Token | by |
Token ID | 416 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
� | +0.008 |
tul | +0.007 |
bal | +0.007 |
Notre | +0.006 |
te | +0.006 |
Neg logit contributions | |
ately | -0.022 |
ting | -0.018 |
upper | -0.018 |
34 | -0.018 |
arsh | -0.018 |
Token | one |
Token ID | 530 |
Feature activation | +6.38e-02 |
Pos logit contributions | |
ately | +0.019 |
nation | +0.016 |
arsh | +0.015 |
ting | +0.015 |
oo | +0.013 |
Neg logit contributions | |
bal | -0.025 |
Ark | -0.024 |
tul | -0.023 |
paramed | -0.023 |
Notre | -0.023 |
Token | � |
Token ID | 6184 |
Feature activation | +2.64e-02 |
Pos logit contributions | |
tul | +0.069 |
Notre | +0.068 |
bal | +0.068 |
esh | +0.067 |
Ark | +0.067 |
Neg logit contributions | |
ately | -0.070 |
nation | -0.056 |
arsh | -0.052 |
34 | -0.051 |
upper | -0.050 |
Token | � |
Token ID | 95 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
bal | +0.027 |
tul | +0.026 |
Notre | +0.026 |
cuc | +0.026 |
mo | +0.025 |
Neg logit contributions | |
ately | -0.030 |
nation | -0.025 |
arsh | -0.024 |
34 | -0.023 |
west | -0.023 |
Token | € |
Token ID | 26391 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
tul | +0.126 |
bal | +0.123 |
Notre | +0.118 |
cuc | +0.116 |
mo | +0.113 |
Neg logit contributions | |
ately | -0.143 |
nation | -0.123 |
arsh | -0.114 |
34 | -0.109 |
west | -0.108 |
Token | � |
Token ID | 447 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
ately | +0.012 |
arsh | +0.010 |
nation | +0.010 |
34 | +0.009 |
west | +0.008 |
Neg logit contributions | |
tul | -0.017 |
cuc | -0.016 |
Notre | -0.016 |
bal | -0.015 |
te | -0.015 |
Token | � |
Token ID | 250 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
tul | +0.282 |
bal | +0.280 |
Notre | +0.271 |
cuc | +0.259 |
te | +0.256 |
Neg logit contributions | |
ately | -0.269 |
nation | -0.222 |
arsh | -0.203 |
34 | -0.194 |
upper | -0.192 |
Token | 1 |
Token ID | 352 |
Feature activation | +2.94e-02 |
Pos logit contributions | |
tul | +0.023 |
bal | +0.023 |
Notre | +0.022 |
cuc | +0.021 |
esh | +0.021 |
Neg logit contributions | |
ately | -0.029 |
nation | -0.025 |
arsh | -0.023 |
34 | -0.022 |
west | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
tul | +0.032 |
bal | +0.030 |
Notre | +0.030 |
esh | +0.029 |
Ark | +0.029 |
Neg logit contributions | |
ately | -0.034 |
nation | -0.028 |
arsh | -0.026 |
34 | -0.026 |
ting | -0.024 |
Token | 2 |
Token ID | 362 |
Feature activation | -2.61e-03 |
Pos logit contributions | |
esh | +0.024 |
Notre | +0.023 |
tul | +0.023 |
bal | +0.023 |
Ark | +0.022 |
Neg logit contributions | |
ately | -0.039 |
nation | -0.033 |
ting | -0.031 |
34 | -0.031 |
arsh | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
ately | +0.003 |
nation | +0.002 |
34 | +0.002 |
arsh | +0.002 |
ting | +0.002 |
Neg logit contributions | |
bal | -0.003 |
tul | -0.003 |
Notre | -0.003 |
esh | -0.003 |
Ark | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
tul | +0.063 |
te | +0.054 |
sprink | +0.052 |
bal | +0.050 |
� | +0.049 |
Neg logit contributions | |
ately | -0.124 |
nation | -0.111 |
arsh | -0.110 |
upper | -0.110 |
west | -0.104 |
Token | da |
Token ID | 12379 |
Feature activation | +7.74e-02 |
Pos logit contributions | |
bal | +0.010 |
tul | +0.010 |
Notre | +0.009 |
esh | +0.009 |
von | +0.009 |
Neg logit contributions | |
ately | -0.017 |
nation | -0.014 |
arsh | -0.013 |
34 | -0.013 |
ting | -0.013 |
Token | is |
Token ID | 271 |
Feature activation | -6.38e-02 |
Pos logit contributions | |
tul | +0.090 |
bal | +0.087 |
Notre | +0.084 |
te | +0.082 |
cuc | +0.081 |
Neg logit contributions | |
ately | -0.082 |
nation | -0.066 |
arsh | -0.063 |
upper | -0.060 |
west | -0.059 |
Token | ies |
Token ID | 444 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
ately | +0.078 |
nation | +0.063 |
arsh | +0.059 |
ting | +0.058 |
34 | +0.057 |
Neg logit contributions | |
bal | -0.061 |
Notre | -0.061 |
tul | -0.061 |
esh | -0.059 |
Ark | -0.057 |
Token | , |
Token ID | 11 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
tul | +0.040 |
bal | +0.040 |
Notre | +0.039 |
esh | +0.037 |
cuc | +0.036 |
Neg logit contributions | |
ately | -0.051 |
nation | -0.043 |
arsh | -0.040 |
ting | -0.039 |
34 | -0.039 |
Token | tul |
Token ID | 48373 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
bal | +0.012 |
tul | +0.011 |
Notre | +0.011 |
esh | +0.011 |
von | +0.011 |
Neg logit contributions | |
ately | -0.017 |
nation | -0.014 |
arsh | -0.013 |
ting | -0.013 |
34 | -0.013 |
Token | ips |
Token ID | 2419 |
Feature activation | -9.10e-02 |
Pos logit contributions | |
tul | +0.247 |
bal | +0.231 |
sprink | +0.229 |
cuc | +0.226 |
te | +0.219 |
Neg logit contributions | |
ately | -0.291 |
nation | -0.277 |
upper | -0.248 |
arsh | -0.237 |
west | -0.232 |
Token | and |
Token ID | 290 |
Feature activation | +4.82e-03 |
Pos logit contributions | |
ately | +0.113 |
nation | +0.097 |
arsh | +0.090 |
34 | +0.087 |
upper | +0.086 |
Neg logit contributions | |
tul | -0.085 |
bal | -0.084 |
Notre | -0.081 |
cuc | -0.077 |
te | -0.076 |
Token | more |
Token ID | 517 |
Feature activation | -4.53e-02 |
Pos logit contributions | |
bal | +0.004 |
esh | +0.004 |
Notre | +0.004 |
Ark | +0.004 |
von | +0.004 |
Neg logit contributions | |
ately | -0.006 |
nation | -0.005 |
ting | -0.005 |
upper | -0.004 |
west | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
ately | +0.045 |
nation | +0.035 |
34 | +0.032 |
ting | +0.032 |
west | +0.031 |
Neg logit contributions | |
esh | -0.055 |
Ark | -0.053 |
Notre | -0.052 |
paramed | -0.051 |
EA | -0.050 |
Token | She |
Token ID | 1375 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
tul | +0.019 |
bal | +0.019 |
esh | +0.019 |
Notre | +0.019 |
mo | +0.018 |
Neg logit contributions | |
ately | -0.017 |
nation | -0.014 |
arsh | -0.013 |
west | -0.013 |
34 | -0.013 |
Token | perfect |
Token ID | 2818 |
Feature activation | +5.45e-02 |
Pos logit contributions | |
esh | +0.053 |
lyn | +0.050 |
muted | +0.049 |
Ark | +0.049 |
ALTH | +0.048 |
Neg logit contributions | |
ately | -0.041 |
usual | -0.034 |
upper | -0.031 |
yet | -0.029 |
first | -0.028 |
Token | bag |
Token ID | 6131 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
lyn | +0.072 |
Ark | +0.072 |
Notre | +0.070 |
muted | +0.069 |
paramed | +0.067 |
Neg logit contributions | |
ately | -0.044 |
usual | -0.034 |
yet | -0.033 |
first | -0.031 |
upper | -0.030 |
Token | � |
Token ID | 6184 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
tul | +0.013 |
Notre | +0.013 |
bal | +0.012 |
Ark | +0.012 |
esh | +0.012 |
Neg logit contributions | |
ately | -0.013 |
nation | -0.010 |
arsh | -0.010 |
ting | -0.009 |
upper | -0.009 |
Token | � |
Token ID | 95 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
mo | +0.022 |
Notre | +0.021 |
tul | +0.021 |
imag | +0.021 |
bal | +0.021 |
Neg logit contributions | |
ately | -0.020 |
nation | -0.017 |
arsh | -0.016 |
ting | -0.016 |
ancer | -0.015 |
Token | € |
Token ID | 26391 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
tul | +0.134 |
bal | +0.125 |
cuc | +0.124 |
Notre | +0.122 |
mo | +0.122 |
Neg logit contributions | |
ately | -0.138 |
nation | -0.124 |
arsh | -0.115 |
ting | -0.109 |
west | -0.108 |
Token | � |
Token ID | 447 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
ately | +0.011 |
arsh | +0.010 |
nation | +0.009 |
oo | +0.009 |
34 | +0.008 |
Neg logit contributions | |
tul | -0.016 |
cuc | -0.016 |
Notre | -0.015 |
thirteen | -0.015 |
cut | -0.015 |
Token | � |
Token ID | 250 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
tul | +0.276 |
bal | +0.266 |
Notre | +0.262 |
esh | +0.255 |
cuc | +0.250 |
Neg logit contributions | |
ately | -0.256 |
nation | -0.216 |
arsh | -0.199 |
west | -0.191 |
34 | -0.188 |
Token | now |
Token ID | 783 |
Feature activation | -4.88e-02 |
Pos logit contributions | |
tul | +0.018 |
bal | +0.018 |
Notre | +0.017 |
cuc | +0.016 |
esh | +0.016 |
Neg logit contributions | |
ately | -0.023 |
nation | -0.019 |
arsh | -0.018 |
34 | -0.017 |
west | -0.017 |
Token | I |
Token ID | 314 |
Feature activation | +8.38e-03 |
Pos logit contributions | |
ately | +0.035 |
nation | +0.024 |
arsh | +0.024 |
yet | +0.021 |
sometime | +0.021 |
Neg logit contributions | |
Notre | -0.068 |
tul | -0.067 |
paramed | -0.066 |
cuc | -0.066 |
stret | -0.065 |
Token | 'm |
Token ID | 1101 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
tul | +0.011 |
esh | +0.010 |
Notre | +0.010 |
cuc | +0.010 |
Ark | +0.010 |
Neg logit contributions | |
ately | -0.008 |
nation | -0.006 |
arsh | -0.006 |
sometime | -0.005 |
ancer | -0.005 |
Token | very |
Token ID | 845 |
Feature activation | +2.07e-02 |
Pos logit contributions | |
Notre | +0.026 |
tul | +0.026 |
esh | +0.025 |
bal | +0.025 |
cosmetics | +0.025 |
Neg logit contributions | |
ately | -0.021 |
nation | -0.018 |
arsh | -0.018 |
oo | -0.016 |
34 | -0.016 |
Token | those |
Token ID | 883 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
bal | +0.021 |
tul | +0.020 |
Notre | +0.020 |
von | +0.019 |
� | +0.019 |
Neg logit contributions | |
ately | -0.030 |
nation | -0.026 |
arsh | -0.023 |
upper | -0.023 |
34 | -0.023 |
Token | in |
Token ID | 287 |
Feature activation | +4.29e-02 |
Pos logit contributions | |
esh | +0.052 |
Ark | +0.052 |
bal | +0.051 |
icol | +0.050 |
Notre | +0.049 |
Neg logit contributions | |
ately | -0.047 |
nation | -0.034 |
34 | -0.033 |
arsh | -0.033 |
usual | -0.032 |
Token | need |
Token ID | 761 |
Feature activation | +5.11e-02 |
Pos logit contributions | |
bal | +0.054 |
tul | +0.051 |
paramed | +0.049 |
Ark | +0.049 |
Notre | +0.049 |
Neg logit contributions | |
ately | -0.039 |
nation | -0.032 |
arsh | -0.029 |
34 | -0.028 |
usual | -0.028 |
Token | â |
Token ID | 22940 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
bal | +0.055 |
tul | +0.055 |
Notre | +0.054 |
esh | +0.051 |
Ark | +0.051 |
Neg logit contributions | |
ately | -0.057 |
nation | -0.046 |
arsh | -0.042 |
ting | -0.041 |
upper | -0.040 |
Token | € |
Token ID | 26391 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
bal | +0.115 |
tul | +0.113 |
Notre | +0.110 |
esh | +0.109 |
� | +0.107 |
Neg logit contributions | |
ately | -0.140 |
nation | -0.113 |
34 | -0.108 |
arsh | -0.106 |
west | -0.104 |
Token | � |
Token ID | 447 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
ately | +0.019 |
arsh | +0.017 |
nation | +0.016 |
oo | +0.015 |
owing | +0.014 |
Neg logit contributions | |
tul | -0.023 |
cuc | -0.022 |
Notre | -0.020 |
thirteen | -0.020 |
imag | -0.019 |
Token | � |
Token ID | 251 |
Feature activation | +4.95e-02 |
Pos logit contributions | |
tul | +0.269 |
bal | +0.256 |
Notre | +0.252 |
esh | +0.245 |
cuc | +0.241 |
Neg logit contributions | |
ately | -0.265 |
nation | -0.230 |
arsh | -0.209 |
west | -0.201 |
ting | -0.198 |
Token | even |
Token ID | 10197 |
Feature activation | +3.43e-02 |
Pos logit contributions | |
tul | +0.056 |
Notre | +0.054 |
bal | +0.053 |
esh | +0.052 |
thirteen | +0.052 |
Neg logit contributions | |
ately | -0.053 |
nation | -0.042 |
arsh | -0.040 |
ting | -0.039 |
west | -0.037 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
tul | +0.037 |
bal | +0.035 |
Notre | +0.035 |
esh | +0.034 |
mo | +0.033 |
Neg logit contributions | |
ately | -0.039 |
nation | -0.032 |
arsh | -0.030 |
ting | -0.029 |
34 | -0.028 |
Token | One |
Token ID | 3198 |
Feature activation | +1.87e-03 |
Pos logit contributions | |
tul | +0.097 |
bal | +0.097 |
Notre | +0.092 |
cuc | +0.086 |
te | +0.086 |
Neg logit contributions | |
ately | -0.168 |
nation | -0.145 |
arsh | -0.137 |
34 | -0.132 |
upper | -0.132 |
Token | day |
Token ID | 1110 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
tul | +0.002 |
Notre | +0.002 |
cuc | +0.002 |
bal | +0.002 |
stret | +0.002 |
Neg logit contributions | |
ately | -0.002 |
nation | -0.001 |
arsh | -0.001 |
usual | -0.001 |
west | -0.001 |
Token | his |
Token ID | 465 |
Feature activation | -4.36e-03 |
Pos logit contributions | |
ately | +0.006 |
nation | +0.005 |
arsh | +0.005 |
34 | +0.005 |
upper | +0.005 |
Neg logit contributions | |
tul | -0.004 |
bal | -0.004 |
Notre | -0.004 |
cuc | -0.004 |
esh | -0.004 |
Token | friends |
Token ID | 2460 |
Feature activation | +5.40e-02 |
Pos logit contributions | |
ately | +0.007 |
nation | +0.006 |
arsh | +0.006 |
west | +0.006 |
ting | +0.006 |
Neg logit contributions | |
bal | -0.003 |
tul | -0.003 |
Notre | -0.003 |
pap | -0.003 |
te | -0.003 |
Token | dropped |
Token ID | 5710 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
bal | +0.043 |
tul | +0.041 |
mo | +0.039 |
� | +0.038 |
Notre | +0.038 |
Neg logit contributions | |
ately | -0.073 |
nation | -0.066 |
arsh | -0.062 |
34 | -0.061 |
upper | -0.061 |
Token | her |
Token ID | 607 |
Feature activation | +7.71e-03 |
Pos logit contributions | |
tul | +0.018 |
bal | +0.017 |
Notre | +0.017 |
Ark | +0.016 |
von | +0.016 |
Neg logit contributions | |
ately | -0.017 |
nation | -0.013 |
ting | -0.013 |
arsh | -0.012 |
34 | -0.012 |
Token | lip |
Token ID | 10645 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
tul | +0.008 |
esh | +0.008 |
Notre | +0.008 |
von | +0.008 |
thirteen | +0.008 |
Neg logit contributions | |
ately | -0.009 |
nation | -0.007 |
upper | -0.006 |
usual | -0.006 |
arsh | -0.006 |
Token | bal |
Token ID | 3652 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
ately | +0.041 |
nation | +0.036 |
arsh | +0.033 |
34 | +0.033 |
west | +0.032 |
Neg logit contributions | |
bal | -0.028 |
tul | -0.025 |
� | -0.023 |
esh | -0.023 |
Notre | -0.023 |
Token | m |
Token ID | 76 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
bal | +0.241 |
ners | +0.211 |
tin | +0.202 |
mo | +0.202 |
tul | +0.196 |
Neg logit contributions | |
34 | -0.236 |
nation | -0.235 |
ately | -0.230 |
upper | -0.226 |
arsh | -0.222 |
Token | on |
Token ID | 319 |
Feature activation | -4.91e-02 |
Pos logit contributions | |
ately | +0.170 |
nation | +0.136 |
arsh | +0.129 |
upper | +0.121 |
34 | +0.121 |
Neg logit contributions | |
tul | -0.176 |
bal | -0.170 |
Notre | -0.169 |
cuc | -0.162 |
esh | -0.160 |
Token | the |
Token ID | 262 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
ately | +0.057 |
nation | +0.049 |
arsh | +0.045 |
ting | +0.044 |
west | +0.043 |
Neg logit contributions | |
bal | -0.050 |
tul | -0.049 |
Notre | -0.047 |
cuc | -0.044 |
von | -0.044 |
Token | ground |
Token ID | 2323 |
Feature activation | -3.95e-02 |
Pos logit contributions | |
tul | +0.017 |
bal | +0.017 |
Notre | +0.016 |
te | +0.015 |
mo | +0.015 |
Neg logit contributions | |
ately | -0.022 |
nation | -0.019 |
arsh | -0.018 |
west | -0.017 |
34 | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
ately | +0.049 |
nation | +0.042 |
arsh | +0.039 |
34 | +0.037 |
upper | +0.037 |
Neg logit contributions | |
tul | -0.037 |
bal | -0.037 |
Notre | -0.036 |
cuc | -0.034 |
esh | -0.034 |
Token | something |
Token ID | 1223 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
ately | +0.049 |
nation | +0.042 |
upper | +0.038 |
arsh | +0.038 |
ting | +0.038 |
Neg logit contributions | |
tul | -0.024 |
bal | -0.024 |
mo | -0.023 |
� | -0.023 |
cuc | -0.023 |
Token | new |
Token ID | 649 |
Feature activation | -7.82e-02 |
Pos logit contributions | |
ately | +0.019 |
nation | +0.016 |
arsh | +0.015 |
34 | +0.014 |
upper | +0.014 |
Neg logit contributions | |
bal | -0.018 |
tul | -0.018 |
Notre | -0.018 |
cuc | -0.017 |
te | -0.017 |
Token | � |
Token ID | 6184 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
ately | +0.097 |
nation | +0.082 |
arsh | +0.076 |
34 | +0.074 |
west | +0.073 |
Neg logit contributions | |
tul | -0.075 |
bal | -0.074 |
Notre | -0.071 |
cuc | -0.068 |
esh | -0.067 |
Token | � |
Token ID | 95 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
tul | +0.022 |
bal | +0.022 |
Notre | +0.021 |
cuc | +0.020 |
esh | +0.020 |
Neg logit contributions | |
ately | -0.027 |
nation | -0.022 |
arsh | -0.021 |
34 | -0.020 |
ting | -0.020 |
Token | € |
Token ID | 26391 |
Feature activation | -4.46e-03 |
Pos logit contributions | |
bal | +0.114 |
Notre | +0.111 |
tul | +0.110 |
� | +0.105 |
esh | +0.105 |
Neg logit contributions | |
ately | -0.161 |
nation | -0.127 |
arsh | -0.120 |
34 | -0.118 |
ting | -0.117 |
Token | � |
Token ID | 447 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
ately | +0.005 |
nation | +0.005 |
arsh | +0.004 |
34 | +0.004 |
upper | +0.004 |
Neg logit contributions | |
tul | -0.004 |
bal | -0.004 |
Notre | -0.004 |
cuc | -0.004 |
te | -0.004 |
Token | � |
Token ID | 250 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
sprink | +0.118 |
� | +0.117 |
bal | +0.110 |
mo | +0.107 |
Es | +0.105 |
Neg logit contributions | |
ately | -0.452 |
nation | -0.369 |
ting | -0.365 |
fully | -0.359 |
arsh | -0.346 |
Token | he |
Token ID | 339 |
Feature activation | +5.61e-02 |
Pos logit contributions | |
bal | +0.011 |
tul | +0.010 |
� | +0.009 |
Notre | +0.009 |
thirteen | +0.009 |
Neg logit contributions | |
ately | -0.015 |
nation | -0.012 |
arsh | -0.012 |
34 | -0.011 |
ting | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
tul | +0.063 |
esh | +0.059 |
stret | +0.059 |
paramed | +0.058 |
Notre | +0.057 |
Neg logit contributions | |
ately | -0.060 |
nation | -0.051 |
arsh | -0.049 |
upper | -0.045 |
rip | -0.044 |
Token | going |
Token ID | 1016 |
Feature activation | +3.90e-02 |
Pos logit contributions | |
tul | +0.018 |
esh | +0.018 |
bal | +0.018 |
Notre | +0.017 |
von | +0.016 |
Neg logit contributions | |
ately | -0.017 |
nation | -0.015 |
arsh | -0.014 |
34 | -0.014 |
upper | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | +3.90e-02 |
Pos logit contributions | |
tul | +0.052 |
bal | +0.050 |
Notre | +0.050 |
esh | +0.048 |
cuc | +0.048 |
Neg logit contributions | |
ately | -0.034 |
nation | -0.027 |
ting | -0.026 |
arsh | -0.025 |
34 | -0.022 |
Token | rare |
Token ID | 4071 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
tul | +0.037 |
bal | +0.036 |
Notre | +0.036 |
esh | +0.035 |
cuc | +0.034 |
Neg logit contributions | |
ately | -0.042 |
nation | -0.035 |
arsh | -0.033 |
upper | -0.032 |
34 | -0.031 |
Token | silence |
Token ID | 9550 |
Feature activation | -6.24e-02 |
Pos logit contributions | |
ately | +0.029 |
nation | +0.023 |
arsh | +0.023 |
west | +0.021 |
upper | +0.021 |
Neg logit contributions | |
Notre | -0.028 |
esh | -0.028 |
Ark | -0.028 |
bal | -0.027 |
tul | -0.027 |
Token | � |
Token ID | 6184 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
ately | +0.067 |
nation | +0.055 |
arsh | +0.051 |
ting | +0.049 |
34 | +0.048 |
Neg logit contributions | |
Ark | -0.067 |
bal | -0.067 |
tul | -0.067 |
Notre | -0.066 |
esh | -0.064 |
Token | � |
Token ID | 95 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
tul | +0.024 |
bal | +0.024 |
Notre | +0.023 |
cuc | +0.022 |
esh | +0.021 |
Neg logit contributions | |
ately | -0.028 |
nation | -0.024 |
arsh | -0.022 |
34 | -0.021 |
west | -0.021 |
Token | € |
Token ID | 26391 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
tul | +0.116 |
bal | +0.115 |
Notre | +0.111 |
esh | +0.105 |
cuc | +0.104 |
Neg logit contributions | |
ately | -0.150 |
nation | -0.127 |
arsh | -0.118 |
34 | -0.114 |
upper | -0.113 |
Token | � |
Token ID | 447 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
ately | +0.012 |
arsh | +0.010 |
nation | +0.010 |
oo | +0.009 |
34 | +0.009 |
Neg logit contributions | |
tul | -0.013 |
cuc | -0.013 |
Notre | -0.012 |
thirteen | -0.012 |
bal | -0.012 |
Token | � |
Token ID | 250 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
tul | +0.244 |
bal | +0.243 |
Notre | +0.234 |
cuc | +0.223 |
te | +0.220 |
Neg logit contributions | |
ately | -0.290 |
nation | -0.242 |
arsh | -0.224 |
34 | -0.215 |
upper | -0.214 |
Token | that |
Token ID | 326 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
tul | +0.011 |
bal | +0.011 |
Notre | +0.010 |
cuc | +0.010 |
esh | +0.010 |
Neg logit contributions | |
ately | -0.014 |
nation | -0.012 |
arsh | -0.011 |
34 | -0.011 |
upper | -0.011 |
Token | added |
Token ID | 2087 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
tul | +0.030 |
bal | +0.030 |
Notre | +0.029 |
cuc | +0.028 |
esh | +0.028 |
Neg logit contributions | |
ately | -0.054 |
nation | -0.048 |
arsh | -0.045 |
west | -0.043 |
ting | -0.043 |
Token | to |
Token ID | 284 |
Feature activation | +3.29e-02 |
Pos logit contributions | |
tul | +0.039 |
bal | +0.039 |
Notre | +0.038 |
esh | +0.037 |
Ark | +0.036 |
Neg logit contributions | |
ately | -0.046 |
nation | -0.039 |
arsh | -0.036 |
ting | -0.035 |
34 | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
tul | +0.032 |
bal | +0.032 |
Notre | +0.031 |
cuc | +0.029 |
esh | +0.029 |
Neg logit contributions | |
ately | -0.040 |
nation | -0.034 |
arsh | -0.031 |
34 | -0.030 |
upper | -0.030 |
Token | moral |
Token ID | 6573 |
Feature activation | +7.31e-02 |
Pos logit contributions | |
esh | +0.026 |
tul | +0.025 |
pter | +0.025 |
thirteen | +0.024 |
Notre | +0.024 |
Neg logit contributions | |
ately | -0.017 |
nation | -0.015 |
upper | -0.015 |
arsh | -0.014 |
usual | -0.014 |
Token | value |
Token ID | 1988 |
Feature activation | -6.89e-03 |
Pos logit contributions | |
tul | +0.075 |
bal | +0.074 |
Notre | +0.071 |
cuc | +0.068 |
esh | +0.068 |
Neg logit contributions | |
ately | -0.085 |
nation | -0.072 |
arsh | -0.066 |
34 | -0.064 |
west | -0.063 |
Token | � |
Token ID | 6184 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
ately | +0.009 |
nation | +0.008 |
upper | +0.008 |
west | +0.008 |
34 | +0.008 |
Neg logit contributions | |
bal | -0.005 |
tul | -0.005 |
Notre | -0.005 |
� | -0.005 |
cuc | -0.005 |
Token | � |
Token ID | 95 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
bal | +0.018 |
tul | +0.018 |
Notre | +0.018 |
esh | +0.017 |
von | +0.017 |
Neg logit contributions | |
ately | -0.023 |
nation | -0.019 |
ting | -0.018 |
34 | -0.017 |
arsh | -0.017 |
Token | € |
Token ID | 26391 |
Feature activation | -4.45e-03 |
Pos logit contributions | |
bal | +0.105 |
Notre | +0.103 |
esh | +0.103 |
� | +0.099 |
von | +0.098 |
Neg logit contributions | |
ately | -0.155 |
nation | -0.123 |
ting | -0.119 |
arsh | -0.118 |
34 | -0.117 |
Token | � |
Token ID | 447 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
ately | +0.005 |
nation | +0.004 |
arsh | +0.004 |
34 | +0.004 |
west | +0.003 |
Neg logit contributions | |
tul | -0.005 |
cuc | -0.005 |
te | -0.005 |
bal | -0.005 |
Notre | -0.005 |
Token | � |
Token ID | 250 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
� | +0.186 |
Notre | +0.183 |
cuc | +0.175 |
foc | +0.175 |
Es | +0.174 |
Neg logit contributions | |
ately | -0.353 |
nation | -0.285 |
fully | -0.271 |
ous | -0.267 |
upper | -0.265 |
Token | to |
Token ID | 284 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
bal | +0.010 |
tul | +0.010 |
Notre | +0.009 |
� | +0.009 |
cuc | +0.009 |
Neg logit contributions | |
ately | -0.015 |
nation | -0.013 |
arsh | -0.012 |
34 | -0.012 |
upper | -0.012 |
Token | be |
Token ID | 307 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
tul | +0.032 |
bal | +0.029 |
Notre | +0.028 |
cuc | +0.028 |
pap | +0.028 |
Neg logit contributions | |
ately | -0.054 |
nation | -0.046 |
34 | -0.043 |
ting | -0.043 |
arsh | -0.042 |
Token | kind |
Token ID | 1611 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
tul | +0.021 |
Notre | +0.021 |
bal | +0.021 |
esh | +0.020 |
imet | +0.020 |
Neg logit contributions | |
ately | -0.016 |
nation | -0.014 |
arsh | -0.014 |
upper | -0.013 |
west | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
ately | +0.013 |
arsh | +0.009 |
ting | +0.009 |
nation | +0.009 |
oo | +0.008 |
Neg logit contributions | |
tul | -0.019 |
mo | -0.019 |
Notre | -0.019 |
stret | -0.019 |
marsh | -0.018 |
Token | ." |
Token ID | 526 |
Feature activation | -7.13e-03 |
Pos logit contributions | |
ately | +0.007 |
upper | +0.006 |
nation | +0.006 |
arsh | +0.005 |
34 | +0.005 |
Neg logit contributions | |
bal | -0.004 |
� | -0.004 |
tul | -0.003 |
te | -0.003 |
Notre | -0.003 |
Token | |
Token ID | 220 |
Feature activation | +3.80e-02 |
Pos logit contributions | |
ately | +0.007 |
nation | +0.005 |
ting | +0.005 |
west | +0.005 |
arsh | +0.005 |
Neg logit contributions | |
Notre | -0.008 |
bal | -0.008 |
tul | -0.008 |
cuc | -0.008 |
thirteen | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
tul | +0.041 |
bal | +0.040 |
Notre | +0.040 |
cuc | +0.038 |
esh | +0.037 |
Neg logit contributions | |
ately | -0.042 |
nation | -0.035 |
arsh | -0.032 |
34 | -0.031 |
west | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
tul | +0.018 |
bal | +0.017 |
Notre | +0.017 |
cuc | +0.017 |
thirteen | +0.017 |
Neg logit contributions | |
ately | -0.014 |
nation | -0.012 |
west | -0.010 |
arsh | -0.010 |
ting | -0.010 |
Token | Ã |
Token ID | 5746 |
Feature activation | -6.13e-03 |
Pos logit contributions | |
majesty | +0.023 |
fault | +0.023 |
pter | +0.023 |
imag | +0.023 |
cuc | +0.023 |
Neg logit contributions | |
nation | -0.009 |
ately | -0.008 |
west | -0.007 |
ting | -0.006 |
expensive | -0.005 |
Token | � |
Token ID | 447 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
ately | +0.007 |
nation | +0.006 |
arsh | +0.005 |
ting | +0.005 |
west | +0.005 |
Neg logit contributions | |
tul | -0.007 |
bal | -0.006 |
Notre | -0.006 |
cuc | -0.006 |
mo | -0.006 |
Token | � |
Token ID | 108 |
Feature activation | +5.94e-02 |
Pos logit contributions | |
bal | +0.143 |
Notre | +0.141 |
tul | +0.141 |
Es | +0.129 |
sprink | +0.128 |
Neg logit contributions | |
ately | -0.390 |
nation | -0.329 |
arsh | -0.316 |
upper | -0.314 |
owing | -0.309 |
Token | mil |
Token ID | 25433 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
tul | +0.071 |
cuc | +0.067 |
Notre | +0.066 |
bal | +0.066 |
imag | +0.065 |
Neg logit contributions | |
ately | -0.061 |
nation | -0.053 |
arsh | -0.047 |
west | -0.046 |
ting | -0.043 |
Token | ie |
Token ID | 494 |
Feature activation | -2.30e-02 |
Pos logit contributions | |
tul | +0.019 |
bal | +0.019 |
Notre | +0.018 |
cuc | +0.017 |
esh | +0.017 |
Neg logit contributions | |
ately | -0.019 |
nation | -0.016 |
arsh | -0.014 |
34 | -0.014 |
upper | -0.013 |
Token | opened |
Token ID | 4721 |
Feature activation | +3.45e-04 |
Pos logit contributions | |
ately | +0.029 |
nation | +0.024 |
arsh | +0.023 |
34 | +0.022 |
upper | +0.022 |
Neg logit contributions | |
tul | -0.022 |
bal | -0.021 |
Notre | -0.021 |
cuc | -0.020 |
esh | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | +3.14e-02 |
Pos logit contributions | |
bal | +0.001 |
Notre | +0.001 |
dod | +0.000 |
lyn | +0.000 |
unbeat | +0.000 |
Neg logit contributions | |
ately | -0.000 |
ting | -0.000 |
west | -0.000 |
nation | -0.000 |
34 | -0.000 |
Token | She |
Token ID | 1375 |
Feature activation | +9.02e-03 |
Pos logit contributions | |
tul | +0.005 |
bal | +0.005 |
Notre | +0.004 |
von | +0.004 |
cuc | +0.004 |
Neg logit contributions | |
ately | -0.006 |
nation | -0.006 |
arsh | -0.005 |
34 | -0.005 |
ting | -0.005 |
Token | also |
Token ID | 635 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
tul | +0.009 |
bal | +0.009 |
Notre | +0.008 |
cuc | +0.008 |
esh | +0.008 |
Neg logit contributions | |
ately | -0.011 |
nation | -0.009 |
arsh | -0.009 |
34 | -0.008 |
west | -0.008 |
Token | put |
Token ID | 1234 |
Feature activation | +4.38e-02 |
Pos logit contributions | |
ately | +0.014 |
nation | +0.012 |
arsh | +0.011 |
34 | +0.011 |
ting | +0.011 |
Neg logit contributions | |
tul | -0.009 |
bal | -0.009 |
Notre | -0.009 |
te | -0.008 |
cuc | -0.008 |
Token | some |
Token ID | 617 |
Feature activation | -3.52e-02 |
Pos logit contributions | |
bal | +0.033 |
tul | +0.033 |
sprink | +0.032 |
te | +0.031 |
mo | +0.030 |
Neg logit contributions | |
ately | -0.065 |
arsh | -0.059 |
nation | -0.058 |
ous | -0.052 |
upper | -0.052 |
Token | lip |
Token ID | 10645 |
Feature activation | -3.61e-02 |
Pos logit contributions | |
ately | +0.044 |
nation | +0.038 |
upper | +0.035 |
ting | +0.034 |
34 | +0.034 |
Neg logit contributions | |
esh | -0.031 |
tul | -0.031 |
Notre | -0.030 |
bal | -0.030 |
Ark | -0.029 |
Token | bal |
Token ID | 3652 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
ately | +0.046 |
nation | +0.039 |
arsh | +0.036 |
ting | +0.035 |
upper | +0.035 |
Neg logit contributions | |
tul | -0.033 |
Notre | -0.032 |
bal | -0.032 |
cuc | -0.030 |
esh | -0.030 |
Token | m |
Token ID | 76 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
bal | +0.225 |
sprink | +0.204 |
pter | +0.193 |
tul | +0.190 |
ners | +0.189 |
Neg logit contributions | |
nation | -0.252 |
arsh | -0.244 |
34 | -0.243 |
arter | -0.239 |
avail | -0.238 |
Token | on |
Token ID | 319 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
ately | +0.201 |
nation | +0.178 |
arsh | +0.161 |
west | +0.159 |
34 | +0.157 |
Neg logit contributions | |
bal | -0.151 |
tul | -0.147 |
Notre | -0.135 |
mo | -0.132 |
cuc | -0.131 |
Token | her |
Token ID | 607 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
ately | +0.068 |
nation | +0.059 |
arsh | +0.056 |
ancer | +0.054 |
34 | +0.054 |
Neg logit contributions | |
tul | -0.040 |
bal | -0.038 |
Notre | -0.035 |
te | -0.035 |
mo | -0.035 |
Token | lips |
Token ID | 11914 |
Feature activation | -4.92e-03 |
Pos logit contributions | |
bal | +0.008 |
tul | +0.008 |
te | +0.007 |
Notre | +0.007 |
mo | +0.007 |
Neg logit contributions | |
ately | -0.015 |
nation | -0.013 |
arsh | -0.013 |
ting | -0.012 |
west | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
ately | +0.008 |
nation | +0.008 |
west | +0.007 |
34 | +0.007 |
xy | +0.007 |
Neg logit contributions | |
bal | -0.003 |
� | -0.002 |
tul | -0.002 |
te | -0.002 |
brush | -0.002 |
Token | in |
Token ID | 287 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
bal | +0.001 |
tul | +0.001 |
Notre | +0.001 |
cuc | +0.001 |
esh | +0.001 |
Neg logit contributions | |
ately | -0.001 |
nation | -0.001 |
arsh | -0.001 |
west | -0.001 |
34 | -0.001 |
Token | return |
Token ID | 1441 |
Feature activation | -4.38e-02 |
Pos logit contributions | |
tul | +0.038 |
bal | +0.038 |
Notre | +0.036 |
cuc | +0.035 |
esh | +0.034 |
Neg logit contributions | |
ately | -0.048 |
nation | -0.041 |
arsh | -0.038 |
34 | -0.036 |
west | -0.036 |
Token | � |
Token ID | 6184 |
Feature activation | +2.19e-02 |
Pos logit contributions | |
ately | +0.057 |
nation | +0.051 |
arsh | +0.047 |
west | +0.046 |
34 | +0.045 |
Neg logit contributions | |
bal | -0.037 |
tul | -0.036 |
Notre | -0.035 |
esh | -0.033 |
cuc | -0.033 |
Token | � |
Token ID | 95 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
tul | +0.021 |
bal | +0.021 |
Notre | +0.021 |
cuc | +0.020 |
esh | +0.020 |
Neg logit contributions | |
ately | -0.026 |
nation | -0.022 |
arsh | -0.020 |
ting | -0.020 |
34 | -0.020 |
Token | € |
Token ID | 26391 |
Feature activation | -4.00e-03 |
Pos logit contributions | |
bal | +0.100 |
Notre | +0.099 |
esh | +0.095 |
tul | +0.094 |
von | +0.092 |
Neg logit contributions | |
ately | -0.168 |
nation | -0.134 |
arsh | -0.126 |
ting | -0.126 |
ancer | -0.125 |
Token | � |
Token ID | 447 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
ately | +0.005 |
nation | +0.004 |
arsh | +0.004 |
34 | +0.004 |
west | +0.004 |
Neg logit contributions | |
tul | -0.004 |
bal | -0.004 |
Notre | -0.004 |
cuc | -0.004 |
te | -0.004 |
Token | � |
Token ID | 250 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
sprink | +0.138 |
� | +0.137 |
bal | +0.129 |
cuc | +0.127 |
EA | +0.125 |
Neg logit contributions | |
ately | -0.416 |
nation | -0.341 |
fully | -0.320 |
34 | -0.320 |
arsh | -0.317 |
Token | a |
Token ID | 257 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
bal | +0.009 |
tul | +0.009 |
Notre | +0.008 |
cuc | +0.008 |
� | +0.008 |
Neg logit contributions | |
ately | -0.014 |
nation | -0.012 |
arsh | -0.011 |
34 | -0.011 |
upper | -0.011 |
Token | wife |
Token ID | 3656 |
Feature activation | +7.23e-03 |
Pos logit contributions | |
esh | +0.030 |
tul | +0.027 |
cosmetics | +0.026 |
Notre | +0.026 |
bal | +0.026 |
Neg logit contributions | |
ately | -0.019 |
nation | -0.016 |
upper | -0.016 |
arsh | -0.016 |
usual | -0.015 |
Token | ! |
Token ID | 0 |
Feature activation | +5.16e-02 |
Pos logit contributions | |
tul | +0.009 |
Notre | +0.008 |
esh | +0.008 |
bal | +0.008 |
mo | +0.008 |
Neg logit contributions | |
ately | -0.007 |
arsh | -0.006 |
nation | -0.006 |
ting | -0.005 |
owing | -0.005 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
tul | +0.055 |
mo | +0.054 |
stret | +0.051 |
Notre | +0.051 |
te | +0.050 |
Neg logit contributions | |
ately | -0.057 |
arsh | -0.047 |
nation | -0.046 |
west | -0.044 |
ting | -0.043 |
Token | lived |
Token ID | 5615 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
bal | +0.001 |
� | +0.001 |
tul | +0.001 |
esh | +0.001 |
mo | +0.001 |
Neg logit contributions | |
ately | -0.003 |
nation | -0.003 |
34 | -0.003 |
ting | -0.003 |
arsh | -0.002 |
Token | many |
Token ID | 867 |
Feature activation | -9.80e-02 |
Pos logit contributions | |
ately | +0.031 |
nation | +0.027 |
arsh | +0.026 |
ous | +0.025 |
ting | +0.025 |
Neg logit contributions | |
tul | -0.011 |
� | -0.010 |
Notre | -0.010 |
cuc | -0.010 |
bal | -0.009 |
Token | animals |
Token ID | 4695 |
Feature activation | -1.97e-04 |
Pos logit contributions | |
ately | +0.146 |
arsh | +0.122 |
nation | +0.122 |
upper | +0.113 |
ancer | +0.111 |
Neg logit contributions | |
tul | -0.079 |
Notre | -0.074 |
mo | -0.073 |
cuc | -0.073 |
bal | -0.070 |
Token | â |
Token ID | 22940 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
ately | +0.000 |
nation | +0.000 |
arsh | +0.000 |
ting | +0.000 |
west | +0.000 |
Neg logit contributions | |
tul | -0.000 |
bal | -0.000 |
von | -0.000 |
Notre | -0.000 |
thirteen | -0.000 |
Token | € |
Token ID | 26391 |
Feature activation | -6.18e-03 |
Pos logit contributions | |
� | +0.102 |
Notre | +0.101 |
bal | +0.099 |
von | +0.097 |
esh | +0.097 |
Neg logit contributions | |
ately | -0.167 |
34 | -0.133 |
ting | -0.121 |
upper | -0.121 |
iring | -0.119 |
Token | � |
Token ID | 447 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
ately | +0.008 |
nation | +0.007 |
ting | +0.006 |
upper | +0.006 |
arsh | +0.006 |
Neg logit contributions | |
bal | -0.005 |
Notre | -0.005 |
tul | -0.005 |
� | -0.005 |
te | -0.005 |
Token | � |
Token ID | 250 |
Feature activation | +9.46e-03 |
Pos logit contributions | |
sprink | +0.182 |
Notre | +0.181 |
tul | +0.176 |
mo | +0.176 |
esh | +0.175 |
Neg logit contributions | |
ately | -0.369 |
upper | -0.302 |
ting | -0.290 |
fully | -0.288 |
34 | -0.277 |
Token | a |
Token ID | 64 |
Feature activation | +6.50e-02 |
Pos logit contributions | |
bal | +0.009 |
tul | +0.008 |
� | +0.008 |
Notre | +0.008 |
thirteen | +0.007 |
Neg logit contributions | |
ately | -0.013 |
arsh | -0.011 |
nation | -0.010 |
34 | -0.010 |
upper | -0.010 |
Token | cow |
Token ID | 9875 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
Notre | +0.084 |
esh | +0.084 |
tul | +0.083 |
bal | +0.083 |
Ark | +0.083 |
Neg logit contributions | |
ately | -0.047 |
oo | -0.040 |
arsh | -0.039 |
nation | -0.038 |
ting | -0.034 |
Token | , |
Token ID | 11 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
bal | +0.033 |
Ark | +0.033 |
Notre | +0.032 |
esh | +0.032 |
stret | +0.031 |
Neg logit contributions | |
ately | -0.018 |
oo | -0.014 |
nation | -0.014 |
arsh | -0.014 |
ting | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
esh | +0.023 |
bal | +0.021 |
pter | +0.021 |
Notre | +0.021 |
icol | +0.021 |
Neg logit contributions | |
ately | -0.018 |
ting | -0.015 |
west | -0.014 |
nation | -0.014 |
ancer | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
bal | +0.034 |
tul | +0.029 |
� | +0.028 |
thirteen | +0.028 |
esh | +0.028 |
Neg logit contributions | |
ately | -0.066 |
nation | -0.058 |
arsh | -0.057 |
west | -0.054 |
34 | -0.053 |
Token | rocked |
Token ID | 36872 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
tul | +0.017 |
bal | +0.016 |
Notre | +0.016 |
esh | +0.015 |
cuc | +0.015 |
Neg logit contributions | |
ately | -0.023 |
nation | -0.019 |
arsh | -0.018 |
ting | -0.018 |
upper | -0.017 |
Token | � |
Token ID | 6184 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
bal | +0.027 |
� | +0.022 |
� | +0.021 |
bra | +0.020 |
tul | +0.020 |
Neg logit contributions | |
ately | -0.065 |
nation | -0.058 |
arsh | -0.058 |
west | -0.056 |
ancer | -0.054 |
Token | � |
Token ID | 95 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
tul | +0.027 |
Notre | +0.026 |
spe | +0.026 |
cuc | +0.026 |
mo | +0.026 |
Neg logit contributions | |
ately | -0.017 |
oo | -0.013 |
arsh | -0.013 |
nation | -0.013 |
our | -0.012 |
Token | € |
Token ID | 26391 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
tul | +0.142 |
bal | +0.128 |
cuc | +0.128 |
mo | +0.126 |
Notre | +0.125 |
Neg logit contributions | |
ately | -0.120 |
nation | -0.112 |
arsh | -0.104 |
upper | -0.099 |
west | -0.097 |
Token | � |
Token ID | 447 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
ately | +0.010 |
arsh | +0.009 |
nation | +0.008 |
34 | +0.008 |
oo | +0.007 |
Neg logit contributions | |
tul | -0.013 |
cuc | -0.012 |
te | -0.012 |
Notre | -0.011 |
thirteen | -0.011 |
Token | � |
Token ID | 250 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
bal | +0.263 |
tul | +0.261 |
Notre | +0.252 |
cuc | +0.242 |
von | +0.238 |
Neg logit contributions | |
ately | -0.266 |
nation | -0.219 |
arsh | -0.200 |
34 | -0.193 |
upper | -0.191 |
Token | higher |
Token ID | 2440 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
bal | +0.011 |
tul | +0.011 |
Notre | +0.010 |
cuc | +0.010 |
thirteen | +0.010 |
Neg logit contributions | |
ately | -0.015 |
nation | -0.012 |
arsh | -0.012 |
34 | -0.011 |
ting | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
tul | +0.055 |
Notre | +0.053 |
bal | +0.052 |
lyn | +0.052 |
paramed | +0.052 |
Neg logit contributions | |
ately | -0.054 |
nation | -0.042 |
arsh | -0.041 |
34 | -0.039 |
ting | -0.039 |
Token | higher |
Token ID | 2440 |
Feature activation | +4.78e-02 |
Pos logit contributions | |
bal | +0.014 |
tul | +0.013 |
Notre | +0.012 |
cuc | +0.012 |
mo | +0.012 |
Neg logit contributions | |
ately | -0.022 |
nation | -0.019 |
arsh | -0.018 |
34 | -0.017 |
west | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
tul | +0.053 |
paramed | +0.053 |
Notre | +0.052 |
bal | +0.051 |
lyn | +0.051 |
Neg logit contributions | |
ately | -0.053 |
nation | -0.040 |
ting | -0.040 |
arsh | -0.039 |
34 | -0.038 |
Token | worried |
Token ID | 7960 |
Feature activation | +3.02e-02 |
Pos logit contributions | |
tul | +0.013 |
bal | +0.013 |
Notre | +0.013 |
esh | +0.012 |
von | +0.012 |
Neg logit contributions | |
ately | -0.013 |
nation | -0.011 |
arsh | -0.010 |
34 | -0.010 |
ting | -0.010 |
Token | at |
Token ID | 379 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
tul | +0.036 |
Notre | +0.034 |
bal | +0.034 |
cuc | +0.033 |
Ark | +0.033 |
Neg logit contributions | |
ately | -0.032 |
nation | -0.025 |
ting | -0.023 |
arsh | -0.023 |
34 | -0.022 |
Token | all |
Token ID | 477 |
Feature activation | -5.57e-03 |
Pos logit contributions | |
tul | +0.013 |
pap | +0.011 |
Flor | +0.010 |
pand | +0.010 |
Wed | +0.010 |
Neg logit contributions | |
arsh | -0.021 |
nation | -0.021 |
upper | -0.021 |
ately | -0.020 |
west | -0.020 |
Token | â |
Token ID | 22940 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
ately | +0.004 |
ting | +0.003 |
yet | +0.002 |
either | +0.002 |
other | +0.002 |
Neg logit contributions | |
icol | -0.009 |
bal | -0.008 |
EA | -0.008 |
von | -0.008 |
paramed | -0.008 |
Token | € |
Token ID | 26391 |
Feature activation | -5.31e-03 |
Pos logit contributions | |
esh | +0.105 |
bal | +0.099 |
von | +0.099 |
� | +0.098 |
Notre | +0.098 |
Neg logit contributions | |
ately | -0.154 |
34 | -0.128 |
nation | -0.123 |
arsh | -0.121 |
ting | -0.121 |
Token | � |
Token ID | 447 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
ately | +0.005 |
nation | +0.005 |
arsh | +0.004 |
34 | +0.004 |
owing | +0.004 |
Neg logit contributions | |
tul | -0.007 |
cuc | -0.006 |
te | -0.006 |
thirteen | -0.006 |
bal | -0.006 |
Token | � |
Token ID | 251 |
Feature activation | +4.58e-02 |
Pos logit contributions | |
� | +0.211 |
Notre | +0.192 |
von | +0.189 |
dod | +0.185 |
te | +0.182 |
Neg logit contributions | |
ately | -0.327 |
nation | -0.274 |
upper | -0.261 |
arsh | -0.258 |
ancer | -0.258 |
Token | it |
Token ID | 270 |
Feature activation | -4.81e-02 |
Pos logit contributions | |
tul | +0.040 |
bal | +0.040 |
Notre | +0.038 |
mo | +0.036 |
esh | +0.036 |
Neg logit contributions | |
ately | -0.059 |
nation | -0.052 |
arsh | -0.048 |
34 | -0.047 |
upper | -0.047 |
Token | just |
Token ID | 655 |
Feature activation | +6.58e-02 |
Pos logit contributions | |
ately | +0.059 |
nation | +0.051 |
arsh | +0.047 |
34 | +0.046 |
upper | +0.045 |
Neg logit contributions | |
tul | -0.045 |
bal | -0.045 |
Notre | -0.043 |
cuc | -0.041 |
esh | -0.041 |
Token | started |
Token ID | 2067 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
tul | +0.062 |
bal | +0.061 |
Notre | +0.059 |
cuc | +0.056 |
esh | +0.055 |
Neg logit contributions | |
ately | -0.082 |
nation | -0.070 |
arsh | -0.065 |
34 | -0.063 |
ting | -0.063 |
Token | to |
Token ID | 284 |
Feature activation | +3.80e-02 |
Pos logit contributions | |
thirteen | +0.020 |
Ark | +0.020 |
Notre | +0.020 |
bal | +0.020 |
ALTH | +0.020 |
Neg logit contributions | |
ately | -0.011 |
nation | -0.007 |
owing | -0.007 |
arsh | -0.007 |
usual | -0.006 |
Token | believe |
Token ID | 1975 |
Feature activation | +7.02e-02 |
Pos logit contributions | |
tul | +0.058 |
bal | +0.057 |
Notre | +0.055 |
cuc | +0.054 |
te | +0.052 |
Neg logit contributions | |
ately | -0.065 |
nation | -0.055 |
arsh | -0.049 |
upper | -0.047 |
west | -0.047 |
Token | it |
Token ID | 340 |
Feature activation | -3.28e-03 |
Pos logit contributions | |
tul | +0.068 |
bal | +0.067 |
Notre | +0.064 |
cuc | +0.061 |
te | +0.061 |
Neg logit contributions | |
ately | -0.088 |
nation | -0.072 |
arsh | -0.069 |
ting | -0.067 |
34 | -0.066 |
Token | � |
Token ID | 6184 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
ately | +0.005 |
upper | +0.005 |
ancer | +0.005 |
west | +0.005 |
nation | +0.005 |
Neg logit contributions | |
� | -0.002 |
bal | -0.001 |
� | -0.001 |
esh | -0.001 |
de | -0.001 |
Token | � |
Token ID | 95 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
tul | +0.020 |
bal | +0.020 |
Notre | +0.019 |
cuc | +0.018 |
esh | +0.018 |
Neg logit contributions | |
ately | -0.024 |
nation | -0.020 |
arsh | -0.019 |
34 | -0.018 |
ting | -0.018 |
Token | € |
Token ID | 26391 |
Feature activation | -2.75e-03 |
Pos logit contributions | |
Notre | +0.107 |
bal | +0.107 |
tul | +0.106 |
esh | +0.102 |
von | +0.101 |
Neg logit contributions | |
ately | -0.157 |
nation | -0.127 |
arsh | -0.120 |
ting | -0.118 |
34 | -0.117 |
Token | � |
Token ID | 447 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
ately | +0.003 |
nation | +0.002 |
arsh | +0.002 |
34 | +0.002 |
west | +0.002 |
Neg logit contributions | |
tul | -0.003 |
bal | -0.003 |
Notre | -0.003 |
cuc | -0.003 |
te | -0.003 |
Token | � |
Token ID | 250 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
sprink | +0.100 |
� | +0.094 |
Notre | +0.090 |
dod | +0.089 |
mo | +0.079 |
Neg logit contributions | |
ately | -0.455 |
fully | -0.382 |
nation | -0.380 |
arsh | -0.372 |
ancer | -0.368 |
Token | their |
Token ID | 511 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
tul | +0.012 |
bal | +0.011 |
Notre | +0.011 |
cuc | +0.011 |
te | +0.011 |
Neg logit contributions | |
ately | -0.014 |
nation | -0.012 |
arsh | -0.010 |
ting | -0.010 |
west | -0.010 |
Token | naughty |
Token ID | 45128 |
Feature activation | +8.30e-03 |
Pos logit contributions | |
ners | +0.035 |
tul | +0.033 |
Ark | +0.033 |
Dun | +0.033 |
thirteen | +0.033 |
Neg logit contributions | |
ately | -0.028 |
upper | -0.021 |
usual | -0.021 |
34 | -0.019 |
arsh | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
tul | +0.009 |
Notre | +0.009 |
thirteen | +0.009 |
esh | +0.009 |
#### | +0.009 |
Neg logit contributions | |
ately | -0.009 |
arsh | -0.007 |
nation | -0.006 |
upper | -0.006 |
rip | -0.006 |
Token | stubborn |
Token ID | 24444 |
Feature activation | +5.40e-03 |
Pos logit contributions | |
tul | +0.015 |
bal | +0.015 |
Notre | +0.015 |
esh | +0.014 |
cuc | +0.014 |
Neg logit contributions | |
ately | -0.018 |
nation | -0.015 |
arsh | -0.014 |
34 | -0.013 |
ting | -0.013 |
Token | was |
Token ID | 373 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
ately | +0.007 |
nation | +0.006 |
arsh | +0.006 |
34 | +0.006 |
upper | +0.006 |
Neg logit contributions | |
bal | -0.005 |
tul | -0.004 |
� | -0.004 |
Notre | -0.004 |
esh | -0.004 |
Token | true |
Token ID | 2081 |
Feature activation | -4.94e-02 |
Pos logit contributions | |
tul | +0.013 |
bal | +0.013 |
Notre | +0.013 |
esh | +0.012 |
Ark | +0.012 |
Neg logit contributions | |
ately | -0.011 |
nation | -0.010 |
34 | -0.009 |
arsh | -0.009 |
west | -0.008 |
Token | � |
Token ID | 6184 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
ately | +0.061 |
nation | +0.051 |
arsh | +0.048 |
34 | +0.046 |
ting | +0.045 |
Neg logit contributions | |
tul | -0.048 |
bal | -0.047 |
Notre | -0.046 |
cuc | -0.044 |
esh | -0.043 |
Token | � |
Token ID | 95 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
tul | +0.019 |
bal | +0.018 |
Notre | +0.017 |
te | +0.017 |
esh | +0.017 |
Neg logit contributions | |
ately | -0.023 |
nation | -0.019 |
ting | -0.017 |
upper | -0.017 |
arsh | -0.017 |
Token | € |
Token ID | 26391 |
Feature activation | -3.88e-03 |
Pos logit contributions | |
bal | +0.105 |
tul | +0.104 |
Notre | +0.104 |
esh | +0.099 |
� | +0.098 |
Neg logit contributions | |
ately | -0.157 |
nation | -0.123 |
ancer | -0.118 |
arsh | -0.117 |
ting | -0.117 |
Token | � |
Token ID | 447 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
ately | +0.005 |
nation | +0.004 |
arsh | +0.004 |
34 | +0.004 |
west | +0.004 |
Neg logit contributions | |
tul | -0.004 |
bal | -0.004 |
Notre | -0.004 |
cuc | -0.003 |
te | -0.003 |
Token | � |
Token ID | 250 |
Feature activation | +8.80e-03 |
Pos logit contributions | |
sprink | +0.112 |
� | +0.110 |
Notre | +0.110 |
cuc | +0.107 |
dod | +0.101 |
Neg logit contributions | |
ately | -0.427 |
nation | -0.362 |
fully | -0.346 |
arsh | -0.344 |
ting | -0.338 |
Token | the |
Token ID | 262 |
Feature activation | +3.08e-02 |
Pos logit contributions | |
bal | +0.007 |
tul | +0.007 |
Notre | +0.007 |
� | +0.006 |
spe | +0.006 |
Neg logit contributions | |
ately | -0.012 |
nation | -0.010 |
arsh | -0.010 |
34 | -0.010 |
upper | -0.010 |
Token | porch |
Token ID | 33179 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
bal | +0.035 |
esh | +0.035 |
stret | +0.035 |
Ark | +0.034 |
Dun | +0.033 |
Neg logit contributions | |
ately | -0.029 |
nation | -0.025 |
34 | -0.024 |
west | -0.022 |
upper | -0.022 |
Token | was |
Token ID | 373 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
ately | +0.018 |
nation | +0.015 |
arsh | +0.014 |
ting | +0.013 |
34 | +0.013 |
Neg logit contributions | |
tul | -0.017 |
Notre | -0.016 |
bal | -0.016 |
Ark | -0.016 |
esh | -0.015 |
Token | lovely |
Token ID | 14081 |
Feature activation | +2.27e-02 |
Pos logit contributions | |
bal | +0.017 |
esh | +0.017 |
Notre | +0.016 |
tul | +0.016 |
von | +0.016 |
Neg logit contributions | |
ately | -0.014 |
nation | -0.013 |
34 | -0.012 |
arsh | -0.012 |
upper | -0.011 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
bal | +0.028 |
mo | +0.026 |
tul | +0.024 |
capt | +0.024 |
Notre | +0.023 |
Neg logit contributions | |
ately | -0.056 |
nation | -0.046 |
34 | -0.043 |
ting | -0.043 |
arsh | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | +4.63e-03 |
Pos logit contributions | |
ately | +0.013 |
ting | +0.010 |
arsh | +0.009 |
ous | +0.008 |
oo | +0.008 |
Neg logit contributions | |
tul | -0.023 |
thirteen | -0.022 |
paramed | -0.022 |
Ark | -0.022 |
stret | -0.022 |
Token | excited |
Token ID | 6568 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
bal | +0.004 |
mo | +0.003 |
tul | +0.003 |
Notre | +0.003 |
capt | +0.003 |
Neg logit contributions | |
ately | -0.007 |
nation | -0.006 |
34 | -0.006 |
arsh | -0.005 |
west | -0.005 |
Token | â |
Token ID | 22940 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
tul | +0.025 |
Notre | +0.023 |
bal | +0.023 |
cuc | +0.022 |
thirteen | +0.022 |
Neg logit contributions | |
ately | -0.026 |
nation | -0.021 |
arsh | -0.020 |
ting | -0.019 |
34 | -0.019 |
Token | € |
Token ID | 26391 |
Feature activation | -7.92e-03 |
Pos logit contributions | |
esh | +0.082 |
� | +0.081 |
von | +0.078 |
bal | +0.077 |
Notre | +0.071 |
Neg logit contributions | |
ately | -0.194 |
34 | -0.160 |
ting | -0.146 |
ancer | -0.143 |
arsh | -0.143 |
Token | � |
Token ID | 447 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
ately | +0.009 |
nation | +0.008 |
arsh | +0.007 |
34 | +0.007 |
west | +0.007 |
Neg logit contributions | |
tul | -0.009 |
bal | -0.008 |
Notre | -0.008 |
cuc | -0.008 |
te | -0.008 |
Token | � |
Token ID | 251 |
Feature activation | +4.17e-02 |
Pos logit contributions | |
icol | +0.149 |
sprink | +0.148 |
de | +0.146 |
marsh | +0.136 |
mo | +0.128 |
Neg logit contributions | |
ately | -0.421 |
nation | -0.347 |
ancer | -0.335 |
ting | -0.328 |
year | -0.319 |
Token | he |
Token ID | 258 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
bal | +0.035 |
tul | +0.033 |
Notre | +0.031 |
� | +0.031 |
mo | +0.030 |
Neg logit contributions | |
ately | -0.061 |
nation | -0.051 |
arsh | -0.047 |
34 | -0.047 |
xy | -0.045 |
Token | 'd |
Token ID | 1549 |
Feature activation | -5.69e-03 |
Pos logit contributions | |
ately | +0.198 |
nation | +0.172 |
arsh | +0.163 |
34 | +0.160 |
upper | +0.157 |
Neg logit contributions | |
bal | -0.137 |
tul | -0.132 |
Notre | -0.124 |
te | -0.123 |
cuc | -0.121 |
Token | just |
Token ID | 655 |
Feature activation | +6.38e-02 |
Pos logit contributions | |
ately | +0.005 |
upper | +0.004 |
arsh | +0.004 |
usual | +0.004 |
nation | +0.004 |
Neg logit contributions | |
tul | -0.007 |
esh | -0.007 |
muted | -0.007 |
Notre | -0.007 |
Ark | -0.007 |
Token | seen |
Token ID | 1775 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
tul | +0.060 |
bal | +0.059 |
Notre | +0.056 |
cuc | +0.054 |
esh | +0.053 |
Neg logit contributions | |
ately | -0.080 |
nation | -0.068 |
arsh | -0.063 |
34 | -0.061 |
west | -0.061 |
Token | made |
Token ID | 925 |
Feature activation | +6.98e-02 |
Pos logit contributions | |
tul | +0.126 |
esh | +0.124 |
stret | +0.122 |
Notre | +0.122 |
Ark | +0.117 |
Neg logit contributions | |
ately | -0.080 |
arsh | -0.059 |
rip | -0.058 |
nation | -0.058 |
west | -0.053 |
Token | this |
Token ID | 428 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
tul | +0.093 |
Notre | +0.091 |
bal | +0.087 |
lyn | +0.087 |
spe | +0.086 |
Neg logit contributions | |
ately | -0.055 |
arsh | -0.044 |
oo | -0.044 |
nation | -0.043 |
west | -0.040 |
Token | � |
Token ID | 6184 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
Ark | +0.049 |
Notre | +0.047 |
stret | +0.045 |
Dun | +0.045 |
esh | +0.045 |
Neg logit contributions | |
ately | -0.033 |
arsh | -0.030 |
nation | -0.027 |
west | -0.025 |
34 | -0.024 |
Token | � |
Token ID | 95 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
tul | +0.022 |
bal | +0.022 |
Notre | +0.022 |
esh | +0.021 |
pap | +0.021 |
Neg logit contributions | |
ately | -0.017 |
nation | -0.014 |
arsh | -0.014 |
west | -0.013 |
34 | -0.013 |
Token | € |
Token ID | 26391 |
Feature activation | -6.23e-03 |
Pos logit contributions | |
tul | +0.115 |
bal | +0.113 |
Notre | +0.108 |
cuc | +0.103 |
mo | +0.102 |
Neg logit contributions | |
ately | -0.145 |
nation | -0.125 |
arsh | -0.116 |
34 | -0.112 |
west | -0.111 |
Token | � |
Token ID | 447 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
ately | +0.006 |
arsh | +0.006 |
34 | +0.005 |
nation | +0.005 |
west | +0.005 |
Neg logit contributions | |
tul | -0.007 |
cuc | -0.006 |
bal | -0.006 |
thirteen | -0.006 |
te | -0.006 |
Token | � |
Token ID | 250 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
bal | +0.201 |
tul | +0.200 |
Notre | +0.193 |
cuc | +0.188 |
� | +0.187 |
Neg logit contributions | |
ately | -0.334 |
nation | -0.282 |
arsh | -0.262 |
ting | -0.254 |
34 | -0.252 |
Token | welcome |
Token ID | 7062 |
Feature activation | -8.12e-04 |
Pos logit contributions | |
bal | +0.009 |
tul | +0.009 |
Notre | +0.009 |
� | +0.008 |
thirteen | +0.008 |
Neg logit contributions | |
ately | -0.013 |
nation | -0.011 |
arsh | -0.010 |
34 | -0.010 |
upper | -0.010 |
Token | ! |
Token ID | 0 |
Feature activation | +4.74e-02 |
Pos logit contributions | |
ately | +0.001 |
oo | +0.001 |
ting | +0.001 |
arsh | +0.001 |
nation | +0.001 |
Neg logit contributions | |
tul | -0.001 |
Notre | -0.001 |
bal | -0.001 |
Ark | -0.001 |
muted | -0.001 |
Token | â |
Token ID | 22940 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
tul | +0.034 |
mo | +0.033 |
stret | +0.032 |
Notre | +0.031 |
cuc | +0.030 |
Neg logit contributions | |
ately | -0.067 |
arsh | -0.058 |
nation | -0.057 |
west | -0.056 |
ting | -0.055 |
Token | € |
Token ID | 26391 |
Feature activation | -4.43e-03 |
Pos logit contributions | |
bal | +0.113 |
Notre | +0.107 |
� | +0.106 |
tul | +0.106 |
esh | +0.105 |
Neg logit contributions | |
ately | -0.158 |
nation | -0.126 |
34 | -0.123 |
upper | -0.120 |
arsh | -0.120 |
Token | her |
Token ID | 607 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
cuc | +0.013 |
Notre | +0.013 |
tul | +0.013 |
Ark | +0.013 |
von | +0.013 |
Neg logit contributions | |
ately | -0.011 |
nation | -0.009 |
ting | -0.008 |
arsh | -0.008 |
usual | -0.008 |
Token | peace |
Token ID | 4167 |
Feature activation | -6.10e-02 |
Pos logit contributions | |
thirteen | +0.012 |
esh | +0.012 |
tul | +0.012 |
Ark | +0.012 |
cuc | +0.012 |
Neg logit contributions | |
ately | -0.013 |
usual | -0.010 |
nation | -0.010 |
upper | -0.010 |
ancer | -0.009 |
Token | � |
Token ID | 6184 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
ately | +0.055 |
nation | +0.040 |
ting | +0.038 |
ous | +0.037 |
arsh | +0.035 |
Neg logit contributions | |
Ark | -0.081 |
tul | -0.079 |
Notre | -0.077 |
lyn | -0.077 |
cuc | -0.077 |
Token | � |
Token ID | 95 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
tul | +0.021 |
bal | +0.021 |
Notre | +0.020 |
esh | +0.019 |
te | +0.019 |
Neg logit contributions | |
ately | -0.027 |
nation | -0.023 |
arsh | -0.022 |
34 | -0.021 |
upper | -0.021 |
Token | € |
Token ID | 26391 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
tul | +0.114 |
bal | +0.114 |
Notre | +0.110 |
esh | +0.104 |
te | +0.103 |
Neg logit contributions | |
ately | -0.156 |
nation | -0.131 |
arsh | -0.122 |
34 | -0.118 |
upper | -0.118 |
Token | � |
Token ID | 447 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
ately | +0.015 |
nation | +0.013 |
arsh | +0.013 |
34 | +0.011 |
oo | +0.011 |
Neg logit contributions | |
tul | -0.019 |
cuc | -0.018 |
Notre | -0.017 |
thirteen | -0.017 |
bal | -0.017 |
Token | � |
Token ID | 250 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
tul | +0.200 |
bal | +0.199 |
Notre | +0.189 |
cuc | +0.177 |
von | +0.177 |
Neg logit contributions | |
ately | -0.328 |
nation | -0.278 |
arsh | -0.262 |
34 | -0.255 |
upper | -0.253 |
Token | she |
Token ID | 673 |
Feature activation | +3.83e-02 |
Pos logit contributions | |
tul | +0.011 |
bal | +0.011 |
Notre | +0.010 |
cuc | +0.010 |
esh | +0.010 |
Neg logit contributions | |
ately | -0.014 |
nation | -0.012 |
arsh | -0.011 |
34 | -0.011 |
west | -0.011 |
Token | just |
Token ID | 655 |
Feature activation | +5.30e-02 |
Pos logit contributions | |
tul | +0.049 |
cuc | +0.048 |
esh | +0.046 |
stret | +0.045 |
Ark | +0.044 |
Neg logit contributions | |
ately | -0.035 |
nation | -0.030 |
arsh | -0.027 |
uch | -0.025 |
oo | -0.024 |
Token | needs |
Token ID | 2476 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
tul | +0.054 |
bal | +0.052 |
Notre | +0.051 |
esh | +0.051 |
cuc | +0.050 |
Neg logit contributions | |
ately | -0.061 |
nation | -0.052 |
arsh | -0.049 |
ting | -0.046 |
west | -0.045 |
Token | to |
Token ID | 284 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
tul | +0.014 |
bal | +0.014 |
Notre | +0.014 |
cuc | +0.013 |
esh | +0.013 |
Neg logit contributions | |
ately | -0.016 |
nation | -0.013 |
arsh | -0.012 |
ting | -0.012 |
34 | -0.011 |
Token | of |
Token ID | 286 |
Feature activation | +4.39e-03 |
Pos logit contributions | |
tul | +0.060 |
Notre | +0.060 |
bal | +0.058 |
Ark | +0.058 |
esh | +0.058 |
Neg logit contributions | |
ately | -0.051 |
nation | -0.037 |
arsh | -0.034 |
34 | -0.034 |
west | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
tul | +0.003 |
mo | +0.002 |
te | +0.002 |
pap | +0.002 |
cuc | +0.002 |
Neg logit contributions | |
ately | -0.007 |
nation | -0.007 |
arsh | -0.006 |
ancer | -0.006 |
upper | -0.006 |
Token | day |
Token ID | 1110 |
Feature activation | +1.01e-03 |
Pos logit contributions | |
tul | +0.027 |
bal | +0.026 |
esh | +0.026 |
Notre | +0.026 |
cuc | +0.025 |
Neg logit contributions | |
ately | -0.023 |
nation | -0.020 |
upper | -0.017 |
usual | -0.017 |
arsh | -0.017 |
Token | â |
Token ID | 22940 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
bal | +0.001 |
� | +0.001 |
mo | +0.001 |
te | +0.001 |
tul | +0.001 |
Neg logit contributions | |
ately | -0.001 |
nation | -0.001 |
west | -0.001 |
arsh | -0.001 |
upper | -0.001 |
Token | € |
Token ID | 26391 |
Feature activation | -6.46e-03 |
Pos logit contributions | |
esh | +0.092 |
� | +0.089 |
bal | +0.082 |
Notre | +0.082 |
von | +0.081 |
Neg logit contributions | |
ately | -0.181 |
34 | -0.147 |
ancer | -0.139 |
ting | -0.138 |
usual | -0.135 |
Token | � |
Token ID | 447 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
ately | +0.008 |
nation | +0.007 |
arsh | +0.006 |
34 | +0.006 |
west | +0.006 |
Neg logit contributions | |
tul | -0.006 |
bal | -0.006 |
Notre | -0.006 |
cuc | -0.006 |
te | -0.006 |
Token | � |
Token ID | 251 |
Feature activation | +4.23e-02 |
Pos logit contributions | |
sprink | +0.133 |
� | +0.129 |
foc | +0.127 |
bal | +0.117 |
mo | +0.115 |
Neg logit contributions | |
ately | -0.408 |
fully | -0.331 |
nation | -0.330 |
upper | -0.328 |
34 | -0.328 |
Token | because |
Token ID | 13893 |
Feature activation | +5.64e-02 |
Pos logit contributions | |
bal | +0.036 |
tul | +0.036 |
Notre | +0.034 |
cuc | +0.033 |
te | +0.032 |
Neg logit contributions | |
ately | -0.057 |
nation | -0.050 |
arsh | -0.045 |
34 | -0.045 |
upper | -0.044 |
Token | that |
Token ID | 326 |
Feature activation | +4.43e-02 |
Pos logit contributions | |
tul | +0.059 |
bal | +0.056 |
Notre | +0.056 |
cuc | +0.054 |
von | +0.053 |
Neg logit contributions | |
ately | -0.066 |
nation | -0.055 |
arsh | -0.051 |
ting | -0.049 |
34 | -0.049 |
Token | 's |
Token ID | 338 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
tul | +0.043 |
bal | +0.043 |
Notre | +0.041 |
cuc | +0.039 |
esh | +0.039 |
Neg logit contributions | |
ately | -0.054 |
nation | -0.046 |
arsh | -0.042 |
34 | -0.041 |
west | -0.040 |
Token | what |
Token ID | 644 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
tul | +0.051 |
cuc | +0.050 |
Notre | +0.049 |
imet | +0.049 |
von | +0.048 |
Neg logit contributions | |
ately | -0.029 |
nation | -0.027 |
usual | -0.024 |
yet | -0.022 |
first | -0.020 |
Token | through |
Token ID | 832 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
ately | +0.135 |
nation | +0.109 |
arsh | +0.102 |
ting | +0.102 |
upper | +0.096 |
Neg logit contributions | |
tul | -0.117 |
Notre | -0.115 |
bal | -0.114 |
von | -0.108 |
cuc | -0.107 |
Token | the |
Token ID | 262 |
Feature activation | -3.51e-03 |
Pos logit contributions | |
ately | +0.017 |
nation | +0.013 |
ting | +0.012 |
arsh | +0.011 |
upper | +0.011 |
Neg logit contributions | |
bal | -0.018 |
Notre | -0.017 |
tul | -0.017 |
Ark | -0.017 |
von | -0.017 |
Token | water |
Token ID | 1660 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
ately | +0.004 |
nation | +0.003 |
ting | +0.003 |
upper | +0.003 |
34 | +0.003 |
Neg logit contributions | |
bal | -0.004 |
tul | -0.004 |
spe | -0.004 |
von | -0.004 |
EA | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
ately | +0.016 |
nation | +0.013 |
ting | +0.012 |
arsh | +0.012 |
34 | +0.011 |
Neg logit contributions | |
tul | -0.015 |
Notre | -0.014 |
bal | -0.014 |
#### | -0.013 |
esh | -0.013 |
Token | delic |
Token ID | 8675 |
Feature activation | +8.45e-02 |
Pos logit contributions | |
ately | +0.022 |
nation | +0.019 |
arsh | +0.018 |
34 | +0.017 |
west | +0.017 |
Neg logit contributions | |
tul | -0.017 |
bal | -0.017 |
Notre | -0.016 |
cuc | -0.016 |
mo | -0.015 |
Token | ately |
Token ID | 1286 |
Feature activation | -3.35e-01 |
Pos logit contributions | |
tul | +0.173 |
bal | +0.172 |
Notre | +0.168 |
esh | +0.166 |
cuc | +0.165 |
Neg logit contributions | |
ately | -0.008 |
nation | +0.002 |
arsh | +0.011 |
west | +0.013 |
34 | +0.013 |
Token | scoop |
Token ID | 23529 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
ately | +0.390 |
nation | +0.316 |
arsh | +0.293 |
ting | +0.286 |
west | +0.261 |
Neg logit contributions | |
Notre | -0.357 |
tul | -0.350 |
bal | -0.345 |
von | -0.339 |
Ark | -0.331 |
Token | ed |
Token ID | 276 |
Feature activation | -9.65e-02 |
Pos logit contributions | |
bal | +0.033 |
tul | +0.033 |
Notre | +0.031 |
cuc | +0.031 |
te | +0.030 |
Neg logit contributions | |
ately | -0.026 |
nation | -0.022 |
arsh | -0.020 |
34 | -0.019 |
upper | -0.018 |
Token | up |
Token ID | 510 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
ately | +0.065 |
ting | +0.059 |
ted | +0.035 |
ing | +0.034 |
nation | +0.031 |
Neg logit contributions | |
Notre | -0.143 |
Ark | -0.141 |
ALTH | -0.139 |
majesty | -0.137 |
paramed | -0.135 |
Token | some |
Token ID | 617 |
Feature activation | -5.16e-02 |
Pos logit contributions | |
Notre | +0.037 |
Ark | +0.036 |
bal | +0.035 |
majesty | +0.035 |
EA | +0.035 |
Neg logit contributions | |
ately | -0.024 |
ting | -0.017 |
nation | -0.015 |
usual | -0.014 |
west | -0.014 |
Token | of |
Token ID | 286 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
ately | +0.057 |
nation | +0.042 |
ting | +0.042 |
upper | +0.039 |
usual | +0.036 |
Neg logit contributions | |
Ark | -0.057 |
Notre | -0.056 |
esh | -0.055 |
von | -0.053 |
bal | -0.053 |
Token | aded |
Token ID | 5286 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
tul | +0.014 |
bal | +0.014 |
Notre | +0.014 |
cuc | +0.013 |
esh | +0.013 |
Neg logit contributions | |
ately | -0.014 |
nation | -0.011 |
arsh | -0.010 |
34 | -0.010 |
upper | -0.010 |
Token | in |
Token ID | 287 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
ately | +0.015 |
ting | +0.012 |
rip | +0.010 |
nation | +0.010 |
fully | +0.010 |
Neg logit contributions | |
tul | -0.024 |
paramed | -0.023 |
styl | -0.023 |
bal | -0.023 |
stret | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | -4.41e-03 |
Pos logit contributions | |
bal | +0.017 |
tul | +0.016 |
Notre | +0.015 |
von | +0.015 |
dod | +0.015 |
Neg logit contributions | |
ately | -0.014 |
nation | -0.011 |
ting | -0.010 |
upper | -0.010 |
arsh | -0.010 |
Token | water |
Token ID | 1660 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
ately | +0.005 |
nation | +0.004 |
arsh | +0.004 |
upper | +0.004 |
ting | +0.004 |
Neg logit contributions | |
tul | -0.004 |
bal | -0.004 |
Notre | -0.004 |
esh | -0.004 |
von | -0.004 |
Token | delic |
Token ID | 8675 |
Feature activation | +8.40e-02 |
Pos logit contributions | |
ately | +0.015 |
nation | +0.013 |
arsh | +0.012 |
ting | +0.011 |
34 | +0.011 |
Neg logit contributions | |
tul | -0.012 |
bal | -0.012 |
Notre | -0.012 |
esh | -0.011 |
von | -0.011 |
Token | ately |
Token ID | 1286 |
Feature activation | -3.35e-01 |
Pos logit contributions | |
tul | +0.103 |
bal | +0.103 |
Notre | +0.100 |
cuc | +0.095 |
te | +0.095 |
Neg logit contributions | |
ately | -0.081 |
nation | -0.065 |
arsh | -0.059 |
34 | -0.056 |
upper | -0.055 |
Token | with |
Token ID | 351 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
ately | +0.339 |
nation | +0.288 |
arsh | +0.275 |
ting | +0.267 |
upper | +0.246 |
Neg logit contributions | |
bal | -0.371 |
tul | -0.369 |
Notre | -0.367 |
von | -0.365 |
paramed | -0.362 |
Token | his |
Token ID | 465 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
bal | +0.027 |
tul | +0.027 |
Notre | +0.027 |
esh | +0.026 |
von | +0.026 |
Neg logit contributions | |
ately | -0.023 |
nation | -0.019 |
upper | -0.018 |
ting | -0.017 |
west | -0.017 |
Token | fishing |
Token ID | 12478 |
Feature activation | -5.99e-02 |
Pos logit contributions | |
ately | +0.033 |
nation | +0.031 |
arsh | +0.028 |
ting | +0.028 |
west | +0.027 |
Neg logit contributions | |
bal | -0.019 |
tul | -0.018 |
mo | -0.018 |
Notre | -0.018 |
cuc | -0.018 |
Token | rod |
Token ID | 15299 |
Feature activation | -3.86e-02 |
Pos logit contributions | |
ately | +0.060 |
nation | +0.048 |
upper | +0.045 |
arsh | +0.045 |
34 | +0.042 |
Neg logit contributions | |
tul | -0.071 |
bal | -0.068 |
Notre | -0.067 |
thirteen | -0.066 |
esh | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
ately | +0.038 |
nation | +0.029 |
arsh | +0.028 |
upper | +0.027 |
ting | +0.026 |
Neg logit contributions | |
tul | -0.047 |
Notre | -0.046 |
Ark | -0.044 |
von | -0.044 |
bal | -0.044 |
Token | day |
Token ID | 1110 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
ately | +0.033 |
nation | +0.027 |
arsh | +0.026 |
ting | +0.024 |
west | +0.024 |
Neg logit contributions | |
tul | -0.034 |
bal | -0.033 |
Notre | -0.032 |
cuc | -0.031 |
Ark | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +2.06e-03 |
Pos logit contributions | |
ately | +0.025 |
nation | +0.022 |
arsh | +0.020 |
34 | +0.019 |
upper | +0.019 |
Neg logit contributions | |
tul | -0.019 |
bal | -0.019 |
Notre | -0.018 |
cuc | -0.017 |
te | -0.017 |
Token | They |
Token ID | 1119 |
Feature activation | +9.08e-03 |
Pos logit contributions | |
tul | +0.002 |
bal | +0.002 |
Notre | +0.002 |
esh | +0.002 |
cuc | +0.002 |
Neg logit contributions | |
ately | -0.002 |
nation | -0.002 |
arsh | -0.002 |
ting | -0.002 |
west | -0.002 |
Token | worked |
Token ID | 3111 |
Feature activation | +5.41e-02 |
Pos logit contributions | |
tul | +0.009 |
bal | +0.009 |
Notre | +0.008 |
esh | +0.008 |
cuc | +0.008 |
Neg logit contributions | |
ately | -0.011 |
nation | -0.009 |
arsh | -0.009 |
west | -0.008 |
upper | -0.008 |
Token | delic |
Token ID | 8675 |
Feature activation | +8.71e-02 |
Pos logit contributions | |
tul | +0.066 |
bal | +0.063 |
Notre | +0.062 |
cuc | +0.059 |
esh | +0.059 |
Neg logit contributions | |
ately | -0.055 |
nation | -0.045 |
arsh | -0.041 |
34 | -0.038 |
ting | -0.038 |
Token | ately |
Token ID | 1286 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
bal | +0.170 |
tul | +0.168 |
esh | +0.162 |
Notre | +0.162 |
thirteen | +0.161 |
Neg logit contributions | |
ately | -0.007 |
nation | -0.007 |
west | +0.004 |
upper | +0.005 |
34 | +0.007 |
Token | and |
Token ID | 290 |
Feature activation | -8.10e-03 |
Pos logit contributions | |
ately | +0.424 |
nation | +0.363 |
arsh | +0.342 |
ting | +0.324 |
west | +0.320 |
Neg logit contributions | |
tul | -0.305 |
bal | -0.293 |
Notre | -0.289 |
cuc | -0.275 |
von | -0.267 |
Token | carefully |
Token ID | 7773 |
Feature activation | -6.87e-02 |
Pos logit contributions | |
ately | +0.011 |
nation | +0.010 |
34 | +0.010 |
arsh | +0.010 |
upper | +0.009 |
Neg logit contributions | |
bal | -0.006 |
tul | -0.006 |
mo | -0.006 |
� | -0.006 |
Notre | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | -7.32e-03 |
Pos logit contributions | |
ately | +0.089 |
nation | +0.076 |
arsh | +0.071 |
34 | +0.068 |
ting | +0.068 |
Neg logit contributions | |
tul | -0.061 |
bal | -0.060 |
Notre | -0.058 |
cuc | -0.055 |
esh | -0.055 |
Token | nothing |
Token ID | 2147 |
Feature activation | +4.61e-02 |
Pos logit contributions | |
ately | +0.009 |
nation | +0.008 |
arsh | +0.008 |
34 | +0.007 |
upper | +0.007 |
Neg logit contributions | |
bal | -0.007 |
tul | -0.007 |
Notre | -0.006 |
mo | -0.006 |
te | -0.006 |
Token | broke |
Token ID | 6265 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
tul | +0.050 |
Notre | +0.047 |
bal | +0.047 |
cuc | +0.045 |
Ark | +0.045 |
Neg logit contributions | |
ately | -0.053 |
nation | -0.044 |
arsh | -0.040 |
ting | -0.038 |
west | -0.037 |
Token | of |
Token ID | 286 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
bal | +0.051 |
tul | +0.051 |
Notre | +0.048 |
cuc | +0.047 |
te | +0.046 |
Neg logit contributions | |
ately | -0.065 |
nation | -0.057 |
arsh | -0.052 |
34 | -0.050 |
west | -0.050 |
Token | sprink |
Token ID | 27562 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
ately | +0.016 |
nation | +0.013 |
upper | +0.012 |
west | +0.012 |
ting | +0.012 |
Neg logit contributions | |
bal | -0.017 |
Notre | -0.017 |
esh | -0.017 |
spe | -0.017 |
tul | -0.016 |
Token | les |
Token ID | 829 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
bal | +0.166 |
tul | +0.160 |
cuc | +0.155 |
sprink | +0.151 |
te | +0.149 |
Neg logit contributions | |
ately | -0.232 |
nation | -0.207 |
arsh | -0.192 |
34 | -0.187 |
upper | -0.184 |
Token | and |
Token ID | 290 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
ately | +0.029 |
ting | +0.022 |
arsh | +0.017 |
fully | +0.017 |
owing | +0.016 |
Neg logit contributions | |
Notre | -0.048 |
majesty | -0.047 |
Ark | -0.047 |
esh | -0.047 |
tul | -0.046 |
Token | delic |
Token ID | 8675 |
Feature activation | +8.85e-02 |
Pos logit contributions | |
ately | +0.012 |
upper | +0.009 |
ting | +0.008 |
usual | +0.008 |
arsh | +0.008 |
Neg logit contributions | |
Notre | -0.013 |
esh | -0.013 |
stret | -0.012 |
#### | -0.012 |
Ark | -0.012 |
Token | ately |
Token ID | 1286 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
tul | +0.142 |
bal | +0.141 |
Notre | +0.137 |
cuc | +0.134 |
esh | +0.133 |
Neg logit contributions | |
ately | -0.051 |
nation | -0.036 |
arsh | -0.029 |
34 | -0.026 |
west | -0.025 |
Token | spread |
Token ID | 4104 |
Feature activation | -5.12e-02 |
Pos logit contributions | |
ately | +0.405 |
nation | +0.340 |
arsh | +0.332 |
west | +0.303 |
ting | +0.303 |
Neg logit contributions | |
Notre | -0.313 |
bal | -0.296 |
tul | -0.294 |
von | -0.286 |
#### | -0.282 |
Token | them |
Token ID | 606 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
ately | +0.027 |
ting | +0.016 |
usual | +0.015 |
other | +0.015 |
rip | +0.014 |
Neg logit contributions | |
Notre | -0.084 |
esh | -0.083 |
lyn | -0.083 |
majesty | -0.081 |
MacDonald | -0.080 |
Token | over |
Token ID | 625 |
Feature activation | +5.03e-02 |
Pos logit contributions | |
tul | +0.028 |
Notre | +0.028 |
paramed | +0.027 |
Ark | +0.027 |
pap | +0.027 |
Neg logit contributions | |
ately | -0.021 |
nation | -0.017 |
usual | -0.015 |
ting | -0.014 |
west | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | +6.27e-03 |
Pos logit contributions | |
bal | +0.054 |
Notre | +0.053 |
tul | +0.053 |
esh | +0.050 |
Ark | +0.049 |
Neg logit contributions | |
ately | -0.058 |
nation | -0.045 |
arsh | -0.042 |
ting | -0.042 |
34 | -0.041 |
Token | ice |
Token ID | 4771 |
Feature activation | -3.92e-02 |
Pos logit contributions | |
esh | +0.008 |
bal | +0.008 |
tin | +0.008 |
Notre | +0.008 |
unbeat | +0.008 |
Neg logit contributions | |
ately | -0.005 |
upper | -0.005 |
usual | -0.004 |
nation | -0.004 |
west | -0.003 |
Token | work |
Token ID | 670 |
Feature activation | +5.58e-02 |
Pos logit contributions | |
tul | +0.009 |
bal | +0.009 |
Notre | +0.009 |
cuc | +0.008 |
esh | +0.008 |
Neg logit contributions | |
ately | -0.016 |
nation | -0.014 |
arsh | -0.013 |
34 | -0.013 |
west | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -3.54e-03 |
Pos logit contributions | |
tul | +0.063 |
bal | +0.060 |
Notre | +0.059 |
cuc | +0.056 |
von | +0.056 |
Neg logit contributions | |
ately | -0.062 |
nation | -0.051 |
arsh | -0.047 |
ting | -0.044 |
34 | -0.044 |
Token | carefully |
Token ID | 7773 |
Feature activation | -7.14e-02 |
Pos logit contributions | |
ately | +0.004 |
nation | +0.003 |
ting | +0.003 |
arsh | +0.003 |
34 | +0.002 |
Neg logit contributions | |
tul | -0.004 |
bal | -0.004 |
Notre | -0.004 |
von | -0.004 |
esh | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
ately | +0.091 |
nation | +0.077 |
arsh | +0.073 |
ting | +0.069 |
upper | +0.068 |
Neg logit contributions | |
tul | -0.066 |
bal | -0.065 |
Notre | -0.064 |
von | -0.060 |
esh | -0.058 |
Token | delic |
Token ID | 8675 |
Feature activation | +8.94e-02 |
Pos logit contributions | |
ately | +0.017 |
nation | +0.015 |
arsh | +0.014 |
34 | +0.014 |
upper | +0.014 |
Neg logit contributions | |
bal | -0.012 |
tul | -0.012 |
mo | -0.011 |
Notre | -0.011 |
cuc | -0.011 |
Token | ately |
Token ID | 1286 |
Feature activation | -3.28e-01 |
Pos logit contributions | |
tul | +0.183 |
bal | +0.182 |
Notre | +0.177 |
cuc | +0.175 |
esh | +0.175 |
Neg logit contributions | |
ately | -0.008 |
nation | +0.002 |
arsh | +0.012 |
west | +0.013 |
upper | +0.014 |
Token | repairing |
Token ID | 39005 |
Feature activation | +7.11e-03 |
Pos logit contributions | |
ately | +0.414 |
nation | +0.353 |
arsh | +0.331 |
ting | +0.313 |
upper | +0.308 |
Neg logit contributions | |
tul | -0.308 |
bal | -0.300 |
Notre | -0.295 |
cuc | -0.277 |
von | -0.274 |
Token | the |
Token ID | 262 |
Feature activation | -2.14e-03 |
Pos logit contributions | |
tul | +0.005 |
marsh | +0.004 |
cuc | +0.004 |
bal | +0.004 |
mo | +0.004 |
Neg logit contributions | |
ately | -0.010 |
nation | -0.010 |
arsh | -0.009 |
ancer | -0.009 |
upper | -0.009 |
Token | butterfly |
Token ID | 35113 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
ately | +0.002 |
nation | +0.002 |
upper | +0.002 |
ting | +0.002 |
34 | +0.002 |
Neg logit contributions | |
esh | -0.002 |
bal | -0.002 |
tul | -0.002 |
Notre | -0.002 |
Ark | -0.002 |
Token | 's |
Token ID | 338 |
Feature activation | +7.11e-03 |
Pos logit contributions | |
ately | +0.020 |
nation | +0.017 |
arsh | +0.015 |
ting | +0.015 |
owing | +0.013 |
Neg logit contributions | |
tul | -0.026 |
Notre | -0.026 |
Ark | -0.025 |
von | -0.024 |
thirteen | -0.024 |
Token | wings |
Token ID | 12098 |
Feature activation | +8.23e-02 |
Pos logit contributions | |
bal | +0.007 |
tul | +0.007 |
Notre | +0.007 |
Ark | +0.007 |
von | +0.007 |
Neg logit contributions | |
ately | -0.009 |
nation | -0.007 |
upper | -0.007 |
arsh | -0.006 |
34 | -0.006 |
Token | uff |
Token ID | 1648 |
Feature activation | -3.95e-02 |
Pos logit contributions | |
bal | +0.095 |
tul | +0.091 |
esh | +0.089 |
von | +0.088 |
Notre | +0.087 |
Neg logit contributions | |
ately | -0.103 |
nation | -0.091 |
arsh | -0.081 |
34 | -0.081 |
west | -0.080 |
Token | led |
Token ID | 992 |
Feature activation | -9.20e-02 |
Pos logit contributions | |
ately | +0.040 |
nation | +0.039 |
34 | +0.036 |
upper | +0.034 |
west | +0.034 |
Neg logit contributions | |
bal | -0.041 |
te | -0.037 |
esh | -0.037 |
tul | -0.036 |
cut | -0.036 |
Token | his |
Token ID | 465 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
ately | +0.093 |
ting | +0.071 |
nation | +0.070 |
upper | +0.062 |
west | +0.058 |
Neg logit contributions | |
Notre | -0.109 |
tul | -0.108 |
#### | -0.104 |
Ark | -0.104 |
bal | -0.103 |
Token | hair |
Token ID | 4190 |
Feature activation | +4.34e-02 |
Pos logit contributions | |
ately | +0.013 |
nation | +0.010 |
upper | +0.010 |
arsh | +0.009 |
34 | +0.009 |
Neg logit contributions | |
tul | -0.010 |
bal | -0.010 |
Notre | -0.010 |
thirteen | -0.010 |
esh | -0.010 |
Token | affection |
Token ID | 17696 |
Feature activation | +5.43e-03 |
Pos logit contributions | |
tul | +0.041 |
bal | +0.041 |
Notre | +0.039 |
cuc | +0.037 |
esh | +0.037 |
Neg logit contributions | |
ately | -0.054 |
nation | -0.046 |
arsh | -0.043 |
34 | -0.041 |
upper | -0.041 |
Token | ately |
Token ID | 1286 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
bal | +0.005 |
cloth | +0.005 |
ener | +0.005 |
pol | +0.005 |
daddy | +0.005 |
Neg logit contributions | |
nation | -0.006 |
west | -0.006 |
Shelter | -0.006 |
weet | -0.006 |
Offic | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +7.24e-03 |
Pos logit contributions | |
ately | +0.353 |
nation | +0.300 |
arsh | +0.271 |
ting | +0.252 |
upper | +0.248 |
Neg logit contributions | |
tul | -0.367 |
Notre | -0.363 |
bal | -0.347 |
von | -0.339 |
mo | -0.339 |
Token | |
Token ID | 220 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
tul | +0.007 |
bal | +0.007 |
Notre | +0.007 |
cuc | +0.007 |
esh | +0.007 |
Neg logit contributions | |
ately | -0.008 |
nation | -0.007 |
arsh | -0.007 |
34 | -0.006 |
west | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -3.86e-03 |
Pos logit contributions | |
bal | +0.018 |
tul | +0.017 |
Notre | +0.017 |
esh | +0.016 |
von | +0.016 |
Neg logit contributions | |
ately | -0.022 |
nation | -0.019 |
arsh | -0.017 |
34 | -0.017 |
upper | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | -3.13e-03 |
Pos logit contributions | |
ately | +0.006 |
nation | +0.005 |
34 | +0.005 |
ting | +0.005 |
upper | +0.005 |
Neg logit contributions | |
� | -0.003 |
bal | -0.002 |
tul | -0.002 |
von | -0.002 |
Ark | -0.002 |
Token | The |
Token ID | 464 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
ately | +0.003 |
nation | +0.003 |
arsh | +0.002 |
west | +0.002 |
upper | +0.002 |
Neg logit contributions | |
tul | -0.004 |
Notre | -0.004 |
bal | -0.003 |
imag | -0.003 |
cuc | -0.003 |
Token | little |
Token ID | 1310 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
bal | +0.021 |
tul | +0.021 |
Notre | +0.021 |
esh | +0.021 |
cuc | +0.019 |
Neg logit contributions | |
ately | -0.029 |
nation | -0.024 |
upper | -0.022 |
arsh | -0.022 |
34 | -0.022 |
Token | girl |
Token ID | 2576 |
Feature activation | +5.36e-02 |
Pos logit contributions | |
ately | +0.021 |
ting | +0.016 |
upper | +0.016 |
nation | +0.014 |
34 | +0.013 |
Neg logit contributions | |
esh | -0.028 |
Notre | -0.027 |
marsh | -0.025 |
stret | -0.025 |
Ark | -0.025 |
Token | 's |
Token ID | 338 |
Feature activation | +3.29e-02 |
Pos logit contributions | |
Notre | +0.070 |
Ark | +0.065 |
tul | +0.065 |
cuc | +0.064 |
#### | +0.064 |
Neg logit contributions | |
ately | -0.051 |
owing | -0.040 |
nation | -0.039 |
ting | -0.035 |
arsh | -0.034 |
Token | shoulders |
Token ID | 12450 |
Feature activation | -4.73e-02 |
Pos logit contributions | |
Notre | +0.038 |
tul | +0.037 |
cuc | +0.036 |
esh | +0.036 |
bal | +0.035 |
Neg logit contributions | |
ately | -0.035 |
nation | -0.028 |
upper | -0.028 |
arsh | -0.025 |
34 | -0.024 |
Token | affection |
Token ID | 17696 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
ately | +0.061 |
nation | +0.052 |
arsh | +0.049 |
34 | +0.047 |
west | +0.047 |
Neg logit contributions | |
bal | -0.042 |
tul | -0.042 |
Notre | -0.039 |
te | -0.038 |
cuc | -0.037 |
Token | ately |
Token ID | 1286 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
bal | +0.017 |
cloth | +0.016 |
tul | +0.015 |
esh | +0.015 |
pter | +0.015 |
Neg logit contributions | |
nation | -0.018 |
west | -0.017 |
arsh | -0.015 |
uch | -0.015 |
film | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
ately | +0.312 |
nation | +0.268 |
arsh | +0.240 |
ting | +0.219 |
upper | +0.208 |
Neg logit contributions | |
Notre | -0.392 |
tul | -0.383 |
von | -0.364 |
pap | -0.360 |
imag | -0.358 |
Token |
|
Token ID | 198 |
Feature activation | +9.10e-03 |
Pos logit contributions | |
bal | +0.017 |
tul | +0.017 |
Notre | +0.016 |
� | +0.016 |
von | +0.015 |
Neg logit contributions | |
ately | -0.027 |
nation | -0.023 |
34 | -0.021 |
arsh | -0.021 |
upper | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | +9.91e-03 |
Pos logit contributions | |
tul | +0.009 |
bal | +0.009 |
Notre | +0.008 |
esh | +0.008 |
von | +0.008 |
Neg logit contributions | |
ately | -0.011 |
nation | -0.010 |
arsh | -0.009 |
34 | -0.009 |
ting | -0.009 |
Token | Sarah |
Token ID | 29284 |
Feature activation | -3.45e-03 |
Pos logit contributions | |
tul | +0.012 |
Notre | +0.012 |
cuc | +0.011 |
bal | +0.011 |
te | +0.011 |
Neg logit contributions | |
ately | -0.010 |
nation | -0.009 |
arsh | -0.008 |
west | -0.007 |
upper | -0.007 |
Token | smiled |
Token ID | 13541 |
Feature activation | +4.44e-02 |
Pos logit contributions | |
ately | +0.003 |
nation | +0.002 |
arsh | +0.002 |
upper | +0.002 |
ting | +0.002 |
Neg logit contributions | |
Notre | -0.005 |
tul | -0.005 |
stret | -0.004 |
cuc | -0.004 |
ALTH | -0.004 |
Token | decided |
Token ID | 3066 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
ately | +0.034 |
nation | +0.023 |
ting | +0.022 |
usual | +0.021 |
arsh | +0.020 |
Neg logit contributions | |
icol | -0.054 |
esh | -0.053 |
tul | -0.053 |
Notre | -0.053 |
Ark | -0.052 |
Token | to |
Token ID | 284 |
Feature activation | +4.53e-03 |
Pos logit contributions | |
ately | +0.024 |
nation | +0.019 |
arsh | +0.017 |
ting | +0.017 |
usual | +0.016 |
Neg logit contributions | |
tul | -0.030 |
bal | -0.029 |
Notre | -0.029 |
von | -0.028 |
Ark | -0.028 |
Token | settle |
Token ID | 12259 |
Feature activation | -2.33e-02 |
Pos logit contributions | |
bal | +0.005 |
tul | +0.005 |
Notre | +0.005 |
cuc | +0.005 |
thirteen | +0.005 |
Neg logit contributions | |
ately | -0.005 |
nation | -0.004 |
arsh | -0.003 |
west | -0.003 |
upper | -0.003 |
Token | into |
Token ID | 656 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
ately | +0.030 |
nation | +0.027 |
arsh | +0.026 |
west | +0.025 |
upper | +0.025 |
Neg logit contributions | |
tul | -0.019 |
bal | -0.019 |
Notre | -0.018 |
mo | -0.018 |
te | -0.017 |
Token | hiber |
Token ID | 46681 |
Feature activation | -7.29e-03 |
Pos logit contributions | |
tul | +0.015 |
bal | +0.014 |
Notre | +0.014 |
te | +0.013 |
mo | +0.013 |
Neg logit contributions | |
ately | -0.024 |
nation | -0.021 |
arsh | -0.019 |
34 | -0.019 |
upper | -0.018 |
Token | nation |
Token ID | 25729 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
ately | +0.010 |
34 | +0.007 |
ancer | +0.007 |
upper | +0.007 |
arsh | +0.007 |
Neg logit contributions | |
esh | -0.007 |
� | -0.007 |
Notre | -0.006 |
unbeat | -0.006 |
stret | -0.006 |
Token | in |
Token ID | 287 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
ately | +0.360 |
nation | +0.309 |
arsh | +0.289 |
34 | +0.279 |
west | +0.277 |
Neg logit contributions | |
tul | -0.259 |
bal | -0.257 |
Notre | -0.245 |
esh | -0.234 |
cuc | -0.233 |
Token | their |
Token ID | 511 |
Feature activation | -1.26e-03 |
Pos logit contributions | |
tul | +0.011 |
bal | +0.010 |
Notre | +0.010 |
mo | +0.009 |
te | +0.009 |
Neg logit contributions | |
ately | -0.017 |
nation | -0.014 |
arsh | -0.014 |
34 | -0.013 |
upper | -0.013 |
Token | den |
Token ID | 2853 |
Feature activation | +2.93e-03 |
Pos logit contributions | |
ately | +0.001 |
nation | +0.001 |
upper | +0.001 |
usual | +0.001 |
34 | +0.001 |
Neg logit contributions | |
bal | -0.001 |
tul | -0.001 |
cuc | -0.001 |
Ark | -0.001 |
esh | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -6.86e-03 |
Pos logit contributions | |
tul | +0.003 |
bal | +0.003 |
Notre | +0.003 |
cuc | +0.003 |
esh | +0.003 |
Neg logit contributions | |
ately | -0.004 |
nation | -0.003 |
arsh | -0.003 |
34 | -0.003 |
upper | -0.003 |
Token | They |
Token ID | 1119 |
Feature activation | -1.76e-03 |
Pos logit contributions | |
ately | +0.007 |
nation | +0.006 |
arsh | +0.006 |
west | +0.005 |
34 | +0.005 |
Neg logit contributions | |
tul | -0.008 |
bal | -0.008 |
Notre | -0.007 |
esh | -0.007 |
cuc | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -5.66e-03 |
Pos logit contributions | |
ately | +0.048 |
upper | +0.041 |
nation | +0.040 |
ancer | +0.040 |
arsh | +0.039 |
Neg logit contributions | |
bal | -0.020 |
� | -0.019 |
te | -0.017 |
tul | -0.016 |
gn | -0.016 |
Token | " |
Token ID | 366 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
ately | +0.007 |
nation | +0.007 |
arsh | +0.006 |
ting | +0.006 |
34 | +0.006 |
Neg logit contributions | |
tul | -0.005 |
bal | -0.005 |
Notre | -0.005 |
von | -0.004 |
te | -0.004 |
Token | It |
Token ID | 1026 |
Feature activation | -3.67e-02 |
Pos logit contributions | |
ately | +0.037 |
nation | +0.030 |
arsh | +0.029 |
34 | +0.027 |
ting | +0.027 |
Neg logit contributions | |
bal | -0.028 |
tul | -0.028 |
Notre | -0.027 |
thirteen | -0.025 |
� | -0.025 |
Token | says |
Token ID | 1139 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
ately | +0.049 |
nation | +0.042 |
arsh | +0.040 |
34 | +0.038 |
upper | +0.038 |
Neg logit contributions | |
tul | -0.031 |
bal | -0.031 |
Notre | -0.029 |
cuc | -0.028 |
te | -0.027 |
Token | ... |
Token ID | 986 |
Feature activation | +6.58e-02 |
Pos logit contributions | |
tul | +0.013 |
bal | +0.012 |
Notre | +0.012 |
cuc | +0.011 |
esh | +0.011 |
Neg logit contributions | |
ately | -0.020 |
nation | -0.017 |
arsh | -0.016 |
34 | -0.015 |
west | -0.015 |
Token | nation |
Token ID | 25729 |
Feature activation | -2.83e-01 |
Pos logit contributions | |
� | +0.060 |
bal | +0.058 |
thirteen | +0.052 |
tul | +0.051 |
sprink | +0.049 |
Neg logit contributions | |
ately | -0.089 |
nation | -0.073 |
upper | -0.069 |
arsh | -0.068 |
arter | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | -7.15e-03 |
Pos logit contributions | |
ately | +0.344 |
nation | +0.297 |
arsh | +0.275 |
34 | +0.265 |
west | +0.260 |
Neg logit contributions | |
bal | -0.276 |
tul | -0.276 |
Notre | -0.264 |
cuc | -0.253 |
mo | -0.249 |
Token | I |
Token ID | 314 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
ately | +0.010 |
nation | +0.009 |
arsh | +0.008 |
usual | +0.008 |
owing | +0.008 |
Neg logit contributions | |
tul | -0.006 |
� | -0.006 |
bal | -0.006 |
Notre | -0.005 |
thirteen | -0.005 |
Token | think |
Token ID | 892 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
ately | +0.015 |
nation | +0.013 |
arsh | +0.012 |
west | +0.011 |
upper | +0.011 |
Neg logit contributions | |
tul | -0.014 |
bal | -0.013 |
Notre | -0.013 |
esh | -0.013 |
cuc | -0.013 |
Token | it |
Token ID | 340 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
bal | +0.036 |
tul | +0.035 |
� | +0.035 |
sprink | +0.032 |
thirteen | +0.032 |
Neg logit contributions | |
ately | -0.063 |
nation | -0.057 |
arsh | -0.052 |
upper | -0.052 |
ting | -0.050 |
Token | means |
Token ID | 1724 |
Feature activation | -4.42e-02 |
Pos logit contributions | |
ately | +0.049 |
nation | +0.040 |
arsh | +0.039 |
ting | +0.038 |
upper | +0.038 |
Neg logit contributions | |
bal | -0.023 |
tul | -0.022 |
Notre | -0.022 |
� | -0.021 |
te | -0.021 |
Token | didn |
Token ID | 1422 |
Feature activation | +2.47e-02 |
Pos logit contributions | |
ately | +0.018 |
nation | +0.015 |
arsh | +0.013 |
west | +0.012 |
upper | +0.012 |
Neg logit contributions | |
tul | -0.021 |
bal | -0.020 |
Notre | -0.020 |
cuc | -0.020 |
thirteen | -0.020 |
Token | 't |
Token ID | 470 |
Feature activation | -3.71e-02 |
Pos logit contributions | |
bal | +0.025 |
� | +0.024 |
tul | +0.024 |
Notre | +0.023 |
esh | +0.023 |
Neg logit contributions | |
ately | -0.030 |
nation | -0.024 |
arsh | -0.022 |
34 | -0.022 |
ting | -0.022 |
Token | understand |
Token ID | 1833 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
ately | +0.043 |
nation | +0.038 |
arsh | +0.035 |
west | +0.033 |
upper | +0.033 |
Neg logit contributions | |
tul | -0.037 |
bal | -0.036 |
Notre | -0.035 |
cuc | -0.035 |
esh | -0.034 |
Token | what |
Token ID | 644 |
Feature activation | +5.41e-04 |
Pos logit contributions | |
ately | +0.030 |
nation | +0.027 |
arsh | +0.025 |
west | +0.025 |
upper | +0.024 |
Neg logit contributions | |
tul | -0.011 |
bal | -0.011 |
Notre | -0.010 |
cuc | -0.009 |
te | -0.009 |
Token | " |
Token ID | 366 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
bal | +0.000 |
tul | +0.000 |
Notre | +0.000 |
te | +0.000 |
cuc | +0.000 |
Neg logit contributions | |
ately | -0.001 |
nation | -0.001 |
arsh | -0.001 |
34 | -0.001 |
upper | -0.001 |
Token | nation |
Token ID | 25729 |
Feature activation | -2.73e-01 |
Pos logit contributions | |
ately | +0.026 |
arter | +0.025 |
34 | +0.024 |
arsh | +0.023 |
Shelter | +0.023 |
Neg logit contributions | |
bal | -0.017 |
bel | -0.014 |
� | -0.013 |
spe | -0.013 |
ners | -0.013 |
Token | " |
Token ID | 1 |
Feature activation | +4.29e-02 |
Pos logit contributions | |
ately | +0.384 |
nation | +0.331 |
arsh | +0.307 |
upper | +0.307 |
34 | +0.303 |
Neg logit contributions | |
tul | -0.209 |
bal | -0.205 |
Notre | -0.193 |
� | -0.188 |
esh | -0.186 |
Token | meant |
Token ID | 4001 |
Feature activation | -7.94e-02 |
Pos logit contributions | |
bal | +0.022 |
bra | +0.014 |
brush | +0.011 |
even | +0.011 |
dri | +0.010 |
Neg logit contributions | |
arter | -0.081 |
Shelter | -0.074 |
ately | -0.071 |
charged | -0.071 |
arsh | -0.070 |
Token | , |
Token ID | 11 |
Feature activation | +3.42e-03 |
Pos logit contributions | |
ately | +0.092 |
nation | +0.075 |
arsh | +0.072 |
34 | +0.068 |
ting | +0.067 |
Neg logit contributions | |
tul | -0.086 |
bal | -0.082 |
Notre | -0.081 |
cuc | -0.078 |
thirteen | -0.077 |
Token | so |
Token ID | 523 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
tul | +0.004 |
bal | +0.004 |
Notre | +0.003 |
cuc | +0.003 |
esh | +0.003 |
Neg logit contributions | |
ately | -0.004 |
nation | -0.003 |
arsh | -0.003 |
34 | -0.003 |
west | -0.003 |
Token | her |
Token ID | 607 |
Feature activation | +3.18e-03 |
Pos logit contributions | |
tul | +0.021 |
bal | +0.020 |
Notre | +0.020 |
cuc | +0.019 |
esh | +0.018 |
Neg logit contributions | |
ately | -0.031 |
nation | -0.026 |
arsh | -0.025 |
34 | -0.024 |
upper | -0.023 |
Token | go |
Token ID | 467 |
Feature activation | +4.25e-03 |
Pos logit contributions | |
bal | +0.004 |
tul | +0.004 |
Notre | +0.004 |
esh | +0.004 |
mo | +0.004 |
Neg logit contributions | |
ately | -0.006 |
nation | -0.005 |
arsh | -0.005 |
34 | -0.005 |
ting | -0.005 |
Token | to |
Token ID | 284 |
Feature activation | +3.62e-03 |
Pos logit contributions | |
tul | +0.005 |
bal | +0.005 |
Notre | +0.005 |
cuc | +0.004 |
thirteen | +0.004 |
Neg logit contributions | |
ately | -0.005 |
nation | -0.004 |
arsh | -0.004 |
ting | -0.003 |
west | -0.003 |
Token | bed |
Token ID | 3996 |
Feature activation | -5.34e-02 |
Pos logit contributions | |
tul | +0.003 |
bal | +0.003 |
mo | +0.003 |
Notre | +0.002 |
te | +0.002 |
Neg logit contributions | |
ately | -0.005 |
nation | -0.005 |
34 | -0.004 |
arsh | -0.004 |
ancer | -0.004 |
Token | without |
Token ID | 1231 |
Feature activation | -3.99e-02 |
Pos logit contributions | |
ately | +0.069 |
nation | +0.060 |
west | +0.055 |
arsh | +0.055 |
34 | +0.055 |
Neg logit contributions | |
bal | -0.047 |
tul | -0.045 |
te | -0.043 |
Notre | -0.043 |
mo | -0.042 |
Token | regret |
Token ID | 13721 |
Feature activation | +7.62e-02 |
Pos logit contributions | |
ately | +0.049 |
nation | +0.042 |
arsh | +0.039 |
34 | +0.038 |
west | +0.037 |
Neg logit contributions | |
tul | -0.038 |
bal | -0.038 |
Notre | -0.036 |
cuc | -0.035 |
esh | -0.034 |
Token | ting |
Token ID | 889 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
tul | +0.072 |
bal | +0.072 |
Notre | +0.068 |
esh | +0.066 |
cuc | +0.065 |
Neg logit contributions | |
ately | -0.092 |
nation | -0.080 |
arsh | -0.073 |
upper | -0.071 |
west | -0.071 |
Token | her |
Token ID | 607 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
ately | +0.267 |
ting | +0.228 |
nation | +0.225 |
arsh | +0.202 |
upper | +0.184 |
Neg logit contributions | |
tul | -0.314 |
Notre | -0.302 |
cuc | -0.296 |
Ark | -0.289 |
bal | -0.286 |
Token | decision |
Token ID | 2551 |
Feature activation | +3.32e-02 |
Pos logit contributions | |
ately | +0.015 |
nation | +0.013 |
upper | +0.011 |
arsh | +0.011 |
34 | +0.011 |
Neg logit contributions | |
tul | -0.014 |
Notre | -0.014 |
cuc | -0.013 |
thirteen | -0.013 |
Ark | -0.013 |
Token | anymore |
Token ID | 7471 |
Feature activation | -8.41e-02 |
Pos logit contributions | |
tul | +0.035 |
Notre | +0.033 |
bal | +0.033 |
cuc | +0.032 |
te | +0.031 |
Neg logit contributions | |
ately | -0.039 |
nation | -0.032 |
arsh | -0.030 |
ting | -0.029 |
34 | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -4.32e-03 |
Pos logit contributions | |
ately | +0.091 |
nation | +0.074 |
arsh | +0.070 |
ting | +0.063 |
34 | +0.063 |
Neg logit contributions | |
tul | -0.098 |
Notre | -0.094 |
bal | -0.091 |
cuc | -0.090 |
thirteen | -0.089 |
Token | And |
Token ID | 843 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
ately | +0.004 |
arsh | +0.004 |
nation | +0.004 |
west | +0.003 |
upper | +0.003 |
Neg logit contributions | |
tul | -0.005 |
Notre | -0.005 |
bal | -0.005 |
mo | -0.005 |
te | -0.005 |
Token | tried |
Token ID | 3088 |
Feature activation | -5.24e-02 |
Pos logit contributions | |
ately | +0.008 |
nation | +0.007 |
arsh | +0.006 |
ting | +0.006 |
west | +0.006 |
Neg logit contributions | |
tul | -0.009 |
bal | -0.008 |
Notre | -0.008 |
thirteen | -0.008 |
cuc | -0.008 |
Token | speaking |
Token ID | 5486 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
ately | +0.058 |
nation | +0.049 |
arsh | +0.045 |
ting | +0.044 |
34 | +0.042 |
Neg logit contributions | |
tul | -0.055 |
bal | -0.054 |
Notre | -0.054 |
cuc | -0.052 |
esh | -0.051 |
Token | softly |
Token ID | 26625 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
ately | +0.136 |
nation | +0.117 |
arsh | +0.110 |
34 | +0.106 |
west | +0.106 |
Neg logit contributions | |
bal | -0.093 |
tul | -0.093 |
Notre | -0.087 |
cuc | -0.082 |
esh | -0.082 |
Token | and |
Token ID | 290 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
ately | +0.153 |
nation | +0.133 |
arsh | +0.124 |
34 | +0.120 |
west | +0.119 |
Neg logit contributions | |
tul | -0.106 |
bal | -0.106 |
Notre | -0.100 |
cuc | -0.095 |
te | -0.094 |
Token | pat |
Token ID | 1458 |
Feature activation | +6.74e-02 |
Pos logit contributions | |
ately | +0.018 |
nation | +0.015 |
arsh | +0.014 |
34 | +0.014 |
west | +0.014 |
Neg logit contributions | |
tul | -0.014 |
bal | -0.013 |
Notre | -0.013 |
cuc | -0.012 |
te | -0.012 |
Token | ting |
Token ID | 889 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
� | +0.064 |
bal | +0.063 |
marsh | +0.060 |
tul | +0.058 |
� | +0.056 |
Neg logit contributions | |
ately | -0.077 |
34 | -0.074 |
arsh | -0.071 |
arter | -0.071 |
upper | -0.068 |
Token | Daisy |
Token ID | 40355 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
ately | +0.254 |
ting | +0.231 |
nation | +0.220 |
arsh | +0.185 |
west | +0.182 |
Neg logit contributions | |
tul | -0.303 |
Notre | -0.296 |
bal | -0.290 |
Ark | -0.289 |
mo | -0.286 |
Token | on |
Token ID | 319 |
Feature activation | -6.75e-02 |
Pos logit contributions | |
ately | +0.018 |
nation | +0.015 |
ting | +0.014 |
arsh | +0.014 |
west | +0.013 |
Neg logit contributions | |
Notre | -0.017 |
tul | -0.017 |
paramed | -0.016 |
Ark | -0.016 |
cuc | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | -2.51e-03 |
Pos logit contributions | |
ately | +0.077 |
nation | +0.065 |
ting | +0.060 |
arsh | +0.059 |
west | +0.057 |
Neg logit contributions | |
bal | -0.070 |
tul | -0.069 |
Notre | -0.067 |
esh | -0.063 |
cuc | -0.063 |
Token | nose |
Token ID | 9686 |
Feature activation | -9.38e-02 |
Pos logit contributions | |
ately | +0.003 |
nation | +0.003 |
arsh | +0.003 |
west | +0.003 |
xy | +0.003 |
Neg logit contributions | |
tul | -0.002 |
bal | -0.002 |
Notre | -0.002 |
stret | -0.002 |
te | -0.002 |
Token | every |
Token ID | 790 |
Feature activation | -4.20e-02 |
Pos logit contributions | |
ately | +0.125 |
nation | +0.110 |
arsh | +0.103 |
34 | +0.101 |
west | +0.101 |
Neg logit contributions | |
bal | -0.082 |
tul | -0.076 |
� | -0.069 |
te | -0.069 |
Notre | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | -9.04e-03 |
Pos logit contributions | |
ately | +0.113 |
ting | +0.086 |
nation | +0.085 |
arsh | +0.079 |
34 | +0.077 |
Neg logit contributions | |
tul | -0.124 |
Notre | -0.120 |
bal | -0.118 |
von | -0.115 |
Ark | -0.115 |
Token | Someone |
Token ID | 17877 |
Feature activation | -4.14e-02 |
Pos logit contributions | |
ately | +0.012 |
nation | +0.011 |
ting | +0.010 |
34 | +0.010 |
arsh | +0.010 |
Neg logit contributions | |
bal | -0.007 |
Notre | -0.007 |
� | -0.006 |
tul | -0.006 |
von | -0.006 |
Token | had |
Token ID | 550 |
Feature activation | +1.85e-03 |
Pos logit contributions | |
ately | +0.037 |
nation | +0.027 |
ting | +0.025 |
usual | +0.022 |
upper | +0.022 |
Neg logit contributions | |
tul | -0.058 |
Notre | -0.057 |
thirteen | -0.056 |
von | -0.055 |
cuc | -0.055 |
Token | put |
Token ID | 1234 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
tul | +0.002 |
Notre | +0.002 |
bal | +0.002 |
cuc | +0.002 |
thirteen | +0.002 |
Neg logit contributions | |
ately | -0.002 |
nation | -0.002 |
arsh | -0.002 |
upper | -0.001 |
usual | -0.001 |
Token | net |
Token ID | 2010 |
Feature activation | -3.15e-03 |
Pos logit contributions | |
tul | +0.030 |
bal | +0.029 |
Notre | +0.028 |
cuc | +0.027 |
von | +0.027 |
Neg logit contributions | |
ately | -0.029 |
nation | -0.024 |
ting | -0.022 |
arsh | -0.021 |
34 | -0.021 |
Token | ting |
Token ID | 889 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
ately | +0.002 |
fully | +0.002 |
ting | +0.002 |
owing | +0.002 |
nation | +0.001 |
Neg logit contributions | |
tul | -0.005 |
Notre | -0.005 |
paramed | -0.005 |
MacDonald | -0.005 |
ALTH | -0.005 |
Token | in |
Token ID | 287 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
ately | +0.196 |
ting | +0.180 |
nation | +0.150 |
upper | +0.117 |
west | +0.116 |
Neg logit contributions | |
tul | -0.384 |
Notre | -0.355 |
cuc | -0.351 |
von | -0.350 |
bal | -0.347 |
Token | the |
Token ID | 262 |
Feature activation | -2.76e-03 |
Pos logit contributions | |
tul | +0.009 |
bal | +0.009 |
Notre | +0.009 |
cuc | +0.008 |
esh | +0.008 |
Neg logit contributions | |
ately | -0.013 |
nation | -0.011 |
arsh | -0.010 |
34 | -0.010 |
west | -0.010 |
Token | hole |
Token ID | 7604 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
ately | +0.003 |
nation | +0.003 |
upper | +0.003 |
34 | +0.003 |
ting | +0.003 |
Neg logit contributions | |
bal | -0.003 |
tul | -0.003 |
esh | -0.003 |
Notre | -0.002 |
von | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -8.29e-03 |
Pos logit contributions | |
tul | +0.027 |
bal | +0.026 |
Notre | +0.025 |
cuc | +0.024 |
Ark | +0.024 |
Neg logit contributions | |
ately | -0.029 |
nation | -0.023 |
arsh | -0.022 |
ting | -0.021 |
34 | -0.021 |
Token | Sam |
Token ID | 3409 |
Feature activation | -4.81e-02 |
Pos logit contributions | |
ately | +0.011 |
nation | +0.010 |
upper | +0.009 |
ting | +0.009 |
34 | +0.009 |
Neg logit contributions | |
bal | -0.007 |
Notre | -0.007 |
� | -0.007 |
dod | -0.006 |
von | -0.006 |
Token | place |
Token ID | 1295 |
Feature activation | -8.96e-02 |
Pos logit contributions | |
ately | +0.049 |
nation | +0.041 |
arsh | +0.037 |
ting | +0.036 |
34 | +0.036 |
Neg logit contributions | |
tul | -0.047 |
bal | -0.046 |
Notre | -0.045 |
cuc | -0.044 |
von | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | -8.44e-03 |
Pos logit contributions | |
ately | +0.078 |
nation | +0.061 |
usual | +0.056 |
owing | +0.053 |
yet | +0.052 |
Neg logit contributions | |
tul | -0.124 |
Ark | -0.118 |
cuc | -0.116 |
Notre | -0.116 |
bal | -0.115 |
Token | He |
Token ID | 679 |
Feature activation | +4.70e-03 |
Pos logit contributions | |
ately | +0.011 |
nation | +0.009 |
arsh | +0.008 |
34 | +0.008 |
ting | +0.008 |
Neg logit contributions | |
bal | -0.008 |
tul | -0.008 |
Notre | -0.007 |
cuc | -0.007 |
von | -0.007 |
Token | started |
Token ID | 2067 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
tul | +0.005 |
bal | +0.005 |
Notre | +0.005 |
cuc | +0.004 |
esh | +0.004 |
Neg logit contributions | |
ately | -0.006 |
nation | -0.005 |
arsh | -0.004 |
34 | -0.004 |
west | -0.004 |
Token | regret |
Token ID | 13721 |
Feature activation | +8.09e-02 |
Pos logit contributions | |
ately | +0.012 |
ting | +0.009 |
nation | +0.008 |
usual | +0.008 |
arsh | +0.008 |
Neg logit contributions | |
cuc | -0.024 |
thirteen | -0.024 |
tul | -0.023 |
von | -0.023 |
bal | -0.023 |
Token | ting |
Token ID | 889 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
bal | +0.071 |
tul | +0.069 |
mo | +0.066 |
Notre | +0.065 |
esh | +0.064 |
Neg logit contributions | |
ately | -0.101 |
nation | -0.093 |
west | -0.084 |
upper | -0.083 |
arsh | -0.082 |
Token | his |
Token ID | 465 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
ately | +0.273 |
ting | +0.229 |
nation | +0.227 |
arsh | +0.200 |
34 | +0.194 |
Neg logit contributions | |
tul | -0.307 |
cuc | -0.289 |
bal | -0.287 |
Notre | -0.285 |
von | -0.281 |
Token | decision |
Token ID | 2551 |
Feature activation | +4.10e-02 |
Pos logit contributions | |
ately | +0.028 |
nation | +0.024 |
arsh | +0.022 |
34 | +0.021 |
west | +0.021 |
Neg logit contributions | |
tul | -0.021 |
bal | -0.021 |
Notre | -0.020 |
cuc | -0.019 |
esh | -0.019 |
Token | to |
Token ID | 284 |
Feature activation | +5.11e-03 |
Pos logit contributions | |
bal | +0.037 |
tul | +0.037 |
Notre | +0.036 |
cuc | +0.034 |
esh | +0.034 |
Neg logit contributions | |
ately | -0.051 |
nation | -0.044 |
arsh | -0.041 |
34 | -0.040 |
west | -0.039 |
Token | go |
Token ID | 467 |
Feature activation | +2.15e-03 |
Pos logit contributions | |
� | +0.003 |
gn | +0.003 |
mo | +0.003 |
esh | +0.002 |
tul | +0.002 |
Neg logit contributions | |
ately | -0.008 |
ancer | -0.008 |
34 | -0.007 |
xy | -0.007 |
arsh | -0.007 |
Token | in |
Token ID | 287 |
Feature activation | +8.74e-03 |
Pos logit contributions | |
tul | +0.002 |
bal | +0.002 |
Notre | +0.002 |
cuc | +0.002 |
te | +0.002 |
Neg logit contributions | |
ately | -0.003 |
nation | -0.002 |
arsh | -0.002 |
ting | -0.002 |
34 | -0.002 |
Token | laugh |
Token ID | 6487 |
Feature activation | +7.85e-02 |
Pos logit contributions | |
bal | +0.002 |
Notre | +0.002 |
Ark | +0.002 |
tul | +0.002 |
#### | +0.001 |
Neg logit contributions | |
ately | -0.001 |
nation | -0.001 |
ting | -0.001 |
upper | -0.001 |
west | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
tul | +0.092 |
Notre | +0.088 |
Ark | +0.087 |
bal | +0.086 |
lyn | +0.084 |
Neg logit contributions | |
ately | -0.078 |
nation | -0.066 |
ting | -0.062 |
arsh | -0.059 |
34 | -0.059 |
Token | They |
Token ID | 1119 |
Feature activation | -3.03e-03 |
Pos logit contributions | |
ately | +0.017 |
nation | +0.014 |
arsh | +0.013 |
west | +0.013 |
ting | +0.013 |
Neg logit contributions | |
tul | -0.018 |
bal | -0.017 |
Notre | -0.017 |
esh | -0.016 |
cuc | -0.016 |
Token | kept |
Token ID | 4030 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
ately | +0.003 |
nation | +0.003 |
ting | +0.003 |
west | +0.002 |
arsh | +0.002 |
Neg logit contributions | |
tul | -0.003 |
bal | -0.003 |
Notre | -0.003 |
esh | -0.003 |
cuc | -0.003 |
Token | pet |
Token ID | 4273 |
Feature activation | -3.10e-03 |
Pos logit contributions | |
ately | +0.032 |
nation | +0.027 |
arsh | +0.025 |
34 | +0.024 |
upper | +0.024 |
Neg logit contributions | |
tul | -0.025 |
bal | -0.025 |
Notre | -0.024 |
cuc | -0.023 |
mo | -0.023 |
Token | ting |
Token ID | 889 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
ately | +0.002 |
nation | +0.002 |
arsh | +0.002 |
34 | +0.001 |
upper | +0.001 |
Neg logit contributions | |
bal | -0.005 |
tul | -0.004 |
Notre | -0.004 |
� | -0.004 |
te | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | -5.16e-03 |
Pos logit contributions | |
ately | +0.297 |
nation | +0.250 |
ting | +0.238 |
arsh | +0.226 |
west | +0.219 |
Neg logit contributions | |
tul | -0.272 |
bal | -0.263 |
Notre | -0.258 |
cuc | -0.252 |
von | -0.245 |
Token | rug |
Token ID | 14477 |
Feature activation | +5.09e-02 |
Pos logit contributions | |
ately | +0.007 |
nation | +0.006 |
34 | +0.006 |
arsh | +0.006 |
upper | +0.006 |
Neg logit contributions | |
bal | -0.004 |
tul | -0.004 |
Notre | -0.004 |
cuc | -0.004 |
esh | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | -2.30e-02 |
Pos logit contributions | |
tul | +0.051 |
bal | +0.050 |
Notre | +0.049 |
cuc | +0.047 |
esh | +0.046 |
Neg logit contributions | |
ately | -0.060 |
nation | -0.050 |
arsh | -0.047 |
ting | -0.045 |
34 | -0.044 |
Token | the |
Token ID | 262 |
Feature activation | -4.22e-03 |
Pos logit contributions | |
ately | +0.030 |
nation | +0.026 |
arsh | +0.025 |
34 | +0.024 |
west | +0.024 |
Neg logit contributions | |
tul | -0.020 |
bal | -0.020 |
Notre | -0.019 |
mo | -0.018 |
te | -0.018 |
Token | rug |
Token ID | 14477 |
Feature activation | +5.06e-02 |
Pos logit contributions | |
ately | +0.004 |
nation | +0.004 |
upper | +0.004 |
ting | +0.004 |
34 | +0.003 |
Neg logit contributions | |
esh | -0.005 |
bal | -0.004 |
Ark | -0.004 |
Dun | -0.004 |
cuc | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
tul | +0.063 |
bal | +0.061 |
Notre | +0.060 |
cuc | +0.057 |
esh | +0.056 |
Neg logit contributions | |
ately | -0.077 |
nation | -0.064 |
arsh | -0.059 |
ting | -0.057 |
upper | -0.057 |
Token | and |
Token ID | 290 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
ately | +0.012 |
nation | +0.010 |
ting | +0.009 |
arsh | +0.009 |
upper | +0.009 |
Neg logit contributions | |
bal | -0.011 |
tul | -0.011 |
Notre | -0.011 |
esh | -0.011 |
von | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | -2.19e-03 |
Pos logit contributions | |
ately | +0.022 |
nation | +0.018 |
ting | +0.017 |
arsh | +0.017 |
west | +0.016 |
Neg logit contributions | |
tul | -0.019 |
Notre | -0.019 |
bal | -0.018 |
esh | -0.018 |
cuc | -0.017 |
Token | gentle |
Token ID | 10296 |
Feature activation | -5.92e-02 |
Pos logit contributions | |
ately | +0.002 |
upper | +0.002 |
usual | +0.002 |
nation | +0.002 |
other | +0.002 |
Neg logit contributions | |
esh | -0.003 |
Dun | -0.002 |
Ark | -0.002 |
ALTH | -0.002 |
tin | -0.002 |
Token | pat |
Token ID | 1458 |
Feature activation | +6.84e-02 |
Pos logit contributions | |
ately | +0.051 |
upper | +0.038 |
nation | +0.037 |
ous | +0.037 |
usual | +0.036 |
Neg logit contributions | |
Notre | -0.073 |
Ark | -0.072 |
Dun | -0.072 |
unbeat | -0.072 |
tin | -0.071 |
Token | ting |
Token ID | 889 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
bal | +0.060 |
� | +0.057 |
te | +0.053 |
tul | +0.051 |
marsh | +0.050 |
Neg logit contributions | |
ately | -0.086 |
nation | -0.077 |
34 | -0.077 |
arsh | -0.076 |
upper | -0.074 |
Token | of |
Token ID | 286 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
ately | +0.282 |
ting | +0.240 |
nation | +0.227 |
west | +0.203 |
upper | +0.203 |
Neg logit contributions | |
tul | -0.275 |
Notre | -0.273 |
Ark | -0.270 |
bal | -0.264 |
esh | -0.262 |
Token | Smoke |
Token ID | 25416 |
Feature activation | -9.88e-02 |
Pos logit contributions | |
ately | +0.029 |
nation | +0.025 |
arsh | +0.024 |
34 | +0.023 |
west | +0.022 |
Neg logit contributions | |
tul | -0.019 |
bal | -0.018 |
Notre | -0.018 |
cuc | -0.017 |
te | -0.017 |
Token | y |
Token ID | 88 |
Feature activation | -8.28e-02 |
Pos logit contributions | |
ately | +0.110 |
nation | +0.089 |
arsh | +0.082 |
ting | +0.081 |
34 | +0.077 |
Neg logit contributions | |
tul | -0.109 |
bal | -0.105 |
Notre | -0.105 |
cuc | -0.101 |
Ark | -0.100 |
Token | 's |
Token ID | 338 |
Feature activation | +3.26e-03 |
Pos logit contributions | |
ately | +0.092 |
nation | +0.072 |
ting | +0.068 |
upper | +0.066 |
arsh | +0.066 |
Neg logit contributions | |
Ark | -0.089 |
Notre | -0.088 |
tul | -0.087 |
esh | -0.087 |
bal | -0.083 |
Token | paw |
Token ID | 41827 |
Feature activation | -7.04e-02 |
Pos logit contributions | |
bal | +0.004 |
tul | +0.003 |
Notre | +0.003 |
Ark | +0.003 |
#### | +0.003 |
Neg logit contributions | |
ately | -0.004 |
nation | -0.003 |
upper | -0.003 |
arsh | -0.003 |
34 | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -6.37e-03 |
Pos logit contributions | |
ately | +0.046 |
nation | +0.036 |
arsh | +0.034 |
ting | +0.034 |
34 | +0.032 |
Neg logit contributions | |
Notre | -0.042 |
tul | -0.041 |
bal | -0.041 |
esh | -0.039 |
von | -0.039 |
Token | " |
Token ID | 366 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
ately | +0.009 |
nation | +0.008 |
upper | +0.008 |
arsh | +0.008 |
west | +0.008 |
Neg logit contributions | |
tul | -0.005 |
bal | -0.004 |
� | -0.004 |
cuc | -0.004 |
pap | -0.004 |
Token | I |
Token ID | 40 |
Feature activation | +4.29e-03 |
Pos logit contributions | |
ately | +0.029 |
nation | +0.025 |
ting | +0.022 |
west | +0.021 |
arsh | +0.021 |
Neg logit contributions | |
tul | -0.028 |
bal | -0.028 |
cuc | -0.027 |
Notre | -0.027 |
mo | -0.026 |
Token | 'm |
Token ID | 1101 |
Feature activation | -2.35e-03 |
Pos logit contributions | |
tul | +0.005 |
cuc | +0.005 |
thirteen | +0.005 |
bal | +0.005 |
Notre | +0.005 |
Neg logit contributions | |
ately | -0.004 |
nation | -0.003 |
arsh | -0.003 |
oo | -0.003 |
west | -0.003 |
Token | regret |
Token ID | 13721 |
Feature activation | +8.25e-02 |
Pos logit contributions | |
ately | +0.003 |
nation | +0.002 |
arsh | +0.002 |
34 | +0.002 |
ting | +0.002 |
Neg logit contributions | |
tul | -0.002 |
bal | -0.002 |
Notre | -0.002 |
cuc | -0.002 |
esh | -0.002 |
Token | ting |
Token ID | 889 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
tul | +0.076 |
bal | +0.075 |
Notre | +0.071 |
mo | +0.068 |
esh | +0.068 |
Neg logit contributions | |
ately | -0.103 |
nation | -0.090 |
arsh | -0.081 |
west | -0.081 |
upper | -0.080 |
Token | not |
Token ID | 407 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
ately | +0.285 |
nation | +0.241 |
ting | +0.234 |
arsh | +0.222 |
34 | +0.211 |
Neg logit contributions | |
tul | -0.283 |
bal | -0.269 |
Notre | -0.269 |
cuc | -0.266 |
Ark | -0.258 |
Token | taking |
Token ID | 2263 |
Feature activation | -3.39e-02 |
Pos logit contributions | |
bal | +0.029 |
tul | +0.029 |
Notre | +0.027 |
� | +0.026 |
#### | +0.026 |
Neg logit contributions | |
ately | -0.037 |
nation | -0.029 |
upper | -0.028 |
arsh | -0.027 |
34 | -0.027 |
Token | one |
Token ID | 530 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
ately | +0.042 |
nation | +0.036 |
arsh | +0.034 |
34 | +0.032 |
upper | +0.032 |
Neg logit contributions | |
tul | -0.032 |
bal | -0.031 |
Notre | -0.030 |
cuc | -0.029 |
esh | -0.028 |
Token | special |
Token ID | 2041 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
bal | +0.028 |
� | +0.025 |
tul | +0.024 |
te | +0.021 |
cuc | +0.020 |
Neg logit contributions | |
ately | -0.039 |
nation | -0.035 |
xy | -0.033 |
arsh | -0.033 |
west | -0.031 |
Token | day |
Token ID | 1110 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
Ark | +0.017 |
tul | +0.016 |
Notre | +0.016 |
esh | +0.016 |
von | +0.016 |
Neg logit contributions | |
ately | -0.018 |
nation | -0.014 |
upper | -0.013 |
arsh | -0.013 |
ting | -0.013 |
Token | followed |
Token ID | 3940 |
Feature activation | -6.75e-02 |
Pos logit contributions | |
tul | +0.033 |
bal | +0.032 |
Notre | +0.032 |
cuc | +0.031 |
esh | +0.031 |
Neg logit contributions | |
ately | -0.032 |
nation | -0.024 |
arsh | -0.022 |
ting | -0.021 |
west | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | +1.76e-03 |
Pos logit contributions | |
ately | +0.079 |
nation | +0.063 |
ting | +0.060 |
arsh | +0.058 |
west | +0.056 |
Neg logit contributions | |
bal | -0.070 |
tul | -0.070 |
Notre | -0.069 |
cuc | -0.066 |
esh | -0.065 |
Token | dog |
Token ID | 3290 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
tul | +0.001 |
bal | +0.001 |
Notre | +0.001 |
te | +0.001 |
mo | +0.001 |
Neg logit contributions | |
ately | -0.002 |
nation | -0.002 |
arsh | -0.002 |
34 | -0.002 |
west | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -6.44e-03 |
Pos logit contributions | |
ately | +0.030 |
nation | +0.025 |
arsh | +0.024 |
34 | +0.023 |
upper | +0.023 |
Neg logit contributions | |
tul | -0.022 |
bal | -0.022 |
Notre | -0.021 |
cuc | -0.020 |
esh | -0.020 |
Token | pat |
Token ID | 1458 |
Feature activation | +7.08e-02 |
Pos logit contributions | |
ately | +0.007 |
nation | +0.006 |
arsh | +0.006 |
ting | +0.006 |
34 | +0.005 |
Neg logit contributions | |
tul | -0.007 |
bal | -0.007 |
Notre | -0.006 |
esh | -0.006 |
cuc | -0.006 |
Token | ting |
Token ID | 889 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
� | +0.078 |
bal | +0.075 |
tul | +0.070 |
bra | +0.070 |
� | +0.070 |
Neg logit contributions | |
ately | -0.068 |
arter | -0.064 |
arsh | -0.064 |
34 | -0.062 |
nation | -0.060 |
Token | its |
Token ID | 663 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
ately | +0.271 |
ting | +0.226 |
nation | +0.224 |
arsh | +0.198 |
west | +0.196 |
Neg logit contributions | |
tul | -0.293 |
bal | -0.283 |
Notre | -0.282 |
cuc | -0.271 |
von | -0.266 |
Token | head |
Token ID | 1182 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
ately | +0.143 |
nation | +0.122 |
arsh | +0.120 |
ancer | +0.117 |
ting | +0.115 |
Neg logit contributions | |
tul | -0.082 |
bal | -0.082 |
Notre | -0.077 |
pap | -0.074 |
mo | -0.073 |
Token | when |
Token ID | 618 |
Feature activation | +9.09e-03 |
Pos logit contributions | |
ately | +0.043 |
nation | +0.040 |
arsh | +0.037 |
west | +0.037 |
34 | +0.036 |
Neg logit contributions | |
bal | -0.024 |
� | -0.022 |
tul | -0.022 |
mo | -0.020 |
esh | -0.019 |
Token | it |
Token ID | 340 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
tul | +0.008 |
bal | +0.008 |
Notre | +0.008 |
cuc | +0.007 |
esh | +0.007 |
Neg logit contributions | |
ately | -0.012 |
nation | -0.010 |
arsh | -0.009 |
34 | -0.009 |
ting | -0.009 |
Token | seemed |
Token ID | 3947 |
Feature activation | -4.10e-02 |
Pos logit contributions | |
ately | +0.043 |
arsh | +0.039 |
nation | +0.038 |
34 | +0.037 |
upper | +0.037 |
Neg logit contributions | |
bal | -0.023 |
tul | -0.022 |
mo | -0.021 |
te | -0.021 |
gn | -0.020 |
Token | precious |
Token ID | 14186 |
Feature activation | -1.34e-03 |
Pos logit contributions | |
ately | +0.026 |
nation | +0.022 |
arsh | +0.021 |
ting | +0.020 |
west | +0.020 |
Neg logit contributions | |
tul | -0.018 |
bal | -0.018 |
Notre | -0.017 |
cuc | -0.016 |
mo | -0.016 |
Token | time |
Token ID | 640 |
Feature activation | +4.45e-02 |
Pos logit contributions | |
ately | +0.002 |
nation | +0.001 |
upper | +0.001 |
arsh | +0.001 |
34 | +0.001 |
Neg logit contributions | |
tul | -0.001 |
bal | -0.001 |
Notre | -0.001 |
Ark | -0.001 |
thirteen | -0.001 |
Token | wisely |
Token ID | 32773 |
Feature activation | +2.68e-03 |
Pos logit contributions | |
tul | +0.045 |
bal | +0.043 |
Notre | +0.043 |
cuc | +0.041 |
esh | +0.041 |
Neg logit contributions | |
ately | -0.053 |
nation | -0.044 |
arsh | -0.041 |
ting | -0.040 |
34 | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | -2.84e-03 |
Pos logit contributions | |
� | +0.002 |
bal | +0.002 |
tul | +0.001 |
Notre | +0.001 |
cuc | +0.001 |
Neg logit contributions | |
ately | -0.004 |
nation | -0.003 |
west | -0.003 |
upper | -0.003 |
rip | -0.003 |
Token | Forget |
Token ID | 29624 |
Feature activation | -6.16e-02 |
Pos logit contributions | |
ately | +0.004 |
nation | +0.003 |
arsh | +0.003 |
34 | +0.003 |
upper | +0.003 |
Neg logit contributions | |
bal | -0.003 |
tul | -0.003 |
Notre | -0.003 |
esh | -0.002 |
cuc | -0.002 |
Token | ting |
Token ID | 889 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
ately | +0.058 |
nation | +0.051 |
west | +0.046 |
upper | +0.046 |
34 | +0.045 |
Neg logit contributions | |
tul | -0.071 |
bal | -0.071 |
thirteen | -0.066 |
Notre | -0.066 |
cuc | -0.065 |
Token | about |
Token ID | 546 |
Feature activation | -3.81e-02 |
Pos logit contributions | |
ately | +0.304 |
nation | +0.256 |
arsh | +0.236 |
ting | +0.235 |
34 | +0.225 |
Neg logit contributions | |
tul | -0.266 |
bal | -0.256 |
Notre | -0.249 |
cuc | -0.242 |
von | -0.237 |
Token | his |
Token ID | 465 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
ately | +0.047 |
nation | +0.040 |
arsh | +0.038 |
34 | +0.036 |
upper | +0.036 |
Neg logit contributions | |
tul | -0.036 |
bal | -0.035 |
Notre | -0.034 |
cuc | -0.032 |
esh | -0.032 |
Token | stumble |
Token ID | 35174 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
ately | +0.026 |
nation | +0.022 |
arsh | +0.021 |
34 | +0.020 |
west | +0.020 |
Neg logit contributions | |
tul | -0.019 |
bal | -0.019 |
Notre | -0.018 |
cuc | -0.018 |
esh | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -7.93e-03 |
Pos logit contributions | |
ately | +0.036 |
nation | +0.030 |
arsh | +0.028 |
ting | +0.027 |
34 | +0.027 |
Neg logit contributions | |
tul | -0.030 |
bal | -0.029 |
Notre | -0.029 |
cuc | -0.027 |
esh | -0.027 |
Token | Johnny |
Token ID | 15470 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
ately | +0.009 |
nation | +0.008 |
arsh | +0.007 |
ting | +0.007 |
34 | +0.007 |
Neg logit contributions | |
tul | -0.008 |
bal | -0.008 |
Notre | -0.008 |
von | -0.007 |
esh | -0.007 |
Token | through |
Token ID | 832 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
ately | +0.044 |
nation | +0.039 |
arsh | +0.038 |
34 | +0.035 |
upper | +0.035 |
Neg logit contributions | |
tul | -0.025 |
bal | -0.025 |
Notre | -0.023 |
mo | -0.023 |
esh | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | -8.32e-04 |
Pos logit contributions | |
ately | +0.023 |
nation | +0.021 |
arsh | +0.021 |
ancer | +0.020 |
upper | +0.019 |
Neg logit contributions | |
tul | -0.012 |
mo | -0.011 |
Notre | -0.010 |
bal | -0.010 |
sprink | -0.010 |
Token | window |
Token ID | 4324 |
Feature activation | -7.19e-02 |
Pos logit contributions | |
ately | +0.001 |
upper | +0.001 |
ting | +0.001 |
34 | +0.001 |
nation | +0.001 |
Neg logit contributions | |
bal | -0.001 |
esh | -0.001 |
Ark | -0.001 |
tul | -0.001 |
unbeat | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
ately | +0.110 |
nation | +0.105 |
west | +0.097 |
arsh | +0.095 |
pay | +0.094 |
Neg logit contributions | |
bal | -0.041 |
mo | -0.040 |
� | -0.039 |
thirteen | -0.038 |
tul | -0.037 |
Token | wet |
Token ID | 9583 |
Feature activation | -9.44e-03 |
Pos logit contributions | |
ately | +0.026 |
nation | +0.021 |
ting | +0.021 |
upper | +0.019 |
arsh | +0.019 |
Neg logit contributions | |
bal | -0.026 |
von | -0.025 |
tul | -0.025 |
Notre | -0.024 |
Ark | -0.024 |
Token | ting |
Token ID | 889 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
ately | +0.012 |
nation | +0.010 |
ting | +0.010 |
arsh | +0.010 |
34 | +0.010 |
Neg logit contributions | |
tul | -0.008 |
bal | -0.008 |
Notre | -0.008 |
cuc | -0.008 |
Ark | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | -1.04e-03 |
Pos logit contributions | |
ately | +0.283 |
ting | +0.235 |
nation | +0.234 |
west | +0.208 |
34 | +0.206 |
Neg logit contributions | |
bal | -0.275 |
tul | -0.272 |
Ark | -0.266 |
Notre | -0.264 |
cuc | -0.260 |
Token | floor |
Token ID | 4314 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
ately | +0.001 |
upper | +0.001 |
nation | +0.001 |
ting | +0.001 |
34 | +0.001 |
Neg logit contributions | |
esh | -0.001 |
bal | -0.001 |
Ark | -0.001 |
tin | -0.001 |
tul | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
tul | +0.012 |
bal | +0.012 |
Notre | +0.011 |
Ark | +0.011 |
cuc | +0.011 |
Neg logit contributions | |
ately | -0.011 |
nation | -0.009 |
arsh | -0.008 |
34 | -0.008 |
upper | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
ately | +0.022 |
nation | +0.020 |
arsh | +0.018 |
ting | +0.018 |
upper | +0.017 |
Neg logit contributions | |
tul | -0.014 |
bal | -0.013 |
Notre | -0.013 |
te | -0.013 |
mo | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
ately | +0.038 |
nation | +0.032 |
ting | +0.031 |
arsh | +0.030 |
upper | +0.030 |
Neg logit contributions | |
tul | -0.018 |
� | -0.018 |
bal | -0.018 |
Notre | -0.018 |
mo | -0.017 |
Token | they |
Token ID | 484 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
tul | +0.003 |
bal | +0.003 |
Notre | +0.003 |
mo | +0.003 |
cuc | +0.003 |
Neg logit contributions | |
ately | -0.005 |
nation | -0.005 |
arsh | -0.004 |
upper | -0.004 |
34 | -0.004 |
Token | put |
Token ID | 1234 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
esh | +0.030 |
iot | +0.028 |
spe | +0.028 |
bal | +0.028 |
tul | +0.028 |
Neg logit contributions | |
eat | -0.006 |
he | -0.006 |
do | -0.006 |
ately | -0.005 |
rip | -0.004 |
Token | away |
Token ID | 1497 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
tul | +0.033 |
bal | +0.032 |
Notre | +0.031 |
cuc | +0.031 |
te | +0.030 |
Neg logit contributions | |
ately | -0.053 |
nation | -0.046 |
arsh | -0.043 |
34 | -0.040 |
upper | -0.040 |
Token | their |
Token ID | 511 |
Feature activation | +9.98e-03 |
Pos logit contributions | |
ately | +0.016 |
nation | +0.015 |
arsh | +0.014 |
upper | +0.013 |
west | +0.013 |
Neg logit contributions | |
bal | -0.007 |
tul | -0.007 |
mo | -0.006 |
Notre | -0.006 |
cuc | -0.006 |
Token | fake |
Token ID | 8390 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
tul | +0.009 |
bal | +0.009 |
Notre | +0.009 |
esh | +0.009 |
von | +0.008 |
Neg logit contributions | |
ately | -0.012 |
nation | -0.011 |
arsh | -0.010 |
upper | -0.009 |
34 | -0.009 |
Token | pizz |
Token ID | 38748 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
ately | +0.012 |
upper | +0.009 |
nation | +0.009 |
34 | +0.009 |
arsh | +0.009 |
Neg logit contributions | |
tul | -0.013 |
esh | -0.013 |
von | -0.013 |
bal | -0.012 |
Notre | -0.012 |
Token | as |
Token ID | 292 |
Feature activation | +4.65e-02 |
Pos logit contributions | |
tul | +0.129 |
bal | +0.122 |
Notre | +0.120 |
esh | +0.115 |
cuc | +0.114 |
Neg logit contributions | |
ately | -0.179 |
nation | -0.151 |
arsh | -0.142 |
34 | -0.139 |
upper | -0.137 |
Token | and |
Token ID | 290 |
Feature activation | -5.57e-03 |
Pos logit contributions | |
bal | +0.042 |
tul | +0.042 |
Notre | +0.040 |
cuc | +0.038 |
te | +0.038 |
Neg logit contributions | |
ately | -0.059 |
nation | -0.051 |
arsh | -0.047 |
34 | -0.046 |
upper | -0.045 |
Token | hats |
Token ID | 23910 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
ately | +0.006 |
nation | +0.005 |
upper | +0.005 |
arsh | +0.005 |
34 | +0.005 |
Neg logit contributions | |
tul | -0.006 |
bal | -0.006 |
esh | -0.005 |
von | -0.005 |
Notre | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +3.01e-03 |
Pos logit contributions | |
tul | +0.097 |
bal | +0.096 |
Notre | +0.092 |
cuc | +0.087 |
esh | +0.087 |
Neg logit contributions | |
ately | -0.131 |
nation | -0.112 |
arsh | -0.104 |
34 | -0.100 |
west | -0.100 |
Token | They |
Token ID | 1119 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
bal | +0.002 |
tul | +0.002 |
Notre | +0.002 |
te | +0.002 |
cuc | +0.002 |
Neg logit contributions | |
ately | -0.005 |
nation | -0.004 |
arsh | -0.004 |
owing | -0.004 |
ting | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | +3.25e-02 |
Pos logit contributions | |
ately | +0.089 |
nation | +0.073 |
arsh | +0.068 |
ting | +0.067 |
34 | +0.064 |
Neg logit contributions | |
tul | -0.083 |
Notre | -0.079 |
bal | -0.079 |
cuc | -0.075 |
von | -0.075 |
Token | play |
Token ID | 711 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
disl | +0.015 |
� | +0.015 |
ot | +0.015 |
mess | +0.015 |
tul | +0.015 |
Neg logit contributions | |
ately | -0.052 |
year | -0.051 |
34 | -0.050 |
ancer | -0.047 |
xy | -0.045 |
Token | with |
Token ID | 351 |
Feature activation | +5.00e-02 |
Pos logit contributions | |
ately | +0.044 |
arsh | +0.037 |
rip | +0.037 |
34 | +0.036 |
west | +0.036 |
Neg logit contributions | |
bal | -0.019 |
te | -0.015 |
tul | -0.013 |
� | -0.013 |
ners | -0.013 |
Token | again |
Token ID | 757 |
Feature activation | -6.35e-02 |
Pos logit contributions | |
bal | +0.033 |
tul | +0.032 |
� | +0.028 |
te | +0.026 |
de | +0.025 |
Neg logit contributions | |
ately | -0.073 |
nation | -0.067 |
34 | -0.065 |
west | -0.064 |
ancer | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
ately | +0.106 |
ancer | +0.097 |
rip | +0.095 |
nation | +0.094 |
west | +0.094 |
Neg logit contributions | |
bal | -0.031 |
� | -0.031 |
tul | -0.027 |
gn | -0.022 |
mo | -0.021 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
tul | +0.028 |
bal | +0.028 |
Notre | +0.027 |
cuc | +0.026 |
te | +0.026 |
Neg logit contributions | |
ately | -0.028 |
nation | -0.023 |
arsh | -0.022 |
west | -0.021 |
34 | -0.020 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
� | +0.062 |
Notre | +0.051 |
bal | +0.051 |
te | +0.049 |
#### | +0.048 |
Neg logit contributions | |
ately | -0.210 |
nation | -0.174 |
arsh | -0.174 |
34 | -0.169 |
upper | -0.168 |
Token | upon |
Token ID | 2402 |
Feature activation | +7.17e-02 |
Pos logit contributions | |
tul | +0.100 |
bal | +0.097 |
Notre | +0.095 |
cuc | +0.089 |
thirteen | +0.088 |
Neg logit contributions | |
ately | -0.153 |
nation | -0.129 |
arsh | -0.124 |
34 | -0.117 |
west | -0.117 |
Token | a |
Token ID | 257 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
bal | +0.042 |
� | +0.040 |
tul | +0.037 |
Notre | +0.037 |
#### | +0.034 |
Neg logit contributions | |
ately | -0.121 |
nation | -0.101 |
arsh | -0.099 |
ting | -0.098 |
34 | -0.097 |
Token | time |
Token ID | 640 |
Feature activation | +7.59e-02 |
Pos logit contributions | |
tul | +0.016 |
bal | +0.016 |
Notre | +0.015 |
cuc | +0.014 |
esh | +0.014 |
Neg logit contributions | |
ately | -0.031 |
nation | -0.027 |
arsh | -0.025 |
34 | -0.025 |
upper | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
tul | +0.084 |
Notre | +0.080 |
bal | +0.080 |
cuc | +0.077 |
thirteen | +0.076 |
Neg logit contributions | |
ately | -0.084 |
nation | -0.067 |
arsh | -0.065 |
34 | -0.060 |
west | -0.059 |
Token | Wow |
Token ID | 22017 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
tul | +0.037 |
bal | +0.034 |
thirteen | +0.031 |
Notre | +0.031 |
� | +0.031 |
Neg logit contributions | |
ately | -0.091 |
arsh | -0.079 |
owing | -0.076 |
nation | -0.074 |
ous | -0.071 |
Token | , |
Token ID | 11 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
paramed | +0.030 |
spe | +0.030 |
tul | +0.030 |
Notre | +0.030 |
capt | +0.029 |
Neg logit contributions | |
ately | -0.020 |
34 | -0.014 |
arsh | -0.013 |
nation | -0.013 |
ting | -0.013 |
Token | look |
Token ID | 804 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
von | +0.020 |
esh | +0.020 |
iot | +0.019 |
bal | +0.019 |
ners | +0.019 |
Neg logit contributions | |
ately | -0.010 |
west | -0.007 |
34 | -0.007 |
yet | -0.006 |
arsh | -0.006 |
Token | at |
Token ID | 379 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
tul | +0.015 |
Notre | +0.015 |
Ark | +0.015 |
esh | +0.015 |
von | +0.015 |
Neg logit contributions | |
ately | -0.012 |
34 | -0.009 |
nation | -0.009 |
arsh | -0.009 |
ting | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
tul | +0.012 |
mo | +0.010 |
sprink | +0.010 |
thirteen | +0.009 |
esh | +0.009 |
Neg logit contributions | |
ately | -0.035 |
nation | -0.031 |
upper | -0.030 |
arsh | -0.030 |
ancer | -0.029 |
Token | k |
Token ID | 479 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
bal | +0.018 |
tul | +0.018 |
Notre | +0.017 |
cuc | +0.016 |
esh | +0.016 |
Neg logit contributions | |
ately | -0.039 |
nation | -0.035 |
34 | -0.033 |
west | -0.032 |
arsh | -0.032 |
Token | ite |
Token ID | 578 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
tul | +0.090 |
bal | +0.090 |
Notre | +0.081 |
te | +0.078 |
sprink | +0.078 |
Neg logit contributions | |
ately | -0.176 |
nation | -0.154 |
arsh | -0.148 |
upper | -0.144 |
34 | -0.135 |
Token | !" |
Token ID | 2474 |
Feature activation | +6.17e-02 |
Pos logit contributions | |
tul | +0.012 |
Notre | +0.012 |
bal | +0.012 |
paramed | +0.012 |
lyn | +0.012 |
Neg logit contributions | |
ately | -0.012 |
nation | -0.010 |
ting | -0.009 |
arsh | -0.009 |
34 | -0.009 |
Token | Sara |
Token ID | 24799 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
esh | +0.085 |
unbeat | +0.082 |
tul | +0.081 |
spe | +0.080 |
iot | +0.079 |
Neg logit contributions | |
ately | -0.049 |
west | -0.034 |
rip | -0.031 |
nation | -0.030 |
ting | -0.030 |
Token | says |
Token ID | 1139 |
Feature activation | +4.45e-02 |
Pos logit contributions | |
tul | +0.038 |
paramed | +0.035 |
spe | +0.035 |
stret | +0.035 |
Notre | +0.035 |
Neg logit contributions | |
ately | -0.026 |
nation | -0.021 |
west | -0.020 |
arsh | -0.019 |
34 | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
bal | +0.032 |
� | +0.029 |
tul | +0.028 |
te | +0.028 |
Notre | +0.026 |
Neg logit contributions | |
ately | -0.063 |
nation | -0.059 |
arsh | -0.054 |
34 | -0.053 |
west | -0.053 |
Token | whist |
Token ID | 32262 |
Feature activation | +2.96e-02 |
Pos logit contributions | |
tul | +0.006 |
bal | +0.006 |
Notre | +0.005 |
cuc | +0.005 |
esh | +0.005 |
Neg logit contributions | |
ately | -0.006 |
nation | -0.005 |
34 | -0.004 |
arsh | -0.004 |
ting | -0.004 |
Token | ling |
Token ID | 1359 |
Feature activation | -3.45e-02 |
Pos logit contributions | |
bal | +0.024 |
ners | +0.021 |
� | +0.021 |
tin | +0.021 |
Notre | +0.021 |
Neg logit contributions | |
ately | -0.042 |
nation | -0.034 |
arsh | -0.033 |
upper | -0.032 |
34 | -0.031 |
Token | too |
Token ID | 1165 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
ately | +0.045 |
nation | +0.038 |
arsh | +0.036 |
34 | +0.034 |
west | +0.034 |
Neg logit contributions | |
bal | -0.031 |
tul | -0.031 |
Notre | -0.029 |
cuc | -0.028 |
esh | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | +9.64e-03 |
Pos logit contributions | |
ately | +0.041 |
nation | +0.036 |
upper | +0.034 |
ting | +0.033 |
ancer | +0.032 |
Neg logit contributions | |
� | -0.016 |
bal | -0.016 |
de | -0.015 |
mo | -0.013 |
capt | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | +9.34e-04 |
Pos logit contributions | |
tul | +0.012 |
bal | +0.012 |
von | +0.012 |
Ark | +0.012 |
Notre | +0.012 |
Neg logit contributions | |
ately | -0.009 |
ting | -0.007 |
nation | -0.007 |
arsh | -0.006 |
34 | -0.006 |
Token | even |
Token ID | 772 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
tul | +0.001 |
bal | +0.001 |
Notre | +0.001 |
cuc | +0.001 |
esh | +0.001 |
Neg logit contributions | |
ately | -0.001 |
nation | -0.001 |
arsh | -0.001 |
34 | -0.001 |
upper | -0.001 |
Token | better |
Token ID | 1365 |
Feature activation | -5.52e-03 |
Pos logit contributions | |
bal | +0.089 |
tul | +0.081 |
bra | +0.077 |
gn | +0.074 |
dod | +0.073 |
Neg logit contributions | |
ately | -0.190 |
nation | -0.168 |
arsh | -0.166 |
xy | -0.164 |
34 | -0.158 |
Token | than |
Token ID | 621 |
Feature activation | +8.23e-02 |
Pos logit contributions | |
ately | +0.007 |
nation | +0.006 |
arsh | +0.006 |
34 | +0.005 |
west | +0.005 |
Neg logit contributions | |
tul | -0.005 |
bal | -0.005 |
Notre | -0.005 |
cuc | -0.004 |
esh | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
bal | +0.057 |
mo | +0.053 |
tul | +0.053 |
de | +0.050 |
cuc | +0.049 |
Neg logit contributions | |
ately | -0.114 |
nation | -0.111 |
arsh | -0.101 |
ivery | -0.099 |
west | -0.099 |
Token | big |
Token ID | 1263 |
Feature activation | -3.37e-04 |
Pos logit contributions | |
tul | +0.022 |
bal | +0.021 |
Notre | +0.021 |
cuc | +0.020 |
te | +0.019 |
Neg logit contributions | |
ately | -0.025 |
nation | -0.021 |
arsh | -0.020 |
34 | -0.018 |
west | -0.018 |
Token | bird |
Token ID | 6512 |
Feature activation | +3.80e-02 |
Pos logit contributions | |
ately | +0.000 |
nation | +0.000 |
arsh | +0.000 |
west | +0.000 |
34 | +0.000 |
Neg logit contributions | |
tul | -0.000 |
bal | -0.000 |
Notre | -0.000 |
cuc | -0.000 |
te | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
esh | +0.001 |
tul | +0.001 |
Ark | +0.001 |
icol | +0.001 |
lyn | +0.001 |
Neg logit contributions | |
ately | -0.001 |
ting | -0.001 |
arsh | -0.000 |
nation | -0.000 |
rip | -0.000 |
Token | You |
Token ID | 921 |
Feature activation | +7.26e-02 |
Pos logit contributions | |
tul | +0.016 |
bal | +0.016 |
Notre | +0.015 |
cuc | +0.015 |
esh | +0.015 |
Neg logit contributions | |
ately | -0.018 |
nation | -0.015 |
arsh | -0.014 |
34 | -0.013 |
west | -0.013 |
Token | found |
Token ID | 1043 |
Feature activation | +3.91e-02 |
Pos logit contributions | |
tul | +0.096 |
bal | +0.091 |
esh | +0.091 |
icol | +0.087 |
cuc | +0.087 |
Neg logit contributions | |
ately | -0.060 |
nation | -0.055 |
west | -0.045 |
rip | -0.045 |
arsh | -0.044 |
Token | a |
Token ID | 257 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
bal | +0.045 |
tul | +0.045 |
Notre | +0.045 |
esh | +0.043 |
spe | +0.043 |
Neg logit contributions | |
ately | -0.039 |
nation | -0.030 |
ting | -0.030 |
34 | -0.029 |
arsh | -0.028 |
Token | nice |
Token ID | 3621 |
Feature activation | -6.03e-02 |
Pos logit contributions | |
esh | +0.012 |
bal | +0.012 |
tul | +0.012 |
Notre | +0.012 |
von | +0.011 |
Neg logit contributions | |
ately | -0.011 |
nation | -0.009 |
upper | -0.008 |
arsh | -0.008 |
34 | -0.008 |
Token | cuc |
Token ID | 38421 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
ately | +0.038 |
usual | +0.030 |
nation | +0.026 |
upper | +0.026 |
oo | +0.026 |
Neg logit contributions | |
esh | -0.082 |
thirteen | -0.082 |
Notre | -0.080 |
pter | -0.079 |
foc | -0.078 |
Token | umber |
Token ID | 4494 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
bal | +0.231 |
tul | +0.228 |
Notre | +0.216 |
cuc | +0.215 |
te | +0.207 |
Neg logit contributions | |
ately | -0.260 |
nation | -0.223 |
arsh | -0.203 |
34 | -0.196 |
west | -0.195 |
Token | . |
Token ID | 13 |
Feature activation | +9.05e-03 |
Pos logit contributions | |
ately | +0.014 |
ting | +0.011 |
arsh | +0.011 |
nation | +0.010 |
owing | +0.010 |
Neg logit contributions | |
tul | -0.021 |
Notre | -0.020 |
imag | -0.020 |
von | -0.020 |
Ark | -0.019 |
Token | It |
Token ID | 632 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
tul | +0.009 |
bal | +0.009 |
Notre | +0.008 |
cuc | +0.008 |
esh | +0.008 |
Neg logit contributions | |
ately | -0.011 |
nation | -0.009 |
arsh | -0.009 |
34 | -0.008 |
upper | -0.008 |
Token | will |
Token ID | 481 |
Feature activation | +4.14e-02 |
Pos logit contributions | |
ately | +0.023 |
nation | +0.019 |
xy | +0.019 |
uch | +0.016 |
ting | +0.015 |
Neg logit contributions | |
stret | -0.041 |
tul | -0.040 |
icol | -0.039 |
bal | -0.039 |
von | -0.038 |
Token | be |
Token ID | 307 |
Feature activation | -2.84e-03 |
Pos logit contributions | |
bal | +0.049 |
tul | +0.046 |
esh | +0.046 |
Notre | +0.045 |
von | +0.045 |
Neg logit contributions | |
ately | -0.040 |
nation | -0.035 |
arsh | -0.030 |
usual | -0.029 |
upper | -0.028 |
Token | there |
Token ID | 612 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
ately | +0.007 |
nation | +0.006 |
arsh | +0.005 |
34 | +0.005 |
upper | +0.005 |
Neg logit contributions | |
tul | -0.005 |
bal | -0.005 |
Notre | -0.005 |
cuc | -0.005 |
esh | -0.005 |
Token | was |
Token ID | 373 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
ately | +0.031 |
nation | +0.027 |
34 | +0.026 |
arsh | +0.026 |
ting | +0.025 |
Neg logit contributions | |
bal | -0.014 |
tul | -0.014 |
� | -0.014 |
Notre | -0.013 |
esh | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | -3.20e-03 |
Pos logit contributions | |
ately | +0.011 |
nation | +0.009 |
arsh | +0.008 |
west | +0.008 |
upper | +0.008 |
Neg logit contributions | |
tul | -0.011 |
bal | -0.011 |
Notre | -0.010 |
esh | -0.010 |
cuc | -0.010 |
Token | poor |
Token ID | 3595 |
Feature activation | -9.89e-02 |
Pos logit contributions | |
ately | +0.004 |
nation | +0.003 |
arsh | +0.003 |
34 | +0.003 |
ting | +0.003 |
Neg logit contributions | |
tul | -0.003 |
bal | -0.003 |
Notre | -0.003 |
cuc | -0.003 |
mo | -0.003 |
Token | family |
Token ID | 1641 |
Feature activation | -3.72e-02 |
Pos logit contributions | |
ately | +0.117 |
nation | +0.099 |
arsh | +0.092 |
34 | +0.088 |
west | +0.087 |
Neg logit contributions | |
tul | -0.097 |
bal | -0.097 |
Notre | -0.094 |
esh | -0.089 |
cuc | -0.088 |
Token | . |
Token ID | 13 |
Feature activation | -1.71e-03 |
Pos logit contributions | |
ately | +0.018 |
ting | +0.017 |
oo | +0.015 |
arsh | +0.012 |
ful | +0.012 |
Neg logit contributions | |
thirteen | -0.062 |
imag | -0.061 |
dod | -0.060 |
imet | -0.060 |
icol | -0.059 |
Token | Every |
Token ID | 3887 |
Feature activation | +3.16e-02 |
Pos logit contributions | |
ately | +0.003 |
nation | +0.002 |
arsh | +0.002 |
ting | +0.002 |
34 | +0.002 |
Neg logit contributions | |
Notre | -0.001 |
tul | -0.001 |
te | -0.001 |
� | -0.001 |
von | -0.001 |
Token | day |
Token ID | 1110 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
bal | +0.020 |
tul | +0.016 |
� | +0.016 |
Notre | +0.016 |
te | +0.015 |
Neg logit contributions | |
ately | -0.054 |
nation | -0.047 |
ting | -0.046 |
34 | -0.046 |
arsh | -0.043 |
Token | was |
Token ID | 373 |
Feature activation | -8.87e-03 |
Pos logit contributions | |
ately | +0.024 |
nation | +0.020 |
arsh | +0.019 |
34 | +0.018 |
west | +0.018 |
Neg logit contributions | |
tul | -0.020 |
bal | -0.019 |
Notre | -0.019 |
esh | -0.018 |
cuc | -0.018 |
Token | so |
Token ID | 523 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
ately | +0.010 |
nation | +0.008 |
arsh | +0.008 |
upper | +0.007 |
ting | +0.007 |
Neg logit contributions | |
tul | -0.009 |
bal | -0.009 |
Notre | -0.009 |
esh | -0.009 |
cuc | -0.009 |
Token | hard |
Token ID | 1327 |
Feature activation | -3.78e-02 |
Pos logit contributions | |
tul | +0.015 |
mo | +0.015 |
bal | +0.015 |
Notre | +0.014 |
te | +0.014 |
Neg logit contributions | |
ately | -0.024 |
nation | -0.021 |
ting | -0.021 |
34 | -0.020 |
arsh | -0.020 |
Token | push |
Token ID | 4574 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
tul | +0.007 |
bal | +0.006 |
Notre | +0.006 |
te | +0.006 |
mo | +0.006 |
Neg logit contributions | |
ately | -0.008 |
nation | -0.007 |
arsh | -0.007 |
34 | -0.006 |
ting | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +4.67e-03 |
Pos logit contributions | |
tul | +0.021 |
Notre | +0.020 |
bal | +0.020 |
cuc | +0.020 |
von | +0.019 |
Neg logit contributions | |
ately | -0.022 |
nation | -0.018 |
ting | -0.017 |
34 | -0.016 |
arsh | -0.016 |
Token | furniture |
Token ID | 13091 |
Feature activation | +2.99e-02 |
Pos logit contributions | |
tul | +0.004 |
bal | +0.004 |
Notre | +0.004 |
te | +0.003 |
cuc | +0.003 |
Neg logit contributions | |
ately | -0.007 |
nation | -0.006 |
arsh | -0.005 |
34 | -0.005 |
west | -0.005 |
Token | around |
Token ID | 1088 |
Feature activation | +8.43e-02 |
Pos logit contributions | |
tul | +0.036 |
Notre | +0.034 |
Ark | +0.034 |
bal | +0.034 |
von | +0.033 |
Neg logit contributions | |
ately | -0.030 |
nation | -0.025 |
ting | -0.023 |
arsh | -0.022 |
upper | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | +4.05e-03 |
Pos logit contributions | |
tul | +0.073 |
bal | +0.073 |
Notre | +0.070 |
cuc | +0.067 |
mo | +0.067 |
Neg logit contributions | |
ately | -0.110 |
nation | -0.095 |
arsh | -0.090 |
34 | -0.086 |
west | -0.086 |
Token | living |
Token ID | 2877 |
Feature activation | +5.65e-02 |
Pos logit contributions | |
bal | +0.004 |
tul | +0.004 |
Notre | +0.004 |
esh | +0.004 |
cuc | +0.004 |
Neg logit contributions | |
ately | -0.004 |
nation | -0.004 |
upper | -0.004 |
ting | -0.004 |
34 | -0.003 |
Token | room |
Token ID | 2119 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
te | +0.031 |
mo | +0.028 |
tul | +0.027 |
piles | +0.027 |
Notre | +0.026 |
Neg logit contributions | |
ately | -0.094 |
nation | -0.085 |
arsh | -0.081 |
west | -0.076 |
ous | -0.076 |
Token | . |
Token ID | 13 |
Feature activation | -4.67e-03 |
Pos logit contributions | |
ately | +0.020 |
nation | +0.017 |
arsh | +0.016 |
34 | +0.016 |
upper | +0.015 |
Neg logit contributions | |
tul | -0.016 |
bal | -0.016 |
Notre | -0.015 |
cuc | -0.014 |
esh | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
ately | +0.005 |
nation | +0.004 |
arsh | +0.004 |
west | +0.004 |
upper | +0.004 |
Neg logit contributions | |
tul | -0.005 |
bal | -0.005 |
Notre | -0.005 |
cuc | -0.005 |
thirteen | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
ately | +0.027 |
nation | +0.022 |
ting | +0.022 |
34 | +0.021 |
upper | +0.021 |
Neg logit contributions | |
bal | -0.013 |
� | -0.012 |
tul | -0.011 |
Notre | -0.011 |
esh | -0.011 |
Token | One |
Token ID | 3198 |
Feature activation | -3.46e-02 |
Pos logit contributions | |
ately | +0.016 |
nation | +0.014 |
arsh | +0.012 |
west | +0.012 |
upper | +0.011 |
Neg logit contributions | |
tul | -0.021 |
Notre | -0.020 |
te | -0.020 |
cuc | -0.020 |
bal | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
ately | +0.029 |
nation | +0.025 |
arsh | +0.023 |
34 | +0.023 |
upper | +0.022 |
Neg logit contributions | |
tul | -0.023 |
bal | -0.023 |
Notre | -0.023 |
cuc | -0.021 |
te | -0.021 |
Token | The |
Token ID | 464 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
ately | +0.021 |
nation | +0.019 |
west | +0.016 |
arsh | +0.016 |
upper | +0.015 |
Neg logit contributions | |
tul | -0.029 |
bal | -0.028 |
imag | -0.028 |
Notre | -0.028 |
icol | -0.028 |
Token | oak |
Token ID | 26210 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
ately | +0.037 |
nation | +0.032 |
arsh | +0.029 |
34 | +0.029 |
west | +0.028 |
Neg logit contributions | |
bal | -0.028 |
tul | -0.028 |
Notre | -0.027 |
esh | -0.026 |
cuc | -0.026 |
Token | tree |
Token ID | 5509 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
ately | +0.014 |
nation | +0.011 |
arsh | +0.010 |
ting | +0.010 |
west | +0.010 |
Neg logit contributions | |
tul | -0.014 |
bal | -0.014 |
Notre | -0.014 |
cuc | -0.013 |
esh | -0.013 |
Token | was |
Token ID | 373 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
ately | +0.039 |
nation | +0.031 |
arsh | +0.026 |
west | +0.024 |
upper | +0.023 |
Neg logit contributions | |
tul | -0.051 |
Notre | -0.051 |
icol | -0.048 |
bal | -0.048 |
Ark | -0.047 |
Token | so |
Token ID | 523 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
nation | +0.010 |
ately | +0.010 |
upper | +0.008 |
arsh | +0.008 |
usual | +0.007 |
Neg logit contributions | |
Notre | -0.019 |
MacDonald | -0.019 |
icol | -0.018 |
bal | -0.018 |
tul | -0.018 |
Token | kind |
Token ID | 1611 |
Feature activation | -4.70e-02 |
Pos logit contributions | |
tul | +0.010 |
bal | +0.010 |
Notre | +0.010 |
cuc | +0.009 |
mo | +0.009 |
Neg logit contributions | |
ately | -0.012 |
nation | -0.010 |
arsh | -0.010 |
34 | -0.009 |
ting | -0.009 |
Token | that |
Token ID | 326 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
ately | +0.047 |
ting | +0.033 |
arsh | +0.032 |
nation | +0.032 |
ous | +0.030 |
Neg logit contributions | |
tul | -0.060 |
Notre | -0.059 |
imag | -0.056 |
thirteen | -0.055 |
Ark | -0.055 |
Token | it |
Token ID | 340 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
tul | +0.011 |
bal | +0.010 |
Notre | +0.010 |
esh | +0.010 |
cuc | +0.010 |
Neg logit contributions | |
ately | -0.015 |
nation | -0.013 |
arsh | -0.012 |
ting | -0.011 |
34 | -0.011 |
Token | allowed |
Token ID | 3142 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
ately | +0.024 |
nation | +0.018 |
xy | +0.014 |
arsh | +0.014 |
west | +0.014 |
Neg logit contributions | |
stret | -0.040 |
icol | -0.039 |
Notre | -0.039 |
tul | -0.038 |
thirteen | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | +1.03e-03 |
Pos logit contributions | |
ately | +0.023 |
nation | +0.020 |
arsh | +0.019 |
34 | +0.018 |
upper | +0.018 |
Neg logit contributions | |
tul | -0.017 |
bal | -0.016 |
Notre | -0.015 |
cuc | -0.015 |
te | -0.015 |
Token | but |
Token ID | 475 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
tul | +0.001 |
Notre | +0.001 |
bal | +0.001 |
Ark | +0.001 |
von | +0.001 |
Neg logit contributions | |
ately | -0.001 |
ting | -0.001 |
nation | -0.000 |
upper | -0.000 |
34 | -0.000 |
Token | with |
Token ID | 351 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
ately | +0.050 |
nation | +0.045 |
arsh | +0.043 |
34 | +0.042 |
west | +0.041 |
Neg logit contributions | |
bal | -0.023 |
� | -0.021 |
mo | -0.021 |
tul | -0.020 |
thirteen | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | +7.04e-03 |
Pos logit contributions | |
bal | +0.026 |
tul | +0.025 |
mo | +0.023 |
Notre | +0.023 |
sprink | +0.022 |
Neg logit contributions | |
ately | -0.047 |
nation | -0.041 |
arsh | -0.039 |
west | -0.037 |
34 | -0.037 |
Token | red |
Token ID | 2266 |
Feature activation | -6.44e-03 |
Pos logit contributions | |
tul | +0.008 |
bal | +0.008 |
Notre | +0.008 |
esh | +0.008 |
cuc | +0.007 |
Neg logit contributions | |
ately | -0.007 |
nation | -0.006 |
upper | -0.006 |
west | -0.005 |
usual | -0.005 |
Token | ball |
Token ID | 2613 |
Feature activation | +3.38e-03 |
Pos logit contributions | |
ately | +0.008 |
nation | +0.007 |
arsh | +0.006 |
west | +0.006 |
ting | +0.006 |
Neg logit contributions | |
tul | -0.006 |
bal | -0.006 |
Notre | -0.006 |
cuc | -0.006 |
esh | -0.006 |
Token | in |
Token ID | 287 |
Feature activation | +2.19e-02 |
Pos logit contributions | |
tul | +0.003 |
bal | +0.003 |
Notre | +0.003 |
cuc | +0.003 |
esh | +0.003 |
Neg logit contributions | |
ately | -0.004 |
nation | -0.004 |
arsh | -0.003 |
34 | -0.003 |
ting | -0.003 |
Token | his |
Token ID | 465 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
tul | +0.021 |
bal | +0.021 |
Notre | +0.020 |
cuc | +0.019 |
esh | +0.019 |
Neg logit contributions | |
ately | -0.027 |
nation | -0.023 |
arsh | -0.022 |
34 | -0.021 |
west | -0.021 |
Token | hands |
Token ID | 2832 |
Feature activation | +4.15e-02 |
Pos logit contributions | |
ately | +0.015 |
nation | +0.013 |
arsh | +0.012 |
ting | +0.012 |
34 | +0.012 |
Neg logit contributions | |
bal | -0.010 |
tul | -0.009 |
Notre | -0.009 |
te | -0.009 |
mo | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +5.60e-03 |
Pos logit contributions | |
bal | +0.032 |
tul | +0.028 |
� | +0.028 |
te | +0.027 |
mo | +0.027 |
Neg logit contributions | |
ately | -0.058 |
nation | -0.053 |
west | -0.049 |
arsh | -0.049 |
34 | -0.049 |
Token | Tom |
Token ID | 4186 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
bal | +0.005 |
Notre | +0.005 |
tul | +0.005 |
von | +0.004 |
thirteen | +0.004 |
Neg logit contributions | |
ately | -0.008 |
nation | -0.006 |
arsh | -0.006 |
ting | -0.006 |
34 | -0.006 |
Token | was |
Token ID | 373 |
Feature activation | -4.32e-03 |
Pos logit contributions | |
ately | +0.032 |
nation | +0.024 |
ting | +0.021 |
upper | +0.021 |
arsh | +0.021 |
Neg logit contributions | |
tul | -0.042 |
Notre | -0.040 |
bal | -0.038 |
von | -0.038 |
cuc | -0.038 |
Token | hear |
Token ID | 3285 |
Feature activation | -3.07e-02 |
Pos logit contributions | |
� | +0.006 |
ot | +0.005 |
rob | +0.005 |
help | +0.005 |
even | +0.005 |
Neg logit contributions | |
ately | -0.061 |
ooky | -0.058 |
aghetti | -0.057 |
leys | -0.057 |
year | -0.056 |
Token | this |
Token ID | 428 |
Feature activation | +4.07e-02 |
Pos logit contributions | |
ately | +0.053 |
nation | +0.047 |
ancer | +0.046 |
upper | +0.046 |
arsh | +0.045 |
Neg logit contributions | |
bal | -0.013 |
� | -0.013 |
mo | -0.013 |
bra | -0.012 |
tul | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | +9.38e-03 |
Pos logit contributions | |
bal | +0.033 |
tul | +0.032 |
Notre | +0.030 |
te | +0.030 |
� | +0.030 |
Neg logit contributions | |
ately | -0.056 |
nation | -0.048 |
arsh | -0.046 |
upper | -0.046 |
34 | -0.045 |
Token | apologised |
Token ID | 46397 |
Feature activation | +8.35e-02 |
Pos logit contributions | |
tul | +0.010 |
bal | +0.010 |
Notre | +0.010 |
thirteen | +0.009 |
von | +0.009 |
Neg logit contributions | |
ately | -0.010 |
nation | -0.008 |
arsh | -0.008 |
ting | -0.008 |
west | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
bal | +0.062 |
tul | +0.056 |
esh | +0.052 |
Notre | +0.052 |
cuc | +0.052 |
Neg logit contributions | |
ately | -0.121 |
nation | -0.109 |
upper | -0.100 |
34 | -0.099 |
west | -0.099 |
Token | the |
Token ID | 262 |
Feature activation | +2.77e-02 |
Pos logit contributions | |
mo | +0.001 |
the | +0.001 |
ot | -0.001 |
pap | -0.003 |
cuc | -0.003 |
Neg logit contributions | |
ately | -0.075 |
nation | -0.071 |
uch | -0.070 |
ivery | -0.068 |
ooky | -0.068 |
Token | old |
Token ID | 1468 |
Feature activation | -6.20e-03 |
Pos logit contributions | |
tul | +0.023 |
bal | +0.022 |
Notre | +0.021 |
te | +0.021 |
cuc | +0.020 |
Neg logit contributions | |
ately | -0.039 |
nation | -0.033 |
arsh | -0.032 |
34 | -0.030 |
upper | -0.030 |
Token | man |
Token ID | 582 |
Feature activation | +5.90e-02 |
Pos logit contributions | |
ately | +0.007 |
nation | +0.006 |
arsh | +0.006 |
west | +0.006 |
ting | +0.005 |
Neg logit contributions | |
tul | -0.006 |
bal | -0.006 |
Notre | -0.006 |
thirteen | -0.006 |
imag | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +2.31e-02 |
Pos logit contributions | |
tul | +0.060 |
bal | +0.059 |
Notre | +0.058 |
cuc | +0.055 |
esh | +0.054 |
Neg logit contributions | |
ately | -0.070 |
nation | -0.059 |
arsh | -0.054 |
ting | -0.052 |
34 | -0.052 |
Token |
|
Token ID | 198 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
bal | +0.020 |
tul | +0.020 |
Notre | +0.020 |
von | +0.019 |
cuc | +0.019 |
Neg logit contributions | |
ately | -0.030 |
nation | -0.026 |
34 | -0.023 |
arsh | -0.023 |
upper | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
� | +0.008 |
Notre | +0.007 |
bal | +0.007 |
esh | +0.007 |
#### | +0.007 |
Neg logit contributions | |
ately | -0.021 |
34 | -0.017 |
nation | -0.017 |
ancer | -0.017 |
ting | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
ately | +0.003 |
nation | +0.003 |
arsh | +0.003 |
west | +0.002 |
34 | +0.002 |
Neg logit contributions | |
tul | -0.004 |
bal | -0.003 |
Notre | -0.003 |
esh | -0.003 |
cuc | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
ately | +0.019 |
nation | +0.016 |
arsh | +0.015 |
west | +0.014 |
34 | +0.014 |
Neg logit contributions | |
tul | -0.021 |
bal | -0.021 |
Notre | -0.020 |
cuc | -0.020 |
thirteen | -0.019 |
Token | So |
Token ID | 2396 |
Feature activation | +6.36e-03 |
Pos logit contributions | |
nation | +0.012 |
ately | +0.012 |
west | +0.010 |
oo | +0.009 |
arsh | +0.009 |
Neg logit contributions | |
imag | -0.027 |
icol | -0.027 |
tul | -0.026 |
cuc | -0.026 |
de | -0.025 |
Token | they |
Token ID | 484 |
Feature activation | +5.82e-03 |
Pos logit contributions | |
stret | +0.009 |
EA | +0.009 |
Notre | +0.009 |
tul | +0.009 |
esh | +0.009 |
Neg logit contributions | |
ately | -0.006 |
ting | -0.003 |
usual | -0.003 |
nation | -0.003 |
arsh | -0.002 |
Token | both |
Token ID | 1111 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
tul | +0.005 |
bal | +0.005 |
Notre | +0.005 |
mo | +0.005 |
cuc | +0.005 |
Neg logit contributions | |
ately | -0.008 |
nation | -0.007 |
arsh | -0.006 |
34 | -0.006 |
upper | -0.006 |
Token | closed |
Token ID | 4838 |
Feature activation | -4.98e-02 |
Pos logit contributions | |
ately | +0.026 |
nation | +0.021 |
arsh | +0.020 |
west | +0.019 |
upper | +0.019 |
Neg logit contributions | |
tul | -0.022 |
Notre | -0.022 |
bal | -0.022 |
cuc | -0.021 |
Ark | -0.021 |
Token | their |
Token ID | 511 |
Feature activation | -2.34e-03 |
Pos logit contributions | |
ately | +0.062 |
nation | +0.051 |
arsh | +0.048 |
ting | +0.046 |
34 | +0.046 |
Neg logit contributions | |
bal | -0.048 |
tul | -0.048 |
Notre | -0.046 |
cuc | -0.044 |
esh | -0.043 |
Token | eyes |
Token ID | 2951 |
Feature activation | -4.72e-02 |
Pos logit contributions | |
ately | +0.003 |
nation | +0.002 |
34 | +0.002 |
arsh | +0.002 |
upper | +0.002 |
Neg logit contributions | |
bal | -0.003 |
tul | -0.003 |
Notre | -0.002 |
esh | -0.002 |
cuc | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
nation | +0.074 |
ately | +0.073 |
ancer | +0.070 |
west | +0.070 |
arsh | +0.070 |
Neg logit contributions | |
� | -0.026 |
bal | -0.024 |
te | -0.018 |
� | -0.018 |
mo | -0.018 |
Token | said |
Token ID | 531 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
ately | +0.021 |
nation | +0.020 |
arsh | +0.019 |
34 | +0.019 |
west | +0.018 |
Neg logit contributions | |
mo | -0.011 |
tul | -0.011 |
bal | -0.011 |
imag | -0.010 |
� | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | +7.66e-04 |
Pos logit contributions | |
ately | +0.045 |
nation | +0.035 |
arsh | +0.034 |
34 | +0.032 |
west | +0.031 |
Neg logit contributions | |
tul | -0.041 |
Notre | -0.041 |
bal | -0.040 |
Ark | -0.039 |
cuc | -0.038 |
Token | end |
Token ID | 886 |
Feature activation | +2.31e-02 |
Pos logit contributions | |
ately | +0.004 |
nation | +0.004 |
arsh | +0.003 |
34 | +0.003 |
west | +0.003 |
Neg logit contributions | |
tul | -0.003 |
bal | -0.003 |
Notre | -0.003 |
cuc | -0.003 |
esh | -0.003 |
Token | of |
Token ID | 286 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
tul | +0.028 |
cuc | +0.028 |
Notre | +0.028 |
bal | +0.028 |
Ark | +0.027 |
Neg logit contributions | |
ately | -0.024 |
nation | -0.017 |
34 | -0.016 |
arsh | -0.015 |
west | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | -3.19e-03 |
Pos logit contributions | |
ately | +0.035 |
nation | +0.032 |
arsh | +0.031 |
upper | +0.029 |
34 | +0.029 |
Neg logit contributions | |
tul | -0.016 |
te | -0.014 |
Notre | -0.013 |
mo | -0.013 |
pap | -0.013 |
Token | maze |
Token ID | 31237 |
Feature activation | -9.83e-03 |
Pos logit contributions | |
ately | +0.004 |
nation | +0.003 |
arsh | +0.003 |
34 | +0.003 |
upper | +0.003 |
Neg logit contributions | |
tul | -0.003 |
bal | -0.003 |
Notre | -0.003 |
cuc | -0.003 |
esh | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -5.49e-03 |
Pos logit contributions | |
ately | +0.012 |
nation | +0.010 |
arsh | +0.009 |
34 | +0.009 |
ting | +0.009 |
Neg logit contributions | |
tul | -0.010 |
bal | -0.010 |
Notre | -0.010 |
cuc | -0.009 |
esh | -0.009 |
Token | The |
Token ID | 383 |
Feature activation | -4.06e-02 |
Pos logit contributions | |
ately | +0.006 |
nation | +0.005 |
arsh | +0.005 |
west | +0.004 |
34 | +0.004 |
Neg logit contributions | |
tul | -0.006 |
bal | -0.006 |
Notre | -0.006 |
esh | -0.006 |
cuc | -0.005 |
Token | little |
Token ID | 1310 |
Feature activation | -4.54e-02 |
Pos logit contributions | |
ately | +0.041 |
nation | +0.034 |
34 | +0.031 |
upper | +0.030 |
ting | +0.030 |
Neg logit contributions | |
tul | -0.046 |
esh | -0.045 |
bal | -0.045 |
Notre | -0.044 |
stret | -0.044 |
Token | girl |
Token ID | 2576 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
ately | +0.039 |
ting | +0.030 |
nation | +0.026 |
usual | +0.025 |
ancer | +0.025 |
Neg logit contributions | |
stret | -0.060 |
esh | -0.058 |
Notre | -0.056 |
Ark | -0.056 |
imet | -0.056 |
Token | was |
Token ID | 373 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
tul | +0.043 |
Notre | +0.042 |
esh | +0.042 |
cuc | +0.041 |
thirteen | +0.041 |
Neg logit contributions | |
ately | -0.029 |
nation | -0.018 |
arsh | -0.017 |
upper | -0.015 |
usual | -0.015 |
Token | sad |
Token ID | 6507 |
Feature activation | -7.72e-04 |
Pos logit contributions | |
ately | +0.010 |
nation | +0.008 |
34 | +0.007 |
arsh | +0.007 |
usual | +0.007 |
Neg logit contributions | |
cuc | -0.017 |
Notre | -0.016 |
tul | -0.016 |
imet | -0.016 |
esh | -0.016 |
Token | to |
Token ID | 284 |
Feature activation | +7.66e-03 |
Pos logit contributions | |
ately | +0.001 |
arsh | +0.001 |
nation | +0.001 |
ting | +0.001 |
34 | +0.001 |
Neg logit contributions | |
tul | -0.001 |
Notre | -0.001 |
cuc | -0.001 |
Ark | -0.001 |
esh | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | -7.71e-03 |
Pos logit contributions | |
ately | +0.019 |
nation | +0.016 |
arsh | +0.015 |
upper | +0.014 |
34 | +0.014 |
Neg logit contributions | |
bal | -0.018 |
tul | -0.018 |
Notre | -0.018 |
esh | -0.017 |
von | -0.017 |
Token | apple |
Token ID | 17180 |
Feature activation | -6.05e-02 |
Pos logit contributions | |
ately | +0.006 |
usual | +0.006 |
upper | +0.005 |
nation | +0.005 |
pay | +0.005 |
Neg logit contributions | |
esh | -0.010 |
icol | -0.009 |
cosmetics | -0.009 |
lyn | -0.009 |
Dun | -0.009 |
Token | piece |
Token ID | 3704 |
Feature activation | -4.96e-02 |
Pos logit contributions | |
ately | +0.067 |
nation | +0.055 |
arsh | +0.051 |
upper | +0.049 |
ting | +0.048 |
Neg logit contributions | |
tul | -0.065 |
Notre | -0.063 |
bal | -0.061 |
esh | -0.061 |
von | -0.060 |
Token | . |
Token ID | 13 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
ately | +0.053 |
nation | +0.042 |
arsh | +0.039 |
upper | +0.037 |
34 | +0.036 |
Neg logit contributions | |
tul | -0.057 |
Notre | -0.056 |
esh | -0.053 |
bal | -0.053 |
Ark | -0.052 |
Token | He |
Token ID | 679 |
Feature activation | +4.54e-03 |
Pos logit contributions | |
ately | +0.018 |
nation | +0.015 |
arsh | +0.014 |
34 | +0.013 |
upper | +0.013 |
Neg logit contributions | |
tul | -0.017 |
bal | -0.016 |
Notre | -0.015 |
esh | -0.015 |
cuc | -0.015 |
Token | said |
Token ID | 531 |
Feature activation | -4.98e-02 |
Pos logit contributions | |
tul | +0.006 |
Notre | +0.006 |
esh | +0.006 |
bal | +0.006 |
Ark | +0.006 |
Neg logit contributions | |
ately | -0.004 |
nation | -0.003 |
upper | -0.002 |
west | -0.002 |
rip | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
ately | +0.052 |
arsh | +0.041 |
nation | +0.040 |
34 | +0.037 |
upper | +0.037 |
Neg logit contributions | |
tul | -0.056 |
Notre | -0.056 |
esh | -0.054 |
bal | -0.052 |
Ark | -0.052 |
Token | " |
Token ID | 366 |
Feature activation | -3.73e-02 |
Pos logit contributions | |
ately | +0.021 |
nation | +0.019 |
arsh | +0.017 |
west | +0.017 |
upper | +0.017 |
Neg logit contributions | |
bal | -0.011 |
tul | -0.011 |
cuc | -0.010 |
Notre | -0.010 |
mo | -0.010 |
Token | OK |
Token ID | 11380 |
Feature activation | -5.04e-03 |
Pos logit contributions | |
ately | +0.045 |
nation | +0.038 |
arsh | +0.035 |
upper | +0.034 |
west | +0.034 |
Neg logit contributions | |
tul | -0.037 |
bal | -0.036 |
Notre | -0.035 |
cuc | -0.034 |
mo | -0.034 |
Token | , |
Token ID | 11 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
ately | +0.004 |
nation | +0.003 |
34 | +0.003 |
arsh | +0.003 |
upper | +0.003 |
Neg logit contributions | |
tul | -0.007 |
Notre | -0.007 |
cuc | -0.006 |
spe | -0.006 |
bal | -0.006 |
Token | mom |
Token ID | 1995 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
ately | +0.012 |
nation | +0.008 |
34 | +0.008 |
upper | +0.008 |
ting | +0.007 |
Neg logit contributions | |
esh | -0.019 |
Notre | -0.018 |
von | -0.018 |
spe | -0.018 |
tul | -0.018 |
Token | " |
Token ID | 366 |
Feature activation | -3.48e-02 |
Pos logit contributions | |
ately | +0.020 |
nation | +0.018 |
arsh | +0.016 |
34 | +0.016 |
west | +0.016 |
Neg logit contributions | |
tul | -0.013 |
bal | -0.013 |
cuc | -0.012 |
Notre | -0.012 |
mo | -0.011 |
Token | No |
Token ID | 2949 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
ately | +0.040 |
nation | +0.034 |
arsh | +0.030 |
west | +0.030 |
ting | +0.030 |
Neg logit contributions | |
tul | -0.036 |
bal | -0.036 |
Notre | -0.034 |
te | -0.033 |
cuc | -0.033 |
Token | , |
Token ID | 11 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
tul | +0.019 |
bal | +0.018 |
Notre | +0.018 |
cuc | +0.017 |
esh | +0.017 |
Neg logit contributions | |
ately | -0.020 |
nation | -0.016 |
arsh | -0.016 |
34 | -0.015 |
upper | -0.015 |
Token | Tim |
Token ID | 5045 |
Feature activation | -2.42e-02 |
Pos logit contributions | |
ately | +0.013 |
34 | +0.009 |
nation | +0.008 |
arsh | +0.008 |
upper | +0.008 |
Neg logit contributions | |
tul | -0.018 |
esh | -0.018 |
von | -0.018 |
Notre | -0.018 |
bal | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
ately | +0.024 |
nation | +0.019 |
arsh | +0.018 |
ting | +0.017 |
34 | +0.016 |
Neg logit contributions | |
tul | -0.031 |
bal | -0.029 |
Notre | -0.029 |
cuc | -0.028 |
capt | -0.027 |
Token | it |
Token ID | 340 |
Feature activation | -3.46e-02 |
Pos logit contributions | |
ately | +0.014 |
34 | +0.010 |
nation | +0.009 |
ting | +0.009 |
arsh | +0.009 |
Neg logit contributions | |
von | -0.018 |
esh | -0.018 |
bal | -0.017 |
tul | -0.017 |
Notre | -0.017 |
Token | 's |
Token ID | 338 |
Feature activation | +8.43e-03 |
Pos logit contributions | |
ately | +0.025 |
nation | +0.022 |
uch | +0.020 |
rip | +0.019 |
xy | +0.018 |
Neg logit contributions | |
tul | -0.051 |
paramed | -0.046 |
cuc | -0.045 |
stret | -0.045 |
bal | -0.045 |
Token | not |
Token ID | 407 |
Feature activation | +2.80e-02 |
Pos logit contributions | |
tul | +0.012 |
cuc | +0.011 |
von | +0.011 |
paramed | +0.011 |
bal | +0.011 |
Neg logit contributions | |
ately | -0.006 |
34 | -0.006 |
yet | -0.005 |
nation | -0.005 |
usual | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
tul | +0.031 |
bal | +0.030 |
Notre | +0.029 |
esh | +0.028 |
von | +0.028 |
Neg logit contributions | |
ately | -0.031 |
nation | -0.026 |
arsh | -0.024 |
34 | -0.023 |
west | -0.023 |
Token | pirate |
Token ID | 25868 |
Feature activation | +4.47e-02 |
Pos logit contributions | |
ately | +0.014 |
nation | +0.012 |
34 | +0.012 |
upper | +0.011 |
arsh | +0.011 |
Neg logit contributions | |
tul | -0.012 |
bal | -0.012 |
von | -0.011 |
esh | -0.011 |
Notre | -0.011 |
Token | ship |
Token ID | 4074 |
Feature activation | -9.54e-03 |
Pos logit contributions | |
tul | +0.049 |
von | +0.046 |
Notre | +0.046 |
bal | +0.046 |
esh | +0.044 |
Neg logit contributions | |
ately | -0.049 |
nation | -0.041 |
34 | -0.038 |
ting | -0.037 |
arsh | -0.037 |
Token | little |
Token ID | 1310 |
Feature activation | -3.92e-02 |
Pos logit contributions | |
tul | +0.001 |
bal | +0.001 |
Notre | +0.001 |
cuc | +0.001 |
mo | +0.001 |
Neg logit contributions | |
ately | -0.001 |
nation | -0.001 |
arsh | -0.001 |
ting | -0.001 |
34 | -0.001 |
Token | boy |
Token ID | 2933 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
ately | +0.047 |
nation | +0.039 |
ting | +0.038 |
arsh | +0.036 |
west | +0.036 |
Neg logit contributions | |
esh | -0.036 |
tul | -0.035 |
Notre | -0.035 |
bal | -0.035 |
thirteen | -0.034 |
Token | went |
Token ID | 1816 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
tul | +0.040 |
bal | +0.039 |
Notre | +0.038 |
cuc | +0.036 |
esh | +0.036 |
Neg logit contributions | |
ately | -0.059 |
nation | -0.050 |
arsh | -0.047 |
ting | -0.045 |
upper | -0.045 |
Token | outside |
Token ID | 2354 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
tul | +0.015 |
bal | +0.015 |
Notre | +0.014 |
thirteen | +0.014 |
spe | +0.014 |
Neg logit contributions | |
ately | -0.013 |
nation | -0.011 |
ting | -0.011 |
arsh | -0.010 |
west | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
ately | +0.024 |
nation | +0.019 |
arsh | +0.018 |
ting | +0.018 |
west | +0.017 |
Neg logit contributions | |
tul | -0.029 |
bal | -0.029 |
thirteen | -0.029 |
Notre | -0.028 |
Ark | -0.028 |
Token | play |
Token ID | 711 |
Feature activation | -4.67e-02 |
Pos logit contributions | |
tul | +0.013 |
bal | +0.013 |
Notre | +0.012 |
cuc | +0.012 |
mo | +0.012 |
Neg logit contributions | |
ately | -0.015 |
nation | -0.013 |
arsh | -0.012 |
34 | -0.012 |
ting | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
ately | +0.063 |
nation | +0.053 |
arsh | +0.051 |
34 | +0.049 |
upper | +0.048 |
Neg logit contributions | |
bal | -0.040 |
tul | -0.040 |
Notre | -0.038 |
cuc | -0.037 |
te | -0.037 |
Token | He |
Token ID | 679 |
Feature activation | +8.37e-03 |
Pos logit contributions | |
ately | +0.011 |
nation | +0.009 |
arsh | +0.009 |
upper | +0.009 |
west | +0.009 |
Neg logit contributions | |
tul | -0.013 |
bal | -0.013 |
Notre | -0.012 |
esh | -0.012 |
Ark | -0.012 |
Token | saw |
Token ID | 2497 |
Feature activation | -8.21e-03 |
Pos logit contributions | |
tul | +0.010 |
bal | +0.010 |
Notre | +0.010 |
thirteen | +0.010 |
esh | +0.010 |
Neg logit contributions | |
ately | -0.008 |
nation | -0.007 |
arsh | -0.006 |
west | -0.006 |
upper | -0.006 |
Token | something |
Token ID | 1223 |
Feature activation | -4.35e-02 |
Pos logit contributions | |
ately | +0.010 |
nation | +0.009 |
arsh | +0.008 |
34 | +0.008 |
west | +0.008 |
Neg logit contributions | |
tul | -0.008 |
bal | -0.008 |
Notre | -0.007 |
cuc | -0.007 |
esh | -0.007 |
Token | funny |
Token ID | 8258 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
ately | +0.043 |
nation | +0.037 |
ting | +0.032 |
west | +0.032 |
usual | +0.031 |
Neg logit contributions | |
tul | -0.047 |
Notre | -0.047 |
von | -0.047 |
esh | -0.047 |
majesty | -0.046 |
Token | They |
Token ID | 1119 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
Notre | +0.003 |
de | +0.003 |
pap | +0.003 |
bal | +0.002 |
marsh | +0.002 |
Neg logit contributions | |
ately | -0.008 |
nation | -0.008 |
owing | -0.007 |
ting | -0.007 |
uch | -0.007 |
Token | said |
Token ID | 531 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
esh | +0.015 |
bal | +0.014 |
Notre | +0.013 |
von | +0.013 |
lyn | +0.013 |
Neg logit contributions | |
ately | -0.009 |
upper | -0.006 |
nation | -0.006 |
usual | -0.006 |
west | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | +4.27e-03 |
Pos logit contributions | |
ately | +0.026 |
arsh | +0.019 |
34 | +0.017 |
nation | +0.017 |
west | +0.017 |
Neg logit contributions | |
esh | -0.034 |
Notre | -0.033 |
cosmetics | -0.032 |
stret | -0.032 |
Ark | -0.032 |
Token | " |
Token ID | 366 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
tul | +0.003 |
bal | +0.003 |
cuc | +0.003 |
Notre | +0.003 |
mo | +0.003 |
Neg logit contributions | |
ately | -0.006 |
nation | -0.005 |
arsh | -0.005 |
upper | -0.005 |
34 | -0.005 |
Token | M |
Token ID | 44 |
Feature activation | -7.25e-03 |
Pos logit contributions | |
ately | +0.022 |
nation | +0.019 |
arsh | +0.018 |
34 | +0.017 |
ting | +0.017 |
Neg logit contributions | |
tul | -0.014 |
bal | -0.013 |
Notre | -0.013 |
cuc | -0.012 |
thirteen | -0.012 |
Token | mm |
Token ID | 3020 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
ately | +0.007 |
nation | +0.007 |
ting | +0.007 |
upper | +0.006 |
west | +0.006 |
Neg logit contributions | |
bal | -0.008 |
tul | -0.008 |
von | -0.008 |
de | -0.008 |
paramed | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | +6.28e-03 |
Pos logit contributions | |
ately | +0.268 |
nation | +0.235 |
arsh | +0.217 |
upper | +0.211 |
west | +0.211 |
Neg logit contributions | |
tul | -0.164 |
bal | -0.163 |
Notre | -0.151 |
cuc | -0.148 |
te | -0.144 |
Token | this |
Token ID | 428 |
Feature activation | +3.35e-02 |
Pos logit contributions | |
tul | +0.005 |
bal | +0.005 |
Notre | +0.005 |
cuc | +0.005 |
te | +0.005 |
Neg logit contributions | |
ately | -0.009 |
nation | -0.008 |
arsh | -0.007 |
upper | -0.007 |
west | -0.007 |
Token | is |
Token ID | 318 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
tul | +0.038 |
Notre | +0.037 |
bal | +0.037 |
esh | +0.036 |
Ark | +0.036 |
Neg logit contributions | |
ately | -0.031 |
nation | -0.027 |
arsh | -0.025 |
upper | -0.024 |
west | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
bal | +0.029 |
tul | +0.029 |
Notre | +0.028 |
esh | +0.028 |
Ark | +0.027 |
Neg logit contributions | |
ately | -0.028 |
nation | -0.024 |
34 | -0.023 |
arsh | -0.022 |
west | -0.021 |
Token | best |
Token ID | 1266 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
esh | +0.027 |
cosmetics | +0.025 |
Hosp | +0.024 |
tin | +0.024 |
Dun | +0.024 |
Neg logit contributions | |
usual | -0.010 |
upper | -0.010 |
first | -0.009 |
34 | -0.009 |
other | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
ately | +0.004 |
nation | +0.003 |
arsh | +0.003 |
upper | +0.003 |
ting | +0.003 |
Neg logit contributions | |
tul | -0.002 |
bal | -0.002 |
Notre | -0.002 |
mo | -0.002 |
te | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
ately | +0.012 |
nation | +0.010 |
arsh | +0.009 |
west | +0.009 |
34 | +0.008 |
Neg logit contributions | |
bal | -0.013 |
tul | -0.013 |
Notre | -0.013 |
cuc | -0.012 |
esh | -0.012 |
Token | Sam |
Token ID | 16305 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
ately | +0.010 |
nation | +0.009 |
west | +0.008 |
34 | +0.008 |
arsh | +0.008 |
Neg logit contributions | |
tul | -0.014 |
bal | -0.014 |
imag | -0.014 |
cuc | -0.013 |
unbeat | -0.013 |
Token | reached |
Token ID | 4251 |
Feature activation | +5.68e-03 |
Pos logit contributions | |
ately | +0.043 |
nation | +0.034 |
ting | +0.034 |
arsh | +0.031 |
rip | +0.030 |
Neg logit contributions | |
tul | -0.056 |
bal | -0.054 |
cuc | -0.052 |
Notre | -0.051 |
Ark | -0.050 |
Token | R |
Token ID | 371 |
Feature activation | -4.43e-02 |
Pos logit contributions | |
tul | +0.008 |
spe | +0.008 |
bal | +0.008 |
Notre | +0.008 |
paramed | +0.008 |
Neg logit contributions | |
ately | -0.004 |
ting | -0.003 |
usual | -0.003 |
arsh | -0.003 |
rip | -0.003 |
Token | oo |
Token ID | 2238 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
ately | +0.047 |
nation | +0.045 |
ting | +0.039 |
arsh | +0.039 |
west | +0.038 |
Neg logit contributions | |
bal | -0.046 |
te | -0.043 |
Notre | -0.042 |
dod | -0.042 |
von | -0.041 |
Token | and |
Token ID | 290 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
ately | +0.254 |
nation | +0.207 |
ting | +0.198 |
arsh | +0.195 |
34 | +0.183 |
Neg logit contributions | |
tul | -0.238 |
Notre | -0.227 |
bal | -0.226 |
cuc | -0.216 |
Ark | -0.212 |
Token | picked |
Token ID | 6497 |
Feature activation | -1.09e-03 |
Pos logit contributions | |
ately | +0.015 |
nation | +0.013 |
ting | +0.012 |
arsh | +0.012 |
west | +0.011 |
Neg logit contributions | |
tul | -0.013 |
bal | -0.013 |
Notre | -0.013 |
esh | -0.012 |
von | -0.012 |
Token | it |
Token ID | 340 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
ately | +0.001 |
nation | +0.001 |
arsh | +0.001 |
34 | +0.001 |
upper | +0.001 |
Neg logit contributions | |
tul | -0.001 |
bal | -0.001 |
Notre | -0.001 |
cuc | -0.001 |
te | -0.001 |
Token | up |
Token ID | 510 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
ately | +0.033 |
nation | +0.028 |
ting | +0.025 |
arsh | +0.025 |
west | +0.025 |
Neg logit contributions | |
tul | -0.031 |
Notre | -0.030 |
bal | -0.028 |
cuc | -0.028 |
thirteen | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -6.54e-03 |
Pos logit contributions | |
tul | +0.029 |
bal | +0.029 |
Notre | +0.028 |
cuc | +0.026 |
esh | +0.026 |
Neg logit contributions | |
ately | -0.036 |
nation | -0.031 |
arsh | -0.029 |
34 | -0.027 |
upper | -0.027 |
Token | had |
Token ID | 550 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
bal | +0.025 |
� | +0.024 |
tul | +0.024 |
cuc | +0.023 |
mo | +0.022 |
Neg logit contributions | |
ately | -0.044 |
nation | -0.039 |
arsh | -0.037 |
34 | -0.037 |
upper | -0.036 |
Token | made |
Token ID | 925 |
Feature activation | +5.41e-02 |
Pos logit contributions | |
tul | +0.014 |
bal | +0.013 |
Notre | +0.013 |
thirteen | +0.012 |
esh | +0.012 |
Neg logit contributions | |
ately | -0.013 |
nation | -0.011 |
arsh | -0.010 |
34 | -0.009 |
upper | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +8.97e-03 |
Pos logit contributions | |
tul | +0.061 |
Notre | +0.058 |
bal | +0.057 |
cuc | +0.055 |
thirteen | +0.055 |
Neg logit contributions | |
ately | -0.060 |
nation | -0.050 |
arsh | -0.047 |
ting | -0.044 |
34 | -0.043 |
Token | And |
Token ID | 843 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
tul | +0.010 |
bal | +0.010 |
Notre | +0.010 |
esh | +0.010 |
cuc | +0.010 |
Neg logit contributions | |
ately | -0.009 |
nation | -0.007 |
arsh | -0.007 |
west | -0.007 |
ting | -0.006 |
Token | as |
Token ID | 355 |
Feature activation | +5.87e-02 |
Pos logit contributions | |
ately | +0.013 |
nation | +0.010 |
arsh | +0.009 |
ting | +0.009 |
west | +0.009 |
Neg logit contributions | |
Notre | -0.011 |
tul | -0.011 |
esh | -0.011 |
bal | -0.011 |
stret | -0.011 |
Token | usual |
Token ID | 6678 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
bal | +0.039 |
tul | +0.036 |
� | +0.033 |
pap | +0.032 |
mo | +0.029 |
Neg logit contributions | |
ately | -0.092 |
nation | -0.086 |
upper | -0.083 |
34 | -0.081 |
west | -0.080 |
Token | , |
Token ID | 11 |
Feature activation | -4.39e-03 |
Pos logit contributions | |
ately | +0.208 |
ting | +0.159 |
nation | +0.136 |
arsh | +0.122 |
usual | +0.113 |
Neg logit contributions | |
von | -0.277 |
Notre | -0.275 |
esh | -0.271 |
EA | -0.270 |
Ark | -0.266 |
Token | her |
Token ID | 607 |
Feature activation | -1.29e-03 |
Pos logit contributions | |
ately | +0.005 |
nation | +0.004 |
ting | +0.004 |
arsh | +0.003 |
usual | +0.003 |
Neg logit contributions | |
Notre | -0.005 |
esh | -0.005 |
tul | -0.005 |
thirteen | -0.005 |
bal | -0.005 |
Token | little |
Token ID | 1310 |
Feature activation | -4.33e-02 |
Pos logit contributions | |
ately | +0.002 |
upper | +0.001 |
nation | +0.001 |
usual | +0.001 |
arsh | +0.001 |
Neg logit contributions | |
esh | -0.001 |
Notre | -0.001 |
thirteen | -0.001 |
tul | -0.001 |
Ark | -0.001 |
Token | brother |
Token ID | 3956 |
Feature activation | +4.55e-02 |
Pos logit contributions | |
ately | +0.039 |
upper | +0.026 |
usual | +0.024 |
ancer | +0.023 |
other | +0.022 |
Neg logit contributions | |
esh | -0.061 |
stret | -0.057 |
Notre | -0.055 |
Dun | -0.055 |
Ark | -0.054 |
Token | was |
Token ID | 373 |
Feature activation | -8.99e-03 |
Pos logit contributions | |
Notre | +0.055 |
tul | +0.054 |
esh | +0.051 |
bal | +0.050 |
Ark | +0.050 |
Neg logit contributions | |
ately | -0.045 |
ting | -0.036 |
nation | -0.033 |
arsh | -0.032 |
rip | -0.028 |
Token | loud |
Token ID | 7812 |
Feature activation | -7.53e-02 |
Pos logit contributions | |
ately | +0.001 |
nation | +0.001 |
ancer | +0.001 |
ting | +0.001 |
arsh | +0.001 |
Neg logit contributions | |
bal | -0.001 |
� | -0.000 |
dod | -0.000 |
de | -0.000 |
thirteen | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
ately | +0.084 |
nation | +0.071 |
arsh | +0.065 |
ting | +0.064 |
34 | +0.063 |
Neg logit contributions | |
tul | -0.081 |
bal | -0.077 |
Notre | -0.077 |
cuc | -0.074 |
Ark | -0.074 |
Token | " |
Token ID | 366 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
ately | +0.029 |
nation | +0.026 |
ting | +0.024 |
owing | +0.024 |
ancer | +0.023 |
Neg logit contributions | |
bal | -0.009 |
mo | -0.007 |
von | -0.007 |
Notre | -0.007 |
thirteen | -0.006 |
Token | A |
Token ID | 32 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
ately | +0.042 |
nation | +0.035 |
arsh | +0.032 |
34 | +0.031 |
west | +0.031 |
Neg logit contributions | |
tul | -0.036 |
bal | -0.036 |
Notre | -0.034 |
cuc | -0.033 |
esh | -0.032 |
Token | ch |
Token ID | 354 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
ately | +0.179 |
nation | +0.151 |
arsh | +0.139 |
ting | +0.138 |
34 | +0.136 |
Neg logit contributions | |
bal | -0.105 |
tul | -0.098 |
mo | -0.092 |
Notre | -0.092 |
te | -0.089 |
Token | oo |
Token ID | 2238 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
ately | +0.009 |
nation | +0.005 |
arsh | +0.005 |
oo | +0.004 |
rip | +0.004 |
Neg logit contributions | |
tul | -0.031 |
Notre | -0.030 |
bal | -0.030 |
cuc | -0.029 |
Ark | -0.029 |
Token | !" |
Token ID | 2474 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
ately | +0.220 |
nation | +0.182 |
oo | +0.181 |
arsh | +0.177 |
ting | +0.172 |
Neg logit contributions | |
tul | -0.263 |
Notre | -0.252 |
Ark | -0.245 |
cuc | -0.241 |
bal | -0.240 |
Token |
|
Token ID | 198 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
tul | +0.050 |
unbeat | +0.046 |
esh | +0.046 |
spe | +0.046 |
imag | +0.045 |
Neg logit contributions | |
ately | -0.025 |
nation | -0.016 |
arsh | -0.015 |
34 | -0.015 |
yet | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
ately | +0.032 |
nation | +0.028 |
ting | +0.026 |
arsh | +0.026 |
34 | +0.026 |
Neg logit contributions | |
bal | -0.018 |
tul | -0.017 |
Notre | -0.016 |
esh | -0.016 |
� | -0.016 |
Token | His |
Token ID | 6653 |
Feature activation | -7.66e-02 |
Pos logit contributions | |
ately | +0.027 |
nation | +0.023 |
arsh | +0.021 |
34 | +0.020 |
upper | +0.020 |
Neg logit contributions | |
tul | -0.021 |
bal | -0.021 |
Notre | -0.020 |
cuc | -0.019 |
te | -0.019 |
Token | sne |
Token ID | 10505 |
Feature activation | +3.39e-02 |
Pos logit contributions | |
ately | +0.097 |
nation | +0.081 |
upper | +0.077 |
arsh | +0.076 |
34 | +0.075 |
Neg logit contributions | |
tul | -0.070 |
bal | -0.068 |
Notre | -0.067 |
esh | -0.066 |
cuc | -0.064 |
Token | noticed |
Token ID | 6810 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
ately | +0.028 |
nation | +0.024 |
arsh | +0.023 |
34 | +0.022 |
upper | +0.022 |
Neg logit contributions | |
tul | -0.021 |
bal | -0.020 |
Notre | -0.020 |
mo | -0.019 |
cuc | -0.019 |
Token | a |
Token ID | 257 |
Feature activation | -9.75e-03 |
Pos logit contributions | |
ately | +0.021 |
nation | +0.019 |
arsh | +0.018 |
upper | +0.018 |
34 | +0.017 |
Neg logit contributions | |
tul | -0.013 |
bal | -0.012 |
esh | -0.011 |
Notre | -0.011 |
mo | -0.011 |
Token | little |
Token ID | 1310 |
Feature activation | -4.94e-02 |
Pos logit contributions | |
ately | +0.013 |
nation | +0.011 |
arsh | +0.010 |
ting | +0.010 |
34 | +0.010 |
Neg logit contributions | |
tul | -0.009 |
bal | -0.009 |
Notre | -0.008 |
mo | -0.008 |
te | -0.008 |
Token | kitten |
Token ID | 42143 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
ately | +0.056 |
nation | +0.045 |
ting | +0.044 |
34 | +0.042 |
upper | +0.041 |
Neg logit contributions | |
bal | -0.048 |
Notre | -0.048 |
esh | -0.047 |
tul | -0.047 |
#### | -0.046 |
Token | me |
Token ID | 502 |
Feature activation | -4.50e-04 |
Pos logit contributions | |
bal | +0.013 |
Notre | +0.013 |
cuc | +0.013 |
thirteen | +0.013 |
von | +0.013 |
Neg logit contributions | |
ately | -0.012 |
ting | -0.010 |
nation | -0.009 |
arsh | -0.008 |
owing | -0.008 |
Token | owing |
Token ID | 7855 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
ately | +0.000 |
34 | +0.000 |
west | +0.000 |
mm | +0.000 |
nation | +0.000 |
Neg logit contributions | |
� | -0.000 |
bal | -0.000 |
gn | -0.000 |
esh | -0.000 |
tul | -0.000 |
Token | for |
Token ID | 329 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
ately | +0.249 |
nation | +0.210 |
ting | +0.190 |
arsh | +0.189 |
west | +0.183 |
Neg logit contributions | |
tul | -0.246 |
bal | -0.243 |
Notre | -0.238 |
cuc | -0.231 |
Ark | -0.225 |
Token | help |
Token ID | 1037 |
Feature activation | +9.61e-03 |
Pos logit contributions | |
ately | +0.032 |
ancer | +0.028 |
nation | +0.028 |
arsh | +0.028 |
34 | +0.027 |
Neg logit contributions | |
tul | -0.014 |
bal | -0.012 |
mo | -0.012 |
gn | -0.012 |
Notre | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
bal | +0.007 |
tul | +0.007 |
� | +0.007 |
Notre | +0.006 |
esh | +0.006 |
Neg logit contributions | |
ately | -0.014 |
nation | -0.012 |
34 | -0.011 |
arsh | -0.011 |
west | -0.011 |
Token | It |
Token ID | 632 |
Feature activation | -4.60e-02 |
Pos logit contributions | |
ately | +0.015 |
nation | +0.013 |
upper | +0.012 |
34 | +0.012 |
arsh | +0.012 |
Neg logit contributions | |
tul | -0.010 |
bal | -0.010 |
� | -0.009 |
Notre | -0.009 |
pap | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
ately | +0.041 |
nation | +0.035 |
ting | +0.030 |
west | +0.028 |
arsh | +0.027 |
Neg logit contributions | |
tul | -0.057 |
bal | -0.056 |
thirteen | -0.056 |
Notre | -0.055 |
cuc | -0.054 |