Token | again |
Token ID | 757 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
amen | +0.111 |
girls | +0.105 |
Rudolph | +0.096 |
Santa | +0.095 |
announcements | +0.095 |
Neg logit contributions | |
run | -0.115 |
bug | -0.107 |
udge | -0.095 |
out | -0.094 |
ours | -0.093 |
Token | . |
Token ID | 13 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
amen | +0.056 |
girls | +0.050 |
announcements | +0.047 |
Rudolph | +0.047 |
hospitality | +0.046 |
Neg logit contributions | |
run | -0.058 |
bug | -0.053 |
udge | -0.047 |
out | -0.047 |
here | -0.046 |
Token | But |
Token ID | 887 |
Feature activation | +6.07e-02 |
Pos logit contributions | |
amen | +0.001 |
girls | +0.001 |
Rudolph | +0.001 |
announcements | +0.001 |
Santa | +0.001 |
Neg logit contributions | |
run | -0.001 |
bug | -0.001 |
ours | -0.001 |
idd | -0.001 |
imb | -0.001 |
Token | unb |
Token ID | 22619 |
Feature activation | +3.11e+00 |
Pos logit contributions | |
run | +0.002 |
bug | +0.002 |
ours | +0.002 |
idd | +0.002 |
imb | +0.002 |
Neg logit contributions | |
amen | -0.001 |
girls | -0.001 |
Rudolph | -0.001 |
announcements | -0.001 |
Santa | -0.001 |
Token | ek |
Token ID | 988 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
run | +0.071 |
bug | +0.068 |
ours | +0.059 |
idd | +0.058 |
imb | +0.057 |
Neg logit contributions | |
amen | -0.130 |
girls | -0.124 |
Rudolph | -0.119 |
announcements | -0.116 |
Santa | -0.115 |
Token | nown |
Token ID | 3408 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
run | +0.033 |
bug | +0.032 |
ours | +0.028 |
idd | +0.028 |
haired | +0.027 |
Neg logit contributions | |
amen | -0.046 |
girls | -0.043 |
Rudolph | -0.041 |
announcements | -0.041 |
hospitality | -0.040 |
Token | to |
Token ID | 284 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
run | +0.218 |
bug | +0.218 |
haired | +0.206 |
ours | +0.203 |
udge | +0.199 |
Neg logit contributions | |
amen | -0.394 |
girls | -0.384 |
Rudolph | -0.369 |
Santa | -0.366 |
sets | -0.358 |
Token | him |
Token ID | 683 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
amen | +0.038 |
girls | +0.037 |
announcements | +0.035 |
sets | +0.034 |
Claus | +0.033 |
Neg logit contributions | |
run | -0.043 |
bug | -0.039 |
imb | -0.035 |
udge | -0.034 |
out | -0.034 |
Token | , |
Token ID | 11 |
Feature activation | -7.08e-01 |
Pos logit contributions | |
amen | +0.036 |
girls | +0.033 |
announcements | +0.033 |
sets | +0.032 |
Santa | +0.031 |
Neg logit contributions | |
run | -0.035 |
bug | -0.033 |
ours | -0.028 |
out | -0.027 |
earing | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
amen | +0.024 |
girls | +0.024 |
announcements | +0.022 |
Rudolph | +0.022 |
sets | +0.022 |
Neg logit contributions | |
run | -0.021 |
bug | -0.020 |
imb | -0.018 |
ours | -0.018 |
idd | -0.017 |
Token | king |
Token ID | 5822 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
amen | +0.070 |
girls | +0.067 |
Rudolph | +0.062 |
announcements | +0.059 |
sets | +0.059 |
Neg logit contributions | |
bug | -0.080 |
run | -0.079 |
ours | -0.071 |
idd | -0.071 |
haired | -0.071 |
Token | day |
Token ID | 1110 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
amen | +0.009 |
girls | +0.009 |
Rudolph | +0.008 |
Santa | +0.008 |
announcements | +0.008 |
Neg logit contributions | |
run | -0.008 |
bug | -0.008 |
ours | -0.007 |
idd | -0.007 |
haired | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
run | +0.074 |
bug | +0.072 |
idd | +0.065 |
eh | +0.065 |
ours | +0.063 |
Neg logit contributions | |
amen | -0.071 |
girls | -0.069 |
announcements | -0.064 |
Santa | -0.063 |
hospitality | -0.062 |
Token | but |
Token ID | 475 |
Feature activation | +3.22e+00 |
Pos logit contributions | |
run | +0.034 |
bug | +0.033 |
ours | +0.030 |
imb | +0.029 |
idd | +0.029 |
Neg logit contributions | |
amen | -0.035 |
girls | -0.033 |
Rudolph | -0.031 |
announcements | -0.031 |
hospitality | -0.030 |
Token | it |
Token ID | 340 |
Feature activation | +3.89e+00 |
Pos logit contributions | |
run | +0.115 |
bug | +0.111 |
ours | +0.103 |
idd | +0.102 |
udge | +0.102 |
Neg logit contributions | |
amen | -0.092 |
girls | -0.085 |
Rudolph | -0.081 |
Santa | -0.077 |
announcements | -0.077 |
Token | was |
Token ID | 373 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
run | +0.119 |
bug | +0.110 |
ours | +0.103 |
udge | +0.102 |
imb | +0.102 |
Neg logit contributions | |
amen | -0.133 |
girls | -0.126 |
Rudolph | -0.121 |
announcements | -0.117 |
Santa | -0.115 |
Token | nowhere |
Token ID | 12062 |
Feature activation | +8.92e+00 |
Pos logit contributions | |
run | +0.079 |
bug | +0.077 |
ours | +0.071 |
udge | +0.070 |
idd | +0.070 |
Neg logit contributions | |
amen | -0.071 |
girls | -0.066 |
Rudolph | -0.064 |
Santa | -0.061 |
announcements | -0.061 |
Token | to |
Token ID | 284 |
Feature activation | -6.61e-01 |
Pos logit contributions | |
ours | +0.306 |
imb | +0.302 |
run | +0.299 |
idd | +0.297 |
bug | +0.295 |
Neg logit contributions | |
amen | -0.230 |
girls | -0.212 |
Rudolph | -0.210 |
mercy | -0.209 |
hospitality | -0.205 |
Token | be |
Token ID | 307 |
Feature activation | -2.88e-01 |
Pos logit contributions | |
amen | +0.019 |
girls | +0.018 |
Rudolph | +0.016 |
Santa | +0.015 |
sets | +0.015 |
Neg logit contributions | |
run | -0.024 |
bug | -0.024 |
haired | -0.021 |
imb | -0.021 |
udge | -0.021 |
Token | found |
Token ID | 1043 |
Feature activation | +5.74e+00 |
Pos logit contributions | |
amen | +0.010 |
girls | +0.010 |
Rudolph | +0.009 |
announcements | +0.009 |
Santa | +0.009 |
Neg logit contributions | |
run | -0.008 |
ours | -0.007 |
bug | -0.007 |
imb | -0.007 |
udge | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
run | +0.239 |
bug | +0.234 |
ours | +0.221 |
idd | +0.217 |
imb | +0.217 |
Neg logit contributions | |
amen | -0.126 |
girls | -0.117 |
Rudolph | -0.108 |
Santa | -0.102 |
announcements | -0.101 |
Token | He |
Token ID | 679 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
run | +0.035 |
bug | +0.035 |
ours | +0.032 |
udge | +0.031 |
idd | +0.031 |
Neg logit contributions | |
amen | -0.029 |
girls | -0.027 |
Rudolph | -0.026 |
Santa | -0.025 |
announcements | -0.024 |
Token | take |
Token ID | 1011 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
amen | +0.021 |
girls | +0.019 |
Rudolph | +0.018 |
Santa | +0.018 |
announcements | +0.018 |
Neg logit contributions | |
run | -0.025 |
bug | -0.025 |
ours | -0.022 |
imb | -0.022 |
udge | -0.022 |
Token | a |
Token ID | 257 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
run | +0.075 |
bug | +0.073 |
ours | +0.068 |
idd | +0.067 |
imb | +0.067 |
Neg logit contributions | |
amen | -0.066 |
girls | -0.063 |
Rudolph | -0.061 |
announcements | -0.057 |
Santa | -0.057 |
Token | trip |
Token ID | 5296 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
ours | +0.007 |
idd | +0.007 |
run | +0.007 |
haired | +0.007 |
ays | +0.007 |
Neg logit contributions | |
amen | -0.005 |
Rudolph | -0.005 |
hospitality | -0.005 |
mercy | -0.005 |
girls | -0.004 |
Token | but |
Token ID | 475 |
Feature activation | +2.91e+00 |
Pos logit contributions | |
run | +0.011 |
bug | +0.011 |
out | +0.010 |
udge | +0.009 |
haired | +0.009 |
Neg logit contributions | |
amen | -0.014 |
girls | -0.013 |
announcements | -0.013 |
Rudolph | -0.012 |
hospitality | -0.012 |
Token | had |
Token ID | 550 |
Feature activation | +4.36e+00 |
Pos logit contributions | |
run | +0.107 |
bug | +0.105 |
ours | +0.095 |
idd | +0.094 |
imb | +0.094 |
Neg logit contributions | |
amen | -0.080 |
girls | -0.074 |
Rudolph | -0.069 |
announcements | -0.068 |
Santa | -0.067 |
Token | nowhere |
Token ID | 12062 |
Feature activation | +8.81e+00 |
Pos logit contributions | |
run | +0.123 |
bug | +0.115 |
ours | +0.100 |
haired | +0.097 |
out | +0.097 |
Neg logit contributions | |
amen | -0.163 |
girls | -0.153 |
Rudolph | -0.142 |
Claus | -0.140 |
Santa | -0.140 |
Token | to |
Token ID | 284 |
Feature activation | -6.21e-01 |
Pos logit contributions | |
imb | +0.293 |
ours | +0.289 |
bug | +0.284 |
idd | +0.283 |
run | +0.282 |
Neg logit contributions | |
amen | -0.230 |
girls | -0.222 |
mercy | -0.218 |
Rudolph | -0.218 |
hospitality | -0.210 |
Token | go |
Token ID | 467 |
Feature activation | +3.14e-01 |
Pos logit contributions | |
amen | +0.020 |
girls | +0.019 |
Rudolph | +0.018 |
sets | +0.018 |
Claus | +0.017 |
Neg logit contributions | |
run | -0.020 |
bug | -0.020 |
udge | -0.017 |
imb | -0.017 |
nail | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
run | +0.012 |
bug | +0.011 |
udge | +0.010 |
imb | +0.010 |
ours | +0.010 |
Neg logit contributions | |
amen | -0.009 |
girls | -0.008 |
Rudolph | -0.007 |
announcements | -0.007 |
hospitality | -0.007 |
Token | One |
Token ID | 1881 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
run | +0.045 |
bug | +0.044 |
ours | +0.041 |
idd | +0.040 |
udge | +0.040 |
Neg logit contributions | |
amen | -0.031 |
girls | -0.029 |
Rudolph | -0.027 |
Santa | -0.026 |
announcements | -0.025 |
Token | day |
Token ID | 1110 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
run | +0.053 |
ours | +0.052 |
idd | +0.051 |
udge | +0.049 |
bug | +0.049 |
Neg logit contributions | |
amen | -0.035 |
Rudolph | -0.034 |
hospitality | -0.031 |
girls | -0.031 |
stories | -0.030 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
run | +0.052 |
bug | +0.049 |
udge | +0.047 |
imb | +0.047 |
idd | +0.046 |
Neg logit contributions | |
amen | -0.056 |
Rudolph | -0.051 |
girls | -0.051 |
announcements | -0.050 |
Santa | -0.048 |
Token | , |
Token ID | 11 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
run | +0.062 |
imb | +0.059 |
bug | +0.058 |
udge | +0.055 |
idd | +0.055 |
Neg logit contributions | |
amen | -0.064 |
Rudolph | -0.059 |
stories | -0.057 |
announcements | -0.056 |
Santa | -0.055 |
Token | but |
Token ID | 475 |
Feature activation | +2.81e+00 |
Pos logit contributions | |
run | +0.008 |
bug | +0.008 |
ours | +0.007 |
idd | +0.007 |
udge | +0.007 |
Neg logit contributions | |
amen | -0.007 |
girls | -0.007 |
Rudolph | -0.006 |
announcements | -0.006 |
Santa | -0.006 |
Token | it |
Token ID | 340 |
Feature activation | +3.74e+00 |
Pos logit contributions | |
run | +0.103 |
udge | +0.097 |
idd | +0.097 |
bug | +0.096 |
imb | +0.095 |
Neg logit contributions | |
amen | -0.074 |
Rudolph | -0.068 |
girls | -0.065 |
Santa | -0.061 |
Santa | -0.061 |
Token | was |
Token ID | 373 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
run | +0.125 |
imb | +0.114 |
bug | +0.114 |
udge | +0.112 |
ours | +0.112 |
Neg logit contributions | |
amen | -0.108 |
girls | -0.105 |
Rudolph | -0.101 |
sets | -0.097 |
announcements | -0.096 |
Token | nowhere |
Token ID | 12062 |
Feature activation | +8.71e+00 |
Pos logit contributions | |
run | +0.070 |
bug | +0.067 |
imb | +0.064 |
ours | +0.063 |
idd | +0.063 |
Neg logit contributions | |
amen | -0.064 |
girls | -0.060 |
Rudolph | -0.059 |
Santa | -0.056 |
announcements | -0.056 |
Token | to |
Token ID | 284 |
Feature activation | -7.67e-01 |
Pos logit contributions | |
imb | +0.381 |
ours | +0.370 |
idd | +0.359 |
ang | +0.355 |
rous | +0.347 |
Neg logit contributions | |
mercy | -0.128 |
Claus | -0.119 |
sets | -0.115 |
girls | -0.114 |
Rudolph | -0.113 |
Token | be |
Token ID | 307 |
Feature activation | -5.24e-01 |
Pos logit contributions | |
amen | +0.018 |
girls | +0.017 |
Rudolph | +0.016 |
announcements | +0.015 |
Santa | +0.015 |
Neg logit contributions | |
run | -0.031 |
bug | -0.030 |
ours | -0.029 |
idd | -0.028 |
imb | -0.028 |
Token | found |
Token ID | 1043 |
Feature activation | +5.60e+00 |
Pos logit contributions | |
amen | +0.017 |
girls | +0.017 |
announcements | +0.016 |
Santa | +0.016 |
Rudolph | +0.016 |
Neg logit contributions | |
ours | -0.014 |
run | -0.014 |
imb | -0.013 |
ang | -0.013 |
bug | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +9.81e-01 |
Pos logit contributions | |
run | +0.243 |
bug | +0.237 |
ours | +0.228 |
imb | +0.224 |
idd | +0.224 |
Neg logit contributions | |
amen | -0.108 |
girls | -0.101 |
Rudolph | -0.092 |
Santa | -0.086 |
announcements | -0.085 |
Token | |
Token ID | 220 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
run | +0.036 |
bug | +0.035 |
ours | +0.034 |
udge | +0.033 |
idd | +0.033 |
Neg logit contributions | |
amen | -0.025 |
girls | -0.023 |
Rudolph | -0.023 |
Santa | -0.022 |
announcements | -0.021 |
Token | around |
Token ID | 1088 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
run | +0.027 |
bug | +0.025 |
imb | +0.025 |
ours | +0.024 |
idd | +0.024 |
Neg logit contributions | |
amen | -0.044 |
Rudolph | -0.041 |
girls | -0.041 |
sets | -0.040 |
stories | -0.040 |
Token | , |
Token ID | 11 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
run | +0.095 |
bug | +0.093 |
ours | +0.084 |
udge | +0.083 |
idd | +0.082 |
Neg logit contributions | |
amen | -0.087 |
girls | -0.082 |
Rudolph | -0.077 |
announcements | -0.075 |
Santa | -0.074 |
Token | but |
Token ID | 475 |
Feature activation | +2.85e+00 |
Pos logit contributions | |
run | +0.009 |
bug | +0.009 |
ours | +0.009 |
idd | +0.008 |
imb | +0.008 |
Neg logit contributions | |
amen | -0.008 |
girls | -0.007 |
Rudolph | -0.007 |
announcements | -0.007 |
Santa | -0.007 |
Token | it |
Token ID | 340 |
Feature activation | +3.79e+00 |
Pos logit contributions | |
run | +0.099 |
bug | +0.094 |
idd | +0.092 |
udge | +0.091 |
ours | +0.091 |
Neg logit contributions | |
amen | -0.080 |
Rudolph | -0.074 |
girls | -0.072 |
announcements | -0.068 |
Santa | -0.068 |
Token | was |
Token ID | 373 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
run | +0.123 |
bug | +0.115 |
ours | +0.112 |
imb | +0.112 |
idd | +0.110 |
Neg logit contributions | |
amen | -0.114 |
girls | -0.109 |
Rudolph | -0.105 |
announcements | -0.102 |
sets | -0.101 |
Token | nowhere |
Token ID | 12062 |
Feature activation | +8.66e+00 |
Pos logit contributions | |
run | +0.069 |
bug | +0.067 |
ours | +0.064 |
imb | +0.064 |
idd | +0.063 |
Neg logit contributions | |
amen | -0.062 |
girls | -0.058 |
Rudolph | -0.057 |
Santa | -0.054 |
announcements | -0.054 |
Token | to |
Token ID | 284 |
Feature activation | -8.08e-01 |
Pos logit contributions | |
ours | +0.346 |
imb | +0.344 |
idd | +0.331 |
ang | +0.322 |
run | +0.318 |
Neg logit contributions | |
mercy | -0.160 |
amen | -0.149 |
sets | -0.148 |
girls | -0.144 |
Rudolph | -0.142 |
Token | be |
Token ID | 307 |
Feature activation | -5.19e-01 |
Pos logit contributions | |
amen | +0.020 |
girls | +0.018 |
Rudolph | +0.017 |
announcements | +0.016 |
Santa | +0.016 |
Neg logit contributions | |
run | -0.032 |
bug | -0.032 |
ours | -0.030 |
idd | -0.029 |
imb | -0.029 |
Token | found |
Token ID | 1043 |
Feature activation | +5.57e+00 |
Pos logit contributions | |
amen | +0.018 |
girls | +0.017 |
announcements | +0.017 |
Rudolph | +0.016 |
Santa | +0.016 |
Neg logit contributions | |
ours | -0.013 |
run | -0.013 |
imb | -0.012 |
bug | -0.012 |
ang | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +8.77e-01 |
Pos logit contributions | |
run | +0.231 |
bug | +0.226 |
ours | +0.216 |
idd | +0.212 |
imb | +0.212 |
Neg logit contributions | |
amen | -0.119 |
girls | -0.111 |
Rudolph | -0.103 |
Santa | -0.097 |
announcements | -0.096 |
Token | |
Token ID | 220 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
run | +0.032 |
bug | +0.032 |
ours | +0.031 |
udge | +0.030 |
idd | +0.030 |
Neg logit contributions | |
amen | -0.022 |
girls | -0.020 |
Rudolph | -0.020 |
Santa | -0.019 |
announcements | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | +9.10e-01 |
Pos logit contributions | |
amen | +0.080 |
girls | +0.080 |
Santa | +0.071 |
announcements | +0.071 |
sets | +0.069 |
Neg logit contributions | |
run | -0.065 |
bug | -0.064 |
out | -0.059 |
elly | -0.059 |
udge | -0.055 |
Token | but |
Token ID | 475 |
Feature activation | +2.97e+00 |
Pos logit contributions | |
run | +0.031 |
bug | +0.030 |
ours | +0.026 |
haired | +0.026 |
imb | +0.026 |
Neg logit contributions | |
amen | -0.027 |
girls | -0.027 |
announcements | -0.024 |
Rudolph | -0.023 |
hospitality | -0.023 |
Token | still |
Token ID | 991 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
run | +0.105 |
bug | +0.104 |
ours | +0.093 |
imb | +0.092 |
idd | +0.091 |
Neg logit contributions | |
amen | -0.083 |
girls | -0.080 |
announcements | -0.073 |
Rudolph | -0.072 |
Santa | -0.071 |
Token | True |
Token ID | 6407 |
Feature activation | +3.81e-01 |
Pos logit contributions | |
run | +0.052 |
bug | +0.051 |
ours | +0.046 |
idd | +0.045 |
imb | +0.045 |
Neg logit contributions | |
amen | -0.046 |
girls | -0.044 |
Rudolph | -0.041 |
announcements | -0.040 |
Santa | -0.040 |
Token | was |
Token ID | 373 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
bug | +0.012 |
run | +0.011 |
ours | +0.010 |
haired | +0.010 |
idd | +0.009 |
Neg logit contributions | |
amen | -0.013 |
girls | -0.012 |
announcements | -0.012 |
Santa | -0.012 |
Rudolph | -0.011 |
Token | nowhere |
Token ID | 12062 |
Feature activation | +8.60e+00 |
Pos logit contributions | |
run | +0.072 |
bug | +0.070 |
ours | +0.064 |
idd | +0.063 |
imb | +0.063 |
Neg logit contributions | |
amen | -0.064 |
girls | -0.060 |
Rudolph | -0.057 |
announcements | -0.055 |
Santa | -0.055 |
Token | to |
Token ID | 284 |
Feature activation | -9.02e-01 |
Pos logit contributions | |
ours | +0.277 |
run | +0.277 |
bug | +0.274 |
imb | +0.273 |
idd | +0.266 |
Neg logit contributions | |
amen | -0.248 |
girls | -0.230 |
Rudolph | -0.224 |
mercy | -0.221 |
hospitality | -0.216 |
Token | be |
Token ID | 307 |
Feature activation | -4.87e-01 |
Pos logit contributions | |
amen | +0.030 |
girls | +0.029 |
Rudolph | +0.027 |
sets | +0.026 |
Claus | +0.026 |
Neg logit contributions | |
run | -0.027 |
bug | -0.027 |
haired | -0.023 |
nail | -0.023 |
imb | -0.023 |
Token | found |
Token ID | 1043 |
Feature activation | +5.55e+00 |
Pos logit contributions | |
amen | +0.016 |
girls | +0.015 |
Rudolph | +0.014 |
announcements | +0.014 |
Santa | +0.014 |
Neg logit contributions | |
run | -0.015 |
bug | -0.015 |
ours | -0.013 |
idd | -0.013 |
udge | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +8.18e-01 |
Pos logit contributions | |
run | +0.210 |
bug | +0.204 |
ours | +0.187 |
udge | +0.185 |
idd | +0.184 |
Neg logit contributions | |
amen | -0.148 |
girls | -0.137 |
Rudolph | -0.128 |
announcements | -0.125 |
Santa | -0.123 |
Token |
|
Token ID | 198 |
Feature activation | -3.91e-02 |
Pos logit contributions | |
run | +0.031 |
bug | +0.030 |
ours | +0.028 |
idd | +0.027 |
udge | +0.027 |
Neg logit contributions | |
amen | -0.022 |
girls | -0.020 |
Rudolph | -0.019 |
Santa | -0.018 |
announcements | -0.018 |
Token | everywhere |
Token ID | 8347 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
run | +0.036 |
bug | +0.035 |
idd | +0.032 |
ours | +0.032 |
imb | +0.032 |
Neg logit contributions | |
amen | -0.032 |
girls | -0.030 |
Rudolph | -0.029 |
announcements | -0.028 |
Santa | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | +2.66e-01 |
Pos logit contributions | |
run | +0.004 |
bug | +0.004 |
ours | +0.004 |
idd | +0.004 |
imb | +0.004 |
Neg logit contributions | |
amen | -0.003 |
girls | -0.002 |
Rudolph | -0.002 |
Santa | -0.002 |
announcements | -0.002 |
Token | but |
Token ID | 475 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
run | +0.011 |
bug | +0.010 |
ours | +0.010 |
idd | +0.010 |
imb | +0.010 |
Neg logit contributions | |
amen | -0.006 |
Rudolph | -0.005 |
girls | -0.005 |
announcements | -0.005 |
Santa | -0.005 |
Token | it |
Token ID | 340 |
Feature activation | +3.51e+00 |
Pos logit contributions | |
run | +0.109 |
idd | +0.107 |
elly | +0.105 |
udge | +0.103 |
ours | +0.103 |
Neg logit contributions | |
amen | -0.055 |
Rudolph | -0.053 |
Santa | -0.048 |
girls | -0.045 |
Santa | -0.044 |
Token | was |
Token ID | 373 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
run | +0.133 |
ours | +0.126 |
imb | +0.125 |
udge | +0.124 |
idd | +0.123 |
Neg logit contributions | |
girls | -0.080 |
amen | -0.078 |
Rudolph | -0.075 |
Claus | -0.072 |
sets | -0.071 |
Token | nowhere |
Token ID | 12062 |
Feature activation | +8.59e+00 |
Pos logit contributions | |
run | +0.056 |
bug | +0.055 |
imb | +0.053 |
ours | +0.052 |
idd | +0.052 |
Neg logit contributions | |
amen | -0.069 |
Rudolph | -0.065 |
girls | -0.065 |
mercy | -0.062 |
Santa | -0.061 |
Token | to |
Token ID | 284 |
Feature activation | -8.89e-01 |
Pos logit contributions | |
imb | +0.382 |
ours | +0.376 |
idd | +0.359 |
ang | +0.352 |
ays | +0.349 |
Neg logit contributions | |
mercy | -0.132 |
Claus | -0.124 |
Rudolph | -0.115 |
girls | -0.113 |
amen | -0.109 |
Token | be |
Token ID | 307 |
Feature activation | -6.68e-01 |
Pos logit contributions | |
amen | +0.022 |
girls | +0.020 |
Rudolph | +0.019 |
announcements | +0.018 |
Santa | +0.018 |
Neg logit contributions | |
run | -0.035 |
bug | -0.034 |
ours | -0.032 |
idd | -0.031 |
imb | -0.031 |
Token | found |
Token ID | 1043 |
Feature activation | +5.55e+00 |
Pos logit contributions | |
amen | +0.024 |
girls | +0.023 |
Rudolph | +0.022 |
announcements | +0.022 |
Santa | +0.022 |
Neg logit contributions | |
ours | -0.016 |
run | -0.015 |
imb | -0.015 |
bug | -0.015 |
ang | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +8.62e-01 |
Pos logit contributions | |
run | +0.244 |
bug | +0.239 |
ours | +0.234 |
imb | +0.232 |
idd | +0.229 |
Neg logit contributions | |
amen | -0.095 |
girls | -0.091 |
Rudolph | -0.082 |
Santa | -0.077 |
hospitality | -0.074 |
Token | Sadly |
Token ID | 27213 |
Feature activation | -2.99e+00 |
Pos logit contributions | |
bug | +0.034 |
run | +0.034 |
ours | +0.033 |
idd | +0.032 |
rous | +0.032 |
Neg logit contributions | |
amen | -0.019 |
girls | -0.017 |
Rudolph | -0.017 |
Santa | -0.016 |
announcements | -0.014 |
Token | big |
Token ID | 1263 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
run | +0.006 |
bug | +0.006 |
ours | +0.005 |
nail | +0.005 |
idd | +0.005 |
Neg logit contributions | |
amen | -0.006 |
Rudolph | -0.005 |
girls | -0.005 |
Santa | -0.005 |
sets | -0.005 |
Token | cats |
Token ID | 11875 |
Feature activation | -4.68e+00 |
Pos logit contributions | |
run | +0.073 |
bug | +0.071 |
ours | +0.065 |
idd | +0.064 |
imb | +0.063 |
Neg logit contributions | |
amen | -0.061 |
girls | -0.057 |
Rudolph | -0.054 |
Santa | -0.052 |
announcements | -0.051 |
Token | came |
Token ID | 1625 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
amen | +0.150 |
girls | +0.140 |
Rudolph | +0.134 |
Santa | +0.130 |
announcements | +0.129 |
Neg logit contributions | |
run | -0.151 |
bug | -0.147 |
ours | -0.132 |
idd | -0.131 |
imb | -0.129 |
Token | out |
Token ID | 503 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
run | +0.031 |
bug | +0.030 |
ours | +0.028 |
udge | +0.027 |
idd | +0.027 |
Neg logit contributions | |
amen | -0.040 |
girls | -0.038 |
Rudolph | -0.036 |
announcements | -0.035 |
Santa | -0.035 |
Token | of |
Token ID | 286 |
Feature activation | -5.90e-02 |
Pos logit contributions | |
run | +0.063 |
bug | +0.061 |
ours | +0.053 |
udge | +0.052 |
idd | +0.052 |
Neg logit contributions | |
amen | -0.088 |
girls | -0.083 |
Rudolph | -0.080 |
Santa | -0.078 |
announcements | -0.078 |
Token | nowhere |
Token ID | 12062 |
Feature activation | +8.58e+00 |
Pos logit contributions | |
amen | +0.001 |
girls | +0.001 |
Rudolph | +0.001 |
announcements | +0.001 |
hospitality | +0.001 |
Neg logit contributions | |
run | -0.002 |
bug | -0.002 |
ours | -0.002 |
imb | -0.002 |
idd | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
run | +0.337 |
bug | +0.330 |
ours | +0.319 |
imb | +0.313 |
idd | +0.310 |
Neg logit contributions | |
amen | -0.201 |
girls | -0.188 |
Rudolph | -0.173 |
hospitality | -0.165 |
announcements | -0.164 |
Token | told |
Token ID | 1297 |
Feature activation | -9.40e-01 |
Pos logit contributions | |
run | +0.051 |
bug | +0.050 |
ours | +0.045 |
idd | +0.044 |
haired | +0.044 |
Neg logit contributions | |
amen | -0.049 |
girls | -0.046 |
Rudolph | -0.043 |
announcements | -0.042 |
Santa | -0.042 |
Token | her |
Token ID | 607 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
amen | +0.031 |
girls | +0.028 |
Santa | +0.027 |
Claus | +0.026 |
sets | +0.026 |
Neg logit contributions | |
run | -0.030 |
bug | -0.028 |
haired | -0.025 |
ours | -0.025 |
udge | -0.024 |
Token | off |
Token ID | 572 |
Feature activation | -2.99e+00 |
Pos logit contributions | |
run | +0.042 |
bug | +0.041 |
ours | +0.036 |
haired | +0.035 |
idd | +0.034 |
Neg logit contributions | |
amen | -0.046 |
girls | -0.044 |
Rudolph | -0.042 |
Santa | -0.041 |
announcements | -0.040 |
Token | for |
Token ID | 329 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
amen | +0.090 |
girls | +0.086 |
Rudolph | +0.080 |
Santa | +0.077 |
announcements | +0.077 |
Neg logit contributions | |
run | -0.106 |
bug | -0.101 |
ours | -0.089 |
haired | -0.088 |
idd | -0.087 |
Token | ball |
Token ID | 2613 |
Feature activation | +4.25e+00 |
Pos logit contributions | |
run | +0.176 |
bug | +0.168 |
ours | +0.154 |
imb | +0.151 |
udge | +0.151 |
Neg logit contributions | |
amen | -0.187 |
girls | -0.176 |
Rudolph | -0.169 |
Santa | -0.162 |
announcements | -0.162 |
Token | suddenly |
Token ID | 6451 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
run | +0.158 |
imb | +0.154 |
udge | +0.154 |
rous | +0.146 |
ours | +0.146 |
Neg logit contributions | |
amen | -0.105 |
girls | -0.096 |
Rudolph | -0.094 |
stories | -0.088 |
thank | -0.086 |
Token | appeared |
Token ID | 4120 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
run | +0.056 |
bug | +0.054 |
ours | +0.050 |
imb | +0.050 |
udge | +0.050 |
Neg logit contributions | |
amen | -0.059 |
girls | -0.056 |
Rudolph | -0.053 |
announcements | -0.052 |
Santa | -0.051 |
Token | out |
Token ID | 503 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
run | +0.041 |
bug | +0.040 |
ours | +0.037 |
idd | +0.036 |
udge | +0.036 |
Neg logit contributions | |
amen | -0.039 |
girls | -0.037 |
Rudolph | -0.035 |
announcements | -0.034 |
Santa | -0.034 |
Token | of |
Token ID | 286 |
Feature activation | +5.09e-03 |
Pos logit contributions | |
run | +0.060 |
bug | +0.058 |
ours | +0.051 |
idd | +0.050 |
imb | +0.050 |
Neg logit contributions | |
amen | -0.092 |
girls | -0.088 |
Rudolph | -0.084 |
announcements | -0.082 |
Santa | -0.082 |
Token | nowhere |
Token ID | 12062 |
Feature activation | +8.54e+00 |
Pos logit contributions | |
run | +0.000 |
bug | +0.000 |
ours | +0.000 |
imb | +0.000 |
idd | +0.000 |
Neg logit contributions | |
amen | -0.000 |
girls | -0.000 |
Rudolph | -0.000 |
announcements | -0.000 |
Santa | -0.000 |
Token | ! |
Token ID | 0 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
run | +0.356 |
ours | +0.354 |
imb | +0.346 |
bug | +0.342 |
idd | +0.338 |
Neg logit contributions | |
amen | -0.164 |
girls | -0.156 |
Rudolph | -0.138 |
hospitality | -0.136 |
mercy | -0.133 |
Token | It |
Token ID | 632 |
Feature activation | +3.61e+00 |
Pos logit contributions | |
run | +0.087 |
bug | +0.086 |
ours | +0.081 |
udge | +0.079 |
idd | +0.079 |
Neg logit contributions | |
amen | -0.068 |
girls | -0.063 |
Rudolph | -0.061 |
Santa | -0.058 |
announcements | -0.056 |
Token | had |
Token ID | 550 |
Feature activation | +4.00e+00 |
Pos logit contributions | |
udge | +0.127 |
imb | +0.127 |
run | +0.123 |
ours | +0.122 |
idd | +0.118 |
Neg logit contributions | |
amen | -0.098 |
girls | -0.091 |
Rudolph | -0.089 |
announcements | -0.087 |
Claus | -0.082 |
Token | been |
Token ID | 587 |
Feature activation | +6.92e-01 |
Pos logit contributions | |
run | +0.132 |
bug | +0.128 |
ours | +0.117 |
idd | +0.115 |
imb | +0.114 |
Neg logit contributions | |
amen | -0.126 |
girls | -0.118 |
Rudolph | -0.112 |
announcements | -0.108 |
Santa | -0.108 |
Token | hiding |
Token ID | 11816 |
Feature activation | +3.68e+00 |
Pos logit contributions | |
run | +0.020 |
bug | +0.020 |
ours | +0.019 |
imb | +0.019 |
udge | +0.018 |
Neg logit contributions | |
amen | -0.023 |
Rudolph | -0.022 |
girls | -0.021 |
Santa | -0.020 |
announcements | -0.020 |
Token | loud |
Token ID | 7812 |
Feature activation | -2.97e+00 |
Pos logit contributions | |
amen | +0.002 |
girls | +0.002 |
Rudolph | +0.002 |
announcements | +0.002 |
Santa | +0.002 |
Neg logit contributions | |
run | -0.002 |
bug | -0.002 |
ours | -0.002 |
imb | -0.002 |
idd | -0.002 |
Token | tune |
Token ID | 14009 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
amen | +0.076 |
girls | +0.072 |
Rudolph | +0.065 |
Santa | +0.064 |
hospitality | +0.063 |
Neg logit contributions | |
run | -0.115 |
bug | -0.113 |
idd | -0.102 |
haired | -0.102 |
ours | -0.102 |
Token | came |
Token ID | 1625 |
Feature activation | +7.75e-01 |
Pos logit contributions | |
amen | +0.064 |
girls | +0.060 |
Rudolph | +0.057 |
announcements | +0.056 |
Santa | +0.055 |
Neg logit contributions | |
run | -0.064 |
bug | -0.062 |
ours | -0.057 |
imb | -0.057 |
udge | -0.056 |
Token | out |
Token ID | 503 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
run | +0.020 |
bug | +0.019 |
imb | +0.016 |
idd | +0.016 |
ours | +0.016 |
Neg logit contributions | |
amen | -0.031 |
girls | -0.029 |
Rudolph | -0.028 |
announcements | -0.027 |
hospitality | -0.027 |
Token | of |
Token ID | 286 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
run | +0.060 |
bug | +0.056 |
udge | +0.049 |
ours | +0.047 |
idd | +0.047 |
Neg logit contributions | |
amen | -0.080 |
girls | -0.076 |
Rudolph | -0.074 |
Santa | -0.070 |
announcements | -0.070 |
Token | nowhere |
Token ID | 12062 |
Feature activation | +8.54e+00 |
Pos logit contributions | |
amen | +0.004 |
girls | +0.003 |
Rudolph | +0.003 |
announcements | +0.003 |
Santa | +0.003 |
Neg logit contributions | |
run | -0.005 |
bug | -0.005 |
ours | -0.004 |
idd | -0.004 |
udge | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
run | +0.326 |
bug | +0.318 |
ours | +0.304 |
imb | +0.298 |
idd | +0.295 |
Neg logit contributions | |
amen | -0.215 |
girls | -0.200 |
Rudolph | -0.185 |
announcements | -0.179 |
Santa | -0.177 |
Token | filled |
Token ID | 5901 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
run | +0.046 |
bug | +0.045 |
ours | +0.041 |
udge | +0.041 |
imb | +0.041 |
Neg logit contributions | |
amen | -0.040 |
girls | -0.037 |
Rudolph | -0.036 |
announcements | -0.034 |
Santa | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
amen | +0.031 |
girls | +0.029 |
Rudolph | +0.028 |
announcements | +0.026 |
Santa | +0.026 |
Neg logit contributions | |
run | -0.044 |
bug | -0.042 |
ours | -0.040 |
imb | -0.039 |
idd | -0.039 |
Token | room |
Token ID | 2119 |
Feature activation | -8.99e-01 |
Pos logit contributions | |
amen | +0.062 |
girls | +0.058 |
Rudolph | +0.055 |
announcements | +0.054 |
Santa | +0.053 |
Neg logit contributions | |
run | -0.072 |
bug | -0.069 |
ours | -0.064 |
idd | -0.063 |
imb | -0.063 |
Token | with |
Token ID | 351 |
Feature activation | +3.31e+00 |
Pos logit contributions | |
amen | +0.028 |
girls | +0.026 |
Rudolph | +0.025 |
announcements | +0.024 |
Santa | +0.024 |
Neg logit contributions | |
run | -0.029 |
bug | -0.028 |
ours | -0.026 |
idd | -0.026 |
imb | -0.026 |
Token | little |
Token ID | 1310 |
Feature activation | +3.60e+00 |
Pos logit contributions | |
amen | +0.073 |
girls | +0.069 |
Rudolph | +0.065 |
announcements | +0.063 |
Santa | +0.062 |
Neg logit contributions | |
run | -0.083 |
bug | -0.081 |
ours | -0.074 |
idd | -0.073 |
imb | -0.073 |
Token | girl |
Token ID | 2576 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
run | +0.119 |
bug | +0.114 |
ours | +0.106 |
udge | +0.104 |
idd | +0.103 |
Neg logit contributions | |
amen | -0.112 |
girls | -0.105 |
Rudolph | -0.101 |
announcements | -0.097 |
Santa | -0.096 |
Token | knew |
Token ID | 2993 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
run | +0.036 |
bug | +0.034 |
ours | +0.033 |
udge | +0.032 |
imb | +0.032 |
Neg logit contributions | |
amen | -0.032 |
girls | -0.030 |
Rudolph | -0.028 |
announcements | -0.027 |
Santa | -0.027 |
Token | there |
Token ID | 612 |
Feature activation | +4.17e+00 |
Pos logit contributions | |
run | +0.064 |
bug | +0.061 |
ours | +0.052 |
haired | +0.050 |
udge | +0.050 |
Neg logit contributions | |
amen | -0.082 |
girls | -0.080 |
announcements | -0.073 |
Rudolph | -0.073 |
Claus | -0.072 |
Token | was |
Token ID | 373 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
run | +0.122 |
bug | +0.118 |
ours | +0.104 |
haired | +0.103 |
imb | +0.102 |
Neg logit contributions | |
amen | -0.147 |
girls | -0.139 |
Rudolph | -0.133 |
Santa | -0.129 |
announcements | -0.129 |
Token | nowhere |
Token ID | 12062 |
Feature activation | +8.52e+00 |
Pos logit contributions | |
run | +0.054 |
bug | +0.052 |
ours | +0.046 |
haired | +0.046 |
imb | +0.045 |
Neg logit contributions | |
amen | -0.055 |
girls | -0.052 |
Rudolph | -0.049 |
announcements | -0.048 |
Santa | -0.047 |
Token | to |
Token ID | 284 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
ours | +0.301 |
imb | +0.291 |
run | +0.290 |
idd | +0.286 |
ang | +0.282 |
Neg logit contributions | |
amen | -0.208 |
girls | -0.200 |
mercy | -0.200 |
Rudolph | -0.196 |
hospitality | -0.194 |
Token | hide |
Token ID | 7808 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
amen | +0.031 |
girls | +0.028 |
Rudolph | +0.027 |
Santa | +0.025 |
Claus | +0.025 |
Neg logit contributions | |
run | -0.040 |
bug | -0.039 |
haired | -0.035 |
udge | -0.035 |
imb | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | +4.98e-01 |
Pos logit contributions | |
amen | +0.032 |
girls | +0.031 |
Rudolph | +0.029 |
announcements | +0.028 |
Santa | +0.028 |
Neg logit contributions | |
run | -0.032 |
bug | -0.031 |
ours | -0.029 |
idd | -0.029 |
imb | -0.028 |
Token | She |
Token ID | 1375 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
run | +0.017 |
bug | +0.017 |
ours | +0.016 |
idd | +0.015 |
udge | +0.015 |
Neg logit contributions | |
amen | -0.015 |
girls | -0.014 |
Rudolph | -0.013 |
announcements | -0.012 |
Santa | -0.012 |
Token | felt |
Token ID | 2936 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
amen | +0.050 |
girls | +0.047 |
Rudolph | +0.045 |
announcements | +0.044 |
Santa | +0.043 |
Neg logit contributions | |
run | -0.054 |
bug | -0.052 |
ours | -0.049 |
imb | -0.049 |
udge | -0.049 |
Token | and |
Token ID | 290 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
run | +0.083 |
bug | +0.080 |
ours | +0.072 |
udge | +0.072 |
haired | +0.071 |
Neg logit contributions | |
amen | -0.079 |
girls | -0.073 |
Rudolph | -0.070 |
Santa | -0.069 |
announcements | -0.068 |
Token | enjoy |
Token ID | 2883 |
Feature activation | -2.97e+00 |
Pos logit contributions | |
run | +0.035 |
bug | +0.034 |
ours | +0.031 |
imb | +0.030 |
haired | +0.030 |
Neg logit contributions | |
amen | -0.035 |
girls | -0.033 |
Rudolph | -0.031 |
announcements | -0.030 |
Santa | -0.030 |
Token | it |
Token ID | 340 |
Feature activation | +3.45e+00 |
Pos logit contributions | |
amen | +0.101 |
girls | +0.094 |
announcements | +0.089 |
Claus | +0.087 |
hospitality | +0.086 |
Neg logit contributions | |
run | -0.091 |
bug | -0.084 |
idd | -0.077 |
udge | -0.077 |
ours | -0.076 |
Token | until |
Token ID | 1566 |
Feature activation | -9.93e-02 |
Pos logit contributions | |
run | +0.115 |
bug | +0.112 |
eh | +0.097 |
out | +0.097 |
udge | +0.097 |
Neg logit contributions | |
amen | -0.108 |
girls | -0.102 |
announcements | -0.098 |
Santa | -0.091 |
Rudolph | -0.090 |
Token | d |
Token ID | 288 |
Feature activation | +5.85e-01 |
Pos logit contributions | |
girls | +0.003 |
amen | +0.003 |
announcements | +0.003 |
hospitality | +0.003 |
MacDonald | +0.003 |
Neg logit contributions | |
run | -0.003 |
bug | -0.003 |
ours | -0.003 |
rous | -0.003 |
imb | -0.003 |
Token | inn |
Token ID | 3732 |
Feature activation | +8.52e+00 |
Pos logit contributions | |
run | +0.015 |
bug | +0.015 |
haired | +0.014 |
idd | +0.014 |
ours | +0.013 |
Neg logit contributions | |
amen | -0.020 |
girls | -0.019 |
Rudolph | -0.019 |
Santa | -0.018 |
announcements | -0.018 |
Token | ert |
Token ID | 861 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
run | +0.220 |
bug | +0.204 |
ours | +0.193 |
out | +0.179 |
idd | +0.178 |
Neg logit contributions | |
amen | -0.334 |
girls | -0.321 |
announcements | -0.301 |
Rudolph | -0.300 |
hospitality | -0.298 |
Token | ime |
Token ID | 524 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
run | +0.003 |
bug | +0.003 |
ours | +0.003 |
imb | +0.003 |
idd | +0.003 |
Neg logit contributions | |
amen | -0.005 |
girls | -0.005 |
Rudolph | -0.004 |
Santa | -0.004 |
announcements | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +8.33e-01 |
Pos logit contributions | |
run | +0.039 |
bug | +0.038 |
idd | +0.034 |
out | +0.033 |
ab | +0.033 |
Neg logit contributions | |
amen | -0.038 |
girls | -0.035 |
hospitality | -0.034 |
mercy | -0.033 |
sets | -0.032 |
Token | |
Token ID | 220 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
run | +0.031 |
bug | +0.030 |
ours | +0.028 |
idd | +0.027 |
imb | +0.027 |
Neg logit contributions | |
amen | -0.023 |
girls | -0.021 |
Rudolph | -0.020 |
announcements | -0.019 |
Santa | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | -7.91e-02 |
Pos logit contributions | |
amen | +0.028 |
girls | +0.026 |
Rudolph | +0.025 |
Santa | +0.024 |
announcements | +0.023 |
Neg logit contributions | |
run | -0.042 |
bug | -0.041 |
ours | -0.038 |
idd | -0.038 |
imb | -0.037 |
Token | just |
Token ID | 655 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
run | +0.121 |
bug | +0.115 |
haired | +0.100 |
udge | +0.098 |
imb | +0.098 |
Neg logit contributions | |
amen | -0.136 |
girls | -0.130 |
Rudolph | -0.122 |
Claus | -0.117 |
Santa | -0.117 |
Token | begun |
Token ID | 9258 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
run | +0.073 |
bug | +0.070 |
ours | +0.064 |
imb | +0.063 |
udge | +0.062 |
Neg logit contributions | |
amen | -0.067 |
girls | -0.064 |
Rudolph | -0.060 |
announcements | -0.058 |
Santa | -0.058 |
Token | ! |
Token ID | 0 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
amen | +0.041 |
girls | +0.039 |
Rudolph | +0.036 |
announcements | +0.035 |
Santa | +0.035 |
Neg logit contributions | |
run | -0.041 |
bug | -0.039 |
ours | -0.035 |
udge | -0.034 |
imb | -0.034 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +5.88e-01 |
Pos logit contributions | |
run | +0.075 |
bug | +0.073 |
ours | +0.068 |
idd | +0.067 |
udge | +0.067 |
Neg logit contributions | |
amen | -0.058 |
girls | -0.053 |
Rudolph | -0.051 |
announcements | -0.048 |
Santa | -0.048 |
Token | F |
Token ID | 37 |
Feature activation | +6.57e-01 |
Pos logit contributions | |
run | +0.019 |
bug | +0.018 |
idd | +0.016 |
ours | +0.016 |
imb | +0.016 |
Neg logit contributions | |
amen | -0.019 |
girls | -0.018 |
Rudolph | -0.017 |
announcements | -0.017 |
hospitality | -0.017 |
Token | inn |
Token ID | 3732 |
Feature activation | +8.50e+00 |
Pos logit contributions | |
run | +0.016 |
bug | +0.015 |
idd | +0.014 |
ours | +0.014 |
haired | +0.013 |
Neg logit contributions | |
amen | -0.025 |
girls | -0.024 |
Rudolph | -0.023 |
hospitality | -0.022 |
Santa | -0.022 |
Token | was |
Token ID | 373 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
run | +0.247 |
bug | +0.241 |
ours | +0.217 |
idd | +0.212 |
imb | +0.209 |
Neg logit contributions | |
amen | -0.302 |
girls | -0.284 |
Rudolph | -0.270 |
announcements | -0.267 |
Santa | -0.264 |
Token | fishing |
Token ID | 12478 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
run | +0.071 |
bug | +0.065 |
ours | +0.064 |
udge | +0.061 |
haired | +0.060 |
Neg logit contributions | |
amen | -0.054 |
Rudolph | -0.052 |
Santa | -0.048 |
Santa | -0.047 |
refreshed | -0.046 |
Token | with |
Token ID | 351 |
Feature activation | +3.29e+00 |
Pos logit contributions | |
run | +0.007 |
bug | +0.007 |
ours | +0.006 |
idd | +0.006 |
udge | +0.006 |
Neg logit contributions | |
amen | -0.008 |
girls | -0.007 |
Rudolph | -0.007 |
Santa | -0.007 |
hospitality | -0.007 |
Token | his |
Token ID | 465 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
run | +0.124 |
bug | +0.121 |
ours | +0.112 |
udge | +0.110 |
imb | +0.110 |
Neg logit contributions | |
amen | -0.087 |
girls | -0.081 |
Rudolph | -0.077 |
announcements | -0.073 |
Santa | -0.073 |
Token | dad |
Token ID | 9955 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
run | +0.052 |
bug | +0.051 |
ours | +0.047 |
idd | +0.046 |
imb | +0.046 |
Neg logit contributions | |
amen | -0.035 |
girls | -0.033 |
Rudolph | -0.030 |
announcements | -0.030 |
Santa | -0.029 |
Token | screaming |
Token ID | 14788 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
run | +0.036 |
bug | +0.035 |
haired | +0.033 |
ours | +0.033 |
udge | +0.032 |
Neg logit contributions | |
amen | -0.031 |
girls | -0.028 |
Rudolph | -0.028 |
announcements | -0.026 |
Santa | -0.026 |
Token | ?" |
Token ID | 1701 |
Feature activation | -9.94e-01 |
Pos logit contributions | |
amen | +0.100 |
girls | +0.094 |
Rudolph | +0.091 |
announcements | +0.087 |
Santa | +0.087 |
Neg logit contributions | |
run | -0.088 |
bug | -0.085 |
ours | -0.077 |
idd | -0.076 |
imb | -0.076 |
Token |
|
Token ID | 198 |
Feature activation | -7.50e-01 |
Pos logit contributions | |
amen | +0.025 |
girls | +0.023 |
Rudolph | +0.023 |
Santa | +0.020 |
announcements | +0.019 |
Neg logit contributions | |
run | -0.038 |
bug | -0.037 |
ours | -0.035 |
haired | -0.035 |
idd | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | -7.34e-01 |
Pos logit contributions | |
amen | +0.017 |
girls | +0.017 |
Rudolph | +0.016 |
Santa | +0.014 |
mercy | +0.013 |
Neg logit contributions | |
run | -0.030 |
bug | -0.030 |
ours | -0.028 |
haired | -0.028 |
imb | -0.027 |
Token | F |
Token ID | 37 |
Feature activation | +5.59e-01 |
Pos logit contributions | |
amen | +0.021 |
girls | +0.021 |
Rudolph | +0.019 |
Santa | +0.019 |
mercy | +0.017 |
Neg logit contributions | |
run | -0.025 |
bug | -0.024 |
ours | -0.023 |
haired | -0.023 |
idd | -0.022 |
Token | inn |
Token ID | 3732 |
Feature activation | +8.50e+00 |
Pos logit contributions | |
bug | +0.012 |
eh | +0.011 |
frame | +0.011 |
brick | +0.011 |
rous | +0.010 |
Neg logit contributions | |
amen | -0.017 |
girls | -0.017 |
Rudolph | -0.017 |
avi | -0.017 |
hospitality | -0.017 |
Token | said |
Token ID | 531 |
Feature activation | -7.28e-01 |
Pos logit contributions | |
run | +0.277 |
bug | +0.268 |
ours | +0.245 |
idd | +0.241 |
imb | +0.241 |
Neg logit contributions | |
amen | -0.268 |
girls | -0.252 |
Rudolph | -0.239 |
announcements | -0.231 |
Santa | -0.229 |
Token | sadly |
Token ID | 21098 |
Feature activation | +9.71e-01 |
Pos logit contributions | |
girls | +0.022 |
amen | +0.021 |
announcements | +0.020 |
Santa | +0.019 |
Rudolph | +0.019 |
Neg logit contributions | |
run | -0.025 |
bug | -0.024 |
ours | -0.022 |
haired | -0.021 |
idd | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | -5.00e-01 |
Pos logit contributions | |
run | +0.030 |
bug | +0.028 |
ours | +0.024 |
earing | +0.024 |
udge | +0.024 |
Neg logit contributions | |
girls | -0.033 |
amen | -0.032 |
announcements | -0.030 |
sets | -0.030 |
Santa | -0.029 |
Token | " |
Token ID | 366 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
girls | +0.021 |
amen | +0.019 |
announcements | +0.019 |
hospitality | +0.018 |
sets | +0.018 |
Neg logit contributions | |
run | -0.012 |
bug | -0.011 |
ours | -0.009 |
haired | -0.009 |
ang | -0.009 |
Token | I |
Token ID | 40 |
Feature activation | +4.05e+00 |
Pos logit contributions | |
run | +0.100 |
bug | +0.094 |
ours | +0.093 |
imb | +0.085 |
idd | +0.084 |
Neg logit contributions | |
amen | -0.097 |
Santa | -0.093 |
girls | -0.093 |
Rudolph | -0.089 |
announcements | -0.085 |
Token | would |
Token ID | 561 |
Feature activation | -3.22e+00 |
Pos logit contributions | |
run | +0.003 |
bug | +0.003 |
ours | +0.002 |
imb | +0.002 |
haired | +0.002 |
Neg logit contributions | |
amen | -0.002 |
girls | -0.002 |
Rudolph | -0.002 |
announcements | -0.002 |
Santa | -0.002 |
Token | lead |
Token ID | 1085 |
Feature activation | -6.81e-01 |
Pos logit contributions | |
amen | +0.099 |
girls | +0.095 |
Santa | +0.089 |
Rudolph | +0.088 |
announcements | +0.087 |
Neg logit contributions | |
run | -0.104 |
bug | -0.104 |
imb | -0.088 |
ours | -0.087 |
haired | -0.087 |
Token | to |
Token ID | 284 |
Feature activation | -6.48e-01 |
Pos logit contributions | |
amen | +0.022 |
girls | +0.021 |
Rudolph | +0.019 |
announcements | +0.019 |
Santa | +0.019 |
Neg logit contributions | |
run | -0.022 |
bug | -0.021 |
udge | -0.019 |
imb | -0.019 |
ours | -0.018 |
Token | an |
Token ID | 281 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
amen | +0.021 |
girls | +0.020 |
Rudolph | +0.018 |
Claus | +0.018 |
Santa | +0.018 |
Neg logit contributions | |
run | -0.021 |
bug | -0.020 |
imb | -0.017 |
udge | -0.017 |
rous | -0.017 |
Token | adventure |
Token ID | 8855 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
run | +0.068 |
bug | +0.063 |
ours | +0.063 |
idd | +0.062 |
udge | +0.059 |
Neg logit contributions | |
amen | -0.050 |
girls | -0.045 |
Rudolph | -0.044 |
hospitality | -0.044 |
mercy | -0.044 |
Token | unlike |
Token ID | 5023 |
Feature activation | +8.47e+00 |
Pos logit contributions | |
amen | +0.045 |
girls | +0.041 |
announcements | +0.039 |
Rudolph | +0.039 |
Santa | +0.038 |
Neg logit contributions | |
run | -0.037 |
bug | -0.035 |
out | -0.030 |
imb | -0.030 |
udge | -0.030 |
Token | any |
Token ID | 597 |
Feature activation | +8.50e-01 |
Pos logit contributions | |
run | +0.283 |
bug | +0.276 |
haired | +0.236 |
rous | +0.233 |
out | +0.231 |
Neg logit contributions | |
amen | -0.263 |
girls | -0.253 |
announcements | -0.227 |
Santa | -0.225 |
sets | -0.224 |
Token | other |
Token ID | 584 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
run | +0.029 |
bug | +0.029 |
ours | +0.026 |
idd | +0.025 |
udge | +0.025 |
Neg logit contributions | |
amen | -0.025 |
girls | -0.024 |
Rudolph | -0.022 |
announcements | -0.022 |
Santa | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +7.20e-01 |
Pos logit contributions | |
amen | +0.062 |
girls | +0.055 |
Rudolph | +0.053 |
Claus | +0.051 |
Santa | +0.051 |
Neg logit contributions | |
run | -0.088 |
bug | -0.086 |
imb | -0.077 |
ours | -0.077 |
idd | -0.076 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +7.15e-01 |
Pos logit contributions | |
run | +0.027 |
bug | +0.026 |
ours | +0.025 |
idd | +0.024 |
udge | +0.024 |
Neg logit contributions | |
amen | -0.019 |
girls | -0.017 |
Rudolph | -0.017 |
Santa | -0.016 |
announcements | -0.015 |
Token | Once |
Token ID | 7454 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
run | +0.024 |
bug | +0.023 |
idd | +0.021 |
ours | +0.021 |
imb | +0.021 |
Neg logit contributions | |
amen | -0.022 |
girls | -0.021 |
announcements | -0.019 |
Rudolph | -0.019 |
hospitality | -0.019 |
Token | room |
Token ID | 2119 |
Feature activation | -7.82e-01 |
Pos logit contributions | |
run | +0.059 |
bug | +0.056 |
udge | +0.054 |
idd | +0.053 |
ours | +0.052 |
Neg logit contributions | |
Rudolph | -0.039 |
amen | -0.037 |
parents | -0.036 |
stories | -0.036 |
bed | -0.035 |
Token | , |
Token ID | 11 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
amen | +0.026 |
girls | +0.025 |
Rudolph | +0.023 |
announcements | +0.022 |
Santa | +0.022 |
Neg logit contributions | |
run | -0.025 |
bug | -0.024 |
out | -0.022 |
udge | -0.022 |
ours | -0.022 |
Token | but |
Token ID | 475 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
amen | +0.005 |
girls | +0.005 |
Rudolph | +0.004 |
announcements | +0.004 |
Santa | +0.004 |
Neg logit contributions | |
run | -0.007 |
bug | -0.007 |
ours | -0.006 |
idd | -0.006 |
imb | -0.006 |
Token | it |
Token ID | 340 |
Feature activation | +3.35e+00 |
Pos logit contributions | |
run | +0.091 |
bug | +0.084 |
idd | +0.084 |
udge | +0.084 |
elly | +0.082 |
Neg logit contributions | |
amen | -0.046 |
Rudolph | -0.044 |
Santa | -0.040 |
Santa | -0.038 |
announcements | -0.036 |
Token | was |
Token ID | 373 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
run | +0.125 |
bug | +0.114 |
ours | +0.112 |
idd | +0.112 |
imb | +0.111 |
Neg logit contributions | |
amen | -0.086 |
girls | -0.084 |
Rudolph | -0.082 |
sets | -0.078 |
Claus | -0.078 |
Token | nowhere |
Token ID | 12062 |
Feature activation | +8.46e+00 |
Pos logit contributions | |
run | +0.042 |
bug | +0.041 |
imb | +0.038 |
ours | +0.037 |
idd | +0.037 |
Neg logit contributions | |
amen | -0.056 |
Rudolph | -0.053 |
girls | -0.053 |
Santa | -0.050 |
announcements | -0.050 |
Token | to |
Token ID | 284 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
imb | +0.381 |
ours | +0.374 |
idd | +0.352 |
ang | +0.348 |
run | +0.345 |
Neg logit contributions | |
Claus | -0.131 |
Rudolph | -0.130 |
mercy | -0.127 |
Santa | -0.114 |
Santa | -0.110 |
Token | be |
Token ID | 307 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
amen | +0.029 |
girls | +0.027 |
Rudolph | +0.025 |
Santa | +0.024 |
sets | +0.024 |
Neg logit contributions | |
run | -0.036 |
bug | -0.036 |
udge | -0.032 |
haired | -0.032 |
imb | -0.032 |
Token | found |
Token ID | 1043 |
Feature activation | +5.29e+00 |
Pos logit contributions | |
amen | +0.043 |
girls | +0.042 |
Rudolph | +0.040 |
announcements | +0.040 |
Santa | +0.040 |
Neg logit contributions | |
run | -0.021 |
bug | -0.020 |
ours | -0.018 |
imb | -0.018 |
ang | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +8.28e-01 |
Pos logit contributions | |
run | +0.246 |
bug | +0.240 |
ours | +0.238 |
imb | +0.235 |
idd | +0.231 |
Neg logit contributions | |
amen | -0.075 |
girls | -0.071 |
Rudolph | -0.064 |
Santa | -0.061 |
stories | -0.060 |
Token |
|
Token ID | 198 |
Feature activation | -9.87e-02 |
Pos logit contributions | |
run | +0.036 |
bug | +0.036 |
idd | +0.033 |
ang | +0.033 |
udge | +0.033 |
Neg logit contributions | |
amen | -0.016 |
girls | -0.014 |
Rudolph | -0.014 |
Santa | -0.013 |
sets | -0.012 |
Token | everywhere |
Token ID | 8347 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
run | +0.069 |
imb | +0.067 |
idd | +0.066 |
udge | +0.065 |
bug | +0.065 |
Neg logit contributions | |
amen | -0.042 |
Rudolph | -0.042 |
stories | -0.041 |
Santa | -0.037 |
announcements | -0.036 |
Token | , |
Token ID | 11 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
amen | +0.003 |
girls | +0.003 |
Rudolph | +0.003 |
announcements | +0.003 |
Santa | +0.003 |
Neg logit contributions | |
run | -0.004 |
bug | -0.004 |
ours | -0.004 |
idd | -0.004 |
imb | -0.004 |
Token | but |
Token ID | 475 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
amen | +0.003 |
girls | +0.003 |
Rudolph | +0.003 |
announcements | +0.003 |
Santa | +0.003 |
Neg logit contributions | |
run | -0.004 |
bug | -0.004 |
ours | -0.004 |
idd | -0.004 |
imb | -0.004 |
Token | it |
Token ID | 340 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
run | +0.092 |
idd | +0.088 |
udge | +0.087 |
bug | +0.086 |
elly | +0.085 |
Neg logit contributions | |
amen | -0.054 |
Rudolph | -0.049 |
Santa | -0.046 |
Santa | -0.043 |
announcements | -0.042 |
Token | was |
Token ID | 373 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
run | +0.135 |
imb | +0.125 |
ours | +0.124 |
idd | +0.124 |
udge | +0.123 |
Neg logit contributions | |
amen | -0.082 |
girls | -0.079 |
Rudolph | -0.078 |
Claus | -0.076 |
sets | -0.072 |
Token | nowhere |
Token ID | 12062 |
Feature activation | +8.45e+00 |
Pos logit contributions | |
run | +0.041 |
bug | +0.040 |
imb | +0.038 |
ours | +0.037 |
ang | +0.037 |
Neg logit contributions | |
amen | -0.064 |
Rudolph | -0.061 |
girls | -0.060 |
Santa | -0.058 |
mercy | -0.058 |
Token | to |
Token ID | 284 |
Feature activation | -8.92e-01 |
Pos logit contributions | |
imb | +0.366 |
ours | +0.358 |
ang | +0.337 |
idd | +0.336 |
run | +0.332 |
Neg logit contributions | |
Rudolph | -0.145 |
Claus | -0.145 |
mercy | -0.144 |
Santa | -0.133 |
amen | -0.126 |
Token | be |
Token ID | 307 |
Feature activation | -9.42e-01 |
Pos logit contributions | |
amen | +0.025 |
girls | +0.024 |
Rudolph | +0.022 |
announcements | +0.021 |
Santa | +0.021 |
Neg logit contributions | |
run | -0.032 |
bug | -0.032 |
udge | -0.029 |
imb | -0.029 |
haired | -0.029 |
Token | found |
Token ID | 1043 |
Feature activation | +5.42e+00 |
Pos logit contributions | |
amen | +0.039 |
girls | +0.038 |
Rudolph | +0.037 |
Santa | +0.036 |
announcements | +0.036 |
Neg logit contributions | |
run | -0.020 |
bug | -0.019 |
ours | -0.018 |
imb | -0.017 |
udge | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +9.29e-01 |
Pos logit contributions | |
run | +0.252 |
bug | +0.246 |
ours | +0.244 |
imb | +0.241 |
idd | +0.237 |
Neg logit contributions | |
amen | -0.077 |
girls | -0.070 |
Rudolph | -0.066 |
Santa | -0.062 |
stories | -0.060 |
Token | |
Token ID | 220 |
Feature activation | -9.65e-01 |
Pos logit contributions | |
run | +0.040 |
bug | +0.039 |
idd | +0.038 |
ang | +0.038 |
udge | +0.037 |
Neg logit contributions | |
amen | -0.018 |
Rudolph | -0.016 |
girls | -0.016 |
Santa | -0.014 |
announcements | -0.013 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
amen | +0.004 |
Santa | +0.004 |
girls | +0.004 |
Rudolph | +0.003 |
announcements | +0.003 |
Neg logit contributions | |
run | -0.006 |
ours | -0.006 |
bug | -0.006 |
ished | -0.005 |
idd | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -1.93e-01 |
Pos logit contributions | |
run | +0.067 |
bug | +0.067 |
ours | +0.059 |
imb | +0.059 |
idd | +0.059 |
Neg logit contributions | |
amen | -0.059 |
girls | -0.057 |
hospitality | -0.051 |
Rudolph | -0.051 |
Santa | -0.050 |
Token | Brown |
Token ID | 4373 |
Feature activation | +5.71e-01 |
Pos logit contributions | |
amen | +0.007 |
girls | +0.006 |
hospitality | +0.006 |
announcements | +0.006 |
Rudolph | +0.006 |
Neg logit contributions | |
bug | -0.006 |
run | -0.006 |
imb | -0.005 |
idd | -0.005 |
ours | -0.005 |
Token | ie |
Token ID | 494 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
bug | +0.013 |
run | +0.012 |
idd | +0.010 |
ours | +0.010 |
udge | +0.010 |
Neg logit contributions | |
amen | -0.025 |
girls | -0.023 |
Rudolph | -0.022 |
hospitality | -0.022 |
announcements | -0.022 |
Token | wh |
Token ID | 348 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
run | +0.051 |
bug | +0.049 |
ours | +0.045 |
udge | +0.044 |
imb | +0.044 |
Neg logit contributions | |
amen | -0.055 |
girls | -0.052 |
Rudolph | -0.049 |
announcements | -0.048 |
Santa | -0.047 |
Token | inn |
Token ID | 3732 |
Feature activation | +8.45e+00 |
Pos logit contributions | |
amen | +0.081 |
girls | +0.078 |
Santa | +0.071 |
announcements | +0.071 |
Jose | +0.070 |
Neg logit contributions | |
run | -0.059 |
bug | -0.056 |
haired | -0.053 |
ours | -0.049 |
imb | -0.048 |
Token | ied |
Token ID | 798 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
run | +0.173 |
bug | +0.162 |
ours | +0.139 |
idd | +0.134 |
udge | +0.133 |
Neg logit contributions | |
amen | -0.374 |
girls | -0.358 |
Rudolph | -0.345 |
announcements | -0.338 |
Santa | -0.336 |
Token | . |
Token ID | 13 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
run | +0.042 |
bug | +0.041 |
ours | +0.037 |
idd | +0.036 |
imb | +0.036 |
Neg logit contributions | |
amen | -0.056 |
girls | -0.053 |
Rudolph | -0.051 |
announcements | -0.050 |
Santa | -0.049 |
Token | � |
Token ID | 6184 |
Feature activation | -5.45e-01 |
Pos logit contributions | |
run | +0.014 |
bug | +0.013 |
ours | +0.012 |
imb | +0.012 |
idd | +0.012 |
Neg logit contributions | |
amen | -0.010 |
girls | -0.010 |
Rudolph | -0.009 |
announcements | -0.009 |
hospitality | -0.009 |
Token | � |
Token ID | 95 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
amen | +0.019 |
girls | +0.018 |
Rudolph | +0.017 |
announcements | +0.016 |
hospitality | +0.016 |
Neg logit contributions | |
run | -0.017 |
bug | -0.016 |
imb | -0.015 |
idd | -0.015 |
ours | -0.015 |
Token | € |
Token ID | 26391 |
Feature activation | -9.38e-01 |
Pos logit contributions | |
amen | +0.047 |
girls | +0.044 |
Rudolph | +0.042 |
announcements | +0.041 |
hospitality | +0.041 |
Neg logit contributions | |
run | -0.033 |
bug | -0.032 |
idd | -0.028 |
haired | -0.028 |
udge | -0.028 |
Token | ups |
Token ID | 4739 |
Feature activation | -7.53e+00 |
Pos logit contributions | |
run | +0.205 |
bug | +0.199 |
ours | +0.187 |
imb | +0.183 |
udge | +0.183 |
Neg logit contributions | |
amen | -0.140 |
girls | -0.133 |
Rudolph | -0.125 |
Santa | -0.120 |
announcements | -0.119 |
Token | . |
Token ID | 13 |
Feature activation | +7.25e-01 |
Pos logit contributions | |
amen | +0.234 |
girls | +0.218 |
announcements | +0.206 |
Rudolph | +0.204 |
mercy | +0.199 |
Neg logit contributions | |
run | -0.256 |
bug | -0.249 |
ours | -0.216 |
idd | -0.214 |
out | -0.213 |
Token | But |
Token ID | 887 |
Feature activation | +6.40e-01 |
Pos logit contributions | |
run | +0.027 |
bug | +0.026 |
ours | +0.024 |
haired | +0.024 |
imb | +0.024 |
Neg logit contributions | |
amen | -0.020 |
girls | -0.018 |
announcements | -0.017 |
Rudolph | -0.017 |
hospitality | -0.017 |
Token | Max |
Token ID | 5436 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
run | +0.023 |
bug | +0.022 |
ours | +0.020 |
imb | +0.020 |
haired | +0.020 |
Neg logit contributions | |
amen | -0.018 |
girls | -0.017 |
announcements | -0.015 |
Rudolph | -0.015 |
Santa | -0.015 |
Token | was |
Token ID | 373 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
amen | +0.068 |
girls | +0.065 |
Rudolph | +0.062 |
announcements | +0.060 |
Santa | +0.059 |
Neg logit contributions | |
run | -0.058 |
bug | -0.055 |
ours | -0.050 |
idd | -0.049 |
imb | -0.049 |
Token | nowhere |
Token ID | 12062 |
Feature activation | +8.44e+00 |
Pos logit contributions | |
run | +0.061 |
bug | +0.059 |
imb | +0.058 |
ours | +0.058 |
udge | +0.057 |
Neg logit contributions | |
amen | -0.049 |
Rudolph | -0.046 |
girls | -0.045 |
announcements | -0.043 |
Santa | -0.042 |
Token | to |
Token ID | 284 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
imb | +0.317 |
ours | +0.316 |
idd | +0.300 |
run | +0.295 |
bug | +0.293 |
Neg logit contributions | |
amen | -0.188 |
Rudolph | -0.187 |
mercy | -0.186 |
girls | -0.179 |
Claus | -0.178 |
Token | be |
Token ID | 307 |
Feature activation | -8.01e-01 |
Pos logit contributions | |
amen | +0.036 |
girls | +0.034 |
sets | +0.032 |
Rudolph | +0.031 |
Claus | +0.030 |
Neg logit contributions | |
run | -0.036 |
bug | -0.034 |
udge | -0.030 |
haired | -0.030 |
nail | -0.030 |
Token | found |
Token ID | 1043 |
Feature activation | +5.37e+00 |
Pos logit contributions | |
amen | +0.030 |
girls | +0.029 |
Rudolph | +0.028 |
announcements | +0.027 |
Santa | +0.027 |
Neg logit contributions | |
bug | -0.020 |
run | -0.020 |
ours | -0.019 |
imb | -0.018 |
udge | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +4.71e-01 |
Pos logit contributions | |
run | +0.201 |
bug | +0.194 |
ours | +0.178 |
udge | +0.177 |
idd | +0.176 |
Neg logit contributions | |
amen | -0.148 |
girls | -0.135 |
Rudolph | -0.127 |
announcements | -0.124 |
Santa | -0.122 |
Token | |
Token ID | 220 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
run | +0.018 |
bug | +0.017 |
haired | +0.016 |
ours | +0.016 |
imb | +0.015 |
Neg logit contributions | |
amen | -0.013 |
girls | -0.012 |
announcements | -0.011 |
hospitality | -0.011 |
Rudolph | -0.011 |
Token | friends |
Token ID | 2460 |
Feature activation | -6.01e+00 |
Pos logit contributions | |
run | +0.056 |
bug | +0.054 |
ours | +0.054 |
udge | +0.052 |
idd | +0.052 |
Neg logit contributions | |
girls | -0.035 |
amen | -0.035 |
stories | -0.030 |
Rudolph | -0.030 |
hospitality | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
amen | +0.177 |
girls | +0.167 |
Rudolph | +0.158 |
announcements | +0.151 |
Santa | +0.150 |
Neg logit contributions | |
run | -0.207 |
bug | -0.200 |
ours | -0.186 |
idd | -0.182 |
imb | -0.182 |
Token | now |
Token ID | 783 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
bug | +0.039 |
run | +0.039 |
ours | +0.034 |
haired | +0.034 |
idd | +0.034 |
Neg logit contributions | |
amen | -0.041 |
girls | -0.038 |
Santa | -0.037 |
Rudolph | -0.037 |
announcements | -0.036 |
Token | she |
Token ID | 673 |
Feature activation | +6.99e-01 |
Pos logit contributions | |
run | +0.058 |
bug | +0.057 |
ours | +0.052 |
idd | +0.051 |
udge | +0.051 |
Neg logit contributions | |
amen | -0.051 |
girls | -0.048 |
Rudolph | -0.046 |
announcements | -0.044 |
Santa | -0.044 |
Token | had |
Token ID | 550 |
Feature activation | +3.67e+00 |
Pos logit contributions | |
run | +0.025 |
imb | +0.024 |
ours | +0.024 |
udge | +0.024 |
rous | +0.023 |
Neg logit contributions | |
girls | -0.018 |
amen | -0.018 |
Claus | -0.016 |
Rudolph | -0.016 |
stories | -0.015 |
Token | nowhere |
Token ID | 12062 |
Feature activation | +8.44e+00 |
Pos logit contributions | |
run | +0.118 |
bug | +0.114 |
ours | +0.102 |
idd | +0.101 |
haired | +0.100 |
Neg logit contributions | |
amen | -0.121 |
girls | -0.113 |
Rudolph | -0.107 |
announcements | -0.103 |
Santa | -0.103 |
Token | to |
Token ID | 284 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
imb | +0.345 |
ours | +0.333 |
ang | +0.332 |
udge | +0.322 |
idd | +0.322 |
Neg logit contributions | |
mercy | -0.159 |
amen | -0.145 |
stories | -0.144 |
sets | -0.142 |
girls | -0.138 |
Token | go |
Token ID | 467 |
Feature activation | -3.10e-02 |
Pos logit contributions | |
amen | +0.030 |
girls | +0.028 |
Rudolph | +0.027 |
Santa | +0.025 |
announcements | +0.025 |
Neg logit contributions | |
bug | -0.036 |
run | -0.036 |
haired | -0.032 |
udge | -0.032 |
imb | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | +8.45e-01 |
Pos logit contributions | |
amen | +0.001 |
girls | +0.001 |
Rudolph | +0.001 |
announcements | +0.001 |
Santa | +0.001 |
Neg logit contributions | |
run | -0.001 |
bug | -0.001 |
ours | -0.001 |
idd | -0.001 |
udge | -0.001 |
Token | She |
Token ID | 1375 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
run | +0.033 |
bug | +0.032 |
ours | +0.030 |
udge | +0.030 |
idd | +0.029 |
Neg logit contributions | |
amen | -0.021 |
girls | -0.020 |
Rudolph | -0.019 |
Santa | -0.018 |
announcements | -0.017 |
Token | needed |
Token ID | 2622 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
amen | +0.042 |
girls | +0.040 |
Rudolph | +0.038 |
announcements | +0.036 |
Claus | +0.035 |
Neg logit contributions | |
run | -0.049 |
imb | -0.045 |
bug | -0.045 |
udge | -0.045 |
ours | -0.044 |
Token | be |
Token ID | 307 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
amen | +0.047 |
girls | +0.044 |
Rudolph | +0.041 |
sets | +0.040 |
announcements | +0.040 |
Neg logit contributions | |
run | -0.046 |
bug | -0.045 |
ours | -0.040 |
imb | -0.040 |
udge | -0.040 |
Token | cow |
Token ID | 9875 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
amen | +0.041 |
girls | +0.037 |
Rudolph | +0.035 |
hospitality | +0.034 |
announcements | +0.034 |
Neg logit contributions | |
run | -0.038 |
bug | -0.038 |
idd | -0.033 |
udge | -0.033 |
ours | -0.033 |
Token | boys |
Token ID | 13202 |
Feature activation | -5.80e+00 |
Pos logit contributions | |
bug | +0.012 |
run | +0.011 |
ang | +0.010 |
out | +0.010 |
c | +0.010 |
Neg logit contributions | |
amen | -0.009 |
Santa | -0.008 |
announcements | -0.007 |
mercy | -0.007 |
Rudolph | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | +6.78e-01 |
Pos logit contributions | |
amen | +0.223 |
announcements | +0.197 |
mercy | +0.197 |
hospitality | +0.192 |
prayers | +0.191 |
Neg logit contributions | |
bug | -0.159 |
run | -0.143 |
out | -0.131 |
ab | -0.125 |
udge | -0.123 |
Token | cow |
Token ID | 9875 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
bug | +0.025 |
run | +0.025 |
eh | +0.021 |
idd | +0.021 |
ours | +0.021 |
Neg logit contributions | |
amen | -0.018 |
announcements | -0.016 |
eness | -0.015 |
Santa | -0.015 |
girls | -0.014 |
Token | girls |
Token ID | 36960 |
Feature activation | -1.39e+01 |
Pos logit contributions | |
bug | +0.016 |
run | +0.015 |
ang | +0.014 |
out | +0.014 |
idd | +0.014 |
Neg logit contributions | |
amen | -0.006 |
announcements | -0.004 |
Santa | -0.004 |
mercy | -0.004 |
hospitality | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +7.36e-01 |
Pos logit contributions | |
amen | +0.419 |
announcements | +0.380 |
mercy | +0.379 |
hospitality | +0.372 |
Santa | +0.366 |
Neg logit contributions | |
bug | -0.474 |
run | -0.449 |
idd | -0.402 |
imb | -0.395 |
ours | -0.393 |
Token | They |
Token ID | 1119 |
Feature activation | -5.55e+00 |
Pos logit contributions | |
run | +0.028 |
bug | +0.028 |
idd | +0.027 |
eh | +0.026 |
haired | +0.026 |
Neg logit contributions | |
amen | -0.018 |
girls | -0.018 |
Rudolph | -0.015 |
announcements | -0.015 |
Santa | -0.014 |
Token | put |
Token ID | 1234 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
amen | +0.196 |
sets | +0.185 |
girls | +0.177 |
Santa | +0.171 |
mercy | +0.170 |
Neg logit contributions | |
bug | -0.160 |
run | -0.150 |
ours | -0.130 |
imb | -0.127 |
nail | -0.125 |
Token | on |
Token ID | 319 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
run | +0.037 |
bug | +0.036 |
ours | +0.033 |
idd | +0.032 |
udge | +0.032 |
Neg logit contributions | |
amen | -0.037 |
girls | -0.034 |
Rudolph | -0.032 |
announcements | -0.032 |
Santa | -0.031 |
Token | hats |
Token ID | 23910 |
Feature activation | -4.19e+00 |
Pos logit contributions | |
run | +0.053 |
bug | +0.052 |
ours | +0.048 |
rous | +0.047 |
eh | +0.046 |
Neg logit contributions | |
amen | -0.052 |
girls | -0.045 |
hospitality | -0.043 |
mercy | -0.042 |
announcements | -0.042 |
Token | my |
Token ID | 616 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
amen | +0.019 |
girls | +0.019 |
Claus | +0.018 |
Rudolph | +0.018 |
Santa | +0.018 |
Neg logit contributions | |
run | -0.015 |
bug | -0.014 |
ours | -0.013 |
imb | -0.012 |
haired | -0.012 |
Token | cow |
Token ID | 9875 |
Feature activation | +4.94e-01 |
Pos logit contributions | |
run | +0.078 |
bug | +0.074 |
ours | +0.072 |
idd | +0.067 |
imb | +0.066 |
Neg logit contributions | |
girls | -0.082 |
amen | -0.079 |
announcements | -0.077 |
sets | -0.074 |
Santa | -0.073 |
Token | boys |
Token ID | 13202 |
Feature activation | -5.46e+00 |
Pos logit contributions | |
bug | +0.016 |
run | +0.015 |
ours | +0.014 |
idd | +0.014 |
ab | +0.013 |
Neg logit contributions | |
amen | -0.016 |
announcements | -0.015 |
girls | -0.014 |
Santa | -0.014 |
Rudolph | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | +5.86e-01 |
Pos logit contributions | |
amen | +0.182 |
announcements | +0.165 |
girls | +0.162 |
mercy | +0.160 |
hospitality | +0.157 |
Neg logit contributions | |
bug | -0.165 |
run | -0.162 |
ab | -0.147 |
out | -0.141 |
idd | -0.138 |
Token | cow |
Token ID | 9875 |
Feature activation | +5.09e-01 |
Pos logit contributions | |
run | +0.020 |
bug | +0.019 |
ours | +0.017 |
idd | +0.016 |
imb | +0.016 |
Neg logit contributions | |
amen | -0.017 |
announcements | -0.016 |
Santa | -0.015 |
girls | -0.015 |
ban | -0.014 |
Token | girls |
Token ID | 36960 |
Feature activation | -1.32e+01 |
Pos logit contributions | |
bug | +0.017 |
run | +0.017 |
idd | +0.015 |
ours | +0.015 |
out | +0.014 |
Neg logit contributions | |
amen | -0.015 |
announcements | -0.014 |
Santa | -0.013 |
Rudolph | -0.013 |
mercy | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +7.07e-01 |
Pos logit contributions | |
amen | +0.406 |
announcements | +0.379 |
mercy | +0.375 |
sets | +0.365 |
girls | +0.363 |
Neg logit contributions | |
bug | -0.418 |
run | -0.401 |
idd | -0.361 |
ours | -0.358 |
ab | -0.357 |
Token | I |
Token ID | 314 |
Feature activation | +3.82e+00 |
Pos logit contributions | |
run | +0.017 |
elly | +0.014 |
imb | +0.014 |
bug | +0.013 |
ours | +0.013 |
Neg logit contributions | |
amen | -0.026 |
girls | -0.026 |
mercy | -0.025 |
sets | -0.025 |
announcements | -0.025 |
Token | love |
Token ID | 1842 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
bug | +0.119 |
run | +0.117 |
idd | +0.101 |
ab | +0.100 |
ours | +0.100 |
Neg logit contributions | |
amen | -0.122 |
girls | -0.116 |
sets | -0.113 |
Rudolph | -0.110 |
announcements | -0.109 |
Token | you |
Token ID | 345 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
amen | +0.052 |
girls | +0.049 |
announcements | +0.049 |
Claus | +0.046 |
Santa | +0.046 |
Neg logit contributions | |
run | -0.038 |
bug | -0.037 |
ours | -0.033 |
idd | -0.031 |
rous | -0.031 |
Token | ." |
Token ID | 526 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
bug | +0.024 |
here | +0.022 |
run | +0.021 |
out | +0.020 |
ab | +0.020 |
Neg logit contributions | |
announcements | -0.038 |
girls | -0.035 |
amen | -0.035 |
sets | -0.034 |
mercy | -0.034 |
Token | play |
Token ID | 711 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
amen | +0.076 |
girls | +0.069 |
Santa | +0.067 |
Rudolph | +0.067 |
hospitality | +0.067 |
Neg logit contributions | |
bug | -0.073 |
run | -0.069 |
ours | -0.061 |
haired | -0.060 |
imb | -0.059 |
Token | cow |
Token ID | 9875 |
Feature activation | +6.34e-01 |
Pos logit contributions | |
amen | +0.043 |
girls | +0.042 |
Rudolph | +0.041 |
announcements | +0.038 |
Santa | +0.038 |
Neg logit contributions | |
run | -0.037 |
bug | -0.035 |
ours | -0.033 |
idd | -0.032 |
udge | -0.031 |
Token | boys |
Token ID | 13202 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
bug | +0.023 |
run | +0.023 |
ours | +0.020 |
imb | +0.020 |
idd | +0.020 |
Neg logit contributions | |
amen | -0.018 |
girls | -0.016 |
Rudolph | -0.015 |
Santa | -0.015 |
announcements | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | +9.87e-01 |
Pos logit contributions | |
amen | +0.189 |
mercy | +0.170 |
girls | +0.167 |
Santa | +0.166 |
announcements | +0.165 |
Neg logit contributions | |
bug | -0.155 |
run | -0.153 |
haired | -0.132 |
ab | -0.131 |
idd | -0.126 |
Token | cow |
Token ID | 9875 |
Feature activation | +6.34e-01 |
Pos logit contributions | |
bug | +0.035 |
run | +0.034 |
eh | +0.031 |
idd | +0.030 |
nail | +0.029 |
Neg logit contributions | |
amen | -0.027 |
eness | -0.022 |
announcements | -0.022 |
Santa | -0.022 |
mercy | -0.021 |
Token | girls |
Token ID | 36960 |
Feature activation | -1.30e+01 |
Pos logit contributions | |
bug | +0.027 |
run | +0.026 |
ours | +0.023 |
ang | +0.023 |
idd | +0.023 |
Neg logit contributions | |
amen | -0.015 |
Santa | -0.012 |
announcements | -0.012 |
mercy | -0.012 |
hospitality | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
amen | +0.432 |
mercy | +0.400 |
announcements | +0.397 |
hospitality | +0.390 |
Santa | +0.390 |
Neg logit contributions | |
bug | -0.395 |
run | -0.381 |
idd | -0.327 |
imb | -0.326 |
ours | -0.325 |
Token | and |
Token ID | 290 |
Feature activation | +9.93e-01 |
Pos logit contributions | |
amen | +0.005 |
girls | +0.004 |
announcements | +0.004 |
mercy | +0.004 |
Santa | +0.004 |
Neg logit contributions | |
bug | -0.005 |
run | -0.005 |
imb | -0.004 |
ours | -0.004 |
haired | -0.004 |
Token | pretend |
Token ID | 16614 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
bug | +0.036 |
run | +0.035 |
idd | +0.031 |
haired | +0.031 |
ours | +0.031 |
Neg logit contributions | |
amen | -0.027 |
girls | -0.024 |
Rudolph | -0.022 |
Santa | -0.022 |
sets | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
run | +0.035 |
bug | +0.034 |
ours | +0.031 |
idd | +0.029 |
haired | +0.029 |
Neg logit contributions | |
amen | -0.035 |
girls | -0.032 |
Rudolph | -0.030 |
Santa | -0.030 |
announcements | -0.030 |
Token | ride |
Token ID | 6594 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
amen | +0.034 |
sets | +0.030 |
girls | +0.029 |
Rudolph | +0.029 |
Santa | +0.028 |
Neg logit contributions | |
bug | -0.034 |
run | -0.034 |
haired | -0.029 |
eh | -0.029 |
udge | -0.029 |
Token | forgot |
Token ID | 16453 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
bug | +0.014 |
run | +0.013 |
ours | +0.011 |
haired | +0.010 |
imb | +0.010 |
Neg logit contributions | |
amen | -0.019 |
girls | -0.019 |
Rudolph | -0.018 |
Santa | -0.018 |
sets | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
amen | +0.086 |
girls | +0.082 |
announcements | +0.077 |
Claus | +0.076 |
Santa | +0.076 |
Neg logit contributions | |
run | -0.067 |
bug | -0.062 |
out | -0.054 |
ours | -0.053 |
haired | -0.051 |
Token | lion |
Token ID | 18744 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
amen | +0.088 |
girls | +0.085 |
Rudolph | +0.079 |
announcements | +0.076 |
Santa | +0.076 |
Neg logit contributions | |
run | -0.087 |
bug | -0.085 |
ours | -0.076 |
haired | -0.076 |
idd | -0.076 |
Token | and |
Token ID | 290 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
bug | +0.007 |
run | +0.007 |
haired | +0.006 |
idd | +0.006 |
udge | +0.006 |
Neg logit contributions | |
amen | -0.007 |
girls | -0.007 |
announcements | -0.006 |
Rudolph | -0.006 |
Santa | -0.006 |
Token | his |
Token ID | 465 |
Feature activation | +7.95e-01 |
Pos logit contributions | |
run | +0.012 |
bug | +0.012 |
ours | +0.010 |
haired | +0.010 |
idd | +0.010 |
Neg logit contributions | |
amen | -0.010 |
girls | -0.009 |
announcements | -0.009 |
Santa | -0.009 |
Rudolph | -0.009 |
Token | mercy |
Token ID | 17703 |
Feature activation | -1.14e+01 |
Pos logit contributions | |
run | +0.026 |
bug | +0.025 |
ours | +0.022 |
idd | +0.022 |
haired | +0.022 |
Neg logit contributions | |
amen | -0.025 |
girls | -0.023 |
Rudolph | -0.023 |
announcements | -0.022 |
Santa | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -9.87e-02 |
Pos logit contributions | |
amen | +0.329 |
girls | +0.310 |
Rudolph | +0.293 |
announcements | +0.283 |
Santa | +0.280 |
Neg logit contributions | |
run | -0.412 |
bug | -0.406 |
udge | -0.363 |
idd | -0.359 |
haired | -0.355 |
Token | He |
Token ID | 679 |
Feature activation | -2.61e+00 |
Pos logit contributions | |
amen | +0.003 |
girls | +0.003 |
announcements | +0.003 |
sets | +0.003 |
hospitality | +0.003 |
Neg logit contributions | |
run | -0.004 |
bug | -0.003 |
udge | -0.003 |
idd | -0.003 |
haired | -0.003 |
Token | always |
Token ID | 1464 |
Feature activation | -4.02e+00 |
Pos logit contributions | |
amen | +0.083 |
girls | +0.077 |
sets | +0.074 |
announcements | +0.074 |
Rudolph | +0.074 |
Neg logit contributions | |
bug | -0.084 |
run | -0.081 |
haired | -0.072 |
idd | -0.071 |
ours | -0.071 |
Token | tried |
Token ID | 3088 |
Feature activation | -2.80e-01 |
Pos logit contributions | |
amen | +0.136 |
girls | +0.134 |
Santa | +0.125 |
announcements | +0.125 |
Rudolph | +0.124 |
Neg logit contributions | |
bug | -0.118 |
run | -0.117 |
ours | -0.101 |
idd | -0.099 |
haired | -0.096 |
Token | to |
Token ID | 284 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
amen | +0.010 |
girls | +0.010 |
announcements | +0.009 |
Rudolph | +0.009 |
Santa | +0.009 |
Neg logit contributions | |
run | -0.008 |
bug | -0.008 |
ours | -0.007 |
udge | -0.007 |
rous | -0.006 |
Token | your |
Token ID | 534 |
Feature activation | +3.34e-01 |
Pos logit contributions | |
girls | +0.090 |
amen | +0.089 |
hospitality | +0.085 |
announcements | +0.083 |
sets | +0.080 |
Neg logit contributions | |
run | -0.067 |
bug | -0.067 |
ours | -0.060 |
out | -0.056 |
here | -0.056 |
Token | fault |
Token ID | 8046 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
run | +0.012 |
bug | +0.012 |
ours | +0.011 |
idd | +0.011 |
imb | +0.011 |
Neg logit contributions | |
amen | -0.009 |
girls | -0.009 |
Rudolph | -0.008 |
announcements | -0.008 |
Santa | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | +3.74e-01 |
Pos logit contributions | |
amen | +0.058 |
announcements | +0.055 |
girls | +0.054 |
Santa | +0.053 |
Rudolph | +0.051 |
Neg logit contributions | |
run | -0.056 |
bug | -0.051 |
haired | -0.045 |
idd | -0.045 |
ang | -0.045 |
Token | ask |
Token ID | 1265 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
run | +0.011 |
bug | +0.010 |
ours | +0.009 |
imb | +0.009 |
idd | +0.009 |
Neg logit contributions | |
amen | -0.013 |
girls | -0.013 |
Rudolph | -0.012 |
announcements | -0.012 |
Santa | -0.012 |
Token | for |
Token ID | 329 |
Feature activation | -4.92e-01 |
Pos logit contributions | |
amen | +0.075 |
girls | +0.072 |
announcements | +0.068 |
Rudolph | +0.068 |
sets | +0.067 |
Neg logit contributions | |
run | -0.059 |
bug | -0.056 |
ours | -0.049 |
imb | -0.048 |
udge | -0.048 |
Token | mercy |
Token ID | 17703 |
Feature activation | -1.13e+01 |
Pos logit contributions | |
girls | +0.013 |
amen | +0.013 |
sets | +0.012 |
Claus | +0.011 |
announcements | +0.011 |
Neg logit contributions | |
run | -0.018 |
bug | -0.017 |
idd | -0.015 |
ours | -0.015 |
eh | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
amen | +0.310 |
girls | +0.290 |
Rudolph | +0.282 |
sets | +0.268 |
announcements | +0.267 |
Neg logit contributions | |
run | -0.424 |
bug | -0.408 |
udge | -0.366 |
ours | -0.364 |
idd | -0.359 |
Token | I |
Token ID | 314 |
Feature activation | +3.31e+00 |
Pos logit contributions | |
run | +0.004 |
bug | +0.004 |
ours | +0.004 |
imb | +0.004 |
idd | +0.004 |
Neg logit contributions | |
amen | -0.006 |
girls | -0.005 |
Rudolph | -0.005 |
announcements | -0.005 |
hospitality | -0.005 |
Token | will |
Token ID | 481 |
Feature activation | -7.68e+00 |
Pos logit contributions | |
run | +0.116 |
bug | +0.113 |
ours | +0.103 |
idd | +0.102 |
imb | +0.101 |
Neg logit contributions | |
amen | -0.097 |
girls | -0.091 |
Rudolph | -0.086 |
announcements | -0.083 |
Santa | -0.082 |
Token | spare |
Token ID | 13952 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
sets | +0.248 |
girls | +0.248 |
amen | +0.244 |
announcements | +0.233 |
MacDonald | +0.229 |
Neg logit contributions | |
bug | -0.207 |
run | -0.202 |
here | -0.194 |
nail | -0.190 |
touched | -0.175 |
Token | you |
Token ID | 345 |
Feature activation | +5.68e-01 |
Pos logit contributions | |
amen | +0.077 |
girls | +0.074 |
Rudolph | +0.070 |
announcements | +0.069 |
Claus | +0.068 |
Neg logit contributions | |
run | -0.061 |
bug | -0.058 |
ours | -0.052 |
udge | -0.051 |
idd | -0.050 |
Token | very |
Token ID | 845 |
Feature activation | +8.92e-01 |
Pos logit contributions | |
run | +0.035 |
bug | +0.034 |
imb | +0.033 |
ours | +0.032 |
idd | +0.032 |
Neg logit contributions | |
amen | -0.038 |
girls | -0.036 |
Rudolph | -0.035 |
announcements | -0.033 |
Santa | -0.033 |
Token | scared |
Token ID | 12008 |
Feature activation | -7.64e-01 |
Pos logit contributions | |
run | +0.029 |
bug | +0.029 |
imb | +0.028 |
ours | +0.027 |
udge | +0.027 |
Neg logit contributions | |
amen | -0.027 |
girls | -0.024 |
Rudolph | -0.024 |
announcements | -0.023 |
mercy | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | +5.38e-01 |
Pos logit contributions | |
amen | +0.027 |
girls | +0.026 |
Claus | +0.024 |
sets | +0.024 |
hospitality | +0.024 |
Neg logit contributions | |
run | -0.025 |
bug | -0.022 |
udge | -0.020 |
elly | -0.019 |
out | -0.019 |
Token | begged |
Token ID | 32946 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
run | +0.017 |
bug | +0.017 |
ours | +0.015 |
idd | +0.015 |
imb | +0.015 |
Neg logit contributions | |
amen | -0.018 |
girls | -0.017 |
Rudolph | -0.016 |
announcements | -0.015 |
Santa | -0.015 |
Token | for |
Token ID | 329 |
Feature activation | -3.96e-01 |
Pos logit contributions | |
sets | +0.041 |
amen | +0.041 |
girls | +0.039 |
announcements | +0.038 |
MacDonald | +0.037 |
Neg logit contributions | |
run | -0.039 |
bug | -0.037 |
udge | -0.032 |
out | -0.032 |
ours | -0.031 |
Token | mercy |
Token ID | 17703 |
Feature activation | -1.11e+01 |
Pos logit contributions | |
sets | +0.013 |
girls | +0.012 |
Claus | +0.012 |
MacDonald | +0.012 |
bows | +0.012 |
Neg logit contributions | |
run | -0.012 |
bug | -0.011 |
here | -0.011 |
nail | -0.010 |
else | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
amen | +0.318 |
Rudolph | +0.296 |
girls | +0.296 |
sets | +0.289 |
Santa | +0.282 |
Neg logit contributions | |
run | -0.409 |
bug | -0.395 |
udge | -0.351 |
ang | -0.337 |
idd | -0.336 |
Token | He |
Token ID | 679 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
amen | +0.005 |
girls | +0.005 |
announcements | +0.004 |
hospitality | +0.004 |
Rudolph | +0.004 |
Neg logit contributions | |
run | -0.005 |
bug | -0.005 |
ours | -0.004 |
idd | -0.004 |
imb | -0.004 |
Token | said |
Token ID | 531 |
Feature activation | -9.47e-01 |
Pos logit contributions | |
amen | +0.060 |
girls | +0.055 |
Rudolph | +0.052 |
Santa | +0.049 |
mercy | +0.049 |
Neg logit contributions | |
run | -0.082 |
bug | -0.078 |
imb | -0.076 |
ours | -0.076 |
udge | -0.075 |
Token | , |
Token ID | 11 |
Feature activation | -6.66e-01 |
Pos logit contributions | |
girls | +0.025 |
amen | +0.024 |
sets | +0.023 |
announcements | +0.022 |
Rudolph | +0.021 |
Neg logit contributions | |
run | -0.036 |
bug | -0.034 |
ours | -0.031 |
haired | -0.030 |
idd | -0.030 |
Token | ' |
Token ID | 705 |
Feature activation | +4.65e+00 |
Pos logit contributions | |
girls | +0.025 |
amen | +0.025 |
announcements | +0.024 |
hospitality | +0.023 |
sets | +0.023 |
Neg logit contributions | |
run | -0.017 |
bug | -0.016 |
ours | -0.013 |
haired | -0.013 |
so | -0.013 |
Token | Tom |
Token ID | 4186 |
Feature activation | -2.96e-01 |
Pos logit contributions | |
bug | +0.010 |
run | +0.009 |
ours | +0.008 |
okes | +0.007 |
haired | +0.007 |
Neg logit contributions | |
amen | -0.023 |
Amen | -0.021 |
sets | -0.021 |
hospitality | -0.021 |
Santa | -0.021 |
Token | cried |
Token ID | 16896 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
sets | +0.011 |
girls | +0.011 |
amen | +0.011 |
Rudolph | +0.011 |
avail | +0.011 |
Neg logit contributions | |
bug | -0.008 |
run | -0.007 |
touched | -0.007 |
found | -0.006 |
out | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +5.61e-01 |
Pos logit contributions | |
amen | +0.075 |
girls | +0.073 |
William | +0.071 |
sets | +0.071 |
MacDonald | +0.070 |
Neg logit contributions | |
run | -0.051 |
at | -0.044 |
bug | -0.042 |
udge | -0.041 |
out | -0.041 |
Token | begged |
Token ID | 32946 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
run | +0.014 |
bug | +0.013 |
touched | +0.011 |
idd | +0.010 |
ours | +0.010 |
Neg logit contributions | |
sets | -0.023 |
amen | -0.022 |
girls | -0.021 |
Santa | -0.020 |
Him | -0.020 |
Token | for |
Token ID | 329 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
sets | +0.040 |
amen | +0.039 |
girls | +0.039 |
Claus | +0.037 |
MacDonald | +0.037 |
Neg logit contributions | |
run | -0.033 |
bug | -0.029 |
udge | -0.028 |
ours | -0.027 |
out | -0.027 |
Token | mercy |
Token ID | 17703 |
Feature activation | -1.10e+01 |
Pos logit contributions | |
sets | +0.006 |
MacDonald | +0.005 |
girls | +0.005 |
bit | +0.005 |
bows | +0.005 |
Neg logit contributions | |
run | -0.004 |
here | -0.004 |
bug | -0.003 |
where | -0.003 |
nail | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
amen | +0.312 |
girls | +0.292 |
Rudolph | +0.289 |
sets | +0.279 |
Santa | +0.273 |
Neg logit contributions | |
run | -0.412 |
bug | -0.389 |
udge | -0.356 |
ours | -0.344 |
out | -0.341 |
Token | They |
Token ID | 1119 |
Feature activation | -5.70e+00 |
Pos logit contributions | |
run | +0.003 |
ours | +0.003 |
bug | +0.003 |
haired | +0.003 |
idd | +0.003 |
Neg logit contributions | |
amen | -0.003 |
hospitality | -0.003 |
girls | -0.003 |
sets | -0.003 |
mercy | -0.003 |
Token | wished |
Token ID | 16555 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
sets | +0.221 |
amen | +0.214 |
girls | +0.205 |
Him | +0.194 |
hospitality | +0.193 |
Neg logit contributions | |
run | -0.152 |
bug | -0.144 |
out | -0.120 |
haired | -0.119 |
touched | -0.117 |
Token | they |
Token ID | 484 |
Feature activation | -2.72e+00 |
Pos logit contributions | |
run | +0.042 |
ours | +0.036 |
bug | +0.036 |
out | +0.032 |
so | +0.031 |
Neg logit contributions | |
girls | -0.049 |
amen | -0.044 |
bit | -0.044 |
sets | -0.044 |
Claus | -0.043 |
Token | had |
Token ID | 550 |
Feature activation | +3.09e+00 |
Pos logit contributions | |
amen | +0.110 |
sets | +0.102 |
girls | +0.100 |
announcements | +0.096 |
Him | +0.094 |
Neg logit contributions | |
run | -0.067 |
bug | -0.062 |
haired | -0.049 |
out | -0.049 |
so | -0.048 |
Token | and |
Token ID | 290 |
Feature activation | +4.88e-01 |
Pos logit contributions | |
amen | +0.043 |
girls | +0.042 |
announcements | +0.039 |
Santa | +0.039 |
Rudolph | +0.039 |
Neg logit contributions | |
run | -0.040 |
bug | -0.039 |
ours | -0.034 |
haired | -0.033 |
out | -0.032 |
Token | told |
Token ID | 1297 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
run | +0.017 |
bug | +0.017 |
ours | +0.015 |
idd | +0.015 |
haired | +0.015 |
Neg logit contributions | |
amen | -0.014 |
girls | -0.013 |
Rudolph | -0.012 |
announcements | -0.012 |
Santa | -0.012 |
Token | him |
Token ID | 683 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
amen | +0.050 |
girls | +0.046 |
sets | +0.043 |
Claus | +0.043 |
Santa | +0.041 |
Neg logit contributions | |
run | -0.070 |
bug | -0.066 |
ours | -0.061 |
out | -0.060 |
udge | -0.060 |
Token | some |
Token ID | 617 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
amen | +0.048 |
girls | +0.047 |
Santa | +0.044 |
announcements | +0.043 |
Rudolph | +0.043 |
Neg logit contributions | |
run | -0.037 |
bug | -0.034 |
ours | -0.031 |
out | -0.029 |
haired | -0.028 |
Token | fun |
Token ID | 1257 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
amen | +0.028 |
girls | +0.027 |
Rudolph | +0.025 |
Claus | +0.024 |
Santa | +0.024 |
Neg logit contributions | |
run | -0.049 |
bug | -0.047 |
ours | -0.044 |
haired | -0.043 |
udge | -0.043 |
Token | facts |
Token ID | 6419 |
Feature activation | -1.04e+01 |
Pos logit contributions | |
amen | +0.081 |
girls | +0.074 |
Rudolph | +0.072 |
Santa | +0.070 |
announcements | +0.068 |
Neg logit contributions | |
run | -0.100 |
bug | -0.095 |
ours | -0.087 |
udge | -0.085 |
idd | -0.084 |
Token | . |
Token ID | 13 |
Feature activation | -3.63e-01 |
Pos logit contributions | |
amen | +0.338 |
girls | +0.324 |
Rudolph | +0.305 |
announcements | +0.297 |
Claus | +0.294 |
Neg logit contributions | |
run | -0.340 |
bug | -0.315 |
haired | -0.277 |
out | -0.277 |
ours | -0.275 |
Token | Ben |
Token ID | 3932 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
amen | +0.010 |
girls | +0.009 |
announcements | +0.009 |
hospitality | +0.009 |
Rudolph | +0.009 |
Neg logit contributions | |
run | -0.013 |
bug | -0.013 |
ours | -0.012 |
idd | -0.012 |
udge | -0.012 |
Token | learned |
Token ID | 4499 |
Feature activation | -6.47e+00 |
Pos logit contributions | |
amen | +0.052 |
girls | +0.050 |
Rudolph | +0.047 |
announcements | +0.046 |
Santa | +0.045 |
Neg logit contributions | |
bug | -0.046 |
run | -0.045 |
ours | -0.040 |
haired | -0.040 |
idd | -0.039 |
Token | a |
Token ID | 257 |
Feature activation | -7.43e-01 |
Pos logit contributions | |
amen | +0.247 |
girls | +0.241 |
Claus | +0.228 |
announcements | +0.224 |
sets | +0.222 |
Neg logit contributions | |
run | -0.171 |
out | -0.137 |
bug | -0.133 |
ours | -0.132 |
udge | -0.126 |
Token | lot |
Token ID | 1256 |
Feature activation | +4.20e-02 |
Pos logit contributions | |
amen | +0.023 |
girls | +0.023 |
Rudolph | +0.021 |
announcements | +0.020 |
Santa | +0.020 |
Neg logit contributions | |
run | -0.024 |
bug | -0.023 |
ours | -0.021 |
haired | -0.021 |
udge | -0.021 |
Token | to |
Token ID | 284 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
amen | +0.066 |
announcements | +0.062 |
girls | +0.058 |
Santa | +0.058 |
Rudolph | +0.055 |
Neg logit contributions | |
run | -0.068 |
bug | -0.065 |
out | -0.055 |
udge | -0.055 |
ang | -0.054 |
Token | explain |
Token ID | 4727 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
amen | +0.041 |
girls | +0.038 |
sets | +0.036 |
Claus | +0.036 |
announcements | +0.036 |
Neg logit contributions | |
run | -0.042 |
bug | -0.039 |
udge | -0.036 |
imb | -0.034 |
rous | -0.034 |
Token | all |
Token ID | 477 |
Feature activation | +6.42e-01 |
Pos logit contributions | |
amen | +0.061 |
girls | +0.059 |
Rudolph | +0.055 |
Claus | +0.055 |
announcements | +0.054 |
Neg logit contributions | |
run | -0.048 |
bug | -0.043 |
ours | -0.039 |
out | -0.038 |
udge | -0.037 |
Token | the |
Token ID | 262 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
run | +0.023 |
bug | +0.022 |
ours | +0.020 |
udge | +0.020 |
haired | +0.020 |
Neg logit contributions | |
amen | -0.019 |
girls | -0.018 |
Rudolph | -0.017 |
announcements | -0.016 |
Santa | -0.016 |
Token | boat |
Token ID | 8848 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
amen | +0.071 |
girls | +0.068 |
Claus | +0.063 |
Rudolph | +0.062 |
Santa | +0.060 |
Neg logit contributions | |
run | -0.077 |
bug | -0.073 |
ours | -0.067 |
rous | -0.067 |
udge | -0.065 |
Token | facts |
Token ID | 6419 |
Feature activation | -1.04e+01 |
Pos logit contributions | |
run | +0.050 |
bug | +0.049 |
haired | +0.042 |
ours | +0.042 |
idd | +0.042 |
Neg logit contributions | |
amen | -0.048 |
girls | -0.044 |
Rudolph | -0.042 |
announcements | -0.041 |
Santa | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
amen | +0.336 |
girls | +0.322 |
Rudolph | +0.301 |
announcements | +0.292 |
Claus | +0.286 |
Neg logit contributions | |
run | -0.339 |
bug | -0.316 |
haired | -0.281 |
ours | -0.278 |
out | -0.276 |
Token | Even |
Token ID | 3412 |
Feature activation | -4.40e-01 |
Pos logit contributions | |
run | +0.008 |
bug | +0.007 |
ours | +0.007 |
idd | +0.007 |
udge | +0.007 |
Neg logit contributions | |
amen | -0.005 |
girls | -0.004 |
Rudolph | -0.004 |
Santa | -0.004 |
announcements | -0.004 |
Token | though |
Token ID | 996 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
amen | +0.015 |
girls | +0.014 |
Rudolph | +0.014 |
announcements | +0.013 |
Santa | +0.013 |
Neg logit contributions | |
run | -0.013 |
bug | -0.013 |
ours | -0.012 |
idd | -0.012 |
imb | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
amen | +0.045 |
girls | +0.041 |
announcements | +0.039 |
Santa | +0.039 |
hospitality | +0.038 |
Neg logit contributions | |
run | -0.047 |
bug | -0.046 |
ours | -0.042 |
imb | -0.040 |
udge | -0.040 |
Token | adventure |
Token ID | 8855 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
amen | +0.081 |
girls | +0.080 |
sets | +0.069 |
mercy | +0.069 |
Rudolph | +0.068 |
Neg logit contributions | |
bug | -0.065 |
run | -0.064 |
nail | -0.063 |
idd | -0.063 |
rous | -0.061 |
Token | always |
Token ID | 1464 |
Feature activation | -3.83e+00 |
Pos logit contributions | |
bug | +0.007 |
run | +0.007 |
ours | +0.006 |
idd | +0.006 |
haired | +0.006 |
Neg logit contributions | |
amen | -0.006 |
girls | -0.005 |
Rudolph | -0.005 |
announcements | -0.005 |
hospitality | -0.005 |
Token | tried |
Token ID | 3088 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
amen | +0.123 |
girls | +0.119 |
Rudolph | +0.111 |
Santa | +0.108 |
announcements | +0.107 |
Neg logit contributions | |
run | -0.124 |
bug | -0.120 |
ours | -0.108 |
idd | -0.106 |
imb | -0.105 |
Token | to |
Token ID | 284 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
amen | +0.006 |
girls | +0.005 |
Rudolph | +0.005 |
announcements | +0.005 |
Santa | +0.005 |
Neg logit contributions | |
run | -0.005 |
bug | -0.005 |
ours | -0.004 |
udge | -0.004 |
imb | -0.004 |
Token | check |
Token ID | 2198 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
amen | +0.049 |
girls | +0.046 |
Rudolph | +0.045 |
Santa | +0.042 |
announcements | +0.042 |
Neg logit contributions | |
bug | -0.047 |
run | -0.046 |
nail | -0.042 |
udge | -0.042 |
imb | -0.041 |
Token | the |
Token ID | 262 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
run | +0.006 |
bug | +0.006 |
ours | +0.005 |
udge | +0.005 |
idd | +0.005 |
Neg logit contributions | |
amen | -0.007 |
girls | -0.006 |
Rudolph | -0.006 |
announcements | -0.006 |
Santa | -0.006 |
Token | facts |
Token ID | 6419 |
Feature activation | -1.03e+01 |
Pos logit contributions | |
amen | +0.064 |
girls | +0.059 |
Rudolph | +0.055 |
announcements | +0.054 |
Santa | +0.053 |
Neg logit contributions | |
run | -0.087 |
bug | -0.084 |
ours | -0.078 |
idd | -0.077 |
udge | -0.076 |
Token | before |
Token ID | 878 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
amen | +0.341 |
girls | +0.322 |
Rudolph | +0.306 |
announcements | +0.296 |
Santa | +0.294 |
Neg logit contributions | |
run | -0.320 |
bug | -0.309 |
ours | -0.279 |
udge | -0.273 |
idd | -0.273 |
Token | she |
Token ID | 673 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
amen | +0.053 |
girls | +0.051 |
mercy | +0.047 |
announcements | +0.047 |
hospitality | +0.047 |
Neg logit contributions | |
run | -0.046 |
bug | -0.042 |
ours | -0.040 |
imb | -0.040 |
idd | -0.039 |
Token | rose |
Token ID | 8278 |
Feature activation | +3.72e+00 |
Pos logit contributions | |
run | +0.010 |
bug | +0.009 |
ours | +0.009 |
udge | +0.009 |
idd | +0.009 |
Neg logit contributions | |
amen | -0.009 |
girls | -0.008 |
Rudolph | -0.008 |
announcements | -0.008 |
Santa | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
run | +0.103 |
bug | +0.100 |
ours | +0.088 |
imb | +0.085 |
idd | +0.084 |
Neg logit contributions | |
amen | -0.138 |
girls | -0.128 |
Rudolph | -0.123 |
announcements | -0.121 |
Santa | -0.120 |
Token | believe |
Token ID | 1975 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
amen | +0.045 |
girls | +0.043 |
Rudolph | +0.039 |
announcements | +0.038 |
sets | +0.038 |
Neg logit contributions | |
run | -0.050 |
bug | -0.050 |
udge | -0.044 |
imb | -0.044 |
ours | -0.043 |
Token | have |
Token ID | 423 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
amen | +0.068 |
girls | +0.063 |
Rudolph | +0.059 |
announcements | +0.059 |
hospitality | +0.058 |
Neg logit contributions | |
bug | -0.075 |
run | -0.074 |
udge | -0.067 |
imb | -0.066 |
ours | -0.066 |
Token | more |
Token ID | 517 |
Feature activation | -5.61e+00 |
Pos logit contributions | |
run | +0.061 |
bug | +0.060 |
ours | +0.053 |
idd | +0.052 |
rous | +0.052 |
Neg logit contributions | |
amen | -0.070 |
girls | -0.069 |
announcements | -0.062 |
Claus | -0.061 |
hospitality | -0.061 |
Token | fun |
Token ID | 1257 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
amen | +0.158 |
girls | +0.147 |
Rudolph | +0.139 |
announcements | +0.134 |
Santa | +0.133 |
Neg logit contributions | |
run | -0.204 |
bug | -0.199 |
ours | -0.183 |
idd | -0.180 |
imb | -0.180 |
Token | when |
Token ID | 618 |
Feature activation | -2.70e+00 |
Pos logit contributions | |
amen | +0.088 |
girls | +0.083 |
Rudolph | +0.079 |
announcements | +0.076 |
hospitality | +0.075 |
Neg logit contributions | |
run | -0.089 |
bug | -0.086 |
ours | -0.079 |
haired | -0.078 |
imb | -0.078 |
Token | you |
Token ID | 345 |
Feature activation | +5.25e-01 |
Pos logit contributions | |
amen | +0.078 |
girls | +0.071 |
Rudolph | +0.069 |
announcements | +0.065 |
Santa | +0.065 |
Neg logit contributions | |
run | -0.097 |
bug | -0.094 |
haired | -0.086 |
udge | -0.086 |
ours | -0.086 |
Token | cooperate |
Token ID | 21270 |
Feature activation | -1.03e+01 |
Pos logit contributions | |
bug | +0.018 |
run | +0.018 |
ours | +0.016 |
imb | +0.016 |
idd | +0.016 |
Neg logit contributions | |
amen | -0.016 |
girls | -0.015 |
Rudolph | -0.014 |
Santa | -0.014 |
announcements | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | +3.75e-01 |
Pos logit contributions | |
amen | +0.319 |
girls | +0.302 |
Rudolph | +0.283 |
announcements | +0.280 |
hospitality | +0.272 |
Neg logit contributions | |
run | -0.341 |
bug | -0.329 |
udge | -0.300 |
ours | -0.300 |
idd | -0.292 |
Token | compromise |
Token ID | 13110 |
Feature activation | -7.34e+00 |
Pos logit contributions | |
run | +0.013 |
bug | +0.013 |
imb | +0.011 |
ours | +0.011 |
haired | +0.011 |
Neg logit contributions | |
amen | -0.010 |
girls | -0.010 |
Rudolph | -0.010 |
Santa | -0.009 |
announcements | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +5.48e-01 |
Pos logit contributions | |
amen | +0.209 |
girls | +0.192 |
Rudolph | +0.187 |
announcements | +0.182 |
Santa | +0.175 |
Neg logit contributions | |
run | -0.266 |
bug | -0.254 |
ours | -0.237 |
udge | -0.231 |
out | -0.227 |
Token | Match |
Token ID | 13225 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
run | +0.015 |
bug | +0.014 |
ours | +0.013 |
imb | +0.013 |
idd | +0.013 |
Neg logit contributions | |
amen | -0.020 |
girls | -0.019 |
announcements | -0.018 |
Rudolph | -0.018 |
hospitality | -0.018 |
Token | ing |
Token ID | 278 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
run | +0.048 |
bug | +0.047 |
ours | +0.044 |
idd | +0.041 |
imb | +0.041 |
Neg logit contributions | |
amen | -0.057 |
girls | -0.053 |
announcements | -0.051 |
Santa | -0.051 |
Rudolph | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
amen | +0.130 |
girls | +0.122 |
Rudolph | +0.116 |
announcements | +0.115 |
Santa | +0.112 |
Neg logit contributions | |
bug | -0.163 |
run | -0.161 |
ours | -0.143 |
idd | -0.140 |
out | -0.140 |
Token | They |
Token ID | 1119 |
Feature activation | -5.98e+00 |
Pos logit contributions | |
run | +0.012 |
bug | +0.012 |
ours | +0.011 |
idd | +0.011 |
imb | +0.011 |
Neg logit contributions | |
amen | -0.014 |
girls | -0.013 |
Rudolph | -0.012 |
announcements | -0.012 |
hospitality | -0.012 |
Token | work |
Token ID | 670 |
Feature activation | -4.63e+00 |
Pos logit contributions | |
amen | +0.184 |
Rudolph | +0.169 |
girls | +0.168 |
Santa | +0.166 |
sets | +0.160 |
Neg logit contributions | |
bug | -0.200 |
run | -0.180 |
haired | -0.163 |
ours | -0.162 |
nail | -0.161 |
Token | hard |
Token ID | 1327 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
amen | +0.172 |
girls | +0.163 |
Rudolph | +0.156 |
announcements | +0.152 |
Santa | +0.151 |
Neg logit contributions | |
run | -0.126 |
bug | -0.121 |
ours | -0.108 |
idd | -0.106 |
udge | -0.105 |
Token | and |
Token ID | 290 |
Feature activation | +3.85e-01 |
Pos logit contributions | |
amen | +0.010 |
girls | +0.009 |
Rudolph | +0.009 |
Santa | +0.009 |
announcements | +0.009 |
Neg logit contributions | |
run | -0.009 |
bug | -0.008 |
ours | -0.008 |
udge | -0.007 |
idd | -0.007 |
Token | cooperate |
Token ID | 21270 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
run | +0.013 |
bug | +0.012 |
ours | +0.011 |
haired | +0.011 |
rous | +0.010 |
Neg logit contributions | |
amen | -0.012 |
girls | -0.011 |
Rudolph | -0.011 |
Santa | -0.011 |
sets | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
amen | +0.334 |
girls | +0.314 |
Rudolph | +0.305 |
announcements | +0.296 |
hospitality | +0.287 |
Neg logit contributions | |
run | -0.322 |
bug | -0.297 |
udge | -0.282 |
out | -0.267 |
ours | -0.266 |
Token | express |
Token ID | 4911 |
Feature activation | -5.65e+00 |
Pos logit contributions | |
run | +0.012 |
bug | +0.012 |
nail | +0.010 |
rous | +0.010 |
haired | +0.010 |
Neg logit contributions | |
Rudolph | -0.011 |
amen | -0.011 |
girls | -0.010 |
Santa | -0.010 |
sets | -0.010 |
Token | themselves |
Token ID | 2405 |
Feature activation | -5.51e+00 |
Pos logit contributions | |
amen | +0.157 |
girls | +0.143 |
Rudolph | +0.137 |
announcements | +0.134 |
hospitality | +0.131 |
Neg logit contributions | |
run | -0.209 |
bug | -0.195 |
udge | -0.189 |
ours | -0.184 |
haired | -0.179 |
Token | . |
Token ID | 13 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
amen | +0.150 |
girls | +0.140 |
Rudolph | +0.134 |
Santa | +0.127 |
announcements | +0.127 |
Neg logit contributions | |
run | -0.205 |
bug | -0.199 |
ours | -0.182 |
udge | -0.177 |
haired | -0.177 |
Token | They |
Token ID | 1119 |
Feature activation | -6.09e+00 |
Pos logit contributions | |
run | +0.007 |
bug | +0.007 |
ours | +0.006 |
idd | +0.006 |
imb | +0.006 |
Neg logit contributions | |
amen | -0.006 |
girls | -0.005 |
announcements | -0.005 |
hospitality | -0.005 |
mercy | -0.005 |
Token | 's |
Token ID | 338 |
Feature activation | -3.01e-01 |
Pos logit contributions | |
run | +0.089 |
bug | +0.088 |
ours | +0.079 |
imb | +0.076 |
idd | +0.076 |
Neg logit contributions | |
amen | -0.109 |
girls | -0.104 |
Rudolph | -0.099 |
announcements | -0.096 |
Santa | -0.096 |
Token | important |
Token ID | 1593 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
amen | +0.010 |
girls | +0.009 |
Rudolph | +0.009 |
announcements | +0.008 |
Santa | +0.008 |
Neg logit contributions | |
run | -0.010 |
bug | -0.009 |
ours | -0.009 |
imb | -0.008 |
udge | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
run | +0.039 |
bug | +0.038 |
ours | +0.034 |
idd | +0.033 |
imb | +0.033 |
Neg logit contributions | |
amen | -0.048 |
girls | -0.046 |
Rudolph | -0.044 |
announcements | -0.043 |
Santa | -0.042 |
Token | check |
Token ID | 2198 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
girls | +0.048 |
amen | +0.047 |
Rudolph | +0.042 |
announcements | +0.042 |
sets | +0.042 |
Neg logit contributions | |
bug | -0.045 |
run | -0.044 |
imb | -0.041 |
nail | -0.041 |
udge | -0.040 |
Token | the |
Token ID | 262 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
run | +0.005 |
bug | +0.005 |
ours | +0.004 |
udge | +0.004 |
imb | +0.004 |
Neg logit contributions | |
amen | -0.006 |
girls | -0.006 |
Rudolph | -0.005 |
announcements | -0.005 |
Santa | -0.005 |
Token | facts |
Token ID | 6419 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
amen | +0.071 |
girls | +0.067 |
Rudolph | +0.063 |
announcements | +0.060 |
Santa | +0.060 |
Neg logit contributions | |
run | -0.087 |
bug | -0.085 |
ours | -0.078 |
imb | -0.077 |
idd | -0.077 |
Token | before |
Token ID | 878 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
amen | +0.330 |
girls | +0.311 |
Rudolph | +0.296 |
announcements | +0.287 |
Santa | +0.285 |
Neg logit contributions | |
run | -0.321 |
bug | -0.312 |
ours | -0.284 |
idd | -0.279 |
imb | -0.278 |
Token | you |
Token ID | 345 |
Feature activation | +5.89e-01 |
Pos logit contributions | |
amen | +0.053 |
girls | +0.052 |
Rudolph | +0.048 |
announcements | +0.047 |
mercy | +0.047 |
Neg logit contributions | |
run | -0.044 |
bug | -0.042 |
imb | -0.041 |
ours | -0.040 |
idd | -0.038 |
Token | believe |
Token ID | 1975 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
run | +0.020 |
bug | +0.020 |
imb | +0.017 |
ours | +0.017 |
udge | +0.017 |
Neg logit contributions | |
amen | -0.018 |
girls | -0.017 |
Rudolph | -0.016 |
Santa | -0.016 |
announcements | -0.016 |
Token | it |
Token ID | 340 |
Feature activation | +3.10e+00 |
Pos logit contributions | |
amen | +0.005 |
girls | +0.005 |
Rudolph | +0.005 |
announcements | +0.005 |
Santa | +0.005 |
Neg logit contributions | |
run | -0.005 |
bug | -0.005 |
ours | -0.004 |
idd | -0.004 |
imb | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +7.19e-02 |
Pos logit contributions | |
run | +0.113 |
bug | +0.109 |
ours | +0.101 |
imb | +0.098 |
udge | +0.098 |
Neg logit contributions | |
amen | -0.086 |
girls | -0.081 |
Rudolph | -0.076 |
announcements | -0.074 |
Santa | -0.073 |
Token | learn |
Token ID | 2193 |
Feature activation | -6.16e+00 |
Pos logit contributions | |
amen | +0.189 |
girls | +0.183 |
sets | +0.176 |
Rudolph | +0.175 |
Santa | +0.174 |
Neg logit contributions | |
bug | -0.199 |
run | -0.171 |
ours | -0.163 |
imb | -0.154 |
nail | -0.153 |
Token | many |
Token ID | 867 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
amen | +0.213 |
girls | +0.208 |
Rudolph | +0.194 |
Claus | +0.191 |
Santa | +0.190 |
Neg logit contributions | |
run | -0.177 |
bug | -0.172 |
ours | -0.156 |
out | -0.146 |
haired | -0.145 |
Token | new |
Token ID | 649 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
amen | +0.040 |
girls | +0.037 |
Rudolph | +0.034 |
Santa | +0.032 |
Claus | +0.031 |
Neg logit contributions | |
run | -0.091 |
bug | -0.089 |
ours | -0.083 |
udge | -0.082 |
haired | -0.082 |
Token | words |
Token ID | 2456 |
Feature activation | -5.51e+00 |
Pos logit contributions | |
run | +0.068 |
bug | +0.066 |
ours | +0.062 |
udge | +0.062 |
haired | +0.062 |
Neg logit contributions | |
amen | -0.029 |
girls | -0.026 |
Rudolph | -0.025 |
Santa | -0.023 |
Claus | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
amen | +0.166 |
girls | +0.154 |
Rudolph | +0.144 |
announcements | +0.143 |
Santa | +0.140 |
Neg logit contributions | |
run | -0.192 |
bug | -0.189 |
ours | -0.167 |
udge | -0.165 |
haired | -0.163 |
Token | facts |
Token ID | 6419 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
run | +0.014 |
bug | +0.014 |
ours | +0.012 |
rous | +0.012 |
haired | +0.012 |
Neg logit contributions | |
amen | -0.010 |
Rudolph | -0.010 |
Santa | -0.010 |
girls | -0.010 |
Jose | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
amen | +0.315 |
girls | +0.301 |
Rudolph | +0.289 |
Claus | +0.273 |
hospitality | +0.272 |
Neg logit contributions | |
run | -0.345 |
bug | -0.321 |
ours | -0.289 |
udge | -0.285 |
out | -0.284 |
Token | They |
Token ID | 1119 |
Feature activation | -6.03e+00 |
Pos logit contributions | |
run | +0.008 |
bug | +0.008 |
ours | +0.008 |
idd | +0.007 |
udge | +0.007 |
Neg logit contributions | |
amen | -0.004 |
girls | -0.004 |
Rudolph | -0.004 |
announcements | -0.003 |
Santa | -0.003 |
Token | enjoy |
Token ID | 2883 |
Feature activation | -3.48e+00 |
Pos logit contributions | |
amen | +0.186 |
Rudolph | +0.175 |
girls | +0.175 |
sets | +0.171 |
Santa | +0.171 |
Neg logit contributions | |
bug | -0.205 |
run | -0.164 |
ours | -0.161 |
nail | -0.158 |
rous | -0.154 |
Token | the |
Token ID | 262 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
amen | +0.115 |
mercy | +0.106 |
girls | +0.106 |
Claus | +0.105 |
Rudolph | +0.103 |
Neg logit contributions | |
bug | -0.095 |
run | -0.092 |
rous | -0.085 |
ours | -0.082 |
udge | -0.081 |
Token | mild |
Token ID | 11607 |
Feature activation | -5.52e-01 |
Pos logit contributions | |
amen | +0.065 |
girls | +0.062 |
Rudolph | +0.058 |
Santa | +0.054 |
announcements | +0.054 |
Neg logit contributions | |
bug | -0.107 |
run | -0.104 |
imb | -0.097 |
haired | -0.097 |
ours | -0.096 |
Token | 's |
Token ID | 338 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
run | +0.090 |
bug | +0.087 |
ours | +0.079 |
udge | +0.078 |
idd | +0.078 |
Neg logit contributions | |
amen | -0.102 |
girls | -0.096 |
Rudolph | -0.092 |
announcements | -0.089 |
Santa | -0.088 |
Token | full |
Token ID | 1336 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
amen | +0.004 |
girls | +0.004 |
Rudolph | +0.004 |
announcements | +0.004 |
Santa | +0.004 |
Neg logit contributions | |
run | -0.004 |
bug | -0.004 |
ours | -0.004 |
imb | -0.004 |
idd | -0.004 |
Token | of |
Token ID | 286 |
Feature activation | -5.09e-01 |
Pos logit contributions | |
run | +0.022 |
bug | +0.021 |
imb | +0.017 |
haired | +0.017 |
ours | +0.017 |
Neg logit contributions | |
amen | -0.048 |
girls | -0.047 |
announcements | -0.044 |
Rudolph | -0.044 |
Santa | -0.043 |
Token | stories |
Token ID | 3923 |
Feature activation | -7.83e+00 |
Pos logit contributions | |
amen | +0.013 |
girls | +0.012 |
Rudolph | +0.011 |
announcements | +0.011 |
Santa | +0.011 |
Neg logit contributions | |
run | -0.020 |
bug | -0.019 |
ours | -0.017 |
udge | -0.017 |
idd | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | +6.86e-01 |
Pos logit contributions | |
amen | +0.260 |
girls | +0.249 |
announcements | +0.233 |
Rudolph | +0.230 |
mercy | +0.227 |
Neg logit contributions | |
run | -0.238 |
bug | -0.232 |
ours | -0.205 |
haired | -0.202 |
imb | -0.200 |
Token | facts |
Token ID | 6419 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
run | +0.024 |
bug | +0.023 |
ours | +0.020 |
haired | +0.020 |
imb | +0.020 |
Neg logit contributions | |
amen | -0.019 |
girls | -0.019 |
Rudolph | -0.017 |
Santa | -0.017 |
announcements | -0.017 |
Token | from |
Token ID | 422 |
Feature activation | -7.46e-01 |
Pos logit contributions | |
amen | +0.344 |
girls | +0.343 |
Rudolph | +0.316 |
announcements | +0.314 |
mercy | +0.310 |
Neg logit contributions | |
run | -0.297 |
bug | -0.270 |
ab | -0.243 |
ours | -0.243 |
out | -0.236 |
Token | all |
Token ID | 477 |
Feature activation | +4.94e-01 |
Pos logit contributions | |
amen | +0.022 |
girls | +0.021 |
Rudolph | +0.020 |
announcements | +0.019 |
Santa | +0.019 |
Neg logit contributions | |
run | -0.026 |
bug | -0.025 |
ours | -0.023 |
idd | -0.023 |
imb | -0.023 |
Token | around |
Token ID | 1088 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
run | +0.015 |
bug | +0.014 |
ours | +0.013 |
haired | +0.012 |
imb | +0.012 |
Neg logit contributions | |
amen | -0.017 |
girls | -0.016 |
Rudolph | -0.015 |
announcements | -0.015 |
Santa | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | -2.54e+00 |
Pos logit contributions | |
run | +0.089 |
bug | +0.087 |
ours | +0.079 |
idd | +0.078 |
imb | +0.078 |
Neg logit contributions | |
amen | -0.086 |
girls | -0.080 |
Rudolph | -0.076 |
announcements | -0.074 |
Santa | -0.073 |
Token | world |
Token ID | 995 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
amen | +0.080 |
girls | +0.070 |
mercy | +0.067 |
hospitality | +0.067 |
Rudolph | +0.066 |
Neg logit contributions | |
bug | -0.083 |
run | -0.080 |
imb | -0.075 |
eh | -0.074 |
idd | -0.074 |
Token | filled |
Token ID | 5901 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
run | +0.045 |
bug | +0.043 |
ours | +0.039 |
haired | +0.038 |
idd | +0.038 |
Neg logit contributions | |
amen | -0.047 |
girls | -0.045 |
Rudolph | -0.042 |
announcements | -0.041 |
Santa | -0.041 |
Token | with |
Token ID | 351 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
amen | +0.055 |
girls | +0.053 |
Rudolph | +0.053 |
Santa | +0.051 |
stories | +0.050 |
Neg logit contributions | |
run | -0.046 |
bug | -0.041 |
ours | -0.041 |
idd | -0.039 |
imb | -0.038 |
Token | different |
Token ID | 1180 |
Feature activation | +2.26e+00 |
Pos logit contributions | |
run | +0.107 |
bug | +0.103 |
haired | +0.091 |
ours | +0.090 |
udge | +0.090 |
Neg logit contributions | |
amen | -0.094 |
girls | -0.087 |
Rudolph | -0.081 |
Claus | -0.079 |
Santa | -0.079 |
Token | stories |
Token ID | 3923 |
Feature activation | -7.74e+00 |
Pos logit contributions | |
run | +0.077 |
bug | +0.073 |
haired | +0.066 |
ours | +0.066 |
udge | +0.065 |
Neg logit contributions | |
amen | -0.069 |
girls | -0.064 |
Rudolph | -0.063 |
Santa | -0.059 |
Claus | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | +9.65e-01 |
Pos logit contributions | |
amen | +0.261 |
girls | +0.244 |
Rudolph | +0.225 |
announcements | +0.223 |
mercy | +0.223 |
Neg logit contributions | |
run | -0.243 |
bug | -0.236 |
haired | -0.205 |
out | -0.203 |
udge | -0.199 |
Token | facts |
Token ID | 6419 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
run | +0.036 |
bug | +0.036 |
haired | +0.031 |
ours | +0.031 |
rous | +0.030 |
Neg logit contributions | |
amen | -0.023 |
girls | -0.021 |
Rudolph | -0.020 |
Santa | -0.020 |
Claus | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
amen | +0.337 |
girls | +0.321 |
Rudolph | +0.305 |
hospitality | +0.291 |
Claus | +0.290 |
Neg logit contributions | |
run | -0.324 |
bug | -0.289 |
out | -0.263 |
haired | -0.257 |
ab | -0.255 |
Token | She |
Token ID | 1375 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
run | +0.013 |
bug | +0.013 |
ours | +0.012 |
udge | +0.012 |
idd | +0.012 |
Neg logit contributions | |
amen | -0.009 |
girls | -0.009 |
Rudolph | -0.008 |
Santa | -0.008 |
announcements | -0.008 |
Token | read |
Token ID | 1100 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
amen | +0.054 |
girls | +0.051 |
Rudolph | +0.048 |
announcements | +0.047 |
Santa | +0.046 |
Neg logit contributions | |
run | -0.058 |
bug | -0.056 |
ours | -0.051 |
idd | -0.051 |
imb | -0.050 |
Token | all |
Token ID | 477 |
Feature activation | +5.00e-01 |
Pos logit contributions | |
amen | +0.049 |
girls | +0.047 |
Rudolph | +0.043 |
announcements | +0.042 |
mercy | +0.042 |
Neg logit contributions | |
run | -0.050 |
bug | -0.049 |
haired | -0.044 |
ours | -0.044 |
udge | -0.042 |
Token | day |
Token ID | 1110 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
run | +0.018 |
bug | +0.018 |
idd | +0.017 |
ours | +0.016 |
imb | +0.016 |
Neg logit contributions | |
amen | -0.014 |
girls | -0.012 |
Rudolph | -0.012 |
stories | -0.012 |
announcements | -0.012 |
Token | also |
Token ID | 635 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
amen | +0.086 |
girls | +0.080 |
Rudolph | +0.076 |
announcements | +0.073 |
Santa | +0.073 |
Neg logit contributions | |
run | -0.101 |
bug | -0.100 |
ours | -0.090 |
idd | -0.089 |
udge | -0.089 |
Token | learned |
Token ID | 4499 |
Feature activation | -6.24e+00 |
Pos logit contributions | |
run | +0.057 |
bug | +0.056 |
ours | +0.050 |
idd | +0.050 |
udge | +0.049 |
Neg logit contributions | |
amen | -0.055 |
girls | -0.052 |
Rudolph | -0.049 |
announcements | -0.048 |
Santa | -0.047 |
Token | new |
Token ID | 649 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
amen | +0.229 |
girls | +0.221 |
Rudolph | +0.206 |
announcements | +0.203 |
hospitality | +0.200 |
Neg logit contributions | |
run | -0.172 |
bug | -0.160 |
ours | -0.148 |
out | -0.143 |
udge | -0.139 |
Token | words |
Token ID | 2456 |
Feature activation | -5.63e+00 |
Pos logit contributions | |
run | +0.051 |
bug | +0.046 |
ours | +0.044 |
udge | +0.043 |
haired | +0.042 |
Neg logit contributions | |
amen | -0.051 |
girls | -0.048 |
Rudolph | -0.047 |
Santa | -0.045 |
Claus | -0.045 |
Token | and |
Token ID | 290 |
Feature activation | +4.65e-01 |
Pos logit contributions | |
amen | +0.195 |
girls | +0.185 |
announcements | +0.175 |
Rudolph | +0.173 |
hospitality | +0.169 |
Neg logit contributions | |
run | -0.170 |
bug | -0.164 |
ours | -0.142 |
haired | -0.141 |
out | -0.138 |
Token | facts |
Token ID | 6419 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
run | +0.015 |
bug | +0.015 |
ours | +0.012 |
haired | +0.012 |
nail | +0.012 |
Neg logit contributions | |
amen | -0.015 |
Rudolph | -0.014 |
Santa | -0.013 |
girls | -0.013 |
William | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | +4.59e-01 |
Pos logit contributions | |
amen | +0.347 |
girls | +0.336 |
Rudolph | +0.319 |
announcements | +0.307 |
hospitality | +0.304 |
Neg logit contributions | |
run | -0.311 |
bug | -0.274 |
out | -0.247 |
ours | -0.247 |
udge | -0.242 |
Token | stories |
Token ID | 3923 |
Feature activation | -7.84e+00 |
Pos logit contributions | |
run | +0.017 |
bug | +0.017 |
haired | +0.014 |
ours | +0.014 |
nail | +0.014 |
Neg logit contributions | |
amen | -0.012 |
Rudolph | -0.011 |
Santa | -0.011 |
girls | -0.010 |
William | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -6.09e-03 |
Pos logit contributions | |
amen | +0.264 |
girls | +0.246 |
Rudolph | +0.234 |
announcements | +0.230 |
Santa | +0.227 |
Neg logit contributions | |
run | -0.243 |
bug | -0.226 |
out | -0.202 |
haired | -0.201 |
ab | -0.196 |
Token | They |
Token ID | 1119 |
Feature activation | -5.91e+00 |
Pos logit contributions | |
amen | +0.000 |
girls | +0.000 |
mercy | +0.000 |
hospitality | +0.000 |
announcements | +0.000 |
Neg logit contributions | |
run | -0.000 |
bug | -0.000 |
ours | -0.000 |
udge | -0.000 |
imb | -0.000 |
Token | had |
Token ID | 550 |
Feature activation | +2.80e+00 |
Pos logit contributions | |
amen | +0.202 |
sets | +0.184 |
girls | +0.183 |
Rudolph | +0.181 |
Santa | +0.178 |
Neg logit contributions | |
bug | -0.184 |
run | -0.152 |
haired | -0.142 |
nail | -0.135 |
ours | -0.135 |
Token | She |
Token ID | 1375 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
run | +0.034 |
bug | +0.033 |
idd | +0.031 |
ours | +0.031 |
udge | +0.030 |
Neg logit contributions | |
amen | -0.022 |
girls | -0.021 |
Rudolph | -0.019 |
announcements | -0.018 |
Santa | -0.018 |
Token | knew |
Token ID | 2993 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
amen | +0.051 |
girls | +0.048 |
Rudolph | +0.046 |
announcements | +0.044 |
Santa | +0.044 |
Neg logit contributions | |
run | -0.051 |
bug | -0.048 |
ours | -0.045 |
imb | -0.045 |
udge | -0.044 |
Token | many |
Token ID | 867 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
run | +0.066 |
bug | +0.064 |
udge | +0.060 |
ours | +0.059 |
idd | +0.059 |
Neg logit contributions | |
amen | -0.056 |
girls | -0.052 |
Rudolph | -0.050 |
hospitality | -0.047 |
mercy | -0.047 |
Token | words |
Token ID | 2456 |
Feature activation | -5.45e+00 |
Pos logit contributions | |
amen | +0.038 |
girls | +0.035 |
Rudolph | +0.032 |
announcements | +0.032 |
stories | +0.031 |
Neg logit contributions | |
run | -0.069 |
bug | -0.069 |
idd | -0.065 |
imb | -0.065 |
ours | -0.064 |
Token | and |
Token ID | 290 |
Feature activation | +7.36e-01 |
Pos logit contributions | |
amen | +0.175 |
girls | +0.164 |
Rudolph | +0.156 |
announcements | +0.151 |
Santa | +0.150 |
Neg logit contributions | |
run | -0.177 |
bug | -0.172 |
ours | -0.154 |
haired | -0.152 |
udge | -0.150 |
Token | facts |
Token ID | 6419 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
run | +0.024 |
bug | +0.023 |
haired | +0.021 |
ours | +0.020 |
udge | +0.020 |
Neg logit contributions | |
amen | -0.024 |
girls | -0.022 |
Rudolph | -0.022 |
Santa | -0.021 |
Claus | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +8.04e-01 |
Pos logit contributions | |
amen | +0.328 |
girls | +0.305 |
Rudolph | +0.298 |
hospitality | +0.284 |
Claus | +0.282 |
Neg logit contributions | |
run | -0.335 |
bug | -0.301 |
haired | -0.270 |
ours | -0.270 |
out | -0.268 |
Token | Sam |
Token ID | 3409 |
Feature activation | -6.65e-01 |
Pos logit contributions | |
run | +0.033 |
bug | +0.032 |
idd | +0.031 |
ours | +0.030 |
eh | +0.030 |
Neg logit contributions | |
girls | -0.017 |
amen | -0.017 |
Rudolph | -0.015 |
sets | -0.014 |
Santa | -0.014 |
Token | was |
Token ID | 373 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
amen | +0.022 |
girls | +0.021 |
Rudolph | +0.020 |
announcements | +0.020 |
Santa | +0.019 |
Neg logit contributions | |
run | -0.020 |
bug | -0.019 |
ours | -0.018 |
udge | -0.018 |
imb | -0.018 |
Token | very |
Token ID | 845 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
run | +0.045 |
bug | +0.043 |
ours | +0.040 |
imb | +0.039 |
udge | +0.039 |
Neg logit contributions | |
amen | -0.039 |
girls | -0.037 |
Rudolph | -0.035 |
announcements | -0.034 |
Santa | -0.034 |
Token | brave |
Token ID | 14802 |
Feature activation | -2.69e+00 |
Pos logit contributions | |
run | +0.035 |
bug | +0.034 |
ours | +0.032 |
imb | +0.032 |
udge | +0.031 |
Neg logit contributions | |
amen | -0.030 |
girls | -0.027 |
Rudolph | -0.027 |
Santa | -0.026 |
announcements | -0.026 |
Token | to |
Token ID | 284 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
run | +0.059 |
bug | +0.057 |
ours | +0.051 |
idd | +0.050 |
imb | +0.050 |
Neg logit contributions | |
amen | -0.073 |
girls | -0.069 |
Rudolph | -0.066 |
announcements | -0.064 |
Santa | -0.064 |
Token | learn |
Token ID | 2193 |
Feature activation | -5.95e+00 |
Pos logit contributions | |
amen | +0.056 |
sets | +0.056 |
girls | +0.056 |
Rudolph | +0.053 |
Claus | +0.049 |
Neg logit contributions | |
udge | -0.035 |
bug | -0.035 |
run | -0.034 |
nail | -0.034 |
imb | -0.033 |
Token | to |
Token ID | 284 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
amen | +0.186 |
girls | +0.171 |
Rudolph | +0.167 |
mercy | +0.162 |
Santa | +0.162 |
Neg logit contributions | |
run | -0.194 |
bug | -0.192 |
idd | -0.176 |
imb | -0.175 |
ang | -0.172 |
Token | share |
Token ID | 2648 |
Feature activation | -4.61e+00 |
Pos logit contributions | |
amen | +0.051 |
girls | +0.050 |
Rudolph | +0.048 |
sets | +0.048 |
Claus | +0.045 |
Neg logit contributions | |
run | -0.039 |
bug | -0.037 |
udge | -0.037 |
nail | -0.037 |
out | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
amen | +0.129 |
girls | +0.121 |
Rudolph | +0.114 |
Santa | +0.110 |
announcements | +0.109 |
Neg logit contributions | |
run | -0.167 |
bug | -0.162 |
ours | -0.149 |
idd | -0.148 |
imb | -0.148 |
Token | cooperate |
Token ID | 21270 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
run | +0.014 |
bug | +0.013 |
ours | +0.012 |
haired | +0.012 |
imb | +0.012 |
Neg logit contributions | |
amen | -0.010 |
girls | -0.010 |
Rudolph | -0.009 |
Santa | -0.009 |
announcements | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
amen | +0.281 |
girls | +0.261 |
Rudolph | +0.247 |
announcements | +0.242 |
hospitality | +0.232 |
Neg logit contributions | |
run | -0.369 |
bug | -0.353 |
udge | -0.329 |
ours | -0.328 |
haired | -0.321 |
Token | How |
Token ID | 1374 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
run | +0.011 |
bug | +0.010 |
ours | +0.009 |
idd | +0.009 |
imb | +0.009 |
Neg logit contributions | |
amen | -0.013 |
girls | -0.012 |
Rudolph | -0.012 |
announcements | -0.011 |
Santa | -0.011 |
Token | about |
Token ID | 546 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
run | +0.054 |
bug | +0.052 |
ours | +0.048 |
imb | +0.047 |
udge | +0.046 |
Neg logit contributions | |
amen | -0.060 |
girls | -0.057 |
Rudolph | -0.054 |
announcements | -0.052 |
Santa | -0.052 |
Token | this |
Token ID | 428 |
Feature activation | +3.02e+00 |
Pos logit contributions | |
amen | +0.035 |
girls | +0.033 |
announcements | +0.032 |
Santa | +0.031 |
hospitality | +0.031 |
Neg logit contributions | |
run | -0.030 |
bug | -0.030 |
ours | -0.028 |
imb | -0.026 |
idd | -0.026 |
Token | : |
Token ID | 25 |
Feature activation | +3.74e+00 |
Pos logit contributions | |
run | +0.101 |
bug | +0.099 |
ours | +0.088 |
udge | +0.088 |
imb | +0.088 |
Neg logit contributions | |
amen | -0.097 |
Rudolph | -0.087 |
girls | -0.086 |
Santa | -0.084 |
hospitality | -0.081 |
Token | . |
Token ID | 13 |
Feature activation | +6.76e-02 |
Pos logit contributions | |
amen | +0.020 |
girls | +0.018 |
Rudolph | +0.017 |
announcements | +0.017 |
Santa | +0.016 |
Neg logit contributions | |
run | -0.028 |
bug | -0.027 |
ours | -0.024 |
udge | -0.024 |
idd | -0.024 |
Token | Instead |
Token ID | 5455 |
Feature activation | -2.86e+00 |
Pos logit contributions | |
run | +0.002 |
bug | +0.002 |
ours | +0.002 |
idd | +0.002 |
imb | +0.002 |
Neg logit contributions | |
amen | -0.002 |
girls | -0.002 |
Rudolph | -0.002 |
announcements | -0.002 |
Santa | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -8.71e-01 |
Pos logit contributions | |
amen | +0.087 |
girls | +0.083 |
Rudolph | +0.078 |
announcements | +0.076 |
Santa | +0.076 |
Neg logit contributions | |
run | -0.099 |
bug | -0.094 |
ours | -0.087 |
imb | -0.084 |
idd | -0.083 |
Token | we |
Token ID | 356 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
amen | +0.027 |
girls | +0.025 |
Rudolph | +0.023 |
announcements | +0.023 |
Santa | +0.023 |
Neg logit contributions | |
run | -0.028 |
bug | -0.028 |
ours | -0.025 |
imb | -0.025 |
idd | -0.024 |
Token | should |
Token ID | 815 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
run | +0.009 |
bug | +0.009 |
ours | +0.008 |
idd | +0.008 |
imb | +0.008 |
Neg logit contributions | |
amen | -0.009 |
girls | -0.009 |
Rudolph | -0.008 |
announcements | -0.008 |
Santa | -0.008 |
Token | cooperate |
Token ID | 21270 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
amen | +0.075 |
girls | +0.072 |
sets | +0.068 |
Rudolph | +0.067 |
announcements | +0.066 |
Neg logit contributions | |
bug | -0.078 |
run | -0.077 |
imb | -0.070 |
ours | -0.068 |
nail | -0.068 |
Token | with |
Token ID | 351 |
Feature activation | +2.80e+00 |
Pos logit contributions | |
amen | +0.346 |
girls | +0.325 |
Rudolph | +0.310 |
announcements | +0.306 |
hospitality | +0.296 |
Neg logit contributions | |
run | -0.307 |
bug | -0.288 |
udge | -0.262 |
ours | -0.259 |
idd | -0.252 |
Token | others |
Token ID | 1854 |
Feature activation | -5.58e+00 |
Pos logit contributions | |
run | +0.099 |
bug | +0.095 |
ours | +0.087 |
udge | +0.086 |
imb | +0.085 |
Neg logit contributions | |
amen | -0.081 |
girls | -0.076 |
Rudolph | -0.070 |
announcements | -0.070 |
Santa | -0.069 |
Token | and |
Token ID | 290 |
Feature activation | +5.68e-01 |
Pos logit contributions | |
amen | +0.174 |
girls | +0.162 |
Rudolph | +0.153 |
announcements | +0.153 |
Santa | +0.150 |
Neg logit contributions | |
run | -0.188 |
bug | -0.181 |
ours | -0.161 |
udge | -0.158 |
haired | -0.158 |
Token | use |
Token ID | 779 |
Feature activation | +8.40e-02 |
Pos logit contributions | |
run | +0.020 |
bug | +0.020 |
ours | +0.017 |
imb | +0.017 |
nail | +0.017 |
Neg logit contributions | |
amen | -0.016 |
girls | -0.015 |
Rudolph | -0.015 |
Santa | -0.014 |
announcements | -0.014 |
Token | teamwork |
Token ID | 48424 |
Feature activation | -5.92e+00 |
Pos logit contributions | |
run | +0.003 |
bug | +0.003 |
ours | +0.003 |
idd | +0.003 |
haired | +0.003 |
Neg logit contributions | |
amen | -0.002 |
girls | -0.002 |
Rudolph | -0.002 |
announcements | -0.002 |
Santa | -0.002 |
Token | too |
Token ID | 1165 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
amen | +0.191 |
girls | +0.176 |
announcements | +0.172 |
Rudolph | +0.171 |
Santa | +0.167 |
Neg logit contributions | |
bug | -0.161 |
run | -0.152 |
idd | -0.132 |
ab | -0.129 |
out | -0.127 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
amen | +0.068 |
girls | +0.064 |
hospitality | +0.062 |
Rudolph | +0.061 |
sets | +0.060 |
Neg logit contributions | |
run | -0.065 |
bug | -0.065 |
idd | -0.057 |
elly | -0.057 |
ang | -0.056 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.69e+00 |
Pos logit contributions | |
amen | +0.041 |
Rudolph | +0.038 |
girls | +0.037 |
announcements | +0.036 |
Santa | +0.035 |
Neg logit contributions | |
run | -0.038 |
bug | -0.037 |
imb | -0.034 |
idd | -0.034 |
ours | -0.033 |
Token | says |
Token ID | 1139 |
Feature activation | -8.76e-01 |
Pos logit contributions | |
run | +0.100 |
bug | +0.097 |
ours | +0.090 |
idd | +0.089 |
imb | +0.089 |
Neg logit contributions | |
amen | -0.073 |
girls | -0.068 |
Rudolph | -0.064 |
announcements | -0.062 |
Santa | -0.061 |
Token | , |
Token ID | 11 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
amen | +0.028 |
girls | +0.027 |
Rudolph | +0.027 |
Santa | +0.026 |
North | +0.025 |
Neg logit contributions | |
run | -0.031 |
bug | -0.029 |
udge | -0.025 |
haired | -0.024 |
earing | -0.024 |
Token | pointing |
Token ID | 10609 |
Feature activation | +5.04e+00 |
Pos logit contributions | |
amen | +0.012 |
girls | +0.011 |
Rudolph | +0.010 |
Santa | +0.010 |
announcements | +0.010 |
Neg logit contributions | |
run | -0.010 |
bug | -0.010 |
haired | -0.009 |
ours | -0.009 |
idd | -0.009 |
Token | to |
Token ID | 284 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
run | +0.158 |
bug | +0.152 |
imb | +0.146 |
idd | +0.146 |
eh | +0.144 |
Neg logit contributions | |
amen | -0.160 |
girls | -0.148 |
Rudolph | -0.146 |
announcements | -0.140 |
stories | -0.138 |
Token | the |
Token ID | 262 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
amen | +0.048 |
girls | +0.045 |
Santa | +0.042 |
Claus | +0.042 |
sets | +0.041 |
Neg logit contributions | |
bug | -0.031 |
run | -0.030 |
haired | -0.025 |
rous | -0.025 |
ours | -0.024 |
Token | book |
Token ID | 1492 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
amen | +0.063 |
girls | +0.058 |
Rudolph | +0.054 |
hospitality | +0.053 |
announcements | +0.052 |
Neg logit contributions | |
bug | -0.074 |
run | -0.074 |
ours | -0.067 |
haired | -0.067 |
idd | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | +6.84e-01 |
Pos logit contributions | |
run | +0.007 |
bug | +0.007 |
ours | +0.006 |
haired | +0.006 |
udge | +0.006 |
Neg logit contributions | |
amen | -0.007 |
girls | -0.007 |
Rudolph | -0.006 |
announcements | -0.006 |
Santa | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -3.74e-01 |
Pos logit contributions | |
run | +0.027 |
bug | +0.026 |
idd | +0.024 |
ours | +0.024 |
haired | +0.024 |
Neg logit contributions | |
amen | -0.017 |
girls | -0.016 |
Rudolph | -0.015 |
Santa | -0.014 |
announcements | -0.014 |
Token | They |
Token ID | 1119 |
Feature activation | -5.78e+00 |
Pos logit contributions | |
run | +0.005 |
bug | +0.005 |
ours | +0.005 |
idd | +0.005 |
imb | +0.005 |
Neg logit contributions | |
amen | -0.004 |
girls | -0.004 |
announcements | -0.004 |
Rudolph | -0.004 |
hospitality | -0.004 |
Token | forgot |
Token ID | 16453 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
amen | +0.184 |
girls | +0.173 |
announcements | +0.165 |
sets | +0.164 |
Santa | +0.163 |
Neg logit contributions | |
bug | -0.195 |
run | -0.167 |
haired | -0.150 |
idd | -0.148 |
ours | -0.147 |
Token | about |
Token ID | 546 |
Feature activation | -9.16e-01 |
Pos logit contributions | |
girls | +0.083 |
amen | +0.082 |
Santa | +0.077 |
Claus | +0.076 |
MacDonald | +0.076 |
Neg logit contributions | |
run | -0.047 |
bug | -0.044 |
ours | -0.039 |
out | -0.037 |
haired | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
amen | +0.029 |
girls | +0.027 |
Santa | +0.026 |
announcements | +0.026 |
Rudolph | +0.025 |
Neg logit contributions | |
run | -0.030 |
bug | -0.029 |
ours | -0.026 |
udge | -0.025 |
rous | -0.025 |
Token | red |
Token ID | 2266 |
Feature activation | +5.03e+00 |
Pos logit contributions | |
amen | +0.080 |
girls | +0.076 |
Rudolph | +0.072 |
announcements | +0.069 |
Santa | +0.069 |
Neg logit contributions | |
run | -0.077 |
bug | -0.075 |
ours | -0.068 |
idd | -0.067 |
haired | -0.067 |
Token | block |
Token ID | 2512 |
Feature activation | +4.42e+00 |
Pos logit contributions | |
run | +0.171 |
bug | +0.163 |
rous | +0.150 |
ang | +0.149 |
imb | +0.149 |
Neg logit contributions | |
amen | -0.145 |
girls | -0.139 |
stories | -0.135 |
sets | -0.134 |
Rudolph | -0.131 |
Token | and |
Token ID | 290 |
Feature activation | +6.36e-01 |
Pos logit contributions | |
run | +0.144 |
bug | +0.140 |
ours | +0.126 |
idd | +0.123 |
udge | +0.123 |
Neg logit contributions | |
amen | -0.141 |
girls | -0.132 |
Rudolph | -0.125 |
announcements | -0.122 |
Santa | -0.121 |
Token | the |
Token ID | 262 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
run | +0.022 |
bug | +0.021 |
ours | +0.019 |
haired | +0.018 |
idd | +0.018 |
Neg logit contributions | |
amen | -0.019 |
girls | -0.018 |
Santa | -0.017 |
Rudolph | -0.017 |
announcements | -0.016 |
Token | fight |
Token ID | 1907 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
amen | +0.082 |
girls | +0.077 |
Rudolph | +0.073 |
Santa | +0.070 |
announcements | +0.070 |
Neg logit contributions | |
bug | -0.075 |
run | -0.074 |
haired | -0.065 |
idd | -0.065 |
nail | -0.064 |
Token | . |
Token ID | 13 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
amen | +0.099 |
girls | +0.093 |
Rudolph | +0.089 |
Santa | +0.086 |
hospitality | +0.085 |
Neg logit contributions | |
run | -0.093 |
bug | -0.088 |
ours | -0.079 |
haired | -0.078 |
udge | -0.077 |
Token | They |
Token ID | 1119 |
Feature activation | -5.81e+00 |
Pos logit contributions | |
run | +0.005 |
bug | +0.005 |
ours | +0.005 |
idd | +0.004 |
imb | +0.004 |
Neg logit contributions | |
amen | -0.003 |
girls | -0.003 |
Rudolph | -0.003 |
announcements | -0.003 |
Santa | -0.003 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
Santa | +0.004 |
amen | +0.003 |
announcements | +0.003 |
Rudolph | +0.003 |
songs | +0.003 |
Neg logit contributions | |
run | -0.008 |
ours | -0.007 |
bug | -0.007 |
ays | -0.007 |
idd | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | -3.81e-01 |
Pos logit contributions | |
run | +0.080 |
bug | +0.077 |
udge | +0.073 |
ours | +0.072 |
idd | +0.072 |
Neg logit contributions | |
amen | -0.040 |
girls | -0.035 |
Rudolph | -0.034 |
announcements | -0.031 |
Santa | -0.031 |
Token | a |
Token ID | 257 |
Feature activation | -3.80e-01 |
Pos logit contributions | |
amen | +0.012 |
girls | +0.011 |
Rudolph | +0.010 |
announcements | +0.010 |
Santa | +0.010 |
Neg logit contributions | |
run | -0.013 |
bug | -0.013 |
ours | -0.012 |
idd | -0.012 |
udge | -0.012 |
Token | graceful |
Token ID | 44363 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
girls | +0.014 |
amen | +0.014 |
sets | +0.012 |
Rudolph | +0.012 |
hospitality | +0.012 |
Neg logit contributions | |
run | -0.010 |
bug | -0.010 |
idd | -0.009 |
haired | -0.009 |
nail | -0.009 |
Token | sw |
Token ID | 1509 |
Feature activation | +3.27e+00 |
Pos logit contributions | |
run | +0.046 |
bug | +0.045 |
ours | +0.041 |
imb | +0.040 |
idd | +0.040 |
Neg logit contributions | |
amen | -0.040 |
girls | -0.037 |
Rudolph | -0.036 |
announcements | -0.034 |
Santa | -0.034 |
Token | an |
Token ID | 272 |
Feature activation | +4.30e+00 |
Pos logit contributions | |
run | +0.089 |
bug | +0.086 |
ours | +0.076 |
imb | +0.074 |
haired | +0.073 |
Neg logit contributions | |
amen | -0.120 |
girls | -0.115 |
Rudolph | -0.110 |
announcements | -0.108 |
Santa | -0.106 |
Token | appeared |
Token ID | 4120 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
run | +0.143 |
bug | +0.135 |
ours | +0.127 |
udge | +0.126 |
imb | +0.126 |
Neg logit contributions | |
amen | -0.134 |
girls | -0.127 |
Rudolph | -0.120 |
announcements | -0.115 |
Santa | -0.114 |
Token | in |
Token ID | 287 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
run | +0.035 |
bug | +0.034 |
ours | +0.031 |
idd | +0.030 |
imb | +0.030 |
Neg logit contributions | |
amen | -0.031 |
girls | -0.029 |
Rudolph | -0.028 |
announcements | -0.027 |
Santa | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
run | +0.038 |
bug | +0.037 |
udge | +0.034 |
idd | +0.033 |
haired | +0.032 |
Neg logit contributions | |
amen | -0.035 |
girls | -0.032 |
Rudolph | -0.030 |
announcements | -0.030 |
Santa | -0.029 |
Token | water |
Token ID | 1660 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
amen | +0.066 |
girls | +0.062 |
hospitality | +0.056 |
announcements | +0.055 |
Rudolph | +0.055 |
Neg logit contributions | |
run | -0.083 |
bug | -0.083 |
ours | -0.077 |
idd | -0.077 |
haired | -0.076 |
Token | . |
Token ID | 13 |
Feature activation | +4.87e-01 |
Pos logit contributions | |
amen | +0.079 |
girls | +0.071 |
Rudolph | +0.068 |
announcements | +0.067 |
hospitality | +0.067 |
Neg logit contributions | |
bug | -0.092 |
run | -0.091 |
idd | -0.082 |
ours | -0.081 |
out | -0.081 |
Token | . |
Token ID | 13 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
run | +0.086 |
bug | +0.085 |
ours | +0.076 |
idd | +0.076 |
udge | +0.075 |
Neg logit contributions | |
amen | -0.068 |
girls | -0.063 |
Rudolph | -0.059 |
announcements | -0.058 |
Santa | -0.057 |
Token | She |
Token ID | 1375 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
run | +0.009 |
bug | +0.009 |
ours | +0.008 |
idd | +0.008 |
imb | +0.008 |
Neg logit contributions | |
amen | -0.008 |
girls | -0.008 |
Rudolph | -0.008 |
announcements | -0.007 |
Santa | -0.007 |
Token | was |
Token ID | 373 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
amen | +0.055 |
Rudolph | +0.051 |
girls | +0.051 |
Claus | +0.049 |
Santa | +0.046 |
Neg logit contributions | |
run | -0.066 |
imb | -0.065 |
ours | -0.065 |
rous | -0.064 |
udge | -0.063 |
Token | standing |
Token ID | 5055 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
run | +0.040 |
ours | +0.039 |
imb | +0.039 |
bug | +0.038 |
elly | +0.038 |
Neg logit contributions | |
amen | -0.033 |
Rudolph | -0.031 |
girls | -0.031 |
announcements | -0.029 |
Santa | -0.027 |
Token | right |
Token ID | 826 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
run | +0.047 |
bug | +0.044 |
ours | +0.040 |
imb | +0.039 |
haired | +0.039 |
Neg logit contributions | |
amen | -0.071 |
girls | -0.067 |
Rudolph | -0.066 |
Santa | -0.064 |
announcements | -0.063 |
Token | behind |
Token ID | 2157 |
Feature activation | +4.96e+00 |
Pos logit contributions | |
run | +0.023 |
bug | +0.022 |
ours | +0.019 |
idd | +0.018 |
udge | +0.018 |
Neg logit contributions | |
amen | -0.046 |
girls | -0.044 |
Rudolph | -0.042 |
announcements | -0.041 |
Santa | -0.041 |
Token | Big |
Token ID | 4403 |
Feature activation | -5.23e+00 |
Pos logit contributions | |
run | +0.162 |
bug | +0.157 |
idd | +0.139 |
ours | +0.139 |
udge | +0.138 |
Neg logit contributions | |
amen | -0.158 |
girls | -0.149 |
announcements | -0.136 |
Rudolph | -0.136 |
Santa | -0.135 |
Token | Bear |
Token ID | 14732 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
amen | +0.164 |
girls | +0.142 |
hospitality | +0.137 |
announcements | +0.136 |
Santa | +0.136 |
Neg logit contributions | |
run | -0.179 |
bug | -0.174 |
idd | -0.156 |
udge | -0.154 |
haired | -0.154 |
Token | , |
Token ID | 11 |
Feature activation | -6.93e-01 |
Pos logit contributions | |
amen | +0.087 |
announcements | +0.076 |
girls | +0.075 |
mercy | +0.074 |
Santa | +0.074 |
Neg logit contributions | |
bug | -0.085 |
run | -0.083 |
out | -0.070 |
udge | -0.069 |
ang | -0.069 |
Token | with |
Token ID | 351 |
Feature activation | +2.86e+00 |
Pos logit contributions | |
amen | +0.022 |
girls | +0.020 |
Rudolph | +0.019 |
announcements | +0.019 |
Santa | +0.019 |
Neg logit contributions | |
run | -0.023 |
bug | -0.022 |
ours | -0.021 |
imb | -0.020 |
udge | -0.020 |
Token | her |
Token ID | 607 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
run | +0.101 |
bug | +0.101 |
ours | +0.096 |
imb | +0.095 |
idd | +0.094 |
Neg logit contributions | |
amen | -0.079 |
girls | -0.074 |
Rudolph | -0.072 |
announcements | -0.068 |
mercy | -0.067 |
Token | away |
Token ID | 1497 |
Feature activation | -6.55e-01 |
Pos logit contributions | |
run | +0.010 |
bug | +0.009 |
udge | +0.008 |
out | +0.008 |
haired | +0.008 |
Neg logit contributions | |
amen | -0.011 |
girls | -0.010 |
Rudolph | -0.010 |
announcements | -0.009 |
hospitality | -0.009 |
Token | ! |
Token ID | 0 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
amen | +0.024 |
girls | +0.023 |
announcements | +0.022 |
Rudolph | +0.021 |
hospitality | +0.021 |
Neg logit contributions | |
run | -0.019 |
bug | -0.017 |
out | -0.016 |
udge | -0.016 |
underneath | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | -2.78e-01 |
Pos logit contributions | |
run | +0.067 |
bug | +0.064 |
ours | +0.059 |
imb | +0.059 |
idd | +0.059 |
Neg logit contributions | |
amen | -0.053 |
girls | -0.050 |
Rudolph | -0.046 |
announcements | -0.046 |
hospitality | -0.045 |
Token |
|
Token ID | 198 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
amen | +0.007 |
girls | +0.006 |
Rudolph | +0.006 |
Santa | +0.006 |
announcements | +0.006 |
Neg logit contributions | |
run | -0.011 |
bug | -0.011 |
ours | -0.010 |
idd | -0.010 |
imb | -0.010 |
Token | The |
Token ID | 464 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
amen | +0.008 |
Santa | +0.007 |
girls | +0.007 |
Rudolph | +0.007 |
announcements | +0.007 |
Neg logit contributions | |
run | -0.012 |
bug | -0.012 |
ours | -0.012 |
idd | -0.011 |
imb | -0.011 |
Token | bug |
Token ID | 5434 |
Feature activation | +4.41e+00 |
Pos logit contributions | |
amen | +0.072 |
girls | +0.066 |
Rudolph | +0.063 |
announcements | +0.061 |
hospitality | +0.061 |
Neg logit contributions | |
run | -0.088 |
bug | -0.084 |
ours | -0.078 |
udge | -0.077 |
idd | -0.076 |
Token | was |
Token ID | 373 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
run | +0.155 |
ours | +0.147 |
udge | +0.144 |
imb | +0.144 |
bug | +0.142 |
Neg logit contributions | |
amen | -0.126 |
girls | -0.116 |
Rudolph | -0.110 |
announcements | -0.110 |
Santa | -0.106 |
Token | very |
Token ID | 845 |
Feature activation | +9.14e-01 |
Pos logit contributions | |
run | +0.045 |
bug | +0.043 |
ours | +0.041 |
imb | +0.040 |
udge | +0.040 |
Neg logit contributions | |
amen | -0.041 |
girls | -0.038 |
Rudolph | -0.037 |
announcements | -0.036 |
Santa | -0.035 |
Token | scared |
Token ID | 12008 |
Feature activation | -7.61e-01 |
Pos logit contributions | |
run | +0.029 |
bug | +0.029 |
ours | +0.027 |
imb | +0.026 |
udge | +0.026 |
Neg logit contributions | |
amen | -0.029 |
girls | -0.026 |
announcements | -0.026 |
Rudolph | -0.026 |
Santa | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | +3.55e-01 |
Pos logit contributions | |
amen | +0.026 |
girls | +0.025 |
Rudolph | +0.023 |
announcements | +0.023 |
hospitality | +0.023 |
Neg logit contributions | |
run | -0.025 |
bug | -0.023 |
udge | -0.021 |
out | -0.020 |
elly | -0.020 |
Token | He |
Token ID | 679 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
run | +0.013 |
bug | +0.012 |
ours | +0.012 |
imb | +0.011 |
haired | +0.011 |
Neg logit contributions | |
amen | -0.010 |
girls | -0.009 |
hospitality | -0.009 |
Rudolph | -0.009 |
announcements | -0.008 |
Token | saw |
Token ID | 2497 |
Feature activation | +3.67e+00 |
Pos logit contributions | |
Rudolph | +0.044 |
amen | +0.044 |
girls | +0.041 |
Santa | +0.040 |
announcements | +0.039 |
Neg logit contributions | |
run | -0.055 |
udge | -0.051 |
bug | -0.050 |
ours | -0.049 |
imb | -0.049 |
Token | a |
Token ID | 257 |
Feature activation | -5.87e-02 |
Pos logit contributions | |
run | +0.140 |
bug | +0.134 |
ours | +0.126 |
imb | +0.126 |
udge | +0.126 |
Neg logit contributions | |
amen | -0.097 |
Rudolph | -0.088 |
girls | -0.088 |
Santa | -0.083 |
mercy | -0.082 |
Token | hunter |
Token ID | 19177 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
amen | +0.002 |
girls | +0.002 |
Rudolph | +0.002 |
Santa | +0.002 |
announcements | +0.002 |
Neg logit contributions | |
run | -0.002 |
bug | -0.002 |
ours | -0.002 |
udge | -0.002 |
imb | -0.002 |
Token | with |
Token ID | 351 |
Feature activation | +3.21e+00 |
Pos logit contributions | |
amen | +0.008 |
girls | +0.008 |
Rudolph | +0.008 |
Claus | +0.007 |
Santa | +0.007 |
Neg logit contributions | |
run | -0.009 |
ours | -0.008 |
bug | -0.008 |
haired | -0.008 |
imb | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | -6.50e-02 |
Pos logit contributions | |
bug | +0.115 |
run | +0.114 |
ours | +0.109 |
imb | +0.107 |
udge | +0.105 |
Neg logit contributions | |
amen | -0.088 |
girls | -0.082 |
Rudolph | -0.080 |
announcements | -0.077 |
Santa | -0.076 |
Token | big |
Token ID | 1263 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
amen | +0.002 |
girls | +0.002 |
Rudolph | +0.002 |
announcements | +0.002 |
Santa | +0.002 |
Neg logit contributions | |
run | -0.002 |
bug | -0.002 |
ours | -0.002 |
idd | -0.002 |
imb | -0.002 |
Token | hat |
Token ID | 6877 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
run | +0.068 |
bug | +0.065 |
idd | +0.060 |
ours | +0.059 |
imb | +0.059 |
Neg logit contributions | |
amen | -0.067 |
girls | -0.064 |
Rudolph | -0.058 |
Santa | -0.058 |
announcements | -0.058 |
Token | and |
Token ID | 290 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
run | +0.006 |
bug | +0.006 |
idd | +0.006 |
ours | +0.005 |
haired | +0.005 |
Neg logit contributions | |
amen | -0.007 |
girls | -0.006 |
announcements | -0.006 |
Rudolph | -0.006 |
hospitality | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | -9.56e-02 |
Pos logit contributions | |
run | +0.033 |
bug | +0.032 |
ours | +0.029 |
idd | +0.029 |
imb | +0.029 |
Neg logit contributions | |
amen | -0.035 |
girls | -0.033 |
Rudolph | -0.032 |
announcements | -0.031 |
Santa | -0.030 |
Token | gun |
Token ID | 2485 |
Feature activation | -5.39e-01 |
Pos logit contributions | |
amen | +0.003 |
girls | +0.003 |
Rudolph | +0.003 |
hospitality | +0.003 |
announcements | +0.003 |
Neg logit contributions | |
run | -0.003 |
bug | -0.003 |
idd | -0.003 |
ours | -0.003 |
haired | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +9.37e-01 |
Pos logit contributions | |
amen | +0.014 |
girls | +0.013 |
Rudolph | +0.012 |
announcements | +0.012 |
Santa | +0.012 |
Neg logit contributions | |
run | -0.020 |
bug | -0.020 |
ours | -0.018 |
idd | -0.018 |
udge | -0.018 |
Token | knowing |
Token ID | 6970 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
run | +0.016 |
bug | +0.015 |
ours | +0.014 |
rous | +0.014 |
imb | +0.014 |
Neg logit contributions | |
girls | -0.016 |
amen | -0.016 |
announcements | -0.015 |
Santa | -0.014 |
Rudolph | -0.014 |
Token | he |
Token ID | 339 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
girls | +0.012 |
amen | +0.012 |
announcements | +0.011 |
sets | +0.011 |
Claus | +0.011 |
Neg logit contributions | |
run | -0.010 |
bug | -0.009 |
ours | -0.008 |
out | -0.008 |
here | -0.008 |
Token | was |
Token ID | 373 |
Feature activation | +9.15e-01 |
Pos logit contributions | |
amen | +0.004 |
girls | +0.004 |
Rudolph | +0.004 |
announcements | +0.004 |
Santa | +0.003 |
Neg logit contributions | |
run | -0.004 |
bug | -0.004 |
ours | -0.003 |
idd | -0.003 |
haired | -0.003 |
Token | loved |
Token ID | 6151 |
Feature activation | -4.24e-01 |
Pos logit contributions | |
run | +0.028 |
bug | +0.028 |
idd | +0.024 |
haired | +0.024 |
ours | +0.024 |
Neg logit contributions | |
amen | -0.030 |
girls | -0.030 |
announcements | -0.027 |
Rudolph | -0.026 |
hospitality | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
amen | +0.014 |
girls | +0.013 |
announcements | +0.013 |
Claus | +0.013 |
Santa | +0.012 |
Neg logit contributions | |
run | -0.014 |
bug | -0.012 |
udge | -0.011 |
idd | -0.010 |
so | -0.010 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
amen | +0.009 |
girls | +0.009 |
announcements | +0.009 |
hospitality | +0.008 |
sets | +0.008 |
Neg logit contributions | |
run | -0.010 |
bug | -0.010 |
ours | -0.009 |
udge | -0.009 |
imb | -0.009 |
Token | Once |
Token ID | 7454 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
run | +0.003 |
bug | +0.003 |
idd | +0.003 |
imb | +0.003 |
udge | +0.003 |
Neg logit contributions | |
amen | -0.004 |
girls | -0.004 |
announcements | -0.004 |
hospitality | -0.004 |
mercy | -0.004 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
bug | +0.059 |
run | +0.058 |
idd | +0.055 |
imb | +0.053 |
udge | +0.052 |
Neg logit contributions | |
amen | -0.077 |
Rudolph | -0.070 |
girls | -0.068 |
Santa | -0.065 |
hospitality | -0.064 |
Token | a |
Token ID | 257 |
Feature activation | -6.66e-02 |
Pos logit contributions | |
run | +0.047 |
bug | +0.046 |
haired | +0.040 |
udge | +0.039 |
eh | +0.039 |
Neg logit contributions | |
amen | -0.059 |
girls | -0.058 |
announcements | -0.052 |
sets | -0.052 |
Rudolph | -0.052 |
Token | time |
Token ID | 640 |
Feature activation | +5.96e-01 |
Pos logit contributions | |
amen | +0.002 |
Rudolph | +0.001 |
girls | +0.001 |
hospitality | +0.001 |
Santa | +0.001 |
Neg logit contributions | |
run | -0.003 |
bug | -0.003 |
ours | -0.003 |
idd | -0.003 |
imb | -0.002 |
Token | there |
Token ID | 612 |
Feature activation | +4.02e+00 |
Pos logit contributions | |
run | +0.030 |
bug | +0.029 |
udge | +0.029 |
ished | +0.029 |
attered | +0.029 |
Neg logit contributions | |
amen | -0.007 |
Rudolph | -0.005 |
girls | -0.004 |
Santa | -0.004 |
stories | -0.003 |
Token | it |
Token ID | 340 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
girls | +0.060 |
amen | +0.059 |
announcements | +0.054 |
hospitality | +0.052 |
Santa | +0.052 |
Neg logit contributions | |
run | -0.059 |
bug | -0.058 |
ours | -0.051 |
imb | -0.051 |
idd | -0.050 |
Token | 's |
Token ID | 338 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
run | +0.104 |
bug | +0.103 |
ours | +0.092 |
imb | +0.088 |
haired | +0.088 |
Neg logit contributions | |
amen | -0.116 |
girls | -0.112 |
Rudolph | -0.106 |
Santa | -0.103 |
hospitality | -0.103 |
Token | not |
Token ID | 407 |
Feature activation | -9.94e-01 |
Pos logit contributions | |
run | +0.007 |
bug | +0.007 |
ours | +0.006 |
udge | +0.006 |
idd | +0.006 |
Neg logit contributions | |
amen | -0.006 |
girls | -0.006 |
Rudolph | -0.006 |
announcements | -0.005 |
Santa | -0.005 |
Token | so |
Token ID | 523 |
Feature activation | -7.31e-01 |
Pos logit contributions | |
amen | +0.032 |
girls | +0.031 |
Rudolph | +0.028 |
Santa | +0.028 |
announcements | +0.028 |
Neg logit contributions | |
run | -0.032 |
bug | -0.031 |
ours | -0.028 |
imb | -0.027 |
eh | -0.027 |
Token | wet |
Token ID | 9583 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
amen | +0.024 |
girls | +0.023 |
Rudolph | +0.022 |
announcements | +0.021 |
Santa | +0.021 |
Neg logit contributions | |
run | -0.023 |
bug | -0.022 |
ours | -0.020 |
idd | -0.020 |
udge | -0.020 |
Token | ". |
Token ID | 1911 |
Feature activation | -8.57e-01 |
Pos logit contributions | |
amen | +0.010 |
girls | +0.010 |
Rudolph | +0.009 |
Santa | +0.009 |
announcements | +0.009 |
Neg logit contributions | |
run | -0.008 |
bug | -0.007 |
ours | -0.007 |
idd | -0.006 |
imb | -0.006 |
Token | The |
Token ID | 383 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
amen | +0.025 |
girls | +0.024 |
Rudolph | +0.022 |
announcements | +0.022 |
hospitality | +0.021 |
Neg logit contributions | |
run | -0.030 |
bug | -0.029 |
ours | -0.026 |
idd | -0.026 |
imb | -0.026 |
Token | duck |
Token ID | 22045 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
amen | +0.067 |
girls | +0.065 |
Rudolph | +0.059 |
announcements | +0.059 |
sets | +0.058 |
Neg logit contributions | |
bug | -0.071 |
run | -0.071 |
idd | -0.063 |
ours | -0.063 |
haired | -0.062 |
Token | smiled |
Token ID | 13541 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
run | +0.041 |
bug | +0.040 |
ours | +0.036 |
idd | +0.035 |
haired | +0.035 |
Neg logit contributions | |
amen | -0.039 |
girls | -0.037 |
Rudolph | -0.035 |
announcements | -0.034 |
Santa | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | +8.07e-01 |
Pos logit contributions | |
girls | +0.103 |
amen | +0.098 |
announcements | +0.093 |
hospitality | +0.090 |
Rudolph | +0.087 |
Neg logit contributions | |
run | -0.073 |
bug | -0.069 |
it | -0.059 |
haired | -0.058 |
out | -0.058 |
Token | thanked |
Token ID | 26280 |
Feature activation | -5.86e+00 |
Pos logit contributions | |
run | +0.027 |
bug | +0.026 |
ours | +0.023 |
idd | +0.023 |
haired | +0.023 |
Neg logit contributions | |
amen | -0.024 |
girls | -0.024 |
announcements | -0.022 |
Santa | -0.022 |
hospitality | -0.021 |
Token | ice |
Token ID | 4771 |
Feature activation | +8.76e-01 |
Pos logit contributions | |
run | +0.070 |
bug | +0.066 |
idd | +0.064 |
udge | +0.063 |
eh | +0.062 |
Neg logit contributions | |
amen | -0.029 |
thank | -0.023 |
hospitality | -0.023 |
mercy | -0.023 |
Rudolph | -0.022 |
Token | cream |
Token ID | 8566 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
run | +0.034 |
bug | +0.032 |
imb | +0.030 |
ours | +0.030 |
udge | +0.030 |
Neg logit contributions | |
amen | -0.022 |
girls | -0.020 |
Rudolph | -0.019 |
hospitality | -0.018 |
Claus | -0.018 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
amen | +0.085 |
girls | +0.079 |
Rudolph | +0.075 |
announcements | +0.072 |
Santa | +0.072 |
Neg logit contributions | |
run | -0.094 |
bug | -0.090 |
ours | -0.083 |
idd | -0.081 |
imb | -0.081 |
Token | she |
Token ID | 673 |
Feature activation | +8.10e-01 |
Pos logit contributions | |
amen | +0.037 |
girls | +0.034 |
Rudolph | +0.033 |
announcements | +0.031 |
Santa | +0.031 |
Neg logit contributions | |
run | -0.051 |
bug | -0.050 |
ours | -0.046 |
imb | -0.046 |
udge | -0.046 |
Token | said |
Token ID | 531 |
Feature activation | -6.84e-01 |
Pos logit contributions | |
run | +0.029 |
bug | +0.028 |
haired | +0.025 |
ours | +0.025 |
idd | +0.025 |
Neg logit contributions | |
amen | -0.023 |
girls | -0.022 |
Rudolph | -0.020 |
hospitality | -0.020 |
Santa | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +4.28e-01 |
Pos logit contributions | |
girls | +0.019 |
amen | +0.019 |
Rudolph | +0.017 |
announcements | +0.017 |
Santa | +0.016 |
Neg logit contributions | |
run | -0.026 |
bug | -0.025 |
haired | -0.022 |
udge | -0.022 |
ours | -0.022 |
Token | " |
Token ID | 366 |
Feature activation | +3.19e+00 |
Pos logit contributions | |
run | +0.014 |
bug | +0.013 |
ours | +0.012 |
idd | +0.012 |
imb | +0.012 |
Neg logit contributions | |
amen | -0.014 |
girls | -0.013 |
Rudolph | -0.012 |
announcements | -0.012 |
Santa | -0.012 |
Token | I |
Token ID | 40 |
Feature activation | +4.07e+00 |
Pos logit contributions | |
run | +0.121 |
ours | +0.117 |
bug | +0.110 |
ool | +0.106 |
idd | +0.106 |
Neg logit contributions | |
amen | -0.083 |
Santa | -0.083 |
announcements | -0.077 |
girls | -0.073 |
Rudolph | -0.071 |
Token | know |
Token ID | 760 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
run | +0.148 |
bug | +0.136 |
ours | +0.135 |
haired | +0.134 |
ang | +0.133 |
Neg logit contributions | |
amen | -0.107 |
Claus | -0.096 |
girls | -0.096 |
thank | -0.094 |
Rudolph | -0.093 |
Token | , |
Token ID | 11 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
amen | +0.011 |
girls | +0.011 |
Rudolph | +0.010 |
announcements | +0.010 |
Santa | +0.010 |
Neg logit contributions | |
run | -0.011 |
bug | -0.011 |
ours | -0.009 |
idd | -0.009 |
imb | -0.009 |
Token | sweet |
Token ID | 6029 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
amen | +0.011 |
girls | +0.011 |
Rudolph | +0.010 |
announcements | +0.010 |
Santa | +0.010 |
Neg logit contributions | |
run | -0.009 |
bug | -0.009 |
ours | -0.008 |
idd | -0.008 |
udge | -0.008 |
Token | match |
Token ID | 2872 |
Feature activation | +3.93e-01 |
Pos logit contributions | |
girls | +0.245 |
amen | +0.241 |
announcements | +0.224 |
sets | +0.217 |
Rudolph | +0.217 |
Neg logit contributions | |
run | -0.233 |
bug | -0.232 |
nail | -0.201 |
here | -0.200 |
ab | -0.197 |
Token | her |
Token ID | 607 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
run | +0.013 |
bug | +0.013 |
ours | +0.012 |
idd | +0.011 |
imb | +0.011 |
Neg logit contributions | |
amen | -0.012 |
girls | -0.011 |
Rudolph | -0.011 |
announcements | -0.010 |
Santa | -0.010 |
Token | every |
Token ID | 790 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
run | +0.037 |
bug | +0.037 |
ours | +0.034 |
idd | +0.033 |
udge | +0.033 |
Neg logit contributions | |
amen | -0.033 |
girls | -0.031 |
Rudolph | -0.030 |
announcements | -0.029 |
hospitality | -0.028 |
Token | step |
Token ID | 2239 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
amen | +0.021 |
hospitality | +0.020 |
service | +0.019 |
prayer | +0.018 |
mercy | +0.017 |
Neg logit contributions | |
ours | -0.042 |
run | -0.041 |
idd | -0.041 |
udge | -0.040 |
okes | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
run | +0.053 |
bug | +0.052 |
ours | +0.047 |
idd | +0.046 |
imb | +0.046 |
Neg logit contributions | |
amen | -0.054 |
girls | -0.051 |
Rudolph | -0.048 |
announcements | -0.047 |
Santa | -0.046 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
run | +0.007 |
bug | +0.007 |
ours | +0.006 |
imb | +0.006 |
udge | +0.006 |
Neg logit contributions | |
amen | -0.005 |
girls | -0.005 |
Rudolph | -0.005 |
announcements | -0.005 |
Santa | -0.005 |
Token | One |
Token ID | 3198 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
run | +0.010 |
bug | +0.009 |
idd | +0.008 |
imb | +0.008 |
ours | +0.008 |
Neg logit contributions | |
amen | -0.011 |
girls | -0.010 |
announcements | -0.010 |
hospitality | -0.010 |
mercy | -0.010 |
Token | mighty |
Token ID | 18680 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
run | +0.071 |
bug | +0.067 |
idd | +0.067 |
ours | +0.066 |
udge | +0.065 |
Neg logit contributions | |
amen | -0.052 |
hospitality | -0.046 |
stories | -0.045 |
Rudolph | -0.045 |
girls | -0.044 |
Token | day |
Token ID | 1110 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
run | +0.055 |
bug | +0.054 |
ours | +0.049 |
idd | +0.049 |
imb | +0.048 |
Neg logit contributions | |
amen | -0.048 |
girls | -0.045 |
Rudolph | -0.043 |
announcements | -0.041 |
Santa | -0.041 |
Token | , |
Token ID | 11 |
Feature activation | -6.22e-03 |
Pos logit contributions | |
run | +0.058 |
bug | +0.056 |
ours | +0.053 |
udge | +0.052 |
imb | +0.052 |
Neg logit contributions | |
amen | -0.030 |
girls | -0.027 |
Rudolph | -0.026 |
Santa | -0.024 |
announcements | -0.024 |
Token | a |
Token ID | 257 |
Feature activation | -5.54e-02 |
Pos logit contributions | |
Rudolph | +0.000 |
amen | +0.000 |
Santa | +0.000 |
Santa | +0.000 |
Arthur | +0.000 |
Neg logit contributions | |
run | -0.000 |
ours | -0.000 |
bug | -0.000 |
ash | -0.000 |
ished | -0.000 |
Token | ever |
Token ID | 1683 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
amen | +0.073 |
girls | +0.068 |
Rudolph | +0.065 |
announcements | +0.064 |
Santa | +0.064 |
Neg logit contributions | |
run | -0.062 |
bug | -0.060 |
idd | -0.053 |
ours | -0.053 |
haired | -0.052 |
Token | ! |
Token ID | 0 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
bug | +0.078 |
run | +0.076 |
out | +0.065 |
elly | +0.064 |
ab | +0.064 |
Neg logit contributions | |
amen | -0.114 |
girls | -0.108 |
Rudolph | -0.104 |
announcements | -0.102 |
Santa | -0.101 |
Token | Thank |
Token ID | 6952 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
run | +0.054 |
bug | +0.052 |
ours | +0.046 |
imb | +0.045 |
idd | +0.045 |
Neg logit contributions | |
amen | -0.079 |
girls | -0.075 |
Rudolph | -0.072 |
announcements | -0.072 |
Santa | -0.070 |
Token | you |
Token ID | 345 |
Feature activation | +5.94e-01 |
Pos logit contributions | |
amen | +0.073 |
announcements | +0.069 |
Claus | +0.068 |
girls | +0.067 |
Rudolph | +0.067 |
Neg logit contributions | |
bug | -0.045 |
run | -0.041 |
imb | -0.039 |
ang | -0.039 |
idd | -0.039 |
Token | for |
Token ID | 329 |
Feature activation | -4.70e-01 |
Pos logit contributions | |
run | +0.012 |
bug | +0.012 |
ash | +0.009 |
ours | +0.009 |
idd | +0.009 |
Neg logit contributions | |
amen | -0.025 |
girls | -0.025 |
announcements | -0.025 |
Santa | -0.023 |
Rudolph | -0.023 |
Token | permitting |
Token ID | 29256 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
amen | +0.015 |
girls | +0.014 |
announcements | +0.013 |
Claus | +0.013 |
Rudolph | +0.013 |
Neg logit contributions | |
run | -0.015 |
bug | -0.015 |
idd | -0.013 |
eh | -0.013 |
imb | -0.013 |
Token | me |
Token ID | 502 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
amen | +0.085 |
girls | +0.080 |
sets | +0.076 |
Claus | +0.075 |
announcements | +0.075 |
Neg logit contributions | |
run | -0.042 |
bug | -0.041 |
out | -0.035 |
idd | -0.034 |
ours | -0.033 |
Token | to |
Token ID | 284 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
run | +0.052 |
bug | +0.050 |
ours | +0.044 |
imb | +0.044 |
idd | +0.044 |
Neg logit contributions | |
amen | -0.060 |
girls | -0.057 |
Rudolph | -0.054 |
announcements | -0.053 |
Santa | -0.053 |
Token | go |
Token ID | 467 |
Feature activation | -4.61e-01 |
Pos logit contributions | |
amen | +0.049 |
Claus | +0.044 |
girls | +0.044 |
announcements | +0.044 |
Rudolph | +0.044 |
Neg logit contributions | |
bug | -0.043 |
run | -0.042 |
imb | -0.038 |
here | -0.035 |
block | -0.034 |
Token | with |
Token ID | 351 |
Feature activation | +2.91e+00 |
Pos logit contributions | |
amen | +0.018 |
girls | +0.017 |
Rudolph | +0.016 |
announcements | +0.016 |
Santa | +0.016 |
Neg logit contributions | |
run | -0.012 |
bug | -0.012 |
ours | -0.010 |
idd | -0.010 |
imb | -0.010 |
Token | you |
Token ID | 345 |
Feature activation | +6.65e-01 |
Pos logit contributions | |
run | +0.094 |
bug | +0.090 |
ours | +0.083 |
udge | +0.081 |
imb | +0.081 |
Neg logit contributions | |
amen | -0.095 |
girls | -0.089 |
Rudolph | -0.086 |
announcements | -0.082 |
Santa | -0.082 |
Token | excited |
Token ID | 6568 |
Feature activation | +9.97e-01 |
Pos logit contributions | |
run | +0.045 |
bug | +0.042 |
ours | +0.042 |
imb | +0.041 |
udge | +0.041 |
Neg logit contributions | |
amen | -0.034 |
girls | -0.030 |
Rudolph | -0.029 |
announcements | -0.028 |
Santa | -0.028 |
Token | to |
Token ID | 284 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
run | +0.025 |
bug | +0.025 |
out | +0.021 |
earing | +0.020 |
ours | +0.020 |
Neg logit contributions | |
girls | -0.039 |
amen | -0.039 |
Rudolph | -0.036 |
Santa | -0.036 |
announcements | -0.035 |
Token | try |
Token ID | 1949 |
Feature activation | -8.29e-01 |
Pos logit contributions | |
amen | +0.040 |
girls | +0.038 |
Rudolph | +0.036 |
announcements | +0.035 |
Santa | +0.034 |
Neg logit contributions | |
run | -0.047 |
bug | -0.046 |
ours | -0.042 |
imb | -0.042 |
udge | -0.041 |
Token | it |
Token ID | 340 |
Feature activation | +3.17e+00 |
Pos logit contributions | |
amen | +0.030 |
girls | +0.029 |
Rudolph | +0.027 |
announcements | +0.027 |
Santa | +0.027 |
Neg logit contributions | |
run | -0.023 |
bug | -0.022 |
ours | -0.020 |
idd | -0.020 |
imb | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
run | +0.113 |
bug | +0.111 |
ours | +0.104 |
idd | +0.103 |
haired | +0.101 |
Neg logit contributions | |
amen | -0.088 |
girls | -0.082 |
Rudolph | -0.079 |
sets | -0.075 |
Santa | -0.075 |
Token | |
Token ID | 220 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
amen | +0.001 |
girls | +0.001 |
Rudolph | +0.001 |
Santa | +0.001 |
sets | +0.001 |
Neg logit contributions | |
run | -0.001 |
bug | -0.001 |
idd | -0.001 |
ours | -0.001 |
eh | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -8.70e-01 |
Pos logit contributions | |
amen | +0.048 |
girls | +0.045 |
Rudolph | +0.042 |
Santa | +0.040 |
announcements | +0.039 |
Neg logit contributions | |
run | -0.076 |
bug | -0.076 |
idd | -0.070 |
ours | -0.070 |
haired | -0.069 |
Token |
|
Token ID | 198 |
Feature activation | -8.71e-01 |
Pos logit contributions | |
amen | +0.022 |
girls | +0.021 |
Rudolph | +0.020 |
Santa | +0.019 |
announcements | +0.019 |
Neg logit contributions | |
run | -0.033 |
bug | -0.033 |
ours | -0.031 |
idd | -0.030 |
imb | -0.030 |
Token | Her |
Token ID | 9360 |
Feature activation | +6.29e-01 |
Pos logit contributions | |
amen | +0.021 |
Santa | +0.019 |
girls | +0.019 |
Rudolph | +0.017 |
announcements | +0.017 |
Neg logit contributions | |
run | -0.036 |
bug | -0.034 |
ours | -0.033 |
idd | -0.032 |
imb | -0.031 |
Token | parents |
Token ID | 3397 |
Feature activation | -4.67e+00 |
Pos logit contributions | |
run | +0.025 |
bug | +0.024 |
ours | +0.024 |
nail | +0.023 |
udge | +0.023 |
Neg logit contributions | |
girls | -0.014 |
amen | -0.013 |
announcements | -0.012 |
Rudolph | -0.012 |
sets | -0.011 |
Token | introduced |
Token ID | 5495 |
Feature activation | +8.88e-02 |
Pos logit contributions | |
amen | +0.138 |
girls | +0.131 |
Rudolph | +0.123 |
announcements | +0.119 |
Santa | +0.118 |
Neg logit contributions | |
run | -0.161 |
bug | -0.159 |
ours | -0.144 |
idd | -0.143 |
imb | -0.141 |
Token | They |
Token ID | 1119 |
Feature activation | -5.98e+00 |
Pos logit contributions | |
amen | +0.001 |
girls | +0.001 |
hospitality | +0.001 |
Rudolph | +0.001 |
announcements | +0.001 |
Neg logit contributions | |
run | -0.001 |
bug | -0.001 |
ours | -0.001 |
udge | -0.001 |
imb | -0.001 |
Token | wanted |
Token ID | 2227 |
Feature activation | +3.72e+00 |
Pos logit contributions | |
sets | +0.219 |
amen | +0.214 |
girls | +0.211 |
hospitality | +0.203 |
MacDonald | +0.193 |
Neg logit contributions | |
bug | -0.169 |
run | -0.146 |
out | -0.132 |
ours | -0.131 |
nail | -0.129 |
Token | to |
Token ID | 284 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
run | +0.094 |
bug | +0.093 |
ours | +0.083 |
out | +0.077 |
udge | +0.075 |
Neg logit contributions | |
amen | -0.142 |
girls | -0.135 |
announcements | -0.128 |
sets | -0.126 |
Claus | -0.124 |
Token | go |
Token ID | 467 |
Feature activation | -3.97e-01 |
Pos logit contributions | |
amen | +0.057 |
sets | +0.057 |
announcements | +0.055 |
girls | +0.054 |
MacDonald | +0.053 |
Neg logit contributions | |
bug | -0.038 |
nail | -0.034 |
block | -0.033 |
run | -0.032 |
imb | -0.030 |
Token | to |
Token ID | 284 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
amen | +0.017 |
girls | +0.016 |
announcements | +0.015 |
Rudolph | +0.015 |
sets | +0.014 |
Neg logit contributions | |
run | -0.009 |
bug | -0.009 |
ours | -0.008 |
idd | -0.007 |
imb | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | -2.54e+00 |
Pos logit contributions | |
amen | +0.050 |
girls | +0.048 |
sets | +0.047 |
announcements | +0.045 |
Santa | +0.044 |
Neg logit contributions | |
run | -0.043 |
bug | -0.042 |
ours | -0.039 |
rous | -0.038 |
udge | -0.037 |
Token | bathroom |
Token ID | 12436 |
Feature activation | -3.09e+00 |
Pos logit contributions | |
amen | +0.075 |
girls | +0.070 |
Rudolph | +0.066 |
announcements | +0.064 |
Santa | +0.063 |
Neg logit contributions | |
run | -0.087 |
bug | -0.086 |
ours | -0.079 |
idd | -0.077 |
imb | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | -6.58e-02 |
Pos logit contributions | |
amen | +0.112 |
girls | +0.106 |
announcements | +0.100 |
Santa | +0.098 |
sets | +0.097 |
Neg logit contributions | |
bug | -0.083 |
run | -0.082 |
out | -0.080 |
ours | -0.071 |
here | -0.068 |
Token |
|
Token ID | 198 |
Feature activation | -7.81e-01 |
Pos logit contributions | |
amen | +0.002 |
girls | +0.002 |
Rudolph | +0.002 |
announcements | +0.001 |
Santa | +0.001 |
Neg logit contributions | |
run | -0.003 |
bug | -0.002 |
ours | -0.002 |
idd | -0.002 |
imb | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -7.94e-01 |
Pos logit contributions | |
amen | +0.019 |
girls | +0.018 |
Rudolph | +0.017 |
Santa | +0.017 |
announcements | +0.016 |
Neg logit contributions | |
run | -0.031 |
bug | -0.030 |
ours | -0.028 |
idd | -0.028 |
imb | -0.028 |
Token | They |
Token ID | 2990 |
Feature activation | -6.29e+00 |
Pos logit contributions | |
amen | +0.018 |
girls | +0.017 |
Santa | +0.016 |
Rudolph | +0.016 |
announcements | +0.015 |
Neg logit contributions | |
run | -0.033 |
bug | -0.032 |
ours | -0.030 |
imb | -0.030 |
idd | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
amen | +0.027 |
girls | +0.025 |
Rudolph | +0.023 |
hospitality | +0.023 |
Claus | +0.023 |
Neg logit contributions | |
run | -0.033 |
bug | -0.030 |
out | -0.028 |
ours | -0.027 |
udge | -0.027 |
Token | They |
Token ID | 1119 |
Feature activation | -5.95e+00 |
Pos logit contributions | |
amen | +0.005 |
girls | +0.004 |
hospitality | +0.004 |
Rudolph | +0.004 |
sets | +0.004 |
Neg logit contributions | |
run | -0.005 |
bug | -0.005 |
ours | -0.005 |
haired | -0.005 |
udge | -0.005 |
Token | said |
Token ID | 531 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
amen | +0.181 |
girls | +0.171 |
Rudolph | +0.161 |
Santa | +0.156 |
hospitality | +0.155 |
Neg logit contributions | |
run | -0.201 |
bug | -0.198 |
ours | -0.178 |
haired | -0.175 |
idd | -0.174 |
Token | it |
Token ID | 340 |
Feature activation | +2.97e+00 |
Pos logit contributions | |
girls | +0.042 |
amen | +0.041 |
Claus | +0.038 |
hospitality | +0.038 |
sets | +0.037 |
Neg logit contributions | |
run | -0.031 |
bug | -0.030 |
ours | -0.028 |
haired | -0.025 |
idd | -0.025 |
Token | gave |
Token ID | 2921 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
run | +0.090 |
bug | +0.087 |
ours | +0.080 |
imb | +0.079 |
udge | +0.079 |
Neg logit contributions | |
amen | -0.100 |
girls | -0.095 |
Rudolph | -0.090 |
announcements | -0.089 |
Santa | -0.087 |
Token | them |
Token ID | 606 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
amen | +0.004 |
girls | +0.004 |
Rudolph | +0.004 |
announcements | +0.004 |
Santa | +0.004 |
Neg logit contributions | |
run | -0.005 |
bug | -0.005 |
ours | -0.004 |
idd | -0.004 |
udge | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | -6.29e-01 |
Pos logit contributions | |
amen | +0.072 |
girls | +0.068 |
Rudolph | +0.064 |
announcements | +0.062 |
Santa | +0.061 |
Neg logit contributions | |
run | -0.075 |
bug | -0.073 |
ours | -0.065 |
haired | -0.064 |
imb | -0.063 |
Token | headache |
Token ID | 24902 |
Feature activation | -6.03e+00 |
Pos logit contributions | |
amen | +0.018 |
girls | +0.016 |
Rudolph | +0.016 |
announcements | +0.015 |
Santa | +0.015 |
Neg logit contributions | |
run | -0.023 |
bug | -0.022 |
ours | -0.021 |
idd | -0.020 |
imb | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | +2.76e-01 |
Pos logit contributions | |
amen | +0.227 |
girls | +0.220 |
Rudolph | +0.204 |
announcements | +0.204 |
Santa | +0.200 |
Neg logit contributions | |
run | -0.165 |
bug | -0.147 |
out | -0.133 |
elly | -0.128 |
haired | -0.128 |
Token | made |
Token ID | 925 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
run | +0.009 |
bug | +0.009 |
ours | +0.008 |
haired | +0.008 |
idd | +0.008 |
Neg logit contributions | |
amen | -0.008 |
girls | -0.008 |
Rudolph | -0.008 |
Santa | -0.007 |
announcements | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
amen | +0.091 |
girls | +0.085 |
Rudolph | +0.082 |
announcements | +0.079 |
Santa | +0.078 |
Neg logit contributions | |
run | -0.101 |
bug | -0.099 |
ours | -0.090 |
idd | -0.089 |
imb | -0.089 |
Token | ," |
Token ID | 553 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
bug | +0.008 |
run | +0.006 |
here | +0.006 |
ang | +0.006 |
eh | +0.006 |
Neg logit contributions | |
amen | -0.011 |
announcements | -0.011 |
girls | -0.010 |
Rudolph | -0.010 |
hospitality | -0.010 |
Token | Ben |
Token ID | 3932 |
Feature activation | -9.62e-01 |
Pos logit contributions | |
girls | +0.074 |
amen | +0.070 |
sets | +0.069 |
hospitality | +0.068 |
Claus | +0.067 |
Neg logit contributions | |
bug | -0.066 |
ours | -0.057 |
run | -0.055 |
haired | -0.051 |
rous | -0.051 |
Token | says |
Token ID | 1139 |
Feature activation | -8.86e-01 |
Pos logit contributions | |
amen | +0.032 |
girls | +0.031 |
hospitality | +0.029 |
Santa | +0.029 |
Rudolph | +0.029 |
Neg logit contributions | |
bug | -0.029 |
run | -0.026 |
ours | -0.026 |
haired | -0.024 |
idd | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +5.36e-01 |
Pos logit contributions | |
girls | +0.029 |
amen | +0.028 |
sets | +0.027 |
Rudolph | +0.027 |
Santa | +0.027 |
Neg logit contributions | |
run | -0.029 |
bug | -0.027 |
ours | -0.024 |
out | -0.023 |
udge | -0.022 |
Token | " |
Token ID | 366 |
Feature activation | +2.99e+00 |
Pos logit contributions | |
run | +0.017 |
bug | +0.016 |
ours | +0.015 |
haired | +0.015 |
idd | +0.015 |
Neg logit contributions | |
amen | -0.017 |
girls | -0.017 |
Rudolph | -0.015 |
announcements | -0.015 |
hospitality | -0.015 |
Token | The |
Token ID | 464 |
Feature activation | -2.72e+00 |
Pos logit contributions | |
run | +0.109 |
bug | +0.102 |
ours | +0.098 |
idd | +0.093 |
imb | +0.093 |
Neg logit contributions | |
amen | -0.084 |
girls | -0.078 |
Santa | -0.077 |
Rudolph | -0.076 |
announcements | -0.070 |
Token | house |
Token ID | 2156 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
amen | +0.088 |
girls | +0.084 |
Rudolph | +0.080 |
hospitality | +0.076 |
announcements | +0.075 |
Neg logit contributions | |
run | -0.085 |
bug | -0.084 |
ours | -0.077 |
idd | -0.076 |
imb | -0.075 |
Token | is |
Token ID | 318 |
Feature activation | +3.25e-01 |
Pos logit contributions | |
amen | +0.010 |
girls | +0.009 |
Rudolph | +0.009 |
announcements | +0.009 |
Santa | +0.009 |
Neg logit contributions | |
bug | -0.006 |
run | -0.006 |
ours | -0.005 |
idd | -0.005 |
haired | -0.005 |
Token | very |
Token ID | 845 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
run | +0.011 |
bug | +0.010 |
udge | +0.010 |
idd | +0.009 |
ours | +0.009 |
Neg logit contributions | |
amen | -0.010 |
girls | -0.009 |
Rudolph | -0.009 |
Santa | -0.009 |
announcements | -0.009 |
Token | pretty |
Token ID | 2495 |
Feature activation | +3.51e+00 |
Pos logit contributions | |
run | +0.036 |
bug | +0.036 |
udge | +0.032 |
imb | +0.032 |
idd | +0.032 |
Neg logit contributions | |
amen | -0.031 |
Santa | -0.027 |
Rudolph | -0.027 |
announcements | -0.026 |
girls | -0.026 |
Token | now |
Token ID | 783 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
bug | +0.109 |
run | +0.102 |
elly | +0.097 |
out | +0.096 |
ours | +0.096 |
Neg logit contributions | |
amen | -0.114 |
girls | -0.110 |
announcements | -0.105 |
hospitality | -0.102 |
mercy | -0.102 |
Token | will |
Token ID | 481 |
Feature activation | -7.36e+00 |
Pos logit contributions | |
bug | +0.035 |
run | +0.035 |
ours | +0.031 |
idd | +0.031 |
imb | +0.031 |
Neg logit contributions | |
amen | -0.035 |
girls | -0.033 |
Rudolph | -0.031 |
announcements | -0.030 |
Santa | -0.030 |
Token | have |
Token ID | 423 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
girls | +0.253 |
amen | +0.240 |
rob | +0.227 |
Rudolph | +0.227 |
mercy | +0.227 |
Neg logit contributions | |
bug | -0.197 |
here | -0.182 |
imb | -0.179 |
run | -0.177 |
block | -0.167 |
Token | cake |
Token ID | 12187 |
Feature activation | -6.86e-01 |
Pos logit contributions | |
run | +0.083 |
bug | +0.081 |
ours | +0.073 |
idd | +0.070 |
rous | +0.070 |
Neg logit contributions | |
amen | -0.096 |
girls | -0.094 |
announcements | -0.085 |
Rudolph | -0.085 |
hospitality | -0.083 |
Token | and |
Token ID | 290 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
amen | +0.026 |
announcements | +0.024 |
girls | +0.024 |
Rudolph | +0.024 |
mercy | +0.024 |
Neg logit contributions | |
bug | -0.017 |
run | -0.016 |
ours | -0.014 |
out | -0.014 |
imb | -0.014 |
Token | sing |
Token ID | 1702 |
Feature activation | -4.26e+00 |
Pos logit contributions | |
bug | +0.036 |
run | +0.035 |
ours | +0.032 |
imb | +0.031 |
idd | +0.031 |
Neg logit contributions | |
amen | -0.032 |
girls | -0.032 |
Rudolph | -0.030 |
announcements | -0.029 |
Santa | -0.029 |
Token | songs |
Token ID | 7259 |
Feature activation | -8.28e+00 |
Pos logit contributions | |
amen | +0.133 |
girls | +0.131 |
mercy | +0.122 |
Rudolph | +0.119 |
Claus | +0.117 |
Neg logit contributions | |
run | -0.129 |
bug | -0.126 |
ab | -0.112 |
haired | -0.112 |
out | -0.111 |
Token | ." |
Token ID | 526 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
amen | +0.300 |
girls | +0.294 |
mercy | +0.285 |
announcements | +0.281 |
Rudolph | +0.273 |
Neg logit contributions | |
bug | -0.205 |
run | -0.204 |
ab | -0.182 |
out | -0.178 |
here | -0.175 |
Token |
|
Token ID | 198 |
Feature activation | -7.07e-02 |
Pos logit contributions | |
amen | +0.050 |
girls | +0.047 |
Rudolph | +0.044 |
announcements | +0.043 |
Santa | +0.043 |
Neg logit contributions | |
run | -0.062 |
bug | -0.060 |
ours | -0.056 |
haired | -0.054 |
imb | -0.054 |
Token |
|
Token ID | 198 |
Feature activation | -8.46e-02 |
Pos logit contributions | |
amen | +0.002 |
girls | +0.002 |
Rudolph | +0.002 |
Santa | +0.001 |
announcements | +0.001 |
Neg logit contributions | |
run | -0.003 |
bug | -0.003 |
ours | -0.003 |
idd | -0.003 |
imb | -0.003 |
Token | L |
Token ID | 43 |
Feature activation | +2.97e+00 |
Pos logit contributions | |
amen | +0.003 |
Santa | +0.003 |
girls | +0.003 |
Rudolph | +0.003 |
announcements | +0.003 |
Neg logit contributions | |
run | -0.003 |
bug | -0.002 |
idd | -0.002 |
ours | -0.002 |
imb | -0.002 |
Token | ily |
Token ID | 813 |
Feature activation | +3.89e+00 |
Pos logit contributions | |
run | +0.106 |
rous | +0.095 |
ang | +0.095 |
idd | +0.093 |
bug | +0.092 |
Neg logit contributions | |
amen | -0.076 |
stories | -0.064 |
parties | -0.064 |
hospitality | -0.061 |
Rudolph | -0.059 |
Token | every |
Token ID | 790 |
Feature activation | -7.19e-01 |
Pos logit contributions | |
run | +0.117 |
idd | +0.113 |
bug | +0.112 |
ours | +0.107 |
udge | +0.106 |
Neg logit contributions | |
amen | -0.075 |
girls | -0.067 |
stories | -0.066 |
Rudolph | -0.066 |
hospitality | -0.063 |
Token | day |
Token ID | 1110 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
amen | +0.013 |
hospitality | +0.012 |
prayer | +0.011 |
service | +0.011 |
stories | +0.011 |
Neg logit contributions | |
run | -0.033 |
idd | -0.033 |
ours | -0.032 |
okes | -0.032 |
imb | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
run | +0.040 |
bug | +0.039 |
udge | +0.034 |
ours | +0.034 |
haired | +0.034 |
Neg logit contributions | |
amen | -0.039 |
girls | -0.037 |
announcements | -0.034 |
Rudolph | -0.034 |
mercy | -0.033 |
Token | sang |
Token ID | 25889 |
Feature activation | -3.69e+00 |
Pos logit contributions | |
run | +0.033 |
bug | +0.032 |
ours | +0.029 |
idd | +0.028 |
imb | +0.028 |
Neg logit contributions | |
amen | -0.033 |
girls | -0.031 |
Rudolph | -0.030 |
announcements | -0.029 |
Santa | -0.029 |
Token | it |
Token ID | 340 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
amen | +0.134 |
girls | +0.133 |
Claus | +0.124 |
announcements | +0.119 |
Rudolph | +0.118 |
Neg logit contributions | |
run | -0.103 |
bug | -0.094 |
haired | -0.085 |
ab | -0.083 |
udge | -0.083 |
Token | songs |
Token ID | 7259 |
Feature activation | -8.56e+00 |
Pos logit contributions | |
run | +0.098 |
bug | +0.095 |
ours | +0.085 |
haired | +0.084 |
udge | +0.083 |
Neg logit contributions | |
amen | -0.097 |
girls | -0.093 |
Rudolph | -0.086 |
announcements | -0.084 |
Santa | -0.084 |
Token | . |
Token ID | 13 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
amen | +0.302 |
girls | +0.290 |
announcements | +0.273 |
Santa | +0.268 |
Rudolph | +0.263 |
Neg logit contributions | |
run | -0.256 |
bug | -0.249 |
haired | -0.216 |
out | -0.212 |
it | -0.209 |
Token |
|
Token ID | 198 |
Feature activation | -4.57e-01 |
Pos logit contributions | |
run | +0.008 |
bug | +0.008 |
ours | +0.007 |
udge | +0.007 |
idd | +0.007 |
Neg logit contributions | |
amen | -0.006 |
girls | -0.005 |
Rudolph | -0.005 |
Santa | -0.005 |
announcements | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -4.72e-01 |
Pos logit contributions | |
amen | +0.009 |
Santa | +0.008 |
Rudolph | +0.008 |
girls | +0.008 |
announcements | +0.007 |
Neg logit contributions | |
bug | -0.020 |
run | -0.020 |
ours | -0.019 |
idd | -0.018 |
ished | -0.018 |
Token | The |
Token ID | 464 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
amen | +0.011 |
Santa | +0.011 |
Rudolph | +0.009 |
girls | +0.009 |
announcements | +0.009 |
Neg logit contributions | |
run | -0.020 |
ours | -0.019 |
bug | -0.018 |
ished | -0.018 |
okes | -0.017 |
Token | sun |
Token ID | 4252 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
amen | +0.066 |
girls | +0.060 |
Rudolph | +0.056 |
announcements | +0.054 |
Santa | +0.054 |
Neg logit contributions | |
run | -0.113 |
bug | -0.108 |
ours | -0.102 |
udge | -0.100 |
idd | -0.100 |
Token | park |
Token ID | 3952 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
amen | +0.073 |
girls | +0.067 |
Rudolph | +0.067 |
Santa | +0.067 |
stories | +0.063 |
Neg logit contributions | |
run | -0.108 |
bug | -0.099 |
ours | -0.090 |
udge | -0.090 |
imb | -0.089 |
Token | . |
Token ID | 13 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
amen | +0.032 |
Rudolph | +0.028 |
girls | +0.028 |
Santa | +0.027 |
mercy | +0.027 |
Neg logit contributions | |
bug | -0.038 |
run | -0.038 |
idd | -0.033 |
out | -0.033 |
haired | -0.032 |
Token | They |
Token ID | 1119 |
Feature activation | -6.13e+00 |
Pos logit contributions | |
amen | +0.009 |
girls | +0.008 |
mercy | +0.008 |
hospitality | +0.008 |
sets | +0.007 |
Neg logit contributions | |
run | -0.010 |
bug | -0.010 |
ours | -0.009 |
haired | -0.009 |
udge | -0.009 |
Token | had |
Token ID | 550 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
amen | +0.218 |
sets | +0.201 |
girls | +0.195 |
Him | +0.193 |
mercy | +0.190 |
Neg logit contributions | |
bug | -0.186 |
run | -0.149 |
haired | -0.144 |
idd | -0.141 |
touched | -0.141 |
Token | open |
Token ID | 1280 |
Feature activation | +2.88e+00 |
Pos logit contributions | |
bug | +0.040 |
found | +0.035 |
run | +0.033 |
touched | +0.032 |
rous | +0.029 |
Neg logit contributions | |
amen | -0.122 |
Amen | -0.119 |
Him | -0.116 |
sets | -0.114 |
eness | -0.110 |
Token | hearts |
Token ID | 11954 |
Feature activation | -9.35e+00 |
Pos logit contributions | |
run | +0.104 |
bug | +0.102 |
ours | +0.094 |
idd | +0.091 |
udge | +0.091 |
Neg logit contributions | |
amen | -0.082 |
girls | -0.075 |
Rudolph | -0.070 |
Santa | -0.068 |
announcements | -0.068 |
Token | and |
Token ID | 290 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
amen | +0.376 |
Amen | +0.353 |
girls | +0.345 |
announcements | +0.341 |
mercy | +0.339 |
Neg logit contributions | |
bug | -0.239 |
run | -0.230 |
out | -0.204 |
ours | -0.185 |
ab | -0.183 |
Token | open |
Token ID | 1280 |
Feature activation | +2.89e+00 |
Pos logit contributions | |
bug | +0.010 |
run | +0.010 |
haired | +0.008 |
ours | +0.008 |
rous | +0.008 |
Neg logit contributions | |
amen | -0.009 |
Rudolph | -0.008 |
girls | -0.008 |
Santa | -0.008 |
mercy | -0.007 |
Token | minds |
Token ID | 9017 |
Feature activation | -5.20e+00 |
Pos logit contributions | |
run | +0.102 |
bug | +0.100 |
ours | +0.092 |
udge | +0.088 |
idd | +0.088 |
Neg logit contributions | |
amen | -0.084 |
girls | -0.077 |
Rudolph | -0.073 |
Santa | -0.072 |
hospitality | -0.071 |
Token | . |
Token ID | 13 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
amen | +0.185 |
girls | +0.162 |
Rudolph | +0.160 |
Amen | +0.155 |
announcements | +0.155 |
Neg logit contributions | |
run | -0.160 |
bug | -0.154 |
out | -0.142 |
udge | -0.134 |
ours | -0.127 |
Token | They |
Token ID | 1119 |
Feature activation | -6.06e+00 |
Pos logit contributions | |
amen | +0.007 |
girls | +0.007 |
mercy | +0.007 |
hospitality | +0.006 |
sets | +0.006 |
Neg logit contributions | |
run | -0.009 |
bug | -0.009 |
ours | -0.008 |
udge | -0.008 |
imb | -0.008 |
Token | by |
Token ID | 416 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
run | +0.000 |
bug | +0.000 |
out | +0.000 |
udge | +0.000 |
ang | +0.000 |
Neg logit contributions | |
amen | -0.000 |
girls | -0.000 |
announcements | -0.000 |
Rudolph | -0.000 |
Santa | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | -7.01e-01 |
Pos logit contributions | |
amen | +0.091 |
girls | +0.084 |
Rudolph | +0.080 |
Santa | +0.077 |
announcements | +0.077 |
Neg logit contributions | |
run | -0.102 |
bug | -0.097 |
udge | -0.088 |
ours | -0.086 |
idd | -0.086 |
Token | offering |
Token ID | 6011 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
amen | +0.021 |
girls | +0.020 |
Rudolph | +0.019 |
announcements | +0.018 |
Santa | +0.018 |
Neg logit contributions | |
run | -0.024 |
bug | -0.023 |
ours | -0.021 |
idd | -0.021 |
haired | -0.020 |
Token | words |
Token ID | 2456 |
Feature activation | -5.52e+00 |
Pos logit contributions | |
run | +0.062 |
bug | +0.059 |
udge | +0.053 |
ours | +0.053 |
out | +0.052 |
Neg logit contributions | |
amen | -0.052 |
girls | -0.048 |
announcements | -0.044 |
Claus | -0.043 |
Rudolph | -0.043 |
Token | of |
Token ID | 286 |
Feature activation | -4.90e-01 |
Pos logit contributions | |
amen | +0.196 |
girls | +0.183 |
announcements | +0.176 |
Rudolph | +0.171 |
Santa | +0.167 |
Neg logit contributions | |
run | -0.169 |
bug | -0.159 |
out | -0.134 |
haired | -0.134 |
ours | -0.132 |
Token | encouragement |
Token ID | 27529 |
Feature activation | -8.13e+00 |
Pos logit contributions | |
amen | +0.016 |
girls | +0.014 |
announcements | +0.014 |
Claus | +0.014 |
Santa | +0.013 |
Neg logit contributions | |
run | -0.016 |
bug | -0.015 |
udge | -0.014 |
ours | -0.013 |
rous | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
amen | +0.273 |
girls | +0.242 |
announcements | +0.232 |
Rudolph | +0.230 |
Santa | +0.227 |
Neg logit contributions | |
run | -0.274 |
bug | -0.243 |
out | -0.219 |
udge | -0.215 |
haired | -0.213 |
Token | |
Token ID | 220 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
run | +0.007 |
bug | +0.007 |
ours | +0.006 |
idd | +0.006 |
udge | +0.006 |
Neg logit contributions | |
amen | -0.005 |
girls | -0.005 |
Rudolph | -0.004 |
Santa | -0.004 |
announcements | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -6.93e-01 |
Pos logit contributions | |
amen | +0.051 |
girls | +0.047 |
Rudolph | +0.045 |
announcements | +0.043 |
Santa | +0.043 |
Neg logit contributions | |
run | -0.069 |
bug | -0.068 |
ours | -0.062 |
idd | -0.061 |
imb | -0.061 |
Token |
|
Token ID | 198 |
Feature activation | -7.13e-01 |
Pos logit contributions | |
amen | +0.020 |
girls | +0.018 |
Rudolph | +0.017 |
announcements | +0.017 |
hospitality | +0.017 |
Neg logit contributions | |
run | -0.025 |
bug | -0.024 |
ours | -0.022 |
idd | -0.022 |
udge | -0.022 |
Token | Finally |
Token ID | 11158 |
Feature activation | -2.98e+00 |
Pos logit contributions | |
amen | +0.018 |
Santa | +0.017 |
girls | +0.017 |
Rudolph | +0.016 |
announcements | +0.015 |
Neg logit contributions | |
run | -0.027 |
bug | -0.026 |
ours | -0.026 |
udge | -0.024 |
idd | -0.024 |
Token | always |
Token ID | 1464 |
Feature activation | -3.81e+00 |
Pos logit contributions | |
amen | +0.165 |
girls | +0.164 |
announcements | +0.154 |
sets | +0.148 |
Rudolph | +0.147 |
Neg logit contributions | |
bug | -0.153 |
run | -0.152 |
imb | -0.134 |
here | -0.126 |
udge | -0.126 |
Token | remember |
Token ID | 3505 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
amen | +0.124 |
girls | +0.122 |
announcements | +0.114 |
Rudolph | +0.114 |
Santa | +0.112 |
Neg logit contributions | |
run | -0.117 |
bug | -0.113 |
imb | -0.096 |
here | -0.094 |
rous | -0.093 |
Token | the |
Token ID | 262 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
amen | +0.046 |
girls | +0.046 |
announcements | +0.043 |
Claus | +0.042 |
MacDonald | +0.040 |
Neg logit contributions | |
run | -0.036 |
bug | -0.032 |
rous | -0.029 |
out | -0.028 |
ours | -0.028 |
Token | kindness |
Token ID | 23887 |
Feature activation | -5.15e+00 |
Pos logit contributions | |
amen | +0.077 |
girls | +0.076 |
Rudolph | +0.069 |
Santa | +0.067 |
Claus | +0.066 |
Neg logit contributions | |
run | -0.072 |
bug | -0.071 |
haired | -0.064 |
idd | -0.063 |
imb | -0.063 |
Token | and |
Token ID | 290 |
Feature activation | +6.40e-01 |
Pos logit contributions | |
amen | +0.192 |
girls | +0.184 |
announcements | +0.172 |
Rudolph | +0.171 |
Santa | +0.170 |
Neg logit contributions | |
run | -0.142 |
bug | -0.135 |
haired | -0.116 |
udge | -0.115 |
ours | -0.113 |
Token | encouragement |
Token ID | 27529 |
Feature activation | -8.19e+00 |
Pos logit contributions | |
run | +0.022 |
bug | +0.020 |
imb | +0.018 |
rous | +0.018 |
haired | +0.018 |
Neg logit contributions | |
amen | -0.019 |
girls | -0.019 |
Santa | -0.017 |
announcements | -0.017 |
Rudolph | -0.017 |
Token | from |
Token ID | 422 |
Feature activation | -5.91e-01 |
Pos logit contributions | |
amen | +0.298 |
girls | +0.282 |
Rudolph | +0.259 |
announcements | +0.259 |
Santa | +0.256 |
Neg logit contributions | |
run | -0.253 |
bug | -0.212 |
out | -0.192 |
udge | -0.188 |
haired | -0.186 |
Token | the |
Token ID | 262 |
Feature activation | -2.43e+00 |
Pos logit contributions | |
amen | +0.017 |
girls | +0.017 |
announcements | +0.015 |
sets | +0.015 |
Rudolph | +0.015 |
Neg logit contributions | |
run | -0.020 |
bug | -0.020 |
ours | -0.018 |
udge | -0.018 |
haired | -0.017 |
Token | clown |
Token ID | 25573 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
amen | +0.068 |
girls | +0.067 |
Rudolph | +0.061 |
Santa | +0.058 |
sets | +0.058 |
Neg logit contributions | |
run | -0.084 |
bug | -0.082 |
idd | -0.076 |
haired | -0.076 |
ours | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
bug | +0.042 |
run | +0.042 |
ang | +0.035 |
ours | +0.034 |
imb | +0.034 |
Neg logit contributions | |
amen | -0.044 |
girls | -0.040 |
announcements | -0.039 |
Rudolph | -0.039 |
Santa | -0.038 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
amen | +0.005 |
girls | +0.005 |
announcements | +0.004 |
Rudolph | +0.004 |
Santa | +0.004 |
Neg logit contributions | |
run | -0.005 |
bug | -0.005 |
ours | -0.005 |
imb | -0.005 |
udge | -0.005 |