Token | it |
Token ID | 340 |
Feature activation | +4.75e-02 |
Pos logit contributions | |
Mic | +0.077 |
ery | +0.075 |
Max | +0.074 |
mice | +0.072 |
oyd | +0.071 |
Neg logit contributions | |
goodbye | -0.062 |
contain | -0.050 |
wave | -0.046 |
attention | -0.045 |
fy | -0.045 |
Token | was |
Token ID | 373 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
Mic | +0.094 |
Charlie | +0.089 |
Max | +0.087 |
ities | +0.083 |
elvet | +0.082 |
Neg logit contributions | |
goodbye | -0.072 |
contain | -0.062 |
wave | -0.061 |
look | -0.058 |
hug | -0.057 |
Token | the |
Token ID | 262 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
Max | +0.070 |
Mic | +0.069 |
ery | +0.066 |
Ted | +0.065 |
Bear | +0.065 |
Neg logit contributions | |
goodbye | -0.049 |
contain | -0.037 |
attention | -0.036 |
umb | -0.035 |
tour | -0.034 |
Token | most |
Token ID | 749 |
Feature activation | +6.42e-02 |
Pos logit contributions | |
Max | +0.039 |
Ted | +0.038 |
ery | +0.038 |
oon | +0.037 |
es | +0.037 |
Neg logit contributions | |
goodbye | -0.036 |
contain | -0.031 |
attention | -0.030 |
plunge | -0.030 |
wave | -0.029 |
Token | famous |
Token ID | 5863 |
Feature activation | +4.85e-02 |
Pos logit contributions | |
Max | +0.111 |
ery | +0.111 |
Mic | +0.111 |
Ted | +0.105 |
mice | +0.104 |
Neg logit contributions | |
goodbye | -0.117 |
contain | -0.099 |
plunge | -0.094 |
attention | -0.093 |
fy | -0.092 |
Token | avocado |
Token ID | 40377 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
Max | +0.065 |
Mic | +0.062 |
ery | +0.062 |
Ted | +0.060 |
Bear | +0.059 |
Neg logit contributions | |
goodbye | -0.111 |
contain | -0.093 |
tour | -0.090 |
plunge | -0.089 |
attention | -0.089 |
Token | in |
Token ID | 287 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
Mic | +0.232 |
Max | +0.223 |
Next | +0.214 |
ery | +0.212 |
Ted | +0.211 |
Neg logit contributions | |
goodbye | -0.178 |
contain | -0.148 |
tour | -0.139 |
toward | -0.135 |
fy | -0.134 |
Token | the |
Token ID | 262 |
Feature activation | +2.41e-02 |
Pos logit contributions | |
Mic | +0.063 |
ery | +0.062 |
Max | +0.060 |
iders | +0.057 |
ters | +0.057 |
Neg logit contributions | |
goodbye | -0.040 |
wave | -0.034 |
contain | -0.033 |
tour | -0.032 |
toward | -0.032 |
Token | world |
Token ID | 995 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
ery | +0.042 |
Max | +0.042 |
Mic | +0.041 |
Ted | +0.041 |
iders | +0.041 |
Neg logit contributions | |
goodbye | -0.043 |
tour | -0.036 |
contain | -0.035 |
wave | -0.034 |
plunge | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -1.44e-03 |
Pos logit contributions | |
Mic | +0.025 |
ery | +0.025 |
Max | +0.024 |
mice | +0.023 |
Ted | +0.023 |
Neg logit contributions | |
goodbye | -0.023 |
contain | -0.019 |
plunge | -0.018 |
toward | -0.018 |
umb | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
goodbye | +0.002 |
contain | +0.002 |
hi | +0.002 |
farewell | +0.002 |
plunge | +0.002 |
Neg logit contributions | |
ery | -0.003 |
iflower | -0.003 |
Mic | -0.003 |
ters | -0.003 |
mice | -0.003 |
Token | not |
Token ID | 407 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
Mic | +0.081 |
Max | +0.078 |
oyd | +0.078 |
Bear | +0.076 |
ery | +0.075 |
Neg logit contributions | |
goodbye | -0.086 |
attention | -0.075 |
umb | -0.071 |
contain | -0.068 |
tour | -0.068 |
Token | nice |
Token ID | 3621 |
Feature activation | +4.83e-02 |
Pos logit contributions | |
Mic | +0.051 |
Max | +0.051 |
oyd | +0.049 |
Mouse | +0.049 |
Ted | +0.049 |
Neg logit contributions | |
goodbye | -0.034 |
attention | -0.030 |
umb | -0.025 |
tickets | -0.025 |
contain | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
ery | +0.088 |
Mic | +0.088 |
Max | +0.086 |
mice | +0.083 |
Ted | +0.082 |
Neg logit contributions | |
goodbye | -0.083 |
contain | -0.069 |
plunge | -0.066 |
umb | -0.065 |
attention | -0.065 |
Token | That |
Token ID | 1320 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
ery | +0.040 |
iflower | +0.038 |
Mic | +0.037 |
oyd | +0.036 |
lf | +0.036 |
Neg logit contributions | |
goodbye | -0.040 |
hi | -0.032 |
hello | -0.031 |
umb | -0.030 |
owl | -0.030 |
Token | is |
Token ID | 318 |
Feature activation | +4.81e-02 |
Pos logit contributions | |
Mic | +0.059 |
Max | +0.054 |
ery | +0.052 |
Bear | +0.051 |
Fox | +0.051 |
Neg logit contributions | |
goodbye | -0.045 |
wave | -0.035 |
hug | -0.035 |
plunge | -0.035 |
attention | -0.034 |
Token | wasteful |
Token ID | 45393 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
Mic | +0.082 |
Max | +0.080 |
oyd | +0.077 |
ase | +0.077 |
oon | +0.077 |
Neg logit contributions | |
goodbye | -0.085 |
attention | -0.077 |
umb | -0.070 |
closer | -0.068 |
tickets | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
Mic | +0.231 |
ery | +0.218 |
Max | +0.216 |
Fox | +0.212 |
Bear | +0.211 |
Neg logit contributions | |
goodbye | -0.171 |
attention | -0.131 |
contain | -0.131 |
umb | -0.126 |
toward | -0.125 |
Token | You |
Token ID | 921 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
ery | +0.047 |
iflower | +0.043 |
Mic | +0.041 |
lf | +0.041 |
oyd | +0.041 |
Neg logit contributions | |
goodbye | -0.046 |
hi | -0.037 |
umb | -0.036 |
owl | -0.035 |
hello | -0.035 |
Token | have |
Token ID | 423 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
goodbye | +0.034 |
contain | +0.031 |
hug | +0.030 |
wave | +0.029 |
umb | +0.028 |
Neg logit contributions | |
Mic | -0.054 |
Charlie | -0.050 |
Max | -0.050 |
iflower | -0.050 |
Ted | -0.050 |
Token | to |
Token ID | 284 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
ery | +0.072 |
Mic | +0.072 |
Max | +0.070 |
Ted | +0.069 |
Bear | +0.068 |
Neg logit contributions | |
goodbye | -0.032 |
umb | -0.023 |
contain | -0.022 |
attention | -0.021 |
plunge | -0.020 |
Token | eat |
Token ID | 4483 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
ery | +0.022 |
Mic | +0.021 |
Max | +0.021 |
mice | +0.021 |
iflower | +0.020 |
Neg logit contributions | |
goodbye | -0.017 |
contain | -0.014 |
plunge | -0.013 |
umb | -0.013 |
attention | -0.013 |
Token | not |
Token ID | 407 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
ery | +0.064 |
Mic | +0.063 |
ase | +0.063 |
Bear | +0.062 |
oon | +0.061 |
Neg logit contributions | |
goodbye | -0.061 |
attention | -0.051 |
closer | -0.048 |
contain | -0.047 |
wave | -0.047 |
Token | dough |
Token ID | 15756 |
Feature activation | +8.13e-02 |
Pos logit contributions | |
Mic | +0.024 |
Max | +0.023 |
Bear | +0.022 |
Ted | +0.022 |
Fox | +0.022 |
Neg logit contributions | |
goodbye | -0.018 |
attention | -0.015 |
contain | -0.014 |
tickets | -0.014 |
umb | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
Mic | +0.180 |
ery | +0.175 |
Max | +0.174 |
Bear | +0.168 |
mice | +0.168 |
Neg logit contributions | |
goodbye | -0.108 |
contain | -0.088 |
plunge | -0.083 |
attention | -0.080 |
umb | -0.077 |
Token | that |
Token ID | 326 |
Feature activation | +5.03e-02 |
Pos logit contributions | |
Mic | +0.040 |
ery | +0.040 |
Max | +0.038 |
Bear | +0.037 |
Ted | +0.037 |
Neg logit contributions | |
goodbye | -0.048 |
contain | -0.040 |
plunge | -0.038 |
umb | -0.038 |
attention | -0.038 |
Token | is |
Token ID | 318 |
Feature activation | +4.93e-02 |
Pos logit contributions | |
Mic | +0.101 |
ascal | +0.091 |
ery | +0.090 |
Max | +0.090 |
oyd | +0.089 |
Neg logit contributions | |
goodbye | -0.081 |
hug | -0.065 |
wave | -0.065 |
umb | -0.063 |
plunge | -0.062 |
Token | clay |
Token ID | 21558 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
Mic | +0.085 |
Max | +0.082 |
ery | +0.080 |
Bear | +0.079 |
Ted | +0.078 |
Neg logit contributions | |
goodbye | -0.092 |
attention | -0.078 |
contain | -0.075 |
umb | -0.074 |
tickets | -0.073 |
Token | . |
Token ID | 13 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
Mic | +0.245 |
Max | +0.235 |
ery | +0.231 |
Bear | +0.226 |
Next | +0.226 |
Neg logit contributions | |
goodbye | -0.135 |
contain | -0.107 |
attention | -0.104 |
plunge | -0.103 |
wave | -0.101 |
Token | You |
Token ID | 921 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
ery | +0.028 |
iflower | +0.026 |
Mic | +0.026 |
lf | +0.024 |
Max | +0.024 |
Neg logit contributions | |
goodbye | -0.032 |
hi | -0.026 |
contain | -0.025 |
umb | -0.025 |
toward | -0.025 |
Token | can |
Token ID | 460 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
goodbye | +0.035 |
contain | +0.032 |
hug | +0.029 |
wave | +0.029 |
plunge | +0.029 |
Neg logit contributions | |
Mic | -0.060 |
Max | -0.057 |
iflower | -0.057 |
Charlie | -0.056 |
Ted | -0.056 |
Token | 't |
Token ID | 470 |
Feature activation | +5.32e-04 |
Pos logit contributions | |
goodbye | +0.025 |
contain | +0.023 |
plunge | +0.022 |
hug | +0.021 |
umb | +0.021 |
Neg logit contributions | |
ery | -0.023 |
Mic | -0.022 |
mice | -0.021 |
iflower | -0.021 |
Max | -0.021 |
Token | play |
Token ID | 711 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
ery | +0.001 |
Mic | +0.001 |
iflower | +0.001 |
mice | +0.001 |
Next | +0.001 |
Neg logit contributions | |
goodbye | -0.001 |
contain | -0.001 |
wave | -0.001 |
plunge | -0.001 |
hug | -0.001 |
Token | The |
Token ID | 383 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
goodbye | +0.025 |
contain | +0.021 |
plunge | +0.020 |
toward | +0.020 |
farewell | +0.020 |
Neg logit contributions | |
ery | -0.018 |
Mic | -0.018 |
Max | -0.017 |
iflower | -0.016 |
mice | -0.016 |
Token | cave |
Token ID | 11527 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
ery | +0.017 |
mice | +0.017 |
Max | +0.016 |
Mic | +0.016 |
Bear | +0.016 |
Neg logit contributions | |
goodbye | -0.021 |
contain | -0.019 |
oured | -0.018 |
umb | -0.018 |
fy | -0.018 |
Token | was |
Token ID | 373 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
ery | +0.027 |
Max | +0.025 |
Mic | +0.025 |
mice | +0.024 |
iflower | +0.023 |
Neg logit contributions | |
goodbye | -0.056 |
contain | -0.050 |
plunge | -0.049 |
umb | -0.048 |
attention | -0.048 |
Token | full |
Token ID | 1336 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
ery | +0.056 |
Mic | +0.055 |
Max | +0.054 |
mice | +0.053 |
iflower | +0.052 |
Neg logit contributions | |
goodbye | -0.053 |
contain | -0.044 |
plunge | -0.043 |
umb | -0.041 |
attention | -0.041 |
Token | of |
Token ID | 286 |
Feature activation | +7.99e-02 |
Pos logit contributions | |
ery | +0.035 |
mice | +0.035 |
Max | +0.034 |
Mic | +0.033 |
Bear | +0.032 |
Neg logit contributions | |
goodbye | -0.041 |
plunge | -0.036 |
contain | -0.035 |
umb | -0.034 |
oured | -0.034 |
Token | clay |
Token ID | 21558 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
ery | +0.126 |
mice | +0.126 |
Mic | +0.115 |
Max | +0.115 |
Next | +0.111 |
Neg logit contributions | |
goodbye | -0.155 |
contain | -0.138 |
oured | -0.132 |
plunge | -0.129 |
toward | -0.128 |
Token | and |
Token ID | 290 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
Mic | +0.193 |
ery | +0.192 |
Max | +0.189 |
Ted | +0.178 |
Bear | +0.177 |
Neg logit contributions | |
goodbye | -0.178 |
contain | -0.149 |
plunge | -0.146 |
umb | -0.142 |
attention | -0.141 |
Token | it |
Token ID | 340 |
Feature activation | +4.47e-02 |
Pos logit contributions | |
Mic | +0.072 |
ery | +0.069 |
Max | +0.064 |
ascal | +0.063 |
Bear | +0.062 |
Neg logit contributions | |
goodbye | -0.067 |
contain | -0.056 |
plunge | -0.055 |
wave | -0.053 |
attention | -0.053 |
Token | was |
Token ID | 373 |
Feature activation | +3.03e-02 |
Pos logit contributions | |
Mic | +0.091 |
ery | +0.081 |
Max | +0.080 |
Next | +0.080 |
Charlie | +0.079 |
Neg logit contributions | |
goodbye | -0.065 |
contain | -0.059 |
wave | -0.059 |
tour | -0.056 |
plunge | -0.056 |
Token | very |
Token ID | 845 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
Mic | +0.061 |
Max | +0.060 |
ery | +0.058 |
Bear | +0.057 |
mice | +0.056 |
Neg logit contributions | |
goodbye | -0.047 |
contain | -0.038 |
umb | -0.037 |
attention | -0.036 |
plunge | -0.036 |
Token | soft |
Token ID | 2705 |
Feature activation | +4.74e-02 |
Pos logit contributions | |
ery | +0.028 |
Mic | +0.026 |
mice | +0.026 |
Max | +0.026 |
iflower | +0.026 |
Neg logit contributions | |
goodbye | -0.035 |
contain | -0.030 |
plunge | -0.029 |
oured | -0.029 |
umb | -0.029 |
Token | It |
Token ID | 632 |
Feature activation | +4.79e-02 |
Pos logit contributions | |
goodbye | +0.001 |
contain | +0.001 |
toward | +0.001 |
hug | +0.001 |
tour | +0.001 |
Neg logit contributions | |
ery | -0.002 |
iflower | -0.002 |
Mic | -0.002 |
ters | -0.002 |
hes | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | +3.64e-02 |
Pos logit contributions | |
Mic | +0.093 |
ities | +0.091 |
ters | +0.090 |
mice | +0.086 |
magnets | +0.083 |
Neg logit contributions | |
goodbye | -0.072 |
tour | -0.061 |
contain | -0.059 |
wave | -0.058 |
hug | -0.058 |
Token | a |
Token ID | 257 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
Mic | +0.069 |
ery | +0.067 |
Max | +0.067 |
Bear | +0.065 |
iflower | +0.064 |
Neg logit contributions | |
goodbye | -0.061 |
contain | -0.050 |
tour | -0.047 |
attention | -0.047 |
plunge | -0.046 |
Token | very |
Token ID | 845 |
Feature activation | -5.42e-03 |
Pos logit contributions | |
ery | +0.027 |
Mic | +0.027 |
Max | +0.026 |
mice | +0.026 |
iflower | +0.025 |
Neg logit contributions | |
goodbye | -0.031 |
contain | -0.027 |
plunge | -0.026 |
umb | -0.025 |
attention | -0.025 |
Token | special |
Token ID | 2041 |
Feature activation | +4.46e-02 |
Pos logit contributions | |
goodbye | +0.011 |
contain | +0.010 |
umb | +0.010 |
plunge | +0.010 |
oured | +0.010 |
Neg logit contributions | |
ery | -0.008 |
Mic | -0.007 |
iflower | -0.007 |
mice | -0.007 |
Max | -0.007 |
Token | avocado |
Token ID | 40377 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
ery | +0.062 |
Mic | +0.061 |
Max | +0.059 |
mice | +0.058 |
iflower | +0.056 |
Neg logit contributions | |
goodbye | -0.096 |
contain | -0.084 |
plunge | -0.081 |
umb | -0.079 |
attention | -0.079 |
Token | because |
Token ID | 780 |
Feature activation | +3.76e-03 |
Pos logit contributions | |
Mic | +0.218 |
Max | +0.216 |
Next | +0.205 |
ery | +0.205 |
Ted | +0.204 |
Neg logit contributions | |
goodbye | -0.151 |
contain | -0.119 |
fy | -0.114 |
tour | -0.114 |
wave | -0.111 |
Token | it |
Token ID | 340 |
Feature activation | +5.27e-02 |
Pos logit contributions | |
Mic | +0.008 |
ery | +0.008 |
Max | +0.008 |
oyd | +0.007 |
Ted | +0.007 |
Neg logit contributions | |
goodbye | -0.006 |
closer | -0.004 |
plunge | -0.004 |
contain | -0.004 |
umb | -0.004 |
Token | was |
Token ID | 373 |
Feature activation | +3.97e-02 |
Pos logit contributions | |
Mic | +0.110 |
ities | +0.109 |
elvet | +0.107 |
Charlie | +0.106 |
Next | +0.103 |
Neg logit contributions | |
look | -0.057 |
contain | -0.054 |
goodbye | -0.053 |
wave | -0.050 |
tour | -0.048 |
Token | spicy |
Token ID | 26880 |
Feature activation | +6.65e-02 |
Pos logit contributions | |
Max | +0.087 |
Mic | +0.085 |
Ted | +0.082 |
iders | +0.081 |
ery | +0.081 |
Neg logit contributions | |
goodbye | -0.056 |
contain | -0.044 |
attention | -0.042 |
umb | -0.040 |
plunge | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -1.28e-03 |
Pos logit contributions | |
Mic | +0.122 |
Max | +0.120 |
ery | +0.119 |
Ted | +0.113 |
Bear | +0.112 |
Neg logit contributions | |
goodbye | -0.115 |
contain | -0.095 |
plunge | -0.093 |
attention | -0.092 |
umb | -0.090 |
Token | , |
Token ID | 11 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
goodbye | +0.046 |
contain | +0.038 |
plunge | +0.036 |
umb | +0.035 |
attention | +0.034 |
Neg logit contributions | |
ery | -0.064 |
Mic | -0.063 |
Max | -0.062 |
mice | -0.061 |
iflower | -0.059 |
Token | " |
Token ID | 366 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
goodbye | +0.028 |
contain | +0.021 |
toward | +0.021 |
farewell | +0.021 |
hi | +0.021 |
Neg logit contributions | |
ery | -0.024 |
Mic | -0.022 |
mice | -0.021 |
iflower | -0.021 |
Max | -0.020 |
Token | That |
Token ID | 2504 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
ery | +0.040 |
mice | +0.040 |
Midnight | +0.036 |
iflower | +0.034 |
Mic | +0.034 |
Neg logit contributions | |
goodbye | -0.067 |
contain | -0.064 |
toward | -0.063 |
des | -0.062 |
fy | -0.061 |
Token | is |
Token ID | 318 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
Mic | +0.042 |
ery | +0.039 |
Max | +0.038 |
mice | +0.036 |
Ted | +0.036 |
Neg logit contributions | |
goodbye | -0.049 |
plunge | -0.041 |
wave | -0.040 |
contain | -0.040 |
tour | -0.040 |
Token | because |
Token ID | 780 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
Mic | +0.060 |
ery | +0.058 |
Max | +0.056 |
ase | +0.055 |
chini | +0.054 |
Neg logit contributions | |
goodbye | -0.064 |
attention | -0.053 |
contain | -0.051 |
closer | -0.051 |
plunge | -0.049 |
Token | clay |
Token ID | 21558 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
goodbye | +0.022 |
contain | +0.018 |
plunge | +0.018 |
umb | +0.018 |
attention | +0.018 |
Neg logit contributions | |
Mic | -0.019 |
ery | -0.018 |
Max | -0.018 |
Ted | -0.017 |
mice | -0.017 |
Token | is |
Token ID | 318 |
Feature activation | +3.63e-02 |
Pos logit contributions | |
Mic | +0.226 |
Max | +0.215 |
Charlie | +0.203 |
Ted | +0.201 |
Doctor | +0.201 |
Neg logit contributions | |
goodbye | -0.149 |
plunge | -0.121 |
attention | -0.121 |
contain | -0.120 |
umb | -0.113 |
Token | very |
Token ID | 845 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
Mic | +0.064 |
Max | +0.062 |
ery | +0.059 |
Ted | +0.058 |
Bear | +0.058 |
Neg logit contributions | |
goodbye | -0.067 |
attention | -0.057 |
umb | -0.056 |
contain | -0.056 |
plunge | -0.054 |
Token | fragile |
Token ID | 21049 |
Feature activation | +7.37e-02 |
Pos logit contributions | |
ery | +0.039 |
Mic | +0.039 |
Max | +0.039 |
mice | +0.037 |
Ted | +0.036 |
Neg logit contributions | |
goodbye | -0.046 |
contain | -0.040 |
plunge | -0.038 |
umb | -0.037 |
attention | -0.037 |
Token | , |
Token ID | 11 |
Feature activation | +2.48e-02 |
Pos logit contributions | |
Mic | +0.136 |
ery | +0.135 |
Max | +0.134 |
Ted | +0.127 |
mice | +0.126 |
Neg logit contributions | |
goodbye | -0.125 |
contain | -0.104 |
plunge | -0.100 |
attention | -0.098 |
umb | -0.097 |
Token | Johnny |
Token ID | 15470 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
ery | +0.043 |
Mic | +0.042 |
Max | +0.041 |
mice | +0.040 |
iflower | +0.039 |
Neg logit contributions | |
goodbye | -0.046 |
contain | -0.039 |
plunge | -0.037 |
umb | -0.037 |
attention | -0.036 |
Token | a |
Token ID | 257 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
Mic | +0.104 |
ery | +0.102 |
iflower | +0.097 |
mice | +0.096 |
Max | +0.095 |
Neg logit contributions | |
goodbye | -0.139 |
contain | -0.119 |
plunge | -0.113 |
attention | -0.108 |
farewell | -0.108 |
Token | time |
Token ID | 640 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
goodbye | +0.023 |
contain | +0.020 |
plunge | +0.019 |
fy | +0.019 |
umb | +0.019 |
Neg logit contributions | |
ery | -0.015 |
Mic | -0.014 |
mice | -0.014 |
Max | -0.014 |
iflower | -0.014 |
Token | there |
Token ID | 612 |
Feature activation | -5.03e-03 |
Pos logit contributions | |
goodbye | +0.023 |
contain | +0.019 |
plunge | +0.018 |
umb | +0.017 |
attention | +0.017 |
Neg logit contributions | |
ery | -0.034 |
Mic | -0.034 |
Max | -0.033 |
mice | -0.033 |
Ted | -0.032 |
Token | were |
Token ID | 547 |
Feature activation | +4.63e-02 |
Pos logit contributions | |
goodbye | +0.007 |
plunge | +0.006 |
contain | +0.006 |
inspection | +0.006 |
look | +0.006 |
Neg logit contributions | |
Mic | -0.011 |
mice | -0.010 |
ery | -0.010 |
iflower | -0.009 |
ters | -0.009 |
Token | two |
Token ID | 734 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
ery | +0.059 |
Mic | +0.055 |
iflower | +0.054 |
Max | +0.053 |
iders | +0.052 |
Neg logit contributions | |
goodbye | -0.106 |
umb | -0.094 |
plunge | -0.092 |
contain | -0.089 |
tickets | -0.087 |
Token | twin |
Token ID | 15203 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
ery | +0.042 |
Mic | +0.041 |
Max | +0.040 |
mice | +0.037 |
iflower | +0.037 |
Neg logit contributions | |
goodbye | -0.090 |
contain | -0.080 |
plunge | -0.078 |
umb | -0.077 |
attention | -0.076 |
Token | boys |
Token ID | 6510 |
Feature activation | +5.70e-02 |
Pos logit contributions | |
ery | +0.167 |
Mic | +0.164 |
chini | +0.160 |
iflower | +0.159 |
Max | +0.158 |
Neg logit contributions | |
goodbye | -0.187 |
hug | -0.159 |
umb | -0.158 |
tour | -0.149 |
ogging | -0.147 |
Token | - |
Token ID | 532 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
ery | +0.114 |
Mic | +0.110 |
iflower | +0.108 |
chini | +0.107 |
Max | +0.106 |
Neg logit contributions | |
goodbye | -0.088 |
contain | -0.071 |
umb | -0.070 |
hug | -0.067 |
toward | -0.067 |
Token | Joe |
Token ID | 5689 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
ery | +0.041 |
iflower | +0.038 |
Mic | +0.038 |
chini | +0.036 |
Max | +0.034 |
Neg logit contributions | |
goodbye | -0.065 |
contain | -0.055 |
umb | -0.055 |
tour | -0.053 |
plunge | -0.053 |
Token | and |
Token ID | 290 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
Mic | +0.044 |
chini | +0.042 |
mice | +0.041 |
Max | +0.041 |
ery | +0.041 |
Neg logit contributions | |
goodbye | -0.027 |
hug | -0.021 |
wave | -0.020 |
contain | -0.020 |
toward | -0.019 |
Token | Sam |
Token ID | 3409 |
Feature activation | +6.74e-03 |
Pos logit contributions | |
ery | +0.052 |
iflower | +0.048 |
chini | +0.045 |
Mic | +0.045 |
oon | +0.044 |
Neg logit contributions | |
goodbye | -0.073 |
contain | -0.062 |
umb | -0.060 |
tour | -0.059 |
plunge | -0.059 |
Token | made |
Token ID | 925 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
Mic | +0.059 |
ery | +0.059 |
mice | +0.056 |
Max | +0.056 |
iflower | +0.053 |
Neg logit contributions | |
goodbye | -0.072 |
contain | -0.063 |
plunge | -0.061 |
attention | -0.058 |
toward | -0.058 |
Token | him |
Token ID | 683 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
Mic | +0.041 |
Max | +0.039 |
ery | +0.039 |
iflower | +0.037 |
Bear | +0.036 |
Neg logit contributions | |
goodbye | -0.040 |
contain | -0.033 |
plunge | -0.033 |
attention | -0.032 |
umb | -0.031 |
Token | very |
Token ID | 845 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
ery | +0.046 |
mice | +0.042 |
iflower | +0.040 |
Max | +0.040 |
Mic | +0.040 |
Neg logit contributions | |
goodbye | -0.050 |
plunge | -0.042 |
oured | -0.042 |
umb | -0.041 |
contain | -0.041 |
Token | happy |
Token ID | 3772 |
Feature activation | +6.35e-02 |
Pos logit contributions | |
goodbye | +0.031 |
oured | +0.029 |
umb | +0.027 |
plunge | +0.026 |
contain | +0.026 |
Neg logit contributions | |
ery | -0.019 |
mice | -0.017 |
iflower | -0.017 |
Fox | -0.017 |
chini | -0.016 |
Token | - |
Token ID | 532 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
Mic | +0.121 |
ery | +0.118 |
Max | +0.117 |
mice | +0.111 |
Ted | +0.110 |
Neg logit contributions | |
goodbye | -0.108 |
contain | -0.090 |
plunge | -0.086 |
attention | -0.084 |
toward | -0.084 |
Token | clay |
Token ID | 21558 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
ery | +0.029 |
mice | +0.028 |
Max | +0.028 |
Mic | +0.028 |
Ted | +0.026 |
Neg logit contributions | |
goodbye | -0.049 |
contain | -0.044 |
plunge | -0.042 |
fy | -0.042 |
oured | -0.041 |
Token | ! |
Token ID | 0 |
Feature activation | -7.77e-03 |
Pos logit contributions | |
Mic | +0.219 |
Max | +0.214 |
ery | +0.212 |
Bear | +0.203 |
Charlie | +0.202 |
Neg logit contributions | |
goodbye | -0.136 |
contain | -0.110 |
plunge | -0.110 |
umb | -0.105 |
toward | -0.105 |
Token | Not |
Token ID | 1892 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
goodbye | +0.017 |
contain | +0.015 |
plunge | +0.015 |
umb | +0.014 |
toward | +0.014 |
Neg logit contributions | |
ery | -0.011 |
Mic | -0.010 |
Max | -0.010 |
mice | -0.010 |
iflower | -0.009 |
Token | just |
Token ID | 655 |
Feature activation | -3.59e-02 |
Pos logit contributions | |
Mic | +0.031 |
ery | +0.031 |
Max | +0.031 |
Doctor | +0.029 |
iflower | +0.028 |
Neg logit contributions | |
goodbye | -0.034 |
contain | -0.029 |
plunge | -0.028 |
attention | -0.027 |
umb | -0.026 |
Token | any |
Token ID | 597 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
goodbye | +0.068 |
fy | +0.061 |
umb | +0.060 |
plunge | +0.059 |
contain | +0.059 |
Neg logit contributions | |
ery | -0.052 |
mice | -0.050 |
Next | -0.047 |
Max | -0.046 |
Ted | -0.046 |
Token | clay |
Token ID | 21558 |
Feature activation | +7.52e-02 |
Pos logit contributions | |
Mic | +0.033 |
Max | +0.033 |
ery | +0.032 |
Bear | +0.030 |
iflower | +0.030 |
Neg logit contributions | |
goodbye | -0.047 |
contain | -0.041 |
attention | -0.040 |
plunge | -0.040 |
toward | -0.039 |
Token | a |
Token ID | 257 |
Feature activation | -9.97e-03 |
Pos logit contributions | |
Mic | +0.115 |
ery | +0.111 |
iflower | +0.106 |
mice | +0.105 |
ters | +0.104 |
Neg logit contributions | |
goodbye | -0.137 |
contain | -0.115 |
plunge | -0.108 |
closer | -0.105 |
attention | -0.104 |
Token | time |
Token ID | 640 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
goodbye | +0.022 |
contain | +0.019 |
plunge | +0.018 |
fy | +0.018 |
umb | +0.018 |
Neg logit contributions | |
ery | -0.014 |
Mic | -0.013 |
mice | -0.013 |
Max | -0.013 |
iflower | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | -6.78e-03 |
Pos logit contributions | |
goodbye | +0.019 |
contain | +0.015 |
plunge | +0.014 |
umb | +0.014 |
attention | +0.014 |
Neg logit contributions | |
ery | -0.031 |
Mic | -0.030 |
Max | -0.030 |
mice | -0.029 |
iflower | -0.029 |
Token | there |
Token ID | 612 |
Feature activation | +1.05e-03 |
Pos logit contributions | |
goodbye | +0.012 |
closer | +0.010 |
plunge | +0.010 |
contain | +0.010 |
toward | +0.009 |
Neg logit contributions | |
Mic | -0.012 |
iflower | -0.011 |
mice | -0.011 |
Doctor | -0.011 |
chini | -0.011 |
Token | were |
Token ID | 547 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
Mic | +0.002 |
mice | +0.002 |
ery | +0.002 |
iflower | +0.002 |
Fox | +0.002 |
Neg logit contributions | |
goodbye | -0.002 |
plunge | -0.001 |
contain | -0.001 |
inspection | -0.001 |
toward | -0.001 |
Token | twin |
Token ID | 15203 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
ery | +0.068 |
Mic | +0.064 |
Max | +0.063 |
iflower | +0.063 |
iders | +0.062 |
Neg logit contributions | |
goodbye | -0.108 |
umb | -0.093 |
plunge | -0.093 |
contain | -0.089 |
tickets | -0.087 |
Token | cats |
Token ID | 11875 |
Feature activation | +6.42e-02 |
Pos logit contributions | |
ery | +0.167 |
Mic | +0.161 |
Next | +0.157 |
iflower | +0.154 |
chini | +0.154 |
Neg logit contributions | |
goodbye | -0.177 |
hug | -0.157 |
umb | -0.148 |
wave | -0.144 |
ogging | -0.144 |
Token | named |
Token ID | 3706 |
Feature activation | +2.77e-03 |
Pos logit contributions | |
ery | +0.132 |
Mic | +0.128 |
chini | +0.127 |
iflower | +0.125 |
Next | +0.122 |
Neg logit contributions | |
goodbye | -0.097 |
contain | -0.074 |
hug | -0.072 |
umb | -0.070 |
toward | -0.070 |
Token | Tom |
Token ID | 4186 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
ery | +0.004 |
iflower | +0.004 |
oon | +0.003 |
chini | +0.003 |
Mic | +0.003 |
Neg logit contributions | |
goodbye | -0.006 |
umb | -0.005 |
tour | -0.005 |
plunge | -0.005 |
tickets | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
chini | +0.061 |
Mic | +0.061 |
ery | +0.059 |
iders | +0.058 |
ceed | +0.058 |
Neg logit contributions | |
goodbye | -0.031 |
hug | -0.024 |
wave | -0.021 |
ogging | -0.019 |
closer | -0.019 |
Token | Tim |
Token ID | 5045 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
ery | +0.049 |
iflower | +0.045 |
Mic | +0.043 |
chini | +0.042 |
oon | +0.042 |
Neg logit contributions | |
goodbye | -0.076 |
contain | -0.064 |
umb | -0.062 |
hug | -0.062 |
tour | -0.061 |
Token | , |
Token ID | 11 |
Feature activation | +3.44e-03 |
Pos logit contributions | |
Mic | +0.047 |
ery | +0.044 |
mice | +0.042 |
Max | +0.042 |
Bear | +0.041 |
Neg logit contributions | |
goodbye | -0.046 |
plunge | -0.039 |
tour | -0.038 |
contain | -0.038 |
tickets | -0.036 |
Token | there |
Token ID | 612 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
Mic | +0.005 |
ery | +0.005 |
Max | +0.005 |
iflower | +0.005 |
mice | +0.005 |
Neg logit contributions | |
goodbye | -0.007 |
contain | -0.006 |
plunge | -0.006 |
umb | -0.006 |
attention | -0.006 |
Token | were |
Token ID | 547 |
Feature activation | +6.26e-02 |
Pos logit contributions | |
Mic | +0.030 |
mice | +0.025 |
Fox | +0.025 |
Midnight | +0.025 |
iflower | +0.024 |
Neg logit contributions | |
goodbye | -0.022 |
plunge | -0.018 |
contain | -0.018 |
tour | -0.018 |
look | -0.017 |
Token | all |
Token ID | 477 |
Feature activation | -3.09e-03 |
Pos logit contributions | |
ery | +0.093 |
Mic | +0.091 |
Max | +0.090 |
mice | +0.089 |
Ted | +0.084 |
Neg logit contributions | |
goodbye | -0.127 |
contain | -0.111 |
plunge | -0.107 |
toward | -0.104 |
umb | -0.104 |
Token | sorts |
Token ID | 10524 |
Feature activation | +6.49e-02 |
Pos logit contributions | |
goodbye | +0.005 |
contain | +0.004 |
plunge | +0.004 |
umb | +0.004 |
attention | +0.004 |
Neg logit contributions | |
ery | -0.006 |
Mic | -0.006 |
Max | -0.006 |
mice | -0.006 |
iflower | -0.005 |
Token | of |
Token ID | 286 |
Feature activation | +9.93e-02 |
Pos logit contributions | |
ery | +0.102 |
mice | +0.101 |
hes | +0.091 |
ters | +0.091 |
chini | +0.091 |
Neg logit contributions | |
goodbye | -0.122 |
plunge | -0.108 |
contain | -0.105 |
icol | -0.103 |
oured | -0.103 |
Token | fun |
Token ID | 1257 |
Feature activation | +5.22e-02 |
Pos logit contributions | |
mice | +0.119 |
ery | +0.119 |
Mic | +0.110 |
Max | +0.107 |
iflower | +0.103 |
Neg logit contributions | |
goodbye | -0.234 |
contain | -0.206 |
plunge | -0.202 |
oured | -0.198 |
toward | -0.195 |
Token | things |
Token ID | 1243 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
ery | +0.057 |
mice | +0.056 |
Max | +0.052 |
iflower | +0.049 |
Mic | +0.049 |
Neg logit contributions | |
goodbye | -0.128 |
contain | -0.114 |
plunge | -0.111 |
oured | -0.110 |
fy | -0.107 |
Token | to |
Token ID | 284 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
ery | +0.070 |
mice | +0.067 |
Max | +0.067 |
Midnight | +0.064 |
Mic | +0.064 |
Neg logit contributions | |
goodbye | -0.063 |
contain | -0.054 |
plunge | -0.053 |
oured | -0.050 |
eal | -0.050 |
Token | do |
Token ID | 466 |
Feature activation | +4.63e-02 |
Pos logit contributions | |
ery | +0.031 |
Max | +0.030 |
mice | +0.029 |
Ted | +0.028 |
chini | +0.028 |
Neg logit contributions | |
goodbye | -0.041 |
oured | -0.037 |
fy | -0.037 |
ounded | -0.035 |
umb | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -3.88e-03 |
Pos logit contributions | |
ery | +0.074 |
Max | +0.070 |
Mic | +0.069 |
mice | +0.069 |
iflower | +0.066 |
Neg logit contributions | |
goodbye | -0.090 |
contain | -0.077 |
plunge | -0.076 |
umb | -0.074 |
oured | -0.073 |
Token | any |
Token ID | 597 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
goodbye | +0.068 |
fy | +0.061 |
umb | +0.060 |
plunge | +0.059 |
contain | +0.059 |
Neg logit contributions | |
ery | -0.052 |
mice | -0.050 |
Next | -0.047 |
Max | -0.046 |
Ted | -0.046 |
Token | clay |
Token ID | 21558 |
Feature activation | +7.52e-02 |
Pos logit contributions | |
Mic | +0.033 |
Max | +0.033 |
ery | +0.032 |
Bear | +0.030 |
iflower | +0.030 |
Neg logit contributions | |
goodbye | -0.047 |
contain | -0.041 |
attention | -0.040 |
plunge | -0.040 |
toward | -0.039 |
Token | , |
Token ID | 11 |
Feature activation | +1.72e-03 |
Pos logit contributions | |
Mic | +0.150 |
Max | +0.142 |
ery | +0.138 |
Doctor | +0.135 |
Bear | +0.134 |
Neg logit contributions | |
goodbye | -0.121 |
contain | -0.102 |
plunge | -0.101 |
toward | -0.097 |
attention | -0.096 |
Token | but |
Token ID | 475 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
ery | +0.003 |
Mic | +0.003 |
mice | +0.002 |
Max | +0.002 |
iflower | +0.002 |
Neg logit contributions | |
goodbye | -0.003 |
contain | -0.003 |
umb | -0.003 |
plunge | -0.003 |
fy | -0.003 |
Token | special |
Token ID | 2041 |
Feature activation | +5.11e-02 |
Pos logit contributions | |
goodbye | +0.026 |
contain | +0.022 |
plunge | +0.021 |
attention | +0.020 |
toward | +0.020 |
Neg logit contributions | |
ery | -0.022 |
Mic | -0.022 |
Max | -0.021 |
mice | -0.021 |
iflower | -0.020 |
Token | clay |
Token ID | 21558 |
Feature activation | +9.91e-02 |
Pos logit contributions | |
Mic | +0.089 |
Max | +0.084 |
ery | +0.082 |
Bear | +0.079 |
Doctor | +0.078 |
Neg logit contributions | |
goodbye | -0.096 |
contain | -0.081 |
plunge | -0.080 |
toward | -0.078 |
attention | -0.078 |
Token | that |
Token ID | 326 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
Mic | +0.187 |
Max | +0.177 |
ery | +0.169 |
Charlie | +0.166 |
Doctor | +0.165 |
Neg logit contributions | |
goodbye | -0.169 |
contain | -0.144 |
plunge | -0.142 |
attention | -0.137 |
toward | -0.137 |
Token | could |
Token ID | 714 |
Feature activation | +4.22e-03 |
Pos logit contributions | |
Mic | +0.073 |
ery | +0.069 |
Max | +0.067 |
Brown | +0.065 |
elvet | +0.064 |
Neg logit contributions | |
goodbye | -0.054 |
wave | -0.046 |
plunge | -0.046 |
contain | -0.045 |
tour | -0.043 |
Token | be |
Token ID | 307 |
Feature activation | +4.15e-02 |
Pos logit contributions | |
Mic | +0.007 |
ery | +0.007 |
Max | +0.007 |
mice | +0.007 |
iflower | +0.007 |
Neg logit contributions | |
goodbye | -0.008 |
contain | -0.007 |
plunge | -0.006 |
umb | -0.006 |
attention | -0.006 |
Token | used |
Token ID | 973 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
Mic | +0.070 |
Max | +0.069 |
ery | +0.065 |
Ted | +0.064 |
Next | +0.064 |
Neg logit contributions | |
goodbye | -0.076 |
contain | -0.065 |
umb | -0.064 |
attention | -0.064 |
plunge | -0.062 |
Token | to |
Token ID | 284 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
ery | +0.027 |
Mic | +0.026 |
Max | +0.026 |
mice | +0.026 |
iflower | +0.025 |
Neg logit contributions | |
goodbye | -0.018 |
contain | -0.015 |
plunge | -0.014 |
umb | -0.014 |
attention | -0.013 |
Token | was |
Token ID | 373 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
Mic | +0.067 |
ery | +0.064 |
mice | +0.061 |
Max | +0.060 |
iflower | +0.059 |
Neg logit contributions | |
goodbye | -0.091 |
contain | -0.078 |
plunge | -0.076 |
umb | -0.076 |
toward | -0.075 |
Token | very |
Token ID | 845 |
Feature activation | +2.42e-02 |
Pos logit contributions | |
Mic | +0.071 |
ery | +0.071 |
Max | +0.069 |
mice | +0.067 |
iflower | +0.067 |
Neg logit contributions | |
goodbye | -0.067 |
contain | -0.056 |
plunge | -0.054 |
umb | -0.053 |
attention | -0.052 |
Token | important |
Token ID | 1593 |
Feature activation | +6.84e-02 |
Pos logit contributions | |
ery | +0.036 |
mice | +0.034 |
Next | +0.034 |
Max | +0.033 |
iflower | +0.033 |
Neg logit contributions | |
goodbye | -0.049 |
oured | -0.043 |
contain | -0.043 |
plunge | -0.042 |
umb | -0.040 |
Token | because |
Token ID | 780 |
Feature activation | -5.37e-03 |
Pos logit contributions | |
mice | +0.110 |
ery | +0.109 |
Max | +0.102 |
Mic | +0.101 |
iflower | +0.101 |
Neg logit contributions | |
goodbye | -0.127 |
contain | -0.114 |
plunge | -0.110 |
oured | -0.108 |
eal | -0.106 |
Token | lots |
Token ID | 6041 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
goodbye | +0.011 |
contain | +0.009 |
plunge | +0.009 |
umb | +0.009 |
attention | +0.009 |
Neg logit contributions | |
Mic | -0.008 |
ery | -0.008 |
Max | -0.008 |
mice | -0.008 |
Ted | -0.008 |
Token | of |
Token ID | 286 |
Feature activation | +9.87e-02 |
Pos logit contributions | |
ery | +0.043 |
Mic | +0.042 |
Max | +0.041 |
mice | +0.041 |
iflower | +0.040 |
Neg logit contributions | |
goodbye | -0.038 |
contain | -0.032 |
plunge | -0.031 |
umb | -0.029 |
oured | -0.029 |
Token | people |
Token ID | 661 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
ery | +0.151 |
Mic | +0.146 |
Max | +0.144 |
mice | +0.143 |
iflower | +0.136 |
Neg logit contributions | |
goodbye | -0.199 |
contain | -0.173 |
plunge | -0.166 |
oured | -0.161 |
umb | -0.161 |
Token | came |
Token ID | 1625 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
Mic | +0.065 |
ery | +0.064 |
Max | +0.062 |
Ted | +0.060 |
mice | +0.059 |
Neg logit contributions | |
goodbye | -0.075 |
contain | -0.065 |
plunge | -0.062 |
umb | -0.061 |
attention | -0.061 |
Token | there |
Token ID | 612 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
goodbye | +0.033 |
contain | +0.027 |
plunge | +0.025 |
oured | +0.025 |
fy | +0.025 |
Neg logit contributions | |
ery | -0.051 |
mice | -0.050 |
Max | -0.049 |
Mic | -0.048 |
iflower | -0.047 |
Token | to |
Token ID | 284 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
goodbye | +0.020 |
contain | +0.016 |
plunge | +0.016 |
toward | +0.015 |
attention | +0.015 |
Neg logit contributions | |
Mic | -0.026 |
ery | -0.026 |
Max | -0.025 |
mice | -0.024 |
iflower | -0.024 |
Token | roll |
Token ID | 4836 |
Feature activation | -3.24e-03 |
Pos logit contributions | |
ery | +0.030 |
Mic | +0.029 |
Max | +0.029 |
mice | +0.028 |
iflower | +0.028 |
Neg logit contributions | |
goodbye | -0.022 |
contain | -0.018 |
plunge | -0.017 |
umb | -0.017 |
attention | -0.016 |
Token | food |
Token ID | 2057 |
Feature activation | +3.79e-02 |
Pos logit contributions | |
ery | +0.037 |
Mic | +0.037 |
Max | +0.036 |
mice | +0.035 |
iflower | +0.034 |
Neg logit contributions | |
goodbye | -0.042 |
contain | -0.036 |
plunge | -0.035 |
umb | -0.034 |
attention | -0.034 |
Token | , |
Token ID | 11 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
Mic | +0.079 |
ery | +0.077 |
Max | +0.077 |
Ted | +0.074 |
Bear | +0.074 |
Neg logit contributions | |
goodbye | -0.056 |
contain | -0.046 |
plunge | -0.044 |
attention | -0.042 |
umb | -0.042 |
Token | it |
Token ID | 340 |
Feature activation | +4.95e-02 |
Pos logit contributions | |
Mic | +0.037 |
ery | +0.036 |
Max | +0.035 |
iflower | +0.034 |
Bear | +0.034 |
Neg logit contributions | |
goodbye | -0.029 |
contain | -0.024 |
umb | -0.022 |
plunge | -0.022 |
attention | -0.022 |
Token | was |
Token ID | 373 |
Feature activation | +3.85e-02 |
Pos logit contributions | |
Mic | +0.100 |
Max | +0.090 |
ery | +0.088 |
Charlie | +0.088 |
elvet | +0.088 |
Neg logit contributions | |
goodbye | -0.072 |
wave | -0.063 |
closer | -0.061 |
toward | -0.060 |
tour | -0.058 |
Token | something |
Token ID | 1223 |
Feature activation | +3.92e-02 |
Pos logit contributions | |
Mic | +0.070 |
Max | +0.069 |
ery | +0.063 |
Bear | +0.063 |
Mouse | +0.062 |
Neg logit contributions | |
goodbye | -0.070 |
attention | -0.057 |
contain | -0.056 |
closer | -0.055 |
umb | -0.055 |
Token | else |
Token ID | 2073 |
Feature activation | +9.81e-02 |
Pos logit contributions | |
ery | +0.067 |
Mic | +0.065 |
Max | +0.064 |
mice | +0.063 |
iflower | +0.061 |
Neg logit contributions | |
goodbye | -0.072 |
contain | -0.062 |
plunge | -0.059 |
umb | -0.058 |
attention | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | -7.50e-03 |
Pos logit contributions | |
ery | +0.206 |
Mic | +0.203 |
Max | +0.200 |
mice | +0.196 |
iflower | +0.191 |
Neg logit contributions | |
goodbye | -0.141 |
contain | -0.115 |
plunge | -0.110 |
umb | -0.105 |
attention | -0.104 |
Token | So |
Token ID | 1406 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
goodbye | +0.012 |
toward | +0.009 |
contain | +0.009 |
hi | +0.009 |
oured | +0.009 |
Neg logit contributions | |
ery | -0.016 |
iflower | -0.015 |
web | -0.014 |
chini | -0.014 |
ters | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | +5.94e-03 |
Pos logit contributions | |
goodbye | +0.021 |
contain | +0.017 |
plunge | +0.016 |
umb | +0.016 |
attention | +0.016 |
Neg logit contributions | |
ery | -0.029 |
Mic | -0.029 |
Max | -0.028 |
mice | -0.028 |
iflower | -0.027 |
Token | he |
Token ID | 339 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
ery | +0.014 |
Mic | +0.013 |
chini | +0.012 |
Max | +0.012 |
iflower | +0.012 |
Neg logit contributions | |
goodbye | -0.009 |
contain | -0.007 |
plunge | -0.006 |
toward | -0.006 |
wave | -0.006 |
Token | followed |
Token ID | 3940 |
Feature activation | -4.72e-02 |
Pos logit contributions | |
ery | +0.027 |
Mic | +0.027 |
Max | +0.026 |
iflower | +0.025 |
mice | +0.025 |
Neg logit contributions | |
goodbye | -0.027 |
contain | -0.023 |
plunge | -0.022 |
umb | -0.021 |
toward | -0.021 |
Token | a |
Token ID | 257 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
Mic | +0.109 |
ery | +0.106 |
iflower | +0.100 |
mice | +0.100 |
ters | +0.098 |
Neg logit contributions | |
goodbye | -0.133 |
contain | -0.112 |
plunge | -0.105 |
attention | -0.102 |
farewell | -0.101 |
Token | time |
Token ID | 640 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
goodbye | +0.025 |
contain | +0.022 |
plunge | +0.021 |
umb | +0.020 |
fy | +0.020 |
Neg logit contributions | |
ery | -0.015 |
Mic | -0.015 |
mice | -0.014 |
Max | -0.014 |
iflower | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | -9.32e-03 |
Pos logit contributions | |
goodbye | +0.022 |
contain | +0.019 |
plunge | +0.018 |
umb | +0.017 |
attention | +0.017 |
Neg logit contributions | |
ery | -0.031 |
Max | -0.030 |
Mic | -0.030 |
mice | -0.029 |
Ted | -0.029 |
Token | there |
Token ID | 612 |
Feature activation | -6.60e-04 |
Pos logit contributions | |
goodbye | +0.017 |
closer | +0.015 |
contain | +0.014 |
plunge | +0.014 |
toward | +0.014 |
Neg logit contributions | |
Mic | -0.016 |
mice | -0.015 |
ery | -0.014 |
iflower | -0.014 |
Doctor | -0.014 |
Token | were |
Token ID | 547 |
Feature activation | +4.86e-02 |
Pos logit contributions | |
goodbye | +0.001 |
plunge | +0.001 |
contain | +0.001 |
inspection | +0.001 |
toward | +0.001 |
Neg logit contributions | |
Mic | -0.001 |
ery | -0.001 |
mice | -0.001 |
ters | -0.001 |
iflower | -0.001 |
Token | twin |
Token ID | 15203 |
Feature activation | +9.78e-02 |
Pos logit contributions | |
ery | +0.068 |
Mic | +0.063 |
iflower | +0.063 |
Max | +0.062 |
iders | +0.061 |
Neg logit contributions | |
goodbye | -0.105 |
umb | -0.091 |
plunge | -0.089 |
contain | -0.087 |
tickets | -0.084 |
Token | sisters |
Token ID | 15153 |
Feature activation | +4.93e-02 |
Pos logit contributions | |
ery | +0.166 |
Mic | +0.162 |
Next | +0.155 |
iflower | +0.154 |
chini | +0.153 |
Neg logit contributions | |
goodbye | -0.173 |
hug | -0.150 |
umb | -0.144 |
wave | -0.140 |
ogging | -0.139 |
Token | , |
Token ID | 11 |
Feature activation | +8.26e-03 |
Pos logit contributions | |
ery | +0.092 |
Mic | +0.089 |
Max | +0.085 |
chini | +0.085 |
iflower | +0.085 |
Neg logit contributions | |
goodbye | -0.086 |
contain | -0.070 |
umb | -0.067 |
toward | -0.065 |
plunge | -0.064 |
Token | named |
Token ID | 3706 |
Feature activation | +1.75e-03 |
Pos logit contributions | |
ery | +0.013 |
iflower | +0.013 |
chini | +0.012 |
Mic | +0.012 |
iders | +0.012 |
Neg logit contributions | |
goodbye | -0.017 |
hug | -0.015 |
contain | -0.014 |
umb | -0.014 |
wave | -0.013 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
ery | +0.002 |
iflower | +0.002 |
Mic | +0.002 |
oon | +0.002 |
chini | +0.002 |
Neg logit contributions | |
goodbye | -0.004 |
umb | -0.003 |
contain | -0.003 |
plunge | -0.003 |
tour | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +3.29e-02 |
Pos logit contributions | |
ery | +0.067 |
chini | +0.065 |
Mic | +0.065 |
iflower | +0.063 |
mice | +0.062 |
Neg logit contributions | |
goodbye | -0.046 |
hug | -0.035 |
contain | -0.033 |
wave | -0.033 |
toward | -0.033 |
Token | " |
Token ID | 366 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
goodbye | +0.014 |
contain | +0.011 |
toward | +0.011 |
plunge | +0.011 |
farewell | +0.011 |
Neg logit contributions | |
ery | -0.010 |
Mic | -0.009 |
mice | -0.009 |
Max | -0.008 |
iflower | -0.008 |
Token | I |
Token ID | 40 |
Feature activation | -4.63e-02 |
Pos logit contributions | |
ery | +0.043 |
mice | +0.042 |
iders | +0.037 |
iflower | +0.035 |
Midnight | +0.035 |
Neg logit contributions | |
goodbye | -0.074 |
contain | -0.066 |
fy | -0.065 |
des | -0.065 |
toward | -0.064 |
Token | love |
Token ID | 1842 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
goodbye | +0.084 |
contain | +0.072 |
umb | +0.070 |
plunge | +0.070 |
attention | +0.067 |
Neg logit contributions | |
ery | -0.084 |
Max | -0.078 |
Mic | -0.077 |
mice | -0.074 |
Bear | -0.073 |
Token | you |
Token ID | 345 |
Feature activation | -7.15e-03 |
Pos logit contributions | |
ery | +0.041 |
Max | +0.041 |
Ted | +0.038 |
oyd | +0.038 |
Mic | +0.038 |
Neg logit contributions | |
goodbye | -0.049 |
contain | -0.040 |
attention | -0.040 |
plunge | -0.040 |
umb | -0.039 |
Token | , |
Token ID | 11 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
goodbye | +0.010 |
contain | +0.008 |
plunge | +0.008 |
attention | +0.007 |
umb | +0.007 |
Neg logit contributions | |
ery | -0.015 |
Mic | -0.015 |
Max | -0.015 |
mice | -0.015 |
Ted | -0.015 |
Token | avocado |
Token ID | 40377 |
Feature activation | +9.72e-02 |
Pos logit contributions | |
ery | +0.030 |
Mic | +0.028 |
Max | +0.027 |
Bear | +0.026 |
Ted | +0.026 |
Neg logit contributions | |
goodbye | -0.043 |
contain | -0.036 |
umb | -0.035 |
attention | -0.034 |
plunge | -0.034 |
Token | !" |
Token ID | 2474 |
Feature activation | +8.61e-03 |
Pos logit contributions | |
Mic | +0.217 |
Max | +0.214 |
ery | +0.210 |
Ted | +0.204 |
Next | +0.202 |
Neg logit contributions | |
goodbye | -0.129 |
contain | -0.099 |
plunge | -0.093 |
fy | -0.092 |
attention | -0.091 |
Token |
|
Token ID | 198 |
Feature activation | +8.20e-03 |
Pos logit contributions | |
ery | +0.017 |
mice | +0.015 |
Mic | +0.015 |
iflower | +0.015 |
chini | +0.014 |
Neg logit contributions | |
goodbye | -0.017 |
hug | -0.012 |
farewell | -0.012 |
contain | -0.012 |
ogging | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | +5.54e-03 |
Pos logit contributions | |
ery | +0.013 |
mice | +0.012 |
ters | +0.011 |
Mic | +0.011 |
ities | +0.011 |
Neg logit contributions | |
goodbye | -0.017 |
contain | -0.014 |
ogging | -0.014 |
farewell | -0.013 |
toward | -0.013 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | +3.05e-03 |
Pos logit contributions | |
ery | +0.009 |
mice | +0.009 |
ters | +0.008 |
iflower | +0.008 |
magnets | +0.008 |
Neg logit contributions | |
goodbye | -0.011 |
contain | -0.009 |
farewell | -0.009 |
ogging | -0.009 |
toward | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +2.42e-03 |
Pos logit contributions | |
ery | +0.006 |
Max | +0.006 |
Mic | +0.006 |
mice | +0.006 |
Midnight | +0.006 |
Neg logit contributions | |
goodbye | -0.004 |
contain | -0.004 |
plunge | -0.004 |
umb | -0.003 |
toward | -0.003 |
Token | broken |
Token ID | 5445 |
Feature activation | +5.32e-02 |
Pos logit contributions | |
ery | +0.080 |
Mic | +0.080 |
Max | +0.078 |
mice | +0.075 |
iflower | +0.074 |
Neg logit contributions | |
goodbye | -0.092 |
contain | -0.079 |
umb | -0.076 |
plunge | -0.076 |
attention | -0.074 |
Token | , |
Token ID | 11 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
Mic | +0.100 |
ery | +0.098 |
Max | +0.097 |
mice | +0.093 |
Ted | +0.093 |
Neg logit contributions | |
goodbye | -0.088 |
contain | -0.073 |
plunge | -0.071 |
umb | -0.070 |
toward | -0.069 |
Token | and |
Token ID | 290 |
Feature activation | +4.27e-02 |
Pos logit contributions | |
Mic | +0.026 |
ery | +0.026 |
Max | +0.025 |
mice | +0.024 |
iflower | +0.024 |
Neg logit contributions | |
goodbye | -0.024 |
contain | -0.020 |
umb | -0.019 |
plunge | -0.019 |
attention | -0.019 |
Token | there |
Token ID | 612 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
ery | +0.075 |
Mic | +0.074 |
iflower | +0.068 |
Max | +0.068 |
chini | +0.067 |
Neg logit contributions | |
goodbye | -0.079 |
contain | -0.065 |
umb | -0.064 |
plunge | -0.064 |
attention | -0.063 |
Token | were |
Token ID | 547 |
Feature activation | +6.32e-02 |
Pos logit contributions | |
Mic | +0.026 |
ery | +0.023 |
iflower | +0.022 |
Max | +0.022 |
mice | +0.022 |
Neg logit contributions | |
goodbye | -0.027 |
plunge | -0.023 |
contain | -0.023 |
umb | -0.022 |
wave | -0.021 |
Token | cob |
Token ID | 22843 |
Feature activation | +9.70e-02 |
Pos logit contributions | |
ery | +0.098 |
Mic | +0.098 |
Max | +0.094 |
iflower | +0.090 |
mice | +0.089 |
Neg logit contributions | |
goodbye | -0.126 |
contain | -0.108 |
umb | -0.106 |
plunge | -0.105 |
attention | -0.102 |
Token | we |
Token ID | 732 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
Mic | +0.103 |
Max | +0.092 |
ery | +0.086 |
Midnight | +0.083 |
Charlie | +0.082 |
Neg logit contributions | |
goodbye | -0.240 |
umb | -0.219 |
plunge | -0.216 |
toward | -0.214 |
contain | -0.213 |
Token | bs |
Token ID | 1443 |
Feature activation | +4.20e-02 |
Pos logit contributions | |
goodbye | +0.016 |
contain | +0.013 |
plunge | +0.012 |
umb | +0.012 |
attention | +0.012 |
Neg logit contributions | |
ery | -0.021 |
Mic | -0.020 |
Max | -0.020 |
mice | -0.019 |
Ted | -0.019 |
Token | everywhere |
Token ID | 8347 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
Mic | +0.081 |
ery | +0.078 |
Max | +0.077 |
iflower | +0.074 |
mice | +0.074 |
Neg logit contributions | |
goodbye | -0.069 |
contain | -0.057 |
plunge | -0.055 |
toward | -0.054 |
umb | -0.054 |
Token | . |
Token ID | 13 |
Feature activation | -3.77e-03 |
Pos logit contributions | |
Mic | +0.067 |
ery | +0.061 |
Max | +0.060 |
Fox | +0.060 |
Doctor | +0.059 |
Neg logit contributions | |
goodbye | -0.039 |
contain | -0.030 |
toward | -0.029 |
attention | -0.029 |
plunge | -0.029 |
Token | Spike |
Token ID | 26309 |
Feature activation | -2.90e-03 |
Pos logit contributions | |
goodbye | +0.007 |
plunge | +0.006 |
contain | +0.006 |
toward | +0.006 |
farewell | +0.006 |
Neg logit contributions | |
ery | -0.007 |
iflower | -0.006 |
Mic | -0.006 |
chini | -0.006 |
mice | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | -7.85e-03 |
Pos logit contributions | |
Mic | +0.124 |
ery | +0.119 |
iflower | +0.114 |
mice | +0.113 |
ters | +0.112 |
Neg logit contributions | |
goodbye | -0.138 |
contain | -0.114 |
plunge | -0.107 |
closer | -0.104 |
farewell | -0.103 |
Token | time |
Token ID | 640 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
goodbye | +0.018 |
contain | +0.016 |
plunge | +0.015 |
umb | +0.015 |
attention | +0.015 |
Neg logit contributions | |
ery | -0.010 |
Mic | -0.010 |
Max | -0.009 |
mice | -0.009 |
iflower | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | -8.53e-03 |
Pos logit contributions | |
goodbye | +0.021 |
contain | +0.017 |
plunge | +0.016 |
umb | +0.016 |
attention | +0.016 |
Neg logit contributions | |
ery | -0.030 |
Mic | -0.029 |
Max | -0.029 |
mice | -0.029 |
Ted | -0.028 |
Token | there |
Token ID | 612 |
Feature activation | +1.69e-03 |
Pos logit contributions | |
goodbye | +0.014 |
closer | +0.012 |
contain | +0.011 |
plunge | +0.011 |
toward | +0.011 |
Neg logit contributions | |
Mic | -0.017 |
chini | -0.016 |
iflower | -0.015 |
Doctor | -0.015 |
mice | -0.015 |
Token | were |
Token ID | 547 |
Feature activation | +5.00e-02 |
Pos logit contributions | |
Mic | +0.004 |
mice | +0.003 |
ery | +0.003 |
ters | +0.003 |
iflower | +0.003 |
Neg logit contributions | |
goodbye | -0.003 |
plunge | -0.002 |
contain | -0.002 |
look | -0.002 |
hug | -0.002 |
Token | twin |
Token ID | 15203 |
Feature activation | +9.70e-02 |
Pos logit contributions | |
ery | +0.074 |
iders | +0.070 |
Max | +0.069 |
Mic | +0.068 |
iflower | +0.067 |
Neg logit contributions | |
goodbye | -0.106 |
umb | -0.090 |
plunge | -0.090 |
contain | -0.085 |
tickets | -0.084 |
Token | sisters |
Token ID | 15153 |
Feature activation | +5.07e-02 |
Pos logit contributions | |
Mic | +0.183 |
ery | +0.181 |
Max | +0.177 |
Next | +0.175 |
chini | +0.171 |
Neg logit contributions | |
goodbye | -0.154 |
hug | -0.128 |
umb | -0.123 |
wave | -0.120 |
tour | -0.119 |
Token | named |
Token ID | 3706 |
Feature activation | +1.77e-03 |
Pos logit contributions | |
ery | +0.109 |
Mic | +0.104 |
chini | +0.102 |
iflower | +0.100 |
Max | +0.099 |
Neg logit contributions | |
goodbye | -0.076 |
contain | -0.059 |
umb | -0.055 |
hug | -0.054 |
toward | -0.054 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
ery | +0.003 |
Mic | +0.002 |
iflower | +0.002 |
chini | +0.002 |
oon | +0.002 |
Neg logit contributions | |
goodbye | -0.004 |
umb | -0.003 |
contain | -0.003 |
tour | -0.003 |
plunge | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
ery | +0.072 |
chini | +0.071 |
Mic | +0.071 |
iflower | +0.068 |
Next | +0.067 |
Neg logit contributions | |
goodbye | -0.045 |
hug | -0.033 |
contain | -0.031 |
toward | -0.030 |
wave | -0.030 |
Token | Lucy |
Token ID | 22162 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
ery | +0.056 |
iflower | +0.050 |
chini | +0.050 |
oon | +0.049 |
Mic | +0.047 |
Neg logit contributions | |
goodbye | -0.072 |
contain | -0.058 |
umb | -0.058 |
hug | -0.057 |
wave | -0.057 |
Token | a |
Token ID | 257 |
Feature activation | -6.77e-03 |
Pos logit contributions | |
Mic | +0.098 |
ery | +0.097 |
iflower | +0.094 |
ters | +0.090 |
iders | +0.089 |
Neg logit contributions | |
goodbye | -0.108 |
contain | -0.089 |
plunge | -0.083 |
attention | -0.081 |
closer | -0.081 |
Token | time |
Token ID | 640 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
goodbye | +0.013 |
contain | +0.011 |
fy | +0.011 |
tour | +0.011 |
farewell | +0.011 |
Neg logit contributions | |
ery | -0.010 |
ters | -0.010 |
mice | -0.010 |
chini | -0.010 |
Mic | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
goodbye | +0.021 |
contain | +0.016 |
plunge | +0.016 |
umb | +0.015 |
attention | +0.015 |
Neg logit contributions | |
ery | -0.037 |
Mic | -0.036 |
Max | -0.035 |
mice | -0.035 |
iflower | -0.035 |
Token | there |
Token ID | 612 |
Feature activation | +5.74e-04 |
Pos logit contributions | |
goodbye | +0.018 |
closer | +0.017 |
ounded | +0.014 |
toward | +0.014 |
deeper | +0.013 |
Neg logit contributions | |
Mic | -0.024 |
mice | -0.023 |
Doctor | -0.023 |
chini | -0.022 |
iflower | -0.022 |
Token | were |
Token ID | 547 |
Feature activation | +4.85e-02 |
Pos logit contributions | |
Mic | +0.001 |
mice | +0.001 |
iflower | +0.001 |
ery | +0.001 |
ities | +0.001 |
Neg logit contributions | |
goodbye | -0.001 |
plunge | -0.001 |
inspection | -0.001 |
look | -0.001 |
contain | -0.001 |
Token | twin |
Token ID | 15203 |
Feature activation | +9.69e-02 |
Pos logit contributions | |
ery | +0.069 |
iders | +0.063 |
iflower | +0.063 |
Mic | +0.061 |
Max | +0.061 |
Neg logit contributions | |
goodbye | -0.105 |
umb | -0.091 |
plunge | -0.089 |
tickets | -0.085 |
contain | -0.085 |
Token | brothers |
Token ID | 9397 |
Feature activation | +2.81e-02 |
Pos logit contributions | |
ery | +0.158 |
Mic | +0.158 |
Next | +0.152 |
Max | +0.149 |
chini | +0.146 |
Neg logit contributions | |
goodbye | -0.177 |
hug | -0.151 |
umb | -0.147 |
wave | -0.140 |
ogging | -0.139 |
Token | who |
Token ID | 508 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
ery | +0.057 |
Mic | +0.056 |
chini | +0.054 |
iflower | +0.054 |
Max | +0.052 |
Neg logit contributions | |
goodbye | -0.044 |
contain | -0.033 |
hug | -0.033 |
umb | -0.033 |
wave | -0.031 |
Token | lived |
Token ID | 5615 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
Mic | +0.026 |
iflower | +0.026 |
ery | +0.025 |
iders | +0.025 |
mice | +0.024 |
Neg logit contributions | |
goodbye | -0.026 |
hug | -0.022 |
contain | -0.021 |
wave | -0.020 |
fy | -0.020 |
Token | in |
Token ID | 287 |
Feature activation | +7.84e-03 |
Pos logit contributions | |
iflower | +0.096 |
Mic | +0.094 |
ery | +0.094 |
chini | +0.093 |
iders | +0.091 |
Neg logit contributions | |
goodbye | -0.073 |
closer | -0.061 |
toward | -0.058 |
contain | -0.056 |
hug | -0.056 |
Token | the |
Token ID | 262 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
ery | +0.018 |
iders | +0.018 |
iflower | +0.017 |
oon | +0.016 |
Mic | +0.016 |
Neg logit contributions | |
goodbye | -0.010 |
hug | -0.008 |
look | -0.008 |
closer | -0.007 |
tickets | -0.007 |
Token | very |
Token ID | 845 |
Feature activation | +2.63e-02 |
Pos logit contributions | |
Mic | +0.064 |
ery | +0.063 |
Max | +0.061 |
mice | +0.059 |
Bear | +0.059 |
Neg logit contributions | |
goodbye | -0.075 |
contain | -0.063 |
plunge | -0.060 |
umb | -0.060 |
attention | -0.059 |
Token | clean |
Token ID | 3424 |
Feature activation | +4.12e-02 |
Pos logit contributions | |
ery | +0.042 |
Max | +0.040 |
Mic | +0.040 |
mice | +0.039 |
iflower | +0.039 |
Neg logit contributions | |
goodbye | -0.051 |
contain | -0.044 |
plunge | -0.043 |
umb | -0.042 |
oured | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
Mic | +0.076 |
ery | +0.075 |
Max | +0.073 |
mice | +0.071 |
Ted | +0.071 |
Neg logit contributions | |
goodbye | -0.071 |
contain | -0.059 |
plunge | -0.056 |
attention | -0.054 |
umb | -0.054 |
Token | showed |
Token ID | 3751 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
Mic | +0.065 |
ery | +0.062 |
Max | +0.060 |
mice | +0.059 |
Ted | +0.058 |
Neg logit contributions | |
goodbye | -0.059 |
contain | -0.048 |
umb | -0.046 |
attention | -0.046 |
plunge | -0.046 |
Token | lots |
Token ID | 6041 |
Feature activation | +6.84e-03 |
Pos logit contributions | |
ery | +0.023 |
Mic | +0.023 |
Max | +0.022 |
mice | +0.021 |
Ted | +0.021 |
Neg logit contributions | |
goodbye | -0.030 |
contain | -0.026 |
plunge | -0.025 |
umb | -0.025 |
attention | -0.025 |
Token | of |
Token ID | 286 |
Feature activation | +9.68e-02 |
Pos logit contributions | |
ery | +0.011 |
mice | +0.011 |
Max | +0.010 |
iflower | +0.010 |
chini | +0.010 |
Neg logit contributions | |
goodbye | -0.013 |
contain | -0.012 |
plunge | -0.011 |
oured | -0.011 |
icol | -0.011 |
Token | happy |
Token ID | 3772 |
Feature activation | +7.87e-02 |
Pos logit contributions | |
ery | +0.135 |
Mic | +0.135 |
Max | +0.131 |
mice | +0.123 |
Ted | +0.122 |
Neg logit contributions | |
goodbye | -0.207 |
contain | -0.179 |
plunge | -0.175 |
umb | -0.173 |
attention | -0.172 |
Token | things |
Token ID | 1243 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
ery | +0.100 |
Mic | +0.096 |
Max | +0.094 |
mice | +0.092 |
iflower | +0.088 |
Neg logit contributions | |
goodbye | -0.179 |
contain | -0.159 |
plunge | -0.154 |
umb | -0.151 |
attention | -0.149 |
Token | . |
Token ID | 13 |
Feature activation | -7.14e-03 |
Pos logit contributions | |
ery | +0.031 |
Max | +0.029 |
mice | +0.029 |
iflower | +0.028 |
Midnight | +0.028 |
Neg logit contributions | |
goodbye | -0.036 |
contain | -0.032 |
plunge | -0.032 |
umb | -0.031 |
oured | -0.031 |
Token | One |
Token ID | 1881 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
goodbye | +0.011 |
contain | +0.008 |
hi | +0.008 |
hello | +0.008 |
farewell | +0.008 |
Neg logit contributions | |
ery | -0.016 |
iflower | -0.015 |
Mic | -0.014 |
web | -0.014 |
ascal | -0.014 |
Token | day |
Token ID | 1110 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
goodbye | +0.018 |
contain | +0.015 |
tour | +0.014 |
fy | +0.014 |
attention | +0.014 |
Neg logit contributions | |
Mic | -0.021 |
ery | -0.020 |
Max | -0.019 |
mice | -0.019 |
ters | -0.019 |
Token | to |
Token ID | 284 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
ery | +0.086 |
Mic | +0.082 |
Max | +0.082 |
Ted | +0.080 |
Bear | +0.079 |
Neg logit contributions | |
goodbye | -0.037 |
umb | -0.026 |
contain | -0.025 |
plunge | -0.024 |
attention | -0.024 |
Token | eat |
Token ID | 4483 |
Feature activation | +3.92e-02 |
Pos logit contributions | |
ery | +0.023 |
Mic | +0.023 |
Max | +0.022 |
mice | +0.022 |
Midnight | +0.022 |
Neg logit contributions | |
goodbye | -0.022 |
contain | -0.019 |
plunge | -0.018 |
toward | -0.018 |
oured | -0.018 |
Token | jelly |
Token ID | 27644 |
Feature activation | +3.67e-02 |
Pos logit contributions | |
Max | +0.065 |
ery | +0.065 |
Mic | +0.064 |
Ted | +0.062 |
oyd | +0.060 |
Neg logit contributions | |
goodbye | -0.074 |
umb | -0.061 |
contain | -0.060 |
plunge | -0.059 |
attention | -0.059 |
Token | , |
Token ID | 11 |
Feature activation | +9.47e-03 |
Pos logit contributions | |
Mic | +0.068 |
Max | +0.067 |
ery | +0.065 |
Ted | +0.063 |
Charlie | +0.063 |
Neg logit contributions | |
goodbye | -0.063 |
contain | -0.053 |
plunge | -0.051 |
attention | -0.050 |
toward | -0.048 |
Token | especially |
Token ID | 2592 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
ery | +0.015 |
Mic | +0.015 |
Max | +0.015 |
mice | +0.014 |
Ted | +0.014 |
Neg logit contributions | |
goodbye | -0.019 |
contain | -0.016 |
plunge | -0.015 |
umb | -0.015 |
attention | -0.015 |
Token | grape |
Token ID | 30777 |
Feature activation | +9.68e-02 |
Pos logit contributions | |
Mic | +0.043 |
oon | +0.043 |
Ted | +0.042 |
Max | +0.042 |
iders | +0.040 |
Neg logit contributions | |
goodbye | -0.040 |
attention | -0.035 |
toward | -0.031 |
closer | -0.030 |
plunge | -0.029 |
Token | jelly |
Token ID | 27644 |
Feature activation | +5.93e-02 |
Pos logit contributions | |
Max | +0.211 |
Charlie | +0.206 |
Mic | +0.206 |
Next | +0.201 |
Ted | +0.201 |
Neg logit contributions | |
goodbye | -0.131 |
contain | -0.100 |
fy | -0.095 |
umb | -0.094 |
ogging | -0.093 |
Token | . |
Token ID | 13 |
Feature activation | +3.28e-04 |
Pos logit contributions | |
Max | +0.121 |
Mic | +0.121 |
Charlie | +0.117 |
Ted | +0.116 |
Next | +0.115 |
Neg logit contributions | |
goodbye | -0.090 |
contain | -0.072 |
fy | -0.068 |
tour | -0.067 |
plunge | -0.066 |
Token | One |
Token ID | 1881 |
Feature activation | -4.07e-03 |
Pos logit contributions | |
ery | +0.001 |
iflower | +0.001 |
Mic | +0.001 |
mice | +0.001 |
Fox | +0.001 |
Neg logit contributions | |
goodbye | -0.001 |
hug | -0.000 |
contain | -0.000 |
umb | -0.000 |
oured | -0.000 |
Token | day |
Token ID | 1110 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
goodbye | +0.008 |
contain | +0.006 |
plunge | +0.006 |
attention | +0.006 |
tour | +0.006 |
Neg logit contributions | |
Mic | -0.007 |
ery | -0.007 |
Max | -0.007 |
mice | -0.007 |
Ted | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
Max | +0.025 |
ery | +0.025 |
Mic | +0.024 |
mice | +0.024 |
Midnight | +0.024 |
Neg logit contributions | |
goodbye | -0.023 |
contain | -0.022 |
plunge | -0.020 |
umb | -0.019 |
oured | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
goodbye | +0.003 |
contain | +0.003 |
plunge | +0.003 |
umb | +0.002 |
attention | +0.002 |
Neg logit contributions | |
ery | -0.003 |
Mic | -0.003 |
Max | -0.003 |
mice | -0.003 |
iflower | -0.003 |
Token | black |
Token ID | 2042 |
Feature activation | -3.95e-02 |
Pos logit contributions | |
goodbye | +0.037 |
contain | +0.032 |
plunge | +0.031 |
toward | +0.030 |
fy | +0.030 |
Neg logit contributions | |
ery | -0.044 |
Mic | -0.042 |
mice | -0.042 |
Max | -0.042 |
iflower | -0.041 |
Token | mascara |
Token ID | 48430 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
goodbye | +0.067 |
contain | +0.061 |
plunge | +0.060 |
ounded | +0.058 |
fy | +0.058 |
Neg logit contributions | |
ery | -0.069 |
mice | -0.062 |
Max | -0.062 |
iflower | -0.062 |
chini | -0.061 |
Token | on |
Token ID | 319 |
Feature activation | +7.21e-03 |
Pos logit contributions | |
goodbye | +0.026 |
contain | +0.022 |
plunge | +0.022 |
umb | +0.021 |
attention | +0.020 |
Neg logit contributions | |
ery | -0.023 |
Max | -0.021 |
Mic | -0.021 |
mice | -0.021 |
iflower | -0.020 |
Token | her |
Token ID | 607 |
Feature activation | -6.73e-03 |
Pos logit contributions | |
ery | +0.013 |
Mic | +0.012 |
Max | +0.012 |
Bear | +0.012 |
mice | +0.012 |
Neg logit contributions | |
goodbye | -0.014 |
contain | -0.010 |
toward | -0.010 |
farewell | -0.010 |
hug | -0.010 |
Token | eyel |
Token ID | 29708 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
goodbye | +0.012 |
contain | +0.010 |
plunge | +0.010 |
umb | +0.009 |
toward | +0.009 |
Neg logit contributions | |
ery | -0.012 |
Mic | -0.012 |
Max | -0.012 |
mice | -0.011 |
iflower | -0.011 |
Token | ashes |
Token ID | 7465 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
goodbye | +0.241 |
plunge | +0.230 |
contain | +0.215 |
umb | +0.202 |
tour | +0.194 |
Neg logit contributions | |
ery | -0.258 |
hes | -0.225 |
Mic | -0.224 |
mice | -0.224 |
Next | -0.221 |
Token | . |
Token ID | 13 |
Feature activation | +9.81e-03 |
Pos logit contributions | |
goodbye | +0.024 |
contain | +0.020 |
plunge | +0.019 |
umb | +0.019 |
attention | +0.019 |
Neg logit contributions | |
Mic | -0.026 |
ery | -0.025 |
Max | -0.025 |
mice | -0.024 |
Ted | -0.024 |
Token | She |
Token ID | 1375 |
Feature activation | +8.33e-03 |
Pos logit contributions | |
ery | +0.017 |
iflower | +0.016 |
Mic | +0.016 |
lf | +0.015 |
mice | +0.014 |
Neg logit contributions | |
goodbye | -0.020 |
hi | -0.015 |
hello | -0.015 |
inspection | -0.015 |
toward | -0.015 |
Token | makes |
Token ID | 1838 |
Feature activation | +4.41e-03 |
Pos logit contributions | |
ery | +0.015 |
Mic | +0.014 |
Max | +0.014 |
mice | +0.014 |
iflower | +0.014 |
Neg logit contributions | |
goodbye | -0.015 |
contain | -0.012 |
plunge | -0.012 |
umb | -0.012 |
attention | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
Mic | +0.007 |
ery | +0.007 |
Max | +0.006 |
iflower | +0.006 |
mice | +0.006 |
Neg logit contributions | |
goodbye | -0.009 |
contain | -0.008 |
plunge | -0.008 |
attention | -0.008 |
umb | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
goodbye | +0.019 |
plunge | +0.018 |
contain | +0.017 |
fy | +0.017 |
icol | +0.016 |
Neg logit contributions | |
ery | -0.014 |
mice | -0.013 |
Max | -0.013 |
ters | -0.013 |
magnets | -0.013 |
Token | pushed |
Token ID | 7121 |
Feature activation | -2.70e-02 |
Pos logit contributions | |
ery | +0.018 |
Mic | +0.017 |
Max | +0.017 |
mice | +0.017 |
iflower | +0.016 |
Neg logit contributions | |
goodbye | -0.020 |
contain | -0.017 |
plunge | -0.016 |
umb | -0.016 |
attention | -0.016 |
Token | her |
Token ID | 607 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
goodbye | +0.059 |
contain | +0.052 |
plunge | +0.050 |
umb | +0.049 |
attention | +0.049 |
Neg logit contributions | |
ery | -0.037 |
Mic | -0.036 |
Max | -0.035 |
mice | -0.034 |
iflower | -0.033 |
Token | off |
Token ID | 572 |
Feature activation | -3.67e-02 |
Pos logit contributions | |
goodbye | +0.037 |
fy | +0.035 |
oured | +0.034 |
contain | +0.034 |
icol | +0.033 |
Neg logit contributions | |
ery | -0.039 |
ters | -0.038 |
Max | -0.038 |
mice | -0.037 |
Mic | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | -4.83e-02 |
Pos logit contributions | |
goodbye | +0.073 |
contain | +0.064 |
plunge | +0.062 |
umb | +0.060 |
attention | +0.059 |
Neg logit contributions | |
ery | -0.056 |
Mic | -0.055 |
Max | -0.054 |
mice | -0.053 |
Ted | -0.051 |
Token | sc |
Token ID | 629 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
goodbye | +0.095 |
oured | +0.090 |
fy | +0.086 |
contain | +0.082 |
ur | +0.080 |
Neg logit contributions | |
Max | -0.067 |
Mic | -0.065 |
You | -0.062 |
Tim | -0.061 |
mice | -0.058 |
Token | ooter |
Token ID | 25141 |
Feature activation | -4.04e-02 |
Pos logit contributions | |
goodbye | +0.202 |
plunge | +0.189 |
contain | +0.179 |
anticipation | +0.177 |
hesitation | +0.163 |
Neg logit contributions | |
house | -0.243 |
ery | -0.240 |
Mic | -0.239 |
Riding | -0.236 |
Max | -0.236 |
Token | . |
Token ID | 13 |
Feature activation | +5.53e-03 |
Pos logit contributions | |
goodbye | +0.082 |
plunge | +0.074 |
contain | +0.072 |
umb | +0.071 |
oured | +0.070 |
Neg logit contributions | |
ery | -0.057 |
Max | -0.056 |
Mic | -0.053 |
mice | -0.053 |
ters | -0.050 |
Token | Anna |
Token ID | 11735 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
ery | +0.012 |
iflower | +0.011 |
Mic | +0.011 |
ters | +0.010 |
hers | +0.010 |
Neg logit contributions | |
goodbye | -0.010 |
farewell | -0.007 |
toward | -0.007 |
inspection | -0.007 |
hug | -0.007 |
Token | fell |
Token ID | 3214 |
Feature activation | -5.40e-02 |
Pos logit contributions | |
goodbye | +0.040 |
contain | +0.033 |
plunge | +0.031 |
oured | +0.030 |
toward | +0.030 |
Neg logit contributions | |
Mic | -0.036 |
ery | -0.036 |
iflower | -0.034 |
mice | -0.033 |
Next | -0.033 |
Token | on |
Token ID | 319 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
goodbye | +0.097 |
contain | +0.097 |
plunge | +0.091 |
tour | +0.088 |
fy | +0.087 |
Neg logit contributions | |
Max | -0.085 |
mice | -0.080 |
ery | -0.077 |
ters | -0.076 |
hes | -0.074 |
Token | She |
Token ID | 1375 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
goodbye | +0.012 |
farewell | +0.009 |
hi | +0.009 |
hug | +0.009 |
toward | +0.009 |
Neg logit contributions | |
ery | -0.012 |
iflower | -0.011 |
Mic | -0.011 |
chini | -0.011 |
web | -0.011 |
Token | put |
Token ID | 1234 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
Mic | +0.036 |
ery | +0.033 |
Charlie | +0.033 |
Max | +0.032 |
mice | +0.032 |
Neg logit contributions | |
goodbye | -0.037 |
contain | -0.031 |
hug | -0.030 |
oured | -0.029 |
wave | -0.029 |
Token | some |
Token ID | 617 |
Feature activation | -9.82e-03 |
Pos logit contributions | |
Mic | +0.026 |
ery | +0.025 |
Max | +0.024 |
mice | +0.024 |
Ted | +0.024 |
Neg logit contributions | |
goodbye | -0.031 |
contain | -0.025 |
attention | -0.025 |
umb | -0.025 |
plunge | -0.025 |
Token | on |
Token ID | 319 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
goodbye | +0.020 |
contain | +0.018 |
plunge | +0.017 |
toward | +0.016 |
farewell | +0.016 |
Neg logit contributions | |
ery | -0.014 |
mice | -0.014 |
Max | -0.013 |
Mic | -0.013 |
dough | -0.013 |
Token | her |
Token ID | 607 |
Feature activation | -3.86e-03 |
Pos logit contributions | |
goodbye | +0.026 |
contain | +0.021 |
umb | +0.020 |
attention | +0.020 |
toward | +0.020 |
Neg logit contributions | |
ery | -0.021 |
Mic | -0.020 |
mice | -0.019 |
Bear | -0.019 |
Max | -0.019 |
Token | eyel |
Token ID | 29708 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
goodbye | +0.007 |
contain | +0.007 |
plunge | +0.006 |
umb | +0.006 |
toward | +0.006 |
Neg logit contributions | |
ery | -0.006 |
Mic | -0.006 |
Max | -0.006 |
mice | -0.006 |
Midnight | -0.006 |
Token | ids |
Token ID | 2340 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
goodbye | +0.270 |
plunge | +0.258 |
contain | +0.239 |
tour | +0.227 |
further | +0.223 |
Neg logit contributions | |
ery | -0.213 |
Cere | -0.190 |
hes | -0.186 |
mice | -0.184 |
Next | -0.181 |
Token | and |
Token ID | 290 |
Feature activation | +6.90e-04 |
Pos logit contributions | |
goodbye | +0.029 |
contain | +0.024 |
plunge | +0.023 |
umb | +0.023 |
attention | +0.022 |
Neg logit contributions | |
ery | -0.029 |
Mic | -0.029 |
Max | -0.028 |
mice | -0.027 |
Ted | -0.027 |
Token | cheeks |
Token ID | 25839 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
ery | +0.001 |
Mic | +0.001 |
Max | +0.001 |
mice | +0.001 |
iflower | +0.001 |
Neg logit contributions | |
goodbye | -0.001 |
contain | -0.001 |
plunge | -0.001 |
umb | -0.001 |
attention | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -5.46e-03 |
Pos logit contributions | |
goodbye | +0.067 |
contain | +0.058 |
plunge | +0.056 |
umb | +0.054 |
attention | +0.053 |
Neg logit contributions | |
ery | -0.060 |
Mic | -0.058 |
Max | -0.057 |
mice | -0.056 |
iflower | -0.054 |
Token | It |
Token ID | 632 |
Feature activation | +4.37e-02 |
Pos logit contributions | |
goodbye | +0.010 |
farewell | +0.008 |
hug | +0.008 |
hi | +0.007 |
toward | +0.007 |
Neg logit contributions | |
ery | -0.010 |
iflower | -0.009 |
Mic | -0.009 |
web | -0.009 |
kin | -0.009 |
Token | She |
Token ID | 1375 |
Feature activation | +6.23e-03 |
Pos logit contributions | |
goodbye | +0.026 |
oured | +0.020 |
toward | +0.020 |
farewell | +0.020 |
ogging | +0.020 |
Neg logit contributions | |
ery | -0.021 |
iflower | -0.018 |
chini | -0.018 |
Mic | -0.017 |
mice | -0.017 |
Token | climbed |
Token ID | 19952 |
Feature activation | -4.00e-02 |
Pos logit contributions | |
ery | +0.010 |
Mic | +0.010 |
Max | +0.010 |
mice | +0.010 |
iflower | +0.009 |
Neg logit contributions | |
goodbye | -0.012 |
contain | -0.010 |
plunge | -0.010 |
umb | -0.010 |
attention | -0.010 |
Token | back |
Token ID | 736 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
goodbye | +0.073 |
contain | +0.071 |
fy | +0.065 |
plunge | +0.063 |
eal | +0.063 |
Neg logit contributions | |
Max | -0.063 |
mice | -0.063 |
ery | -0.062 |
Mic | -0.062 |
Midnight | -0.059 |
Token | on |
Token ID | 319 |
Feature activation | -4.66e-02 |
Pos logit contributions | |
goodbye | +0.036 |
contain | +0.036 |
fy | +0.032 |
plunge | +0.031 |
eal | +0.029 |
Neg logit contributions | |
mice | -0.035 |
ery | -0.033 |
Max | -0.033 |
Next | -0.032 |
house | -0.032 |
Token | her |
Token ID | 607 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
goodbye | +0.092 |
contain | +0.074 |
toward | +0.072 |
plunge | +0.072 |
attention | +0.071 |
Neg logit contributions | |
ery | -0.080 |
Mic | -0.074 |
iflower | -0.073 |
chini | -0.071 |
Max | -0.070 |
Token | sc |
Token ID | 629 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
goodbye | +0.034 |
plunge | +0.033 |
eal | +0.033 |
oured | +0.032 |
contain | +0.032 |
Neg logit contributions | |
Mic | -0.035 |
mice | -0.035 |
ery | -0.034 |
iflower | -0.034 |
Max | -0.034 |
Token | ooter |
Token ID | 25141 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
plunge | +0.187 |
goodbye | +0.173 |
contain | +0.168 |
umb | +0.166 |
anticipation | +0.161 |
Neg logit contributions | |
house | -0.239 |
ery | -0.236 |
Mic | -0.233 |
ase | -0.231 |
iny | -0.230 |
Token | and |
Token ID | 290 |
Feature activation | -1.51e-03 |
Pos logit contributions | |
goodbye | +0.083 |
plunge | +0.076 |
contain | +0.075 |
umb | +0.073 |
attention | +0.070 |
Neg logit contributions | |
Max | -0.061 |
ery | -0.061 |
Mic | -0.059 |
mice | -0.059 |
Midnight | -0.056 |
Token | went |
Token ID | 1816 |
Feature activation | -5.31e-02 |
Pos logit contributions | |
goodbye | +0.003 |
contain | +0.002 |
plunge | +0.002 |
attention | +0.002 |
umb | +0.002 |
Neg logit contributions | |
Max | -0.002 |
ery | -0.002 |
Mic | -0.002 |
mice | -0.002 |
Midnight | -0.002 |
Token | back |
Token ID | 736 |
Feature activation | -3.67e-03 |
Pos logit contributions | |
contain | +0.102 |
plunge | +0.097 |
ounded | +0.096 |
anticipation | +0.096 |
hesitation | +0.092 |
Neg logit contributions | |
Max | -0.060 |
Midnight | -0.059 |
hes | -0.057 |
with | -0.056 |
Max | -0.056 |
Token | to |
Token ID | 284 |
Feature activation | -3.52e-02 |
Pos logit contributions | |
goodbye | +0.006 |
contain | +0.006 |
plunge | +0.005 |
fy | +0.005 |
eal | +0.005 |
Neg logit contributions | |
mice | -0.007 |
Max | -0.007 |
Mic | -0.006 |
Ted | -0.006 |
Midnight | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
goodbye | +0.083 |
plunge | +0.074 |
contain | +0.073 |
umb | +0.073 |
oured | +0.072 |
Neg logit contributions | |
ery | -0.064 |
Max | -0.062 |
mice | -0.059 |
Mic | -0.057 |
ters | -0.057 |
Token | He |
Token ID | 679 |
Feature activation | +2.15e-02 |
Pos logit contributions | |
goodbye | +0.032 |
toward | +0.026 |
contain | +0.025 |
farewell | +0.025 |
ogging | +0.024 |
Neg logit contributions | |
ery | -0.028 |
iflower | -0.025 |
mice | -0.025 |
Mic | -0.024 |
chini | -0.023 |
Token | hopped |
Token ID | 45230 |
Feature activation | -6.44e-02 |
Pos logit contributions | |
ery | +0.028 |
Max | +0.027 |
Mic | +0.026 |
Bear | +0.025 |
mice | +0.025 |
Neg logit contributions | |
goodbye | -0.045 |
contain | -0.042 |
plunge | -0.041 |
attention | -0.041 |
umb | -0.040 |
Token | on |
Token ID | 319 |
Feature activation | -4.26e-02 |
Pos logit contributions | |
umb | +0.113 |
contain | +0.111 |
goodbye | +0.109 |
des | +0.107 |
fy | +0.107 |
Neg logit contributions | |
Max | -0.094 |
ery | -0.083 |
Ted | -0.078 |
Midnight | -0.077 |
Mic | -0.077 |
Token | his |
Token ID | 465 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
goodbye | +0.090 |
contain | +0.087 |
eal | +0.084 |
oured | +0.084 |
fy | +0.084 |
Neg logit contributions | |
Max | -0.050 |
mice | -0.049 |
Mic | -0.046 |
ery | -0.045 |
Ted | -0.044 |
Token | sc |
Token ID | 629 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
oured | +0.032 |
oms | +0.032 |
eal | +0.030 |
plunge | +0.030 |
utt | +0.030 |
Neg logit contributions | |
Max | -0.033 |
Mic | -0.032 |
mice | -0.031 |
lf | -0.030 |
Charlie | -0.030 |
Token | ooter |
Token ID | 25141 |
Feature activation | -4.34e-02 |
Pos logit contributions | |
plunge | +0.166 |
umb | +0.154 |
goodbye | +0.148 |
inspection | +0.139 |
contain | +0.137 |
Neg logit contributions | |
Riding | -0.229 |
ery | -0.224 |
her | -0.221 |
house | -0.219 |
Max | -0.219 |
Token | and |
Token ID | 290 |
Feature activation | -2.86e-03 |
Pos logit contributions | |
goodbye | +0.086 |
plunge | +0.081 |
contain | +0.080 |
umb | +0.079 |
eal | +0.074 |
Neg logit contributions | |
Max | -0.058 |
ery | -0.055 |
Mic | -0.053 |
mice | -0.051 |
house | -0.050 |
Token | made |
Token ID | 925 |
Feature activation | -4.61e-03 |
Pos logit contributions | |
goodbye | +0.005 |
contain | +0.005 |
plunge | +0.005 |
attention | +0.004 |
umb | +0.004 |
Neg logit contributions | |
Max | -0.005 |
ery | -0.004 |
Mic | -0.004 |
Ted | -0.004 |
mice | -0.004 |
Token | his |
Token ID | 465 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
goodbye | +0.010 |
contain | +0.009 |
plunge | +0.009 |
umb | +0.009 |
toward | +0.009 |
Neg logit contributions | |
ery | -0.006 |
Max | -0.006 |
Mic | -0.006 |
mice | -0.006 |
Ted | -0.005 |
Token | way |
Token ID | 835 |
Feature activation | -4.04e-02 |
Pos logit contributions | |
mice | +0.028 |
Max | +0.028 |
ery | +0.028 |
Next | +0.026 |
house | +0.026 |
Neg logit contributions | |
goodbye | -0.034 |
oms | -0.034 |
oured | -0.034 |
contain | -0.033 |
plunge | -0.033 |
Token | , |
Token ID | 11 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
ery | +0.012 |
Mic | +0.012 |
Max | +0.012 |
mice | +0.012 |
iflower | +0.011 |
Neg logit contributions | |
goodbye | -0.010 |
contain | -0.008 |
plunge | -0.008 |
umb | -0.007 |
attention | -0.007 |
Token | Tim |
Token ID | 5045 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
ery | +0.028 |
Mic | +0.028 |
iflower | +0.026 |
Max | +0.026 |
mice | +0.026 |
Neg logit contributions | |
goodbye | -0.035 |
plunge | -0.029 |
contain | -0.029 |
umb | -0.028 |
attention | -0.027 |
Token | fell |
Token ID | 3214 |
Feature activation | -3.84e-02 |
Pos logit contributions | |
ery | +0.019 |
Mic | +0.018 |
mice | +0.017 |
chini | +0.017 |
iflower | +0.017 |
Neg logit contributions | |
goodbye | -0.019 |
contain | -0.016 |
plunge | -0.016 |
umb | -0.015 |
oured | -0.015 |
Token | off |
Token ID | 572 |
Feature activation | -3.95e-02 |
Pos logit contributions | |
goodbye | +0.067 |
contain | +0.065 |
plunge | +0.061 |
tour | +0.058 |
tickets | +0.058 |
Neg logit contributions | |
Max | -0.065 |
ery | -0.063 |
Mic | -0.060 |
mice | -0.059 |
Midnight | -0.059 |
Token | the |
Token ID | 262 |
Feature activation | -4.19e-02 |
Pos logit contributions | |
goodbye | +0.088 |
plunge | +0.083 |
contain | +0.083 |
oured | +0.080 |
ounded | +0.080 |
Neg logit contributions | |
Max | -0.044 |
ery | -0.043 |
mice | -0.042 |
Midnight | -0.040 |
Mic | -0.040 |
Token | sc |
Token ID | 629 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
goodbye | +0.084 |
oured | +0.079 |
eal | +0.078 |
inspection | +0.078 |
oms | +0.077 |
Neg logit contributions | |
Max | -0.056 |
Mic | -0.055 |
iflower | -0.050 |
lf | -0.050 |
mice | -0.050 |
Token | ooter |
Token ID | 25141 |
Feature activation | -3.95e-02 |
Pos logit contributions | |
plunge | +0.187 |
goodbye | +0.186 |
anticipation | +0.175 |
contain | +0.174 |
inspection | +0.169 |
Neg logit contributions | |
house | -0.221 |
her | -0.214 |
Max | -0.211 |
Riding | -0.211 |
hes | -0.208 |
Token | . |
Token ID | 13 |
Feature activation | -2.53e-03 |
Pos logit contributions | |
goodbye | +0.083 |
plunge | +0.079 |
umb | +0.077 |
contain | +0.074 |
tickets | +0.073 |
Neg logit contributions | |
Max | -0.049 |
ery | -0.043 |
house | -0.043 |
Mic | -0.042 |
Midnight | -0.040 |
Token | He |
Token ID | 679 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
goodbye | +0.004 |
hi | +0.003 |
toward | +0.003 |
hug | +0.003 |
plunge | +0.003 |
Neg logit contributions | |
ery | -0.005 |
iflower | -0.005 |
Mic | -0.004 |
mice | -0.004 |
ters | -0.004 |
Token | felt |
Token ID | 2936 |
Feature activation | -9.22e-03 |
Pos logit contributions | |
ery | +0.033 |
Max | +0.032 |
Mic | +0.032 |
mice | +0.030 |
Ted | +0.029 |
Neg logit contributions | |
goodbye | -0.051 |
contain | -0.046 |
plunge | -0.045 |
attention | -0.043 |
umb | -0.043 |
Token | hurt |
Token ID | 5938 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
goodbye | +0.019 |
contain | +0.017 |
plunge | +0.017 |
tour | +0.017 |
umb | +0.017 |
Neg logit contributions | |
ery | -0.013 |
mice | -0.012 |
Max | -0.012 |
Mic | -0.012 |
Midnight | -0.012 |
Token | him |
Token ID | 683 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
goodbye | +0.052 |
contain | +0.043 |
attention | +0.042 |
toward | +0.042 |
plunge | +0.040 |
Neg logit contributions | |
ery | -0.072 |
iflower | -0.069 |
Mic | -0.069 |
chini | -0.065 |
Next | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
ery | +0.023 |
Mic | +0.022 |
Max | +0.022 |
mice | +0.021 |
iflower | +0.021 |
Neg logit contributions | |
goodbye | -0.025 |
contain | -0.021 |
plunge | -0.020 |
umb | -0.020 |
fy | -0.019 |
Token | She |
Token ID | 1375 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
goodbye | +0.029 |
toward | +0.022 |
oured | +0.021 |
ounded | +0.021 |
hug | +0.021 |
Neg logit contributions | |
ery | -0.028 |
iflower | -0.025 |
chini | -0.024 |
Mic | -0.022 |
web | -0.022 |
Token | pulled |
Token ID | 5954 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
ery | +0.020 |
Mic | +0.019 |
iflower | +0.018 |
chini | +0.018 |
mice | +0.018 |
Neg logit contributions | |
goodbye | -0.018 |
oured | -0.014 |
umb | -0.014 |
contain | -0.014 |
closer | -0.013 |
Token | his |
Token ID | 465 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
goodbye | +0.035 |
attention | +0.029 |
toward | +0.028 |
plunge | +0.028 |
closer | +0.028 |
Neg logit contributions | |
ery | -0.037 |
iflower | -0.035 |
Mic | -0.034 |
chini | -0.034 |
oyd | -0.032 |
Token | sc |
Token ID | 629 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
goodbye | +0.042 |
contain | +0.041 |
fy | +0.039 |
ogging | +0.037 |
plunge | +0.037 |
Neg logit contributions | |
Mic | -0.030 |
ery | -0.030 |
Max | -0.030 |
mice | -0.030 |
magnets | -0.028 |
Token | ooter |
Token ID | 25141 |
Feature activation | -4.56e-02 |
Pos logit contributions | |
goodbye | +0.175 |
plunge | +0.148 |
contain | +0.145 |
icol | +0.133 |
anticipation | +0.132 |
Neg logit contributions | |
ery | -0.258 |
Mic | -0.255 |
Max | -0.250 |
house | -0.246 |
hes | -0.246 |
Token | out |
Token ID | 503 |
Feature activation | -3.91e-02 |
Pos logit contributions | |
goodbye | +0.075 |
contain | +0.066 |
plunge | +0.063 |
umb | +0.060 |
oured | +0.059 |
Neg logit contributions | |
Max | -0.083 |
ery | -0.081 |
Mic | -0.079 |
mice | -0.076 |
ters | -0.075 |
Token | of |
Token ID | 286 |
Feature activation | +2.06e-03 |
Pos logit contributions | |
goodbye | +0.075 |
plunge | +0.066 |
contain | +0.066 |
oured | +0.062 |
eal | +0.062 |
Neg logit contributions | |
ery | -0.060 |
mice | -0.059 |
Max | -0.059 |
Ted | -0.057 |
Mic | -0.056 |
Token | the |
Token ID | 262 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
ery | +0.003 |
Mic | +0.003 |
iflower | +0.003 |
Max | +0.003 |
mice | +0.003 |
Neg logit contributions | |
goodbye | -0.004 |
contain | -0.004 |
plunge | -0.003 |
toward | -0.003 |
umb | -0.003 |
Token | flower |
Token ID | 15061 |
Feature activation | +3.92e-03 |
Pos logit contributions | |
goodbye | +0.065 |
contain | +0.058 |
oured | +0.054 |
fy | +0.053 |
plunge | +0.053 |
Neg logit contributions | |
mice | -0.046 |
Mic | -0.045 |
ery | -0.045 |
Max | -0.043 |
iflower | -0.042 |
Token | cried |
Token ID | 16896 |
Feature activation | -3.72e-02 |
Pos logit contributions | |
goodbye | +0.033 |
contain | +0.028 |
plunge | +0.027 |
umb | +0.027 |
attention | +0.027 |
Neg logit contributions | |
ery | -0.022 |
Mic | -0.021 |
Max | -0.021 |
mice | -0.020 |
iflower | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
goodbye | +0.069 |
contain | +0.060 |
plunge | +0.058 |
umb | +0.056 |
oured | +0.055 |
Neg logit contributions | |
ery | -0.063 |
Mic | -0.061 |
Max | -0.061 |
mice | -0.059 |
Ted | -0.057 |
Token | Tommy |
Token ID | 19919 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
ery | +0.030 |
iflower | +0.027 |
chini | +0.025 |
web | +0.024 |
hers | +0.024 |
Neg logit contributions | |
goodbye | -0.023 |
toward | -0.017 |
hug | -0.017 |
hello | -0.017 |
hi | -0.017 |
Token | grabbed |
Token ID | 13176 |
Feature activation | -9.22e-03 |
Pos logit contributions | |
goodbye | +0.021 |
wave | +0.017 |
hug | +0.017 |
contain | +0.016 |
eal | +0.016 |
Neg logit contributions | |
Mic | -0.025 |
ery | -0.025 |
chini | -0.024 |
iflower | -0.024 |
Next | -0.023 |
Token | her |
Token ID | 607 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
goodbye | +0.021 |
contain | +0.018 |
plunge | +0.018 |
umb | +0.017 |
attention | +0.017 |
Neg logit contributions | |
ery | -0.012 |
Mic | -0.012 |
Max | -0.012 |
mice | -0.011 |
iflower | -0.011 |
Token | sc |
Token ID | 629 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
goodbye | +0.047 |
contain | +0.041 |
plunge | +0.040 |
umb | +0.039 |
fy | +0.038 |
Neg logit contributions | |
ery | -0.040 |
Mic | -0.038 |
Max | -0.038 |
mice | -0.037 |
iflower | -0.036 |
Token | ooter |
Token ID | 25141 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
goodbye | +0.185 |
plunge | +0.173 |
umb | +0.172 |
anticipation | +0.166 |
further | +0.165 |
Neg logit contributions | |
ery | -0.231 |
Mic | -0.222 |
Max | -0.216 |
iflower | -0.209 |
Blood | -0.209 |
Token | and |
Token ID | 290 |
Feature activation | +2.49e-03 |
Pos logit contributions | |
goodbye | +0.079 |
contain | +0.072 |
plunge | +0.071 |
umb | +0.068 |
tour | +0.067 |
Neg logit contributions | |
ery | -0.052 |
Max | -0.052 |
Mic | -0.050 |
mice | -0.049 |
Midnight | -0.046 |
Token | rode |
Token ID | 22075 |
Feature activation | -4.87e-02 |
Pos logit contributions | |
ery | +0.004 |
Mic | +0.004 |
Max | +0.004 |
mice | +0.004 |
iflower | +0.004 |
Neg logit contributions | |
goodbye | -0.005 |
contain | -0.004 |
plunge | -0.004 |
umb | -0.004 |
attention | -0.004 |
Token | away |
Token ID | 1497 |
Feature activation | -6.23e-05 |
Pos logit contributions | |
goodbye | +0.080 |
contain | +0.075 |
plunge | +0.071 |
umb | +0.070 |
oured | +0.069 |
Neg logit contributions | |
Max | -0.088 |
ery | -0.084 |
mice | -0.082 |
Mic | -0.082 |
Ted | -0.079 |
Token | . |
Token ID | 13 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
goodbye | +0.000 |
contain | +0.000 |
plunge | +0.000 |
umb | +0.000 |
attention | +0.000 |
Neg logit contributions | |
ery | -0.000 |
Mic | -0.000 |
Max | -0.000 |
mice | -0.000 |
iflower | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | +5.95e-03 |
Pos logit contributions | |
ery | +0.012 |
mice | +0.012 |
Mic | +0.012 |
iflower | +0.011 |
ters | +0.011 |
Neg logit contributions | |
goodbye | -0.022 |
contain | -0.020 |
plunge | -0.019 |
toward | -0.019 |
ogging | -0.019 |
Token | Tim |
Token ID | 14967 |
Feature activation | +4.31e-03 |
Pos logit contributions | |
mice | +0.008 |
ery | +0.008 |
Midnight | +0.008 |
ters | +0.008 |
iflower | +0.008 |
Neg logit contributions | |
goodbye | -0.012 |
contain | -0.011 |
toward | -0.011 |
tour | -0.011 |
fy | -0.011 |
Token | my |
Token ID | 1820 |
Feature activation | +7.68e-03 |
Pos logit contributions | |
ery | +0.006 |
Mic | +0.006 |
mice | +0.006 |
chini | +0.006 |
Midnight | +0.006 |
Neg logit contributions | |
goodbye | -0.008 |
contain | -0.007 |
fy | -0.007 |
attention | -0.007 |
oured | -0.007 |
Token | rode |
Token ID | 22075 |
Feature activation | -4.28e-02 |
Pos logit contributions | |
ery | +0.011 |
Mic | +0.011 |
Max | +0.011 |
mice | +0.011 |
iflower | +0.010 |
Neg logit contributions | |
goodbye | -0.016 |
contain | -0.014 |
plunge | -0.013 |
umb | -0.013 |
toward | -0.013 |
Token | his |
Token ID | 465 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
goodbye | +0.092 |
contain | +0.085 |
plunge | +0.085 |
umb | +0.084 |
oured | +0.082 |
Neg logit contributions | |
Max | -0.055 |
ery | -0.052 |
Mic | -0.051 |
mice | -0.051 |
Ted | -0.048 |
Token | sc |
Token ID | 629 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
goodbye | +0.042 |
contain | +0.039 |
fy | +0.037 |
oured | +0.035 |
plunge | +0.035 |
Neg logit contributions | |
Mic | -0.045 |
Max | -0.044 |
mice | -0.040 |
lf | -0.040 |
ery | -0.040 |
Token | ooter |
Token ID | 25141 |
Feature activation | -4.34e-02 |
Pos logit contributions | |
goodbye | +0.174 |
plunge | +0.157 |
contain | +0.147 |
anticipation | +0.140 |
icol | +0.138 |
Neg logit contributions | |
Max | -0.245 |
ery | -0.242 |
Mic | -0.242 |
hes | -0.239 |
house | -0.237 |
Token | all |
Token ID | 477 |
Feature activation | -7.20e-02 |
Pos logit contributions | |
goodbye | +0.082 |
contain | +0.076 |
plunge | +0.076 |
umb | +0.076 |
oured | +0.071 |
Neg logit contributions | |
Max | -0.064 |
ery | -0.061 |
Mic | -0.061 |
mice | -0.058 |
Ted | -0.055 |
Token | around |
Token ID | 1088 |
Feature activation | +8.48e-03 |
Pos logit contributions | |
goodbye | +0.151 |
contain | +0.133 |
umb | +0.132 |
plunge | +0.131 |
oured | +0.128 |
Neg logit contributions | |
Max | -0.100 |
ery | -0.100 |
Ted | -0.093 |
Mic | -0.093 |
mice | -0.089 |
Token | the |
Token ID | 262 |
Feature activation | -6.85e-03 |
Pos logit contributions | |
ery | +0.014 |
Mic | +0.014 |
iflower | +0.013 |
Max | +0.013 |
mice | +0.013 |
Neg logit contributions | |
goodbye | -0.017 |
contain | -0.014 |
toward | -0.014 |
plunge | -0.014 |
attention | -0.013 |
Token | park |
Token ID | 3952 |
Feature activation | -7.95e-03 |
Pos logit contributions | |
goodbye | +0.015 |
contain | +0.013 |
plunge | +0.012 |
umb | +0.012 |
toward | +0.012 |
Neg logit contributions | |
ery | -0.010 |
Mic | -0.009 |
Max | -0.009 |
mice | -0.009 |
iflower | -0.009 |
Token | them |
Token ID | 606 |
Feature activation | +8.68e-03 |
Pos logit contributions | |
ery | +0.062 |
Mic | +0.061 |
iflower | +0.058 |
Max | +0.056 |
chini | +0.055 |
Neg logit contributions | |
goodbye | -0.066 |
contain | -0.054 |
plunge | -0.054 |
umb | -0.054 |
attention | -0.054 |
Token | . |
Token ID | 13 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
Mic | +0.018 |
ery | +0.017 |
iflower | +0.017 |
Max | +0.017 |
Ted | +0.016 |
Neg logit contributions | |
goodbye | -0.014 |
contain | -0.011 |
attention | -0.010 |
plunge | -0.010 |
toward | -0.010 |
Token | She |
Token ID | 1375 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
ery | +0.033 |
iflower | +0.032 |
chini | +0.029 |
Mic | +0.028 |
lf | +0.027 |
Neg logit contributions | |
goodbye | -0.026 |
toward | -0.021 |
attention | -0.020 |
hi | -0.020 |
hello | -0.020 |
Token | rode |
Token ID | 22075 |
Feature activation | -4.54e-02 |
Pos logit contributions | |
ery | +0.024 |
Mic | +0.023 |
Max | +0.023 |
mice | +0.022 |
iflower | +0.021 |
Neg logit contributions | |
goodbye | -0.029 |
contain | -0.025 |
plunge | -0.024 |
umb | -0.024 |
attention | -0.023 |
Token | her |
Token ID | 607 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
goodbye | +0.094 |
contain | +0.084 |
plunge | +0.082 |
umb | +0.080 |
oured | +0.079 |
Neg logit contributions | |
ery | -0.064 |
Max | -0.063 |
Mic | -0.062 |
mice | -0.061 |
Ted | -0.058 |
Token | sc |
Token ID | 629 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
goodbye | +0.049 |
contain | +0.047 |
oured | +0.046 |
fy | +0.046 |
eal | +0.045 |
Neg logit contributions | |
Max | -0.045 |
mice | -0.044 |
ery | -0.043 |
Mic | -0.042 |
magnets | -0.041 |
Token | ooter |
Token ID | 25141 |
Feature activation | -4.20e-02 |
Pos logit contributions | |
plunge | +0.154 |
goodbye | +0.150 |
anticipation | +0.141 |
contain | +0.140 |
hesitation | +0.136 |
Neg logit contributions | |
Max | -0.258 |
ery | -0.251 |
house | -0.245 |
Next | -0.241 |
ase | -0.240 |
Token | to |
Token ID | 284 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
goodbye | +0.075 |
contain | +0.067 |
plunge | +0.066 |
umb | +0.065 |
oured | +0.062 |
Neg logit contributions | |
Max | -0.069 |
ery | -0.069 |
Mic | -0.067 |
mice | -0.066 |
Midnight | -0.063 |
Token | them |
Token ID | 606 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
goodbye | +0.024 |
contain | +0.021 |
plunge | +0.021 |
attention | +0.020 |
umb | +0.020 |
Neg logit contributions | |
ery | -0.016 |
Mic | -0.016 |
Max | -0.015 |
iflower | -0.015 |
mice | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
goodbye | +0.018 |
contain | +0.014 |
plunge | +0.014 |
toward | +0.013 |
attention | +0.013 |
Neg logit contributions | |
ery | -0.019 |
Mic | -0.019 |
Max | -0.018 |
iflower | -0.018 |
Ted | -0.018 |
Token | She |
Token ID | 1375 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
ery | +0.024 |
iflower | +0.024 |
Mic | +0.021 |
chini | +0.020 |
lf | +0.020 |
Neg logit contributions | |
goodbye | -0.020 |
toward | -0.017 |
inspection | -0.015 |
hug | -0.015 |
hi | -0.015 |
Token | zoom |
Token ID | 19792 |
Feature activation | -3.88e-02 |
Pos logit contributions | |
goodbye | +0.026 |
contain | +0.022 |
plunge | +0.021 |
umb | +0.020 |
toward | +0.020 |
Neg logit contributions | |
ery | -0.025 |
Mic | -0.024 |
Max | -0.024 |
mice | -0.023 |
iflower | -0.023 |
Token | ed |
Token ID | 276 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
goodbye | +0.068 |
contain | +0.061 |
plunge | +0.058 |
umb | +0.057 |
attention | +0.055 |
Neg logit contributions | |
ery | -0.067 |
Max | -0.065 |
Mic | -0.063 |
mice | -0.063 |
Ted | -0.061 |
Token | off |
Token ID | 572 |
Feature activation | -4.53e-02 |
Pos logit contributions | |
goodbye | +0.027 |
contain | +0.023 |
plunge | +0.022 |
umb | +0.021 |
attention | +0.021 |
Neg logit contributions | |
ery | -0.032 |
Mic | -0.032 |
Max | -0.031 |
mice | -0.031 |
iflower | -0.030 |
Token | on |
Token ID | 319 |
Feature activation | -3.67e-02 |
Pos logit contributions | |
goodbye | +0.082 |
contain | +0.068 |
toward | +0.064 |
plunge | +0.064 |
umb | +0.062 |
Neg logit contributions | |
ery | -0.083 |
Mic | -0.081 |
Max | -0.079 |
iflower | -0.076 |
mice | -0.075 |
Token | his |
Token ID | 465 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
goodbye | +0.071 |
contain | +0.057 |
toward | +0.055 |
attention | +0.055 |
farewell | +0.053 |
Neg logit contributions | |
ery | -0.066 |
Mic | -0.065 |
Max | -0.062 |
Bear | -0.060 |
iflower | -0.059 |
Token | sc |
Token ID | 629 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
goodbye | +0.021 |
plunge | +0.019 |
contain | +0.019 |
oured | +0.018 |
toward | +0.018 |
Neg logit contributions | |
ery | -0.022 |
mice | -0.021 |
Mic | -0.021 |
iflower | -0.021 |
Max | -0.020 |
Token | ooter |
Token ID | 25141 |
Feature activation | -3.99e-02 |
Pos logit contributions | |
plunge | +0.167 |
goodbye | +0.163 |
contain | +0.146 |
umb | +0.145 |
toward | +0.135 |
Neg logit contributions | |
ery | -0.266 |
Mic | -0.249 |
iflower | -0.248 |
house | -0.247 |
ase | -0.245 |
Token | with |
Token ID | 351 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
goodbye | +0.078 |
plunge | +0.071 |
contain | +0.069 |
umb | +0.069 |
oured | +0.065 |
Neg logit contributions | |
ery | -0.056 |
Max | -0.056 |
Mic | -0.053 |
mice | -0.052 |
house | -0.050 |
Token | a |
Token ID | 257 |
Feature activation | -4.79e-02 |
Pos logit contributions | |
ery | +0.021 |
Mic | +0.020 |
Max | +0.020 |
iflower | +0.019 |
mice | +0.019 |
Neg logit contributions | |
goodbye | -0.021 |
contain | -0.017 |
plunge | -0.016 |
attention | -0.016 |
umb | -0.016 |
Token | big |
Token ID | 1263 |
Feature activation | -4.43e-02 |
Pos logit contributions | |
goodbye | +0.066 |
contain | +0.057 |
plunge | +0.055 |
umb | +0.053 |
attention | +0.052 |
Neg logit contributions | |
ery | -0.098 |
Mic | -0.097 |
Max | -0.097 |
mice | -0.093 |
Ted | -0.092 |
Token | smile |
Token ID | 8212 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
contain | +0.052 |
plunge | +0.051 |
goodbye | +0.050 |
ooters | +0.048 |
ounded | +0.047 |
Neg logit contributions | |
dough | -0.088 |
ery | -0.087 |
mice | -0.086 |
Max | -0.086 |
Mic | -0.085 |
Token | and |
Token ID | 290 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
Mic | +0.026 |
ery | +0.025 |
mice | +0.024 |
Max | +0.024 |
Bear | +0.024 |
Neg logit contributions | |
goodbye | -0.039 |
attention | -0.033 |
contain | -0.033 |
toward | -0.032 |
plunge | -0.031 |
Token | never |
Token ID | 1239 |
Feature activation | -3.25e-02 |
Pos logit contributions | |
ery | +0.037 |
Mic | +0.036 |
Max | +0.035 |
mice | +0.034 |
iflower | +0.033 |
Neg logit contributions | |
goodbye | -0.059 |
contain | -0.052 |
plunge | -0.051 |
umb | -0.049 |
attention | -0.049 |
Token | fell |
Token ID | 3214 |
Feature activation | -3.24e-02 |
Pos logit contributions | |
goodbye | +0.067 |
contain | +0.059 |
plunge | +0.058 |
umb | +0.056 |
attention | +0.055 |
Neg logit contributions | |
Mic | -0.046 |
ery | -0.046 |
Max | -0.045 |
mice | -0.044 |
Ted | -0.042 |
Token | off |
Token ID | 572 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
goodbye | +0.058 |
contain | +0.055 |
plunge | +0.052 |
eal | +0.050 |
fy | +0.049 |
Neg logit contributions | |
Max | -0.051 |
ery | -0.050 |
mice | -0.049 |
ters | -0.048 |
Mic | -0.048 |
Token | the |
Token ID | 262 |
Feature activation | -4.42e-02 |
Pos logit contributions | |
goodbye | +0.056 |
contain | +0.051 |
plunge | +0.050 |
eal | +0.047 |
umb | +0.047 |
Neg logit contributions | |
ery | -0.043 |
mice | -0.042 |
Max | -0.041 |
Mic | -0.040 |
Midnight | -0.039 |
Token | sc |
Token ID | 629 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
goodbye | +0.085 |
oured | +0.078 |
contain | +0.078 |
eal | +0.076 |
fy | +0.072 |
Neg logit contributions | |
Max | -0.066 |
Mic | -0.065 |
mice | -0.061 |
Brown | -0.059 |
ery | -0.058 |
Token | ooter |
Token ID | 25141 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
goodbye | +0.187 |
contain | +0.168 |
plunge | +0.168 |
anticipation | +0.162 |
hesitation | +0.150 |
Neg logit contributions | |
house | -0.231 |
Riding | -0.224 |
Max | -0.223 |
Mic | -0.222 |
her | -0.220 |
Token | again |
Token ID | 757 |
Feature activation | -9.11e-03 |
Pos logit contributions | |
goodbye | +0.077 |
plunge | +0.071 |
umb | +0.068 |
contain | +0.067 |
attention | +0.064 |
Neg logit contributions | |
Max | -0.065 |
ery | -0.063 |
Mic | -0.062 |
house | -0.061 |
mice | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
goodbye | +0.014 |
contain | +0.012 |
toward | +0.011 |
attention | +0.011 |
umb | +0.011 |
Neg logit contributions | |
ery | -0.018 |
Mic | -0.018 |
Max | -0.017 |
mice | -0.017 |
chini | -0.017 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +5.09e-02 |
Pos logit contributions | |
goodbye | +0.023 |
tour | +0.018 |
contain | +0.018 |
attention | +0.018 |
toward | +0.018 |
Neg logit contributions | |
ery | -0.017 |
iflower | -0.015 |
Mic | -0.015 |
mice | -0.015 |
chini | -0.014 |
Token | Once |
Token ID | 7454 |
Feature activation | -3.26e-02 |
Pos logit contributions | |
ery | +0.085 |
Mic | +0.083 |
iflower | +0.083 |
mice | +0.079 |
iders | +0.077 |
Neg logit contributions | |
goodbye | -0.091 |
contain | -0.073 |
toward | -0.073 |
plunge | -0.073 |
hug | -0.070 |
Token | helmet |
Token ID | 14335 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
goodbye | +0.004 |
contain | +0.004 |
plunge | +0.004 |
ogging | +0.004 |
oms | +0.004 |
Neg logit contributions | |
Max | -0.003 |
ery | -0.003 |
mice | -0.003 |
Mic | -0.003 |
Ted | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +8.34e-03 |
Pos logit contributions | |
goodbye | +0.043 |
contain | +0.038 |
plunge | +0.036 |
umb | +0.035 |
attention | +0.035 |
Neg logit contributions | |
ery | -0.041 |
Max | -0.040 |
Mic | -0.040 |
mice | -0.039 |
iflower | -0.037 |
Token | hopped |
Token ID | 45230 |
Feature activation | -5.79e-02 |
Pos logit contributions | |
Max | +0.012 |
ery | +0.012 |
Mic | +0.012 |
mice | +0.012 |
Ted | +0.011 |
Neg logit contributions | |
goodbye | -0.016 |
contain | -0.015 |
plunge | -0.014 |
ogging | -0.014 |
oured | -0.014 |
Token | on |
Token ID | 319 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
goodbye | +0.119 |
oured | +0.119 |
umb | +0.119 |
contain | +0.119 |
eal | +0.118 |
Neg logit contributions | |
Max | -0.060 |
ery | -0.060 |
hes | -0.049 |
You | -0.048 |
mice | -0.048 |
Token | his |
Token ID | 465 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
goodbye | +0.079 |
contain | +0.077 |
ounded | +0.076 |
eal | +0.075 |
fy | +0.075 |
Neg logit contributions | |
mice | -0.032 |
Max | -0.031 |
magnets | -0.029 |
Mic | -0.029 |
ery | -0.028 |
Token | sc |
Token ID | 629 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
goodbye | +0.023 |
oured | +0.023 |
oms | +0.022 |
contain | +0.022 |
plunge | +0.022 |
Neg logit contributions | |
Mic | -0.025 |
Max | -0.024 |
ery | -0.024 |
lf | -0.024 |
mice | -0.024 |
Token | ooter |
Token ID | 25141 |
Feature activation | -3.68e-02 |
Pos logit contributions | |
plunge | +0.181 |
goodbye | +0.169 |
umb | +0.161 |
hesitation | +0.154 |
contain | +0.154 |
Neg logit contributions | |
ery | -0.215 |
Riding | -0.210 |
her | -0.207 |
iny | -0.200 |
house | -0.200 |
Token | . |
Token ID | 13 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
goodbye | +0.075 |
plunge | +0.071 |
contain | +0.068 |
umb | +0.068 |
eal | +0.065 |
Neg logit contributions | |
Max | -0.048 |
ery | -0.047 |
Mic | -0.044 |
house | -0.043 |
mice | -0.042 |
Token | He |
Token ID | 679 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
goodbye | +0.031 |
contain | +0.025 |
toward | +0.025 |
plunge | +0.024 |
oured | +0.024 |
Neg logit contributions | |
ery | -0.027 |
Mic | -0.026 |
iflower | -0.026 |
Max | -0.025 |
mice | -0.024 |
Token | zoom |
Token ID | 19792 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
ery | +0.041 |
Mic | +0.039 |
Max | +0.039 |
mice | +0.038 |
Ted | +0.037 |
Neg logit contributions | |
goodbye | -0.055 |
contain | -0.048 |
plunge | -0.047 |
attention | -0.046 |
umb | -0.045 |
Token | ed |
Token ID | 276 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
goodbye | +0.057 |
contain | +0.052 |
plunge | +0.051 |
umb | +0.050 |
attention | +0.048 |
Neg logit contributions | |
Max | -0.048 |
ery | -0.048 |
mice | -0.044 |
Mic | -0.044 |
Ted | -0.044 |
Token | her |
Token ID | 607 |
Feature activation | -7.90e-03 |
Pos logit contributions | |
goodbye | +0.013 |
contain | +0.011 |
plunge | +0.011 |
umb | +0.011 |
attention | +0.011 |
Neg logit contributions | |
ery | -0.010 |
Mic | -0.010 |
Max | -0.009 |
iflower | -0.009 |
mice | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +7.51e-03 |
Pos logit contributions | |
goodbye | +0.014 |
contain | +0.012 |
plunge | +0.012 |
umb | +0.011 |
attention | +0.011 |
Neg logit contributions | |
ery | -0.014 |
Mic | -0.014 |
Max | -0.013 |
mice | -0.013 |
iflower | -0.013 |
Token | They |
Token ID | 1119 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
ery | +0.015 |
iflower | +0.014 |
Mic | +0.013 |
chini | +0.012 |
lf | +0.011 |
Neg logit contributions | |
goodbye | -0.015 |
toward | -0.012 |
umb | -0.011 |
contain | -0.011 |
farewell | -0.011 |
Token | rode |
Token ID | 22075 |
Feature activation | -4.66e-02 |
Pos logit contributions | |
goodbye | +0.037 |
plunge | +0.033 |
attention | +0.032 |
contain | +0.032 |
ogging | +0.032 |
Neg logit contributions | |
Max | -0.024 |
ery | -0.023 |
Mic | -0.023 |
mice | -0.023 |
Midnight | -0.022 |
Token | their |
Token ID | 511 |
Feature activation | -3.25e-02 |
Pos logit contributions | |
goodbye | +0.076 |
contain | +0.069 |
plunge | +0.068 |
oured | +0.067 |
umb | +0.064 |
Neg logit contributions | |
Max | -0.083 |
mice | -0.081 |
ery | -0.080 |
Mic | -0.078 |
Ted | -0.076 |
Token | sc |
Token ID | 629 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
goodbye | +0.056 |
contain | +0.052 |
oured | +0.051 |
plunge | +0.050 |
eal | +0.049 |
Neg logit contributions | |
mice | -0.053 |
Max | -0.053 |
Mic | -0.053 |
ery | -0.050 |
Midnight | -0.049 |
Token | ooters |
Token ID | 48316 |
Feature activation | -5.73e-02 |
Pos logit contributions | |
goodbye | +0.154 |
plunge | +0.149 |
contain | +0.136 |
anticipation | +0.135 |
icol | +0.131 |
Neg logit contributions | |
Max | -0.256 |
ery | -0.247 |
Mic | -0.245 |
house | -0.240 |
Next | -0.236 |
Token | to |
Token ID | 284 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
goodbye | +0.106 |
contain | +0.093 |
plunge | +0.092 |
umb | +0.089 |
oured | +0.088 |
Neg logit contributions | |
Max | -0.092 |
ery | -0.092 |
Mic | -0.089 |
mice | -0.086 |
Ted | -0.082 |
Token | the |
Token ID | 262 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
goodbye | +0.055 |
contain | +0.049 |
plunge | +0.047 |
umb | +0.046 |
attention | +0.046 |
Neg logit contributions | |
ery | -0.034 |
Mic | -0.033 |
Max | -0.033 |
mice | -0.032 |
Ted | -0.030 |
Token | other |
Token ID | 584 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
goodbye | +0.052 |
contain | +0.047 |
fy | +0.044 |
oured | +0.044 |
plunge | +0.044 |
Neg logit contributions | |
ery | -0.038 |
mice | -0.038 |
Max | -0.038 |
Mic | -0.037 |
iflower | -0.035 |
Token | side |
Token ID | 1735 |
Feature activation | -4.30e-02 |
Pos logit contributions | |
goodbye | +0.052 |
contain | +0.045 |
plunge | +0.043 |
umb | +0.042 |
attention | +0.042 |
Neg logit contributions | |
ery | -0.042 |
Mic | -0.041 |
Max | -0.040 |
mice | -0.038 |
iflower | -0.038 |
Token | her |
Token ID | 607 |
Feature activation | -5.19e-03 |
Pos logit contributions | |
goodbye | +0.043 |
contain | +0.037 |
plunge | +0.035 |
umb | +0.035 |
attention | +0.035 |
Neg logit contributions | |
ery | -0.032 |
Mic | -0.031 |
iflower | -0.029 |
Max | -0.029 |
mice | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | +7.14e-03 |
Pos logit contributions | |
goodbye | +0.010 |
plunge | +0.009 |
contain | +0.009 |
oured | +0.009 |
fy | +0.009 |
Neg logit contributions | |
ery | -0.007 |
mice | -0.007 |
Max | -0.007 |
Mic | -0.007 |
ters | -0.007 |
Token | They |
Token ID | 1119 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
iflower | +0.016 |
ery | +0.015 |
Mic | +0.014 |
if | +0.013 |
lf | +0.013 |
Neg logit contributions | |
goodbye | -0.012 |
toward | -0.009 |
farewell | -0.009 |
umb | -0.009 |
hello | -0.009 |
Token | rode |
Token ID | 22075 |
Feature activation | -4.58e-02 |
Pos logit contributions | |
goodbye | +0.038 |
plunge | +0.033 |
contain | +0.033 |
attention | +0.032 |
toward | +0.032 |
Neg logit contributions | |
ery | -0.025 |
Max | -0.025 |
Mic | -0.024 |
mice | -0.024 |
Midnight | -0.023 |
Token | their |
Token ID | 511 |
Feature activation | -3.03e-02 |
Pos logit contributions | |
goodbye | +0.080 |
contain | +0.073 |
plunge | +0.072 |
oured | +0.071 |
umb | +0.068 |
Neg logit contributions | |
Max | -0.077 |
mice | -0.074 |
ery | -0.073 |
Mic | -0.071 |
Ted | -0.069 |
Token | sc |
Token ID | 629 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
goodbye | +0.061 |
contain | +0.060 |
oured | +0.059 |
fy | +0.057 |
eal | +0.055 |
Neg logit contributions | |
Max | -0.040 |
mice | -0.038 |
Mic | -0.036 |
magnets | -0.035 |
animals | -0.032 |
Token | ooters |
Token ID | 48316 |
Feature activation | -5.25e-02 |
Pos logit contributions | |
goodbye | +0.157 |
plunge | +0.155 |
anticipation | +0.140 |
contain | +0.138 |
hesitation | +0.135 |
Neg logit contributions | |
Max | -0.258 |
ery | -0.243 |
house | -0.242 |
Mic | -0.242 |
hes | -0.240 |
Token | away |
Token ID | 1497 |
Feature activation | -4.06e-03 |
Pos logit contributions | |
goodbye | +0.097 |
contain | +0.085 |
plunge | +0.085 |
umb | +0.081 |
oured | +0.080 |
Neg logit contributions | |
Max | -0.086 |
ery | -0.084 |
Mic | -0.081 |
mice | -0.079 |
Ted | -0.076 |
Token | from |
Token ID | 422 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
goodbye | +0.008 |
contain | +0.007 |
plunge | +0.007 |
umb | +0.006 |
oured | +0.006 |
Neg logit contributions | |
ery | -0.006 |
Max | -0.006 |
mice | -0.006 |
Mic | -0.006 |
Ted | -0.006 |
Token | her |
Token ID | 607 |
Feature activation | -3.84e-03 |
Pos logit contributions | |
ery | +0.020 |
Mic | +0.019 |
iflower | +0.018 |
Max | +0.018 |
Bear | +0.018 |
Neg logit contributions | |
goodbye | -0.025 |
contain | -0.021 |
plunge | -0.021 |
attention | -0.020 |
umb | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +5.18e-03 |
Pos logit contributions | |
goodbye | +0.008 |
contain | +0.007 |
plunge | +0.007 |
oured | +0.007 |
fy | +0.007 |
Neg logit contributions | |
mice | -0.005 |
Max | -0.005 |
ery | -0.005 |
Mic | -0.005 |
Midnight | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
goodbye | +0.049 |
umb | +0.044 |
plunge | +0.043 |
contain | +0.043 |
oured | +0.042 |
Neg logit contributions | |
ery | -0.038 |
Max | -0.037 |
mice | -0.035 |
ters | -0.034 |
Mic | -0.034 |
Token | He |
Token ID | 679 |
Feature activation | +3.01e-02 |
Pos logit contributions | |
goodbye | +0.025 |
contain | +0.021 |
plunge | +0.021 |
toward | +0.021 |
attention | +0.020 |
Neg logit contributions | |
ery | -0.019 |
Mic | -0.018 |
Max | -0.018 |
iflower | -0.018 |
mice | -0.017 |
Token | hopped |
Token ID | 45230 |
Feature activation | -6.15e-02 |
Pos logit contributions | |
ery | +0.044 |
Max | +0.043 |
Mic | +0.041 |
Ted | +0.040 |
mice | +0.040 |
Neg logit contributions | |
goodbye | -0.063 |
contain | -0.055 |
attention | -0.053 |
plunge | -0.053 |
umb | -0.051 |
Token | onto |
Token ID | 4291 |
Feature activation | -4.54e-02 |
Pos logit contributions | |
goodbye | +0.122 |
umb | +0.121 |
contain | +0.118 |
des | +0.116 |
fy | +0.114 |
Neg logit contributions | |
Max | -0.082 |
ery | -0.071 |
Ted | -0.069 |
mice | -0.064 |
You | -0.064 |
Token | his |
Token ID | 465 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
goodbye | +0.103 |
contain | +0.091 |
plunge | +0.088 |
umb | +0.087 |
fy | +0.087 |
Neg logit contributions | |
ery | -0.056 |
Mic | -0.055 |
Max | -0.055 |
mice | -0.053 |
Ted | -0.051 |
Token | sc |
Token ID | 629 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
goodbye | +0.030 |
inspection | +0.028 |
contain | +0.027 |
des | +0.027 |
oured | +0.027 |
Neg logit contributions | |
Max | -0.024 |
ery | -0.024 |
mice | -0.022 |
lf | -0.022 |
iflower | -0.022 |
Token | ooter |
Token ID | 25141 |
Feature activation | -3.19e-02 |
Pos logit contributions | |
plunge | +0.158 |
goodbye | +0.154 |
umb | +0.151 |
hesitation | +0.145 |
inspection | +0.140 |
Neg logit contributions | |
ery | -0.241 |
house | -0.233 |
Max | -0.231 |
her | -0.230 |
iny | -0.228 |
Token | and |
Token ID | 290 |
Feature activation | +4.19e-03 |
Pos logit contributions | |
goodbye | +0.068 |
plunge | +0.063 |
umb | +0.063 |
contain | +0.061 |
eal | +0.059 |
Neg logit contributions | |
Max | -0.039 |
ery | -0.037 |
Mic | -0.036 |
mice | -0.034 |
house | -0.034 |
Token | zoom |
Token ID | 19792 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
Max | +0.006 |
ery | +0.006 |
Ted | +0.006 |
mice | +0.006 |
Mic | +0.005 |
Neg logit contributions | |
goodbye | -0.008 |
contain | -0.007 |
ogging | -0.007 |
plunge | -0.007 |
attention | -0.007 |
Token | ed |
Token ID | 276 |
Feature activation | -4.51e-03 |
Pos logit contributions | |
goodbye | +0.054 |
contain | +0.050 |
plunge | +0.049 |
umb | +0.049 |
ogging | +0.047 |
Neg logit contributions | |
Max | -0.044 |
ery | -0.043 |
Ted | -0.040 |
mice | -0.039 |
ters | -0.039 |
Token | all |
Token ID | 477 |
Feature activation | -6.28e-02 |
Pos logit contributions | |
contain | +0.008 |
goodbye | +0.008 |
oured | +0.008 |
fy | +0.008 |
plunge | +0.008 |
Neg logit contributions | |
Max | -0.006 |
mice | -0.006 |
Ted | -0.006 |
ery | -0.006 |
Mic | -0.005 |
Token | helmet |
Token ID | 14335 |
Feature activation | -4.12e-02 |
Pos logit contributions | |
goodbye | +0.013 |
contain | +0.013 |
fy | +0.012 |
toward | +0.011 |
plunge | +0.011 |
Neg logit contributions | |
ery | -0.013 |
Max | -0.012 |
Mic | -0.012 |
magnets | -0.012 |
Next | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | -5.79e-03 |
Pos logit contributions | |
goodbye | +0.081 |
contain | +0.071 |
plunge | +0.069 |
umb | +0.068 |
attention | +0.066 |
Neg logit contributions | |
ery | -0.063 |
Mic | -0.061 |
Max | -0.061 |
mice | -0.059 |
iflower | -0.057 |
Token | jumped |
Token ID | 11687 |
Feature activation | -4.42e-02 |
Pos logit contributions | |
goodbye | +0.011 |
contain | +0.009 |
plunge | +0.009 |
umb | +0.009 |
attention | +0.009 |
Neg logit contributions | |
ery | -0.009 |
Mic | -0.009 |
Max | -0.009 |
mice | -0.009 |
Ted | -0.009 |
Token | on |
Token ID | 319 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
goodbye | +0.076 |
contain | +0.072 |
plunge | +0.067 |
fy | +0.066 |
umb | +0.065 |
Neg logit contributions | |
ery | -0.072 |
Max | -0.071 |
Mic | -0.066 |
mice | -0.066 |
ters | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | -3.76e-02 |
Pos logit contributions | |
goodbye | +0.057 |
contain | +0.053 |
plunge | +0.051 |
fy | +0.050 |
ounded | +0.050 |
Neg logit contributions | |
Max | -0.029 |
ery | -0.028 |
Mic | -0.027 |
mice | -0.027 |
Ted | -0.026 |
Token | sc |
Token ID | 629 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
goodbye | +0.067 |
ogging | +0.062 |
fy | +0.061 |
inspection | +0.060 |
contain | +0.060 |
Neg logit contributions | |
Max | -0.057 |
ery | -0.055 |
Mic | -0.054 |
mice | -0.052 |
dough | -0.051 |
Token | ooter |
Token ID | 25141 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
plunge | +0.177 |
goodbye | +0.177 |
contain | +0.163 |
hesitation | +0.159 |
umb | +0.159 |
Neg logit contributions | |
her | -0.209 |
Riding | -0.207 |
ery | -0.206 |
Max | -0.205 |
ep | -0.200 |
Token | and |
Token ID | 290 |
Feature activation | +1.11e-03 |
Pos logit contributions | |
goodbye | +0.067 |
plunge | +0.060 |
contain | +0.059 |
umb | +0.059 |
attention | +0.056 |
Neg logit contributions | |
Max | -0.044 |
ery | -0.044 |
Mic | -0.041 |
mice | -0.040 |
ters | -0.039 |
Token | started |
Token ID | 2067 |
Feature activation | -9.81e-03 |
Pos logit contributions | |
Max | +0.002 |
ery | +0.002 |
Mic | +0.002 |
Ted | +0.002 |
mice | +0.002 |
Neg logit contributions | |
goodbye | -0.002 |
contain | -0.002 |
plunge | -0.002 |
umb | -0.002 |
attention | -0.002 |
Token | to |
Token ID | 284 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
goodbye | +0.016 |
contain | +0.015 |
fy | +0.014 |
plunge | +0.014 |
ounded | +0.013 |
Neg logit contributions | |
Max | -0.016 |
mice | -0.016 |
ery | -0.015 |
hes | -0.015 |
Mic | -0.015 |
Token | roll |
Token ID | 4836 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
goodbye | +0.020 |
contain | +0.016 |
fy | +0.016 |
oured | +0.016 |
plunge | +0.016 |
Neg logit contributions | |
Max | -0.022 |
ery | -0.022 |
Mic | -0.021 |
Ted | -0.020 |
mice | -0.020 |
Token | heard |
Token ID | 2982 |
Feature activation | -3.16e-02 |
Pos logit contributions | |
goodbye | +0.008 |
contain | +0.007 |
plunge | +0.007 |
oured | +0.007 |
toward | +0.007 |
Neg logit contributions | |
Mic | -0.008 |
ery | -0.008 |
Max | -0.007 |
mice | -0.007 |
Ted | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | -5.66e-02 |
Pos logit contributions | |
goodbye | +0.071 |
contain | +0.063 |
plunge | +0.061 |
umb | +0.060 |
attention | +0.059 |
Neg logit contributions | |
ery | -0.040 |
Mic | -0.039 |
Max | -0.038 |
mice | -0.037 |
Ted | -0.036 |
Token | knock |
Token ID | 10643 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
goodbye | +0.116 |
oured | +0.102 |
contain | +0.101 |
plunge | +0.097 |
umb | +0.097 |
Neg logit contributions | |
Mic | -0.080 |
Max | -0.080 |
mice | -0.080 |
ery | -0.078 |
Ted | -0.077 |
Token | at |
Token ID | 379 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
goodbye | +0.066 |
contain | +0.059 |
plunge | +0.057 |
oured | +0.055 |
umb | +0.055 |
Neg logit contributions | |
ery | -0.050 |
Max | -0.048 |
mice | -0.048 |
Mic | -0.046 |
Next | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | -4.93e-02 |
Pos logit contributions | |
goodbye | +0.025 |
contain | +0.022 |
plunge | +0.021 |
attention | +0.021 |
umb | +0.021 |
Neg logit contributions | |
ery | -0.018 |
Mic | -0.017 |
Max | -0.017 |
mice | -0.016 |
iflower | -0.016 |
Token | front |
Token ID | 2166 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
goodbye | +0.081 |
contain | +0.070 |
fy | +0.067 |
plunge | +0.066 |
oured | +0.065 |
Neg logit contributions | |
mice | -0.088 |
ery | -0.088 |
Max | -0.086 |
dough | -0.084 |
chini | -0.083 |
Token | door |
Token ID | 3420 |
Feature activation | -5.30e-02 |
Pos logit contributions | |
goodbye | +0.219 |
plunge | +0.197 |
oured | +0.187 |
fy | +0.180 |
contain | +0.180 |
Neg logit contributions | |
mice | -0.208 |
Max | -0.206 |
ery | -0.205 |
dough | -0.199 |
house | -0.197 |
Token | . |
Token ID | 13 |
Feature activation | -2.42e-02 |
Pos logit contributions | |
goodbye | +0.119 |
plunge | +0.110 |
farewell | +0.105 |
icol | +0.105 |
umb | +0.104 |
Neg logit contributions | |
Max | -0.055 |
house | -0.054 |
ery | -0.052 |
lf | -0.050 |
ters | -0.049 |
Token |
|
Token ID | 198 |
Feature activation | +2.04e-03 |
Pos logit contributions | |
goodbye | +0.047 |
contain | +0.037 |
toward | +0.036 |
hi | +0.036 |
farewell | +0.035 |
Neg logit contributions | |
ery | -0.047 |
iflower | -0.042 |
Mic | -0.040 |
mice | -0.039 |
chini | -0.038 |
Token |
|
Token ID | 198 |
Feature activation | +1.70e-03 |
Pos logit contributions | |
ery | +0.003 |
mice | +0.003 |
iflower | +0.002 |
Mic | +0.002 |
ters | +0.002 |
Neg logit contributions | |
goodbye | -0.005 |
contain | -0.004 |
toward | -0.004 |
farewell | -0.004 |
plunge | -0.004 |
Token | His |
Token ID | 6653 |
Feature activation | +7.17e-03 |
Pos logit contributions | |
ery | +0.002 |
mice | +0.002 |
iflower | +0.002 |
ters | +0.002 |
Midnight | +0.002 |
Neg logit contributions | |
goodbye | -0.004 |
contain | -0.003 |
toward | -0.003 |
farewell | -0.003 |
fy | -0.003 |
Token | The |
Token ID | 464 |
Feature activation | +5.78e-03 |
Pos logit contributions | |
ery | +0.010 |
iflower | +0.009 |
mice | +0.009 |
magnets | +0.008 |
Mic | +0.008 |
Neg logit contributions | |
goodbye | -0.016 |
contain | -0.015 |
fy | -0.014 |
toward | -0.014 |
umb | -0.014 |
Token | girl |
Token ID | 2576 |
Feature activation | +2.44e-02 |
Pos logit contributions | |
ery | +0.009 |
Mic | +0.009 |
Max | +0.009 |
mice | +0.008 |
iflower | +0.008 |
Neg logit contributions | |
goodbye | -0.012 |
contain | -0.010 |
plunge | -0.010 |
umb | -0.009 |
attention | -0.009 |
Token | put |
Token ID | 1234 |
Feature activation | +7.18e-03 |
Pos logit contributions | |
ery | +0.040 |
Mic | +0.039 |
Max | +0.038 |
mice | +0.037 |
iflower | +0.037 |
Neg logit contributions | |
goodbye | -0.047 |
contain | -0.040 |
plunge | -0.038 |
umb | -0.038 |
toward | -0.037 |
Token | on |
Token ID | 319 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
Mic | +0.013 |
ery | +0.012 |
Max | +0.012 |
iflower | +0.012 |
oyd | +0.012 |
Neg logit contributions | |
goodbye | -0.013 |
attention | -0.010 |
umb | -0.010 |
closer | -0.010 |
plunge | -0.010 |
Token | her |
Token ID | 607 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
goodbye | +0.031 |
contain | +0.026 |
umb | +0.026 |
plunge | +0.025 |
attention | +0.025 |
Neg logit contributions | |
ery | -0.023 |
Mic | -0.022 |
Max | -0.021 |
iflower | -0.020 |
Bear | -0.020 |
Token | sc |
Token ID | 629 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
goodbye | +0.028 |
contain | +0.026 |
fy | +0.025 |
toward | +0.025 |
plunge | +0.025 |
Neg logit contributions | |
ery | -0.024 |
Mic | -0.023 |
Max | -0.023 |
iflower | -0.023 |
mice | -0.023 |
Token | uba |
Token ID | 22013 |
Feature activation | -3.33e-02 |
Pos logit contributions | |
goodbye | +0.169 |
plunge | +0.159 |
umb | +0.153 |
contain | +0.147 |
further | +0.146 |
Neg logit contributions | |
ery | -0.246 |
iflower | -0.224 |
Mic | -0.223 |
Next | -0.223 |
hes | -0.221 |
Token | gear |
Token ID | 7733 |
Feature activation | -3.99e-02 |
Pos logit contributions | |
goodbye | +0.060 |
contain | +0.053 |
plunge | +0.051 |
attention | +0.049 |
umb | +0.049 |
Neg logit contributions | |
ery | -0.055 |
Mic | -0.054 |
Max | -0.053 |
mice | -0.053 |
iflower | -0.050 |
Token | and |
Token ID | 290 |
Feature activation | -3.35e-03 |
Pos logit contributions | |
goodbye | +0.076 |
contain | +0.065 |
plunge | +0.063 |
umb | +0.061 |
attention | +0.060 |
Neg logit contributions | |
ery | -0.065 |
Mic | -0.064 |
Max | -0.063 |
mice | -0.061 |
iflower | -0.060 |
Token | jumped |
Token ID | 11687 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
goodbye | +0.006 |
contain | +0.005 |
plunge | +0.005 |
umb | +0.005 |
attention | +0.005 |
Neg logit contributions | |
ery | -0.006 |
Mic | -0.005 |
Max | -0.005 |
mice | -0.005 |
iflower | -0.005 |
Token | into |
Token ID | 656 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
goodbye | +0.096 |
contain | +0.090 |
fy | +0.086 |
plunge | +0.086 |
oured | +0.086 |
Neg logit contributions | |
ery | -0.076 |
mice | -0.071 |
Max | -0.069 |
ters | -0.066 |
chini | -0.066 |
Token | to |
Token ID | 284 |
Feature activation | -4.13e-03 |
Pos logit contributions | |
ery | +0.025 |
Mic | +0.024 |
Max | +0.024 |
mice | +0.024 |
iflower | +0.023 |
Neg logit contributions | |
goodbye | -0.024 |
contain | -0.020 |
plunge | -0.020 |
umb | -0.019 |
attention | -0.019 |
Token | get |
Token ID | 651 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
goodbye | +0.006 |
contain | +0.005 |
plunge | +0.005 |
umb | +0.005 |
attention | +0.005 |
Neg logit contributions | |
ery | -0.008 |
Mic | -0.008 |
Max | -0.008 |
mice | -0.008 |
iflower | -0.008 |
Token | back |
Token ID | 736 |
Feature activation | -7.81e-05 |
Pos logit contributions | |
goodbye | +0.026 |
contain | +0.024 |
plunge | +0.022 |
tour | +0.022 |
ounded | +0.022 |
Neg logit contributions | |
ery | -0.016 |
mice | -0.015 |
Max | -0.015 |
Midnight | -0.015 |
Mic | -0.015 |
Token | on |
Token ID | 319 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
goodbye | +0.000 |
contain | +0.000 |
plunge | +0.000 |
fy | +0.000 |
eal | +0.000 |
Neg logit contributions | |
ery | -0.000 |
mice | -0.000 |
Max | -0.000 |
Mic | -0.000 |
Midnight | -0.000 |
Token | the |
Token ID | 262 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
goodbye | +0.065 |
contain | +0.059 |
plunge | +0.057 |
ounded | +0.055 |
fy | +0.055 |
Neg logit contributions | |
Max | -0.035 |
Mic | -0.035 |
mice | -0.034 |
ery | -0.034 |
Midnight | -0.033 |
Token | sc |
Token ID | 629 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
oured | +0.059 |
goodbye | +0.058 |
eal | +0.057 |
fy | +0.057 |
toward | +0.056 |
Neg logit contributions | |
Max | -0.049 |
Mic | -0.048 |
lf | -0.047 |
dough | -0.045 |
Midnight | -0.044 |
Token | ooter |
Token ID | 25141 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
goodbye | +0.181 |
plunge | +0.178 |
inspection | +0.170 |
anticipation | +0.170 |
contain | +0.165 |
Neg logit contributions | |
house | -0.214 |
ase | -0.202 |
her | -0.199 |
ery | -0.195 |
Riding | -0.193 |
Token | . |
Token ID | 13 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
goodbye | +0.066 |
contain | +0.057 |
plunge | +0.057 |
umb | +0.054 |
attention | +0.054 |
Neg logit contributions | |
ery | -0.055 |
Max | -0.053 |
Mic | -0.053 |
mice | -0.051 |
iflower | -0.050 |
Token | She |
Token ID | 1375 |
Feature activation | +7.15e-03 |
Pos logit contributions | |
goodbye | +0.043 |
contain | +0.037 |
plunge | +0.036 |
umb | +0.036 |
toward | +0.036 |
Neg logit contributions | |
ery | -0.028 |
Mic | -0.026 |
Max | -0.025 |
iflower | -0.025 |
mice | -0.025 |
Token | said |
Token ID | 531 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
ery | +0.012 |
Mic | +0.012 |
Max | +0.012 |
mice | +0.011 |
iflower | +0.011 |
Neg logit contributions | |
goodbye | -0.013 |
contain | -0.011 |
plunge | -0.011 |
umb | -0.011 |
oured | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -8.91e-03 |
Pos logit contributions | |
contain | +0.045 |
goodbye | +0.043 |
plunge | +0.041 |
tour | +0.040 |
toward | +0.040 |
Neg logit contributions | |
Midnight | -0.056 |
ery | -0.056 |
mice | -0.056 |
Max | -0.055 |
Mic | -0.054 |
Token | , |
Token ID | 11 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
goodbye | +0.024 |
contain | +0.020 |
plunge | +0.019 |
umb | +0.018 |
attention | +0.018 |
Neg logit contributions | |
ery | -0.034 |
Mic | -0.034 |
Max | -0.033 |
mice | -0.033 |
Ted | -0.032 |
Token | there |
Token ID | 612 |
Feature activation | -2.57e-03 |
Pos logit contributions | |
goodbye | +0.020 |
closer | +0.017 |
contain | +0.017 |
plunge | +0.017 |
toward | +0.016 |
Neg logit contributions | |
Mic | -0.020 |
mice | -0.018 |
ery | -0.018 |
Doctor | -0.018 |
chini | -0.018 |
Token | was |
Token ID | 373 |
Feature activation | +3.17e-02 |
Pos logit contributions | |
goodbye | +0.004 |
plunge | +0.003 |
contain | +0.003 |
wave | +0.003 |
tour | +0.003 |
Neg logit contributions | |
Mic | -0.005 |
mice | -0.005 |
ery | -0.005 |
Fox | -0.004 |
iflower | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | +5.92e-03 |
Pos logit contributions | |
ery | +0.053 |
Mic | +0.051 |
Mouse | +0.051 |
iders | +0.051 |
ters | +0.051 |
Neg logit contributions | |
goodbye | -0.056 |
plunge | -0.044 |
contain | -0.044 |
umb | -0.043 |
tour | -0.042 |
Token | big |
Token ID | 1263 |
Feature activation | -5.93e-03 |
Pos logit contributions | |
ery | +0.010 |
Mic | +0.009 |
ters | +0.009 |
Max | +0.009 |
chini | +0.009 |
Neg logit contributions | |
goodbye | -0.011 |
wave | -0.009 |
tour | -0.009 |
contain | -0.008 |
plunge | -0.008 |
Token | dragon |
Token ID | 10441 |
Feature activation | +5.40e-02 |
Pos logit contributions | |
goodbye | +0.012 |
contain | +0.010 |
plunge | +0.010 |
umb | +0.010 |
attention | +0.009 |
Neg logit contributions | |
ery | -0.009 |
Mic | -0.009 |
Max | -0.009 |
mice | -0.009 |
Ted | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -1.57e-04 |
Pos logit contributions | |
Mic | +0.105 |
ery | +0.102 |
Max | +0.102 |
Charlie | +0.097 |
Next | +0.096 |
Neg logit contributions | |
goodbye | -0.086 |
contain | -0.069 |
wave | -0.068 |
tour | -0.066 |
toward | -0.065 |
Token | The |
Token ID | 383 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
goodbye | +0.000 |
contain | +0.000 |
hello | +0.000 |
hi | +0.000 |
hug | +0.000 |
Neg logit contributions | |
ery | -0.000 |
iflower | -0.000 |
Mic | -0.000 |
ters | -0.000 |
iders | -0.000 |
Token | dragon |
Token ID | 10441 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
Mic | +0.018 |
Max | +0.018 |
iders | +0.017 |
Ted | +0.017 |
Midnight | +0.017 |
Neg logit contributions | |
goodbye | -0.021 |
wave | -0.017 |
tour | -0.016 |
plunge | -0.016 |
farewell | -0.015 |
Token | loved |
Token ID | 6151 |
Feature activation | +3.31e-02 |
Pos logit contributions | |
Mic | +0.060 |
Max | +0.056 |
Midnight | +0.054 |
ters | +0.054 |
Charlie | +0.053 |
Neg logit contributions | |
goodbye | -0.064 |
wave | -0.054 |
tour | -0.053 |
contain | -0.051 |
hug | -0.050 |
Token | to |
Token ID | 284 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
ery | +0.086 |
Max | +0.086 |
Mic | +0.085 |
Ted | +0.081 |
Midnight | +0.081 |
Neg logit contributions | |
goodbye | -0.034 |
contain | -0.021 |
attention | -0.021 |
plunge | -0.020 |
tour | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
goodbye | +0.019 |
contain | +0.014 |
plunge | +0.014 |
umb | +0.013 |
toward | +0.013 |
Neg logit contributions | |
ery | -0.035 |
Mic | -0.034 |
mice | -0.034 |
Max | -0.034 |
iflower | -0.033 |
Token | there |
Token ID | 612 |
Feature activation | +2.39e-03 |
Pos logit contributions | |
goodbye | +0.022 |
closer | +0.021 |
toward | +0.018 |
ounded | +0.018 |
plunge | +0.018 |
Neg logit contributions | |
mice | -0.027 |
Mic | -0.026 |
Doctor | -0.026 |
chini | -0.025 |
ities | -0.025 |
Token | was |
Token ID | 373 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
Mic | +0.005 |
mice | +0.005 |
iflower | +0.004 |
ery | +0.004 |
magnets | +0.004 |
Neg logit contributions | |
goodbye | -0.004 |
plunge | -0.003 |
inspection | -0.003 |
contain | -0.003 |
wave | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | +9.99e-03 |
Pos logit contributions | |
iders | +0.062 |
ery | +0.061 |
iflower | +0.060 |
Mouse | +0.060 |
ters | +0.060 |
Neg logit contributions | |
goodbye | -0.061 |
plunge | -0.045 |
farewell | -0.045 |
contain | -0.045 |
umb | -0.044 |
Token | little |
Token ID | 1310 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
ery | +0.017 |
Mic | +0.015 |
Max | +0.015 |
Next | +0.015 |
mice | +0.015 |
Neg logit contributions | |
goodbye | -0.019 |
wave | -0.015 |
tour | -0.015 |
contain | -0.015 |
plunge | -0.015 |
Token | girl |
Token ID | 2576 |
Feature activation | +5.52e-02 |
Pos logit contributions | |
ery | +0.047 |
Mic | +0.046 |
Max | +0.045 |
mice | +0.044 |
iflower | +0.042 |
Neg logit contributions | |
goodbye | -0.075 |
contain | -0.066 |
plunge | -0.064 |
umb | -0.063 |
attention | -0.062 |
Token | named |
Token ID | 3706 |
Feature activation | -2.27e-03 |
Pos logit contributions | |
ery | +0.091 |
Mic | +0.089 |
Max | +0.087 |
mice | +0.084 |
iflower | +0.082 |
Neg logit contributions | |
goodbye | -0.105 |
contain | -0.089 |
plunge | -0.085 |
umb | -0.084 |
toward | -0.083 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.01e-02 |
Pos logit contributions | |
goodbye | +0.006 |
contain | +0.005 |
plunge | +0.005 |
umb | +0.005 |
attention | +0.005 |
Neg logit contributions | |
ery | -0.002 |
Mic | -0.002 |
Max | -0.002 |
mice | -0.002 |
iflower | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -3.27e-04 |
Pos logit contributions | |
Mic | +0.061 |
ery | +0.060 |
mice | +0.060 |
chini | +0.059 |
Max | +0.058 |
Neg logit contributions | |
goodbye | -0.042 |
contain | -0.034 |
toward | -0.032 |
attention | -0.031 |
hug | -0.031 |
Token | She |
Token ID | 1375 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
goodbye | +0.001 |
contain | +0.000 |
fy | +0.000 |
toward | +0.000 |
umb | +0.000 |
Neg logit contributions | |
ery | -0.001 |
iflower | -0.001 |
chini | -0.001 |
Mic | -0.001 |
oyd | -0.001 |
Token | loved |
Token ID | 6151 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
Mic | +0.041 |
ters | +0.041 |
mice | +0.041 |
ery | +0.040 |
chini | +0.038 |
Neg logit contributions | |
goodbye | -0.045 |
contain | -0.036 |
wave | -0.035 |
tour | -0.035 |
hug | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | +3.43e-03 |
Pos logit contributions | |
goodbye | +0.057 |
contain | +0.048 |
plunge | +0.046 |
umb | +0.045 |
attention | +0.045 |
Neg logit contributions | |
ery | -0.056 |
Mic | -0.056 |
Max | -0.055 |
mice | -0.053 |
iflower | -0.052 |
Token | They |
Token ID | 1119 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
ery | +0.007 |
iflower | +0.007 |
Mic | +0.006 |
web | +0.005 |
lf | +0.005 |
Neg logit contributions | |
goodbye | -0.006 |
umb | -0.005 |
toward | -0.005 |
hi | -0.005 |
contain | -0.005 |
Token | had |
Token ID | 550 |
Feature activation | +6.40e-03 |
Pos logit contributions | |
goodbye | +0.029 |
contain | +0.025 |
plunge | +0.024 |
umb | +0.023 |
attention | +0.023 |
Neg logit contributions | |
ery | -0.026 |
Mic | -0.026 |
Max | -0.025 |
mice | -0.025 |
iflower | -0.024 |
Token | a |
Token ID | 257 |
Feature activation | -9.78e-03 |
Pos logit contributions | |
ery | +0.010 |
Max | +0.010 |
Mic | +0.010 |
mice | +0.010 |
Ted | +0.009 |
Neg logit contributions | |
goodbye | -0.012 |
contain | -0.011 |
plunge | -0.010 |
toward | -0.010 |
oured | -0.010 |
Token | lot |
Token ID | 1256 |
Feature activation | +2.17e-03 |
Pos logit contributions | |
goodbye | +0.017 |
contain | +0.015 |
plunge | +0.014 |
toward | +0.014 |
oured | +0.014 |
Neg logit contributions | |
ery | -0.017 |
Mic | -0.017 |
Max | -0.017 |
mice | -0.016 |
Ted | -0.016 |
Token | of |
Token ID | 286 |
Feature activation | +6.34e-02 |
Pos logit contributions | |
ery | +0.004 |
Max | +0.004 |
mice | +0.004 |
Mic | +0.004 |
Ted | +0.003 |
Neg logit contributions | |
goodbye | -0.004 |
contain | -0.003 |
plunge | -0.003 |
umb | -0.003 |
oured | -0.003 |
Token | fun |
Token ID | 1257 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
ery | +0.103 |
mice | +0.099 |
Max | +0.098 |
Mic | +0.096 |
dough | +0.092 |
Neg logit contributions | |
goodbye | -0.125 |
contain | -0.109 |
oured | -0.104 |
plunge | -0.102 |
toward | -0.101 |
Token | with |
Token ID | 351 |
Feature activation | +4.71e-02 |
Pos logit contributions | |
Mic | +0.030 |
ery | +0.029 |
Max | +0.028 |
mice | +0.027 |
Bear | +0.027 |
Neg logit contributions | |
goodbye | -0.029 |
contain | -0.025 |
plunge | -0.024 |
umb | -0.024 |
toward | -0.023 |
Token | their |
Token ID | 511 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
Mic | +0.093 |
ery | +0.088 |
Bear | +0.083 |
iflower | +0.082 |
Max | +0.082 |
Neg logit contributions | |
goodbye | -0.081 |
umb | -0.069 |
attention | -0.066 |
tickets | -0.063 |
toward | -0.063 |
Token | hoop |
Token ID | 44792 |
Feature activation | -1.84e-03 |
Pos logit contributions | |
ery | +0.021 |
Mic | +0.021 |
Max | +0.020 |
mice | +0.019 |
Bear | +0.019 |
Neg logit contributions | |
goodbye | -0.027 |
contain | -0.023 |
plunge | -0.023 |
umb | -0.022 |
attention | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
goodbye | +0.003 |
contain | +0.003 |
plunge | +0.003 |
umb | +0.003 |
attention | +0.003 |
Neg logit contributions | |
ery | -0.003 |
Mic | -0.003 |
Max | -0.003 |
mice | -0.003 |
iflower | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -8.57e-03 |
Pos logit contributions | |
ery | +0.046 |
Mic | +0.045 |
Max | +0.044 |
mice | +0.043 |
iflower | +0.042 |
Neg logit contributions | |
goodbye | -0.045 |
contain | -0.038 |
plunge | -0.036 |
umb | -0.035 |
attention | -0.035 |
Token | |
Token ID | 220 |
Feature activation | -2.96e-02 |
Pos logit contributions | |
goodbye | +0.016 |
contain | +0.013 |
toward | +0.013 |
umb | +0.013 |
attention | +0.012 |
Neg logit contributions | |
ery | -0.016 |
iflower | -0.015 |
Mic | -0.014 |
chini | -0.014 |
mice | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +6.65e-03 |
Pos logit contributions | |
goodbye | +0.052 |
oms | +0.044 |
attention | +0.044 |
ogging | +0.043 |
toward | +0.043 |
Neg logit contributions | |
Mic | -0.049 |
mice | -0.048 |
ery | -0.048 |
iflower | -0.047 |
lf | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | +4.10e-03 |
Pos logit contributions | |
ery | +0.011 |
mice | +0.010 |
iflower | +0.010 |
ities | +0.009 |
chini | +0.009 |
Neg logit contributions | |
goodbye | -0.013 |
contain | -0.011 |
toward | -0.011 |
ogging | -0.011 |
closer | -0.011 |
Token | " |
Token ID | 1 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
ery | +0.006 |
mice | +0.006 |
iflower | +0.006 |
chini | +0.005 |
Midnight | +0.005 |
Neg logit contributions | |
goodbye | -0.008 |
contain | -0.007 |
toward | -0.007 |
ogging | -0.007 |
farewell | -0.007 |
Token | Wow |
Token ID | 22017 |
Feature activation | +5.31e-02 |
Pos logit contributions | |
ery | +0.037 |
iflower | +0.036 |
mice | +0.036 |
es | +0.035 |
Mickey | +0.034 |
Neg logit contributions | |
contain | -0.036 |
umb | -0.035 |
closer | -0.035 |
goodbye | -0.035 |
toward | -0.033 |
Token | !" |
Token ID | 2474 |
Feature activation | +8.26e-03 |
Pos logit contributions | |
ery | +0.110 |
Max | +0.107 |
Mic | +0.107 |
mice | +0.104 |
iflower | +0.102 |
Neg logit contributions | |
goodbye | -0.077 |
contain | -0.065 |
plunge | -0.061 |
umb | -0.059 |
toward | -0.058 |
Token | said |
Token ID | 531 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
ery | +0.015 |
Mic | +0.014 |
iflower | +0.013 |
mice | +0.013 |
Max | +0.013 |
Neg logit contributions | |
goodbye | -0.015 |
contain | -0.013 |
plunge | -0.012 |
umb | -0.012 |
attention | -0.012 |
Token | Sarah |
Token ID | 10490 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
goodbye | +0.059 |
contain | +0.050 |
plunge | +0.047 |
umb | +0.047 |
toward | +0.047 |
Neg logit contributions | |
ery | -0.052 |
Mic | -0.049 |
iflower | -0.047 |
Max | -0.046 |
mice | -0.046 |
Token | , |
Token ID | 11 |
Feature activation | -8.80e-03 |
Pos logit contributions | |
goodbye | +0.038 |
contain | +0.032 |
plunge | +0.030 |
toward | +0.030 |
attention | +0.029 |
Neg logit contributions | |
ery | -0.041 |
Mic | -0.040 |
Max | -0.039 |
mice | -0.038 |
iflower | -0.038 |
Token | her |
Token ID | 607 |
Feature activation | +8.21e-03 |
Pos logit contributions | |
goodbye | +0.018 |
contain | +0.015 |
toward | +0.014 |
plunge | +0.013 |
wave | +0.013 |
Neg logit contributions | |
ery | -0.015 |
Mic | -0.015 |
Max | -0.014 |
mice | -0.014 |
iflower | -0.014 |
Token | green |
Token ID | 4077 |
Feature activation | +6.66e-02 |
Pos logit contributions | |
ery | +0.086 |
Mic | +0.086 |
Max | +0.085 |
Ted | +0.081 |
mice | +0.081 |
Neg logit contributions | |
goodbye | -0.075 |
contain | -0.063 |
umb | -0.060 |
plunge | -0.059 |
attention | -0.058 |
Token | and |
Token ID | 290 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
ery | +0.103 |
mice | +0.097 |
Mic | +0.097 |
Max | +0.096 |
iflower | +0.092 |
Neg logit contributions | |
goodbye | -0.134 |
contain | -0.117 |
plunge | -0.113 |
umb | -0.109 |
attention | -0.108 |
Token | yellow |
Token ID | 7872 |
Feature activation | +4.83e-02 |
Pos logit contributions | |
ery | +0.070 |
Mic | +0.069 |
Max | +0.067 |
mice | +0.065 |
iflower | +0.064 |
Neg logit contributions | |
goodbye | -0.063 |
contain | -0.053 |
umb | -0.051 |
plunge | -0.050 |
attention | -0.049 |
Token | and |
Token ID | 290 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
ery | +0.075 |
mice | +0.072 |
Max | +0.068 |
Mic | +0.067 |
ters | +0.066 |
Neg logit contributions | |
goodbye | -0.096 |
contain | -0.085 |
plunge | -0.081 |
fy | -0.078 |
oured | -0.078 |
Token | look |
Token ID | 804 |
Feature activation | -3.89e-03 |
Pos logit contributions | |
ery | +0.061 |
Mic | +0.061 |
Max | +0.058 |
iflower | +0.056 |
Ted | +0.056 |
Neg logit contributions | |
goodbye | -0.058 |
contain | -0.048 |
umb | -0.047 |
plunge | -0.046 |
attention | -0.045 |
Token | y |
Token ID | 331 |
Feature activation | +5.03e-02 |
Pos logit contributions | |
goodbye | +0.008 |
plunge | +0.008 |
ogging | +0.008 |
contain | +0.008 |
oured | +0.008 |
Neg logit contributions | |
ery | -0.005 |
mice | -0.005 |
dough | -0.004 |
hes | -0.004 |
magnets | -0.004 |
Token | ummy |
Token ID | 13513 |
Feature activation | +7.58e-02 |
Pos logit contributions | |
ery | +0.062 |
mice | +0.057 |
Mic | +0.054 |
iflower | +0.054 |
Next | +0.053 |
Neg logit contributions | |
goodbye | -0.121 |
contain | -0.107 |
plunge | -0.103 |
tour | -0.101 |
farewell | -0.100 |
Token | . |
Token ID | 13 |
Feature activation | +8.00e-03 |
Pos logit contributions | |
ery | +0.120 |
mice | +0.115 |
Mic | +0.110 |
Max | +0.108 |
iflower | +0.108 |
Neg logit contributions | |
goodbye | -0.149 |
contain | -0.132 |
plunge | -0.127 |
oured | -0.121 |
umb | -0.121 |
Token |
|
Token ID | 198 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
ery | +0.011 |
Mic | +0.010 |
iflower | +0.010 |
mice | +0.009 |
Ted | +0.009 |
Neg logit contributions | |
goodbye | -0.019 |
contain | -0.016 |
tickets | -0.015 |
umb | -0.015 |
toward | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
ery | +0.027 |
Mic | +0.026 |
mice | +0.024 |
iflower | +0.024 |
Max | +0.023 |
Neg logit contributions | |
goodbye | -0.057 |
contain | -0.051 |
plunge | -0.049 |
toward | -0.048 |
attention | -0.048 |
Token | " |
Token ID | 1 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
ery | +0.019 |
Mic | +0.018 |
mice | +0.018 |
iflower | +0.017 |
Midnight | +0.016 |
Neg logit contributions | |
goodbye | -0.041 |
contain | -0.037 |
fy | -0.035 |
toward | -0.034 |
plunge | -0.034 |
Token | , |
Token ID | 11 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
goodbye | +0.025 |
contain | +0.021 |
plunge | +0.020 |
umb | +0.019 |
attention | +0.019 |
Neg logit contributions | |
ery | -0.037 |
Mic | -0.037 |
Max | -0.036 |
mice | -0.035 |
Ted | -0.035 |
Token | there |
Token ID | 612 |
Feature activation | -1.24e-03 |
Pos logit contributions | |
goodbye | +0.017 |
closer | +0.016 |
plunge | +0.014 |
toward | +0.013 |
contain | +0.013 |
Neg logit contributions | |
Mic | -0.022 |
iflower | -0.021 |
Doctor | -0.020 |
chini | -0.020 |
mice | -0.020 |
Token | was |
Token ID | 373 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
goodbye | +0.002 |
plunge | +0.002 |
contain | +0.001 |
inspection | +0.001 |
look | +0.001 |
Neg logit contributions | |
Mic | -0.003 |
mice | -0.002 |
ery | -0.002 |
iflower | -0.002 |
Fox | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | +5.49e-03 |
Pos logit contributions | |
iders | +0.062 |
ery | +0.061 |
iflower | +0.060 |
Mouse | +0.059 |
Fox | +0.058 |
Neg logit contributions | |
goodbye | -0.056 |
plunge | -0.042 |
umb | -0.041 |
contain | -0.041 |
farewell | -0.040 |
Token | cute |
Token ID | 13779 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
ery | +0.009 |
Mic | +0.008 |
Max | +0.008 |
chini | +0.008 |
Next | +0.008 |
Neg logit contributions | |
goodbye | -0.010 |
wave | -0.009 |
tour | -0.008 |
contain | -0.008 |
plunge | -0.008 |
Token | little |
Token ID | 1310 |
Feature activation | +3.78e-02 |
Pos logit contributions | |
ery | +0.026 |
Mic | +0.025 |
Max | +0.024 |
hes | +0.024 |
Ted | +0.024 |
Neg logit contributions | |
goodbye | -0.035 |
contain | -0.029 |
tour | -0.028 |
plunge | -0.028 |
wave | -0.028 |
Token | puppy |
Token ID | 26188 |
Feature activation | +5.55e-02 |
Pos logit contributions | |
ery | +0.046 |
Mic | +0.045 |
Max | +0.044 |
mice | +0.043 |
iflower | +0.040 |
Neg logit contributions | |
goodbye | -0.088 |
contain | -0.079 |
plunge | -0.076 |
umb | -0.074 |
attention | -0.074 |
Token | named |
Token ID | 3706 |
Feature activation | +7.27e-03 |
Pos logit contributions | |
ery | +0.100 |
Mic | +0.098 |
Max | +0.094 |
mice | +0.091 |
chini | +0.091 |
Neg logit contributions | |
goodbye | -0.097 |
contain | -0.078 |
toward | -0.074 |
fy | -0.073 |
plunge | -0.073 |
Token | Max |
Token ID | 5436 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
ery | +0.010 |
iflower | +0.008 |
chini | +0.008 |
hes | +0.008 |
mice | +0.008 |
Neg logit contributions | |
goodbye | -0.017 |
plunge | -0.014 |
contain | -0.014 |
umb | -0.014 |
fy | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -8.31e-04 |
Pos logit contributions | |
ery | +0.048 |
Mic | +0.048 |
chini | +0.046 |
mice | +0.046 |
Max | +0.046 |
Neg logit contributions | |
goodbye | -0.034 |
contain | -0.026 |
toward | -0.025 |
hug | -0.025 |
tour | -0.024 |
Token | Max |
Token ID | 5436 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
goodbye | +0.002 |
hug | +0.002 |
fy | +0.002 |
toward | +0.002 |
farewell | +0.002 |
Neg logit contributions | |
ery | -0.001 |
iflower | -0.001 |
chini | -0.001 |
ters | -0.001 |
web | -0.001 |
Token | out |
Token ID | 503 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
ery | +0.028 |
mice | +0.027 |
Max | +0.027 |
Mic | +0.027 |
Midnight | +0.026 |
Neg logit contributions | |
goodbye | -0.031 |
contain | -0.028 |
plunge | -0.026 |
fy | -0.025 |
ogging | -0.025 |
Token | what |
Token ID | 644 |
Feature activation | +7.77e-03 |
Pos logit contributions | |
mice | +0.024 |
ery | +0.023 |
Max | +0.023 |
Midnight | +0.022 |
Mic | +0.021 |
Neg logit contributions | |
goodbye | -0.033 |
plunge | -0.031 |
contain | -0.031 |
oured | -0.030 |
eal | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
ery | +0.015 |
Mic | +0.015 |
iflower | +0.013 |
chini | +0.013 |
ascal | +0.013 |
Neg logit contributions | |
goodbye | -0.014 |
umb | -0.011 |
contain | -0.011 |
tour | -0.011 |
tickets | -0.011 |
Token | mechanic |
Token ID | 21600 |
Feature activation | +2.84e-02 |
Pos logit contributions | |
ery | +0.018 |
Mic | +0.018 |
Max | +0.017 |
Ted | +0.017 |
iflower | +0.016 |
Neg logit contributions | |
goodbye | -0.026 |
contain | -0.023 |
plunge | -0.022 |
attention | -0.022 |
umb | -0.022 |
Token | was |
Token ID | 373 |
Feature activation | +6.94e-04 |
Pos logit contributions | |
Mic | +0.050 |
ery | +0.047 |
iflower | +0.046 |
mice | +0.045 |
Max | +0.045 |
Neg logit contributions | |
goodbye | -0.051 |
contain | -0.044 |
oured | -0.042 |
fy | -0.041 |
plunge | -0.041 |
Token | doing |
Token ID | 1804 |
Feature activation | +2.31e-02 |
Pos logit contributions | |
ery | +0.001 |
Mic | +0.001 |
mice | +0.001 |
Max | +0.001 |
iflower | +0.001 |
Neg logit contributions | |
goodbye | -0.001 |
contain | -0.001 |
plunge | -0.001 |
umb | -0.001 |
farewell | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
ery | +0.039 |
Mic | +0.038 |
Max | +0.037 |
mice | +0.037 |
iflower | +0.036 |
Neg logit contributions | |
goodbye | -0.042 |
contain | -0.036 |
plunge | -0.035 |
umb | -0.034 |
attention | -0.033 |
Token | She |
Token ID | 1375 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
goodbye | +0.021 |
toward | +0.018 |
contain | +0.017 |
oured | +0.016 |
plunge | +0.016 |
Neg logit contributions | |
ery | -0.021 |
iflower | -0.019 |
Mic | -0.018 |
chini | -0.018 |
lf | -0.018 |
Token | opened |
Token ID | 4721 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
ery | +0.026 |
Mic | +0.026 |
Max | +0.025 |
mice | +0.025 |
iflower | +0.024 |
Neg logit contributions | |
goodbye | -0.030 |
contain | -0.025 |
plunge | -0.024 |
umb | -0.024 |
toward | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | -3.68e-02 |
Pos logit contributions | |
goodbye | +0.077 |
contain | +0.070 |
plunge | +0.067 |
fy | +0.066 |
umb | +0.064 |
Neg logit contributions | |
ery | -0.041 |
Max | -0.040 |
Mic | -0.039 |
mice | -0.039 |
Ted | -0.036 |
Token | door |
Token ID | 3420 |
Feature activation | -6.14e-02 |
Pos logit contributions | |
goodbye | +0.063 |
contain | +0.059 |
ogging | +0.055 |
plunge | +0.054 |
fy | +0.053 |
Neg logit contributions | |
ery | -0.062 |
mice | -0.062 |
magnets | -0.062 |
Mic | -0.060 |
Max | -0.059 |
Token | were |
Token ID | 547 |
Feature activation | -2.29e-03 |
Pos logit contributions | |
goodbye | +0.108 |
contain | +0.093 |
umb | +0.088 |
plunge | +0.087 |
oured | +0.085 |
Neg logit contributions | |
ery | -0.097 |
Mic | -0.095 |
Max | -0.092 |
iflower | -0.091 |
mice | -0.090 |
Token | on |
Token ID | 319 |
Feature activation | -2.41e-03 |
Pos logit contributions | |
goodbye | +0.004 |
contain | +0.004 |
umb | +0.003 |
plunge | +0.003 |
attention | +0.003 |
Neg logit contributions | |
ery | -0.004 |
Mic | -0.004 |
Max | -0.004 |
iflower | -0.004 |
Ted | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | -4.44e-02 |
Pos logit contributions | |
goodbye | +0.005 |
contain | +0.004 |
umb | +0.004 |
plunge | +0.004 |
attention | +0.004 |
Neg logit contributions | |
ery | -0.004 |
Mic | -0.004 |
Max | -0.004 |
iflower | -0.004 |
Bear | -0.004 |
Token | leash |
Token ID | 40529 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
goodbye | +0.076 |
contain | +0.064 |
plunge | +0.062 |
attention | +0.061 |
oured | +0.060 |
Neg logit contributions | |
Mic | -0.081 |
ery | -0.080 |
Max | -0.080 |
mice | -0.077 |
iflower | -0.075 |
Token | , |
Token ID | 11 |
Feature activation | +8.99e-03 |
Pos logit contributions | |
goodbye | +0.058 |
contain | +0.049 |
plunge | +0.048 |
umb | +0.046 |
attention | +0.045 |
Neg logit contributions | |
ery | -0.061 |
Mic | -0.059 |
Max | -0.059 |
mice | -0.057 |
iflower | -0.056 |
Token | but |
Token ID | 475 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
ery | +0.017 |
Mic | +0.016 |
Max | +0.016 |
iflower | +0.016 |
mice | +0.015 |
Neg logit contributions | |
goodbye | -0.016 |
contain | -0.013 |
umb | -0.012 |
plunge | -0.012 |
toward | -0.012 |
Token | they |
Token ID | 484 |
Feature activation | -8.52e-03 |
Pos logit contributions | |
goodbye | +0.022 |
contain | +0.017 |
plunge | +0.017 |
umb | +0.017 |
attention | +0.016 |
Neg logit contributions | |
ery | -0.020 |
Mic | -0.020 |
iflower | -0.019 |
Max | -0.018 |
chini | -0.018 |
Token | still |
Token ID | 991 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
goodbye | +0.016 |
contain | +0.013 |
umb | +0.012 |
plunge | +0.012 |
toward | +0.012 |
Neg logit contributions | |
Mic | -0.015 |
ery | -0.015 |
iflower | -0.014 |
Max | -0.014 |
Next | -0.014 |
Token | couldn |
Token ID | 3521 |
Feature activation | -8.49e-03 |
Pos logit contributions | |
goodbye | +0.028 |
contain | +0.025 |
plunge | +0.024 |
umb | +0.023 |
attention | +0.023 |
Neg logit contributions | |
ery | -0.020 |
Mic | -0.019 |
Max | -0.019 |
mice | -0.018 |
iflower | -0.018 |
Token | 't |
Token ID | 470 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
goodbye | +0.019 |
plunge | +0.016 |
contain | +0.016 |
umb | +0.016 |
toward | +0.015 |
Neg logit contributions | |
ery | -0.011 |
Mic | -0.011 |
mice | -0.010 |
iflower | -0.010 |
Max | -0.010 |
Token | be |
Token ID | 307 |
Feature activation | +8.26e-03 |
Pos logit contributions | |
goodbye | +0.023 |
contain | +0.021 |
plunge | +0.019 |
toward | +0.018 |
umb | +0.018 |
Neg logit contributions | |
ery | -0.031 |
Mic | -0.030 |
Max | -0.029 |
iflower | -0.029 |
mice | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -5.06e-03 |
Pos logit contributions | |
Mic | +0.077 |
Fox | +0.073 |
chini | +0.072 |
Mouse | +0.071 |
Ted | +0.070 |
Neg logit contributions | |
goodbye | -0.042 |
fy | -0.027 |
attention | -0.026 |
hug | -0.025 |
contain | -0.025 |
Token | It |
Token ID | 632 |
Feature activation | +4.25e-02 |
Pos logit contributions | |
goodbye | +0.010 |
umb | +0.008 |
fy | +0.008 |
hi | +0.007 |
toward | +0.007 |
Neg logit contributions | |
ery | -0.009 |
Mic | -0.008 |
iflower | -0.008 |
Max | -0.007 |
chini | -0.007 |
Token | 's |
Token ID | 338 |
Feature activation | +2.77e-02 |
Pos logit contributions | |
Mic | +0.083 |
Charlie | +0.076 |
ery | +0.075 |
iflower | +0.074 |
chini | +0.074 |
Neg logit contributions | |
goodbye | -0.066 |
hug | -0.057 |
plunge | -0.056 |
ounded | -0.055 |
closer | -0.054 |
Token | not |
Token ID | 407 |
Feature activation | +7.52e-03 |
Pos logit contributions | |
Mic | +0.038 |
Ted | +0.038 |
Bear | +0.037 |
ery | +0.037 |
Max | +0.037 |
Neg logit contributions | |
goodbye | -0.062 |
attention | -0.054 |
contain | -0.051 |
umb | -0.051 |
closer | -0.049 |
Token | nice |
Token ID | 3621 |
Feature activation | +2.41e-02 |
Pos logit contributions | |
Mic | +0.015 |
Max | +0.014 |
ery | +0.014 |
Ted | +0.014 |
iflower | +0.014 |
Neg logit contributions | |
goodbye | -0.012 |
contain | -0.010 |
attention | -0.010 |
fy | -0.009 |
umb | -0.009 |
Token | to |
Token ID | 284 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
ery | +0.052 |
mice | +0.051 |
Mic | +0.050 |
Max | +0.050 |
iflower | +0.048 |
Neg logit contributions | |
goodbye | -0.032 |
contain | -0.027 |
plunge | -0.026 |
umb | -0.024 |
oured | -0.024 |
Token | ignore |
Token ID | 8856 |
Feature activation | +6.46e-03 |
Pos logit contributions | |
ery | +0.023 |
Mic | +0.021 |
iflower | +0.021 |
Max | +0.021 |
chini | +0.021 |
Neg logit contributions | |
goodbye | -0.016 |
contain | -0.014 |
umb | -0.013 |
plunge | -0.013 |
attention | -0.013 |
Token | her |
Token ID | 607 |
Feature activation | -1.21e-03 |
Pos logit contributions | |
ery | +0.012 |
Mic | +0.012 |
chini | +0.012 |
Ted | +0.011 |
iflower | +0.011 |
Neg logit contributions | |
goodbye | -0.011 |
attention | -0.009 |
umb | -0.009 |
hug | -0.008 |
plunge | -0.008 |
Token | ." |
Token ID | 526 |
Feature activation | -5.79e-03 |
Pos logit contributions | |
goodbye | +0.002 |
contain | +0.002 |
attention | +0.002 |
plunge | +0.002 |
umb | +0.002 |
Neg logit contributions | |
Mic | -0.002 |
Max | -0.002 |
ery | -0.002 |
Ted | -0.002 |
Charlie | -0.002 |
Token | Tim |
Token ID | 5045 |
Feature activation | -4.92e-03 |
Pos logit contributions | |
goodbye | +0.012 |
umb | +0.010 |
hug | +0.009 |
toward | +0.009 |
contain | +0.009 |
Neg logit contributions | |
ery | -0.011 |
iflower | -0.010 |
chini | -0.009 |
Mic | -0.009 |
Fox | -0.009 |
Token | felt |
Token ID | 2936 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
goodbye | +0.008 |
eal | +0.006 |
contain | +0.006 |
hug | +0.006 |
oured | +0.006 |
Neg logit contributions | |
ery | -0.010 |
Mic | -0.010 |
iflower | -0.009 |
chini | -0.009 |
mice | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
ery | +0.027 |
Mic | +0.027 |
Max | +0.026 |
mice | +0.025 |
chini | +0.025 |
Neg logit contributions | |
goodbye | -0.031 |
contain | -0.025 |
plunge | -0.024 |
umb | -0.024 |
oured | -0.024 |
Token | sad |
Token ID | 6507 |
Feature activation | +5.03e-02 |
Pos logit contributions | |
ery | +0.041 |
Mic | +0.040 |
Max | +0.039 |
Ted | +0.037 |
Bear | +0.037 |
Neg logit contributions | |
goodbye | -0.038 |
contain | -0.032 |
plunge | -0.030 |
umb | -0.030 |
attention | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
mice | +0.063 |
ery | +0.059 |
Midnight | +0.057 |
Next | +0.057 |
Max | +0.056 |
Neg logit contributions | |
goodbye | -0.109 |
contain | -0.103 |
plunge | -0.100 |
oured | -0.099 |
eal | -0.097 |
Token | lost |
Token ID | 2626 |
Feature activation | +2.64e-02 |
Pos logit contributions | |
ery | +0.023 |
Mic | +0.022 |
Max | +0.022 |
mice | +0.021 |
iflower | +0.021 |
Neg logit contributions | |
goodbye | -0.028 |
contain | -0.024 |
plunge | -0.023 |
umb | -0.023 |
attention | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +1.17e-03 |
Pos logit contributions | |
ery | +0.033 |
mice | +0.033 |
Max | +0.032 |
Mic | +0.032 |
Midnight | +0.030 |
Neg logit contributions | |
goodbye | -0.058 |
contain | -0.053 |
plunge | -0.052 |
oured | -0.051 |
fy | -0.050 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
ery | +0.002 |
iflower | +0.002 |
chini | +0.002 |
Mic | +0.002 |
oyd | +0.002 |
Neg logit contributions | |
goodbye | -0.002 |
contain | -0.002 |
toward | -0.002 |
fy | -0.002 |
umb | -0.002 |
Token | wanted |
Token ID | 2227 |
Feature activation | -8.29e-03 |
Pos logit contributions | |
ery | +0.033 |
Mic | +0.032 |
Max | +0.030 |
iflower | +0.030 |
mice | +0.030 |
Neg logit contributions | |
goodbye | -0.040 |
contain | -0.033 |
plunge | -0.032 |
umb | -0.031 |
toward | -0.031 |
Token | to |
Token ID | 284 |
Feature activation | +3.11e-03 |
Pos logit contributions | |
goodbye | +0.011 |
contain | +0.009 |
plunge | +0.008 |
umb | +0.008 |
oured | +0.008 |
Neg logit contributions | |
ery | -0.018 |
Mic | -0.018 |
Max | -0.018 |
mice | -0.018 |
Midnight | -0.017 |
Token | help |
Token ID | 1037 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
ery | +0.006 |
Mic | +0.006 |
mice | +0.006 |
Max | +0.006 |
Midnight | +0.005 |
Neg logit contributions | |
goodbye | -0.005 |
oured | -0.004 |
fy | -0.004 |
umb | -0.004 |
oms | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | -1.01e-03 |
Pos logit contributions | |
goodbye | +0.039 |
contain | +0.033 |
attention | +0.032 |
plunge | +0.032 |
toward | +0.031 |
Neg logit contributions | |
ery | -0.039 |
Mic | -0.037 |
Max | -0.036 |
iflower | -0.035 |
Ted | -0.035 |
Token | fairy |
Token ID | 25607 |
Feature activation | +8.64e-03 |
Pos logit contributions | |
goodbye | +0.002 |
contain | +0.002 |
plunge | +0.002 |
umb | +0.002 |
attention | +0.002 |
Neg logit contributions | |
ery | -0.001 |
Mic | -0.001 |
Max | -0.001 |
mice | -0.001 |
iflower | -0.001 |
Token | see |
Token ID | 766 |
Feature activation | +8.36e-03 |
Pos logit contributions | |
goodbye | +0.053 |
oured | +0.045 |
fy | +0.044 |
plunge | +0.044 |
umb | +0.043 |
Neg logit contributions | |
ery | -0.061 |
Max | -0.060 |
mice | -0.059 |
Mic | -0.058 |
Ted | -0.056 |
Token | something |
Token ID | 1223 |
Feature activation | +7.19e-03 |
Pos logit contributions | |
ery | +0.010 |
Mic | +0.010 |
Max | +0.010 |
mice | +0.010 |
iflower | +0.009 |
Neg logit contributions | |
goodbye | -0.019 |
contain | -0.017 |
plunge | -0.017 |
umb | -0.016 |
attention | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +6.36e-03 |
Pos logit contributions | |
ery | +0.011 |
Max | +0.011 |
mice | +0.011 |
iflower | +0.010 |
hes | +0.010 |
Neg logit contributions | |
goodbye | -0.014 |
contain | -0.013 |
oured | -0.012 |
fy | -0.012 |
plunge | -0.012 |
Token | They |
Token ID | 1119 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
ery | +0.011 |
Mic | +0.010 |
iflower | +0.010 |
Max | +0.009 |
Ted | +0.009 |
Neg logit contributions | |
goodbye | -0.013 |
contain | -0.011 |
umb | -0.010 |
toward | -0.010 |
plunge | -0.010 |
Token | see |
Token ID | 766 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
goodbye | +0.021 |
fy | +0.018 |
oured | +0.017 |
plunge | +0.017 |
contain | +0.017 |
Neg logit contributions | |
ery | -0.027 |
Max | -0.026 |
mice | -0.026 |
Mic | -0.026 |
Ted | -0.025 |
Token | a |
Token ID | 257 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
ery | +0.017 |
Mic | +0.017 |
Max | +0.016 |
mice | +0.015 |
iflower | +0.015 |
Neg logit contributions | |
goodbye | -0.030 |
contain | -0.026 |
plunge | -0.026 |
umb | -0.025 |
attention | -0.025 |
Token | big |
Token ID | 1263 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
goodbye | +0.056 |
contain | +0.050 |
plunge | +0.047 |
oured | +0.046 |
attention | +0.045 |
Neg logit contributions | |
ery | -0.044 |
Mic | -0.044 |
Max | -0.043 |
mice | -0.043 |
iflower | -0.041 |
Token | red |
Token ID | 2266 |
Feature activation | -7.69e-03 |
Pos logit contributions | |
goodbye | +0.063 |
contain | +0.057 |
oured | +0.054 |
plunge | +0.054 |
fy | +0.053 |
Neg logit contributions | |
mice | -0.042 |
ery | -0.041 |
Max | -0.040 |
Mic | -0.040 |
iflower | -0.038 |
Token | truck |
Token ID | 7779 |
Feature activation | +2.44e-03 |
Pos logit contributions | |
goodbye | +0.016 |
contain | +0.015 |
fy | +0.014 |
attention | +0.014 |
oured | +0.014 |
Neg logit contributions | |
ery | -0.011 |
mice | -0.010 |
Max | -0.010 |
Mic | -0.010 |
dough | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +4.30e-03 |
Pos logit contributions | |
Max | +0.003 |
ery | +0.003 |
Mic | +0.003 |
mice | +0.003 |
Ted | +0.003 |
Neg logit contributions | |
goodbye | -0.005 |
contain | -0.004 |
plunge | -0.004 |
umb | -0.004 |
fy | -0.004 |
Token | The |
Token ID | 383 |
Feature activation | +6.52e-03 |
Pos logit contributions | |
ery | +0.007 |
Mic | +0.006 |
iflower | +0.006 |
Max | +0.006 |
mice | +0.006 |
Neg logit contributions | |
goodbye | -0.009 |
contain | -0.008 |
umb | -0.007 |
plunge | -0.007 |
toward | -0.007 |
Token | not |
Token ID | 407 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
goodbye | +0.009 |
contain | +0.008 |
plunge | +0.008 |
umb | +0.008 |
oured | +0.008 |
Neg logit contributions | |
ery | -0.002 |
Max | -0.002 |
Mic | -0.002 |
mice | -0.002 |
Midnight | -0.001 |
Token | like |
Token ID | 588 |
Feature activation | -2.30e-03 |
Pos logit contributions | |
goodbye | +0.044 |
contain | +0.035 |
plunge | +0.034 |
umb | +0.033 |
oured | +0.032 |
Neg logit contributions | |
ery | -0.061 |
Mic | -0.059 |
Max | -0.059 |
mice | -0.058 |
Ted | -0.056 |
Token | this |
Token ID | 428 |
Feature activation | -8.74e-03 |
Pos logit contributions | |
goodbye | +0.004 |
attention | +0.003 |
umb | +0.003 |
contain | +0.003 |
plunge | +0.003 |
Neg logit contributions | |
ery | -0.005 |
Mic | -0.004 |
Max | -0.004 |
Bear | -0.004 |
Mouse | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
goodbye | +0.015 |
contain | +0.013 |
attention | +0.012 |
plunge | +0.012 |
toward | +0.012 |
Neg logit contributions | |
Mic | -0.016 |
ery | -0.016 |
Max | -0.016 |
iflower | -0.015 |
Bear | -0.015 |
Token | became |
Token ID | 2627 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
Mic | +0.034 |
ery | +0.032 |
chini | +0.031 |
Max | +0.031 |
iflower | +0.030 |
Neg logit contributions | |
goodbye | -0.028 |
wave | -0.021 |
contain | -0.020 |
toward | -0.020 |
paddle | -0.020 |
Token | very |
Token ID | 845 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
oon | +0.076 |
Mic | +0.075 |
Max | +0.075 |
iflower | +0.075 |
ery | +0.074 |
Neg logit contributions | |
goodbye | -0.060 |
attention | -0.051 |
closer | -0.049 |
umb | -0.047 |
further | -0.046 |
Token | sad |
Token ID | 6507 |
Feature activation | +4.20e-02 |
Pos logit contributions | |
goodbye | +0.037 |
contain | +0.033 |
plunge | +0.031 |
umb | +0.031 |
oured | +0.030 |
Neg logit contributions | |
ery | -0.026 |
Mic | -0.025 |
Max | -0.025 |
mice | -0.025 |
iflower | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -3.82e-03 |
Pos logit contributions | |
Mic | +0.077 |
ery | +0.077 |
Max | +0.075 |
iflower | +0.071 |
mice | +0.071 |
Neg logit contributions | |
goodbye | -0.073 |
contain | -0.061 |
plunge | -0.058 |
attention | -0.057 |
umb | -0.057 |
Token | The |
Token ID | 383 |
Feature activation | +7.03e-03 |
Pos logit contributions | |
goodbye | +0.006 |
farewell | +0.004 |
hi | +0.004 |
hello | +0.004 |
toward | +0.004 |
Neg logit contributions | |
ery | -0.009 |
iflower | -0.008 |
Mic | -0.008 |
web | -0.008 |
kin | -0.007 |
Token | dog |
Token ID | 3290 |
Feature activation | -4.96e-03 |
Pos logit contributions | |
ery | +0.012 |
iflower | +0.011 |
chini | +0.011 |
Mic | +0.010 |
Ted | +0.010 |
Neg logit contributions | |
goodbye | -0.014 |
hug | -0.012 |
attention | -0.012 |
wave | -0.012 |
tour | -0.012 |
Token | did |
Token ID | 750 |
Feature activation | +2.09e-03 |
Pos logit contributions | |
goodbye | +0.008 |
wave | +0.006 |
contain | +0.006 |
hug | +0.006 |
tour | +0.006 |
Neg logit contributions | |
ery | -0.010 |
Mic | -0.010 |
chini | -0.010 |
iflower | -0.010 |
Next | -0.009 |
Token | dad |
Token ID | 9955 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
ery | +0.001 |
Mic | +0.001 |
Max | +0.001 |
mice | +0.001 |
Ted | +0.001 |
Neg logit contributions | |
goodbye | -0.001 |
contain | -0.001 |
plunge | -0.001 |
umb | -0.001 |
fy | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -5.35e-04 |
Pos logit contributions | |
goodbye | +0.045 |
umb | +0.040 |
plunge | +0.040 |
fy | +0.039 |
contain | +0.039 |
Neg logit contributions | |
ery | -0.040 |
mice | -0.039 |
Max | -0.039 |
Mic | -0.038 |
Midnight | -0.037 |
Token | but |
Token ID | 475 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
goodbye | +0.001 |
contain | +0.001 |
toward | +0.001 |
closer | +0.001 |
attention | +0.001 |
Neg logit contributions | |
ery | -0.001 |
iflower | -0.001 |
Mic | -0.001 |
chini | -0.001 |
mice | -0.001 |
Token | they |
Token ID | 484 |
Feature activation | -6.79e-03 |
Pos logit contributions | |
goodbye | +0.039 |
contain | +0.033 |
plunge | +0.032 |
umb | +0.031 |
attention | +0.031 |
Neg logit contributions | |
ery | -0.029 |
Mic | -0.029 |
Max | -0.027 |
mice | -0.027 |
iflower | -0.026 |
Token | did |
Token ID | 750 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
goodbye | +0.015 |
contain | +0.013 |
plunge | +0.013 |
umb | +0.013 |
attention | +0.013 |
Neg logit contributions | |
ery | -0.009 |
Max | -0.009 |
Mic | -0.009 |
mice | -0.008 |
Ted | -0.008 |
Token | not |
Token ID | 407 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
ery | +0.010 |
Mic | +0.010 |
Max | +0.009 |
mice | +0.009 |
iflower | +0.008 |
Neg logit contributions | |
goodbye | -0.033 |
contain | -0.030 |
plunge | -0.029 |
umb | -0.028 |
attention | -0.028 |
Token | know |
Token ID | 760 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
goodbye | +0.038 |
oured | +0.034 |
umb | +0.033 |
plunge | +0.033 |
icol | +0.031 |
Neg logit contributions | |
Max | -0.038 |
ery | -0.038 |
mice | -0.037 |
Mic | -0.036 |
Doctor | -0.035 |
Token | where |
Token ID | 810 |
Feature activation | -7.81e-03 |
Pos logit contributions | |
ery | +0.025 |
Mic | +0.025 |
Max | +0.024 |
mice | +0.023 |
Ted | +0.023 |
Neg logit contributions | |
goodbye | -0.028 |
contain | -0.024 |
plunge | -0.023 |
umb | -0.023 |
attention | -0.022 |
Token | it |
Token ID | 340 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
goodbye | +0.016 |
contain | +0.015 |
oured | +0.015 |
plunge | +0.015 |
fy | +0.014 |
Neg logit contributions | |
mice | -0.010 |
Max | -0.010 |
ery | -0.010 |
Midnight | -0.010 |
Ted | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
Mic | +0.027 |
ery | +0.026 |
Max | +0.024 |
iflower | +0.024 |
mice | +0.024 |
Neg logit contributions | |
goodbye | -0.036 |
contain | -0.032 |
plunge | -0.031 |
tour | -0.030 |
toward | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -1.34e-03 |
Pos logit contributions | |
ery | +0.029 |
mice | +0.027 |
Max | +0.027 |
Midnight | +0.026 |
Mic | +0.026 |
Neg logit contributions | |
goodbye | -0.039 |
contain | -0.034 |
oured | -0.033 |
plunge | -0.033 |
eal | -0.033 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
ery | +0.004 |
Mic | +0.004 |
iflower | +0.004 |
chini | +0.004 |
Fox | +0.004 |
Neg logit contributions | |
goodbye | -0.005 |
umb | -0.004 |
contain | -0.004 |
attention | -0.004 |
plunge | -0.004 |
Token | took |
Token ID | 1718 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
ery | +0.022 |
Mic | +0.022 |
mice | +0.021 |
Max | +0.021 |
chini | +0.021 |
Neg logit contributions | |
goodbye | -0.023 |
contain | -0.018 |
oured | -0.018 |
umb | -0.018 |
plunge | -0.018 |
Token | a |
Token ID | 257 |
Feature activation | -6.26e-02 |
Pos logit contributions | |
goodbye | +0.030 |
contain | +0.026 |
plunge | +0.025 |
umb | +0.025 |
attention | +0.025 |
Neg logit contributions | |
ery | -0.023 |
Mic | -0.022 |
Max | -0.021 |
mice | -0.021 |
iflower | -0.020 |
Token | nap |
Token ID | 25422 |
Feature activation | -9.06e-03 |
Pos logit contributions | |
goodbye | +0.116 |
contain | +0.108 |
umb | +0.108 |
oured | +0.105 |
utt | +0.103 |
Neg logit contributions | |
Max | -0.092 |
Mic | -0.088 |
Midnight | -0.087 |
dough | -0.081 |
ery | -0.081 |
Token | and |
Token ID | 290 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
goodbye | +0.018 |
contain | +0.015 |
plunge | +0.015 |
umb | +0.014 |
attention | +0.014 |
Neg logit contributions | |
ery | -0.014 |
Mic | -0.014 |
Max | -0.014 |
mice | -0.013 |
iflower | -0.013 |
Token | felt |
Token ID | 2936 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
ery | +0.017 |
Max | +0.016 |
Mic | +0.016 |
Midnight | +0.016 |
mice | +0.016 |
Neg logit contributions | |
goodbye | -0.021 |
contain | -0.018 |
plunge | -0.018 |
toward | -0.018 |
umb | -0.017 |
Token | much |
Token ID | 881 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
goodbye | +0.026 |
plunge | +0.024 |
contain | +0.024 |
umb | +0.024 |
tour | +0.023 |
Neg logit contributions | |
ery | -0.021 |
mice | -0.019 |
Midnight | -0.019 |
hes | -0.019 |
Max | -0.019 |
Token | better |
Token ID | 1365 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
ery | +0.022 |
hes | +0.020 |
mice | +0.019 |
Midnight | +0.019 |
Max | +0.018 |
Neg logit contributions | |
goodbye | -0.037 |
umb | -0.036 |
plunge | -0.036 |
contain | -0.036 |
ooters | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
ery | +0.040 |
mice | +0.038 |
Max | +0.037 |
iflower | +0.037 |
Mic | +0.036 |
Neg logit contributions | |
goodbye | -0.060 |
plunge | -0.056 |
umb | -0.055 |
contain | -0.055 |
oured | -0.053 |
Token | She |
Token ID | 1375 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
goodbye | +0.028 |
contain | +0.023 |
toward | +0.022 |
oured | +0.022 |
attention | +0.022 |
Neg logit contributions | |
ery | -0.021 |
Mic | -0.019 |
iflower | -0.019 |
mice | -0.018 |
Max | -0.018 |
Token | learned |
Token ID | 4499 |
Feature activation | -1.57e-04 |
Pos logit contributions | |
ery | +0.023 |
Mic | +0.023 |
Max | +0.022 |
mice | +0.022 |
iflower | +0.021 |
Neg logit contributions | |
goodbye | -0.028 |
contain | -0.024 |
plunge | -0.023 |
umb | -0.022 |
attention | -0.022 |
Token | up |
Token ID | 510 |
Feature activation | +6.52e-03 |
Pos logit contributions | |
goodbye | +0.015 |
contain | +0.015 |
fy | +0.014 |
umb | +0.013 |
plunge | +0.013 |
Neg logit contributions | |
Midnight | -0.013 |
ery | -0.013 |
mice | -0.013 |
Max | -0.013 |
iflower | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -9.56e-03 |
Pos logit contributions | |
mice | +0.009 |
Midnight | +0.009 |
Max | +0.009 |
Mickey | +0.009 |
ery | +0.009 |
Neg logit contributions | |
goodbye | -0.011 |
contain | -0.011 |
fy | -0.011 |
ooters | -0.011 |
eal | -0.010 |
Token | she |
Token ID | 673 |
Feature activation | -1.29e-03 |
Pos logit contributions | |
goodbye | +0.020 |
contain | +0.018 |
plunge | +0.017 |
umb | +0.017 |
fy | +0.016 |
Neg logit contributions | |
ery | -0.013 |
Max | -0.013 |
Mic | -0.013 |
mice | -0.013 |
iflower | -0.012 |
Token | had |
Token ID | 550 |
Feature activation | -5.41e-03 |
Pos logit contributions | |
goodbye | +0.003 |
contain | +0.002 |
plunge | +0.002 |
umb | +0.002 |
attention | +0.002 |
Neg logit contributions | |
ery | -0.002 |
Mic | -0.002 |
Max | -0.002 |
mice | -0.002 |
iflower | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
goodbye | +0.011 |
contain | +0.010 |
plunge | +0.010 |
umb | +0.009 |
attention | +0.009 |
Neg logit contributions | |
ery | -0.008 |
Mic | -0.008 |
Max | -0.008 |
mice | -0.007 |
Ted | -0.007 |
Token | bright |
Token ID | 6016 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
goodbye | +0.053 |
contain | +0.048 |
toward | +0.045 |
plunge | +0.044 |
umb | +0.044 |
Neg logit contributions | |
ery | -0.045 |
Mic | -0.045 |
Max | -0.045 |
mice | -0.043 |
Midnight | -0.042 |
Token | idea |
Token ID | 2126 |
Feature activation | -9.48e-03 |
Pos logit contributions | |
goodbye | +0.051 |
contain | +0.047 |
plunge | +0.044 |
toward | +0.043 |
attention | +0.043 |
Neg logit contributions | |
ery | -0.047 |
Max | -0.043 |
mice | -0.043 |
iflower | -0.042 |
dough | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
goodbye | +0.019 |
contain | +0.017 |
plunge | +0.017 |
fy | +0.017 |
oured | +0.017 |
Neg logit contributions | |
ery | -0.013 |
Max | -0.013 |
mice | -0.012 |
Ted | -0.012 |
Mic | -0.012 |
Token | She |
Token ID | 1375 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
goodbye | +0.034 |
contain | +0.032 |
plunge | +0.031 |
umb | +0.030 |
fy | +0.030 |
Neg logit contributions | |
mice | -0.012 |
ery | -0.012 |
Max | -0.012 |
Mic | -0.012 |
Next | -0.012 |
Token | decided |
Token ID | 3066 |
Feature activation | -2.96e-02 |
Pos logit contributions | |
ery | +0.025 |
Mic | +0.024 |
Max | +0.024 |
mice | +0.023 |
Ted | +0.023 |
Neg logit contributions | |
goodbye | -0.032 |
contain | -0.028 |
plunge | -0.027 |
umb | -0.026 |
attention | -0.026 |
Token | to |
Token ID | 284 |
Feature activation | -8.43e-03 |
Pos logit contributions | |
contain | +0.044 |
goodbye | +0.044 |
fy | +0.043 |
eal | +0.042 |
plunge | +0.041 |
Neg logit contributions | |
Max | -0.051 |
mice | -0.049 |
Midnight | -0.049 |
You | -0.047 |
Mic | -0.047 |
Token | to |
Token ID | 284 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
goodbye | +0.001 |
contain | +0.001 |
plunge | +0.001 |
umb | +0.001 |
attention | +0.001 |
Neg logit contributions | |
ery | -0.001 |
Mic | -0.001 |
Max | -0.001 |
mice | -0.001 |
iflower | -0.001 |
Token | stand |
Token ID | 1302 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
ery | +0.021 |
Mic | +0.021 |
Max | +0.021 |
mice | +0.020 |
iflower | +0.020 |
Neg logit contributions | |
goodbye | -0.015 |
contain | -0.012 |
plunge | -0.011 |
umb | -0.011 |
attention | -0.011 |
Token | up |
Token ID | 510 |
Feature activation | -7.64e-04 |
Pos logit contributions | |
goodbye | +0.030 |
contain | +0.026 |
plunge | +0.025 |
umb | +0.025 |
fy | +0.024 |
Neg logit contributions | |
ery | -0.023 |
Max | -0.022 |
Mic | -0.022 |
mice | -0.021 |
iflower | -0.021 |
Token | in |
Token ID | 287 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
goodbye | +0.002 |
plunge | +0.001 |
contain | +0.001 |
umb | +0.001 |
eal | +0.001 |
Neg logit contributions | |
ery | -0.001 |
Max | -0.001 |
Mic | -0.001 |
mice | -0.001 |
Next | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
goodbye | +0.079 |
contain | +0.069 |
plunge | +0.067 |
umb | +0.065 |
attention | +0.065 |
Neg logit contributions | |
ery | -0.053 |
Mic | -0.051 |
Max | -0.050 |
mice | -0.048 |
iflower | -0.047 |
Token | front |
Token ID | 2166 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
goodbye | +0.044 |
contain | +0.041 |
plunge | +0.039 |
inspection | +0.039 |
oured | +0.039 |
Neg logit contributions | |
Max | -0.030 |
ery | -0.029 |
mice | -0.029 |
Mic | -0.028 |
dough | -0.027 |
Token | of |
Token ID | 286 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
goodbye | +0.187 |
plunge | +0.177 |
oured | +0.168 |
ounded | +0.163 |
umb | +0.161 |
Neg logit contributions | |
mice | -0.141 |
ery | -0.140 |
Max | -0.138 |
dough | -0.135 |
house | -0.134 |
Token | the |
Token ID | 262 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
ery | +0.023 |
Mic | +0.022 |
Max | +0.021 |
iflower | +0.021 |
Ted | +0.020 |
Neg logit contributions | |
goodbye | -0.029 |
contain | -0.025 |
plunge | -0.024 |
umb | -0.024 |
attention | -0.023 |
Token | room |
Token ID | 2119 |
Feature activation | -7.75e-04 |
Pos logit contributions | |
goodbye | +0.035 |
contain | +0.031 |
plunge | +0.030 |
oured | +0.029 |
umb | +0.029 |
Neg logit contributions | |
Max | -0.027 |
ery | -0.027 |
mice | -0.026 |
Mic | -0.026 |
Ted | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -7.37e-03 |
Pos logit contributions | |
goodbye | +0.001 |
plunge | +0.001 |
contain | +0.001 |
umb | +0.001 |
attention | +0.001 |
Neg logit contributions | |
ery | -0.001 |
Max | -0.001 |
mice | -0.001 |
Mic | -0.001 |
iflower | -0.001 |
Token | All |
Token ID | 1439 |
Feature activation | -2.30e-02 |
Pos logit contributions | |
goodbye | +0.012 |
plunge | +0.010 |
toward | +0.010 |
inspection | +0.010 |
contain | +0.009 |
Neg logit contributions | |
ery | -0.016 |
iflower | -0.015 |
Mic | -0.014 |
web | -0.014 |
ities | -0.013 |
Token | thanked |
Token ID | 26280 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
ery | +0.013 |
Mic | +0.012 |
Max | +0.012 |
mice | +0.012 |
iflower | +0.011 |
Neg logit contributions | |
goodbye | -0.015 |
contain | -0.013 |
plunge | -0.013 |
umb | -0.013 |
attention | -0.012 |
Token | God |
Token ID | 1793 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
goodbye | +0.050 |
contain | +0.042 |
plunge | +0.040 |
attention | +0.039 |
umb | +0.039 |
Neg logit contributions | |
ery | -0.041 |
Mic | -0.039 |
Max | -0.037 |
iflower | -0.036 |
mice | -0.036 |
Token | for |
Token ID | 329 |
Feature activation | +2.27e-02 |
Pos logit contributions | |
goodbye | +0.041 |
contain | +0.037 |
plunge | +0.037 |
umb | +0.035 |
eal | +0.033 |
Neg logit contributions | |
ery | -0.056 |
mice | -0.054 |
Max | -0.053 |
Mic | -0.053 |
Bear | -0.052 |
Token | answering |
Token ID | 18877 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
ery | +0.036 |
Mic | +0.035 |
Max | +0.034 |
Ted | +0.033 |
Bear | +0.032 |
Neg logit contributions | |
goodbye | -0.047 |
contain | -0.039 |
plunge | -0.037 |
attention | -0.037 |
umb | -0.036 |
Token | his |
Token ID | 465 |
Feature activation | -1.88e-03 |
Pos logit contributions | |
goodbye | +0.060 |
contain | +0.052 |
plunge | +0.051 |
umb | +0.049 |
toward | +0.048 |
Neg logit contributions | |
ery | -0.047 |
Mic | -0.045 |
Max | -0.045 |
mice | -0.044 |
iflower | -0.042 |
Token | prayer |
Token ID | 11443 |
Feature activation | -8.60e-02 |
Pos logit contributions | |
goodbye | +0.003 |
contain | +0.003 |
plunge | +0.003 |
oured | +0.003 |
umb | +0.003 |
Neg logit contributions | |
ery | -0.003 |
mice | -0.003 |
Mic | -0.003 |
Max | -0.003 |
iflower | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
goodbye | +0.158 |
contain | +0.137 |
plunge | +0.136 |
umb | +0.133 |
oured | +0.127 |
Neg logit contributions | |
ery | -0.142 |
Max | -0.134 |
Mic | -0.133 |
mice | -0.131 |
iflower | -0.129 |
Token | He |
Token ID | 679 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
goodbye | +0.036 |
farewell | +0.028 |
toward | +0.026 |
contain | +0.026 |
hi | +0.026 |
Neg logit contributions | |
ery | -0.033 |
Mic | -0.029 |
mice | -0.029 |
iflower | -0.028 |
iders | -0.028 |
Token | said |
Token ID | 531 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
ery | +0.022 |
Mic | +0.022 |
Max | +0.022 |
mice | +0.021 |
iflower | +0.020 |
Neg logit contributions | |
goodbye | -0.026 |
contain | -0.022 |
plunge | -0.021 |
umb | -0.021 |
oured | -0.020 |
Token | that |
Token ID | 326 |
Feature activation | +4.04e-03 |
Pos logit contributions | |
goodbye | +0.037 |
contain | +0.030 |
plunge | +0.028 |
umb | +0.027 |
attention | +0.026 |
Neg logit contributions | |
ery | -0.062 |
Mic | -0.061 |
Max | -0.060 |
mice | -0.059 |
iflower | -0.057 |
Token | no |
Token ID | 645 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
Mic | +0.008 |
ery | +0.008 |
Max | +0.008 |
Ted | +0.007 |
oyd | +0.007 |
Neg logit contributions | |
goodbye | -0.007 |
contain | -0.005 |
attention | -0.005 |
tickets | -0.005 |
plunge | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +3.05e-03 |
Pos logit contributions | |
ery | +0.000 |
iflower | +0.000 |
chini | +0.000 |
Mic | +0.000 |
ities | +0.000 |
Neg logit contributions | |
goodbye | -0.000 |
contain | -0.000 |
farewell | -0.000 |
hug | -0.000 |
ogging | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | +1.54e-03 |
Pos logit contributions | |
ery | +0.005 |
ities | +0.005 |
mice | +0.004 |
ters | +0.004 |
magnets | +0.004 |
Neg logit contributions | |
goodbye | -0.006 |
contain | -0.005 |
farewell | -0.005 |
ogging | -0.005 |
toward | -0.005 |
Token | Bob |
Token ID | 18861 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
ery | +0.002 |
ities | +0.002 |
magnets | +0.002 |
iflower | +0.002 |
mice | +0.002 |
Neg logit contributions | |
goodbye | -0.003 |
contain | -0.003 |
fy | -0.003 |
farewell | -0.003 |
further | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | -1.67e-03 |
Pos logit contributions | |
goodbye | +0.032 |
contain | +0.027 |
oured | +0.025 |
plunge | +0.025 |
farewell | +0.024 |
Neg logit contributions | |
ery | -0.033 |
Mic | -0.033 |
mice | -0.031 |
Max | -0.031 |
iflower | -0.030 |
Token | Max |
Token ID | 5436 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
goodbye | +0.003 |
contain | +0.002 |
hug | +0.002 |
wave | +0.002 |
farewell | +0.002 |
Neg logit contributions | |
ery | -0.003 |
iflower | -0.003 |
chini | -0.003 |
iders | -0.003 |
hes | -0.003 |
Token | quickly |
Token ID | 2952 |
Feature activation | -8.12e-02 |
Pos logit contributions | |
goodbye | +0.028 |
attention | +0.025 |
plunge | +0.025 |
ogging | +0.024 |
toward | +0.024 |
Neg logit contributions | |
Max | -0.018 |
Mic | -0.017 |
ery | -0.017 |
mice | -0.017 |
Bear | -0.016 |
Token | became |
Token ID | 2627 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
plunge | +0.159 |
goodbye | +0.157 |
attention | +0.155 |
ogging | +0.148 |
contain | +0.148 |
Neg logit contributions | |
mice | -0.095 |
Max | -0.093 |
Mic | -0.089 |
ery | -0.089 |
Midnight | -0.089 |
Token | friends |
Token ID | 2460 |
Feature activation | -2.72e-03 |
Pos logit contributions | |
iflower | +0.025 |
ery | +0.025 |
iders | +0.025 |
oon | +0.025 |
Mic | +0.024 |
Neg logit contributions | |
goodbye | -0.028 |
attention | -0.022 |
umb | -0.022 |
closer | -0.022 |
further | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
plunge | +0.005 |
goodbye | +0.005 |
umb | +0.005 |
contain | +0.005 |
tour | +0.005 |
Neg logit contributions | |
mice | -0.004 |
Max | -0.004 |
ery | -0.003 |
Mic | -0.003 |
Midnight | -0.003 |
Token | Every |
Token ID | 3887 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
goodbye | +0.033 |
contain | +0.026 |
toward | +0.026 |
farewell | +0.026 |
umb | +0.025 |
Neg logit contributions | |
ery | -0.032 |
iflower | -0.030 |
Mic | -0.027 |
web | -0.027 |
chini | -0.026 |
Token | day |
Token ID | 1110 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
goodbye | +0.044 |
contain | +0.038 |
plunge | +0.037 |
umb | +0.036 |
attention | +0.035 |
Neg logit contributions | |
ery | -0.036 |
Mic | -0.036 |
Max | -0.035 |
mice | -0.034 |
iflower | -0.033 |
Token | She |
Token ID | 1375 |
Feature activation | +2.05e-03 |
Pos logit contributions | |
ery | +0.005 |
Mic | +0.004 |
iflower | +0.004 |
chini | +0.004 |
Max | +0.004 |
Neg logit contributions | |
goodbye | -0.006 |
toward | -0.004 |
contain | -0.004 |
umb | -0.004 |
attention | -0.004 |
Token | kissed |
Token ID | 27946 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
Mic | +0.004 |
ery | +0.004 |
mice | +0.004 |
iflower | +0.004 |
Max | +0.004 |
Neg logit contributions | |
goodbye | -0.003 |
contain | -0.003 |
umb | -0.003 |
oured | -0.003 |
wave | -0.003 |
Token | her |
Token ID | 607 |
Feature activation | -2.66e-02 |
Pos logit contributions | |
goodbye | +0.069 |
contain | +0.055 |
attention | +0.054 |
umb | +0.054 |
farewell | +0.053 |
Neg logit contributions | |
ery | -0.054 |
Mic | -0.052 |
iflower | -0.051 |
chini | -0.048 |
Ted | -0.048 |
Token | on |
Token ID | 319 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
goodbye | +0.046 |
contain | +0.039 |
plunge | +0.038 |
umb | +0.037 |
attention | +0.036 |
Neg logit contributions | |
ery | -0.049 |
Mic | -0.048 |
Max | -0.047 |
mice | -0.046 |
iflower | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | -4.00e-02 |
Pos logit contributions | |
goodbye | +0.045 |
contain | +0.034 |
farewell | +0.034 |
attention | +0.033 |
umb | +0.033 |
Neg logit contributions | |
ery | -0.041 |
Mic | -0.038 |
Ted | -0.036 |
iflower | -0.036 |
Bear | -0.035 |
Token | head |
Token ID | 1182 |
Feature activation | -8.21e-02 |
Pos logit contributions | |
goodbye | +0.071 |
oured | +0.067 |
umb | +0.066 |
contain | +0.066 |
fy | +0.065 |
Neg logit contributions | |
ery | -0.063 |
iflower | -0.062 |
Max | -0.060 |
mice | -0.059 |
Mic | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | +1.32e-03 |
Pos logit contributions | |
goodbye | +0.154 |
contain | +0.136 |
plunge | +0.134 |
umb | +0.129 |
attention | +0.125 |
Neg logit contributions | |
ery | -0.133 |
Max | -0.126 |
Mic | -0.125 |
mice | -0.123 |
iflower | -0.118 |
Token | She |
Token ID | 1375 |
Feature activation | -1.25e-03 |
Pos logit contributions | |
ery | +0.002 |
iflower | +0.002 |
Mic | +0.002 |
Max | +0.002 |
Ted | +0.002 |
Neg logit contributions | |
goodbye | -0.003 |
contain | -0.002 |
umb | -0.002 |
toward | -0.002 |
hug | -0.002 |
Token | said |
Token ID | 531 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
goodbye | +0.002 |
contain | +0.002 |
umb | +0.002 |
wave | +0.002 |
oured | +0.002 |
Neg logit contributions | |
Mic | -0.002 |
ery | -0.002 |
mice | -0.002 |
Max | -0.002 |
iflower | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -1.68e-03 |
Pos logit contributions | |
goodbye | +0.030 |
contain | +0.027 |
plunge | +0.025 |
toward | +0.024 |
oured | +0.024 |
Neg logit contributions | |
ery | -0.041 |
mice | -0.040 |
Max | -0.040 |
Mic | -0.040 |
Midnight | -0.039 |
Token | " |
Token ID | 366 |
Feature activation | +3.67e-02 |
Pos logit contributions | |
goodbye | +0.004 |
hi | +0.003 |
hug | +0.003 |
hello | +0.002 |
toward | +0.002 |
Neg logit contributions | |
ery | -0.003 |
Mic | -0.003 |
iflower | -0.003 |
mice | -0.003 |
chini | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | -6.64e-03 |
Pos logit contributions | |
ery | +0.019 |
Mic | +0.018 |
Max | +0.018 |
mice | +0.017 |
iflower | +0.017 |
Neg logit contributions | |
goodbye | -0.020 |
contain | -0.017 |
plunge | -0.016 |
umb | -0.016 |
attention | -0.016 |
Token | big |
Token ID | 1263 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
goodbye | +0.013 |
contain | +0.011 |
plunge | +0.011 |
oured | +0.011 |
fy | +0.011 |
Neg logit contributions | |
ery | -0.010 |
mice | -0.010 |
Max | -0.010 |
Mic | -0.009 |
iflower | -0.009 |
Token | cross |
Token ID | 3272 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
goodbye | +0.074 |
contain | +0.068 |
plunge | +0.066 |
oured | +0.066 |
fy | +0.065 |
Neg logit contributions | |
mice | -0.062 |
ery | -0.058 |
dough | -0.057 |
iflower | -0.056 |
Max | -0.054 |
Token | at |
Token ID | 379 |
Feature activation | -8.17e-03 |
Pos logit contributions | |
goodbye | +0.037 |
contain | +0.032 |
plunge | +0.030 |
umb | +0.030 |
toward | +0.029 |
Neg logit contributions | |
Mic | -0.029 |
ery | -0.029 |
Max | -0.028 |
mice | -0.027 |
iflower | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
goodbye | +0.017 |
contain | +0.015 |
plunge | +0.014 |
umb | +0.014 |
attention | +0.014 |
Neg logit contributions | |
ery | -0.012 |
Mic | -0.012 |
Max | -0.012 |
mice | -0.011 |
iflower | -0.011 |
Token | front |
Token ID | 2166 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
goodbye | +0.036 |
contain | +0.031 |
umb | +0.030 |
plunge | +0.030 |
oured | +0.029 |
Neg logit contributions | |
ery | -0.034 |
mice | -0.033 |
Max | -0.032 |
Mic | -0.032 |
iflower | -0.031 |
Token | of |
Token ID | 286 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
goodbye | +0.180 |
plunge | +0.157 |
contain | +0.152 |
oured | +0.147 |
umb | +0.145 |
Neg logit contributions | |
ery | -0.172 |
mice | -0.171 |
Max | -0.168 |
Mic | -0.164 |
iflower | -0.159 |
Token | the |
Token ID | 262 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
ery | +0.026 |
Mic | +0.025 |
Max | +0.024 |
iflower | +0.024 |
Ted | +0.023 |
Neg logit contributions | |
goodbye | -0.035 |
contain | -0.030 |
umb | -0.029 |
plunge | -0.029 |
attention | -0.028 |
Token | room |
Token ID | 2119 |
Feature activation | +5.76e-03 |
Pos logit contributions | |
goodbye | +0.037 |
contain | +0.031 |
plunge | +0.030 |
umb | +0.030 |
oured | +0.029 |
Neg logit contributions | |
ery | -0.036 |
Max | -0.035 |
mice | -0.035 |
Mic | -0.035 |
iflower | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
ery | +0.009 |
Max | +0.009 |
mice | +0.009 |
Mic | +0.009 |
iflower | +0.008 |
Neg logit contributions | |
goodbye | -0.011 |
contain | -0.009 |
plunge | -0.009 |
umb | -0.009 |
attention | -0.009 |
Token | Her |
Token ID | 2332 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
goodbye | +0.032 |
toward | +0.024 |
umb | +0.024 |
farewell | +0.024 |
hi | +0.023 |
Neg logit contributions | |
ery | -0.032 |
iflower | -0.029 |
Mic | -0.029 |
Max | -0.027 |
Ted | -0.026 |