Token | � |
Token ID | 95 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
Ho | +0.064 |
Rib | +0.062 |
H | +0.057 |
Bub | +0.053 |
Bun | +0.052 |
Neg logit contributions | |
eks | -0.051 |
oos | -0.050 |
ays | -0.049 |
cases | -0.049 |
ases | -0.049 |
Token | € |
Token ID | 26391 |
Feature activation | -8.98e-02 |
Pos logit contributions | |
Ho | +0.072 |
Rib | +0.071 |
H | +0.065 |
Bub | +0.059 |
Bun | +0.059 |
Neg logit contributions | |
eks | -0.057 |
oos | -0.057 |
ays | -0.056 |
cases | -0.056 |
ases | -0.055 |
Token | � |
Token ID | 129 |
Feature activation | -3.91e-02 |
Pos logit contributions | |
Ho | +0.152 |
Rib | +0.148 |
H | +0.134 |
Bun | +0.122 |
Bub | +0.122 |
Neg logit contributions | |
eks | -0.144 |
oos | -0.143 |
elevator | -0.140 |
ays | -0.139 |
ases | -0.139 |
Token | � |
Token ID | 241 |
Feature activation | +5.11e-02 |
Pos logit contributions | |
Ho | +0.081 |
Rib | +0.079 |
H | +0.073 |
Bub | +0.068 |
Bun | +0.068 |
Neg logit contributions | |
eks | -0.048 |
oos | -0.048 |
ays | -0.047 |
cases | -0.046 |
elevator | -0.046 |
Token | H |
Token ID | 39 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
he | +0.077 |
oos | +0.073 |
ases | +0.073 |
hed | +0.072 |
elevator | +0.072 |
Neg logit contributions | |
Ho | -0.092 |
Rib | -0.080 |
Bun | -0.075 |
Ros | -0.075 |
Bub | -0.073 |
Token | oor |
Token ID | 2675 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
Ho | +0.041 |
Rib | +0.041 |
H | +0.038 |
Bub | +0.035 |
Bun | +0.035 |
Neg logit contributions | |
eks | -0.022 |
oos | -0.022 |
cases | -0.022 |
elevator | -0.021 |
ays | -0.021 |
Token | ay |
Token ID | 323 |
Feature activation | +7.10e-02 |
Pos logit contributions | |
eks | +0.133 |
oos | +0.118 |
ases | +0.117 |
elevator | +0.110 |
cks | +0.108 |
Neg logit contributions | |
Ho | -0.437 |
Rib | -0.427 |
H | -0.413 |
Bun | -0.387 |
Lucky | -0.383 |
Token | ! |
Token ID | 0 |
Feature activation | +3.73e-02 |
Pos logit contributions | |
oos | +0.100 |
eks | +0.099 |
ays | +0.097 |
cases | +0.097 |
agged | +0.097 |
Neg logit contributions | |
Ho | -0.134 |
Rib | -0.132 |
H | -0.121 |
Bun | -0.112 |
Bub | -0.111 |
Token | I |
Token ID | 314 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
ays | +0.044 |
ases | +0.042 |
he | +0.040 |
ay | +0.040 |
are | +0.038 |
Neg logit contributions | |
Ho | -0.065 |
accidentally | -0.065 |
Rib | -0.063 |
onna | -0.060 |
please | -0.060 |
Token | found |
Token ID | 1043 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
Ho | +0.049 |
Rib | +0.048 |
H | +0.043 |
Bub | +0.040 |
Bun | +0.039 |
Neg logit contributions | |
oos | -0.053 |
eks | -0.053 |
ays | -0.052 |
cases | -0.052 |
elevator | -0.052 |
Token | a |
Token ID | 257 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
eks | +0.017 |
elevator | +0.016 |
ays | +0.016 |
oos | +0.016 |
ases | +0.016 |
Neg logit contributions | |
Rib | -0.022 |
Ho | -0.022 |
H | -0.020 |
accidentally | -0.019 |
Bun | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
eks | +0.061 |
oos | +0.059 |
ays | +0.059 |
ases | +0.058 |
elevator | +0.058 |
Neg logit contributions | |
Ho | -0.069 |
Rib | -0.067 |
H | -0.061 |
Bub | -0.057 |
Bun | -0.056 |
Token | cheered |
Token ID | 34767 |
Feature activation | +4.23e-02 |
Pos logit contributions | |
eks | +0.020 |
led | +0.019 |
hed | +0.019 |
cases | +0.019 |
oos | +0.019 |
Neg logit contributions | |
Rib | -0.027 |
Ho | -0.027 |
H | -0.024 |
Butter | -0.023 |
Bub | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | -2.78e-03 |
Pos logit contributions | |
oos | +0.063 |
eks | +0.062 |
cases | +0.060 |
ays | +0.060 |
ases | +0.060 |
Neg logit contributions | |
Ho | -0.077 |
Rib | -0.075 |
H | -0.067 |
Bub | -0.063 |
Bun | -0.062 |
Token | " |
Token ID | 366 |
Feature activation | +4.83e-03 |
Pos logit contributions | |
Ho | +0.005 |
Rib | +0.005 |
H | +0.004 |
Bub | +0.004 |
Bun | +0.004 |
Neg logit contributions | |
cases | -0.004 |
eks | -0.004 |
hed | -0.004 |
elevator | -0.004 |
oos | -0.004 |
Token | H |
Token ID | 39 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
hed | +0.007 |
he | +0.007 |
oos | +0.006 |
ases | +0.006 |
eks | +0.006 |
Neg logit contributions | |
Ho | -0.009 |
Rib | -0.009 |
H | -0.008 |
accidentally | -0.008 |
onna | -0.008 |
Token | oor |
Token ID | 2675 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
Ho | +0.046 |
Rib | +0.046 |
H | +0.044 |
Bub | +0.041 |
Bun | +0.040 |
Neg logit contributions | |
cases | -0.026 |
elevator | -0.025 |
eks | -0.025 |
ays | -0.024 |
oos | -0.023 |
Token | ay |
Token ID | 323 |
Feature activation | +6.85e-02 |
Pos logit contributions | |
eks | +0.122 |
oos | +0.108 |
ases | +0.107 |
elevator | +0.107 |
cases | +0.098 |
Neg logit contributions | |
Ho | -0.448 |
Rib | -0.438 |
H | -0.420 |
Bun | -0.400 |
Bub | -0.393 |
Token | ! |
Token ID | 0 |
Feature activation | +3.92e-02 |
Pos logit contributions | |
cases | +0.104 |
ays | +0.102 |
roads | +0.102 |
elevator | +0.101 |
oos | +0.101 |
Neg logit contributions | |
Ho | -0.124 |
Rib | -0.121 |
H | -0.109 |
Bun | -0.103 |
Tw | -0.101 |
Token | The |
Token ID | 383 |
Feature activation | -7.38e-03 |
Pos logit contributions | |
ays | +0.047 |
ases | +0.045 |
he | +0.044 |
hed | +0.044 |
oos | +0.042 |
Neg logit contributions | |
Ho | -0.074 |
Rib | -0.071 |
accidentally | -0.067 |
onna | -0.062 |
please | -0.061 |
Token | fairy |
Token ID | 25607 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
Rib | +0.013 |
Ho | +0.013 |
H | +0.012 |
Stri | +0.011 |
Butter | +0.011 |
Neg logit contributions | |
elevator | -0.011 |
eks | -0.011 |
oos | -0.010 |
he | -0.010 |
ases | -0.010 |
Token | has |
Token ID | 468 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
Ho | +0.194 |
Rib | +0.190 |
H | +0.172 |
Bub | +0.160 |
Bun | +0.160 |
Neg logit contributions | |
oos | -0.149 |
eks | -0.147 |
ays | -0.144 |
cases | -0.143 |
elevator | -0.143 |
Token | . |
Token ID | 13 |
Feature activation | -9.95e-03 |
Pos logit contributions | |
oos | +0.056 |
cases | +0.054 |
eks | +0.053 |
elevator | +0.053 |
ays | +0.053 |
Neg logit contributions | |
Ho | -0.084 |
Rib | -0.084 |
H | -0.074 |
Bub | -0.070 |
Butter | -0.070 |
Token |
|
Token ID | 198 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
Ho | +0.020 |
Rib | +0.019 |
H | +0.017 |
Bun | +0.017 |
accidentally | +0.016 |
Neg logit contributions | |
oos | -0.012 |
agged | -0.012 |
ays | -0.012 |
ases | -0.012 |
ags | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -2.42e-02 |
Pos logit contributions | |
Ho | +0.032 |
Rib | +0.032 |
H | +0.026 |
Butter | +0.024 |
accidentally | +0.023 |
Neg logit contributions | |
he | -0.047 |
ases | -0.046 |
eks | -0.046 |
oos | -0.046 |
ays | -0.045 |
Token | " |
Token ID | 1 |
Feature activation | +4.47e-02 |
Pos logit contributions | |
Ho | +0.036 |
Rib | +0.035 |
H | +0.031 |
Bub | +0.029 |
Bun | +0.029 |
Neg logit contributions | |
eks | -0.044 |
oos | -0.044 |
ays | -0.043 |
cases | -0.043 |
elevator | -0.043 |
Token | H |
Token ID | 39 |
Feature activation | -2.33e-02 |
Pos logit contributions | |
agged | +0.060 |
ases | +0.059 |
oos | +0.059 |
hed | +0.058 |
he | +0.058 |
Neg logit contributions | |
Ho | -0.086 |
Rib | -0.079 |
H | -0.075 |
Bun | -0.070 |
accidentally | -0.069 |
Token | oor |
Token ID | 2675 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
Rib | +0.038 |
Ho | +0.037 |
H | +0.035 |
Bub | +0.034 |
Fro | +0.032 |
Neg logit contributions | |
eks | -0.038 |
cases | -0.036 |
elevator | -0.036 |
ps | -0.035 |
bags | -0.035 |
Token | ay |
Token ID | 323 |
Feature activation | +7.08e-02 |
Pos logit contributions | |
eks | +0.181 |
cks | +0.172 |
ps | +0.167 |
ases | +0.159 |
oos | +0.155 |
Neg logit contributions | |
Ho | -0.377 |
Rib | -0.364 |
H | -0.362 |
Bun | -0.338 |
Lucky | -0.331 |
Token | , |
Token ID | 11 |
Feature activation | -5.37e-03 |
Pos logit contributions | |
ays | +0.120 |
agged | +0.119 |
oos | +0.118 |
cases | +0.118 |
eks | +0.118 |
Neg logit contributions | |
Ho | -0.114 |
Rib | -0.112 |
H | -0.103 |
Bun | -0.094 |
Bub | -0.091 |
Token | h |
Token ID | 289 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
Ho | +0.008 |
Rib | +0.008 |
H | +0.007 |
Bun | +0.006 |
accidentally | +0.006 |
Neg logit contributions | |
elevator | -0.010 |
eks | -0.010 |
oos | -0.009 |
cases | -0.009 |
ases | -0.009 |
Token | oor |
Token ID | 2675 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
Rib | +0.061 |
Ho | +0.059 |
H | +0.055 |
Bub | +0.053 |
Stri | +0.051 |
Neg logit contributions | |
eks | -0.052 |
led | -0.049 |
aks | -0.049 |
roads | -0.049 |
igators | -0.048 |
Token | ay |
Token ID | 323 |
Feature activation | +7.05e-02 |
Pos logit contributions | |
eks | +0.176 |
cks | +0.156 |
ases | +0.153 |
ps | +0.150 |
oos | +0.148 |
Neg logit contributions | |
Ho | -0.388 |
Rib | -0.377 |
H | -0.372 |
Bun | -0.352 |
Lucky | -0.340 |
Token | mom |
Token ID | 1995 |
Feature activation | +6.17e-02 |
Pos logit contributions | |
Ho | +0.012 |
Rib | +0.012 |
H | +0.011 |
Bun | +0.010 |
Bub | +0.010 |
Neg logit contributions | |
eks | -0.012 |
elevator | -0.012 |
ays | -0.012 |
oos | -0.012 |
cases | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +9.52e-03 |
Pos logit contributions | |
eks | +0.088 |
oos | +0.087 |
ays | +0.085 |
elevator | +0.085 |
cases | +0.084 |
Neg logit contributions | |
Ho | -0.116 |
Rib | -0.114 |
H | -0.103 |
Bub | -0.096 |
Bun | -0.095 |
Token | dad |
Token ID | 9955 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
eks | +0.015 |
oos | +0.015 |
ays | +0.015 |
cases | +0.015 |
elevator | +0.015 |
Neg logit contributions | |
Ho | -0.016 |
Rib | -0.016 |
H | -0.014 |
Bub | -0.013 |
Bun | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -6.19e-03 |
Pos logit contributions | |
oos | +0.170 |
ays | +0.163 |
agged | +0.163 |
eks | +0.163 |
cases | +0.160 |
Neg logit contributions | |
Ho | -0.181 |
Rib | -0.174 |
H | -0.160 |
Bub | -0.148 |
accidentally | -0.147 |
Token | h |
Token ID | 289 |
Feature activation | -3.63e-02 |
Pos logit contributions | |
Ho | +0.011 |
Rib | +0.011 |
H | +0.010 |
Bub | +0.009 |
Bun | +0.009 |
Neg logit contributions | |
eks | -0.009 |
oos | -0.009 |
elevator | -0.009 |
ases | -0.009 |
ays | -0.009 |
Token | oor |
Token ID | 2675 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
Rib | +0.065 |
Ho | +0.063 |
H | +0.057 |
Tw | +0.053 |
Bub | +0.053 |
Neg logit contributions | |
eks | -0.050 |
roads | -0.050 |
igators | -0.050 |
aks | -0.047 |
elevator | -0.047 |
Token | ay |
Token ID | 323 |
Feature activation | +7.06e-02 |
Pos logit contributions | |
eks | +0.183 |
cks | +0.175 |
ases | +0.169 |
oos | +0.168 |
ps | +0.161 |
Neg logit contributions | |
Ho | -0.358 |
H | -0.339 |
Rib | -0.339 |
Lucky | -0.311 |
Bun | -0.310 |
Token | ! |
Token ID | 0 |
Feature activation | +3.25e-02 |
Pos logit contributions | |
eks | +0.106 |
agged | +0.102 |
oos | +0.101 |
roads | +0.101 |
ays | +0.101 |
Neg logit contributions | |
Ho | -0.130 |
Rib | -0.128 |
H | -0.115 |
Tw | -0.106 |
Bun | -0.105 |
Token |
|
Token ID | 198 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
ays | +0.058 |
ases | +0.055 |
he | +0.055 |
cases | +0.054 |
agged | +0.054 |
Neg logit contributions | |
Ho | -0.047 |
Rib | -0.046 |
H | -0.041 |
Bun | -0.039 |
Bub | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
Ho | +0.039 |
Rib | +0.038 |
H | +0.033 |
Bub | +0.031 |
accidentally | +0.030 |
Neg logit contributions | |
he | -0.047 |
eks | -0.045 |
ays | -0.044 |
ases | -0.044 |
elevator | -0.044 |
Token | They |
Token ID | 2990 |
Feature activation | +7.00e-02 |
Pos logit contributions | |
Ho | +0.044 |
Rib | +0.043 |
H | +0.039 |
Bub | +0.036 |
Bun | +0.035 |
Neg logit contributions | |
eks | -0.042 |
he | -0.042 |
oos | -0.041 |
ays | -0.041 |
ases | -0.041 |
Token | H |
Token ID | 39 |
Feature activation | -2.33e-02 |
Pos logit contributions | |
agged | +0.060 |
ases | +0.059 |
oos | +0.059 |
hed | +0.058 |
he | +0.058 |
Neg logit contributions | |
Ho | -0.086 |
Rib | -0.079 |
H | -0.075 |
Bun | -0.070 |
accidentally | -0.069 |
Token | oor |
Token ID | 2675 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
Rib | +0.038 |
Ho | +0.037 |
H | +0.035 |
Bub | +0.034 |
Fro | +0.032 |
Neg logit contributions | |
eks | -0.038 |
cases | -0.036 |
elevator | -0.036 |
ps | -0.035 |
bags | -0.035 |
Token | ay |
Token ID | 323 |
Feature activation | +7.08e-02 |
Pos logit contributions | |
eks | +0.181 |
cks | +0.172 |
ps | +0.167 |
ases | +0.159 |
oos | +0.155 |
Neg logit contributions | |
Ho | -0.377 |
Rib | -0.364 |
H | -0.362 |
Bun | -0.338 |
Lucky | -0.331 |
Token | , |
Token ID | 11 |
Feature activation | -5.37e-03 |
Pos logit contributions | |
ays | +0.120 |
agged | +0.119 |
oos | +0.118 |
cases | +0.118 |
eks | +0.118 |
Neg logit contributions | |
Ho | -0.114 |
Rib | -0.112 |
H | -0.103 |
Bun | -0.094 |
Bub | -0.091 |
Token | h |
Token ID | 289 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
Ho | +0.008 |
Rib | +0.008 |
H | +0.007 |
Bun | +0.006 |
accidentally | +0.006 |
Neg logit contributions | |
elevator | -0.010 |
eks | -0.010 |
oos | -0.009 |
cases | -0.009 |
ases | -0.009 |
Token | oor |
Token ID | 2675 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
Rib | +0.061 |
Ho | +0.059 |
H | +0.055 |
Bub | +0.053 |
Stri | +0.051 |
Neg logit contributions | |
eks | -0.052 |
led | -0.049 |
aks | -0.049 |
roads | -0.049 |
igators | -0.048 |
Token | ay |
Token ID | 323 |
Feature activation | +7.05e-02 |
Pos logit contributions | |
eks | +0.176 |
cks | +0.156 |
ases | +0.153 |
ps | +0.150 |
oos | +0.148 |
Neg logit contributions | |
Ho | -0.388 |
Rib | -0.377 |
H | -0.372 |
Bun | -0.352 |
Lucky | -0.340 |
Token | , |
Token ID | 11 |
Feature activation | -5.60e-03 |
Pos logit contributions | |
ays | +0.113 |
roads | +0.113 |
eks | +0.111 |
cases | +0.110 |
agged | +0.110 |
Neg logit contributions | |
Ho | -0.119 |
Rib | -0.118 |
H | -0.109 |
Bun | -0.100 |
Bub | -0.097 |
Token | h |
Token ID | 289 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
Ho | +0.009 |
Rib | +0.009 |
H | +0.008 |
Bun | +0.007 |
accidentally | +0.007 |
Neg logit contributions | |
elevator | -0.009 |
eks | -0.009 |
cases | -0.009 |
oos | -0.009 |
he | -0.009 |
Token | oor |
Token ID | 2675 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
Rib | +0.061 |
Ho | +0.059 |
H | +0.055 |
Bub | +0.053 |
Stri | +0.051 |
Neg logit contributions | |
eks | -0.052 |
roads | -0.049 |
led | -0.048 |
aks | -0.048 |
igators | -0.048 |
Token | ay |
Token ID | 323 |
Feature activation | +7.05e-02 |
Pos logit contributions | |
eks | +0.174 |
ases | +0.152 |
cks | +0.152 |
oos | +0.147 |
ps | +0.145 |
Neg logit contributions | |
Ho | -0.390 |
Rib | -0.380 |
H | -0.375 |
Bun | -0.357 |
Lucky | -0.342 |
Token | everyone |
Token ID | 2506 |
Feature activation | +6.32e-02 |
Pos logit contributions | |
eks | +0.025 |
cases | +0.025 |
elevator | +0.024 |
oos | +0.024 |
led | +0.024 |
Neg logit contributions | |
Rib | -0.025 |
Ho | -0.025 |
H | -0.022 |
Bub | -0.020 |
Po | -0.019 |
Token | cheered |
Token ID | 34767 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
eks | +0.103 |
oos | +0.103 |
ays | +0.100 |
elevator | +0.100 |
cases | +0.100 |
Neg logit contributions | |
Ho | -0.105 |
Rib | -0.103 |
H | -0.092 |
Bub | -0.084 |
Bun | -0.083 |
Token | . |
Token ID | 13 |
Feature activation | -3.90e-03 |
Pos logit contributions | |
eks | +0.062 |
oos | +0.061 |
ays | +0.060 |
elevator | +0.059 |
cases | +0.059 |
Neg logit contributions | |
Ho | -0.080 |
Rib | -0.078 |
H | -0.071 |
Bub | -0.066 |
Bun | -0.066 |
Token | " |
Token ID | 366 |
Feature activation | +5.00e-03 |
Pos logit contributions | |
Rib | +0.005 |
Ho | +0.005 |
H | +0.004 |
accidentally | +0.004 |
Fro | +0.004 |
Neg logit contributions | |
cases | -0.007 |
agged | -0.007 |
hed | -0.007 |
ays | -0.007 |
oos | -0.007 |
Token | H |
Token ID | 39 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
hed | +0.007 |
he | +0.007 |
eks | +0.007 |
agged | +0.007 |
oos | +0.006 |
Neg logit contributions | |
Ho | -0.010 |
Rib | -0.009 |
H | -0.008 |
onna | -0.008 |
Fro | -0.008 |
Token | oor |
Token ID | 2675 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
Ho | +0.045 |
Rib | +0.044 |
H | +0.043 |
Bun | +0.041 |
Bub | +0.040 |
Neg logit contributions | |
elevator | -0.030 |
ays | -0.030 |
eks | -0.030 |
oos | -0.029 |
cases | -0.029 |
Token | ay |
Token ID | 323 |
Feature activation | +6.77e-02 |
Pos logit contributions | |
eks | +0.163 |
cks | +0.148 |
ases | +0.144 |
elevator | +0.144 |
ps | +0.144 |
Neg logit contributions | |
Ho | -0.402 |
Rib | -0.388 |
H | -0.382 |
Bun | -0.367 |
Bub | -0.351 |
Token | for |
Token ID | 329 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
cases | +0.106 |
ays | +0.104 |
oos | +0.103 |
elevator | +0.102 |
roads | +0.100 |
Neg logit contributions | |
Ho | -0.121 |
Rib | -0.117 |
H | -0.109 |
Bun | -0.103 |
Bub | -0.098 |
Token | Blue |
Token ID | 4518 |
Feature activation | -7.49e-02 |
Pos logit contributions | |
eks | +0.045 |
oos | +0.044 |
ases | +0.043 |
elevator | +0.043 |
ays | +0.042 |
Neg logit contributions | |
Ho | -0.050 |
Rib | -0.050 |
H | -0.044 |
Bub | -0.041 |
Bun | -0.040 |
Token | and |
Token ID | 290 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
Ho | +0.145 |
Rib | +0.142 |
H | +0.129 |
Bun | +0.121 |
Bub | +0.120 |
Neg logit contributions | |
eks | -0.103 |
oos | -0.103 |
ays | -0.100 |
elevator | -0.100 |
cases | -0.099 |
Token | his |
Token ID | 465 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
eks | +0.023 |
elevator | +0.023 |
oos | +0.022 |
ases | +0.022 |
led | +0.022 |
Neg logit contributions | |
Rib | -0.025 |
Ho | -0.025 |
H | -0.022 |
Stri | -0.020 |
Tw | -0.020 |
Token | |
Token ID | 220 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
Ho | +0.006 |
Rib | +0.006 |
H | +0.005 |
Bun | +0.005 |
Tw | +0.005 |
Neg logit contributions | |
cases | -0.005 |
ays | -0.005 |
elevator | -0.005 |
oos | -0.005 |
agged | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
ays | +0.032 |
cases | +0.031 |
he | +0.031 |
oos | +0.031 |
hed | +0.031 |
Neg logit contributions | |
Rib | -0.024 |
Ho | -0.023 |
H | -0.019 |
Bub | -0.018 |
Po | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
Ho | +0.024 |
Rib | +0.024 |
H | +0.020 |
Po | +0.019 |
Bub | +0.018 |
Neg logit contributions | |
he | -0.030 |
cases | -0.028 |
hed | -0.027 |
irens | -0.026 |
agged | -0.026 |
Token | " |
Token ID | 1 |
Feature activation | +5.02e-02 |
Pos logit contributions | |
Ho | +0.028 |
Rib | +0.027 |
H | +0.024 |
Bub | +0.022 |
Bun | +0.022 |
Neg logit contributions | |
oos | -0.027 |
eks | -0.026 |
cases | -0.026 |
elevator | -0.026 |
ays | -0.026 |
Token | H |
Token ID | 39 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
eks | +0.076 |
he | +0.075 |
oos | +0.075 |
ays | +0.073 |
ases | +0.073 |
Neg logit contributions | |
Ho | -0.090 |
Rib | -0.086 |
H | -0.078 |
Bub | -0.073 |
Bun | -0.073 |
Token | oor |
Token ID | 2675 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
Rib | +0.048 |
Ho | +0.047 |
H | +0.046 |
onna | +0.042 |
accidentally | +0.041 |
Neg logit contributions | |
eks | -0.027 |
cases | -0.025 |
elevator | -0.024 |
led | -0.024 |
roads | -0.023 |
Token | ay |
Token ID | 323 |
Feature activation | +6.81e-02 |
Pos logit contributions | |
eks | +0.171 |
cks | +0.152 |
ases | +0.151 |
ps | +0.146 |
oos | +0.146 |
Neg logit contributions | |
Ho | -0.381 |
H | -0.368 |
Rib | -0.368 |
Bun | -0.346 |
Lucky | -0.333 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
ays | +0.108 |
roads | +0.107 |
oos | +0.107 |
ags | +0.107 |
agged | +0.107 |
Neg logit contributions | |
Ho | -0.112 |
Rib | -0.110 |
H | -0.101 |
Bun | -0.094 |
Bub | -0.091 |
Token | said |
Token ID | 531 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
cases | +0.042 |
elevator | +0.041 |
he | +0.041 |
eks | +0.040 |
oos | +0.040 |
Neg logit contributions | |
Ho | -0.041 |
Rib | -0.041 |
H | -0.035 |
Bun | -0.033 |
Bub | -0.032 |
Token | Tommy |
Token ID | 19919 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
Ho | +0.027 |
Rib | +0.026 |
H | +0.023 |
Bun | +0.022 |
Butter | +0.021 |
Neg logit contributions | |
elevator | -0.021 |
oos | -0.021 |
he | -0.021 |
cases | -0.021 |
ays | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | -4.75e-03 |
Pos logit contributions | |
cases | +0.033 |
hed | +0.030 |
he | +0.029 |
led | +0.028 |
ow | +0.027 |
Neg logit contributions | |
Ho | -0.068 |
Rib | -0.067 |
H | -0.058 |
Bub | -0.058 |
Butter | -0.058 |
Token | time |
Token ID | 640 |
Feature activation | -8.24e-03 |
Pos logit contributions | |
elevator | +0.015 |
eks | +0.015 |
ags | +0.015 |
agged | +0.014 |
oos | +0.014 |
Neg logit contributions | |
Rib | -0.028 |
Ho | -0.026 |
H | -0.023 |
Bub | -0.022 |
Stri | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
Ho | +0.012 |
Rib | +0.011 |
H | +0.010 |
Bub | +0.009 |
Bun | +0.009 |
Neg logit contributions | |
oos | -0.016 |
eks | -0.016 |
ays | -0.015 |
ases | -0.015 |
agged | -0.015 |
Token | go |
Token ID | 467 |
Feature activation | +2.58e-02 |
Pos logit contributions | |
Ho | +0.031 |
Rib | +0.030 |
H | +0.028 |
Bub | +0.026 |
Bun | +0.026 |
Neg logit contributions | |
oos | -0.019 |
eks | -0.019 |
ays | -0.019 |
cases | -0.018 |
agged | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -4.35e-03 |
Pos logit contributions | |
oos | +0.043 |
agged | +0.042 |
eks | +0.042 |
ays | +0.042 |
ases | +0.042 |
Neg logit contributions | |
Ho | -0.043 |
Rib | -0.040 |
H | -0.039 |
Bun | -0.035 |
Bub | -0.035 |
Token | H |
Token ID | 367 |
Feature activation | -5.32e-02 |
Pos logit contributions | |
Ho | +0.006 |
Rib | +0.006 |
H | +0.005 |
Bun | +0.005 |
Tooth | +0.005 |
Neg logit contributions | |
ays | -0.007 |
oos | -0.007 |
ases | -0.007 |
eks | -0.007 |
hed | -0.007 |
Token | oor |
Token ID | 2675 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
Ho | +0.085 |
Rib | +0.082 |
H | +0.073 |
Bub | +0.067 |
Bun | +0.067 |
Neg logit contributions | |
oos | -0.092 |
eks | -0.091 |
ays | -0.090 |
ases | -0.089 |
cases | -0.088 |
Token | ay |
Token ID | 323 |
Feature activation | +6.70e-02 |
Pos logit contributions | |
eks | +0.158 |
ases | +0.145 |
cks | +0.144 |
oos | +0.139 |
elevator | +0.130 |
Neg logit contributions | |
Ho | -0.390 |
Rib | -0.377 |
H | -0.376 |
Bun | -0.352 |
Lucky | -0.341 |
Token | ! |
Token ID | 0 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
oos | +0.094 |
eks | +0.094 |
cases | +0.092 |
elevator | +0.090 |
ays | +0.090 |
Neg logit contributions | |
Ho | -0.126 |
Rib | -0.124 |
H | -0.112 |
Bub | -0.104 |
Butter | -0.103 |
Token | |
Token ID | 220 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
ays | +0.055 |
oos | +0.052 |
ases | +0.051 |
he | +0.051 |
agged | +0.049 |
Neg logit contributions | |
Ho | -0.056 |
Rib | -0.055 |
accidentally | -0.050 |
please | -0.049 |
Bun | -0.047 |
Token |
|
Token ID | 198 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
ays | +0.026 |
he | +0.026 |
oos | +0.025 |
cases | +0.024 |
ases | +0.024 |
Neg logit contributions | |
Rib | -0.026 |
Ho | -0.025 |
fle | -0.021 |
H | -0.020 |
accidentally | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
Ho | +0.027 |
Rib | +0.027 |
H | +0.023 |
accidentally | +0.022 |
Butter | +0.022 |
Neg logit contributions | |
he | -0.034 |
eks | -0.031 |
oos | -0.030 |
ases | -0.030 |
hed | -0.030 |
Token | it |
Token ID | 340 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
eks | +0.013 |
cases | +0.013 |
elevator | +0.013 |
led | +0.013 |
oos | +0.013 |
Neg logit contributions | |
Ho | -0.013 |
Rib | -0.013 |
H | -0.012 |
Butter | -0.011 |
Tw | -0.010 |
Token | felt |
Token ID | 2936 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
eks | +0.021 |
oos | +0.021 |
cases | +0.020 |
elevator | +0.020 |
ays | +0.020 |
Neg logit contributions | |
Ho | -0.018 |
Rib | -0.017 |
H | -0.015 |
Bub | -0.014 |
Bun | -0.014 |
Token | wonderful |
Token ID | 7932 |
Feature activation | -3.61e-04 |
Pos logit contributions | |
eks | +0.020 |
oos | +0.020 |
ays | +0.019 |
elevator | +0.019 |
cases | +0.019 |
Neg logit contributions | |
Ho | -0.020 |
Rib | -0.020 |
H | -0.018 |
Bub | -0.016 |
Bun | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -9.41e-03 |
Pos logit contributions | |
Ho | +0.001 |
Rib | +0.001 |
H | +0.001 |
Bub | +0.001 |
Bun | +0.001 |
Neg logit contributions | |
eks | -0.001 |
oos | -0.001 |
ays | -0.001 |
elevator | -0.001 |
ases | -0.001 |
Token | H |
Token ID | 367 |
Feature activation | -5.56e-02 |
Pos logit contributions | |
Rib | +0.016 |
Ho | +0.015 |
accidentally | +0.014 |
H | +0.013 |
Ros | +0.013 |
Neg logit contributions | |
hed | -0.014 |
he | -0.014 |
ays | -0.014 |
oos | -0.014 |
agged | -0.014 |
Token | oor |
Token ID | 2675 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
Ho | +0.097 |
Rib | +0.095 |
H | +0.092 |
Bun | +0.086 |
Bub | +0.083 |
Neg logit contributions | |
elevator | -0.085 |
eks | -0.083 |
oos | -0.077 |
lobby | -0.077 |
ags | -0.077 |
Token | ay |
Token ID | 323 |
Feature activation | +6.53e-02 |
Pos logit contributions | |
eks | +0.123 |
oos | +0.117 |
ases | +0.114 |
elevator | +0.111 |
ays | +0.109 |
Neg logit contributions | |
Ho | -0.427 |
Rib | -0.418 |
H | -0.396 |
Bun | -0.376 |
Bub | -0.374 |
Token | ! |
Token ID | 0 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
eks | +0.075 |
oos | +0.075 |
cases | +0.072 |
ases | +0.071 |
elevator | +0.071 |
Neg logit contributions | |
Ho | -0.139 |
Rib | -0.137 |
H | -0.125 |
Bub | -0.118 |
Butter | -0.117 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
he | +0.052 |
hed | +0.049 |
ays | +0.047 |
agged | +0.046 |
are | +0.045 |
Neg logit contributions | |
Rib | -0.049 |
accidentally | -0.047 |
Ho | -0.047 |
please | -0.043 |
onna | -0.042 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
he | +0.160 |
hed | +0.148 |
ays | +0.140 |
oos | +0.138 |
agged | +0.130 |
Neg logit contributions | |
Rib | -0.227 |
Ho | -0.224 |
H | -0.197 |
accidentally | -0.194 |
Butter | -0.192 |
Token | upon |
Token ID | 2402 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
eks | +0.021 |
oos | +0.021 |
ays | +0.021 |
elevator | +0.021 |
cases | +0.021 |
Neg logit contributions | |
Ho | -0.017 |
Rib | -0.017 |
H | -0.015 |
Bub | -0.013 |
Bun | -0.013 |
Token | too |
Token ID | 1165 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
Ho | +0.064 |
Rib | +0.063 |
H | +0.056 |
Bub | +0.053 |
Bun | +0.052 |
Neg logit contributions | |
oos | -0.052 |
eks | -0.052 |
elevator | -0.051 |
ays | -0.051 |
cases | -0.051 |
Token | spicy |
Token ID | 26880 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
Ho | +0.056 |
Rib | +0.054 |
H | +0.049 |
accidentally | +0.045 |
Bub | +0.044 |
Neg logit contributions | |
eks | -0.053 |
oos | -0.053 |
ays | -0.051 |
agged | -0.051 |
ases | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | -8.17e-03 |
Pos logit contributions | |
Ho | +0.034 |
Rib | +0.032 |
H | +0.029 |
Bub | +0.027 |
Bun | +0.027 |
Neg logit contributions | |
eks | -0.034 |
agged | -0.032 |
oos | -0.032 |
ays | -0.032 |
elevator | -0.031 |
Token | " |
Token ID | 366 |
Feature activation | +2.27e-03 |
Pos logit contributions | |
Ho | +0.014 |
Rib | +0.013 |
H | +0.012 |
Bub | +0.011 |
Bun | +0.011 |
Neg logit contributions | |
cases | -0.013 |
oos | -0.013 |
ays | -0.013 |
elevator | -0.013 |
agged | -0.012 |
Token | H |
Token ID | 39 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
he | +0.003 |
oos | +0.003 |
eks | +0.003 |
ays | +0.003 |
ases | +0.003 |
Neg logit contributions | |
Ho | -0.004 |
Rib | -0.004 |
H | -0.004 |
Bub | -0.003 |
Bun | -0.003 |
Token | oor |
Token ID | 2675 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
Rib | +0.041 |
Ho | +0.041 |
H | +0.037 |
onna | +0.034 |
accidentally | +0.033 |
Neg logit contributions | |
eks | -0.043 |
cases | -0.042 |
elevator | -0.041 |
roads | -0.040 |
oos | -0.040 |
Token | ay |
Token ID | 323 |
Feature activation | +6.55e-02 |
Pos logit contributions | |
eks | +0.162 |
cks | +0.147 |
ases | +0.143 |
ps | +0.137 |
oos | +0.136 |
Neg logit contributions | |
Ho | -0.384 |
Rib | -0.367 |
H | -0.366 |
Bun | -0.347 |
Lucky | -0.337 |
Token | !" |
Token ID | 2474 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
cases | +0.097 |
ays | +0.097 |
eks | +0.097 |
oos | +0.097 |
agged | +0.096 |
Neg logit contributions | |
Ho | -0.116 |
Rib | -0.112 |
H | -0.105 |
Bun | -0.097 |
Bub | -0.093 |
Token | said |
Token ID | 531 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
eks | +0.064 |
oos | +0.064 |
ases | +0.062 |
ays | +0.062 |
elevator | +0.062 |
Neg logit contributions | |
Ho | -0.050 |
Rib | -0.048 |
H | -0.042 |
Bub | -0.038 |
Bun | -0.038 |
Token | Mom |
Token ID | 11254 |
Feature activation | +7.46e-02 |
Pos logit contributions | |
Ho | +0.032 |
Rib | +0.032 |
H | +0.028 |
Bub | +0.025 |
Bun | +0.025 |
Neg logit contributions | |
elevator | -0.032 |
oos | -0.032 |
eks | -0.031 |
he | -0.031 |
cases | -0.031 |
Token | my |
Token ID | 1820 |
Feature activation | -3.37e-02 |
Pos logit contributions | |
elevator | +0.095 |
cases | +0.095 |
he | +0.094 |
ays | +0.093 |
oos | +0.093 |
Neg logit contributions | |
Ho | -0.149 |
Rib | -0.147 |
H | -0.124 |
Butter | -0.124 |
Bub | -0.123 |
Token | . |
Token ID | 13 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
eks | +0.027 |
agged | +0.026 |
oos | +0.026 |
ases | +0.025 |
ays | +0.025 |
Neg logit contributions | |
Ho | -0.023 |
Rib | -0.023 |
H | -0.021 |
Bun | -0.019 |
Tw | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
Ho | +0.039 |
Rib | +0.039 |
Bun | +0.035 |
H | +0.035 |
accidentally | +0.033 |
Neg logit contributions | |
ays | -0.033 |
he | -0.031 |
irens | -0.030 |
elevator | -0.030 |
oos | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
Rib | +0.054 |
Ho | +0.053 |
H | +0.044 |
Butter | +0.042 |
accidentally | +0.041 |
Neg logit contributions | |
he | -0.062 |
ases | -0.059 |
eks | -0.058 |
ays | -0.058 |
elevator | -0.057 |
Token | " |
Token ID | 1 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
Ho | +0.060 |
Rib | +0.059 |
H | +0.053 |
Bub | +0.049 |
Bun | +0.049 |
Neg logit contributions | |
eks | -0.054 |
oos | -0.054 |
ases | -0.053 |
elevator | -0.053 |
cases | -0.053 |
Token | H |
Token ID | 39 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
eks | +0.063 |
oos | +0.063 |
ays | +0.061 |
elevator | +0.061 |
cases | +0.061 |
Neg logit contributions | |
Ho | -0.065 |
Rib | -0.063 |
H | -0.057 |
Bub | -0.052 |
Bun | -0.052 |
Token | oor |
Token ID | 2675 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
Ho | +0.052 |
Rib | +0.052 |
H | +0.048 |
onna | +0.043 |
Fro | +0.043 |
Neg logit contributions | |
eks | -0.050 |
elevator | -0.050 |
cases | -0.050 |
roads | -0.048 |
ags | -0.048 |
Token | ay |
Token ID | 323 |
Feature activation | +6.59e-02 |
Pos logit contributions | |
eks | +0.177 |
led | +0.170 |
oos | +0.167 |
cks | +0.165 |
ps | +0.160 |
Neg logit contributions | |
Ho | -0.356 |
Rib | -0.336 |
H | -0.335 |
Lucky | -0.309 |
Bun | -0.307 |
Token | ! |
Token ID | 0 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
roads | +0.116 |
agged | +0.115 |
cases | +0.113 |
ags | +0.112 |
ays | +0.111 |
Neg logit contributions | |
Ho | -0.103 |
Rib | -0.100 |
H | -0.090 |
Bun | -0.079 |
Stri | -0.079 |
Token | We |
Token ID | 775 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
ays | +0.040 |
ases | +0.037 |
oos | +0.037 |
elevator | +0.037 |
cases | +0.035 |
Neg logit contributions | |
Ho | -0.044 |
Rib | -0.042 |
Bun | -0.038 |
H | -0.038 |
accidentally | -0.037 |
Token | did |
Token ID | 750 |
Feature activation | -3.99e-02 |
Pos logit contributions | |
oos | +0.032 |
ays | +0.032 |
agged | +0.031 |
cases | +0.030 |
ags | +0.030 |
Neg logit contributions | |
Ho | -0.029 |
Rib | -0.028 |
H | -0.026 |
accidentally | -0.025 |
Bub | -0.025 |
Token | it |
Token ID | 340 |
Feature activation | +4.90e-03 |
Pos logit contributions | |
Ho | +0.063 |
Rib | +0.063 |
H | +0.053 |
Bub | +0.051 |
Bun | +0.049 |
Neg logit contributions | |
eks | -0.067 |
elevator | -0.066 |
oos | -0.064 |
tickets | -0.063 |
cases | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
Ho | +0.108 |
Rib | +0.106 |
H | +0.093 |
Bub | +0.085 |
Lucky | +0.085 |
Neg logit contributions | |
eks | -0.114 |
oos | -0.111 |
agged | -0.111 |
ases | -0.106 |
elevator | -0.106 |
Token |
|
Token ID | 198 |
Feature activation | -2.96e-02 |
Pos logit contributions | |
Ho | +0.027 |
Rib | +0.027 |
H | +0.024 |
Bun | +0.024 |
accidentally | +0.023 |
Neg logit contributions | |
ays | -0.017 |
oos | -0.017 |
ases | -0.016 |
cases | -0.016 |
tickets | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
Rib | +0.044 |
Ho | +0.044 |
H | +0.036 |
Butter | +0.034 |
Bub | +0.033 |
Neg logit contributions | |
he | -0.053 |
ases | -0.050 |
oos | -0.050 |
ays | -0.050 |
eks | -0.049 |
Token | " |
Token ID | 1 |
Feature activation | +4.09e-02 |
Pos logit contributions | |
Ho | +0.045 |
Rib | +0.044 |
H | +0.039 |
Bub | +0.035 |
Bun | +0.035 |
Neg logit contributions | |
eks | -0.052 |
oos | -0.052 |
ays | -0.050 |
elevator | -0.050 |
agged | -0.050 |
Token | H |
Token ID | 39 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
eks | +0.068 |
oos | +0.068 |
elevator | +0.066 |
ases | +0.066 |
ays | +0.066 |
Neg logit contributions | |
Ho | -0.067 |
Rib | -0.066 |
H | -0.059 |
Bub | -0.054 |
Bun | -0.054 |
Token | oor |
Token ID | 2675 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
Ho | +0.054 |
Rib | +0.052 |
H | +0.049 |
Bub | +0.045 |
Stri | +0.044 |
Neg logit contributions | |
eks | -0.036 |
cases | -0.036 |
elevator | -0.036 |
led | -0.036 |
ays | -0.035 |
Token | ay |
Token ID | 323 |
Feature activation | +6.67e-02 |
Pos logit contributions | |
eks | +0.166 |
oos | +0.155 |
cks | +0.154 |
ps | +0.154 |
led | +0.153 |
Neg logit contributions | |
Ho | -0.362 |
Rib | -0.341 |
H | -0.341 |
Bun | -0.314 |
Lucky | -0.314 |
Token | , |
Token ID | 11 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
roads | +0.110 |
agged | +0.110 |
cases | +0.109 |
ays | +0.107 |
ags | +0.106 |
Neg logit contributions | |
Ho | -0.112 |
Rib | -0.108 |
H | -0.099 |
Stri | -0.089 |
Bun | -0.089 |
Token | we |
Token ID | 356 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
Ho | +0.016 |
Rib | +0.016 |
H | +0.014 |
Bun | +0.013 |
accidentally | +0.013 |
Neg logit contributions | |
elevator | -0.016 |
eks | -0.016 |
cases | -0.016 |
oos | -0.016 |
ases | -0.015 |
Token | found |
Token ID | 1043 |
Feature activation | +6.79e-03 |
Pos logit contributions | |
ays | +0.024 |
oos | +0.023 |
agged | +0.023 |
irens | +0.022 |
cases | +0.022 |
Neg logit contributions | |
Ho | -0.027 |
Rib | -0.025 |
H | -0.024 |
accidentally | -0.023 |
Bub | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
eks | +0.010 |
oos | +0.010 |
ays | +0.010 |
elevator | +0.010 |
ases | +0.010 |
Neg logit contributions | |
Ho | -0.012 |
Rib | -0.012 |
H | -0.011 |
Bub | -0.010 |
Bun | -0.010 |
Token | ! |
Token ID | 0 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
Ho | +0.167 |
Rib | +0.165 |
H | +0.144 |
Bub | +0.137 |
Butter | +0.136 |
Neg logit contributions | |
oos | -0.110 |
he | -0.108 |
eks | -0.106 |
ases | -0.106 |
cases | -0.105 |
Token |
|
Token ID | 198 |
Feature activation | -2.70e-02 |
Pos logit contributions | |
he | +0.043 |
oos | +0.042 |
agged | +0.042 |
hed | +0.041 |
ays | +0.041 |
Neg logit contributions | |
Rib | -0.048 |
Ho | -0.048 |
H | -0.041 |
please | -0.038 |
Stri | -0.038 |
Token |
|
Token ID | 198 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
Ho | +0.041 |
Rib | +0.041 |
H | +0.035 |
Butter | +0.032 |
fle | +0.031 |
Neg logit contributions | |
he | -0.050 |
oos | -0.047 |
eks | -0.046 |
agged | -0.045 |
hed | -0.045 |
Token | " |
Token ID | 1 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
Ho | +0.044 |
Rib | +0.043 |
H | +0.038 |
Bub | +0.034 |
Butter | +0.034 |
Neg logit contributions | |
oos | -0.047 |
eks | -0.047 |
he | -0.045 |
ases | -0.045 |
cases | -0.045 |
Token | H |
Token ID | 39 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
he | +0.055 |
hed | +0.054 |
hing | +0.050 |
ited | +0.050 |
oos | +0.049 |
Neg logit contributions | |
Ho | -0.078 |
Rib | -0.072 |
H | -0.069 |
Bub | -0.065 |
Fro | -0.064 |
Token | oor |
Token ID | 2675 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
Ho | +0.055 |
Rib | +0.055 |
H | +0.051 |
Bub | +0.048 |
Bun | +0.047 |
Neg logit contributions | |
eks | -0.039 |
cases | -0.038 |
elevator | -0.037 |
ays | -0.036 |
oos | -0.036 |
Token | ay |
Token ID | 323 |
Feature activation | +6.59e-02 |
Pos logit contributions | |
eks | +0.142 |
ases | +0.124 |
oos | +0.122 |
cks | +0.121 |
elevator | +0.120 |
Neg logit contributions | |
Ho | -0.396 |
Rib | -0.385 |
H | -0.375 |
Bun | -0.357 |
Lucky | -0.347 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
ays | +0.096 |
cases | +0.095 |
oos | +0.095 |
eks | +0.094 |
agged | +0.093 |
Neg logit contributions | |
Ho | -0.121 |
Rib | -0.119 |
H | -0.109 |
Bun | -0.101 |
Bub | -0.100 |
Token | shouted |
Token ID | 17293 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
led | +0.028 |
eks | +0.028 |
oos | +0.027 |
cases | +0.027 |
he | +0.027 |
Neg logit contributions | |
Rib | -0.029 |
Ho | -0.029 |
H | -0.025 |
Butter | -0.023 |
Stri | -0.022 |
Token | Ted |
Token ID | 11396 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
oos | +0.049 |
led | +0.048 |
elevator | +0.046 |
hed | +0.046 |
ued | +0.045 |
Neg logit contributions | |
Rib | -0.044 |
Ho | -0.044 |
H | -0.038 |
Stri | -0.035 |
Butter | -0.034 |
Token | as |
Token ID | 355 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
Ho | +0.021 |
Rib | +0.020 |
H | +0.018 |
Butter | +0.018 |
Bub | +0.017 |
Neg logit contributions | |
cases | -0.011 |
hed | -0.011 |
led | -0.011 |
he | -0.011 |
oos | -0.011 |
Token | Alice |
Token ID | 44484 |
Feature activation | +7.25e-03 |
Pos logit contributions | |
Ho | +0.038 |
Rib | +0.038 |
H | +0.034 |
Bun | +0.032 |
Tw | +0.031 |
Neg logit contributions | |
oos | -0.038 |
eks | -0.037 |
ays | -0.037 |
ases | -0.037 |
elevator | -0.036 |
Token | said |
Token ID | 531 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
oos | +0.011 |
eks | +0.011 |
cases | +0.011 |
led | +0.010 |
hed | +0.010 |
Neg logit contributions | |
Ho | -0.013 |
Rib | -0.013 |
H | -0.011 |
Bub | -0.011 |
Butter | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -9.81e-03 |
Pos logit contributions | |
Ho | +0.042 |
Rib | +0.041 |
H | +0.037 |
Bub | +0.035 |
Bun | +0.035 |
Neg logit contributions | |
eks | -0.031 |
oos | -0.031 |
ays | -0.030 |
ases | -0.030 |
elevator | -0.030 |
Token | " |
Token ID | 366 |
Feature activation | -1.41e-03 |
Pos logit contributions | |
Ho | +0.015 |
Rib | +0.014 |
H | +0.012 |
Bub | +0.011 |
Fro | +0.011 |
Neg logit contributions | |
cases | -0.017 |
elevator | -0.016 |
he | -0.016 |
hed | -0.016 |
oos | -0.016 |
Token | H |
Token ID | 39 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
Ho | +0.003 |
Rib | +0.003 |
H | +0.002 |
Bub | +0.002 |
Fro | +0.002 |
Neg logit contributions | |
he | -0.002 |
hed | -0.002 |
oos | -0.002 |
agged | -0.002 |
ases | -0.002 |
Token | oor |
Token ID | 2675 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
Rib | +0.055 |
Ho | +0.054 |
H | +0.053 |
Bub | +0.048 |
Bun | +0.047 |
Neg logit contributions | |
eks | -0.033 |
led | -0.031 |
elevator | -0.030 |
ays | -0.030 |
oos | -0.030 |
Token | ay |
Token ID | 323 |
Feature activation | +6.48e-02 |
Pos logit contributions | |
eks | +0.139 |
ases | +0.121 |
cks | +0.121 |
oos | +0.117 |
elevator | +0.114 |
Neg logit contributions | |
Ho | -0.393 |
Rib | -0.384 |
H | -0.377 |
Bun | -0.355 |
Lucky | -0.345 |
Token | ! |
Token ID | 0 |
Feature activation | +2.96e-02 |
Pos logit contributions | |
eks | +0.097 |
oos | +0.096 |
ays | +0.096 |
agged | +0.095 |
cases | +0.094 |
Neg logit contributions | |
Ho | -0.115 |
Rib | -0.113 |
H | -0.105 |
Bun | -0.096 |
Bub | -0.094 |
Token | Let |
Token ID | 3914 |
Feature activation | -3.41e-02 |
Pos logit contributions | |
ays | +0.037 |
cases | +0.036 |
elevator | +0.035 |
ases | +0.035 |
he | +0.034 |
Neg logit contributions | |
Ho | -0.054 |
Rib | -0.051 |
accidentally | -0.048 |
Fro | -0.045 |
H | -0.044 |
Token | 's |
Token ID | 338 |
Feature activation | -5.02e-02 |
Pos logit contributions | |
Ho | +0.055 |
Rib | +0.054 |
H | +0.048 |
Bun | +0.045 |
Bub | +0.044 |
Neg logit contributions | |
eks | -0.057 |
oos | -0.057 |
ases | -0.056 |
ays | -0.056 |
agged | -0.056 |
Token | throw |
Token ID | 3714 |
Feature activation | +2.29e-03 |
Pos logit contributions | |
Ho | +0.104 |
Rib | +0.102 |
H | +0.093 |
Bub | +0.087 |
Bun | +0.086 |
Neg logit contributions | |
eks | -0.062 |
oos | -0.061 |
elevator | -0.060 |
ases | -0.059 |
cases | -0.058 |
Token | We |
Token ID | 1135 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
Ho | +0.009 |
Rib | +0.008 |
H | +0.007 |
Fro | +0.007 |
Bub | +0.007 |
Neg logit contributions | |
hed | -0.005 |
he | -0.005 |
ases | -0.005 |
oos | -0.005 |
agged | -0.005 |
Token | did |
Token ID | 750 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
cases | +0.045 |
oos | +0.045 |
ays | +0.044 |
agged | +0.044 |
elevator | +0.042 |
Neg logit contributions | |
Ho | -0.046 |
Rib | -0.043 |
H | -0.041 |
accidentally | -0.039 |
Bub | -0.038 |
Token | it |
Token ID | 340 |
Feature activation | +2.44e-03 |
Pos logit contributions | |
Ho | +0.083 |
Rib | +0.082 |
H | +0.071 |
Butter | +0.067 |
Bub | +0.066 |
Neg logit contributions | |
elevator | -0.061 |
eks | -0.061 |
oos | -0.060 |
cases | -0.058 |
ases | -0.058 |
Token | ! |
Token ID | 0 |
Feature activation | +2.64e-02 |
Pos logit contributions | |
eks | +0.003 |
oos | +0.003 |
ays | +0.003 |
elevator | +0.003 |
cases | +0.003 |
Neg logit contributions | |
Ho | -0.005 |
Rib | -0.005 |
H | -0.004 |
Bub | -0.004 |
Bun | -0.004 |
Token | H |
Token ID | 367 |
Feature activation | -6.31e-02 |
Pos logit contributions | |
ays | +0.032 |
ases | +0.031 |
hed | +0.028 |
oos | +0.027 |
ay | +0.027 |
Neg logit contributions | |
Ho | -0.051 |
Rib | -0.048 |
accidentally | -0.045 |
onna | -0.044 |
Fro | -0.044 |
Token | oor |
Token ID | 2675 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
Ho | +0.131 |
Rib | +0.130 |
H | +0.119 |
Bun | +0.112 |
Bub | +0.111 |
Neg logit contributions | |
eks | -0.078 |
elevator | -0.076 |
oos | -0.074 |
cases | -0.073 |
agged | -0.072 |
Token | ay |
Token ID | 323 |
Feature activation | +6.27e-02 |
Pos logit contributions | |
eks | +0.118 |
oos | +0.104 |
ases | +0.103 |
elevator | +0.101 |
cases | +0.095 |
Neg logit contributions | |
Ho | -0.411 |
Rib | -0.404 |
H | -0.389 |
Bun | -0.370 |
Bub | -0.361 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
eks | +0.089 |
oos | +0.088 |
ays | +0.087 |
elevator | +0.087 |
cases | +0.086 |
Neg logit contributions | |
Ho | -0.116 |
Rib | -0.115 |
H | -0.106 |
Bun | -0.098 |
Bub | -0.097 |
Token | he |
Token ID | 339 |
Feature activation | -9.08e-03 |
Pos logit contributions | |
hed | +0.038 |
he | +0.035 |
cases | +0.035 |
ays | +0.034 |
led | +0.034 |
Neg logit contributions | |
Rib | -0.035 |
onna | -0.035 |
please | -0.032 |
Ho | -0.032 |
Acc | -0.030 |
Token | said |
Token ID | 531 |
Feature activation | -2.76e-02 |
Pos logit contributions | |
Ho | +0.018 |
Rib | +0.017 |
H | +0.016 |
Butter | +0.016 |
Tooth | +0.015 |
Neg logit contributions | |
led | -0.011 |
cases | -0.011 |
hed | -0.010 |
eks | -0.010 |
tickets | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
Ho | +0.056 |
Rib | +0.054 |
H | +0.049 |
Butter | +0.046 |
Stri | +0.046 |
Neg logit contributions | |
cases | -0.034 |
oos | -0.034 |
he | -0.033 |
led | -0.033 |
elevator | -0.033 |
Token | Elsa |
Token ID | 49050 |
Feature activation | -9.25e-02 |
Pos logit contributions | |
Ho | +0.004 |
Rib | +0.004 |
H | +0.004 |
Bub | +0.003 |
Butter | +0.003 |
Neg logit contributions | |
he | -0.003 |
eks | -0.003 |
oos | -0.003 |
ases | -0.003 |
roads | -0.003 |
Token | smiled |
Token ID | 13541 |
Feature activation | +3.03e-03 |
Pos logit contributions | |
Ho | +0.151 |
Rib | +0.147 |
H | +0.132 |
Bub | +0.121 |
Bun | +0.120 |
Neg logit contributions | |
oos | -0.153 |
eks | -0.153 |
ays | -0.150 |
elevator | -0.149 |
cases | -0.149 |
Token | . |
Token ID | 13 |
Feature activation | -4.41e-03 |
Pos logit contributions | |
oos | +0.004 |
agged | +0.003 |
ays | +0.003 |
led | +0.003 |
cases | +0.003 |
Neg logit contributions | |
Ho | -0.006 |
Rib | -0.006 |
H | -0.006 |
Bub | -0.005 |
Stri | -0.005 |
Token | " |
Token ID | 366 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
Rib | +0.009 |
Ho | +0.009 |
accidentally | +0.009 |
please | +0.008 |
Bub | +0.008 |
Neg logit contributions | |
ays | -0.004 |
elevator | -0.003 |
he | -0.003 |
oos | -0.003 |
downtown | -0.003 |
Token | H |
Token ID | 39 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
he | +0.015 |
hed | +0.014 |
agged | +0.013 |
ited | +0.013 |
hing | +0.013 |
Neg logit contributions | |
Ho | -0.027 |
Rib | -0.024 |
Bub | -0.023 |
Fro | -0.023 |
accidentally | -0.023 |
Token | oor |
Token ID | 2675 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
Ho | +0.048 |
Rib | +0.047 |
H | +0.043 |
Bub | +0.039 |
Bun | +0.039 |
Neg logit contributions | |
eks | -0.043 |
oos | -0.042 |
cases | -0.042 |
elevator | -0.042 |
ays | -0.041 |
Token | ay |
Token ID | 323 |
Feature activation | +7.26e-02 |
Pos logit contributions | |
eks | +0.101 |
oos | +0.093 |
ases | +0.089 |
elevator | +0.088 |
cases | +0.085 |
Neg logit contributions | |
Ho | -0.424 |
Rib | -0.416 |
H | -0.397 |
Bun | -0.376 |
Bub | -0.370 |
Token | , |
Token ID | 11 |
Feature activation | -8.68e-03 |
Pos logit contributions | |
oos | +0.101 |
eks | +0.101 |
ays | +0.100 |
cases | +0.100 |
agged | +0.098 |
Neg logit contributions | |
Ho | -0.138 |
Rib | -0.135 |
H | -0.125 |
Bub | -0.115 |
Bun | -0.114 |
Token | we |
Token ID | 356 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
Rib | +0.014 |
Ho | +0.013 |
H | +0.012 |
Bub | +0.011 |
accidentally | +0.011 |
Neg logit contributions | |
elevator | -0.014 |
he | -0.013 |
are | -0.013 |
eks | -0.013 |
cases | -0.012 |
Token | took |
Token ID | 1718 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
oos | +0.014 |
ays | +0.014 |
cases | +0.014 |
eks | +0.014 |
agged | +0.013 |
Neg logit contributions | |
Ho | -0.020 |
Rib | -0.020 |
H | -0.018 |
Bub | -0.017 |
Bun | -0.017 |
Token | a |
Token ID | 257 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
oos | +0.016 |
eks | +0.016 |
ays | +0.016 |
cases | +0.016 |
elevator | +0.016 |
Neg logit contributions | |
Ho | -0.022 |
Rib | -0.021 |
H | -0.019 |
Bub | -0.018 |
Bun | -0.018 |
Token | flew |
Token ID | 13112 |
Feature activation | -2.66e-02 |
Pos logit contributions | |
led | +0.017 |
eks | +0.016 |
elevator | +0.015 |
cases | +0.015 |
oos | +0.015 |
Neg logit contributions | |
Ho | -0.020 |
Rib | -0.020 |
H | -0.018 |
Ros | -0.016 |
Tw | -0.016 |
Token | away |
Token ID | 1497 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
Ho | +0.066 |
Rib | +0.065 |
H | +0.061 |
accidentally | +0.057 |
Butter | +0.057 |
Neg logit contributions | |
oos | -0.021 |
eks | -0.021 |
agged | -0.020 |
led | -0.020 |
out | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -7.46e-03 |
Pos logit contributions | |
oos | +0.052 |
eks | +0.052 |
cases | +0.050 |
ays | +0.050 |
elevator | +0.050 |
Neg logit contributions | |
Ho | -0.063 |
Rib | -0.062 |
H | -0.056 |
Bub | -0.052 |
Bun | -0.051 |
Token | " |
Token ID | 366 |
Feature activation | +6.76e-03 |
Pos logit contributions | |
Ho | +0.014 |
Rib | +0.013 |
Ros | +0.012 |
needle | +0.012 |
accidentally | +0.012 |
Neg logit contributions | |
he | -0.008 |
hed | -0.008 |
cases | -0.007 |
oos | -0.007 |
ays | -0.007 |
Token | H |
Token ID | 39 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
he | +0.010 |
eks | +0.009 |
hed | +0.009 |
oos | +0.009 |
agged | +0.009 |
Neg logit contributions | |
Ho | -0.013 |
Rib | -0.012 |
H | -0.011 |
Bub | -0.011 |
onna | -0.010 |
Token | oor |
Token ID | 2675 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
Rib | +0.054 |
Ho | +0.053 |
H | +0.050 |
Bub | +0.046 |
Butter | +0.045 |
Neg logit contributions | |
cases | -0.036 |
eks | -0.035 |
ays | -0.035 |
elevator | -0.035 |
oos | -0.034 |
Token | ay |
Token ID | 323 |
Feature activation | +6.58e-02 |
Pos logit contributions | |
eks | +0.116 |
oos | +0.106 |
ases | +0.105 |
elevator | +0.102 |
cases | +0.098 |
Neg logit contributions | |
Ho | -0.409 |
Rib | -0.400 |
H | -0.384 |
Bun | -0.365 |
Bub | -0.358 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
ays | +0.093 |
oos | +0.092 |
cases | +0.091 |
eks | +0.090 |
elevator | +0.089 |
Neg logit contributions | |
Ho | -0.124 |
Rib | -0.122 |
H | -0.112 |
Bun | -0.104 |
Bub | -0.103 |
Token | Jane |
Token ID | 12091 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
hed | +0.039 |
he | +0.038 |
cases | +0.038 |
elevator | +0.038 |
led | +0.037 |
Neg logit contributions | |
Rib | -0.042 |
Ho | -0.041 |
H | -0.035 |
onna | -0.035 |
Ros | -0.035 |
Token | shouted |
Token ID | 17293 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
led | +0.030 |
hed | +0.030 |
he | +0.029 |
eks | +0.027 |
ered | +0.026 |
Neg logit contributions | |
Ho | -0.045 |
Rib | -0.043 |
H | -0.040 |
Butter | -0.038 |
Bub | -0.038 |
Token | with |
Token ID | 351 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
oos | +0.035 |
cases | +0.034 |
eks | +0.034 |
elevator | +0.033 |
he | +0.033 |
Neg logit contributions | |
Ho | -0.052 |
Rib | -0.051 |
H | -0.046 |
Bub | -0.043 |
Stri | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
ases | +0.017 |
oos | +0.017 |
elevator | +0.017 |
ays | +0.017 |
cases | +0.016 |
Neg logit contributions | |
Ho | -0.040 |
Rib | -0.040 |
H | -0.036 |
Butter | -0.035 |
Bub | -0.035 |
Token | They |
Token ID | 1119 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
Ho | +0.042 |
Rib | +0.041 |
H | +0.036 |
Bun | +0.035 |
Bub | +0.034 |
Neg logit contributions | |
oos | -0.029 |
agged | -0.029 |
ases | -0.029 |
ays | -0.029 |
elevator | -0.028 |
Token | say |
Token ID | 910 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
oos | +0.045 |
ays | +0.044 |
eks | +0.044 |
cases | +0.043 |
agged | +0.043 |
Neg logit contributions | |
Ho | -0.064 |
Rib | -0.062 |
H | -0.057 |
Bub | -0.053 |
Bun | -0.052 |
Token | " |
Token ID | 366 |
Feature activation | -9.90e-03 |
Pos logit contributions | |
Ho | +0.038 |
Rib | +0.037 |
H | +0.032 |
Bun | +0.031 |
Butter | +0.030 |
Neg logit contributions | |
elevator | -0.044 |
tickets | -0.044 |
agged | -0.043 |
ases | -0.043 |
oos | -0.042 |
Token | H |
Token ID | 39 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
Ho | +0.019 |
Rib | +0.017 |
H | +0.016 |
Bub | +0.015 |
Po | +0.015 |
Neg logit contributions | |
he | -0.015 |
ases | -0.015 |
oos | -0.015 |
hed | -0.014 |
eks | -0.014 |
Token | oor |
Token ID | 2675 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
Rib | +0.081 |
Ho | +0.079 |
H | +0.077 |
Bub | +0.071 |
onna | +0.070 |
Neg logit contributions | |
eks | -0.045 |
elevator | -0.043 |
cases | -0.043 |
led | -0.042 |
roads | -0.041 |
Token | ay |
Token ID | 323 |
Feature activation | +6.16e-02 |
Pos logit contributions | |
eks | +0.165 |
cks | +0.146 |
led | +0.141 |
ps | +0.140 |
oos | +0.138 |
Neg logit contributions | |
Ho | -0.346 |
Rib | -0.337 |
H | -0.332 |
Lucky | -0.308 |
Bun | -0.303 |
Token | ! |
Token ID | 0 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
eks | +0.103 |
roads | +0.101 |
cases | +0.100 |
agged | +0.100 |
oos | +0.099 |
Neg logit contributions | |
Ho | -0.100 |
Rib | -0.100 |
H | -0.092 |
Stri | -0.081 |
Bun | -0.081 |
Token | We |
Token ID | 775 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
ays | +0.028 |
ases | +0.028 |
oos | +0.025 |
elevator | +0.024 |
hed | +0.024 |
Neg logit contributions | |
Ho | -0.040 |
Rib | -0.037 |
Bun | -0.034 |
accidentally | -0.034 |
H | -0.034 |
Token | are |
Token ID | 389 |
Feature activation | +4.64e-02 |
Pos logit contributions | |
ays | +0.021 |
oos | +0.021 |
cases | +0.021 |
agged | +0.020 |
irens | +0.020 |
Neg logit contributions | |
Ho | -0.023 |
Rib | -0.022 |
H | -0.020 |
Bub | -0.019 |
accidentally | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
eks | +0.066 |
elevator | +0.065 |
ases | +0.065 |
ags | +0.063 |
agged | +0.062 |
Neg logit contributions | |
Ho | -0.086 |
Rib | -0.085 |
H | -0.075 |
Butter | -0.072 |
Bun | -0.071 |
Token | special |
Token ID | 2041 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
Ho | +0.023 |
Rib | +0.022 |
H | +0.019 |
Bub | +0.017 |
Bun | +0.017 |
Neg logit contributions | |
oos | -0.028 |
eks | -0.028 |
elevator | -0.028 |
ays | -0.028 |
ases | -0.027 |
Token | polish |
Token ID | 25245 |
Feature activation | -8.28e-02 |
Pos logit contributions | |
elevator | +0.038 |
oos | +0.038 |
eks | +0.036 |
ays | +0.036 |
ases | +0.036 |
Neg logit contributions | |
Ho | -0.053 |
Rib | -0.052 |
H | -0.047 |
Bub | -0.043 |
Bun | -0.043 |
Token | he |
Token ID | 339 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
Ho | +0.149 |
Rib | +0.146 |
H | +0.131 |
Bub | +0.122 |
Bun | +0.121 |
Neg logit contributions | |
oos | -0.123 |
eks | -0.123 |
ays | -0.120 |
cases | -0.119 |
elevator | -0.119 |
Token | couldn |
Token ID | 3521 |
Feature activation | -4.42e-02 |
Pos logit contributions | |
eks | +0.025 |
oos | +0.025 |
cases | +0.024 |
elevator | +0.024 |
led | +0.024 |
Neg logit contributions | |
Ho | -0.021 |
Rib | -0.021 |
H | -0.019 |
Bub | -0.017 |
Butter | -0.017 |
Token | 't |
Token ID | 470 |
Feature activation | -6.19e-03 |
Pos logit contributions | |
Ho | +0.072 |
Rib | +0.071 |
H | +0.064 |
Bub | +0.059 |
Bun | +0.059 |
Neg logit contributions | |
eks | -0.074 |
oos | -0.073 |
agged | -0.072 |
ays | -0.071 |
ases | -0.071 |
Token | contain |
Token ID | 3994 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
Ho | +0.011 |
Rib | +0.011 |
H | +0.010 |
Bub | +0.009 |
Bun | +0.009 |
Neg logit contributions | |
oos | -0.010 |
eks | -0.010 |
ays | -0.009 |
cases | -0.009 |
ases | -0.009 |
Token | his |
Token ID | 465 |
Feature activation | +4.72e-02 |
Pos logit contributions | |
eks | +0.279 |
oos | +0.277 |
ays | +0.276 |
cases | +0.275 |
agged | +0.272 |
Neg logit contributions | |
Ho | -0.222 |
Rib | -0.212 |
H | -0.190 |
Bub | -0.176 |
Bun | -0.173 |
Token | happiness |
Token ID | 12157 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
eks | +0.068 |
oos | +0.068 |
elevator | +0.065 |
cases | +0.065 |
ases | +0.065 |
Neg logit contributions | |
Ho | -0.088 |
Rib | -0.087 |
H | -0.079 |
Bub | -0.073 |
Bun | -0.072 |
Token | . |
Token ID | 13 |
Feature activation | -2.91e-04 |
Pos logit contributions | |
cases | +0.037 |
oos | +0.037 |
he | +0.036 |
eks | +0.036 |
led | +0.036 |
Neg logit contributions | |
Ho | -0.057 |
Rib | -0.056 |
H | -0.051 |
Butter | -0.047 |
Bub | -0.047 |
Token | He |
Token ID | 679 |
Feature activation | +3.11e-02 |
Pos logit contributions | |
Ho | +0.001 |
Rib | +0.001 |
onna | +0.000 |
accidentally | +0.000 |
please | +0.000 |
Neg logit contributions | |
ays | -0.000 |
he | -0.000 |
roads | -0.000 |
oos | -0.000 |
elevator | -0.000 |
Token | began |
Token ID | 2540 |
Feature activation | -8.04e-02 |
Pos logit contributions | |
eks | +0.051 |
oos | +0.051 |
elevator | +0.050 |
cases | +0.049 |
led | +0.049 |
Neg logit contributions | |
Ho | -0.052 |
Rib | -0.051 |
H | -0.046 |
Bub | -0.042 |
Butter | -0.041 |
Token | excited |
Token ID | 6568 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
eks | +0.027 |
elevator | +0.026 |
cases | +0.026 |
oos | +0.026 |
led | +0.026 |
Neg logit contributions | |
Ho | -0.028 |
Rib | -0.027 |
H | -0.024 |
Bun | -0.022 |
Bub | -0.022 |
Token | - |
Token ID | 532 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
eks | +0.185 |
oos | +0.185 |
ays | +0.180 |
elevator | +0.179 |
cases | +0.179 |
Neg logit contributions | |
Ho | -0.202 |
Rib | -0.197 |
H | -0.177 |
Bub | -0.164 |
Bun | -0.162 |
Token | she |
Token ID | 673 |
Feature activation | -3.09e-02 |
Pos logit contributions | |
elevator | +0.034 |
cases | +0.033 |
oos | +0.032 |
eks | +0.032 |
tickets | +0.031 |
Neg logit contributions | |
Ho | -0.035 |
Rib | -0.035 |
H | -0.030 |
Bub | -0.028 |
Butter | -0.027 |
Token | could |
Token ID | 714 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
Ho | +0.050 |
Rib | +0.050 |
H | +0.044 |
Butter | +0.040 |
Bub | +0.040 |
Neg logit contributions | |
eks | -0.052 |
oos | -0.051 |
elevator | -0.051 |
cases | -0.050 |
led | -0.049 |
Token | hardly |
Token ID | 8941 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
Ho | +0.192 |
Rib | +0.187 |
H | +0.167 |
accidentally | +0.158 |
Bun | +0.157 |
Neg logit contributions | |
ays | -0.226 |
agged | -0.225 |
oos | -0.224 |
ases | -0.215 |
led | -0.215 |
Token | contain |
Token ID | 3994 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
eks | +0.022 |
he | +0.021 |
oos | +0.021 |
elevator | +0.021 |
cases | +0.021 |
Neg logit contributions | |
Ho | -0.054 |
Rib | -0.053 |
H | -0.049 |
Bun | -0.046 |
Bub | -0.045 |
Token | herself |
Token ID | 5223 |
Feature activation | +1.81e-04 |
Pos logit contributions | |
cases | +0.287 |
ays | +0.286 |
eks | +0.282 |
oos | +0.282 |
agged | +0.279 |
Neg logit contributions | |
Ho | -0.210 |
Rib | -0.200 |
H | -0.182 |
Bun | -0.166 |
Bub | -0.164 |
Token | . |
Token ID | 13 |
Feature activation | +6.90e-03 |
Pos logit contributions | |
eks | +0.000 |
oos | +0.000 |
ases | +0.000 |
agged | +0.000 |
ays | +0.000 |
Neg logit contributions | |
Ho | -0.000 |
Rib | -0.000 |
H | -0.000 |
Bun | -0.000 |
Bub | -0.000 |
Token | She |
Token ID | 1375 |
Feature activation | -5.32e-03 |
Pos logit contributions | |
he | +0.009 |
oos | +0.008 |
ays | +0.008 |
downtown | +0.008 |
roads | +0.008 |
Neg logit contributions | |
Rib | -0.013 |
Ho | -0.012 |
onna | -0.011 |
Stri | -0.010 |
accidentally | -0.010 |
Token | started |
Token ID | 2067 |
Feature activation | -7.43e-02 |
Pos logit contributions | |
Ho | +0.009 |
Rib | +0.008 |
H | +0.008 |
Bub | +0.007 |
Bun | +0.007 |
Neg logit contributions | |
eks | -0.009 |
oos | -0.009 |
elevator | -0.009 |
cases | -0.009 |
ays | -0.009 |
Token | to |
Token ID | 284 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
Ho | +0.108 |
Rib | +0.105 |
H | +0.093 |
Bub | +0.084 |
Bun | +0.083 |
Neg logit contributions | |
eks | -0.136 |
oos | -0.136 |
ays | -0.133 |
cases | -0.133 |
agged | -0.132 |
Token | the |
Token ID | 262 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
Rib | +0.048 |
Ho | +0.047 |
H | +0.042 |
Bub | +0.040 |
accidentally | +0.039 |
Neg logit contributions | |
oos | -0.046 |
elevator | -0.046 |
ays | -0.046 |
eks | -0.045 |
ases | -0.045 |
Token | herb |
Token ID | 21047 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
Rib | +0.035 |
Ho | +0.034 |
H | +0.031 |
Bun | +0.030 |
Bub | +0.030 |
Neg logit contributions | |
elevator | -0.026 |
are | -0.024 |
eks | -0.023 |
ps | -0.023 |
oos | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
Ho | +0.189 |
Rib | +0.186 |
H | +0.169 |
Bub | +0.157 |
Bun | +0.156 |
Neg logit contributions | |
oos | -0.143 |
eks | -0.142 |
elevator | -0.140 |
ays | -0.140 |
cases | -0.139 |
Token | It |
Token ID | 632 |
Feature activation | -3.83e-03 |
Pos logit contributions | |
Rib | +0.043 |
Ho | +0.043 |
Bun | +0.038 |
accidentally | +0.037 |
Bub | +0.037 |
Neg logit contributions | |
he | -0.026 |
ays | -0.025 |
ases | -0.023 |
elevator | -0.022 |
otion | -0.022 |
Token | is |
Token ID | 318 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
Ho | +0.008 |
Rib | +0.008 |
H | +0.008 |
Bub | +0.007 |
Butter | +0.007 |
Neg logit contributions | |
oos | -0.004 |
eks | -0.004 |
cases | -0.004 |
agged | -0.004 |
ays | -0.004 |
Token | lav |
Token ID | 21606 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
ps | +0.017 |
elevator | +0.016 |
ags | +0.015 |
cks | +0.015 |
led | +0.015 |
Neg logit contributions | |
Rib | -0.026 |
Ho | -0.025 |
H | -0.024 |
Bun | -0.023 |
Bub | -0.022 |
Token | ender |
Token ID | 2194 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
Ho | +0.248 |
Rib | +0.243 |
H | +0.217 |
Bub | +0.193 |
Bun | +0.192 |
Neg logit contributions | |
ays | -0.400 |
oos | -0.398 |
eks | -0.389 |
ases | -0.384 |
agged | -0.384 |
Token | . |
Token ID | 13 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
Ho | +0.319 |
Rib | +0.309 |
H | +0.275 |
Butter | +0.256 |
Bub | +0.254 |
Neg logit contributions | |
eks | -0.296 |
oos | -0.287 |
agged | -0.282 |
cases | -0.280 |
ases | -0.278 |
Token | " |
Token ID | 366 |
Feature activation | -5.65e-03 |
Pos logit contributions | |
Ho | +0.044 |
Rib | +0.044 |
Bun | +0.040 |
accidentally | +0.039 |
Bub | +0.039 |
Neg logit contributions | |
ays | -0.025 |
he | -0.024 |
ases | -0.022 |
otion | -0.022 |
are | -0.021 |
Token | That |
Token ID | 2504 |
Feature activation | +6.33e-02 |
Pos logit contributions | |
Ho | +0.010 |
Rib | +0.009 |
H | +0.008 |
Bun | +0.008 |
Bub | +0.008 |
Neg logit contributions | |
ases | -0.009 |
eks | -0.009 |
ays | -0.009 |
he | -0.009 |
oos | -0.008 |
Token | is |
Token ID | 318 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
oos | +0.063 |
elevator | +0.063 |
ays | +0.063 |
eks | +0.063 |
ases | +0.059 |
Neg logit contributions | |
Ho | -0.143 |
Rib | -0.141 |
H | -0.130 |
Bub | -0.124 |
Bun | -0.124 |
Token | source |
Token ID | 2723 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
Rib | +0.034 |
Ho | +0.034 |
H | +0.031 |
Bub | +0.029 |
Bun | +0.029 |
Neg logit contributions | |
elevator | -0.023 |
eks | -0.021 |
oos | -0.021 |
tickets | -0.020 |
he | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
Ho | +0.040 |
Rib | +0.039 |
H | +0.036 |
Bun | +0.034 |
Lucky | +0.034 |
Neg logit contributions | |
eks | -0.032 |
oos | -0.030 |
agged | -0.030 |
ases | -0.029 |
ags | -0.029 |
Token | It |
Token ID | 632 |
Feature activation | -2.86e-03 |
Pos logit contributions | |
Rib | +0.033 |
Ho | +0.032 |
please | +0.031 |
H | +0.030 |
Cle | +0.029 |
Neg logit contributions | |
agged | -0.020 |
ays | -0.019 |
ags | -0.018 |
he | -0.018 |
oos | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | +5.51e-03 |
Pos logit contributions | |
Ho | +0.005 |
Rib | +0.005 |
H | +0.004 |
Bub | +0.004 |
Bun | +0.004 |
Neg logit contributions | |
eks | -0.005 |
oos | -0.005 |
elevator | -0.005 |
led | -0.005 |
cases | -0.004 |
Token | their |
Token ID | 511 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
elevator | +0.007 |
ps | +0.006 |
ags | +0.006 |
eks | +0.006 |
ases | +0.006 |
Neg logit contributions | |
Rib | -0.011 |
Ho | -0.011 |
H | -0.010 |
Bun | -0.010 |
Bub | -0.009 |
Token | ham |
Token ID | 8891 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
elevator | +0.029 |
oos | +0.028 |
eks | +0.028 |
ases | +0.028 |
ays | +0.027 |
Neg logit contributions | |
Ho | -0.036 |
Rib | -0.036 |
H | -0.033 |
Bun | -0.031 |
Bub | -0.030 |
Token | ster |
Token ID | 1706 |
Feature activation | -5.39e-02 |
Pos logit contributions | |
Ho | +0.333 |
Rib | +0.331 |
H | +0.306 |
Bub | +0.284 |
Bun | +0.281 |
Neg logit contributions | |
oos | -0.267 |
eks | -0.265 |
ays | -0.261 |
cases | -0.260 |
ases | -0.259 |
Token | cage |
Token ID | 17186 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
Ho | +0.099 |
Rib | +0.098 |
H | +0.089 |
Bub | +0.083 |
Bun | +0.081 |
Neg logit contributions | |
eks | -0.078 |
oos | -0.077 |
elevator | -0.075 |
cases | -0.075 |
ays | -0.075 |
Token | . |
Token ID | 13 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
eks | +0.020 |
ases | +0.019 |
agged | +0.019 |
ays | +0.019 |
oos | +0.019 |
Neg logit contributions | |
Ho | -0.016 |
Rib | -0.015 |
H | -0.014 |
Bun | -0.013 |
Lucky | -0.013 |
Token | Their |
Token ID | 5334 |
Feature activation | -4.07e-02 |
Pos logit contributions | |
Rib | +0.030 |
Ho | +0.030 |
please | +0.030 |
Bun | +0.028 |
H | +0.028 |
Neg logit contributions | |
ays | -0.023 |
agged | -0.022 |
he | -0.022 |
ases | -0.021 |
ags | -0.021 |
Token | ham |
Token ID | 8891 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
Rib | +0.082 |
Ho | +0.082 |
H | +0.077 |
Bun | +0.073 |
Bub | +0.072 |
Neg logit contributions | |
oos | -0.050 |
eks | -0.050 |
elevator | -0.049 |
ases | -0.048 |
tickets | -0.047 |
Token | ham |
Token ID | 8891 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
elevator | +0.029 |
oos | +0.028 |
eks | +0.028 |
ases | +0.028 |
ays | +0.027 |
Neg logit contributions | |
Ho | -0.036 |
Rib | -0.036 |
H | -0.033 |
Bun | -0.031 |
Bub | -0.030 |
Token | ster |
Token ID | 1706 |
Feature activation | -5.39e-02 |
Pos logit contributions | |
Ho | +0.333 |
Rib | +0.331 |
H | +0.306 |
Bub | +0.284 |
Bun | +0.281 |
Neg logit contributions | |
oos | -0.267 |
eks | -0.265 |
ays | -0.261 |
cases | -0.260 |
ases | -0.259 |
Token | cage |
Token ID | 17186 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
Ho | +0.099 |
Rib | +0.098 |
H | +0.089 |
Bub | +0.083 |
Bun | +0.081 |
Neg logit contributions | |
eks | -0.078 |
oos | -0.077 |
elevator | -0.075 |
cases | -0.075 |
ays | -0.075 |
Token | . |
Token ID | 13 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
eks | +0.020 |
ases | +0.019 |
agged | +0.019 |
ays | +0.019 |
oos | +0.019 |
Neg logit contributions | |
Ho | -0.016 |
Rib | -0.015 |
H | -0.014 |
Bun | -0.013 |
Lucky | -0.013 |
Token | Their |
Token ID | 5334 |
Feature activation | -4.07e-02 |
Pos logit contributions | |
Rib | +0.030 |
Ho | +0.030 |
please | +0.030 |
Bun | +0.028 |
H | +0.028 |
Neg logit contributions | |
ays | -0.023 |
agged | -0.022 |
he | -0.022 |
ases | -0.021 |
ags | -0.021 |
Token | ham |
Token ID | 8891 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
Rib | +0.082 |
Ho | +0.082 |
H | +0.077 |
Bun | +0.073 |
Bub | +0.072 |
Neg logit contributions | |
oos | -0.050 |
eks | -0.050 |
elevator | -0.049 |
ases | -0.048 |
tickets | -0.047 |
Token | ster |
Token ID | 1706 |
Feature activation | -5.41e-02 |
Pos logit contributions | |
Ho | +0.322 |
Rib | +0.317 |
H | +0.292 |
Bub | +0.267 |
Bun | +0.267 |
Neg logit contributions | |
eks | -0.282 |
oos | -0.281 |
ays | -0.275 |
ases | -0.273 |
cases | -0.272 |
Token | , |
Token ID | 11 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
Ho | +0.088 |
Rib | +0.087 |
H | +0.080 |
Bub | +0.071 |
Bun | +0.071 |
Neg logit contributions | |
eks | -0.089 |
oos | -0.087 |
cases | -0.087 |
elevator | -0.086 |
led | -0.086 |
Token | Fl |
Token ID | 1610 |
Feature activation | -9.69e-02 |
Pos logit contributions | |
Rib | +0.022 |
Ho | +0.022 |
H | +0.020 |
Bub | +0.018 |
Bun | +0.018 |
Neg logit contributions | |
eks | -0.021 |
elevator | -0.020 |
oos | -0.020 |
ases | -0.020 |
ays | -0.020 |
Token | uffy |
Token ID | 15352 |
Feature activation | -4.76e-02 |
Pos logit contributions | |
Ho | +0.166 |
Rib | +0.163 |
H | +0.147 |
Bub | +0.137 |
Bun | +0.136 |
Neg logit contributions | |
oos | -0.154 |
eks | -0.153 |
ays | -0.150 |
ases | -0.149 |
elevator | -0.147 |
Token | , |
Token ID | 11 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
Rib | +0.091 |
Ho | +0.091 |
H | +0.084 |
Bub | +0.078 |
Bun | +0.076 |
Neg logit contributions | |
oos | -0.062 |
eks | -0.060 |
elevator | -0.060 |
cases | -0.060 |
led | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
Ho | +0.170 |
Rib | +0.166 |
H | +0.151 |
Bub | +0.140 |
Bun | +0.139 |
Neg logit contributions | |
oos | -0.136 |
eks | -0.136 |
ays | -0.132 |
elevator | -0.131 |
cases | -0.131 |
Token | He |
Token ID | 679 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
Ho | +0.049 |
Rib | +0.049 |
accidentally | +0.047 |
Ros | +0.047 |
Bun | +0.046 |
Neg logit contributions | |
ags | -0.021 |
agged | -0.021 |
ases | -0.020 |
he | -0.020 |
ays | -0.020 |
Token | likes |
Token ID | 7832 |
Feature activation | +2.31e-02 |
Pos logit contributions | |
eks | +0.017 |
oos | +0.017 |
ays | +0.016 |
cases | +0.016 |
elevator | +0.016 |
Neg logit contributions | |
Ho | -0.033 |
Rib | -0.033 |
H | -0.030 |
Bub | -0.028 |
Bun | -0.028 |
Token | cheese |
Token ID | 9891 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
ases | +0.030 |
elevator | +0.030 |
oos | +0.029 |
tickets | +0.028 |
out | +0.028 |
Neg logit contributions | |
Ho | -0.042 |
Bun | -0.042 |
H | -0.041 |
Rib | -0.040 |
Ros | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | -3.13e-03 |
Pos logit contributions | |
Ho | +0.199 |
Rib | +0.196 |
H | +0.184 |
Bun | +0.171 |
Bub | +0.169 |
Neg logit contributions | |
ases | -0.129 |
oos | -0.128 |
ays | -0.125 |
eks | -0.125 |
he | -0.125 |
Token | ham |
Token ID | 8891 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
Ho | +0.006 |
H | +0.006 |
Rib | +0.006 |
Bun | +0.005 |
Bub | +0.005 |
Neg logit contributions | |
ases | -0.004 |
elevator | -0.004 |
eks | -0.004 |
oos | -0.004 |
are | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | -1.98e-03 |
Pos logit contributions | |
Ho | +0.360 |
Rib | +0.353 |
H | +0.335 |
Bun | +0.318 |
Bub | +0.312 |
Neg logit contributions | |
ases | -0.239 |
ays | -0.238 |
oos | -0.237 |
eks | -0.233 |
agged | -0.230 |
Token | lettuce |
Token ID | 39406 |
Feature activation | -9.98e-02 |
Pos logit contributions | |
Ho | +0.003 |
Rib | +0.003 |
H | +0.003 |
Bun | +0.003 |
Bub | +0.003 |
Neg logit contributions | |
ases | -0.003 |
elevator | -0.003 |
eks | -0.003 |
oos | -0.003 |
led | -0.003 |
Token | too |
Token ID | 1165 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
Ho | +0.190 |
Rib | +0.187 |
H | +0.172 |
Bun | +0.160 |
Bub | +0.160 |
Neg logit contributions | |
oos | -0.134 |
eks | -0.134 |
ays | -0.132 |
ases | -0.132 |
cases | -0.131 |
Token | . |
Token ID | 13 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
Ho | +0.051 |
Rib | +0.048 |
H | +0.044 |
accidentally | +0.039 |
Lucky | +0.038 |
Neg logit contributions | |
eks | -0.074 |
oos | -0.072 |
ays | -0.071 |
agged | -0.068 |
ases | -0.068 |
Token | He |
Token ID | 679 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
Rib | +0.044 |
Ho | +0.044 |
accidentally | +0.043 |
Ros | +0.043 |
Bun | +0.042 |
Neg logit contributions | |
ays | -0.020 |
ases | -0.019 |
ags | -0.019 |
agged | -0.019 |
are | -0.018 |
Token | because |
Token ID | 780 |
Feature activation | +4.25e-02 |
Pos logit contributions | |
oos | +0.030 |
eks | +0.030 |
ays | +0.029 |
elevator | +0.029 |
cases | +0.029 |
Neg logit contributions | |
Ho | -0.037 |
Rib | -0.036 |
H | -0.032 |
Bub | -0.030 |
Bun | -0.030 |
Token | I |
Token ID | 314 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
elevator | +0.068 |
ays | +0.063 |
ases | +0.063 |
cases | +0.062 |
oos | +0.062 |
Neg logit contributions | |
Ho | -0.073 |
Rib | -0.072 |
H | -0.062 |
Bun | -0.059 |
Fro | -0.059 |
Token | like |
Token ID | 588 |
Feature activation | -5.46e-03 |
Pos logit contributions | |
Ho | +0.066 |
Rib | +0.065 |
H | +0.059 |
Bub | +0.054 |
Bun | +0.054 |
Neg logit contributions | |
oos | -0.057 |
eks | -0.057 |
ays | -0.055 |
elevator | -0.055 |
cases | -0.055 |
Token | cheese |
Token ID | 9891 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
Ho | +0.010 |
Rib | +0.010 |
H | +0.009 |
Bun | +0.009 |
Ros | +0.009 |
Neg logit contributions | |
elevator | -0.007 |
ases | -0.007 |
tickets | -0.007 |
oos | -0.007 |
ays | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | +4.98e-04 |
Pos logit contributions | |
Ho | +0.183 |
Rib | +0.179 |
H | +0.161 |
Bub | +0.150 |
Bun | +0.149 |
Neg logit contributions | |
eks | -0.151 |
oos | -0.150 |
ays | -0.146 |
elevator | -0.145 |
cases | -0.145 |
Token | ham |
Token ID | 8891 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
elevator | +0.001 |
ases | +0.001 |
bags | +0.001 |
oos | +0.001 |
tickets | +0.001 |
Neg logit contributions | |
Ho | -0.001 |
Rib | -0.001 |
H | -0.001 |
Bun | -0.001 |
Bub | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | +1.39e-03 |
Pos logit contributions | |
Ho | +0.337 |
Rib | +0.330 |
H | +0.300 |
Bub | +0.278 |
Bun | +0.277 |
Neg logit contributions | |
oos | -0.269 |
eks | -0.269 |
ays | -0.262 |
ases | -0.261 |
cases | -0.260 |
Token | yellow |
Token ID | 7872 |
Feature activation | +2.67e-03 |
Pos logit contributions | |
elevator | +0.002 |
ases | +0.002 |
bags | +0.002 |
are | +0.002 |
eks | +0.002 |
Neg logit contributions | |
Rib | -0.002 |
Ho | -0.002 |
H | -0.002 |
Bun | -0.002 |
Bub | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | +2.15e-03 |
Pos logit contributions | |
oos | +0.004 |
eks | +0.004 |
ays | +0.004 |
elevator | +0.004 |
cases | +0.004 |
Neg logit contributions | |
Ho | -0.005 |
Rib | -0.005 |
H | -0.004 |
Bub | -0.004 |
Bun | -0.004 |
Token | red |
Token ID | 2266 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
elevator | +0.003 |
are | +0.003 |
bags | +0.003 |
ases | +0.003 |
eks | +0.003 |
Neg logit contributions | |
Ho | -0.004 |
Rib | -0.003 |
H | -0.003 |
Bun | -0.003 |
Bub | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
eks | +0.053 |
oos | +0.051 |
cases | +0.051 |
elevator | +0.050 |
ays | +0.050 |
Neg logit contributions | |
Ho | -0.064 |
Rib | -0.062 |
H | -0.056 |
Bub | -0.052 |
Bun | -0.051 |
Token | herb |
Token ID | 21047 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
Rib | +0.035 |
Ho | +0.034 |
H | +0.031 |
Bun | +0.030 |
Bub | +0.030 |
Neg logit contributions | |
elevator | -0.026 |
are | -0.024 |
eks | -0.023 |
ps | -0.023 |
oos | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
Ho | +0.189 |
Rib | +0.186 |
H | +0.169 |
Bub | +0.157 |
Bun | +0.156 |
Neg logit contributions | |
oos | -0.143 |
eks | -0.142 |
elevator | -0.140 |
ays | -0.140 |
cases | -0.139 |
Token | It |
Token ID | 632 |
Feature activation | -3.83e-03 |
Pos logit contributions | |
Rib | +0.043 |
Ho | +0.043 |
Bun | +0.038 |
accidentally | +0.037 |
Bub | +0.037 |
Neg logit contributions | |
he | -0.026 |
ays | -0.025 |
ases | -0.023 |
elevator | -0.022 |
otion | -0.022 |
Token | is |
Token ID | 318 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
Ho | +0.008 |
Rib | +0.008 |
H | +0.008 |
Bub | +0.007 |
Butter | +0.007 |
Neg logit contributions | |
oos | -0.004 |
eks | -0.004 |
cases | -0.004 |
agged | -0.004 |
ays | -0.004 |
Token | lav |
Token ID | 21606 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
ps | +0.017 |
elevator | +0.016 |
ags | +0.015 |
cks | +0.015 |
led | +0.015 |
Neg logit contributions | |
Rib | -0.026 |
Ho | -0.025 |
H | -0.024 |
Bun | -0.023 |
Bub | -0.022 |
Token | ender |
Token ID | 2194 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
Ho | +0.248 |
Rib | +0.243 |
H | +0.217 |
Bub | +0.193 |
Bun | +0.192 |
Neg logit contributions | |
ays | -0.400 |
oos | -0.398 |
eks | -0.389 |
ases | -0.384 |
agged | -0.384 |
Token | . |
Token ID | 13 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
Ho | +0.319 |
Rib | +0.309 |
H | +0.275 |
Butter | +0.256 |
Bub | +0.254 |
Neg logit contributions | |
eks | -0.296 |
oos | -0.287 |
agged | -0.282 |
cases | -0.280 |
ases | -0.278 |
Token | " |
Token ID | 366 |
Feature activation | -5.65e-03 |
Pos logit contributions | |
Ho | +0.044 |
Rib | +0.044 |
Bun | +0.040 |
accidentally | +0.039 |
Bub | +0.039 |
Neg logit contributions | |
ays | -0.025 |
he | -0.024 |
ases | -0.022 |
otion | -0.022 |
are | -0.021 |
Token | That |
Token ID | 2504 |
Feature activation | +6.33e-02 |
Pos logit contributions | |
Ho | +0.010 |
Rib | +0.009 |
H | +0.008 |
Bun | +0.008 |
Bub | +0.008 |
Neg logit contributions | |
ases | -0.009 |
eks | -0.009 |
ays | -0.009 |
he | -0.009 |
oos | -0.008 |
Token | is |
Token ID | 318 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
oos | +0.063 |
elevator | +0.063 |
ays | +0.063 |
eks | +0.063 |
ases | +0.059 |
Neg logit contributions | |
Ho | -0.143 |
Rib | -0.141 |
H | -0.130 |
Bub | -0.124 |
Bun | -0.124 |
Token | not |
Token ID | 407 |
Feature activation | -9.81e-03 |
Pos logit contributions | |
ps | +0.017 |
elevator | +0.016 |
eks | +0.016 |
are | +0.016 |
cks | +0.016 |
Neg logit contributions | |
Rib | -0.027 |
Ho | -0.025 |
Bun | -0.023 |
H | -0.023 |
stub | -0.022 |
Token | cheeks |
Token ID | 25839 |
Feature activation | +4.80e-02 |
Pos logit contributions | |
oos | +0.093 |
eks | +0.093 |
elevator | +0.091 |
ays | +0.090 |
ases | +0.090 |
Neg logit contributions | |
Ho | -0.124 |
Rib | -0.121 |
H | -0.110 |
Bub | -0.103 |
Bun | -0.103 |
Token | . |
Token ID | 13 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
eks | +0.090 |
oos | +0.087 |
agged | +0.086 |
cases | +0.084 |
elevator | +0.083 |
Neg logit contributions | |
Ho | -0.071 |
Rib | -0.068 |
H | -0.060 |
Tw | -0.055 |
Lucky | -0.055 |
Token | That |
Token ID | 1320 |
Feature activation | +3.06e-02 |
Pos logit contributions | |
Ho | +0.030 |
Rib | +0.029 |
accidentally | +0.026 |
H | +0.026 |
Bun | +0.026 |
Neg logit contributions | |
ays | -0.024 |
ases | -0.023 |
oos | -0.022 |
elevator | -0.022 |
eks | -0.021 |
Token | is |
Token ID | 318 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
eks | +0.031 |
elevator | +0.031 |
oos | +0.031 |
ays | +0.031 |
ases | +0.030 |
Neg logit contributions | |
Ho | -0.069 |
Rib | -0.068 |
H | -0.063 |
Bub | -0.060 |
Bun | -0.059 |
Token | a |
Token ID | 257 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
elevator | +0.019 |
cks | +0.019 |
ps | +0.018 |
tickets | +0.017 |
eks | +0.017 |
Neg logit contributions | |
Rib | -0.033 |
Ho | -0.032 |
Bun | -0.030 |
H | -0.030 |
Ros | -0.029 |
Token | ham |
Token ID | 8891 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
Rib | +0.035 |
Ho | +0.034 |
H | +0.033 |
Stri | +0.031 |
Bun | +0.031 |
Neg logit contributions | |
elevator | -0.021 |
tickets | -0.018 |
eks | -0.018 |
ases | -0.018 |
stadium | -0.018 |
Token | ster |
Token ID | 1706 |
Feature activation | -5.56e-02 |
Pos logit contributions | |
Ho | +0.320 |
Rib | +0.313 |
H | +0.303 |
Bun | +0.284 |
Bub | +0.279 |
Neg logit contributions | |
ays | -0.266 |
ases | -0.265 |
oos | -0.257 |
eks | -0.255 |
out | -0.254 |
Token | . |
Token ID | 13 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
Ho | +0.107 |
Rib | +0.106 |
H | +0.098 |
Bub | +0.091 |
Bun | +0.091 |
Neg logit contributions | |
elevator | -0.072 |
ases | -0.072 |
ays | -0.072 |
eks | -0.071 |
oos | -0.071 |
Token | A |
Token ID | 317 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
Ho | +0.028 |
Rib | +0.027 |
accidentally | +0.026 |
Bun | +0.025 |
H | +0.025 |
Neg logit contributions | |
ases | -0.025 |
ays | -0.025 |
eks | -0.023 |
oos | -0.023 |
elevator | -0.022 |
Token | rabbit |
Token ID | 22746 |
Feature activation | -6.44e-02 |
Pos logit contributions | |
Rib | +0.055 |
H | +0.055 |
Ho | +0.054 |
Bun | +0.053 |
Bub | +0.050 |
Neg logit contributions | |
elevator | -0.033 |
tickets | -0.032 |
are | -0.032 |
men | -0.030 |
envelope | -0.030 |
Token | has |
Token ID | 468 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
Ho | +0.126 |
Rib | +0.123 |
H | +0.111 |
Bub | +0.104 |
Bun | +0.103 |
Neg logit contributions | |
oos | -0.088 |
eks | -0.087 |
cases | -0.085 |
ays | -0.084 |
elevator | -0.084 |
Token | cheese |
Token ID | 9891 |
Feature activation | -9.84e-02 |
Pos logit contributions | |
ays | +0.047 |
oos | +0.046 |
eks | +0.046 |
ases | +0.046 |
elevator | +0.046 |
Neg logit contributions | |
Ho | -0.063 |
Rib | -0.062 |
H | -0.057 |
Bun | -0.053 |
Bub | -0.052 |
Token | and |
Token ID | 290 |
Feature activation | +2.33e-03 |
Pos logit contributions | |
Ho | +0.180 |
Rib | +0.176 |
H | +0.160 |
Bub | +0.148 |
Bun | +0.147 |
Neg logit contributions | |
eks | -0.144 |
oos | -0.144 |
ays | -0.139 |
elevator | -0.139 |
cases | -0.139 |
Token | ol |
Token ID | 25776 |
Feature activation | -3.24e-02 |
Pos logit contributions | |
elevator | +0.003 |
eks | +0.003 |
ases | +0.003 |
oos | +0.003 |
led | +0.003 |
Neg logit contributions | |
Ho | -0.004 |
Rib | -0.004 |
H | -0.004 |
Bun | -0.004 |
Bub | -0.004 |
Token | ives |
Token ID | 1083 |
Feature activation | -4.18e-02 |
Pos logit contributions | |
Ho | +0.083 |
Rib | +0.081 |
H | +0.075 |
Bub | +0.072 |
Lucky | +0.071 |
Neg logit contributions | |
eks | -0.025 |
oos | -0.024 |
elevator | -0.024 |
cases | -0.022 |
ays | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | +3.73e-03 |
Pos logit contributions | |
Ho | +0.074 |
Rib | +0.072 |
H | +0.064 |
Bub | +0.060 |
Lucky | +0.059 |
Neg logit contributions | |
eks | -0.065 |
oos | -0.064 |
cases | -0.063 |
elevator | -0.063 |
agged | -0.062 |
Token | ham |
Token ID | 8891 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
elevator | +0.005 |
ases | +0.005 |
eks | +0.005 |
ps | +0.005 |
cases | +0.005 |
Neg logit contributions | |
Ho | -0.007 |
Rib | -0.007 |
H | -0.007 |
Bun | -0.006 |
Bub | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
Ho | +0.350 |
Rib | +0.342 |
H | +0.321 |
Bun | +0.304 |
Bub | +0.298 |
Neg logit contributions | |
ays | -0.247 |
oos | -0.246 |
ases | -0.244 |
eks | -0.242 |
cases | -0.237 |
Token | They |
Token ID | 1119 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
Ho | +0.045 |
Rib | +0.044 |
Bun | +0.042 |
shadow | +0.042 |
H | +0.042 |
Neg logit contributions | |
ays | -0.013 |
ases | -0.012 |
oos | -0.012 |
ags | -0.010 |
are | -0.010 |
Token | see |
Token ID | 766 |
Feature activation | +3.52e-02 |
Pos logit contributions | |
oos | +0.060 |
ays | +0.058 |
agged | +0.058 |
cases | +0.057 |
eks | +0.057 |
Neg logit contributions | |
Ho | -0.070 |
Rib | -0.068 |
H | -0.060 |
Bub | -0.057 |
accidentally | -0.056 |
Token | mom |
Token ID | 1995 |
Feature activation | +5.96e-02 |
Pos logit contributions | |
ays | +0.047 |
oos | +0.047 |
eks | +0.047 |
elevator | +0.047 |
ases | +0.047 |
Neg logit contributions | |
Ho | -0.067 |
Rib | -0.066 |
H | -0.061 |
Bun | -0.057 |
Bub | -0.056 |
Token | 's |
Token ID | 338 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
oos | +0.098 |
eks | +0.096 |
agged | +0.094 |
cases | +0.093 |
elevator | +0.093 |
Neg logit contributions | |
Ho | -0.101 |
Rib | -0.099 |
H | -0.087 |
Bub | -0.081 |
Butter | -0.079 |
Token | . |
Token ID | 13 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
elevator | +0.003 |
oos | +0.003 |
eks | +0.003 |
ays | +0.003 |
cases | +0.003 |
Neg logit contributions | |
Ho | -0.004 |
Rib | -0.004 |
H | -0.003 |
Bub | -0.003 |
Bun | -0.003 |
Token | It |
Token ID | 632 |
Feature activation | -2.46e-03 |
Pos logit contributions | |
Ho | +0.042 |
Rib | +0.041 |
Bun | +0.039 |
please | +0.038 |
accidentally | +0.038 |
Neg logit contributions | |
ays | -0.017 |
ases | -0.016 |
ags | -0.016 |
agged | -0.016 |
oos | -0.015 |
Token | had |
Token ID | 550 |
Feature activation | -3.67e-02 |
Pos logit contributions | |
Ho | +0.005 |
Rib | +0.004 |
H | +0.004 |
Bun | +0.004 |
Bub | +0.004 |
Neg logit contributions | |
led | -0.003 |
eks | -0.003 |
hed | -0.003 |
elevator | -0.003 |
oos | -0.003 |
Token | cheese |
Token ID | 9891 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
Ho | +0.070 |
Rib | +0.068 |
H | +0.064 |
Bun | +0.060 |
Bub | +0.059 |
Neg logit contributions | |
eks | -0.049 |
oos | -0.049 |
ases | -0.048 |
elevator | -0.048 |
agged | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | -2.28e-03 |
Pos logit contributions | |
Ho | +0.182 |
Rib | +0.177 |
H | +0.158 |
Bub | +0.148 |
Bun | +0.146 |
Neg logit contributions | |
eks | -0.153 |
oos | -0.152 |
elevator | -0.147 |
cases | -0.147 |
ays | -0.146 |
Token | ham |
Token ID | 8891 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
Ho | +0.004 |
Rib | +0.004 |
H | +0.004 |
Bun | +0.004 |
Bub | +0.003 |
Neg logit contributions | |
elevator | -0.003 |
led | -0.003 |
oos | -0.003 |
ases | -0.003 |
ps | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | -8.13e-04 |
Pos logit contributions | |
Ho | +0.347 |
Rib | +0.341 |
H | +0.317 |
Bun | +0.296 |
Bub | +0.294 |
Neg logit contributions | |
oos | -0.248 |
eks | -0.246 |
ays | -0.245 |
ases | -0.243 |
cases | -0.239 |
Token | bread |
Token ID | 8509 |
Feature activation | -5.97e-02 |
Pos logit contributions | |
Ho | +0.001 |
Rib | +0.001 |
H | +0.001 |
Bun | +0.001 |
Bub | +0.001 |
Neg logit contributions | |
elevator | -0.001 |
led | -0.001 |
ases | -0.001 |
are | -0.001 |
bags | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
Ho | +0.114 |
Rib | +0.113 |
H | +0.104 |
Bun | +0.097 |
Bub | +0.096 |
Neg logit contributions | |
oos | -0.080 |
eks | -0.079 |
ays | -0.079 |
ases | -0.079 |
cases | -0.077 |
Token | He |
Token ID | 679 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
Ho | +0.039 |
Rib | +0.039 |
Bun | +0.038 |
please | +0.037 |
Ros | +0.036 |
Neg logit contributions | |
ays | -0.020 |
ases | -0.019 |
ags | -0.018 |
agged | -0.018 |
oos | -0.016 |
Token | also |
Token ID | 635 |
Feature activation | -7.67e-02 |
Pos logit contributions | |
led | +0.030 |
eks | +0.029 |
oos | +0.028 |
elevator | +0.028 |
cases | +0.027 |
Neg logit contributions | |
Ho | -0.041 |
Rib | -0.040 |
H | -0.038 |
Bun | -0.035 |
Bub | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | +2.93e-03 |
Pos logit contributions | |
Ho | +0.103 |
Rib | +0.101 |
H | +0.090 |
Bub | +0.084 |
Bun | +0.083 |
Neg logit contributions | |
eks | -0.091 |
oos | -0.091 |
elevator | -0.088 |
cases | -0.088 |
ays | -0.088 |
Token | the |
Token ID | 262 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
led | +0.004 |
eks | +0.004 |
elevator | +0.004 |
oos | +0.004 |
cases | +0.004 |
Neg logit contributions | |
Rib | -0.005 |
Ho | -0.005 |
H | -0.005 |
Bun | -0.005 |
Bub | -0.004 |
Token | cheese |
Token ID | 9891 |
Feature activation | -9.95e-02 |
Pos logit contributions | |
Ho | +0.034 |
Rib | +0.034 |
H | +0.032 |
Bun | +0.030 |
Bub | +0.030 |
Neg logit contributions | |
elevator | -0.022 |
oos | -0.022 |
eks | -0.021 |
ays | -0.021 |
ases | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | +4.14e-03 |
Pos logit contributions | |
Ho | +0.170 |
Rib | +0.165 |
H | +0.143 |
Bub | +0.135 |
Butter | +0.133 |
Neg logit contributions | |
eks | -0.162 |
oos | -0.159 |
agged | -0.155 |
elevator | -0.154 |
cases | -0.154 |
Token | the |
Token ID | 262 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
led | +0.006 |
eks | +0.006 |
elevator | +0.006 |
oos | +0.006 |
cases | +0.005 |
Neg logit contributions | |
Rib | -0.008 |
Ho | -0.007 |
H | -0.007 |
Bun | -0.006 |
Bub | -0.006 |
Token | ham |
Token ID | 8891 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
Ho | +0.031 |
Rib | +0.031 |
H | +0.030 |
Bun | +0.028 |
Bub | +0.027 |
Neg logit contributions | |
elevator | -0.021 |
oos | -0.020 |
eks | -0.020 |
ases | -0.019 |
tickets | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
Ho | +0.349 |
Rib | +0.342 |
H | +0.324 |
Bun | +0.302 |
Bub | +0.299 |
Neg logit contributions | |
oos | -0.247 |
ays | -0.244 |
eks | -0.243 |
ases | -0.241 |
cases | -0.237 |
Token | They |
Token ID | 1119 |
Feature activation | +4.40e-02 |
Pos logit contributions | |
Rib | +0.036 |
Ho | +0.035 |
H | +0.032 |
Bun | +0.032 |
accidentally | +0.031 |
Neg logit contributions | |
ays | -0.016 |
ags | -0.015 |
agged | -0.014 |
oos | -0.014 |
he | -0.014 |
Token | put |
Token ID | 1234 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
eks | +0.072 |
oos | +0.072 |
ays | +0.070 |
cases | +0.070 |
elevator | +0.070 |
Neg logit contributions | |
Ho | -0.073 |
Rib | -0.071 |
H | -0.064 |
Bub | -0.059 |
Bun | -0.058 |
Token | them |
Token ID | 606 |
Feature activation | +4.22e-02 |
Pos logit contributions | |
Ho | +0.055 |
Rib | +0.053 |
H | +0.047 |
Bub | +0.044 |
Bun | +0.043 |
Neg logit contributions | |
eks | -0.052 |
oos | -0.052 |
cases | -0.051 |
elevator | -0.050 |
ases | -0.050 |
Token | on |
Token ID | 319 |
Feature activation | +6.71e-04 |
Pos logit contributions | |
roads | +0.074 |
ays | +0.073 |
eks | +0.072 |
elevator | +0.072 |
ases | +0.072 |
Neg logit contributions | |
Ho | -0.068 |
Rib | -0.062 |
H | -0.054 |
Bub | -0.052 |
Lucky | -0.051 |
Token | out |
Token ID | 503 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
oos | +0.010 |
eks | +0.010 |
ays | +0.010 |
ases | +0.010 |
elevator | +0.010 |
Neg logit contributions | |
Ho | -0.015 |
Rib | -0.015 |
H | -0.014 |
Bun | -0.013 |
Bub | -0.013 |
Token | some |
Token ID | 617 |
Feature activation | +7.25e-02 |
Pos logit contributions | |
eks | +0.044 |
oos | +0.044 |
cases | +0.044 |
agged | +0.044 |
ays | +0.043 |
Neg logit contributions | |
Ho | -0.045 |
Rib | -0.044 |
H | -0.039 |
Bub | -0.036 |
Po | -0.036 |
Token | cheese |
Token ID | 9891 |
Feature activation | -9.87e-02 |
Pos logit contributions | |
eks | +0.101 |
oos | +0.100 |
cases | +0.097 |
ays | +0.097 |
agged | +0.096 |
Neg logit contributions | |
Ho | -0.140 |
Rib | -0.137 |
H | -0.123 |
Bub | -0.116 |
Bun | -0.114 |
Token | and |
Token ID | 290 |
Feature activation | +3.89e-03 |
Pos logit contributions | |
Ho | +0.178 |
Rib | +0.175 |
H | +0.158 |
Bub | +0.146 |
Bun | +0.145 |
Neg logit contributions | |
oos | -0.147 |
eks | -0.147 |
ays | -0.143 |
cases | -0.142 |
elevator | -0.142 |
Token | some |
Token ID | 617 |
Feature activation | +7.28e-02 |
Pos logit contributions | |
led | +0.005 |
eks | +0.005 |
oos | +0.005 |
elevator | +0.005 |
cases | +0.005 |
Neg logit contributions | |
Rib | -0.007 |
Ho | -0.007 |
H | -0.007 |
Bun | -0.006 |
Fay | -0.006 |
Token | ham |
Token ID | 8891 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
oos | +0.101 |
eks | +0.098 |
elevator | +0.098 |
led | +0.098 |
ases | +0.098 |
Neg logit contributions | |
Ho | -0.135 |
Rib | -0.133 |
H | -0.126 |
Bun | -0.115 |
Bub | -0.113 |
Token | . |
Token ID | 13 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
Ho | +0.354 |
Rib | +0.347 |
H | +0.331 |
Bun | +0.310 |
Bub | +0.303 |
Neg logit contributions | |
oos | -0.240 |
ays | -0.237 |
ases | -0.235 |
eks | -0.232 |
cases | -0.230 |
Token | She |
Token ID | 1375 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
Rib | +0.029 |
Ho | +0.028 |
H | +0.025 |
accidentally | +0.024 |
Bun | +0.024 |
Neg logit contributions | |
ays | -0.015 |
oos | -0.014 |
cases | -0.013 |
hed | -0.013 |
he | -0.013 |
Token | put |
Token ID | 1234 |
Feature activation | -3.64e-02 |
Pos logit contributions | |
Ho | +0.027 |
Rib | +0.027 |
H | +0.026 |
Bun | +0.023 |
Butter | +0.023 |
Neg logit contributions | |
led | -0.022 |
eks | -0.022 |
oos | -0.021 |
hed | -0.021 |
elevator | -0.021 |
Token | them |
Token ID | 606 |
Feature activation | +4.01e-02 |
Pos logit contributions | |
Ho | +0.065 |
Rib | +0.065 |
H | +0.060 |
Bun | +0.055 |
Bub | +0.054 |
Neg logit contributions | |
eks | -0.053 |
oos | -0.053 |
ays | -0.052 |
elevator | -0.051 |
ases | -0.050 |
Token | on |
Token ID | 319 |
Feature activation | -3.27e-03 |
Pos logit contributions | |
roads | +0.066 |
elevator | +0.065 |
ases | +0.065 |
cases | +0.065 |
ays | +0.065 |
Neg logit contributions | |
Ho | -0.069 |
Rib | -0.063 |
H | -0.054 |
Bub | -0.054 |
Po | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | +1.30e-03 |
Pos logit contributions | |
eks | +0.055 |
agged | +0.054 |
oos | +0.052 |
hing | +0.051 |
ags | +0.051 |
Neg logit contributions | |
Rib | -0.027 |
Ho | -0.027 |
onna | -0.022 |
accidentally | -0.022 |
Po | -0.021 |
Token | It |
Token ID | 632 |
Feature activation | -4.89e-04 |
Pos logit contributions | |
ays | +0.002 |
oos | +0.002 |
elevator | +0.002 |
ases | +0.002 |
eks | +0.002 |
Neg logit contributions | |
Ho | -0.002 |
Rib | -0.002 |
H | -0.002 |
Bun | -0.002 |
Bub | -0.002 |
Token | has |
Token ID | 468 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
Ho | +0.001 |
Rib | +0.001 |
H | +0.001 |
Bub | +0.001 |
Butter | +0.001 |
Neg logit contributions | |
oos | -0.001 |
eks | -0.001 |
agged | -0.001 |
ases | -0.001 |
cases | -0.001 |
Token | cheese |
Token ID | 9891 |
Feature activation | -9.75e-02 |
Pos logit contributions | |
Ho | +0.053 |
Rib | +0.052 |
H | +0.046 |
Bub | +0.043 |
Bun | +0.043 |
Neg logit contributions | |
eks | -0.044 |
oos | -0.044 |
ays | -0.042 |
cases | -0.042 |
ases | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
Ho | +0.167 |
Rib | +0.158 |
H | +0.136 |
Lucky | +0.131 |
Po | +0.129 |
Neg logit contributions | |
eks | -0.165 |
oos | -0.162 |
agged | -0.156 |
elevator | -0.155 |
ases | -0.153 |
Token | ham |
Token ID | 8891 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
oos | +0.017 |
eks | +0.017 |
elevator | +0.016 |
ays | +0.016 |
cases | +0.016 |
Neg logit contributions | |
Ho | -0.017 |
Rib | -0.016 |
H | -0.015 |
Bun | -0.014 |
Bub | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
Ho | +0.339 |
Rib | +0.332 |
H | +0.300 |
Bub | +0.280 |
Bun | +0.277 |
Neg logit contributions | |
eks | -0.260 |
oos | -0.260 |
ays | -0.251 |
elevator | -0.251 |
cases | -0.251 |
Token | lettuce |
Token ID | 39406 |
Feature activation | -9.48e-02 |
Pos logit contributions | |
elevator | +0.017 |
eks | +0.017 |
oos | +0.016 |
cases | +0.016 |
ays | +0.016 |
Neg logit contributions | |
Ho | -0.019 |
Rib | -0.019 |
H | -0.018 |
Bun | -0.016 |
Bub | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +9.82e-04 |
Pos logit contributions | |
Ho | +0.163 |
Rib | +0.158 |
H | +0.137 |
Bub | +0.128 |
Lucky | +0.127 |
Neg logit contributions | |
oos | -0.154 |
eks | -0.154 |
elevator | -0.150 |
agged | -0.148 |
cases | -0.146 |
Token | It |
Token ID | 632 |
Feature activation | -1.72e-04 |
Pos logit contributions | |
ays | +0.002 |
oos | +0.002 |
elevator | +0.002 |
ases | +0.002 |
agged | +0.002 |
Neg logit contributions | |
Ho | -0.001 |
Rib | -0.001 |
H | -0.001 |
Bun | -0.001 |
Ros | -0.001 |
Token | looks |
Token ID | 3073 |
Feature activation | +3.52e-02 |
Pos logit contributions | |
Ho | +0.000 |
Rib | +0.000 |
H | +0.000 |
Bub | +0.000 |
Bun | +0.000 |
Neg logit contributions | |
oos | -0.000 |
eks | -0.000 |
ays | -0.000 |
cases | -0.000 |
elevator | -0.000 |
Token | Do |
Token ID | 2141 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
accidentally | +0.035 |
stub | +0.031 |
Ho | +0.031 |
Rib | +0.030 |
Bun | +0.030 |
Neg logit contributions | |
ases | -0.026 |
ays | -0.024 |
icks | -0.024 |
oops | -0.024 |
are | -0.023 |
Token | you |
Token ID | 345 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
Ho | +0.223 |
Rib | +0.222 |
accidentally | +0.210 |
H | +0.206 |
Bun | +0.202 |
Neg logit contributions | |
eks | -0.140 |
are | -0.136 |
tickets | -0.126 |
elevator | -0.123 |
ases | -0.122 |
Token | like |
Token ID | 588 |
Feature activation | -7.24e-03 |
Pos logit contributions | |
Ho | +0.044 |
Rib | +0.043 |
H | +0.040 |
Bun | +0.038 |
Bub | +0.037 |
Neg logit contributions | |
eks | -0.021 |
ases | -0.020 |
oos | -0.019 |
he | -0.019 |
agged | -0.019 |
Token | cheese |
Token ID | 9891 |
Feature activation | -9.87e-02 |
Pos logit contributions | |
Ho | +0.012 |
H | +0.012 |
Rib | +0.012 |
Bun | +0.011 |
Cle | +0.011 |
Neg logit contributions | |
ases | -0.010 |
ays | -0.010 |
cks | -0.009 |
tickets | -0.009 |
eks | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | +1.26e-03 |
Pos logit contributions | |
Ho | +0.206 |
Rib | +0.203 |
H | +0.190 |
Bub | +0.179 |
Bun | +0.178 |
Neg logit contributions | |
ays | -0.112 |
ases | -0.111 |
oos | -0.111 |
he | -0.110 |
eks | -0.109 |
Token | ham |
Token ID | 8891 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
are | +0.001 |
bags | +0.001 |
tickets | +0.001 |
eks | +0.001 |
candles | +0.001 |
Neg logit contributions | |
H | -0.002 |
Ho | -0.002 |
Rib | -0.002 |
Bun | -0.002 |
Ros | -0.002 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
Ho | +0.356 |
Rib | +0.353 |
H | +0.339 |
Bun | +0.326 |
Bub | +0.319 |
Neg logit contributions | |
ays | -0.231 |
ases | -0.225 |
oos | -0.215 |
eks | -0.212 |
agged | -0.211 |
Token |
|
Token ID | 198 |
Feature activation | -3.17e-02 |
Pos logit contributions | |
die | +0.025 |
accidentally | +0.025 |
Ho | +0.024 |
ccoli | +0.024 |
Rib | +0.024 |
Neg logit contributions | |
he | -0.017 |
ases | -0.016 |
ays | -0.015 |
oos | -0.015 |
eks | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -3.19e-02 |
Pos logit contributions | |
accidentally | +0.052 |
Rib | +0.052 |
Ho | +0.051 |
Butter | +0.048 |
Tooth | +0.047 |
Neg logit contributions | |
he | -0.044 |
ases | -0.042 |
eks | -0.041 |
oos | -0.039 |
ays | -0.039 |
Token | L |
Token ID | 43 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
Ho | +0.046 |
Rib | +0.045 |
accidentally | +0.043 |
Benny | +0.042 |
Butter | +0.041 |
Neg logit contributions | |
ases | -0.052 |
eks | -0.051 |
he | -0.050 |
oos | -0.048 |
out | -0.047 |
Token | ila |
Token ID | 10102 |
Feature activation | -5.00e-02 |
Pos logit contributions | |
Ho | +0.033 |
Rib | +0.032 |
H | +0.029 |
Bun | +0.028 |
Bub | +0.027 |
Neg logit contributions | |
eks | -0.024 |
oos | -0.024 |
ases | -0.024 |
ays | -0.024 |
agged | -0.023 |
Token | do |
Token ID | 466 |
Feature activation | -5.16e-02 |
Pos logit contributions | |
Ho | +0.067 |
Rib | +0.066 |
H | +0.060 |
Bun | +0.057 |
Bub | +0.055 |
Neg logit contributions | |
eks | -0.047 |
oos | -0.045 |
he | -0.045 |
ases | -0.045 |
out | -0.044 |
Token | not |
Token ID | 407 |
Feature activation | -7.13e-03 |
Pos logit contributions | |
Ho | +0.087 |
Rib | +0.085 |
H | +0.077 |
Bun | +0.073 |
Bub | +0.071 |
Neg logit contributions | |
eks | -0.081 |
ases | -0.081 |
elevator | -0.079 |
oos | -0.079 |
out | -0.078 |
Token | like |
Token ID | 588 |
Feature activation | -3.61e-03 |
Pos logit contributions | |
Ho | +0.017 |
Rib | +0.016 |
H | +0.015 |
Bun | +0.014 |
Bub | +0.014 |
Neg logit contributions | |
eks | -0.007 |
ases | -0.007 |
he | -0.007 |
oos | -0.006 |
elevator | -0.006 |
Token | cheese |
Token ID | 9891 |
Feature activation | -9.75e-02 |
Pos logit contributions | |
Ho | +0.006 |
Rib | +0.006 |
H | +0.006 |
Bun | +0.005 |
Bub | +0.005 |
Neg logit contributions | |
ays | -0.006 |
oos | -0.006 |
elevator | -0.006 |
ases | -0.005 |
eks | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +6.73e-03 |
Pos logit contributions | |
Ho | +0.186 |
Rib | +0.183 |
H | +0.169 |
Bub | +0.158 |
Bun | +0.156 |
Neg logit contributions | |
oos | -0.131 |
ays | -0.129 |
ases | -0.129 |
eks | -0.128 |
elevator | -0.127 |
Token | ham |
Token ID | 8891 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
ases | +0.010 |
elevator | +0.010 |
eks | +0.010 |
oos | +0.009 |
ays | +0.009 |
Neg logit contributions | |
Ho | -0.011 |
Rib | -0.011 |
H | -0.011 |
Bun | -0.010 |
Bub | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
Ho | +0.342 |
Rib | +0.336 |
H | +0.314 |
Bun | +0.296 |
Bub | +0.293 |
Neg logit contributions | |
oos | -0.248 |
ays | -0.248 |
ases | -0.247 |
eks | -0.244 |
agged | -0.236 |
Token | They |
Token ID | 1119 |
Feature activation | +4.37e-02 |
Pos logit contributions | |
Ho | +0.020 |
Rib | +0.020 |
accidentally | +0.019 |
H | +0.019 |
Bun | +0.019 |
Neg logit contributions | |
ays | -0.016 |
ases | -0.016 |
icks | -0.015 |
oos | -0.015 |
oops | -0.014 |
Token | are |
Token ID | 389 |
Feature activation | +5.34e-02 |
Pos logit contributions | |
oos | +0.040 |
agged | +0.037 |
ays | +0.037 |
cases | +0.036 |
eks | +0.035 |
Neg logit contributions | |
Ho | -0.109 |
Rib | -0.106 |
H | -0.097 |
Bub | -0.094 |
Butter | -0.093 |
Token | too |
Token ID | 1165 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
ps | +0.064 |
cks | +0.064 |
elevator | +0.063 |
eks | +0.061 |
ases | +0.061 |
Neg logit contributions | |
Rib | -0.103 |
Ho | -0.100 |
H | -0.098 |
Bun | -0.094 |
Ros | -0.091 |
Token | salty |
Token ID | 36021 |
Feature activation | +7.21e-03 |
Pos logit contributions | |
Ho | +0.051 |
Rib | +0.048 |
H | +0.043 |
accidentally | +0.039 |
Butter | +0.039 |
Neg logit contributions | |
eks | -0.056 |
oos | -0.056 |
ays | -0.053 |
cases | -0.053 |
ases | -0.053 |
Token | my |
Token ID | 1820 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
ays | +0.095 |
he | +0.094 |
eks | +0.094 |
oos | +0.093 |
ases | +0.091 |
Neg logit contributions | |
Ho | -0.113 |
Rib | -0.110 |
H | -0.097 |
Bub | -0.093 |
Bun | -0.090 |
Token | hurts |
Token ID | 20406 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
Rib | +0.071 |
Ho | +0.071 |
H | +0.064 |
Bub | +0.061 |
Po | +0.059 |
Neg logit contributions | |
eks | -0.049 |
cases | -0.048 |
led | -0.048 |
oos | -0.048 |
ays | -0.047 |
Token | from |
Token ID | 422 |
Feature activation | -9.26e-03 |
Pos logit contributions | |
Ho | +0.040 |
Rib | +0.039 |
H | +0.036 |
Bun | +0.033 |
Bub | +0.033 |
Neg logit contributions | |
eks | -0.034 |
oos | -0.034 |
ays | -0.033 |
cases | -0.033 |
agged | -0.033 |
Token | the |
Token ID | 262 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
Rib | +0.017 |
Ho | +0.016 |
H | +0.014 |
Ros | +0.014 |
Bub | +0.014 |
Neg logit contributions | |
elevator | -0.014 |
eks | -0.013 |
oos | -0.013 |
ags | -0.013 |
ases | -0.013 |
Token | bad |
Token ID | 2089 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
Rib | +0.031 |
Ho | +0.031 |
H | +0.028 |
Bub | +0.027 |
Bun | +0.026 |
Neg logit contributions | |
elevator | -0.022 |
oos | -0.021 |
eks | -0.021 |
he | -0.021 |
cases | -0.020 |
Token | ham |
Token ID | 8891 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
Ho | +0.025 |
Rib | +0.024 |
H | +0.022 |
Bun | +0.020 |
Lucky | +0.020 |
Neg logit contributions | |
eks | -0.024 |
oos | -0.024 |
cases | -0.024 |
ays | -0.024 |
agged | -0.024 |
Token | ," |
Token ID | 553 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
Ho | +0.340 |
Rib | +0.336 |
H | +0.309 |
Bub | +0.290 |
Bun | +0.285 |
Neg logit contributions | |
oos | -0.253 |
eks | -0.252 |
ays | -0.249 |
cases | -0.247 |
elevator | -0.246 |
Token | Lily |
Token ID | 20037 |
Feature activation | -5.17e-02 |
Pos logit contributions | |
Ho | +0.049 |
Rib | +0.048 |
Bun | +0.040 |
H | +0.040 |
onna | +0.040 |
Neg logit contributions | |
elevator | -0.047 |
eks | -0.044 |
ases | -0.041 |
oos | -0.040 |
are | -0.040 |
Token | said |
Token ID | 531 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
Ho | +0.091 |
Rib | +0.091 |
H | +0.079 |
Bub | +0.077 |
Butter | +0.075 |
Neg logit contributions | |
cases | -0.078 |
led | -0.075 |
eks | -0.075 |
elevator | -0.073 |
hed | -0.073 |
Token | , |
Token ID | 11 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
Ho | +0.055 |
Rib | +0.054 |
H | +0.049 |
Bub | +0.048 |
Butter | +0.046 |
Neg logit contributions | |
led | -0.028 |
oos | -0.028 |
cases | -0.028 |
agged | -0.028 |
out | -0.027 |
Token | rubbing |
Token ID | 27714 |
Feature activation | -9.51e-02 |
Pos logit contributions | |
Ho | +0.023 |
Rib | +0.023 |
H | +0.020 |
Bub | +0.019 |
Bun | +0.019 |
Neg logit contributions | |
ps | -0.019 |
elevator | -0.019 |
eks | -0.018 |
led | -0.018 |
agged | -0.018 |
Token | had |
Token ID | 550 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
Ho | +0.016 |
Rib | +0.016 |
H | +0.014 |
Bun | +0.013 |
Bub | +0.013 |
Neg logit contributions | |
eks | -0.015 |
elevator | -0.015 |
oos | -0.015 |
cases | -0.015 |
led | -0.015 |
Token | already |
Token ID | 1541 |
Feature activation | -4.38e-02 |
Pos logit contributions | |
Ho | +0.044 |
Rib | +0.043 |
H | +0.038 |
Bub | +0.035 |
Bun | +0.035 |
Neg logit contributions | |
eks | -0.049 |
oos | -0.049 |
ays | -0.048 |
cases | -0.048 |
elevator | -0.048 |
Token | put |
Token ID | 1234 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
Ho | +0.068 |
Rib | +0.066 |
H | +0.060 |
accidentally | +0.059 |
Bub | +0.053 |
Neg logit contributions | |
ays | -0.074 |
cases | -0.073 |
oos | -0.073 |
eks | -0.070 |
agged | -0.070 |
Token | cheese |
Token ID | 9891 |
Feature activation | -9.89e-02 |
Pos logit contributions | |
Ho | +0.060 |
Rib | +0.057 |
H | +0.052 |
Bub | +0.049 |
Butter | +0.049 |
Neg logit contributions | |
eks | -0.044 |
oos | -0.043 |
cases | -0.043 |
ases | -0.043 |
ays | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
Ho | +0.180 |
Rib | +0.174 |
H | +0.155 |
Bub | +0.146 |
Bun | +0.145 |
Neg logit contributions | |
eks | -0.149 |
oos | -0.147 |
elevator | -0.143 |
ays | -0.143 |
ases | -0.142 |
Token | ham |
Token ID | 8891 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
elevator | +0.015 |
eks | +0.015 |
oos | +0.015 |
cases | +0.014 |
led | +0.014 |
Neg logit contributions | |
Ho | -0.020 |
Rib | -0.020 |
H | -0.019 |
Bun | -0.018 |
Bub | -0.017 |
Token | in |
Token ID | 287 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
Ho | +0.349 |
Rib | +0.344 |
H | +0.318 |
Bun | +0.296 |
Bub | +0.295 |
Neg logit contributions | |
oos | -0.245 |
eks | -0.244 |
ays | -0.240 |
cases | -0.237 |
ases | -0.237 |
Token | some |
Token ID | 617 |
Feature activation | +7.61e-02 |
Pos logit contributions | |
Ho | +0.032 |
Rib | +0.031 |
H | +0.028 |
Bub | +0.025 |
Bun | +0.025 |
Neg logit contributions | |
eks | -0.033 |
oos | -0.033 |
ays | -0.032 |
cases | -0.032 |
elevator | -0.032 |
Token | of |
Token ID | 286 |
Feature activation | -4.26e-02 |
Pos logit contributions | |
eks | +0.109 |
oos | +0.108 |
ays | +0.105 |
cases | +0.105 |
ases | +0.104 |
Neg logit contributions | |
Ho | -0.143 |
Rib | -0.140 |
H | -0.126 |
Bub | -0.118 |
Bun | -0.117 |
Token | the |
Token ID | 262 |
Feature activation | -8.87e-03 |
Pos logit contributions | |
Ho | +0.091 |
Rib | +0.090 |
H | +0.083 |
Bun | +0.078 |
Bub | +0.076 |
Neg logit contributions | |
eks | -0.047 |
oos | -0.047 |
elevator | -0.047 |
ays | -0.045 |
ases | -0.045 |
Token | eggs |
Token ID | 9653 |
Feature activation | -5.60e-02 |
Pos logit contributions | |
Ho | +0.019 |
Rib | +0.019 |
H | +0.018 |
Bub | +0.016 |
Bun | +0.016 |
Neg logit contributions | |
eks | -0.010 |
oos | -0.010 |
ays | -0.010 |
cases | -0.009 |
elevator | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
eks | +0.071 |
oos | +0.070 |
agged | +0.068 |
ays | +0.066 |
ases | +0.066 |
Neg logit contributions | |
Ho | -0.049 |
Rib | -0.046 |
H | -0.039 |
Lucky | -0.038 |
Bub | -0.036 |
Token | They |
Token ID | 1119 |
Feature activation | +4.61e-02 |
Pos logit contributions | |
Rib | +0.021 |
Ho | +0.020 |
H | +0.018 |
Bun | +0.018 |
Bub | +0.017 |
Neg logit contributions | |
ays | -0.017 |
he | -0.016 |
oos | -0.016 |
ases | -0.016 |
agged | -0.016 |
Token | were |
Token ID | 547 |
Feature activation | +5.28e-02 |
Pos logit contributions | |
oos | +0.078 |
eks | +0.076 |
ays | +0.076 |
agged | +0.075 |
ases | +0.074 |
Neg logit contributions | |
Ho | -0.077 |
Rib | -0.074 |
H | -0.066 |
Bub | -0.061 |
Bun | -0.061 |
Token | cheese |
Token ID | 9891 |
Feature activation | -9.85e-02 |
Pos logit contributions | |
eks | +0.072 |
elevator | +0.072 |
agged | +0.071 |
ags | +0.070 |
led | +0.069 |
Neg logit contributions | |
Rib | -0.098 |
Ho | -0.098 |
H | -0.091 |
Bun | -0.085 |
Bub | -0.083 |
Token | and |
Token ID | 290 |
Feature activation | +5.60e-03 |
Pos logit contributions | |
Ho | +0.169 |
Rib | +0.161 |
H | +0.137 |
Lucky | +0.133 |
Bun | +0.130 |
Neg logit contributions | |
eks | -0.169 |
oos | -0.166 |
agged | -0.159 |
ays | -0.159 |
elevator | -0.159 |
Token | ham |
Token ID | 8891 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
elevator | +0.008 |
eks | +0.008 |
oos | +0.008 |
led | +0.007 |
cases | +0.007 |
Neg logit contributions | |
Ho | -0.010 |
Rib | -0.010 |
H | -0.009 |
Bun | -0.008 |
Bub | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | +6.67e-03 |
Pos logit contributions | |
Ho | +0.341 |
Rib | +0.335 |
H | +0.306 |
Bub | +0.285 |
Bun | +0.283 |
Neg logit contributions | |
oos | -0.254 |
eks | -0.253 |
ays | -0.247 |
cases | -0.246 |
ases | -0.245 |
Token | lettuce |
Token ID | 39406 |
Feature activation | -9.48e-02 |
Pos logit contributions | |
elevator | +0.009 |
eks | +0.009 |
ases | +0.009 |
led | +0.009 |
cases | +0.009 |
Neg logit contributions | |
Ho | -0.012 |
Rib | -0.012 |
H | -0.011 |
Bun | -0.010 |
Bub | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
Ho | +0.168 |
Rib | +0.163 |
H | +0.144 |
Bub | +0.135 |
Bun | +0.133 |
Neg logit contributions | |
eks | -0.149 |
oos | -0.148 |
elevator | -0.143 |
agged | -0.142 |
ays | -0.142 |
Token | They |
Token ID | 1119 |
Feature activation | +4.61e-02 |
Pos logit contributions | |
Rib | +0.024 |
Ho | +0.024 |
H | +0.021 |
Bun | +0.021 |
accidentally | +0.020 |
Neg logit contributions | |
ays | -0.017 |
he | -0.016 |
ases | -0.015 |
ags | -0.015 |
agged | -0.015 |
Token | were |
Token ID | 547 |
Feature activation | +5.40e-02 |
Pos logit contributions | |
oos | +0.072 |
eks | +0.072 |
ays | +0.070 |
elevator | +0.070 |
cases | +0.070 |
Neg logit contributions | |
Ho | -0.080 |
Rib | -0.078 |
H | -0.070 |
Bub | -0.065 |
Bun | -0.064 |
Token | . |
Token ID | 13 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
Ho | +0.054 |
Rib | +0.054 |
H | +0.049 |
Bub | +0.045 |
Bun | +0.045 |
Neg logit contributions | |
oos | -0.040 |
eks | -0.039 |
ays | -0.039 |
ases | -0.039 |
cases | -0.039 |
Token | It |
Token ID | 632 |
Feature activation | -2.51e-03 |
Pos logit contributions | |
accidentally | +0.032 |
Ho | +0.029 |
Rib | +0.028 |
stub | +0.028 |
Bun | +0.028 |
Neg logit contributions | |
ases | -0.024 |
ays | -0.022 |
oops | -0.022 |
are | -0.022 |
ags | -0.022 |
Token | is |
Token ID | 318 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
Ho | +0.005 |
Rib | +0.005 |
H | +0.005 |
Bun | +0.005 |
Bub | +0.005 |
Neg logit contributions | |
eks | -0.003 |
ays | -0.003 |
elevator | -0.003 |
ases | -0.003 |
oos | -0.003 |
Token | cheese |
Token ID | 9891 |
Feature activation | -9.86e-02 |
Pos logit contributions | |
cks | +0.017 |
ags | +0.016 |
ps | +0.015 |
candles | +0.015 |
elevator | +0.015 |
Neg logit contributions | |
Rib | -0.032 |
H | -0.031 |
Ho | -0.030 |
Bun | -0.030 |
Ros | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | +8.55e-04 |
Pos logit contributions | |
Ho | +0.179 |
Rib | +0.175 |
H | +0.158 |
Bub | +0.146 |
Bun | +0.145 |
Neg logit contributions | |
eks | -0.147 |
oos | -0.147 |
ays | -0.142 |
elevator | -0.142 |
cases | -0.141 |
Token | ham |
Token ID | 8891 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
elevator | +0.001 |
eks | +0.001 |
are | +0.001 |
bags | +0.001 |
ases | +0.001 |
Neg logit contributions | |
Ho | -0.001 |
Rib | -0.001 |
H | -0.001 |
Bun | -0.001 |
Bub | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
Ho | +0.340 |
Rib | +0.336 |
H | +0.318 |
Bun | +0.299 |
Bub | +0.295 |
Neg logit contributions | |
ases | -0.243 |
ays | -0.243 |
oos | -0.240 |
eks | -0.236 |
cases | -0.232 |
Token | Do |
Token ID | 2141 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
accidentally | +0.035 |
stub | +0.031 |
Ho | +0.031 |
Rib | +0.030 |
Bun | +0.030 |
Neg logit contributions | |
ases | -0.026 |
ays | -0.024 |
icks | -0.024 |
oops | -0.024 |
are | -0.023 |
Token | you |
Token ID | 345 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
Ho | +0.223 |
Rib | +0.222 |
accidentally | +0.210 |
H | +0.206 |
Bun | +0.202 |
Neg logit contributions | |
eks | -0.140 |
are | -0.136 |
tickets | -0.126 |
elevator | -0.123 |
ases | -0.122 |
Token | like |
Token ID | 588 |
Feature activation | -7.24e-03 |
Pos logit contributions | |
Ho | +0.044 |
Rib | +0.043 |
H | +0.040 |
Bun | +0.038 |
Bub | +0.037 |
Neg logit contributions | |
eks | -0.021 |
ases | -0.020 |
oos | -0.019 |
he | -0.019 |
agged | -0.019 |
Token | cheese |
Token ID | 9891 |
Feature activation | -9.87e-02 |
Pos logit contributions | |
Ho | +0.012 |
H | +0.012 |
Rib | +0.012 |
Bun | +0.011 |
Cle | +0.011 |
Neg logit contributions | |
ases | -0.010 |
ays | -0.010 |
cks | -0.009 |
tickets | -0.009 |
eks | -0.009 |
Token | watching |
Token ID | 4964 |
Feature activation | -4.02e-02 |
Pos logit contributions | |
Ho | +0.065 |
Rib | +0.064 |
H | +0.058 |
Bub | +0.054 |
Bun | +0.053 |
Neg logit contributions | |
eks | -0.047 |
oos | -0.046 |
elevator | -0.046 |
led | -0.045 |
ases | -0.044 |
Token | her |
Token ID | 607 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
Rib | +0.073 |
Ho | +0.073 |
H | +0.065 |
accidentally | +0.062 |
Bub | +0.060 |
Neg logit contributions | |
oos | -0.055 |
elevator | -0.052 |
eks | -0.051 |
ases | -0.051 |
cases | -0.050 |
Token | eat |
Token ID | 4483 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
Ho | +0.033 |
Rib | +0.032 |
H | +0.029 |
accidentally | +0.028 |
Bub | +0.027 |
Neg logit contributions | |
oos | -0.022 |
eks | -0.022 |
cases | -0.022 |
ases | -0.022 |
he | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
Ho | +0.030 |
Rib | +0.030 |
H | +0.027 |
Bub | +0.025 |
Bun | +0.025 |
Neg logit contributions | |
oos | -0.023 |
eks | -0.023 |
elevator | -0.023 |
ays | -0.022 |
ases | -0.022 |
Token | pine |
Token ID | 20161 |
Feature activation | -5.51e-02 |
Pos logit contributions | |
Ho | +0.020 |
Rib | +0.020 |
H | +0.018 |
Bub | +0.016 |
Bun | +0.016 |
Neg logit contributions | |
oos | -0.015 |
elevator | -0.015 |
eks | -0.015 |
cases | -0.015 |
he | -0.015 |
Token | con |
Token ID | 1102 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
Ho | +0.106 |
Rib | +0.105 |
H | +0.095 |
Bub | +0.090 |
accidentally | +0.086 |
Neg logit contributions | |
oos | -0.074 |
elevator | -0.074 |
cases | -0.073 |
he | -0.073 |
eks | -0.071 |
Token | es |
Token ID | 274 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
Ho | +0.383 |
Rib | +0.376 |
H | +0.345 |
Bub | +0.324 |
Bun | +0.320 |
Neg logit contributions | |
oos | -0.214 |
eks | -0.212 |
cases | -0.207 |
ays | -0.205 |
elevator | -0.204 |
Token | . |
Token ID | 13 |
Feature activation | -9.97e-03 |
Pos logit contributions | |
Ho | +0.019 |
Rib | +0.019 |
H | +0.017 |
Bun | +0.016 |
Bub | +0.016 |
Neg logit contributions | |
eks | -0.018 |
oos | -0.018 |
agged | -0.018 |
ays | -0.018 |
ases | -0.018 |
Token | The |
Token ID | 383 |
Feature activation | -5.96e-03 |
Pos logit contributions | |
Rib | +0.019 |
Ho | +0.018 |
please | +0.016 |
Stri | +0.016 |
H | +0.015 |
Neg logit contributions | |
cases | -0.011 |
oos | -0.010 |
he | -0.010 |
hed | -0.010 |
elevator | -0.009 |
Token | squirrel |
Token ID | 33039 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
Rib | +0.009 |
Ho | +0.009 |
onna | +0.008 |
accidentally | +0.007 |
H | +0.007 |
Neg logit contributions | |
elevator | -0.010 |
eks | -0.010 |
he | -0.010 |
oos | -0.009 |
agged | -0.009 |
Token | said |
Token ID | 531 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
Ho | +0.045 |
Rib | +0.044 |
H | +0.039 |
Bub | +0.035 |
Butter | +0.035 |
Neg logit contributions | |
eks | -0.044 |
oos | -0.043 |
cases | -0.043 |
led | -0.042 |
elevator | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | +3.99e-03 |
Pos logit contributions | |
Ho | +0.095 |
Rib | +0.092 |
H | +0.081 |
Bub | +0.076 |
Butter | +0.074 |
Neg logit contributions | |
eks | -0.097 |
oos | -0.096 |
agged | -0.094 |
elevator | -0.092 |
ases | -0.092 |
Token | the |
Token ID | 262 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
eks | +0.006 |
elevator | +0.005 |
oos | +0.005 |
ases | +0.005 |
cases | +0.005 |
Neg logit contributions | |
Rib | -0.007 |
Ho | -0.007 |
H | -0.007 |
Bun | -0.006 |
Bub | -0.006 |
Token | cheese |
Token ID | 9891 |
Feature activation | -9.79e-02 |
Pos logit contributions | |
Ho | +0.027 |
Rib | +0.026 |
H | +0.026 |
Bun | +0.024 |
Bub | +0.024 |
Neg logit contributions | |
elevator | -0.016 |
oos | -0.016 |
eks | -0.016 |
tickets | -0.015 |
ases | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | +5.27e-03 |
Pos logit contributions | |
Ho | +0.156 |
Rib | +0.149 |
H | +0.126 |
accidentally | +0.120 |
Lucky | +0.120 |
Neg logit contributions | |
eks | -0.170 |
oos | -0.167 |
agged | -0.166 |
ases | -0.159 |
ays | -0.158 |
Token | the |
Token ID | 262 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
eks | +0.008 |
elevator | +0.007 |
oos | +0.007 |
ases | +0.007 |
ays | +0.007 |
Neg logit contributions | |
Ho | -0.009 |
Rib | -0.009 |
H | -0.009 |
Bun | -0.008 |
Bub | -0.008 |
Token | ham |
Token ID | 8891 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
Rib | +0.025 |
Ho | +0.025 |
H | +0.024 |
Bun | +0.022 |
Bub | +0.022 |
Neg logit contributions | |
elevator | -0.016 |
oos | -0.015 |
eks | -0.015 |
tickets | -0.015 |
ases | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
Ho | +0.334 |
Rib | +0.327 |
H | +0.297 |
Bub | +0.276 |
Bun | +0.274 |
Neg logit contributions | |
eks | -0.262 |
oos | -0.262 |
ays | -0.255 |
cases | -0.253 |
elevator | -0.253 |
Token | Can |
Token ID | 1680 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
Ho | +0.022 |
Rib | +0.022 |
accidentally | +0.021 |
Bun | +0.021 |
H | +0.020 |
Neg logit contributions | |
ays | -0.016 |
ases | -0.016 |
ags | -0.015 |
are | -0.014 |
oos | -0.014 |
Token | you |
Token ID | 345 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
Ho | +0.178 |
Rib | +0.177 |
H | +0.161 |
Bub | +0.150 |
Bun | +0.149 |
Neg logit contributions | |
elevator | -0.141 |
eks | -0.136 |
ays | -0.135 |
oos | -0.133 |
cases | -0.132 |
Token | help |
Token ID | 1037 |
Feature activation | -6.46e-03 |
Pos logit contributions | |
Ho | +0.036 |
Rib | +0.036 |
H | +0.032 |
Bun | +0.030 |
Bub | +0.029 |
Neg logit contributions | |
eks | -0.025 |
he | -0.024 |
oos | -0.023 |
elevator | -0.023 |
ases | -0.023 |
Token | me |
Token ID | 502 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
Rib | +0.012 |
Ho | +0.012 |
H | +0.011 |
Bub | +0.010 |
Bun | +0.010 |
Neg logit contributions | |
eks | -0.009 |
oos | -0.009 |
ases | -0.009 |
he | -0.009 |
elevator | -0.009 |
Token | light |
Token ID | 1657 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
elevator | +0.082 |
oos | +0.077 |
he | +0.076 |
eks | +0.076 |
ays | +0.074 |
Neg logit contributions | |
Ho | -0.124 |
Rib | -0.124 |
H | -0.109 |
Bub | -0.102 |
Stri | -0.102 |
Token | way |
Token ID | 835 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
Ho | +0.034 |
Rib | +0.033 |
H | +0.030 |
Bun | +0.028 |
Bub | +0.028 |
Neg logit contributions | |
eks | -0.028 |
oos | -0.028 |
ases | -0.027 |
agged | -0.027 |
cases | -0.027 |
Token | off |
Token ID | 572 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
eks | +0.021 |
oos | +0.021 |
ases | +0.020 |
agged | +0.020 |
cases | +0.020 |
Neg logit contributions | |
Ho | -0.022 |
Rib | -0.021 |
H | -0.020 |
Bub | -0.018 |
Bun | -0.018 |
Token | in |
Token ID | 287 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
eks | +0.028 |
oos | +0.027 |
cases | +0.027 |
ases | +0.027 |
ays | +0.027 |
Neg logit contributions | |
Ho | -0.021 |
Rib | -0.020 |
H | -0.018 |
Bub | -0.017 |
Bun | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | -1.31e-03 |
Pos logit contributions | |
Ho | +0.017 |
Rib | +0.016 |
H | +0.015 |
Bub | +0.013 |
Po | +0.013 |
Neg logit contributions | |
cases | -0.023 |
agged | -0.022 |
oos | -0.022 |
led | -0.021 |
ays | -0.021 |
Token | distance |
Token ID | 5253 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
Ho | +0.002 |
Rib | +0.002 |
H | +0.002 |
Bun | +0.002 |
Bub | +0.002 |
Neg logit contributions | |
ays | -0.002 |
eks | -0.002 |
oos | -0.002 |
ases | -0.002 |
cases | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +8.79e-03 |
Pos logit contributions | |
oos | +0.141 |
eks | +0.138 |
elevator | +0.137 |
ays | +0.136 |
cases | +0.136 |
Neg logit contributions | |
Ho | -0.202 |
Rib | -0.198 |
H | -0.178 |
Bub | -0.167 |
Bun | -0.166 |
Token | He |
Token ID | 679 |
Feature activation | +3.20e-02 |
Pos logit contributions | |
oos | +0.013 |
elevator | +0.013 |
cases | +0.013 |
agged | +0.012 |
eks | +0.012 |
Neg logit contributions | |
Rib | -0.016 |
Ho | -0.016 |
H | -0.013 |
Bub | -0.012 |
Lucky | -0.012 |
Token | ran |
Token ID | 4966 |
Feature activation | +7.44e-02 |
Pos logit contributions | |
eks | +0.042 |
led | +0.041 |
oos | +0.039 |
elevator | +0.037 |
out | +0.037 |
Neg logit contributions | |
Rib | -0.064 |
Ho | -0.063 |
H | -0.055 |
onna | -0.054 |
Butter | -0.054 |
Token | over |
Token ID | 625 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
ays | +0.120 |
oos | +0.119 |
eks | +0.118 |
ases | +0.117 |
agged | +0.115 |
Neg logit contributions | |
Ho | -0.129 |
Rib | -0.124 |
H | -0.115 |
Bub | -0.108 |
Bun | -0.107 |
Token | to |
Token ID | 284 |
Feature activation | -5.35e-03 |
Pos logit contributions | |
oos | +0.073 |
eks | +0.073 |
elevator | +0.071 |
ases | +0.070 |
ays | +0.069 |
Neg logit contributions | |
Ho | -0.060 |
Rib | -0.059 |
H | -0.051 |
Bub | -0.047 |
Bun | -0.046 |
Token | were |
Token ID | 547 |
Feature activation | +5.31e-02 |
Pos logit contributions | |
Ho | +0.090 |
Rib | +0.087 |
H | +0.079 |
Bub | +0.075 |
Bun | +0.075 |
Neg logit contributions | |
ays | -0.073 |
oos | -0.073 |
eks | -0.072 |
ases | -0.071 |
elevator | -0.070 |
Token | ready |
Token ID | 3492 |
Feature activation | +6.86e-04 |
Pos logit contributions | |
oos | +0.088 |
eks | +0.088 |
ays | +0.086 |
cases | +0.085 |
ases | +0.085 |
Neg logit contributions | |
Ho | -0.088 |
Rib | -0.086 |
H | -0.076 |
Bub | -0.070 |
Bun | -0.070 |
Token | for |
Token ID | 329 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
eks | +0.001 |
oos | +0.001 |
ays | +0.001 |
ases | +0.001 |
elevator | +0.001 |
Neg logit contributions | |
Ho | -0.001 |
Rib | -0.001 |
H | -0.001 |
Bub | -0.001 |
Bun | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
eks | +0.037 |
oos | +0.037 |
elevator | +0.036 |
ays | +0.035 |
cases | +0.035 |
Neg logit contributions | |
Ho | -0.047 |
Rib | -0.047 |
H | -0.042 |
Bub | -0.039 |
Bun | -0.039 |
Token | next |
Token ID | 1306 |
Feature activation | +7.46e-02 |
Pos logit contributions | |
Ho | +0.023 |
Rib | +0.023 |
H | +0.021 |
Bub | +0.019 |
Bun | +0.019 |
Neg logit contributions | |
eks | -0.017 |
oos | -0.017 |
ays | -0.016 |
cases | -0.016 |
ases | -0.016 |
Token | journey |
Token ID | 7002 |
Feature activation | +8.85e-02 |
Pos logit contributions | |
eks | +0.107 |
oos | +0.107 |
ays | +0.104 |
cases | +0.103 |
ases | +0.103 |
Neg logit contributions | |
Ho | -0.140 |
Rib | -0.137 |
H | -0.124 |
Bub | -0.115 |
Bun | -0.115 |
Token | ! |
Token ID | 0 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
eks | +0.143 |
oos | +0.139 |
ays | +0.137 |
ases | +0.137 |
agged | +0.136 |
Neg logit contributions | |
Ho | -0.151 |
Rib | -0.147 |
H | -0.133 |
Bun | -0.124 |
Bub | -0.123 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
cases | +0.060 |
ays | +0.059 |
he | +0.058 |
oos | +0.058 |
elevator | +0.057 |
Neg logit contributions | |
Ho | -0.049 |
Rib | -0.049 |
H | -0.042 |
accidentally | -0.040 |
Bub | -0.039 |
Token | Sam |
Token ID | 16305 |
Feature activation | +2.84e-02 |
Pos logit contributions | |
he | +0.153 |
hed | +0.145 |
ays | +0.141 |
oos | +0.139 |
agged | +0.132 |
Neg logit contributions | |
Ho | -0.226 |
Rib | -0.226 |
H | -0.203 |
accidentally | -0.194 |
Butter | -0.193 |
Token | my |
Token ID | 1820 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
eks | +0.052 |
oos | +0.051 |
cases | +0.051 |
agged | +0.051 |
ays | +0.051 |
Neg logit contributions | |
Ho | -0.042 |
Rib | -0.040 |
H | -0.036 |
Bun | -0.033 |
Bub | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
Ho | +0.020 |
Rib | +0.019 |
H | +0.017 |
Bun | +0.016 |
accidentally | +0.016 |
Neg logit contributions | |
eks | -0.021 |
oos | -0.021 |
agged | -0.021 |
ases | -0.021 |
cases | -0.021 |
Token | was |
Token ID | 373 |
Feature activation | +8.21e-03 |
Pos logit contributions | |
eks | +0.002 |
oos | +0.002 |
elevator | +0.002 |
cases | +0.002 |
ays | +0.002 |
Neg logit contributions | |
Ho | -0.002 |
Rib | -0.002 |
H | -0.002 |
Bub | -0.002 |
Bun | -0.002 |
Token | living |
Token ID | 2877 |
Feature activation | -5.04e-02 |
Pos logit contributions | |
elevator | +0.010 |
oos | +0.010 |
eks | +0.010 |
ays | +0.010 |
ases | +0.010 |
Neg logit contributions | |
Ho | -0.016 |
Rib | -0.016 |
H | -0.015 |
Bub | -0.014 |
Bun | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -9.36e-03 |
Pos logit contributions | |
Ho | +0.091 |
Rib | +0.089 |
H | +0.081 |
Bub | +0.074 |
Bun | +0.073 |
Neg logit contributions | |
eks | -0.076 |
oos | -0.075 |
cases | -0.075 |
agged | -0.074 |
ays | -0.073 |
Token | He |
Token ID | 679 |
Feature activation | +2.55e-02 |
Pos logit contributions | |
Rib | +0.018 |
Ho | +0.018 |
please | +0.015 |
Ros | +0.015 |
accidentally | +0.015 |
Neg logit contributions | |
cases | -0.011 |
elevator | -0.010 |
oos | -0.010 |
he | -0.010 |
roads | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | +7.21e-03 |
Pos logit contributions | |
eks | +0.036 |
cases | +0.035 |
led | +0.034 |
oos | +0.034 |
elevator | +0.033 |
Neg logit contributions | |
Rib | -0.047 |
Ho | -0.047 |
H | -0.042 |
Butter | -0.039 |
Bub | -0.038 |
Token | eager |
Token ID | 11069 |
Feature activation | +8.63e-02 |
Pos logit contributions | |
eks | +0.010 |
elevator | +0.010 |
cases | +0.009 |
oos | +0.009 |
ases | +0.009 |
Neg logit contributions | |
Ho | -0.013 |
Rib | -0.013 |
H | -0.012 |
Bub | -0.011 |
Bun | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
oos | +0.153 |
eks | +0.151 |
elevator | +0.150 |
cases | +0.150 |
ases | +0.147 |
Neg logit contributions | |
Ho | -0.130 |
Rib | -0.128 |
H | -0.110 |
Bub | -0.102 |
Butter | -0.101 |
Token | find |
Token ID | 1064 |
Feature activation | -1.76e-03 |
Pos logit contributions | |
Ho | +0.034 |
Rib | +0.033 |
H | +0.030 |
Bub | +0.028 |
Bun | +0.028 |
Neg logit contributions | |
oos | -0.024 |
eks | -0.024 |
ays | -0.024 |
cases | -0.023 |
elevator | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
Ho | +0.003 |
Rib | +0.003 |
H | +0.003 |
Bub | +0.003 |
Bun | +0.003 |
Neg logit contributions | |
oos | -0.003 |
agged | -0.003 |
ays | -0.003 |
cases | -0.003 |
eks | -0.002 |
Token | creature |
Token ID | 7185 |
Feature activation | +7.93e-04 |
Pos logit contributions | |
Ho | +0.029 |
Rib | +0.029 |
H | +0.026 |
Bun | +0.024 |
Bub | +0.024 |
Neg logit contributions | |
elevator | -0.022 |
eks | -0.021 |
oos | -0.020 |
he | -0.020 |
cases | -0.019 |
Token | ! |
Token ID | 0 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
eks | +0.001 |
oos | +0.001 |
ays | +0.001 |
ases | +0.001 |
agged | +0.001 |
Neg logit contributions | |
Ho | -0.001 |
Rib | -0.001 |
H | -0.001 |
Bub | -0.001 |
Bun | -0.001 |
Token | we |
Token ID | 356 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
Rib | +0.054 |
Ho | +0.054 |
H | +0.047 |
Bub | +0.045 |
Fro | +0.045 |
Neg logit contributions | |
elevator | -0.045 |
oos | -0.043 |
cases | -0.043 |
eks | -0.043 |
tickets | -0.042 |
Token | can |
Token ID | 460 |
Feature activation | -9.79e-02 |
Pos logit contributions | |
eks | +0.029 |
oos | +0.028 |
ays | +0.028 |
elevator | +0.028 |
cases | +0.028 |
Neg logit contributions | |
Ho | -0.034 |
Rib | -0.033 |
H | -0.030 |
Bub | -0.028 |
Bun | -0.028 |
Token | fix |
Token ID | 4259 |
Feature activation | -8.45e-02 |
Pos logit contributions | |
Ho | +0.183 |
Rib | +0.181 |
H | +0.161 |
Butter | +0.150 |
Bub | +0.149 |
Neg logit contributions | |
he | -0.140 |
eks | -0.138 |
oos | -0.137 |
ases | -0.136 |
ays | -0.134 |
Token | it |
Token ID | 340 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
Rib | +0.123 |
Ho | +0.122 |
H | +0.104 |
Bub | +0.097 |
Bun | +0.095 |
Neg logit contributions | |
ays | -0.153 |
oos | -0.152 |
eks | -0.151 |
elevator | -0.150 |
ases | -0.149 |
Token | ." |
Token ID | 526 |
Feature activation | -7.09e-03 |
Pos logit contributions | |
eks | +0.019 |
oos | +0.019 |
ays | +0.018 |
elevator | +0.018 |
cases | +0.018 |
Neg logit contributions | |
Ho | -0.021 |
Rib | -0.020 |
H | -0.018 |
Bub | -0.017 |
Bun | -0.017 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
Rib | +0.011 |
Ho | +0.011 |
H | +0.009 |
Bub | +0.009 |
Bun | +0.009 |
Neg logit contributions | |
he | -0.012 |
ays | -0.012 |
oos | -0.011 |
agged | -0.011 |
ags | -0.011 |
Token | Sam |
Token ID | 16305 |
Feature activation | +3.64e-02 |
Pos logit contributions | |
he | +0.129 |
ays | +0.124 |
hed | +0.120 |
oos | +0.119 |
ases | +0.117 |
Neg logit contributions | |
Ho | -0.255 |
Rib | -0.255 |
H | -0.232 |
Bub | -0.229 |
accidentally | -0.227 |
Token | likes |
Token ID | 7832 |
Feature activation | +3.99e-02 |
Pos logit contributions | |
eks | +0.067 |
oos | +0.067 |
cases | +0.066 |
ays | +0.066 |
elevator | +0.065 |
Neg logit contributions | |
Ho | -0.053 |
Rib | -0.050 |
H | -0.046 |
Bun | -0.041 |
Bub | -0.041 |
Token | to |
Token ID | 284 |
Feature activation | -2.38e-03 |
Pos logit contributions | |
elevator | +0.057 |
oos | +0.055 |
tickets | +0.051 |
bags | +0.050 |
ases | +0.050 |
Neg logit contributions | |
Ho | -0.072 |
Rib | -0.071 |
H | -0.062 |
Ros | -0.060 |
Tw | -0.060 |
Token | play |
Token ID | 711 |
Feature activation | +8.16e-03 |
Pos logit contributions | |
Ho | +0.004 |
Rib | +0.004 |
H | +0.004 |
Bub | +0.003 |
Bun | +0.003 |
Neg logit contributions | |
eks | -0.004 |
oos | -0.004 |
ays | -0.004 |
elevator | -0.004 |
cases | -0.004 |
Token | with |
Token ID | 351 |
Feature activation | +4.39e-02 |
Pos logit contributions | |
oos | +0.012 |
eks | +0.012 |
elevator | +0.012 |
ays | +0.012 |
ases | +0.012 |
Neg logit contributions | |
Ho | -0.015 |
Rib | -0.015 |
H | -0.013 |
Bub | -0.012 |
Butter | -0.012 |
Token | a |
Token ID | 257 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
eks | +0.033 |
oos | +0.032 |
elevator | +0.032 |
ases | +0.032 |
he | +0.030 |
Neg logit contributions | |
Ho | -0.046 |
Rib | -0.046 |
H | -0.040 |
Bub | -0.039 |
Butter | -0.038 |
Token | year |
Token ID | 614 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
Rib | +0.029 |
Ho | +0.028 |
H | +0.024 |
Stri | +0.024 |
Fro | +0.023 |
Neg logit contributions | |
elevator | -0.024 |
cases | -0.024 |
he | -0.023 |
ued | -0.023 |
tickets | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | +3.02e-03 |
Pos logit contributions | |
eks | +0.157 |
oos | +0.156 |
ays | +0.152 |
ases | +0.151 |
agged | +0.150 |
Neg logit contributions | |
Ho | -0.179 |
Rib | -0.175 |
H | -0.159 |
Bub | -0.146 |
Bun | -0.146 |
Token | more |
Token ID | 517 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
cases | +0.004 |
eks | +0.004 |
elevator | +0.004 |
are | +0.004 |
oos | +0.004 |
Neg logit contributions | |
Rib | -0.005 |
Ho | -0.005 |
H | -0.005 |
Bub | -0.004 |
Bun | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
Ho | +0.202 |
Rib | +0.201 |
H | +0.179 |
Bub | +0.170 |
Butter | +0.165 |
Neg logit contributions | |
oos | -0.135 |
elevator | -0.134 |
ases | -0.131 |
led | -0.130 |
cases | -0.128 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
Ho | +0.027 |
Rib | +0.027 |
accidentally | +0.025 |
H | +0.024 |
Bun | +0.023 |
Neg logit contributions | |
ays | -0.022 |
ases | -0.022 |
he | -0.022 |
agged | -0.021 |
ags | -0.021 |
Token | Anna |
Token ID | 31160 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
he | +0.196 |
ays | +0.185 |
oos | +0.185 |
hed | +0.183 |
ases | +0.182 |
Neg logit contributions | |
Ho | -0.169 |
Rib | -0.166 |
H | -0.144 |
Bub | -0.137 |
accidentally | -0.136 |
Token | and |
Token ID | 290 |
Feature activation | +1.93e-02 |
Pos logit contributions | |
Ho | +0.053 |
Rib | +0.052 |
H | +0.047 |
Bub | +0.043 |
Bun | +0.043 |
Neg logit contributions | |
eks | -0.048 |
oos | -0.048 |
ays | -0.047 |
cases | -0.047 |
elevator | -0.047 |
Token | Ben |
Token ID | 3932 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
oos | +0.034 |
eks | +0.033 |
ays | +0.033 |
cases | +0.033 |
elevator | +0.032 |
Neg logit contributions | |
Ho | -0.031 |
Rib | -0.030 |
H | -0.027 |
Bub | -0.024 |
Bun | -0.024 |
Token | love |
Token ID | 1842 |
Feature activation | +3.32e-02 |
Pos logit contributions | |
cases | +0.022 |
ays | +0.022 |
oos | +0.021 |
agged | +0.021 |
irens | +0.021 |
Neg logit contributions | |
Ho | -0.025 |
Rib | -0.022 |
H | -0.022 |
Bub | -0.020 |
accidentally | -0.020 |
Token | to |
Token ID | 284 |
Feature activation | -6.92e-03 |
Pos logit contributions | |
oos | +0.054 |
elevator | +0.054 |
ases | +0.052 |
ays | +0.051 |
tickets | +0.050 |
Neg logit contributions | |
Ho | -0.055 |
Rib | -0.053 |
H | -0.048 |
Bub | -0.046 |
Bun | -0.045 |
Token | vendor |
Token ID | 18371 |
Feature activation | +6.92e-02 |
Pos logit contributions | |
cks | +0.010 |
eks | +0.010 |
ases | +0.010 |
ays | +0.010 |
oos | +0.010 |
Neg logit contributions | |
Lily | -0.004 |
Rib | -0.004 |
" | -0.004 |
| -0.003 |
friend | -0.003 |
Token | is |
Token ID | 318 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
agged | +0.094 |
oos | +0.094 |
ags | +0.088 |
cases | +0.087 |
ays | +0.087 |
Neg logit contributions | |
Ho | -0.133 |
Rib | -0.130 |
H | -0.112 |
Tw | -0.108 |
Will | -0.106 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
elevator | +0.031 |
eks | +0.031 |
agged | +0.030 |
ases | +0.030 |
ags | +0.030 |
Neg logit contributions | |
Ho | -0.037 |
Rib | -0.036 |
H | -0.032 |
Bun | -0.031 |
Bub | -0.031 |
Token | to |
Token ID | 284 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
ases | +0.017 |
oos | +0.017 |
ays | +0.017 |
elevator | +0.017 |
out | +0.017 |
Neg logit contributions | |
Ho | -0.030 |
Rib | -0.030 |
H | -0.027 |
Bub | -0.026 |
Bun | -0.026 |
Token | see |
Token ID | 766 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
Ho | +0.023 |
Rib | +0.023 |
H | +0.021 |
Bub | +0.019 |
Bun | +0.018 |
Neg logit contributions | |
ays | -0.025 |
cases | -0.024 |
oos | -0.024 |
led | -0.023 |
elevator | -0.023 |
Token | them |
Token ID | 606 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
oos | +0.065 |
eks | +0.065 |
ays | +0.064 |
elevator | +0.063 |
ases | +0.063 |
Neg logit contributions | |
Ho | -0.051 |
Rib | -0.050 |
H | -0.044 |
Bub | -0.039 |
Bun | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
oos | +0.080 |
eks | +0.079 |
agged | +0.078 |
ays | +0.076 |
elevator | +0.075 |
Neg logit contributions | |
Ho | -0.072 |
Rib | -0.071 |
H | -0.063 |
Bub | -0.058 |
Bun | -0.057 |
Token | He |
Token ID | 679 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
Rib | +0.029 |
Ho | +0.029 |
Bun | +0.026 |
H | +0.025 |
Ros | +0.025 |
Neg logit contributions | |
ays | -0.020 |
elevator | -0.019 |
cases | -0.019 |
agged | -0.019 |
ags | -0.019 |
Token | gives |
Token ID | 3607 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
oos | +0.028 |
eks | +0.027 |
ays | +0.027 |
cases | +0.026 |
elevator | +0.026 |
Neg logit contributions | |
Ho | -0.047 |
Rib | -0.046 |
H | -0.042 |
Bub | -0.039 |
Bun | -0.039 |
Token | them |
Token ID | 606 |
Feature activation | +4.47e-02 |
Pos logit contributions | |
Ho | +0.047 |
Rib | +0.047 |
H | +0.043 |
Bun | +0.040 |
Bub | +0.040 |
Neg logit contributions | |
oos | -0.027 |
eks | -0.027 |
ays | -0.026 |
elevator | -0.026 |
ases | -0.025 |
Token | each |
Token ID | 1123 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
eks | +0.084 |
roads | +0.083 |
ays | +0.082 |
cks | +0.081 |
ement | +0.079 |
Neg logit contributions | |
Ho | -0.071 |
Rib | -0.067 |
H | -0.059 |
Bub | -0.056 |
Lucky | -0.055 |
Token | asleep |
Token ID | 16039 |
Feature activation | -4.95e-02 |
Pos logit contributions | |
Rib | +0.024 |
Ho | +0.024 |
H | +0.020 |
Butter | +0.019 |
Bub | +0.019 |
Neg logit contributions | |
oos | -0.028 |
eks | -0.028 |
ays | -0.027 |
elevator | -0.027 |
cases | -0.027 |
Token | with |
Token ID | 351 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
Rib | +0.093 |
Ho | +0.091 |
H | +0.079 |
Butter | +0.077 |
Bub | +0.077 |
Neg logit contributions | |
he | -0.070 |
led | -0.066 |
eks | -0.066 |
elevator | -0.065 |
cases | -0.065 |
Token | a |
Token ID | 257 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
eks | +0.046 |
oos | +0.046 |
ays | +0.044 |
elevator | +0.044 |
cases | +0.044 |
Neg logit contributions | |
Ho | -0.050 |
Rib | -0.050 |
H | -0.044 |
Bub | -0.041 |
Bun | -0.041 |
Token | tum |
Token ID | 11814 |
Feature activation | +6.23e-02 |
Pos logit contributions | |
Rib | +0.044 |
Ho | +0.042 |
H | +0.040 |
Bub | +0.037 |
Stri | +0.037 |
Neg logit contributions | |
he | -0.022 |
elevator | -0.020 |
cases | -0.019 |
eks | -0.019 |
lobby | -0.018 |
Token | my |
Token ID | 1820 |
Feature activation | -3.59e-02 |
Pos logit contributions | |
ays | +0.098 |
oos | +0.098 |
he | +0.097 |
eks | +0.097 |
agged | +0.096 |
Neg logit contributions | |
Ho | -0.108 |
Rib | -0.106 |
H | -0.093 |
Bub | -0.088 |
Bun | -0.086 |
Token | full |
Token ID | 1336 |
Feature activation | +4.68e-02 |
Pos logit contributions | |
Rib | +0.067 |
Ho | +0.066 |
H | +0.060 |
Bub | +0.056 |
Fro | +0.054 |
Neg logit contributions | |
he | -0.050 |
led | -0.049 |
cases | -0.049 |
elevator | -0.048 |
oos | -0.048 |
Token | of |
Token ID | 286 |
Feature activation | -4.73e-02 |
Pos logit contributions | |
eks | +0.058 |
oos | +0.057 |
ays | +0.055 |
cases | +0.055 |
ases | +0.055 |
Neg logit contributions | |
Ho | -0.098 |
Rib | -0.096 |
H | -0.088 |
Bub | -0.083 |
Bun | -0.082 |
Token | y |
Token ID | 331 |
Feature activation | -9.38e-03 |
Pos logit contributions | |
Ho | +0.087 |
Rib | +0.086 |
H | +0.077 |
Bub | +0.072 |
Bun | +0.071 |
Neg logit contributions | |
eks | -0.068 |
oos | -0.068 |
elevator | -0.066 |
ays | -0.066 |
ases | -0.065 |
Token | ummy |
Token ID | 13513 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
Ho | +0.015 |
Rib | +0.014 |
H | +0.012 |
Po | +0.012 |
Benny | +0.012 |
Neg logit contributions | |
eks | -0.017 |
roads | -0.017 |
cases | -0.016 |
elevator | -0.016 |
ued | -0.016 |
Token | cookies |
Token ID | 14746 |
Feature activation | -5.46e-02 |
Pos logit contributions | |
eks | +0.023 |
oos | +0.022 |
ays | +0.022 |
ases | +0.022 |
cases | +0.021 |
Neg logit contributions | |
Ho | -0.025 |
Rib | -0.024 |
H | -0.021 |
Bun | -0.020 |
Bub | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
Ho | +0.095 |
Rib | +0.091 |
H | +0.082 |
Bun | +0.076 |
Bub | +0.076 |
Neg logit contributions | |
eks | -0.088 |
oos | -0.087 |
ases | -0.084 |
ays | -0.083 |
agged | -0.083 |
Token | to |
Token ID | 284 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
oos | +0.037 |
eks | +0.037 |
agged | +0.037 |
ays | +0.036 |
ases | +0.036 |
Neg logit contributions | |
Ho | -0.034 |
Rib | -0.033 |
H | -0.031 |
Bub | -0.027 |
Bun | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
Ho | +0.027 |
Rib | +0.027 |
H | +0.023 |
Bub | +0.022 |
Bun | +0.022 |
Neg logit contributions | |
eks | -0.023 |
oos | -0.022 |
elevator | -0.022 |
ases | -0.022 |
ays | -0.021 |
Token | theater |
Token ID | 13766 |
Feature activation | +9.01e-03 |
Pos logit contributions | |
Ho | +0.025 |
Rib | +0.024 |
H | +0.023 |
Bub | +0.021 |
Tw | +0.021 |
Neg logit contributions | |
agged | -0.014 |
eks | -0.014 |
ays | -0.014 |
oos | -0.014 |
ases | -0.014 |
Token | again |
Token ID | 757 |
Feature activation | -3.61e-02 |
Pos logit contributions | |
agged | +0.017 |
eks | +0.017 |
ays | +0.016 |
oos | +0.016 |
ases | +0.016 |
Neg logit contributions | |
Ho | -0.013 |
Rib | -0.012 |
H | -0.012 |
Tw | -0.010 |
Bun | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
Ho | +0.075 |
Rib | +0.074 |
H | +0.065 |
Bub | +0.064 |
Bun | +0.062 |
Neg logit contributions | |
oos | -0.044 |
eks | -0.043 |
he | -0.042 |
ases | -0.042 |
elevator | -0.041 |
Token | They |
Token ID | 1119 |
Feature activation | +4.37e-02 |
Pos logit contributions | |
Ho | +0.028 |
Rib | +0.027 |
accidentally | +0.025 |
please | +0.025 |
H | +0.025 |
Neg logit contributions | |
ays | -0.013 |
oos | -0.012 |
cases | -0.011 |
ases | -0.011 |
he | -0.011 |
Token | never |
Token ID | 1239 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
oos | +0.075 |
eks | +0.074 |
ays | +0.074 |
agged | +0.073 |
ases | +0.073 |
Neg logit contributions | |
Ho | -0.070 |
Rib | -0.067 |
H | -0.061 |
Bub | -0.056 |
Bun | -0.055 |
Token | missed |
Token ID | 6825 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
eks | +0.021 |
oos | +0.021 |
ases | +0.020 |
cases | +0.020 |
elevator | +0.020 |
Neg logit contributions | |
Ho | -0.021 |
Rib | -0.020 |
H | -0.018 |
Bub | -0.017 |
Bun | -0.017 |
Token | one |
Token ID | 530 |
Feature activation | +6.54e-02 |
Pos logit contributions | |
eks | +0.031 |
oos | +0.031 |
ays | +0.030 |
elevator | +0.030 |
cases | +0.030 |
Neg logit contributions | |
Ho | -0.033 |
Rib | -0.032 |
H | -0.029 |
Bub | -0.026 |
Bun | -0.026 |
Token | show |
Token ID | 905 |
Feature activation | +2.16e-03 |
Pos logit contributions | |
eks | +0.108 |
oos | +0.107 |
ays | +0.105 |
cases | +0.104 |
agged | +0.104 |
Neg logit contributions | |
Ho | -0.109 |
Rib | -0.107 |
H | -0.095 |
Bub | -0.088 |
Bun | -0.087 |
Token | ! |
Token ID | 0 |
Feature activation | +3.28e-02 |
Pos logit contributions | |
eks | +0.003 |
oos | +0.003 |
ays | +0.003 |
cases | +0.003 |
elevator | +0.003 |
Neg logit contributions | |
Ho | -0.004 |
Rib | -0.004 |
H | -0.003 |
Bub | -0.003 |
Bun | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | -3.13e-03 |
Pos logit contributions | |
oos | +0.071 |
ays | +0.070 |
ases | +0.069 |
eks | +0.069 |
agged | +0.068 |
Neg logit contributions | |
Ho | -0.070 |
Rib | -0.065 |
H | -0.064 |
Bun | -0.058 |
Bub | -0.057 |
Token | the |
Token ID | 262 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
Rib | +0.006 |
Ho | +0.006 |
Ros | +0.005 |
Bub | +0.005 |
H | +0.005 |
Neg logit contributions | |
elevator | -0.004 |
tickets | -0.004 |
eks | -0.004 |
oos | -0.004 |
ases | -0.004 |
Token | beach |
Token ID | 10481 |
Feature activation | +3.99e-02 |
Pos logit contributions | |
elevator | +0.015 |
oos | +0.013 |
tickets | +0.012 |
lobby | +0.012 |
out | +0.012 |
Neg logit contributions | |
Rib | -0.032 |
Ho | -0.031 |
H | -0.028 |
Bub | -0.028 |
Bun | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
eks | +0.076 |
agged | +0.074 |
ases | +0.072 |
ays | +0.071 |
oos | +0.071 |
Neg logit contributions | |
Ho | -0.059 |
Rib | -0.057 |
H | -0.051 |
Bun | -0.045 |
Bub | -0.045 |
Token | One |
Token ID | 1881 |
Feature activation | +6.13e-02 |
Pos logit contributions | |
elevator | +0.013 |
ays | +0.013 |
he | +0.013 |
downtown | +0.012 |
oos | +0.012 |
Neg logit contributions | |
Rib | -0.026 |
Ho | -0.026 |
onna | -0.023 |
accidentally | -0.022 |
fle | -0.022 |
Token | day |
Token ID | 1110 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
eks | +0.120 |
ases | +0.118 |
oos | +0.118 |
agged | +0.117 |
cases | +0.117 |
Neg logit contributions | |
Ho | -0.084 |
Rib | -0.083 |
H | -0.073 |
Bub | -0.066 |
Butter | -0.064 |
Token | , |
Token ID | 11 |
Feature activation | +6.69e-03 |
Pos logit contributions | |
oos | +0.015 |
elevator | +0.014 |
cases | +0.014 |
eks | +0.014 |
led | +0.014 |
Neg logit contributions | |
Rib | -0.030 |
Ho | -0.029 |
H | -0.026 |
Bub | -0.025 |
Butter | -0.025 |
Token | he |
Token ID | 339 |
Feature activation | +9.93e-03 |
Pos logit contributions | |
elevator | +0.010 |
cases | +0.009 |
he | +0.009 |
oos | +0.009 |
downtown | +0.009 |
Neg logit contributions | |
Rib | -0.012 |
Ho | -0.011 |
onna | -0.010 |
Tw | -0.009 |
H | -0.009 |
Token | saw |
Token ID | 2497 |
Feature activation | +6.39e-02 |
Pos logit contributions | |
eks | +0.015 |
oos | +0.015 |
ays | +0.015 |
cases | +0.015 |
elevator | +0.015 |
Neg logit contributions | |
Ho | -0.018 |
Rib | -0.017 |
H | -0.015 |
Bub | -0.014 |
Bun | -0.014 |
Token | a |
Token ID | 257 |
Feature activation | -3.58e-03 |
Pos logit contributions | |
eks | +0.100 |
oos | +0.100 |
ays | +0.097 |
cases | +0.097 |
ases | +0.097 |
Neg logit contributions | |
Ho | -0.112 |
Rib | -0.109 |
H | -0.099 |
Bub | -0.091 |
Bun | -0.090 |
Token | big |
Token ID | 1263 |
Feature activation | +7.64e-02 |
Pos logit contributions | |
Rib | +0.008 |
Ho | +0.007 |
H | +0.007 |
Bub | +0.006 |
Bun | +0.006 |
Neg logit contributions | |
elevator | -0.004 |
he | -0.004 |
lobby | -0.004 |
eks | -0.004 |
tickets | -0.004 |
Token | it |
Token ID | 340 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
elevator | +0.017 |
eks | +0.017 |
led | +0.017 |
cases | +0.016 |
oos | +0.016 |
Neg logit contributions | |
Rib | -0.017 |
Ho | -0.016 |
H | -0.013 |
onna | -0.013 |
please | -0.013 |
Token | wouldn |
Token ID | 3636 |
Feature activation | -9.85e-02 |
Pos logit contributions | |
eks | +0.020 |
led | +0.020 |
elevator | +0.019 |
oos | +0.019 |
cases | +0.019 |
Neg logit contributions | |
Rib | -0.025 |
Ho | -0.024 |
H | -0.021 |
Bub | -0.020 |
Butter | -0.019 |
Token | 't |
Token ID | 470 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
Ho | +0.174 |
Rib | +0.170 |
H | +0.154 |
Bub | +0.142 |
Bun | +0.142 |
Neg logit contributions | |
oos | -0.151 |
eks | -0.151 |
ays | -0.146 |
elevator | -0.146 |
cases | -0.146 |
Token | bud |
Token ID | 3600 |
Feature activation | -5.37e-02 |
Pos logit contributions | |
Ho | +0.029 |
Rib | +0.028 |
H | +0.026 |
Bun | +0.024 |
Bub | +0.024 |
Neg logit contributions | |
oos | -0.021 |
eks | -0.021 |
ays | -0.020 |
ases | -0.020 |
agged | -0.020 |
Token | ge |
Token ID | 469 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
Ho | +0.095 |
Rib | +0.094 |
H | +0.081 |
Bub | +0.077 |
Butter | +0.077 |
Neg logit contributions | |
cases | -0.082 |
eks | -0.082 |
oos | -0.081 |
ays | -0.081 |
ases | -0.079 |
Token | . |
Token ID | 13 |
Feature activation | +5.41e-03 |
Pos logit contributions | |
Ho | +0.052 |
Rib | +0.049 |
H | +0.046 |
Bun | +0.042 |
Bub | +0.041 |
Neg logit contributions | |
eks | -0.050 |
oos | -0.049 |
ays | -0.048 |
agged | -0.048 |
elevator | -0.048 |
Token | Then |
Token ID | 3244 |
Feature activation | -7.72e-02 |
Pos logit contributions | |
oos | +0.007 |
cases | +0.007 |
agged | +0.007 |
elevator | +0.007 |
ued | +0.007 |
Neg logit contributions | |
Rib | -0.009 |
Ho | -0.008 |
onna | -0.008 |
please | -0.008 |
Tooth | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | -4.22e-04 |
Pos logit contributions | |
Rib | +0.172 |
Ho | +0.164 |
onna | +0.150 |
H | +0.146 |
Butter | +0.142 |
Neg logit contributions | |
eks | -0.078 |
oos | -0.073 |
led | -0.073 |
elevator | -0.072 |
cases | -0.070 |
Token | Bun |
Token ID | 28515 |
Feature activation | -7.75e-02 |
Pos logit contributions | |
Rib | +0.001 |
onna | +0.001 |
Ho | +0.001 |
please | +0.001 |
die | +0.001 |
Neg logit contributions | |
eks | -0.001 |
led | -0.001 |
elevator | -0.001 |
cases | -0.001 |
ued | -0.001 |
Token | remembered |
Token ID | 12086 |
Feature activation | -4.30e-02 |
Pos logit contributions | |
Rib | +0.145 |
onna | +0.132 |
Ho | +0.126 |
Butter | +0.126 |
Pract | +0.117 |
Neg logit contributions | |
led | -0.112 |
eks | -0.096 |
ed | -0.094 |
out | -0.093 |
hed | -0.093 |
Token | that |
Token ID | 326 |
Feature activation | -5.69e-03 |
Pos logit contributions | |
Rib | +0.094 |
Ho | +0.091 |
H | +0.079 |
Butter | +0.079 |
Bub | +0.078 |
Neg logit contributions | |
eks | -0.045 |
oos | -0.045 |
elevator | -0.044 |
out | -0.043 |
cases | -0.040 |
Token | and |
Token ID | 290 |
Feature activation | +8.56e-03 |
Pos logit contributions | |
eks | +0.002 |
ases | +0.002 |
elevator | +0.002 |
oos | +0.002 |
agged | +0.002 |
Neg logit contributions | |
Ho | -0.003 |
Rib | -0.002 |
H | -0.002 |
Bun | -0.002 |
Bub | -0.002 |
Token | move |
Token ID | 1445 |
Feature activation | -7.55e-03 |
Pos logit contributions | |
eks | +0.011 |
oos | +0.010 |
are | +0.010 |
elevator | +0.010 |
cases | +0.010 |
Neg logit contributions | |
Ho | -0.017 |
Rib | -0.016 |
H | -0.015 |
Tooth | -0.014 |
Stri | -0.014 |
Token | in |
Token ID | 287 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
Ho | +0.014 |
Rib | +0.014 |
H | +0.013 |
Bub | +0.012 |
Bun | +0.012 |
Neg logit contributions | |
oos | -0.010 |
eks | -0.010 |
ays | -0.010 |
elevator | -0.010 |
cases | -0.010 |
Token | circles |
Token ID | 13332 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
Ho | +0.032 |
Rib | +0.031 |
H | +0.029 |
Bun | +0.026 |
Bub | +0.026 |
Neg logit contributions | |
eks | -0.037 |
cases | -0.037 |
oos | -0.036 |
agged | -0.036 |
ays | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
eks | +0.018 |
oos | +0.017 |
ays | +0.017 |
ases | +0.017 |
elevator | +0.017 |
Neg logit contributions | |
Ho | -0.020 |
Rib | -0.019 |
H | -0.017 |
Bun | -0.016 |
Bub | -0.016 |
Token | Stella |
Token ID | 45856 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
Rib | +0.021 |
Ho | +0.021 |
H | +0.019 |
Stri | +0.018 |
Fro | +0.017 |
Neg logit contributions | |
oos | -0.017 |
cases | -0.017 |
agged | -0.017 |
ays | -0.016 |
hed | -0.016 |
Token | followed |
Token ID | 3940 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
Rib | +0.031 |
Ho | +0.031 |
H | +0.028 |
Bub | +0.027 |
Butter | +0.026 |
Neg logit contributions | |
hed | -0.022 |
led | -0.022 |
eks | -0.021 |
cases | -0.021 |
oos | -0.021 |
Token | her |
Token ID | 607 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
oos | +0.051 |
eks | +0.050 |
elevator | +0.049 |
he | +0.048 |
cases | +0.048 |
Neg logit contributions | |
Rib | -0.062 |
Ho | -0.062 |
H | -0.055 |
Bub | -0.051 |
Stri | -0.050 |
Token | mom |
Token ID | 1995 |
Feature activation | +5.53e-02 |
Pos logit contributions | |
Rib | +0.040 |
Ho | +0.039 |
H | +0.034 |
Bub | +0.033 |
Butter | +0.033 |
Neg logit contributions | |
elevator | -0.032 |
oos | -0.032 |
cases | -0.032 |
eks | -0.031 |
ases | -0.031 |
Token | 's |
Token ID | 338 |
Feature activation | -4.83e-02 |
Pos logit contributions | |
oos | +0.064 |
cases | +0.063 |
eks | +0.062 |
elevator | +0.062 |
out | +0.060 |
Neg logit contributions | |
Rib | -0.118 |
Ho | -0.116 |
H | -0.104 |
Bub | -0.101 |
Butter | -0.098 |
Token | lead |
Token ID | 1085 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
Ho | +0.097 |
Rib | +0.094 |
H | +0.086 |
Bub | +0.080 |
Bun | +0.080 |
Neg logit contributions | |
eks | -0.064 |
oos | -0.064 |
ays | -0.062 |
cases | -0.062 |
ases | -0.061 |
Token | against |
Token ID | 1028 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
Ho | +0.020 |
Rib | +0.019 |
H | +0.017 |
Bub | +0.016 |
Bun | +0.016 |
Neg logit contributions | |
oos | -0.016 |
eks | -0.015 |
cases | -0.015 |
ays | -0.015 |
elevator | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
oos | +0.032 |
eks | +0.031 |
elevator | +0.031 |
ases | +0.030 |
he | +0.030 |
Neg logit contributions | |
Ho | -0.041 |
Rib | -0.040 |
H | -0.036 |
Bub | -0.034 |
Bun | -0.033 |
Token | fat |
Token ID | 3735 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
Ho | +0.027 |
Rib | +0.027 |
H | +0.025 |
Stri | +0.022 |
Bub | +0.022 |
Neg logit contributions | |
elevator | -0.029 |
oos | -0.028 |
eks | -0.027 |
ases | -0.027 |
ps | -0.027 |
Token | frog |
Token ID | 21264 |
Feature activation | -5.58e-02 |
Pos logit contributions | |
Ho | +0.021 |
Rib | +0.020 |
H | +0.018 |
Bun | +0.016 |
Bub | +0.016 |
Neg logit contributions | |
oos | -0.027 |
elevator | -0.026 |
ases | -0.026 |
eks | -0.026 |
agged | -0.026 |
Token | â |
Token ID | 22940 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
Ho | +0.110 |
Rib | +0.109 |
H | +0.101 |
Bub | +0.094 |
Bun | +0.091 |
Neg logit contributions | |
oos | -0.071 |
cases | -0.070 |
elevator | -0.068 |
eks | -0.067 |
he | -0.067 |
Token | € |
Token ID | 26391 |
Feature activation | -8.61e-02 |
Pos logit contributions | |
oos | +0.017 |
eks | +0.017 |
agged | +0.016 |
ases | +0.016 |
elevator | +0.016 |
Neg logit contributions | |
Ho | -0.022 |
Rib | -0.021 |
H | -0.019 |
Bub | -0.018 |
Bun | -0.018 |
Token | ™ |
Token ID | 8151 |
Feature activation | +9.57e-03 |
Pos logit contributions | |
Ho | +0.149 |
Rib | +0.146 |
H | +0.131 |
Bub | +0.124 |
Bun | +0.118 |
Neg logit contributions | |
oos | -0.136 |
eks | -0.132 |
cases | -0.132 |
ays | -0.131 |
ases | -0.128 |
Token | s |
Token ID | 82 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
oos | +0.013 |
eks | +0.013 |
ays | +0.013 |
cases | +0.013 |
elevator | +0.013 |
Neg logit contributions | |
Ho | -0.018 |
Rib | -0.018 |
H | -0.016 |
Bub | -0.015 |
Bun | -0.015 |
Token | shell |
Token ID | 7582 |
Feature activation | -8.88e-03 |
Pos logit contributions | |
Ho | +0.046 |
Rib | +0.046 |
H | +0.041 |
Bub | +0.039 |
Bun | +0.038 |
Neg logit contributions | |
oos | -0.035 |
eks | -0.035 |
cases | -0.034 |
ases | -0.034 |
ays | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -2.52e-03 |
Pos logit contributions | |
Ho | +0.017 |
Rib | +0.016 |
H | +0.015 |
Bub | +0.014 |
Bun | +0.014 |
Neg logit contributions | |
oos | -0.013 |
eks | -0.013 |
cases | -0.012 |
ays | -0.012 |
elevator | -0.012 |
Token | The |
Token ID | 383 |
Feature activation | -7.06e-03 |
Pos logit contributions | |
Rib | +0.005 |
Ho | +0.005 |
accidentally | +0.004 |
Tw | +0.004 |
Ros | +0.004 |
Neg logit contributions | |
he | -0.003 |
hed | -0.003 |
ays | -0.003 |
cases | -0.003 |
agged | -0.003 |
Token | himself |
Token ID | 2241 |
Feature activation | +2.63e-02 |
Pos logit contributions | |
oos | +0.073 |
elevator | +0.072 |
eks | +0.072 |
he | +0.069 |
ases | +0.069 |
Neg logit contributions | |
Rib | -0.109 |
Ho | -0.109 |
H | -0.095 |
Butter | -0.090 |
Bub | -0.088 |
Token | in |
Token ID | 287 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
cases | +0.030 |
he | +0.028 |
oos | +0.028 |
elevator | +0.028 |
ays | +0.028 |
Neg logit contributions | |
Ho | -0.056 |
Rib | -0.056 |
H | -0.050 |
Bub | -0.048 |
Butter | -0.048 |
Token | the |
Token ID | 262 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
Ho | +0.034 |
Rib | +0.033 |
H | +0.030 |
Bub | +0.027 |
Bun | +0.027 |
Neg logit contributions | |
oos | -0.031 |
eks | -0.031 |
cases | -0.031 |
ays | -0.031 |
agged | -0.030 |
Token | mirror |
Token ID | 10162 |
Feature activation | -5.72e-02 |
Pos logit contributions | |
Rib | +0.020 |
Ho | +0.020 |
H | +0.018 |
Bub | +0.018 |
Bun | +0.017 |
Neg logit contributions | |
elevator | -0.014 |
oos | -0.013 |
eks | -0.013 |
he | -0.012 |
ags | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -4.50e-03 |
Pos logit contributions | |
Ho | +0.111 |
Rib | +0.108 |
H | +0.098 |
Bub | +0.093 |
Bun | +0.092 |
Neg logit contributions | |
oos | -0.078 |
cases | -0.078 |
eks | -0.077 |
elevator | -0.077 |
ays | -0.076 |
Token |
|
Token ID | 198 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
Rib | +0.008 |
Ho | +0.008 |
fle | +0.007 |
Tw | +0.007 |
onna | +0.007 |
Neg logit contributions | |
ays | -0.006 |
oos | -0.006 |
hed | -0.006 |
he | -0.006 |
cases | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
Ho | +0.033 |
Rib | +0.033 |
H | +0.028 |
Butter | +0.027 |
fle | +0.025 |
Neg logit contributions | |
he | -0.038 |
oos | -0.033 |
hed | -0.032 |
eks | -0.032 |
agged | -0.032 |
Token | His |
Token ID | 6653 |
Feature activation | +4.52e-02 |
Pos logit contributions | |
Ho | +0.036 |
Rib | +0.035 |
H | +0.031 |
Bub | +0.028 |
Butter | +0.028 |
Neg logit contributions | |
oos | -0.038 |
he | -0.038 |
eks | -0.037 |
cases | -0.037 |
elevator | -0.036 |
Token | Mom |
Token ID | 11254 |
Feature activation | +7.47e-02 |
Pos logit contributions | |
elevator | +0.065 |
tickets | +0.060 |
bags | +0.057 |
oos | +0.057 |
eks | +0.057 |
Neg logit contributions | |
Rib | -0.088 |
Ho | -0.084 |
H | -0.076 |
Butter | -0.075 |
Bub | -0.074 |
Token | my |
Token ID | 1820 |
Feature activation | -3.18e-02 |
Pos logit contributions | |
eks | +0.106 |
hed | +0.106 |
he | +0.104 |
elevator | +0.104 |
oos | +0.104 |
Neg logit contributions | |
Rib | -0.139 |
Ho | -0.138 |
H | -0.119 |
Butter | -0.115 |
Bub | -0.114 |
Token | thought |
Token ID | 1807 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
Rib | +0.059 |
Ho | +0.058 |
H | +0.051 |
Butter | +0.049 |
Bub | +0.048 |
Neg logit contributions | |
hed | -0.044 |
cases | -0.044 |
eks | -0.044 |
led | -0.043 |
oos | -0.043 |
Token | She |
Token ID | 1375 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
Ho | +0.016 |
Rib | +0.016 |
H | +0.014 |
Bub | +0.013 |
Bun | +0.013 |
Neg logit contributions | |
elevator | -0.015 |
oos | -0.014 |
ays | -0.014 |
cases | -0.014 |
eks | -0.014 |
Token | saw |
Token ID | 2497 |
Feature activation | +5.06e-02 |
Pos logit contributions | |
Ho | +0.027 |
Rib | +0.027 |
H | +0.024 |
Bun | +0.023 |
Bub | +0.023 |
Neg logit contributions | |
eks | -0.018 |
elevator | -0.017 |
oos | -0.017 |
cases | -0.017 |
led | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
agged | +0.108 |
eks | +0.106 |
oos | +0.105 |
ags | +0.102 |
cases | +0.101 |
Neg logit contributions | |
Ho | -0.067 |
Rib | -0.061 |
H | -0.056 |
Bub | -0.052 |
Teddy | -0.051 |
Token | dough |
Token ID | 15756 |
Feature activation | -8.63e-02 |
Pos logit contributions | |
Ho | +0.031 |
Rib | +0.030 |
H | +0.027 |
Bub | +0.025 |
accidentally | +0.025 |
Neg logit contributions | |
eks | -0.024 |
oos | -0.023 |
cases | -0.023 |
ases | -0.023 |
agged | -0.022 |
Token | on |
Token ID | 319 |
Feature activation | +8.73e-05 |
Pos logit contributions | |
Ho | +0.133 |
Rib | +0.129 |
H | +0.110 |
accidentally | +0.109 |
Lucky | +0.106 |
Neg logit contributions | |
eks | -0.152 |
agged | -0.151 |
oos | -0.147 |
cases | -0.145 |
ases | -0.141 |
Token | the |
Token ID | 262 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
oos | +0.000 |
elevator | +0.000 |
eks | +0.000 |
ays | +0.000 |
ases | +0.000 |
Neg logit contributions | |
Ho | -0.000 |
Rib | -0.000 |
H | -0.000 |
Bub | -0.000 |
Bun | -0.000 |
Token | table |
Token ID | 3084 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
Ho | +0.029 |
Rib | +0.028 |
H | +0.025 |
Bub | +0.024 |
Bun | +0.024 |
Neg logit contributions | |
eks | -0.024 |
oos | -0.023 |
agged | -0.023 |
ays | -0.023 |
ases | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
eks | +0.045 |
agged | +0.043 |
oos | +0.042 |
ued | +0.041 |
ases | +0.040 |
Neg logit contributions | |
Ho | -0.024 |
Rib | -0.023 |
accidentally | -0.021 |
H | -0.021 |
Tw | -0.018 |
Token | She |
Token ID | 1375 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
Ho | +0.021 |
Rib | +0.021 |
H | +0.018 |
Bun | +0.017 |
Bub | +0.017 |
Neg logit contributions | |
elevator | -0.017 |
ays | -0.017 |
oos | -0.017 |
cases | -0.017 |
he | -0.016 |
Token | saw |
Token ID | 2497 |
Feature activation | +5.02e-02 |
Pos logit contributions | |
Ho | +0.026 |
Rib | +0.026 |
H | +0.023 |
Bun | +0.021 |
Bub | +0.021 |
Neg logit contributions | |
eks | -0.024 |
elevator | -0.024 |
oos | -0.023 |
cases | -0.023 |
led | -0.023 |
Token | Lily |
Token ID | 20037 |
Feature activation | -4.58e-02 |
Pos logit contributions | |
agged | +0.105 |
eks | +0.104 |
oos | +0.103 |
ags | +0.100 |
cases | +0.099 |
Neg logit contributions | |
Ho | -0.067 |
Rib | -0.062 |
H | -0.057 |
Bub | -0.052 |
Teddy | -0.051 |
Token |
|
Token ID | 198 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
Rib | +0.003 |
Ho | +0.003 |
H | +0.003 |
accidentally | +0.003 |
Po | +0.003 |
Neg logit contributions | |
he | -0.003 |
oos | -0.003 |
elevator | -0.003 |
cases | -0.003 |
hed | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
Rib | +0.024 |
Ho | +0.024 |
H | +0.020 |
Butter | +0.019 |
Po | +0.018 |
Neg logit contributions | |
he | -0.032 |
eks | -0.029 |
oos | -0.028 |
hed | -0.028 |
cases | -0.028 |
Token | The |
Token ID | 464 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
Ho | +0.025 |
Rib | +0.024 |
H | +0.021 |
Bub | +0.019 |
Bun | +0.019 |
Neg logit contributions | |
eks | -0.032 |
oos | -0.032 |
cases | -0.031 |
elevator | -0.031 |
ases | -0.031 |
Token | people |
Token ID | 661 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
Ho | +0.016 |
Rib | +0.014 |
H | +0.014 |
Bub | +0.013 |
Bun | +0.013 |
Neg logit contributions | |
ays | -0.019 |
oos | -0.018 |
ases | -0.018 |
cases | -0.017 |
eks | -0.017 |
Token | of |
Token ID | 286 |
Feature activation | -4.28e-02 |
Pos logit contributions | |
oos | +0.051 |
ays | +0.051 |
ags | +0.051 |
agged | +0.049 |
ases | +0.048 |
Neg logit contributions | |
Ho | -0.046 |
Rib | -0.043 |
H | -0.042 |
accidentally | -0.042 |
Bub | -0.040 |
Token | the |
Token ID | 262 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
Ho | +0.063 |
Rib | +0.060 |
H | +0.054 |
Bub | +0.050 |
Bun | +0.049 |
Neg logit contributions | |
cases | -0.082 |
ays | -0.080 |
oos | -0.079 |
eks | -0.078 |
agged | -0.077 |
Token | city |
Token ID | 1748 |
Feature activation | +7.84e-02 |
Pos logit contributions | |
Ho | +0.017 |
H | +0.015 |
Rib | +0.015 |
Bun | +0.014 |
accidentally | +0.014 |
Neg logit contributions | |
ays | -0.019 |
oos | -0.018 |
ases | -0.018 |
agged | -0.017 |
eks | -0.017 |
Token | cheered |
Token ID | 34767 |
Feature activation | +4.00e-02 |
Pos logit contributions | |
ags | +0.145 |
agged | +0.143 |
ays | +0.143 |
ases | +0.143 |
eks | +0.143 |
Neg logit contributions | |
Ho | -0.108 |
Rib | -0.104 |
accidentally | -0.098 |
H | -0.098 |
Bub | -0.091 |
Token | and |
Token ID | 290 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
eks | +0.069 |
oos | +0.067 |
agged | +0.067 |
ays | +0.067 |
elevator | +0.067 |
Neg logit contributions | |
Ho | -0.065 |
Rib | -0.062 |
H | -0.058 |
Bun | -0.053 |
Bub | -0.052 |
Token | cl |
Token ID | 537 |
Feature activation | -2.76e-02 |
Pos logit contributions | |
eks | +0.018 |
oos | +0.018 |
elevator | +0.017 |
cases | +0.017 |
ases | +0.017 |
Neg logit contributions | |
Ho | -0.021 |
Rib | -0.021 |
H | -0.019 |
Bun | -0.017 |
Butter | -0.017 |
Token | apped |
Token ID | 6320 |
Feature activation | +5.24e-02 |
Pos logit contributions | |
Rib | +0.053 |
Ho | +0.051 |
H | +0.048 |
Bub | +0.046 |
Fro | +0.042 |
Neg logit contributions | |
elevator | -0.041 |
eks | -0.039 |
roads | -0.037 |
cases | -0.037 |
tickets | -0.036 |
Token | messy |
Token ID | 24097 |
Feature activation | +9.01e-03 |
Pos logit contributions | |
Ho | +0.035 |
Rib | +0.034 |
H | +0.032 |
Bun | +0.031 |
Bub | +0.030 |
Neg logit contributions | |
elevator | -0.022 |
cases | -0.020 |
eks | -0.020 |
he | -0.020 |
ays | -0.020 |
Token | pile |
Token ID | 14540 |
Feature activation | -1.64e-03 |
Pos logit contributions | |
eks | +0.013 |
cases | +0.013 |
oos | +0.013 |
ases | +0.013 |
ays | +0.013 |
Neg logit contributions | |
Ho | -0.016 |
Rib | -0.016 |
H | -0.014 |
accidentally | -0.014 |
Bub | -0.013 |
Token | of |
Token ID | 286 |
Feature activation | -4.43e-02 |
Pos logit contributions | |
Rib | +0.003 |
Ho | +0.003 |
H | +0.002 |
Po | +0.002 |
Tw | +0.002 |
Neg logit contributions | |
eks | -0.003 |
oos | -0.003 |
agged | -0.003 |
cases | -0.003 |
ags | -0.002 |
Token | flour |
Token ID | 10601 |
Feature activation | -6.66e-02 |
Pos logit contributions | |
Ho | +0.090 |
Rib | +0.088 |
H | +0.080 |
Bub | +0.076 |
Butter | +0.075 |
Neg logit contributions | |
eks | -0.056 |
oos | -0.056 |
cases | -0.054 |
ays | -0.054 |
agged | -0.054 |
Token | and |
Token ID | 290 |
Feature activation | +4.70e-03 |
Pos logit contributions | |
Ho | +0.099 |
Rib | +0.095 |
H | +0.080 |
accidentally | +0.074 |
Butter | +0.073 |
Neg logit contributions | |
eks | -0.125 |
agged | -0.120 |
oos | -0.120 |
ases | -0.116 |
ags | -0.112 |
Token | sugar |
Token ID | 7543 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
oos | +0.007 |
eks | +0.007 |
elevator | +0.006 |
cases | +0.006 |
ays | +0.006 |
Neg logit contributions | |
Ho | -0.009 |
Rib | -0.009 |
H | -0.008 |
Bun | -0.007 |
Bub | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
Ho | +0.233 |
Rib | +0.225 |
H | +0.197 |
Bub | +0.183 |
Butter | +0.181 |
Neg logit contributions | |
eks | -0.224 |
oos | -0.217 |
agged | -0.214 |
ases | -0.212 |
elevator | -0.209 |
Token | " |
Token ID | 366 |
Feature activation | +1.02e-03 |
Pos logit contributions | |
Rib | +0.026 |
Ho | +0.025 |
Tooth | +0.022 |
Bun | +0.022 |
please | +0.022 |
Neg logit contributions | |
ays | -0.019 |
oos | -0.017 |
ases | -0.016 |
agged | -0.016 |
ags | -0.015 |
Token | Do |
Token ID | 5211 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
eks | +0.002 |
oos | +0.002 |
ases | +0.002 |
elevator | +0.002 |
ays | +0.002 |
Neg logit contributions | |
Ho | -0.002 |
Rib | -0.002 |
H | -0.001 |
Bub | -0.001 |
Bun | -0.001 |
Token | you |
Token ID | 345 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
Ho | +0.182 |
Rib | +0.179 |
H | +0.161 |
Bun | +0.150 |
Bub | +0.150 |
Neg logit contributions | |
eks | -0.158 |
elevator | -0.155 |
oos | -0.153 |
cases | -0.151 |
ases | -0.150 |
Token | want |
Token ID | 765 |
Feature activation | +6.98e-02 |
Pos logit contributions | |
Ho | +0.039 |
Rib | +0.038 |
H | +0.035 |
Bun | +0.033 |
Bub | +0.033 |
Neg logit contributions | |
eks | -0.021 |
oos | -0.021 |
ases | -0.020 |
elevator | -0.020 |
ays | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
Ho | +0.071 |
Rib | +0.070 |
H | +0.064 |
Bun | +0.059 |
Bub | +0.059 |
Neg logit contributions | |
oos | -0.055 |
eks | -0.054 |
ays | -0.053 |
cases | -0.053 |
ases | -0.053 |
Token | They |
Token ID | 1119 |
Feature activation | +5.94e-02 |
Pos logit contributions | |
oos | +0.018 |
elevator | +0.017 |
cases | +0.017 |
ays | +0.017 |
agged | +0.016 |
Neg logit contributions | |
Rib | -0.020 |
Ho | -0.019 |
H | -0.016 |
Fro | -0.016 |
Stri | -0.016 |
Token | also |
Token ID | 635 |
Feature activation | -6.62e-02 |
Pos logit contributions | |
oos | +0.076 |
eks | +0.075 |
ays | +0.074 |
ases | +0.073 |
agged | +0.073 |
Neg logit contributions | |
Ho | -0.122 |
Rib | -0.118 |
H | -0.108 |
Bub | -0.102 |
Bun | -0.101 |
Token | got |
Token ID | 1392 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
Ho | +0.110 |
Rib | +0.107 |
H | +0.096 |
Bub | +0.088 |
accidentally | +0.088 |
Neg logit contributions | |
oos | -0.110 |
eks | -0.109 |
ays | -0.107 |
cases | -0.107 |
agged | -0.107 |
Token | some |
Token ID | 617 |
Feature activation | +8.11e-02 |
Pos logit contributions | |
eks | +0.027 |
oos | +0.026 |
elevator | +0.026 |
ays | +0.026 |
ases | +0.025 |
Neg logit contributions | |
Ho | -0.038 |
Rib | -0.037 |
H | -0.034 |
Bun | -0.032 |
Bub | -0.032 |
Token | milk |
Token ID | 7545 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
eks | +0.137 |
oos | +0.132 |
ays | +0.130 |
cases | +0.128 |
ases | +0.128 |
Neg logit contributions | |
Ho | -0.136 |
Rib | -0.130 |
H | -0.114 |
Bub | -0.110 |
Butter | -0.108 |
Token | and |
Token ID | 290 |
Feature activation | +1.86e-02 |
Pos logit contributions | |
Ho | +0.190 |
Rib | +0.179 |
H | +0.160 |
Lucky | +0.150 |
Bub | +0.149 |
Neg logit contributions | |
eks | -0.202 |
oos | -0.191 |
agged | -0.191 |
ases | -0.188 |
elevator | -0.186 |
Token | bread |
Token ID | 8509 |
Feature activation | -5.21e-02 |
Pos logit contributions | |
oos | +0.025 |
elevator | +0.025 |
eks | +0.025 |
cases | +0.025 |
led | +0.024 |
Neg logit contributions | |
Ho | -0.035 |
Rib | -0.035 |
H | -0.032 |
Bub | -0.029 |
Bun | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | +9.40e-03 |
Pos logit contributions | |
Ho | +0.093 |
Rib | +0.091 |
H | +0.082 |
Bub | +0.077 |
Bun | +0.076 |
Neg logit contributions | |
eks | -0.079 |
oos | -0.078 |
elevator | -0.076 |
ays | -0.076 |
cases | -0.076 |
Token |
|
Token ID | 198 |
Feature activation | -7.39e-03 |
Pos logit contributions | |
oos | +0.015 |
ays | +0.015 |
cases | +0.014 |
elevator | +0.014 |
hed | +0.014 |
Neg logit contributions | |
Rib | -0.016 |
Ho | -0.015 |
Fro | -0.013 |
H | -0.013 |
Stri | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -8.10e-03 |
Pos logit contributions | |
Rib | +0.012 |
Ho | +0.012 |
H | +0.010 |
Butter | +0.009 |
fle | +0.009 |
Neg logit contributions | |
he | -0.014 |
cases | -0.012 |
oos | -0.012 |
eks | -0.011 |
hed | -0.011 |
Token | cl |
Token ID | 537 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
oos | +0.121 |
ays | +0.121 |
ags | +0.120 |
eks | +0.120 |
agged | +0.119 |
Neg logit contributions | |
Ho | -0.075 |
Rib | -0.073 |
H | -0.062 |
accidentally | -0.058 |
Bun | -0.058 |
Token | apped |
Token ID | 6320 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
Rib | +0.068 |
Ho | +0.067 |
H | +0.061 |
Bub | +0.057 |
Lucky | +0.056 |
Neg logit contributions | |
elevator | -0.044 |
cases | -0.041 |
eks | -0.040 |
tickets | -0.039 |
roads | -0.039 |
Token | as |
Token ID | 355 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
eks | +0.080 |
oos | +0.079 |
ays | +0.078 |
cases | +0.077 |
elevator | +0.077 |
Neg logit contributions | |
Ho | -0.094 |
Rib | -0.092 |
H | -0.083 |
Bub | -0.077 |
Bun | -0.076 |
Token | the |
Token ID | 262 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
Ho | +0.053 |
Rib | +0.052 |
H | +0.046 |
Ros | +0.044 |
Bub | +0.043 |
Neg logit contributions | |
elevator | -0.038 |
he | -0.035 |
cases | -0.035 |
led | -0.035 |
oos | -0.035 |
Token | cow |
Token ID | 9875 |
Feature activation | -7.25e-02 |
Pos logit contributions | |
Ho | +0.020 |
Rib | +0.020 |
H | +0.018 |
Bub | +0.017 |
Bun | +0.017 |
Neg logit contributions | |
eks | -0.017 |
oos | -0.017 |
elevator | -0.017 |
cases | -0.017 |
ays | -0.017 |
Token | gently |
Token ID | 15165 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
Ho | +0.124 |
Rib | +0.122 |
H | +0.110 |
Bub | +0.101 |
Bun | +0.100 |
Neg logit contributions | |
oos | -0.113 |
eks | -0.113 |
cases | -0.111 |
elevator | -0.109 |
ays | -0.109 |
Token | carried |
Token ID | 5281 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
Ho | +0.192 |
Rib | +0.188 |
H | +0.169 |
Bub | +0.156 |
Bun | +0.154 |
Neg logit contributions | |
oos | -0.176 |
eks | -0.175 |
cases | -0.171 |
ays | -0.170 |
elevator | -0.170 |
Token | the |
Token ID | 262 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
Ho | +0.025 |
Rib | +0.025 |
H | +0.023 |
Bub | +0.021 |
accidentally | +0.021 |
Neg logit contributions | |
oos | -0.020 |
eks | -0.019 |
elevator | -0.019 |
ases | -0.019 |
cases | -0.018 |
Token | sheep |
Token ID | 15900 |
Feature activation | -3.47e-02 |
Pos logit contributions | |
Ho | +0.022 |
Rib | +0.021 |
H | +0.019 |
Bub | +0.018 |
Bun | +0.018 |
Neg logit contributions | |
elevator | -0.017 |
oos | -0.017 |
eks | -0.017 |
cases | -0.016 |
he | -0.016 |
Token | to |
Token ID | 284 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
Ho | +0.060 |
Rib | +0.057 |
H | +0.052 |
Bun | +0.049 |
Bub | +0.048 |
Neg logit contributions | |
eks | -0.056 |
agged | -0.055 |
oos | -0.055 |
ays | -0.054 |
ases | -0.053 |
Token | safety |
Token ID | 3747 |
Feature activation | +1.81e-03 |
Pos logit contributions | |
Ho | +0.035 |
Rib | +0.035 |
H | +0.031 |
Bub | +0.029 |
accidentally | +0.029 |
Neg logit contributions | |
eks | -0.025 |
oos | -0.025 |
elevator | -0.025 |
ases | -0.024 |
he | -0.024 |
Token | like |
Token ID | 588 |
Feature activation | -4.87e-03 |
Pos logit contributions | |
Ho | +0.019 |
Rib | +0.019 |
Ros | +0.017 |
H | +0.016 |
Bub | +0.016 |
Neg logit contributions | |
eks | -0.011 |
ps | -0.010 |
out | -0.010 |
led | -0.010 |
elevator | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
Rib | +0.009 |
Ho | +0.009 |
H | +0.008 |
Bub | +0.007 |
Ros | +0.007 |
Neg logit contributions | |
elevator | -0.007 |
oos | -0.007 |
eks | -0.006 |
ases | -0.006 |
ays | -0.006 |
Token | butterfly |
Token ID | 35113 |
Feature activation | -5.03e-02 |
Pos logit contributions | |
Ho | +0.020 |
Rib | +0.019 |
H | +0.017 |
Bub | +0.016 |
Bun | +0.016 |
Neg logit contributions | |
eks | -0.021 |
elevator | -0.021 |
oos | -0.021 |
ays | -0.020 |
he | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | +9.63e-03 |
Pos logit contributions | |
Ho | +0.097 |
Rib | +0.096 |
H | +0.086 |
Bub | +0.081 |
Stri | +0.078 |
Neg logit contributions | |
oos | -0.066 |
eks | -0.065 |
elevator | -0.064 |
ays | -0.064 |
cases | -0.064 |
Token | she |
Token ID | 673 |
Feature activation | -4.04e-02 |
Pos logit contributions | |
eks | +0.017 |
oos | +0.017 |
elevator | +0.016 |
ays | +0.016 |
cases | +0.016 |
Neg logit contributions | |
Ho | -0.015 |
Rib | -0.015 |
H | -0.013 |
Bub | -0.012 |
Bun | -0.012 |
Token | can |
Token ID | 460 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
Ho | +0.065 |
Rib | +0.064 |
H | +0.056 |
Bub | +0.053 |
Bun | +0.050 |
Neg logit contributions | |
eks | -0.069 |
oos | -0.067 |
led | -0.067 |
cases | -0.066 |
ays | -0.066 |
Token | 't |
Token ID | 470 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
Ho | +0.179 |
Rib | +0.175 |
H | +0.157 |
Bub | +0.145 |
Bun | +0.143 |
Neg logit contributions | |
eks | -0.162 |
oos | -0.160 |
ases | -0.156 |
he | -0.155 |
elevator | -0.154 |
Token | fly |
Token ID | 6129 |
Feature activation | -6.71e-02 |
Pos logit contributions | |
Ho | +0.033 |
Rib | +0.032 |
H | +0.028 |
Bub | +0.026 |
Stri | +0.026 |
Neg logit contributions | |
eks | -0.025 |
he | -0.024 |
oos | -0.024 |
ases | -0.023 |
elevator | -0.023 |
Token | away |
Token ID | 1497 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
Ho | +0.132 |
Rib | +0.130 |
H | +0.117 |
Bub | +0.110 |
Bun | +0.108 |
Neg logit contributions | |
eks | -0.089 |
oos | -0.089 |
ases | -0.085 |
ays | -0.085 |
elevator | -0.084 |
Token | whenever |
Token ID | 8797 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
eks | +0.052 |
oos | +0.052 |
ays | +0.051 |
cases | +0.050 |
elevator | +0.050 |
Neg logit contributions | |
Ho | -0.060 |
Rib | -0.059 |
H | -0.053 |
Bub | -0.049 |
Bun | -0.049 |
Token | she |
Token ID | 673 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
Ho | +0.025 |
Rib | +0.025 |
H | +0.021 |
Stri | +0.021 |
Bub | +0.020 |
Neg logit contributions | |
oos | -0.024 |
elevator | -0.024 |
eks | -0.023 |
are | -0.022 |
led | -0.022 |
Token | skirt |
Token ID | 23967 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
Rib | +0.070 |
Ho | +0.069 |
H | +0.061 |
Bub | +0.057 |
Butter | +0.056 |
Neg logit contributions | |
oos | -0.067 |
eks | -0.066 |
elevator | -0.066 |
cases | -0.064 |
ays | -0.064 |
Token | is |
Token ID | 318 |
Feature activation | +2.19e-02 |
Pos logit contributions | |
Ho | +0.024 |
Rib | +0.022 |
H | +0.021 |
accidentally | +0.020 |
Bun | +0.020 |
Neg logit contributions | |
agged | -0.020 |
oos | -0.019 |
eks | -0.019 |
ags | -0.019 |
ases | -0.018 |
Token | broken |
Token ID | 5445 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
elevator | +0.025 |
eks | +0.025 |
ays | +0.024 |
oos | +0.023 |
ps | +0.023 |
Neg logit contributions | |
Rib | -0.046 |
Ho | -0.045 |
H | -0.040 |
Bub | -0.038 |
Stri | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | -3.80e-03 |
Pos logit contributions | |
Ho | +0.035 |
Rib | +0.034 |
H | +0.031 |
Bun | +0.029 |
Bub | +0.028 |
Neg logit contributions | |
eks | -0.029 |
oos | -0.029 |
cases | -0.028 |
agged | -0.028 |
ases | -0.028 |
Token | What |
Token ID | 1867 |
Feature activation | -4.27e-04 |
Pos logit contributions | |
Rib | +0.007 |
Ho | +0.007 |
Fro | +0.006 |
accidentally | +0.006 |
Bub | +0.006 |
Neg logit contributions | |
ays | -0.005 |
elevator | -0.005 |
ases | -0.005 |
cases | -0.005 |
oos | -0.005 |
Token | can |
Token ID | 460 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
Ho | +0.001 |
Rib | +0.001 |
H | +0.001 |
Bub | +0.001 |
Bun | +0.001 |
Neg logit contributions | |
eks | -0.001 |
oos | -0.001 |
elevator | -0.001 |
ases | -0.001 |
ays | -0.001 |
Token | I |
Token ID | 314 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
Ho | +0.140 |
Rib | +0.136 |
H | +0.118 |
Bub | +0.106 |
Bun | +0.105 |
Neg logit contributions | |
eks | -0.202 |
oos | -0.202 |
ays | -0.197 |
elevator | -0.197 |
cases | -0.197 |
Token | do |
Token ID | 466 |
Feature activation | -5.11e-02 |
Pos logit contributions | |
Ho | +0.062 |
Rib | +0.062 |
H | +0.054 |
Butter | +0.050 |
Bub | +0.050 |
Neg logit contributions | |
eks | -0.045 |
he | -0.044 |
elevator | -0.043 |
oos | -0.043 |
ases | -0.043 |
Token | ?" |
Token ID | 1701 |
Feature activation | -6.26e-03 |
Pos logit contributions | |
Ho | +0.087 |
Rib | +0.084 |
H | +0.078 |
Bun | +0.072 |
Bub | +0.071 |
Neg logit contributions | |
eks | -0.082 |
oos | -0.081 |
cases | -0.080 |
ays | -0.079 |
agged | -0.078 |
Token | Her |
Token ID | 2332 |
Feature activation | -7.08e-02 |
Pos logit contributions | |
Rib | +0.012 |
Ho | +0.011 |
onna | +0.010 |
Stri | +0.010 |
die | +0.010 |
Neg logit contributions | |
oos | -0.008 |
agged | -0.008 |
he | -0.008 |
hed | -0.007 |
elevator | -0.007 |
Token | mom |
Token ID | 1995 |
Feature activation | +6.19e-02 |
Pos logit contributions | |
Rib | +0.148 |
Ho | +0.136 |
Butter | +0.127 |
onna | +0.125 |
Bub | +0.125 |
Neg logit contributions | |
elevator | -0.085 |
ags | -0.072 |
bags | -0.071 |
oos | -0.071 |
ays | -0.071 |