | -0.011 |
product | -0.011 |
shell | -0.010 |
Token | case |
Token ID | 1339 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
girls | +0.002 |
oof | +0.002 |
Spot | +0.002 |
Zip | +0.002 |
ver | +0.002 |
Neg logit contributions | |
canvas | -0.002 |
tide | -0.001 |
product | -0.001 |
harbor | -0.001 |
shell | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
girls | +0.006 |
oof | +0.005 |
Wh | +0.005 |
Enjoy | +0.005 |
Buddy | +0.005 |
Neg logit contributions | |
line | -0.005 |
canvas | -0.005 |
shell | -0.004 |
tide | -0.004 |
harbor | -0.004 |
Token | It |
Token ID | 632 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
girls | +0.011 |
oof | +0.010 |
Deal | +0.009 |
ever | +0.009 |
losers | +0.009 |
Neg logit contributions | |
line | -0.013 |
lee | -0.013 |
canvas | -0.012 |
product | -0.012 |
tide | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
girls | +0.031 |
oof | +0.031 |
Spot | +0.030 |
uddy | +0.029 |
Buddy | +0.029 |
Neg logit contributions | |
product | -0.036 |
canvas | -0.036 |
tide | -0.035 |
shell | -0.035 |
lee | -0.034 |
Token | glad |
Token ID | 9675 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
girls | +0.039 |
uddy | +0.039 |
Deal | +0.037 |
oof | +0.035 |
Enjoy | +0.035 |
Neg logit contributions | |
product | -0.034 |
tide | -0.033 |
canvas | -0.033 |
lee | -0.032 |
shell | -0.032 |
Token | to |
Token ID | 284 |
Feature activation | +5.98e-01 |
Pos logit contributions | |
girls | +0.009 |
oof | +0.008 |
uddy | +0.008 |
Spot | +0.008 |
Zip | +0.008 |
Neg logit contributions | |
canvas | -0.009 |
tide | -0.008 |
line | -0.008 |
product | -0.008 |
lee | -0.008 |
Token | be |
Token ID | 307 |
Feature activation | -5.91e-01 |
Pos logit contributions | |
Spot | +0.017 |
girls | +0.017 |
Buddy | +0.017 |
Zip | +0.016 |
uddy | +0.016 |
Neg logit contributions | |
canvas | -0.023 |
line | -0.023 |
product | -0.022 |
tide | -0.022 |
lee | -0.021 |
Token | free |
Token ID | 1479 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
canvas | +0.022 |
line | +0.022 |
product | +0.022 |
tide | +0.021 |
lee | +0.021 |
Neg logit contributions | |
Spot | -0.018 |
girls | -0.017 |
Zip | -0.017 |
Buddy | -0.017 |
uddy | -0.017 |
Token | from |
Token ID | 422 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
oof | +0.009 |
girls | +0.009 |
uddy | +0.009 |
Deal | +0.009 |
Zip | +0.008 |
Neg logit contributions | |
line | -0.008 |
canvas | -0.008 |
product | -0.007 |
shell | -0.007 |
lee | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | -4.92e-01 |
Pos logit contributions | |
line | +0.017 |
canvas | +0.016 |
lee | +0.016 |
tide | +0.015 |
product | +0.015 |
Neg logit contributions | |
girls | -0.016 |
oof | -0.015 |
losers | -0.014 |
uddy | -0.014 |
isk | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -6.56e-01 |
Pos logit contributions | |
line | +0.019 |
canvas | +0.019 |
product | +0.019 |
lee | +0.019 |
tide | +0.018 |
Neg logit contributions | |
girls | -0.013 |
Spot | -0.012 |
Monkey | -0.012 |
Buddy | -0.012 |
Zip | -0.011 |
Token | " |
Token ID | 1 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
canvas | +0.023 |
line | +0.022 |
product | +0.022 |
tide | +0.022 |
lee | +0.021 |
Neg logit contributions | |
Spot | -0.022 |
girls | -0.021 |
Buddy | -0.021 |
Zip | -0.021 |
Max | -0.020 |
Token | Give |
Token ID | 23318 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
canvas | +0.052 |
line | +0.051 |
product | +0.050 |
tide | +0.049 |
lee | +0.049 |
Neg logit contributions | |
Spot | -0.038 |
girls | -0.036 |
Zip | -0.036 |
Buddy | -0.036 |
Fl | -0.035 |
Token | it |
Token ID | 340 |
Feature activation | +5.54e-01 |
Pos logit contributions | |
Spot | +0.004 |
Buddy | +0.004 |
Zip | +0.004 |
girls | +0.004 |
Max | +0.004 |
Neg logit contributions | |
canvas | -0.007 |
line | -0.007 |
product | -0.007 |
tide | -0.007 |
harbor | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
Spot | +0.015 |
girls | +0.015 |
Buddy | +0.014 |
Zip | +0.014 |
Max | +0.014 |
Neg logit contributions | |
canvas | -0.022 |
line | -0.022 |
product | -0.022 |
tide | -0.021 |
lee | -0.020 |
Token | me |
Token ID | 502 |
Feature activation | +2.50e-01 |
Pos logit contributions | |
Spot | +0.005 |
Buddy | +0.005 |
Max | +0.005 |
Zip | +0.005 |
Wh | +0.005 |
Neg logit contributions | |
line | -0.010 |
canvas | -0.010 |
product | -0.009 |
tide | -0.009 |
lee | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
girls | +0.008 |
uddy | +0.007 |
Buddy | +0.007 |
Spot | +0.007 |
Zip | +0.007 |
Neg logit contributions | |
canvas | -0.009 |
line | -0.009 |
product | -0.009 |
shell | -0.009 |
pumpkin | -0.009 |
Token | Tom |
Token ID | 4186 |
Feature activation | -8.75e-01 |
Pos logit contributions | |
girls | +0.006 |
uddy | +0.005 |
Spot | +0.005 |
Buddy | +0.005 |
Zip | +0.005 |
Neg logit contributions | |
canvas | -0.009 |
line | -0.008 |
tide | -0.008 |
product | -0.008 |
shell | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +3.76e-01 |
Pos logit contributions | |
canvas | +0.034 |
line | +0.033 |
product | +0.033 |
tide | +0.032 |
lee | +0.031 |
Neg logit contributions | |
Spot | -0.024 |
girls | -0.024 |
Buddy | -0.023 |
Zip | -0.023 |
uddy | -0.022 |
Token | It |
Token ID | 632 |
Feature activation | +8.71e-01 |
Pos logit contributions | |
Spot | +0.010 |
girls | +0.010 |
Buddy | +0.010 |
Zip | +0.010 |
uddy | +0.009 |
Neg logit contributions | |
canvas | -0.015 |
line | -0.015 |
product | -0.015 |
tide | -0.014 |
lee | -0.014 |
Token | Who |
Token ID | 8241 |
Feature activation | -9.47e-01 |
Pos logit contributions | |
canvas | +0.023 |
line | +0.022 |
product | +0.022 |
lee | +0.022 |
tide | +0.022 |
Neg logit contributions | |
Spot | -0.019 |
Zip | -0.018 |
girls | -0.018 |
Buddy | -0.018 |
Wh | -0.017 |
Token | is |
Token ID | 318 |
Feature activation | -8.34e-01 |
Pos logit contributions | |
canvas | +0.036 |
product | +0.034 |
tide | +0.034 |
shell | +0.033 |
line | +0.033 |
Neg logit contributions | |
girls | -0.027 |
Buddy | -0.026 |
Spot | -0.025 |
Johnson | -0.025 |
Monkey | -0.025 |
Token | there |
Token ID | 612 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
canvas | +0.032 |
line | +0.032 |
product | +0.032 |
lee | +0.031 |
tide | +0.030 |
Neg logit contributions | |
Spot | -0.025 |
Max | -0.024 |
Buddy | -0.024 |
Zip | -0.023 |
Fl | -0.023 |
Token | ?" |
Token ID | 1701 |
Feature activation | -3.33e-01 |
Pos logit contributions | |
girls | +0.011 |
uddy | +0.011 |
Zip | +0.011 |
Spot | +0.011 |
Buddy | +0.011 |
Neg logit contributions | |
canvas | -0.007 |
line | -0.006 |
tide | -0.006 |
product | -0.006 |
harbor | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
canvas | +0.017 |
line | +0.017 |
product | +0.017 |
tide | +0.017 |
lee | +0.016 |
Neg logit contributions | |
Spot | -0.005 |
Buddy | -0.005 |
Zip | -0.005 |
Max | -0.005 |
girls | -0.005 |
Token | voice |
Token ID | 3809 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
canvas | +0.000 |
line | +0.000 |
product | +0.000 |
lee | +0.000 |
tide | +0.000 |
Neg logit contributions | |
Spot | -0.000 |
Buddy | -0.000 |
Max | -0.000 |
Zip | -0.000 |
Fl | -0.000 |
Token | said |
Token ID | 531 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
canvas | +0.016 |
pulp | +0.016 |
product | +0.016 |
harbor | +0.016 |
tide | +0.015 |
Neg logit contributions | |
Spot | -0.008 |
Max | -0.008 |
Buddy | -0.008 |
Zip | -0.008 |
Kitty | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
Spot | +0.008 |
girls | +0.007 |
Buddy | +0.007 |
Zip | +0.007 |
Max | +0.007 |
Neg logit contributions | |
canvas | -0.010 |
line | -0.009 |
product | -0.009 |
tide | -0.009 |
harbor | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
girls | +0.005 |
Spot | +0.005 |
Buddy | +0.005 |
oof | +0.005 |
Zip | +0.005 |
Neg logit contributions | |
canvas | -0.009 |
line | -0.009 |
product | -0.009 |
tide | -0.009 |
lee | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
canvas | +0.002 |
line | +0.002 |
product | +0.002 |
tide | +0.002 |
lee | +0.002 |
Neg logit contributions | |
Spot | -0.001 |
girls | -0.001 |
Buddy | -0.001 |
Zip | -0.001 |
Max | -0.001 |
Token | Tom |
Token ID | 13787 |
Feature activation | -9.92e-02 |
Pos logit contributions | |
canvas | +0.009 |
tide | +0.009 |
harbor | +0.009 |
pumpkin | +0.009 |
pulp | +0.009 |
Neg logit contributions | |
Spot | -0.004 |
Max | -0.003 |
Zip | -0.003 |
Wh | -0.003 |
Max | -0.003 |
Token | throw |
Token ID | 3714 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
canvas | +0.012 |
line | +0.012 |
tide | +0.011 |
shell | +0.011 |
harbor | +0.011 |
Neg logit contributions | |
uddy | -0.011 |
Johnson | -0.010 |
oof | -0.010 |
Buddy | -0.010 |
Spot | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | -4.99e-01 |
Pos logit contributions | |
ll | +0.010 |
uddy | +0.010 |
ver | +0.010 |
Deal | +0.010 |
Johnson | +0.009 |
Neg logit contributions | |
shell | -0.006 |
waves | -0.006 |
tide | -0.006 |
shells | -0.006 |
pumpkin | -0.006 |
Token | ball |
Token ID | 2613 |
Feature activation | -4.61e-01 |
Pos logit contributions | |
canvas | +0.018 |
shell | +0.017 |
line | +0.017 |
tide | +0.017 |
product | +0.016 |
Neg logit contributions | |
girls | -0.016 |
Kitty | -0.015 |
isk | -0.015 |
Buddy | -0.015 |
Fl | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | +6.81e-01 |
Pos logit contributions | |
line | +0.016 |
shell | +0.015 |
canvas | +0.015 |
tide | +0.015 |
pumpkin | +0.015 |
Neg logit contributions | |
girls | -0.014 |
oof | -0.014 |
Johnson | -0.014 |
Monkey | -0.014 |
Kenny | -0.014 |
Token | Rex |
Token ID | 17853 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
uddy | +0.021 |
oof | +0.020 |
ver | +0.020 |
Johnson | +0.020 |
girls | +0.020 |
Neg logit contributions | |
canvas | -0.024 |
line | -0.023 |
tide | -0.022 |
product | -0.022 |
shell | -0.022 |
Token | catches |
Token ID | 17591 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
line | +0.006 |
canvas | +0.006 |
product | +0.006 |
shell | +0.006 |
tide | +0.006 |
Neg logit contributions | |
Johnson | -0.005 |
uddy | -0.005 |
oof | -0.005 |
girls | -0.005 |
Buddy | -0.005 |
Token | it |
Token ID | 340 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
line | +0.006 |
waves | +0.006 |
tide | +0.005 |
shell | +0.005 |
pumpkin | +0.005 |
Neg logit contributions | |
uddy | -0.009 |
Deal | -0.009 |
ver | -0.009 |
ll | -0.009 |
Johnson | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +5.11e-01 |
Pos logit contributions | |
oof | +0.009 |
Kenny | +0.009 |
girls | +0.009 |
Monkey | +0.009 |
uddy | +0.009 |
Neg logit contributions | |
shell | -0.005 |
line | -0.005 |
canvas | -0.005 |
product | -0.005 |
tide | -0.005 |
Token | They |
Token ID | 1119 |
Feature activation | -3.81e-01 |
Pos logit contributions | |
atum | +0.018 |
isk | +0.018 |
Deal | +0.017 |
ll | +0.017 |
oof | +0.017 |
Neg logit contributions | |
line | -0.015 |
tide | -0.014 |
lee | -0.013 |
harbor | -0.013 |
tidal | -0.013 |
Token | have |
Token ID | 423 |
Feature activation | +2.99e-02 |
Pos logit contributions | |
canvas | +0.013 |
line | +0.013 |
tide | +0.013 |
shell | +0.013 |
harbor | +0.012 |
Neg logit contributions | |
uddy | -0.012 |
Johnson | -0.012 |
oof | -0.011 |
Buddy | -0.011 |
Kitty | -0.011 |
Token | fun |
Token ID | 1257 |
Feature activation | +6.62e-01 |
Pos logit contributions | |
uddy | +0.001 |
Monkey | +0.001 |
Johnson | +0.001 |
ver | +0.001 |
girls | +0.001 |
Neg logit contributions | |
canvas | -0.001 |
tide | -0.001 |
shell | -0.001 |
product | -0.001 |
harbor | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | -7.37e-01 |
Pos logit contributions | |
line | +0.026 |
canvas | +0.024 |
product | +0.022 |
shell | +0.022 |
harbor | +0.022 |
Neg logit contributions | |
Zip | -0.026 |
oof | -0.025 |
girls | -0.025 |
Kitty | -0.024 |
uddy | -0.024 |
Token | apologize |
Token ID | 16521 |
Feature activation | +6.94e-02 |
Pos logit contributions | |
canvas | +0.029 |
line | +0.027 |
product | +0.027 |
shell | +0.027 |
tide | +0.027 |
Neg logit contributions | |
girls | -0.021 |
uddy | -0.021 |
Zip | -0.020 |
oof | -0.020 |
Johnson | -0.019 |
Token | when |
Token ID | 618 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
girls | +0.002 |
oof | +0.002 |
Zip | +0.002 |
uddy | +0.002 |
Johnson | +0.002 |
Neg logit contributions | |
line | -0.003 |
canvas | -0.003 |
product | -0.002 |
lee | -0.002 |
shell | -0.002 |
Token | we |
Token ID | 356 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
canvas | +0.041 |
tide | +0.040 |
product | +0.040 |
line | +0.040 |
harbor | +0.038 |
Neg logit contributions | |
girls | -0.035 |
uddy | -0.034 |
oof | -0.033 |
Spot | -0.033 |
Zip | -0.033 |
Token | make |
Token ID | 787 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
canvas | +0.009 |
line | +0.009 |
tide | +0.008 |
harbor | +0.008 |
product | +0.008 |
Neg logit contributions | |
Zip | -0.007 |
Spot | -0.007 |
uddy | -0.007 |
girls | -0.007 |
Kitty | -0.007 |
Token | mistakes |
Token ID | 10135 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
canvas | +0.051 |
product | +0.050 |
line | +0.049 |
tide | +0.049 |
lee | +0.048 |
Neg logit contributions | |
Spot | -0.040 |
girls | -0.040 |
Zip | -0.039 |
uddy | -0.039 |
Buddy | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
line | +0.009 |
canvas | +0.009 |
product | +0.008 |
tide | +0.008 |
lee | +0.008 |
Neg logit contributions | |
oof | -0.009 |
Zip | -0.009 |
girls | -0.009 |
uddy | -0.009 |
Spot | -0.009 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
girls | +0.007 |
Zip | +0.006 |
uddy | +0.006 |
Spot | +0.006 |
Buddy | +0.006 |
Neg logit contributions | |
line | -0.009 |
lee | -0.009 |
canvas | -0.009 |
product | -0.009 |
pumpkin | -0.008 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
girls | +0.007 |
Deal | +0.007 |
losers | +0.007 |
oof | +0.007 |
uddy | +0.007 |
Neg logit contributions | |
line | -0.005 |
product | -0.005 |
tide | -0.005 |
lee | -0.005 |
harbor | -0.005 |
Token | upon |
Token ID | 2402 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
canvas | +0.056 |
line | +0.055 |
product | +0.055 |
tide | +0.054 |
lee | +0.053 |
Neg logit contributions | |
Spot | -0.017 |
girls | -0.016 |
Buddy | -0.016 |
Zip | -0.015 |
Max | -0.015 |
Token | a |
Token ID | 257 |
Feature activation | +6.31e-01 |
Pos logit contributions | |
canvas | +0.009 |
line | +0.009 |
tide | +0.009 |
product | +0.009 |
shell | +0.009 |
Neg logit contributions | |
girls | -0.006 |
uddy | -0.006 |
Monkey | -0.006 |
Zip | -0.006 |
Buddy | -0.006 |
Token | it |
Token ID | 340 |
Feature activation | +4.26e-01 |
Pos logit contributions | |
uddy | +0.010 |
girls | +0.010 |
Buddy | +0.010 |
Zip | +0.010 |
Spot | +0.009 |
Neg logit contributions | |
product | -0.011 |
canvas | -0.011 |
line | -0.011 |
tide | -0.011 |
lee | -0.010 |
Token | ." |
Token ID | 526 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
girls | +0.016 |
Zip | +0.015 |
Buddy | +0.015 |
uddy | +0.015 |
oof | +0.014 |
Neg logit contributions | |
line | -0.013 |
product | -0.012 |
canvas | -0.012 |
tide | -0.012 |
shell | -0.011 |
Token | So |
Token ID | 1406 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
girls | +0.006 |
losers | +0.005 |
ver | +0.005 |
Hero | +0.005 |
oof | +0.005 |
Neg logit contributions | |
canvas | -0.008 |
lee | -0.008 |
line | -0.008 |
product | -0.008 |
pumpkin | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | -3.20e-02 |
Pos logit contributions | |
girls | +0.008 |
oof | +0.007 |
isk | +0.007 |
Hero | +0.007 |
uddy | +0.007 |
Neg logit contributions | |
canvas | -0.007 |
tide | -0.006 |
pumpkin | -0.006 |
line | -0.006 |
product | -0.006 |
Token | Tim |
Token ID | 5045 |
Feature activation | +2.74e-01 |
Pos logit contributions | |
canvas | +0.001 |
tide | +0.001 |
pumpkin | +0.001 |
line | +0.001 |
product | +0.001 |
Neg logit contributions | |
girls | -0.001 |
oof | -0.001 |
uddy | -0.001 |
isk | -0.001 |
Hero | -0.001 |
Token | decided |
Token ID | 3066 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
girls | +0.010 |
Zip | +0.010 |
Spot | +0.009 |
Buddy | +0.009 |
oof | +0.009 |
Neg logit contributions | |
canvas | -0.008 |
product | -0.008 |
line | -0.008 |
lee | -0.008 |
shell | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
canvas | +0.013 |
product | +0.013 |
line | +0.013 |
tide | +0.013 |
lee | +0.012 |
Neg logit contributions | |
Spot | -0.005 |
Buddy | -0.004 |
Zip | -0.004 |
Max | -0.004 |
girls | -0.004 |
Token | help |
Token ID | 1037 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
canvas | +0.011 |
product | +0.011 |
line | +0.011 |
tide | +0.010 |
pumpkin | +0.010 |
Neg logit contributions | |
girls | -0.013 |
uddy | -0.012 |
Zip | -0.012 |
Buddy | -0.012 |
oof | -0.012 |
Token | his |
Token ID | 465 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
girls | +0.036 |
Spot | +0.035 |
Zip | +0.034 |
Buddy | +0.034 |
uddy | +0.034 |
Neg logit contributions | |
canvas | -0.054 |
product | -0.054 |
line | -0.053 |
tide | -0.051 |
lee | -0.050 |
Token | neighbors |
Token ID | 12020 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
Spot | +0.001 |
girls | +0.001 |
Buddy | +0.001 |
Zip | +0.001 |
Max | +0.001 |
Neg logit contributions | |
canvas | -0.001 |
line | -0.001 |
tide | -0.001 |
product | -0.001 |
lee | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
oof | +0.007 |
Zip | +0.006 |
girls | +0.006 |
Spot | +0.006 |
Buddy | +0.006 |
Neg logit contributions | |
line | -0.006 |
canvas | -0.006 |
product | -0.006 |
harbor | -0.005 |
pumpkin | -0.005 |
Token | good |
Token ID | 922 |
Feature activation | -6.39e-01 |
Pos logit contributions | |
canvas | +0.017 |
product | +0.016 |
pulp | +0.016 |
line | +0.016 |
lee | +0.015 |
Neg logit contributions | |
Spot | -0.011 |
Zip | -0.010 |
Max | -0.010 |
Buddy | -0.010 |
Fl | -0.010 |
Token | at |
Token ID | 379 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
line | +0.025 |
canvas | +0.024 |
tide | +0.024 |
product | +0.024 |
lee | +0.023 |
Neg logit contributions | |
girls | -0.018 |
Spot | -0.018 |
Buddy | -0.017 |
Zip | -0.017 |
uddy | -0.017 |
Token | balancing |
Token ID | 22486 |
Feature activation | -3.78e-01 |
Pos logit contributions | |
canvas | +0.014 |
pulp | +0.013 |
product | +0.013 |
line | +0.013 |
lee | +0.013 |
Neg logit contributions | |
Spot | -0.012 |
Max | -0.011 |
first | -0.011 |
Zip | -0.011 |
Fl | -0.011 |
Token | on |
Token ID | 319 |
Feature activation | -4.63e-01 |
Pos logit contributions | |
canvas | +0.015 |
line | +0.015 |
product | +0.015 |
tide | +0.015 |
lee | +0.014 |
Neg logit contributions | |
Spot | -0.010 |
girls | -0.010 |
Buddy | -0.010 |
Zip | -0.010 |
uddy | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | -7.98e-02 |
Pos logit contributions | |
tide | +0.015 |
canvas | +0.015 |
line | +0.015 |
product | +0.014 |
shell | +0.014 |
Neg logit contributions | |
uddy | -0.016 |
girls | -0.016 |
ver | -0.015 |
Buddy | -0.015 |
Monkey | -0.015 |
Token | board |
Token ID | 3096 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
product | +0.003 |
lee | +0.002 |
pulp | +0.002 |
rys | +0.002 |
canvas | +0.002 |
Neg logit contributions | |
first | -0.002 |
Spot | -0.002 |
Max | -0.002 |
for | -0.002 |
love | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
line | +0.061 |
canvas | +0.060 |
tide | +0.059 |
product | +0.058 |
shell | +0.056 |
Neg logit contributions | |
Spot | -0.050 |
girls | -0.050 |
Buddy | -0.049 |
Kitty | -0.048 |
Wh | -0.048 |
Token | How |
Token ID | 1374 |
Feature activation | -4.82e-01 |
Pos logit contributions | |
girls | +0.007 |
oof | +0.006 |
uddy | +0.006 |
Spot | +0.006 |
Buddy | +0.006 |
Neg logit contributions | |
line | -0.009 |
canvas | -0.009 |
tide | -0.009 |
lee | -0.008 |
product | -0.008 |
Token | does |
Token ID | 857 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
canvas | +0.021 |
line | +0.021 |
product | +0.020 |
tide | +0.020 |
lee | +0.020 |
Neg logit contributions | |
Spot | -0.011 |
girls | -0.011 |
Buddy | -0.011 |
Zip | -0.011 |
Wh | -0.010 |
Token | he |
Token ID | 339 |
Feature activation | +9.78e-01 |
Pos logit contributions | |
canvas | +0.017 |
pulp | +0.017 |
lee | +0.017 |
line | +0.016 |
product | +0.016 |
Neg logit contributions | |
Spot | -0.010 |
Max | -0.009 |
Zip | -0.009 |
Buddy | -0.009 |
Fl | -0.008 |
Token | do |
Token ID | 466 |
Feature activation | -5.34e-01 |
Pos logit contributions | |
Spot | +0.030 |
girls | +0.029 |
Buddy | +0.029 |
Zip | +0.028 |
Fl | +0.028 |
Neg logit contributions | |
canvas | -0.036 |
line | -0.035 |
product | -0.035 |
tide | -0.034 |
lee | -0.033 |
Token | Be |
Token ID | 1355 |
Feature activation | -7.10e-01 |
Pos logit contributions | |
girls | +0.030 |
uddy | +0.026 |
isk | +0.026 |
ll | +0.026 |
oof | +0.026 |
Neg logit contributions | |
line | -0.030 |
canvas | -0.030 |
lee | -0.029 |
tide | -0.028 |
rys | -0.027 |
Token | careful |
Token ID | 8161 |
Feature activation | +7.88e-02 |
Pos logit contributions | |
product | +0.023 |
canvas | +0.023 |
line | +0.022 |
tide | +0.022 |
shell | +0.022 |
Neg logit contributions | |
girls | -0.024 |
uddy | -0.024 |
Zip | -0.023 |
Spot | -0.023 |
Monkey | -0.023 |
Token | not |
Token ID | 407 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
girls | +0.002 |
Spot | +0.002 |
Zip | +0.002 |
oof | +0.002 |
Monkey | +0.002 |
Neg logit contributions | |
canvas | -0.003 |
line | -0.003 |
product | -0.003 |
tide | -0.003 |
lee | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
girls | +0.001 |
Spot | +0.001 |
Zip | +0.001 |
Buddy | +0.001 |
oof | +0.001 |
Neg logit contributions | |
canvas | -0.001 |
line | -0.001 |
tide | -0.001 |
product | -0.001 |
shell | -0.001 |
Token | hurt |
Token ID | 5938 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
Spot | +0.001 |
girls | +0.001 |
Buddy | +0.001 |
Zip | +0.001 |
uddy | +0.001 |
Neg logit contributions | |
canvas | -0.001 |
line | -0.001 |
product | -0.001 |
tide | -0.001 |
harbor | -0.001 |
Token | yourself |
Token ID | 3511 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
girls | +0.034 |
uddy | +0.033 |
oof | +0.032 |
ver | +0.031 |
Deal | +0.031 |
Neg logit contributions | |
canvas | -0.033 |
line | -0.033 |
tide | -0.032 |
product | -0.031 |
shell | -0.030 |
Token | again |
Token ID | 757 |
Feature activation | +8.48e-01 |
Pos logit contributions | |
line | +0.068 |
canvas | +0.067 |
product | +0.064 |
tide | +0.062 |
harbor | +0.061 |
Neg logit contributions | |
girls | -0.060 |
oof | -0.059 |
uddy | -0.059 |
Zip | -0.059 |
Buddy | -0.058 |
Token | ," |
Token ID | 553 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
uddy | +0.027 |
girls | +0.027 |
Zip | +0.027 |
Spot | +0.027 |
Buddy | +0.026 |
Neg logit contributions | |
canvas | -0.029 |
line | -0.028 |
product | -0.028 |
tide | -0.027 |
lee | -0.027 |
Token | said |
Token ID | 531 |
Feature activation | +4.73e-01 |
Pos logit contributions | |
girls | +0.036 |
uddy | +0.034 |
Zip | +0.033 |
Spot | +0.033 |
oof | +0.032 |
Neg logit contributions | |
canvas | -0.036 |
product | -0.035 |
line | -0.035 |
tide | -0.034 |
pumpkin | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
girls | +0.014 |
oof | +0.013 |
uddy | +0.013 |
Deal | +0.013 |
Zip | +0.013 |
Neg logit contributions | |
canvas | -0.017 |
product | -0.017 |
tide | -0.017 |
line | -0.017 |
pumpkin | -0.017 |
Token | nurse |
Token ID | 15849 |
Feature activation | +3.78e-02 |
Pos logit contributions | |
canvas | +0.033 |
product | +0.032 |
tide | +0.032 |
line | +0.030 |
harbor | +0.030 |
Neg logit contributions | |
girls | -0.043 |
Buddy | -0.041 |
oof | -0.041 |
Zip | -0.041 |
uddy | -0.040 |
Token | his |
Token ID | 465 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
uddy | +0.039 |
girls | +0.038 |
oof | +0.037 |
Johnson | +0.037 |
Wh | +0.035 |
Neg logit contributions | |
tide | -0.032 |
lee | -0.031 |
shell | -0.030 |
tides | -0.030 |
anchor | -0.030 |
Token | friends |
Token ID | 2460 |
Feature activation | +8.92e-01 |
Pos logit contributions | |
lee | +0.001 |
shell | +0.001 |
pumpkin | +0.001 |
tide | +0.001 |
harbor | +0.001 |
Neg logit contributions | |
Wh | -0.001 |
isk | -0.001 |
Johnson | -0.001 |
Spot | -0.001 |
Fl | -0.001 |
Token | about |
Token ID | 546 |
Feature activation | +7.17e-01 |
Pos logit contributions | |
girls | +0.024 |
oof | +0.023 |
Wh | +0.022 |
Spot | +0.022 |
uddy | +0.022 |
Neg logit contributions | |
line | -0.036 |
canvas | -0.035 |
tide | -0.034 |
lee | -0.034 |
harbor | -0.034 |
Token | how |
Token ID | 703 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
uddy | +0.025 |
girls | +0.024 |
oof | +0.024 |
ver | +0.022 |
Wh | +0.022 |
Neg logit contributions | |
tide | -0.023 |
canvas | -0.023 |
product | -0.023 |
pumpkin | -0.023 |
harbor | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
canvas | +0.006 |
line | +0.005 |
product | +0.005 |
tide | +0.005 |
harbor | +0.005 |
Neg logit contributions | |
Spot | -0.004 |
Buddy | -0.004 |
girls | -0.004 |
Zip | -0.004 |
Max | -0.003 |
Token | big |
Token ID | 1263 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
tide | +0.006 |
harbor | +0.006 |
canvas | +0.005 |
product | +0.005 |
pumpkin | +0.005 |
Neg logit contributions | |
Spot | -0.009 |
girls | -0.009 |
oof | -0.009 |
Fl | -0.008 |
uddy | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | -5.24e-01 |
Pos logit contributions | |
canvas | +0.055 |
tide | +0.054 |
product | +0.053 |
line | +0.052 |
harbor | +0.051 |
Neg logit contributions | |
Spot | -0.065 |
girls | -0.065 |
uddy | -0.063 |
oof | -0.063 |
Zip | -0.062 |
Token | strong |
Token ID | 1913 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
canvas | +0.016 |
tide | +0.015 |
line | +0.015 |
product | +0.014 |
harbor | +0.014 |
Neg logit contributions | |
girls | -0.020 |
uddy | -0.020 |
oof | -0.020 |
Spot | -0.019 |
Zip | -0.019 |
Token | rabbit |
Token ID | 22746 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
tide | +0.033 |
canvas | +0.032 |
line | +0.031 |
harbor | +0.031 |
product | +0.030 |
Neg logit contributions | |
girls | -0.041 |
Spot | -0.041 |
oof | -0.039 |
uddy | -0.039 |
Fl | -0.039 |
Token | helped |
Token ID | 4193 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
oof | +0.005 |
girls | +0.005 |
uddy | +0.005 |
Wh | +0.005 |
Spot | +0.004 |
Neg logit contributions | |
line | -0.006 |
shell | -0.006 |
harbor | -0.005 |
product | -0.005 |
tide | -0.005 |
Token | him |
Token ID | 683 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
oof | +0.014 |
uddy | +0.013 |
girls | +0.013 |
Bones | +0.013 |
Wh | +0.013 |
Neg logit contributions | |
line | -0.024 |
canvas | -0.024 |
pumpkin | -0.023 |
product | -0.023 |
tide | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | +2.79e-01 |
Pos logit contributions | |
girls | +0.008 |
uddy | +0.008 |
oof | +0.007 |
isk | +0.007 |
Monkey | +0.007 |
Neg logit contributions | |
line | -0.008 |
tide | -0.008 |
canvas | -0.008 |
harbor | -0.008 |
shell | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
uddy | +0.011 |
girls | +0.011 |
oof | +0.010 |
isk | +0.010 |
Zip | +0.010 |
Neg logit contributions | |
tide | -0.008 |
canvas | -0.008 |
line | -0.008 |
pumpkin | -0.008 |
shell | -0.008 |
Token | w |
Token ID | 266 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
tide | +0.014 |
canvas | +0.013 |
line | +0.013 |
product | +0.013 |
shell | +0.013 |
Neg logit contributions | |
girls | -0.014 |
Zip | -0.014 |
ver | -0.014 |
uddy | -0.014 |
Spot | -0.014 |
Token | affle |
Token ID | 30697 |
Feature activation | +8.73e-01 |
Pos logit contributions | |
canvas | +0.013 |
line | +0.013 |
product | +0.013 |
tide | +0.013 |
lee | +0.013 |
Neg logit contributions | |
Spot | -0.012 |
girls | -0.012 |
Buddy | -0.011 |
Zip | -0.011 |
Max | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
girls | +0.026 |
Zip | +0.026 |
oof | +0.026 |
Spot | +0.025 |
Buddy | +0.025 |
Neg logit contributions | |
canvas | -0.031 |
line | -0.031 |
product | -0.030 |
tide | -0.030 |
shell | -0.030 |
Token | feeling |
Token ID | 4203 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
canvas | +0.007 |
product | +0.007 |
tide | +0.007 |
harbor | +0.006 |
line | +0.006 |
Neg logit contributions | |
girls | -0.006 |
uddy | -0.005 |
Zip | -0.005 |
Monkey | -0.005 |
Bones | -0.005 |
Token | very |
Token ID | 845 |
Feature activation | +3.22e-01 |
Pos logit contributions | |
canvas | +0.076 |
line | +0.075 |
product | +0.075 |
tide | +0.074 |
lee | +0.071 |
Neg logit contributions | |
Spot | -0.047 |
girls | -0.046 |
Zip | -0.045 |
Buddy | -0.045 |
uddy | -0.043 |
Token | cold |
Token ID | 4692 |
Feature activation | +9.85e-02 |
Pos logit contributions | |
Spot | +0.009 |
Buddy | +0.009 |
girls | +0.009 |
Fl | +0.009 |
Zip | +0.009 |
Neg logit contributions | |
canvas | -0.012 |
product | -0.012 |
line | -0.012 |
tide | -0.012 |
lee | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
girls | +0.003 |
Spot | +0.003 |
Zip | +0.003 |
Buddy | +0.003 |
uddy | +0.003 |
Neg logit contributions | |
canvas | -0.003 |
line | -0.003 |
tide | -0.003 |
product | -0.003 |
harbor | -0.003 |
Token | It |
Token ID | 632 |
Feature activation | +8.04e-01 |
Pos logit contributions | |
canvas | +0.004 |
line | +0.004 |
tide | +0.004 |
lee | +0.003 |
harbor | +0.003 |
Neg logit contributions | |
girls | -0.003 |
isk | -0.003 |
uddy | -0.003 |
oof | -0.003 |
Johnson | -0.003 |
Token | wanted |
Token ID | 2227 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
girls | +0.024 |
Spot | +0.024 |
Zip | +0.024 |
Buddy | +0.023 |
uddy | +0.023 |
Neg logit contributions | |
canvas | -0.029 |
product | -0.028 |
tide | -0.028 |
line | -0.028 |
lee | -0.027 |
Token | ily |
Token ID | 813 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
canvas | +0.004 |
line | +0.004 |
lee | +0.003 |
product | +0.003 |
tide | +0.003 |
Neg logit contributions | |
Spot | -0.005 |
girls | -0.005 |
Buddy | -0.005 |
Zip | -0.004 |
Monkey | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
canvas | +0.009 |
line | +0.009 |
product | +0.009 |
lee | +0.009 |
tide | +0.009 |
Neg logit contributions | |
girls | -0.007 |
Spot | -0.007 |
Buddy | -0.007 |
Zip | -0.007 |
uddy | -0.007 |
Token | Tom |
Token ID | 4186 |
Feature activation | +6.90e-02 |
Pos logit contributions | |
uddy | +0.007 |
oof | +0.007 |
girls | +0.007 |
ver | +0.007 |
Deal | +0.007 |
Neg logit contributions | |
canvas | -0.010 |
lee | -0.010 |
pumpkin | -0.010 |
line | -0.009 |
tide | -0.009 |
Token | held |
Token ID | 2714 |
Feature activation | -6.79e-01 |
Pos logit contributions | |
girls | +0.002 |
Spot | +0.002 |
Buddy | +0.002 |
Zip | +0.002 |
uddy | +0.002 |
Neg logit contributions | |
canvas | -0.002 |
line | -0.002 |
tide | -0.002 |
lee | -0.002 |
product | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | -5.95e-01 |
Pos logit contributions | |
canvas | +0.015 |
lee | +0.015 |
pumpkin | +0.015 |
shell | +0.014 |
product | +0.014 |
Neg logit contributions | |
ll | -0.029 |
uddy | -0.029 |
isk | -0.028 |
girls | -0.028 |
ver | -0.027 |
Token | old |
Token ID | 1468 |
Feature activation | -2.88e+00 |
Pos logit contributions | |
canvas | +0.018 |
tide | +0.017 |
product | +0.017 |
pumpkin | +0.017 |
shell | +0.016 |
Neg logit contributions | |
girls | -0.023 |
Spot | -0.023 |
Buddy | -0.023 |
Zip | -0.022 |
Wh | -0.022 |
Token | man |
Token ID | 582 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
canvas | +0.088 |
line | +0.085 |
product | +0.085 |
tide | +0.083 |
lee | +0.079 |
Neg logit contributions | |
Spot | -0.106 |
girls | -0.104 |
Buddy | -0.103 |
Zip | -0.102 |
Fl | -0.100 |
Token | 's |
Token ID | 338 |
Feature activation | -9.61e-01 |
Pos logit contributions | |
canvas | +0.056 |
product | +0.056 |
line | +0.055 |
shell | +0.053 |
pumpkin | +0.052 |
Neg logit contributions | |
girls | -0.052 |
Spot | -0.051 |
Buddy | -0.050 |
Wh | -0.050 |
Zip | -0.050 |
Token | hands |
Token ID | 2832 |
Feature activation | -7.44e-01 |
Pos logit contributions | |
canvas | +0.026 |
product | +0.025 |
line | +0.024 |
tide | +0.024 |
shell | +0.024 |
Neg logit contributions | |
girls | -0.039 |
Spot | -0.039 |
Zip | -0.038 |
Buddy | -0.038 |
Wh | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | -9.40e-02 |
Pos logit contributions | |
line | +0.024 |
canvas | +0.022 |
pumpkin | +0.021 |
shell | +0.021 |
harbor | +0.020 |
Neg logit contributions | |
girls | -0.028 |
oof | -0.028 |
Zip | -0.027 |
uddy | -0.027 |
isk | -0.027 |
Token | led |
Token ID | 2957 |
Feature activation | -7.68e-01 |
Pos logit contributions | |
canvas | +0.003 |
line | +0.003 |
product | +0.003 |
tide | +0.003 |
lee | +0.003 |
Neg logit contributions | |
uddy | -0.003 |
oof | -0.003 |
girls | -0.003 |
Zip | -0.003 |
Spot | -0.003 |