Token | accidentally |
Token ID | 14716 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
much | +0.063 |
lessly | +0.054 |
wonderful | +0.053 |
br | +0.052 |
On | +0.051 |
Neg logit contributions | |
Cr | -0.065 |
ados | -0.065 |
Majesty | -0.063 |
ishment | -0.062 |
Fro | -0.060 |
Token | broke |
Token ID | 6265 |
Feature activation | +9.88e-03 |
Pos logit contributions | |
ados | +0.025 |
Cr | +0.024 |
Majesty | +0.024 |
ishment | +0.023 |
lee | +0.022 |
Neg logit contributions | |
much | -0.018 |
lessly | -0.015 |
wonderful | -0.015 |
On | -0.014 |
br | -0.014 |
Token | it |
Token ID | 340 |
Feature activation | +4.37e-02 |
Pos logit contributions | |
much | +0.019 |
lessly | +0.016 |
wonderful | +0.016 |
On | +0.015 |
hard | +0.015 |
Neg logit contributions | |
Majesty | -0.017 |
Cr | -0.017 |
ados | -0.017 |
ishment | -0.017 |
icans | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | +3.06e-03 |
Pos logit contributions | |
wonderful | +0.078 |
br | +0.078 |
much | +0.077 |
In | +0.075 |
arest | +0.072 |
Neg logit contributions | |
ados | -0.071 |
gy | -0.067 |
stick | -0.063 |
Cr | -0.062 |
ishment | -0.062 |
Token | the |
Token ID | 262 |
Feature activation | +4.28e-02 |
Pos logit contributions | |
much | +0.005 |
wonderful | +0.004 |
lessly | +0.004 |
On | +0.004 |
br | +0.004 |
Neg logit contributions | |
ados | -0.006 |
Cr | -0.006 |
Majesty | -0.006 |
ishment | -0.006 |
lee | -0.005 |
Token | things |
Token ID | 1243 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
much | +0.070 |
lic | +0.065 |
br | +0.063 |
lessly | +0.062 |
On | +0.061 |
Neg logit contributions | |
ados | -0.081 |
Majesty | -0.077 |
knees | -0.075 |
keeping | -0.071 |
heads | -0.071 |
Token | inside |
Token ID | 2641 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
much | +0.153 |
On | +0.149 |
wonderful | +0.144 |
br | +0.143 |
One | +0.141 |
Neg logit contributions | |
ados | -0.186 |
Cr | -0.171 |
knees | -0.166 |
lee | -0.164 |
stick | -0.163 |
Token | the |
Token ID | 262 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
much | +0.046 |
lessly | +0.039 |
wonderful | +0.039 |
br | +0.039 |
On | +0.038 |
Neg logit contributions | |
ados | -0.080 |
Cr | -0.080 |
ishment | -0.077 |
lee | -0.075 |
gy | -0.073 |
Token | box |
Token ID | 3091 |
Feature activation | +6.12e-02 |
Pos logit contributions | |
much | +0.041 |
lessly | +0.039 |
ching | +0.038 |
lic | +0.037 |
br | +0.036 |
Neg logit contributions | |
ados | -0.030 |
Majesty | -0.027 |
ishment | -0.026 |
ASH | -0.025 |
keeping | -0.025 |
Token | got |
Token ID | 1392 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
much | +0.087 |
wonderful | +0.075 |
lessly | +0.072 |
On | +0.071 |
br | +0.070 |
Neg logit contributions | |
ados | -0.130 |
Cr | -0.129 |
gy | -0.126 |
ishment | -0.124 |
lee | -0.123 |
Token | ruined |
Token ID | 20484 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
much | +0.058 |
lessly | +0.056 |
br | +0.056 |
In | +0.055 |
On | +0.055 |
Neg logit contributions | |
Cr | -0.097 |
ados | -0.096 |
ishment | -0.089 |
gy | -0.089 |
lee | -0.087 |
Token |
|
Token ID | 198 |
Feature activation | +5.67e-02 |
Pos logit contributions | |
ados | +0.017 |
Majesty | +0.016 |
ishment | +0.016 |
Cr | +0.016 |
gy | +0.015 |
Neg logit contributions | |
much | -0.009 |
On | -0.007 |
lessly | -0.007 |
wonderful | -0.007 |
In | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +4.76e-02 |
Pos logit contributions | |
much | +0.091 |
lessly | +0.083 |
br | +0.083 |
wonderful | +0.081 |
ching | +0.073 |
Neg logit contributions | |
Cr | -0.106 |
Fro | -0.101 |
ados | -0.097 |
lee | -0.096 |
Rib | -0.096 |
Token | But |
Token ID | 1537 |
Feature activation | +3.52e-02 |
Pos logit contributions | |
much | +0.065 |
lessly | +0.057 |
wonderful | +0.055 |
br | +0.054 |
hard | +0.051 |
Neg logit contributions | |
Cr | -0.097 |
Fro | -0.090 |
ishment | -0.090 |
Rib | -0.090 |
ados | -0.089 |
Token | the |
Token ID | 262 |
Feature activation | +4.71e-02 |
Pos logit contributions | |
much | +0.059 |
wonderful | +0.048 |
On | +0.047 |
lessly | +0.046 |
hard | +0.046 |
Neg logit contributions | |
ados | -0.072 |
Majesty | -0.070 |
Cr | -0.068 |
ishment | -0.068 |
lee | -0.066 |
Token | shiny |
Token ID | 22441 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
much | +0.101 |
lessly | +0.096 |
ching | +0.090 |
br | +0.089 |
On | +0.086 |
Neg logit contributions | |
ados | -0.064 |
Majesty | -0.060 |
ishment | -0.058 |
Cr | -0.057 |
knees | -0.057 |
Token | things |
Token ID | 1243 |
Feature activation | +9.76e-02 |
Pos logit contributions | |
much | +0.019 |
lessly | +0.016 |
wonderful | +0.016 |
On | +0.016 |
In | +0.016 |
Neg logit contributions | |
ados | -0.020 |
Cr | -0.019 |
ishment | -0.019 |
Majesty | -0.019 |
gy | -0.018 |
Token | were |
Token ID | 547 |
Feature activation | +5.93e-02 |
Pos logit contributions | |
much | +0.149 |
On | +0.134 |
br | +0.131 |
Finally | +0.129 |
wonderful | +0.128 |
Neg logit contributions | |
ados | -0.182 |
Cr | -0.178 |
lee | -0.165 |
knees | -0.165 |
ishment | -0.164 |
Token | not |
Token ID | 407 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
arest | +0.082 |
lessly | +0.080 |
much | +0.077 |
Finally | +0.077 |
br | +0.077 |
Neg logit contributions | |
Cr | -0.118 |
Fro | -0.111 |
ados | -0.108 |
ishment | -0.105 |
Rib | -0.103 |
Token | stars |
Token ID | 5788 |
Feature activation | +6.79e-02 |
Pos logit contributions | |
much | +0.088 |
In | +0.088 |
Finally | +0.087 |
On | +0.087 |
Over | +0.085 |
Neg logit contributions | |
ados | -0.099 |
Cr | -0.099 |
ishment | -0.094 |
Harry | -0.092 |
gy | -0.088 |
Token | . |
Token ID | 13 |
Feature activation | -4.91e-03 |
Pos logit contributions | |
much | +0.119 |
On | +0.109 |
In | +0.108 |
lessly | +0.105 |
wonderful | +0.104 |
Neg logit contributions | |
ados | -0.118 |
Cr | -0.113 |
ishment | -0.113 |
gy | -0.112 |
lee | -0.107 |
Token | They |
Token ID | 1119 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
Majesty | +0.008 |
ados | +0.008 |
ci | +0.008 |
keeping | +0.007 |
gy | +0.007 |
Neg logit contributions | |
much | -0.011 |
On | -0.008 |
In | -0.008 |
lessly | -0.008 |
lot | -0.008 |
Token | was |
Token ID | 373 |
Feature activation | +5.47e-02 |
Pos logit contributions | |
much | +0.096 |
lessly | +0.081 |
wonderful | +0.081 |
br | +0.079 |
On | +0.079 |
Neg logit contributions | |
ados | -0.146 |
gy | -0.142 |
Cr | -0.140 |
ishment | -0.138 |
lee | -0.137 |
Token | covered |
Token ID | 5017 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
br | +0.080 |
arest | +0.078 |
lessly | +0.073 |
ching | +0.069 |
hest | +0.069 |
Neg logit contributions | |
gy | -0.097 |
ishment | -0.091 |
stick | -0.090 |
lee | -0.090 |
Cr | -0.088 |
Token | in |
Token ID | 287 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
much | +0.042 |
br | +0.036 |
wonderful | +0.036 |
lessly | +0.035 |
On | +0.035 |
Neg logit contributions | |
ados | -0.056 |
Cr | -0.054 |
ishment | -0.054 |
gy | -0.051 |
Majesty | -0.050 |
Token | flames |
Token ID | 16209 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
br | +0.030 |
lessly | +0.030 |
much | +0.029 |
In | +0.028 |
ching | +0.028 |
Neg logit contributions | |
ados | -0.028 |
ishment | -0.027 |
Cr | -0.026 |
Majesty | -0.025 |
keeping | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | +4.76e-03 |
Pos logit contributions | |
much | +0.077 |
In | +0.070 |
wonderful | +0.068 |
br | +0.067 |
On | +0.067 |
Neg logit contributions | |
ados | -0.078 |
ishment | -0.076 |
gy | -0.073 |
Cr | -0.073 |
Majesty | -0.072 |
Token | everything |
Token ID | 2279 |
Feature activation | +9.68e-02 |
Pos logit contributions | |
much | +0.008 |
br | +0.007 |
lessly | +0.007 |
On | +0.007 |
wonderful | +0.007 |
Neg logit contributions | |
ados | -0.009 |
Cr | -0.008 |
Majesty | -0.008 |
ishment | -0.008 |
lee | -0.008 |
Token | was |
Token ID | 373 |
Feature activation | +5.66e-02 |
Pos logit contributions | |
On | +0.142 |
much | +0.142 |
br | +0.135 |
Finally | +0.134 |
In | +0.133 |
Neg logit contributions | |
ishment | -0.152 |
lee | -0.152 |
ados | -0.151 |
ci | -0.150 |
Cr | -0.150 |
Token | gone |
Token ID | 3750 |
Feature activation | +3.02e-02 |
Pos logit contributions | |
br | +0.087 |
arest | +0.081 |
Over | +0.080 |
cro | +0.075 |
lessly | +0.073 |
Neg logit contributions | |
gy | -0.099 |
ishment | -0.098 |
lee | -0.096 |
Harry | -0.092 |
urg | -0.091 |
Token | . |
Token ID | 13 |
Feature activation | +7.30e-03 |
Pos logit contributions | |
much | +0.053 |
br | +0.052 |
In | +0.052 |
On | +0.049 |
Over | +0.049 |
Neg logit contributions | |
ishment | -0.049 |
gy | -0.048 |
ados | -0.046 |
Majesty | -0.043 |
keeping | -0.043 |
Token | |
Token ID | 220 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
much | +0.015 |
lessly | +0.012 |
wonderful | +0.012 |
On | +0.012 |
In | +0.012 |
Neg logit contributions | |
ados | -0.013 |
Majesty | -0.012 |
ishment | -0.012 |
Cr | -0.012 |
icans | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +6.43e-02 |
Pos logit contributions | |
much | +0.037 |
wonderful | +0.029 |
lessly | +0.029 |
On | +0.028 |
br | +0.028 |
Neg logit contributions | |
ados | -0.063 |
Cr | -0.062 |
Majesty | -0.061 |
ishment | -0.061 |
lee | -0.058 |
Token | was |
Token ID | 373 |
Feature activation | +4.67e-02 |
Pos logit contributions | |
much | +0.056 |
lessly | +0.044 |
wonderful | +0.044 |
On | +0.042 |
br | +0.042 |
Neg logit contributions | |
ados | -0.082 |
Cr | -0.082 |
ishment | -0.081 |
Majesty | -0.081 |
gy | -0.080 |
Token | gone |
Token ID | 3750 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
arest | +0.072 |
br | +0.071 |
cro | +0.068 |
lessly | +0.065 |
hest | +0.061 |
Neg logit contributions | |
gy | -0.080 |
Harry | -0.079 |
ishment | -0.078 |
Cr | -0.078 |
Fro | -0.076 |
Token | , |
Token ID | 11 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
br | +0.055 |
much | +0.054 |
In | +0.053 |
arest | +0.053 |
lic | +0.051 |
Neg logit contributions | |
gy | -0.037 |
ishment | -0.036 |
Harry | -0.033 |
ados | -0.031 |
Sammy | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
much | +0.031 |
br | +0.028 |
lessly | +0.027 |
wonderful | +0.026 |
In | +0.026 |
Neg logit contributions | |
Cr | -0.032 |
ishment | -0.031 |
ados | -0.031 |
Majesty | -0.030 |
Fro | -0.029 |
Token | suddenly |
Token ID | 6451 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
much | +0.027 |
br | +0.023 |
lessly | +0.023 |
wonderful | +0.023 |
On | +0.023 |
Neg logit contributions | |
Cr | -0.023 |
ados | -0.023 |
ishment | -0.022 |
Majesty | -0.022 |
lee | -0.021 |
Token | everything |
Token ID | 2279 |
Feature activation | +9.67e-02 |
Pos logit contributions | |
much | +0.059 |
wonderful | +0.049 |
lessly | +0.048 |
br | +0.047 |
On | +0.047 |
Neg logit contributions | |
ados | -0.066 |
Cr | -0.065 |
ishment | -0.063 |
Majesty | -0.063 |
lee | -0.059 |
Token | felt |
Token ID | 2936 |
Feature activation | +5.97e-02 |
Pos logit contributions | |
much | +0.157 |
br | +0.145 |
On | +0.139 |
lessly | +0.134 |
Finally | +0.133 |
Neg logit contributions | |
Cr | -0.159 |
lee | -0.159 |
knees | -0.157 |
ishment | -0.155 |
ados | -0.149 |
Token | good |
Token ID | 922 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
br | +0.085 |
arest | +0.084 |
cro | +0.079 |
ris | +0.076 |
Finally | +0.076 |
Neg logit contributions | |
Nem | -0.108 |
Cr | -0.108 |
D | -0.103 |
Fro | -0.100 |
Sammy | -0.098 |
Token | . |
Token ID | 13 |
Feature activation | +8.93e-03 |
Pos logit contributions | |
much | +0.023 |
On | +0.021 |
In | +0.020 |
wonderful | +0.020 |
Finally | +0.020 |
Neg logit contributions | |
ados | -0.027 |
Cr | -0.026 |
lee | -0.025 |
ishment | -0.025 |
Fro | -0.023 |
Token | Joe |
Token ID | 5689 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
much | +0.019 |
wonderful | +0.016 |
lessly | +0.016 |
In | +0.016 |
On | +0.015 |
Neg logit contributions | |
ados | -0.015 |
Majesty | -0.014 |
ishment | -0.014 |
Cr | -0.014 |
Rib | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | -2.55e-02 |
Pos logit contributions | |
much | +0.019 |
wonderful | +0.016 |
lessly | +0.016 |
On | +0.015 |
In | +0.015 |
Neg logit contributions | |
ados | -0.026 |
Cr | -0.025 |
Majesty | -0.025 |
ishment | -0.025 |
icans | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +3.02e-02 |
Pos logit contributions | |
much | +0.039 |
lessly | +0.034 |
br | +0.033 |
On | +0.033 |
In | +0.033 |
Neg logit contributions | |
ados | -0.044 |
Cr | -0.043 |
ishment | -0.042 |
Majesty | -0.042 |
lee | -0.039 |
Token | dots |
Token ID | 22969 |
Feature activation | +7.82e-02 |
Pos logit contributions | |
much | +0.055 |
lessly | +0.050 |
br | +0.049 |
On | +0.048 |
lic | +0.047 |
Neg logit contributions | |
Majesty | -0.053 |
Cr | -0.052 |
ishment | -0.051 |
ados | -0.051 |
Harry | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | -9.73e-03 |
Pos logit contributions | |
much | +0.124 |
wonderful | +0.111 |
In | +0.110 |
On | +0.108 |
br | +0.105 |
Neg logit contributions | |
ados | -0.149 |
Cr | -0.141 |
ishment | -0.139 |
Majesty | -0.137 |
lee | -0.136 |
Token | But |
Token ID | 887 |
Feature activation | +4.44e-02 |
Pos logit contributions | |
ados | +0.016 |
Cr | +0.016 |
Majesty | +0.015 |
ishment | +0.015 |
lee | +0.014 |
Neg logit contributions | |
much | -0.021 |
On | -0.018 |
In | -0.017 |
wonderful | -0.017 |
lessly | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | +3.92e-02 |
Pos logit contributions | |
much | +0.076 |
lessly | +0.066 |
wonderful | +0.065 |
br | +0.064 |
On | +0.062 |
Neg logit contributions | |
Cr | -0.084 |
ados | -0.084 |
ishment | -0.081 |
Majesty | -0.080 |
Fro | -0.076 |
Token | dots |
Token ID | 22969 |
Feature activation | +9.66e-02 |
Pos logit contributions | |
much | +0.076 |
lessly | +0.067 |
On | +0.064 |
br | +0.064 |
In | +0.064 |
Neg logit contributions | |
Majesty | -0.067 |
ados | -0.067 |
Cr | -0.066 |
ishment | -0.065 |
Fro | -0.058 |
Token | were |
Token ID | 547 |
Feature activation | +5.55e-02 |
Pos logit contributions | |
much | +0.141 |
wonderful | +0.123 |
On | +0.116 |
br | +0.112 |
lessly | +0.112 |
Neg logit contributions | |
ados | -0.192 |
Cr | -0.185 |
lee | -0.182 |
Fro | -0.180 |
knees | -0.179 |
Token | hard |
Token ID | 1327 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
arest | +0.081 |
br | +0.080 |
much | +0.076 |
ching | +0.075 |
lessly | +0.075 |
Neg logit contributions | |
ishment | -0.101 |
Cr | -0.099 |
ados | -0.098 |
Fro | -0.097 |
lee | -0.093 |
Token | to |
Token ID | 284 |
Feature activation | -2.39e-04 |
Pos logit contributions | |
much | +0.023 |
In | +0.022 |
wonderful | +0.022 |
Over | +0.021 |
arest | +0.021 |
Neg logit contributions | |
ados | -0.015 |
ishment | -0.014 |
gy | -0.013 |
Cr | -0.013 |
keeping | -0.013 |
Token | see |
Token ID | 766 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
Cr | +0.000 |
ados | +0.000 |
ishment | +0.000 |
gy | +0.000 |
Majesty | +0.000 |
Neg logit contributions | |
much | -0.000 |
lessly | -0.000 |
wonderful | -0.000 |
br | -0.000 |
In | -0.000 |
Token | and |
Token ID | 290 |
Feature activation | -4.65e-04 |
Pos logit contributions | |
much | +0.047 |
lessly | +0.042 |
On | +0.041 |
In | +0.040 |
wonderful | +0.040 |
Neg logit contributions | |
ados | -0.050 |
Cr | -0.048 |
ishment | -0.048 |
Majesty | -0.048 |
lee | -0.045 |
Token | are |
Token ID | 389 |
Feature activation | +7.05e-02 |
Pos logit contributions | |
much | +0.105 |
wonderful | +0.101 |
br | +0.094 |
On | +0.091 |
lessly | +0.090 |
Neg logit contributions | |
ados | -0.129 |
ishment | -0.117 |
heads | -0.117 |
Cr | -0.116 |
knees | -0.116 |
Token | too |
Token ID | 1165 |
Feature activation | +8.38e-02 |
Pos logit contributions | |
arest | +0.115 |
ching | +0.108 |
br | +0.104 |
hest | +0.099 |
lessly | +0.099 |
Neg logit contributions | |
gy | -0.112 |
Cr | -0.107 |
ados | -0.107 |
heads | -0.103 |
ishment | -0.102 |
Token | full |
Token ID | 1336 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
lessly | +0.096 |
br | +0.095 |
arest | +0.092 |
hest | +0.091 |
ching | +0.090 |
Neg logit contributions | |
gy | -0.157 |
ados | -0.150 |
knees | -0.147 |
ishment | -0.144 |
lee | -0.144 |
Token | . |
Token ID | 13 |
Feature activation | -6.08e-03 |
Pos logit contributions | |
much | +0.041 |
wonderful | +0.038 |
lessly | +0.036 |
In | +0.036 |
arest | +0.035 |
Neg logit contributions | |
ados | -0.051 |
Cr | -0.048 |
ishment | -0.046 |
lee | -0.046 |
Fro | -0.045 |
Token | Their |
Token ID | 5334 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
Majesty | +0.010 |
ados | +0.009 |
Cr | +0.009 |
ishment | +0.009 |
ci | +0.008 |
Neg logit contributions | |
much | -0.013 |
On | -0.011 |
lessly | -0.011 |
In | -0.011 |
lot | -0.011 |
Token | things |
Token ID | 1243 |
Feature activation | +9.62e-02 |
Pos logit contributions | |
much | +0.023 |
wonderful | +0.020 |
hard | +0.019 |
lessly | +0.019 |
On | +0.019 |
Neg logit contributions | |
Cr | -0.019 |
ados | -0.018 |
ishment | -0.018 |
Majesty | -0.017 |
Fro | -0.016 |
Token | do |
Token ID | 466 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
much | +0.152 |
br | +0.146 |
lic | +0.145 |
On | +0.144 |
Finally | +0.142 |
Neg logit contributions | |
ados | -0.154 |
heads | -0.153 |
knees | -0.152 |
ishment | -0.139 |
lee | -0.138 |
Token | not |
Token ID | 407 |
Feature activation | +4.05e-02 |
Pos logit contributions | |
lessly | +0.056 |
much | +0.054 |
wonderful | +0.049 |
br | +0.049 |
importantly | +0.048 |
Neg logit contributions | |
gy | -0.059 |
Cr | -0.055 |
ishment | -0.052 |
Av | -0.050 |
ci | -0.050 |
Token | fit |
Token ID | 4197 |
Feature activation | +1.75e-03 |
Pos logit contributions | |
much | +0.072 |
ching | +0.071 |
br | +0.070 |
million | +0.068 |
In | +0.068 |
Neg logit contributions | |
Cr | -0.051 |
ados | -0.048 |
ishment | -0.047 |
Harry | -0.046 |
gy | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | -2.12e-03 |
Pos logit contributions | |
much | +0.003 |
On | +0.002 |
wonderful | +0.002 |
lessly | +0.002 |
In | +0.002 |
Neg logit contributions | |
ados | -0.003 |
Cr | -0.003 |
ishment | -0.003 |
Majesty | -0.003 |
lee | -0.003 |
Token | Sara |
Token ID | 24799 |
Feature activation | +2.42e-03 |
Pos logit contributions | |
Majesty | +0.003 |
ados | +0.003 |
Cr | +0.003 |
ishment | +0.003 |
lee | +0.003 |
Neg logit contributions | |
much | -0.005 |
lessly | -0.004 |
On | -0.004 |
wonderful | -0.004 |
In | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +1.56e-03 |
Pos logit contributions | |
much | +0.021 |
wonderful | +0.019 |
In | +0.019 |
On | +0.018 |
br | +0.018 |
Neg logit contributions | |
ados | -0.019 |
Cr | -0.018 |
ishment | -0.018 |
Majesty | -0.017 |
lee | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | +6.07e-02 |
Pos logit contributions | |
much | +0.004 |
wonderful | +0.003 |
lessly | +0.003 |
br | +0.003 |
On | +0.003 |
Neg logit contributions | |
Cr | -0.002 |
ados | -0.002 |
ishment | -0.002 |
Majesty | -0.002 |
Fro | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +5.60e-02 |
Pos logit contributions | |
much | +0.099 |
lessly | +0.099 |
wonderful | +0.098 |
br | +0.097 |
ching | +0.089 |
Neg logit contributions | |
Cr | -0.106 |
Fro | -0.097 |
Rib | -0.093 |
gy | -0.091 |
lee | -0.090 |
Token | Then |
Token ID | 6423 |
Feature activation | +4.68e-02 |
Pos logit contributions | |
lessly | +0.073 |
much | +0.073 |
wonderful | +0.073 |
iest | +0.061 |
br | +0.061 |
Neg logit contributions | |
Cr | -0.116 |
Rib | -0.105 |
Fro | -0.103 |
gy | -0.101 |
Majesty | -0.097 |
Token | , |
Token ID | 11 |
Feature activation | +4.60e-02 |
Pos logit contributions | |
much | +0.087 |
lessly | +0.077 |
wonderful | +0.075 |
On | +0.074 |
In | +0.072 |
Neg logit contributions | |
Cr | -0.082 |
ados | -0.082 |
Majesty | -0.079 |
ishment | -0.078 |
Fro | -0.074 |
Token | something |
Token ID | 1223 |
Feature activation | +9.52e-02 |
Pos logit contributions | |
much | +0.086 |
lessly | +0.079 |
ching | +0.076 |
wonderful | +0.076 |
br | +0.075 |
Neg logit contributions | |
Cr | -0.080 |
ados | -0.072 |
Majesty | -0.072 |
ishment | -0.071 |
Fro | -0.071 |
Token | bad |
Token ID | 2089 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
ifying | +0.124 |
lessly | +0.121 |
On | +0.110 |
much | +0.110 |
arest | +0.108 |
Neg logit contributions | |
Cr | -0.170 |
Fro | -0.168 |
Nem | -0.158 |
slur | -0.155 |
Freddy | -0.154 |
Token | happened |
Token ID | 3022 |
Feature activation | +7.14e-03 |
Pos logit contributions | |
much | +0.035 |
wonderful | +0.029 |
lessly | +0.029 |
On | +0.029 |
In | +0.027 |
Neg logit contributions | |
Cr | -0.047 |
ados | -0.046 |
ishment | -0.043 |
Fro | -0.043 |
lee | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | +8.82e-03 |
Pos logit contributions | |
much | +0.012 |
wonderful | +0.011 |
lessly | +0.011 |
br | +0.011 |
In | +0.011 |
Neg logit contributions | |
Cr | -0.012 |
ados | -0.012 |
gy | -0.012 |
Majesty | -0.012 |
ishment | -0.012 |
Token | Z |
Token ID | 1168 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
much | +0.020 |
lessly | +0.017 |
wonderful | +0.017 |
On | +0.017 |
br | +0.017 |
Neg logit contributions | |
ados | -0.012 |
Majesty | -0.012 |
ishment | -0.012 |
Cr | -0.012 |
lee | -0.011 |
Token | ee |
Token ID | 1453 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
ados | +0.020 |
Cr | +0.020 |
Majesty | +0.019 |
ishment | +0.019 |
lee | +0.018 |
Neg logit contributions | |
much | -0.020 |
wonderful | -0.017 |
lessly | -0.017 |
On | -0.017 |
br | -0.017 |
Token | When |
Token ID | 1649 |
Feature activation | +5.89e-02 |
Pos logit contributions | |
much | +0.007 |
wonderful | +0.006 |
lessly | +0.006 |
br | +0.006 |
On | +0.006 |
Neg logit contributions | |
ados | -0.005 |
Cr | -0.005 |
Majesty | -0.005 |
ishment | -0.005 |
lee | -0.005 |
Token | they |
Token ID | 484 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
much | +0.108 |
lessly | +0.102 |
br | +0.102 |
ching | +0.099 |
On | +0.098 |
Neg logit contributions | |
Cr | -0.093 |
ados | -0.092 |
Majesty | -0.084 |
lee | -0.084 |
ishment | -0.082 |
Token | got |
Token ID | 1392 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
much | +0.057 |
lessly | +0.055 |
br | +0.054 |
ching | +0.050 |
valuable | +0.050 |
Neg logit contributions | |
Cr | -0.057 |
Fro | -0.055 |
lee | -0.052 |
Av | -0.049 |
D | -0.048 |
Token | there |
Token ID | 612 |
Feature activation | +4.43e-02 |
Pos logit contributions | |
much | +0.076 |
wonderful | +0.065 |
lessly | +0.064 |
On | +0.063 |
In | +0.062 |
Neg logit contributions | |
ados | -0.057 |
Majesty | -0.056 |
Cr | -0.056 |
ishment | -0.055 |
lee | -0.049 |
Token | , |
Token ID | 11 |
Feature activation | +3.52e-02 |
Pos logit contributions | |
much | +0.082 |
lessly | +0.070 |
wonderful | +0.070 |
br | +0.068 |
On | +0.068 |
Neg logit contributions | |
ados | -0.079 |
Cr | -0.079 |
Majesty | -0.076 |
ishment | -0.076 |
lee | -0.071 |
Token | everything |
Token ID | 2279 |
Feature activation | +9.48e-02 |
Pos logit contributions | |
much | +0.076 |
br | +0.074 |
lessly | +0.071 |
ching | +0.070 |
On | +0.068 |
Neg logit contributions | |
ados | -0.046 |
Cr | -0.045 |
ishment | -0.044 |
Majesty | -0.041 |
gy | -0.039 |
Token | was |
Token ID | 373 |
Feature activation | +5.89e-02 |
Pos logit contributions | |
much | +0.143 |
br | +0.127 |
On | +0.126 |
lessly | +0.122 |
He | +0.118 |
Neg logit contributions | |
lee | -0.168 |
Cr | -0.166 |
ci | -0.164 |
ishment | -0.164 |
knees | -0.158 |
Token | safe |
Token ID | 3338 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
cro | +0.100 |
br | +0.099 |
arest | +0.095 |
lessly | +0.086 |
vere | +0.083 |
Neg logit contributions | |
Twe | -0.095 |
gy | -0.093 |
Harry | -0.091 |
ishment | -0.091 |
Sammy | -0.088 |
Token | and |
Token ID | 290 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
much | +0.047 |
br | +0.043 |
On | +0.042 |
In | +0.041 |
wonderful | +0.040 |
Neg logit contributions | |
ados | -0.038 |
ishment | -0.037 |
Cr | -0.036 |
gy | -0.036 |
lee | -0.035 |
Token | Lucy |
Token ID | 22162 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
br | +0.040 |
ching | +0.038 |
much | +0.037 |
lessly | +0.036 |
arest | +0.035 |
Neg logit contributions | |
ishment | -0.053 |
Cr | -0.051 |
ados | -0.048 |
ci | -0.048 |
lee | -0.047 |
Token | felt |
Token ID | 2936 |
Feature activation | +6.92e-02 |
Pos logit contributions | |
much | +0.019 |
wonderful | +0.016 |
lessly | +0.015 |
On | +0.015 |
br | +0.015 |
Neg logit contributions | |
ados | -0.032 |
Cr | -0.031 |
Majesty | -0.031 |
ishment | -0.031 |
lee | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | +6.03e-02 |
Pos logit contributions | |
Cr | +0.010 |
ados | +0.009 |
Majesty | +0.009 |
ishment | +0.009 |
Fro | +0.008 |
Neg logit contributions | |
much | -0.011 |
wonderful | -0.010 |
lessly | -0.010 |
br | -0.010 |
On | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +5.39e-02 |
Pos logit contributions | |
much | +0.090 |
lessly | +0.084 |
br | +0.083 |
wonderful | +0.079 |
On | +0.073 |
Neg logit contributions | |
Cr | -0.120 |
Fro | -0.114 |
ados | -0.110 |
lee | -0.109 |
Majesty | -0.107 |
Token | But |
Token ID | 1537 |
Feature activation | +4.28e-02 |
Pos logit contributions | |
much | +0.085 |
lessly | +0.085 |
wonderful | +0.082 |
br | +0.079 |
ier | +0.078 |
Neg logit contributions | |
Cr | -0.096 |
Fro | -0.088 |
roc | -0.082 |
Harry | -0.082 |
ados | -0.081 |
Token | then |
Token ID | 788 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
much | +0.088 |
lessly | +0.082 |
br | +0.081 |
wonderful | +0.080 |
ching | +0.078 |
Neg logit contributions | |
Cr | -0.067 |
ados | -0.061 |
Fro | -0.059 |
ishment | -0.058 |
Majesty | -0.056 |
Token | , |
Token ID | 11 |
Feature activation | +4.76e-02 |
Pos logit contributions | |
much | +0.054 |
wonderful | +0.044 |
hard | +0.042 |
br | +0.042 |
On | +0.041 |
Neg logit contributions | |
ados | -0.066 |
ishment | -0.066 |
Majesty | -0.064 |
Cr | -0.064 |
lee | -0.061 |
Token | something |
Token ID | 1223 |
Feature activation | +9.42e-02 |
Pos logit contributions | |
much | +0.093 |
lessly | +0.084 |
wonderful | +0.081 |
br | +0.081 |
On | +0.079 |
Neg logit contributions | |
Cr | -0.080 |
ados | -0.078 |
ishment | -0.073 |
Majesty | -0.073 |
Fro | -0.071 |
Token | bad |
Token ID | 2089 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
ifying | +0.111 |
lessly | +0.104 |
million | +0.103 |
arest | +0.100 |
On | +0.096 |
Neg logit contributions | |
Cr | -0.182 |
Fro | -0.174 |
Nem | -0.164 |
Freddy | -0.163 |
lee | -0.162 |
Token | happened |
Token ID | 3022 |
Feature activation | +7.53e-04 |
Pos logit contributions | |
much | +0.025 |
wonderful | +0.021 |
lessly | +0.021 |
On | +0.021 |
In | +0.020 |
Neg logit contributions | |
Cr | -0.035 |
ados | -0.034 |
Fro | -0.031 |
ishment | -0.031 |
lee | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +7.18e-03 |
Pos logit contributions | |
much | +0.001 |
wonderful | +0.001 |
lessly | +0.001 |
br | +0.001 |
In | +0.001 |
Neg logit contributions | |
Cr | -0.001 |
ados | -0.001 |
ishment | -0.001 |
Majesty | -0.001 |
gy | -0.001 |
Token | Tim |
Token ID | 5045 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
much | +0.017 |
On | +0.014 |
In | +0.014 |
lessly | +0.014 |
br | +0.013 |
Neg logit contributions | |
Majesty | -0.011 |
ishment | -0.010 |
ados | -0.010 |
ishes | -0.009 |
ans | -0.009 |
Token | hit |
Token ID | 2277 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
much | +0.034 |
lessly | +0.030 |
wonderful | +0.030 |
br | +0.028 |
On | +0.027 |
Neg logit contributions | |
ados | -0.036 |
lee | -0.035 |
Cr | -0.035 |
ishment | -0.035 |
gy | -0.035 |
Token | his |
Token ID | 465 |
Feature activation | +8.06e-03 |
Pos logit contributions | |
ados | +0.020 |
Majesty | +0.020 |
Cr | +0.020 |
ishment | +0.020 |
lee | +0.018 |
Neg logit contributions | |
much | -0.023 |
wonderful | -0.020 |
lessly | -0.019 |
On | -0.019 |
hard | -0.019 |
Token | carrot |
Token ID | 40562 |
Feature activation | +6.12e-03 |
Pos logit contributions | |
much | +0.016 |
wonderful | +0.014 |
lessly | +0.014 |
On | +0.013 |
br | +0.013 |
Neg logit contributions | |
ados | -0.013 |
Cr | -0.013 |
Majesty | -0.013 |
ishment | -0.013 |
lee | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
much | +0.009 |
wonderful | +0.007 |
lessly | +0.007 |
hard | +0.007 |
On | +0.007 |
Neg logit contributions | |
Cr | -0.013 |
Majesty | -0.013 |
ados | -0.013 |
ishment | -0.013 |
lee | -0.012 |
Token | But |
Token ID | 887 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
ishment | +0.022 |
ados | +0.022 |
Majesty | +0.021 |
Cr | +0.021 |
ican | +0.021 |
Neg logit contributions | |
much | -0.020 |
lessly | -0.016 |
In | -0.016 |
hard | -0.015 |
On | -0.015 |
Token | then |
Token ID | 788 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
much | +0.071 |
lessly | +0.062 |
wonderful | +0.062 |
br | +0.061 |
On | +0.059 |
Neg logit contributions | |
Cr | -0.072 |
ados | -0.071 |
ishment | -0.068 |
Majesty | -0.068 |
Fro | -0.065 |
Token | something |
Token ID | 1223 |
Feature activation | +9.38e-02 |
Pos logit contributions | |
much | +0.039 |
wonderful | +0.031 |
lessly | +0.030 |
hard | +0.029 |
br | +0.029 |
Neg logit contributions | |
ados | -0.055 |
ishment | -0.054 |
Cr | -0.054 |
Majesty | -0.053 |
lee | -0.050 |
Token | strange |
Token ID | 6283 |
Feature activation | +3.77e-02 |
Pos logit contributions | |
lessly | +0.094 |
much | +0.092 |
On | +0.088 |
ifying | +0.088 |
arest | +0.087 |
Neg logit contributions | |
Cr | -0.204 |
Fro | -0.200 |
Nem | -0.190 |
Freddy | -0.186 |
lee | -0.184 |
Token | happened |
Token ID | 3022 |
Feature activation | +1.46e-03 |
Pos logit contributions | |
much | +0.065 |
lessly | +0.054 |
wonderful | +0.052 |
On | +0.052 |
Finally | +0.052 |
Neg logit contributions | |
Cr | -0.069 |
ados | -0.066 |
lee | -0.065 |
ishment | -0.065 |
Harry | -0.064 |
Token | . |
Token ID | 13 |
Feature activation | +6.28e-03 |
Pos logit contributions | |
much | +0.002 |
br | +0.002 |
wonderful | +0.002 |
lessly | +0.002 |
In | +0.002 |
Neg logit contributions | |
Cr | -0.003 |
ados | -0.003 |
ishment | -0.003 |
Majesty | -0.003 |
Fro | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +6.09e-02 |
Pos logit contributions | |
much | +0.014 |
In | +0.011 |
On | +0.011 |
lessly | +0.011 |
wonderful | +0.011 |
Neg logit contributions | |
Majesty | -0.010 |
ishment | -0.010 |
ados | -0.010 |
Cr | -0.009 |
ci | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +5.30e-02 |
Pos logit contributions | |
much | +0.098 |
lessly | +0.094 |
wonderful | +0.093 |
br | +0.089 |
terrible | +0.083 |
Neg logit contributions | |
Cr | -0.108 |
Fro | -0.102 |
ados | -0.098 |
Rib | -0.096 |
lee | -0.096 |
Token | After |
Token ID | 2293 |
Feature activation | +4.53e-02 |
Pos logit contributions | |
ados | +0.001 |
Cr | +0.000 |
Majesty | +0.000 |
ishment | +0.000 |
lee | +0.000 |
Neg logit contributions | |
much | -0.001 |
wonderful | -0.001 |
lessly | -0.001 |
On | -0.001 |
br | -0.001 |
Token | a |
Token ID | 257 |
Feature activation | +5.59e-02 |
Pos logit contributions | |
much | +0.067 |
lessly | +0.061 |
br | +0.059 |
wonderful | +0.058 |
On | +0.058 |
Neg logit contributions | |
Cr | -0.093 |
ados | -0.092 |
Majesty | -0.087 |
ishment | -0.084 |
lee | -0.084 |
Token | few |
Token ID | 1178 |
Feature activation | +4.89e-02 |
Pos logit contributions | |
hes | +0.091 |
lessly | +0.088 |
ier | +0.084 |
eps | +0.081 |
ooky | +0.076 |
Neg logit contributions | |
Cr | -0.086 |
Harry | -0.084 |
Po | -0.083 |
Av | -0.082 |
fro | -0.082 |
Token | minutes |
Token ID | 2431 |
Feature activation | +4.52e-02 |
Pos logit contributions | |
lessly | +0.083 |
much | +0.083 |
On | +0.082 |
In | +0.081 |
lic | +0.078 |
Neg logit contributions | |
ados | -0.076 |
breaths | -0.075 |
Majesty | -0.073 |
lee | -0.072 |
knees | -0.069 |
Token | , |
Token ID | 11 |
Feature activation | +4.43e-02 |
Pos logit contributions | |
much | +0.078 |
wonderful | +0.068 |
lessly | +0.067 |
On | +0.065 |
br | +0.064 |
Neg logit contributions | |
ados | -0.085 |
Cr | -0.085 |
ishment | -0.080 |
Majesty | -0.079 |
lee | -0.077 |
Token | something |
Token ID | 1223 |
Feature activation | +9.36e-02 |
Pos logit contributions | |
br | +0.091 |
ching | +0.088 |
much | +0.087 |
lessly | +0.084 |
lic | +0.083 |
Neg logit contributions | |
ados | -0.063 |
Cr | -0.061 |
Harry | -0.058 |
ishment | -0.057 |
Majesty | -0.052 |
Token | incredible |
Token ID | 8082 |
Feature activation | +2.64e-02 |
Pos logit contributions | |
On | +0.109 |
lessly | +0.107 |
ifying | +0.106 |
Finally | +0.103 |
much | +0.101 |
Neg logit contributions | |
Cr | -0.174 |
Harry | -0.170 |
Nem | -0.168 |
lee | -0.166 |
Sammy | -0.161 |
Token | happened |
Token ID | 3022 |
Feature activation | +1.64e-03 |
Pos logit contributions | |
much | +0.041 |
On | +0.033 |
lessly | +0.033 |
In | +0.032 |
lot | +0.031 |
Neg logit contributions | |
Cr | -0.054 |
ados | -0.053 |
ishment | -0.052 |
Harry | -0.051 |
lee | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
much | +0.003 |
wonderful | +0.002 |
br | +0.002 |
lessly | +0.002 |
In | +0.002 |
Neg logit contributions | |
Cr | -0.003 |
ados | -0.003 |
ishment | -0.003 |
Majesty | -0.003 |
lee | -0.003 |
Token | John |
Token ID | 1757 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
much | +0.035 |
br | +0.033 |
wonderful | +0.033 |
lessly | +0.033 |
ching | +0.032 |
Neg logit contributions | |
Cr | -0.018 |
ados | -0.018 |
ishment | -0.016 |
Majesty | -0.016 |
lee | -0.015 |
Token | saw |
Token ID | 2497 |
Feature activation | +2.91e-02 |
Pos logit contributions | |
much | +0.018 |
lessly | +0.015 |
wonderful | +0.015 |
br | +0.014 |
On | +0.014 |
Neg logit contributions | |
ados | -0.028 |
Cr | -0.028 |
Majesty | -0.028 |
ishment | -0.027 |
lee | -0.026 |
Token | shoes |
Token ID | 10012 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
ados | +0.018 |
Cr | +0.018 |
ishment | +0.017 |
Majesty | +0.017 |
lee | +0.016 |
Neg logit contributions | |
much | -0.022 |
wonderful | -0.019 |
lessly | -0.019 |
On | -0.019 |
br | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
much | +0.053 |
wonderful | +0.049 |
br | +0.047 |
In | +0.045 |
lessly | +0.045 |
Neg logit contributions | |
ados | -0.056 |
Cr | -0.056 |
ishment | -0.055 |
gy | -0.052 |
Fro | -0.051 |
Token | But |
Token ID | 887 |
Feature activation | +4.05e-02 |
Pos logit contributions | |
ados | +0.022 |
Cr | +0.022 |
Majesty | +0.021 |
ishment | +0.021 |
lee | +0.019 |
Neg logit contributions | |
much | -0.030 |
lessly | -0.026 |
wonderful | -0.026 |
On | -0.026 |
br | -0.025 |
Token | then |
Token ID | 788 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
much | +0.086 |
lessly | +0.080 |
wonderful | +0.080 |
br | +0.079 |
ching | +0.078 |
Neg logit contributions | |
Cr | -0.058 |
ados | -0.056 |
Fro | -0.052 |
ishment | -0.050 |
Majesty | -0.048 |
Token | , |
Token ID | 11 |
Feature activation | +4.09e-02 |
Pos logit contributions | |
much | +0.048 |
lessly | +0.041 |
wonderful | +0.041 |
On | +0.039 |
br | +0.039 |
Neg logit contributions | |
ados | -0.049 |
Cr | -0.049 |
Majesty | -0.047 |
ishment | -0.047 |
lee | -0.044 |
Token | something |
Token ID | 1223 |
Feature activation | +9.34e-02 |
Pos logit contributions | |
much | +0.080 |
lessly | +0.073 |
wonderful | +0.071 |
br | +0.070 |
ching | +0.069 |
Neg logit contributions | |
Cr | -0.065 |
ados | -0.065 |
Majesty | -0.060 |
Fro | -0.058 |
ishment | -0.058 |
Token | happened |
Token ID | 3022 |
Feature activation | +1.40e-03 |
Pos logit contributions | |
ifying | +0.112 |
lessly | +0.107 |
much | +0.102 |
arest | +0.102 |
On | +0.101 |
Neg logit contributions | |
Fro | -0.170 |
Cr | -0.165 |
lee | -0.164 |
Freddy | -0.160 |
slur | -0.158 |
Token | . |
Token ID | 13 |
Feature activation | +4.57e-03 |
Pos logit contributions | |
much | +0.002 |
wonderful | +0.002 |
lessly | +0.002 |
br | +0.002 |
On | +0.002 |
Neg logit contributions | |
ados | -0.003 |
Cr | -0.003 |
Majesty | -0.003 |
ishment | -0.003 |
lee | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +6.27e-02 |
Pos logit contributions | |
much | +0.010 |
wonderful | +0.009 |
lessly | +0.009 |
On | +0.009 |
br | +0.009 |
Neg logit contributions | |
ados | -0.006 |
Cr | -0.006 |
Majesty | -0.006 |
ishment | -0.006 |
lee | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +5.94e-02 |
Pos logit contributions | |
much | +0.096 |
lessly | +0.090 |
wonderful | +0.088 |
br | +0.086 |
On | +0.080 |
Neg logit contributions | |
Cr | -0.119 |
ados | -0.114 |
Fro | -0.112 |
lee | -0.108 |
Majesty | -0.107 |
Token | Tom |
Token ID | 13787 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
wonderful | +0.085 |
much | +0.084 |
lessly | +0.083 |
hard | +0.073 |
valuable | +0.073 |
Neg logit contributions | |
ados | -0.108 |
Cr | -0.106 |
Fro | -0.097 |
Rib | -0.096 |
ree | -0.095 |
Token | |
Token ID | 220 |
Feature activation | +9.09e-03 |
Pos logit contributions | |
ados | +0.018 |
ican | +0.018 |
ishment | +0.017 |
Majesty | +0.017 |
ans | +0.017 |
Neg logit contributions | |
much | -0.015 |
On | -0.013 |
In | -0.012 |
lessly | -0.012 |
wonderful | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +5.30e-02 |
Pos logit contributions | |
much | +0.013 |
wonderful | +0.011 |
lessly | +0.011 |
br | +0.011 |
On | +0.010 |
Neg logit contributions | |
Cr | -0.019 |
ados | -0.019 |
Majesty | -0.018 |
ishment | -0.018 |
Fro | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
much | +0.084 |
wonderful | +0.075 |
br | +0.073 |
lessly | +0.073 |
terrible | +0.069 |
Neg logit contributions | |
Cr | -0.099 |
Fro | -0.096 |
ados | -0.094 |
Rib | -0.092 |
lee | -0.092 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | +4.82e-02 |
Pos logit contributions | |
much | +0.058 |
wonderful | +0.049 |
lessly | +0.048 |
br | +0.048 |
hard | +0.046 |
Neg logit contributions | |
Cr | -0.096 |
ados | -0.095 |
Majesty | -0.093 |
Fro | -0.093 |
ishment | -0.092 |
Token | , |
Token ID | 11 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
much | +0.081 |
wonderful | +0.066 |
lessly | +0.064 |
br | +0.063 |
hard | +0.063 |
Neg logit contributions | |
ados | -0.099 |
Cr | -0.098 |
ishment | -0.097 |
Majesty | -0.096 |
keeping | -0.090 |
Token | something |
Token ID | 1223 |
Feature activation | +9.34e-02 |
Pos logit contributions | |
much | +0.113 |
wonderful | +0.100 |
lessly | +0.099 |
On | +0.097 |
br | +0.097 |
Neg logit contributions | |
ados | -0.054 |
Cr | -0.053 |
Majesty | -0.051 |
ishment | -0.051 |
lee | -0.045 |
Token | loud |
Token ID | 7812 |
Feature activation | +9.80e-03 |
Pos logit contributions | |
ris | +0.109 |
ifying | +0.107 |
much | +0.104 |
million | +0.104 |
arest | +0.098 |
Neg logit contributions | |
lee | -0.159 |
Fro | -0.158 |
Harry | -0.154 |
Sammy | -0.153 |
Freddy | -0.150 |
Token | happened |
Token ID | 3022 |
Feature activation | -8.97e-04 |
Pos logit contributions | |
much | +0.017 |
wonderful | +0.014 |
In | +0.014 |
lessly | +0.014 |
On | +0.014 |
Neg logit contributions | |
ados | -0.018 |
Cr | -0.018 |
ishment | -0.017 |
lee | -0.017 |
icans | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +8.41e-03 |
Pos logit contributions | |
Cr | +0.002 |
ados | +0.002 |
gy | +0.001 |
ishment | +0.001 |
lee | +0.001 |
Neg logit contributions | |
much | -0.002 |
br | -0.001 |
wonderful | -0.001 |
lessly | -0.001 |
In | -0.001 |
Token | It |
Token ID | 632 |
Feature activation | +5.82e-02 |
Pos logit contributions | |
much | +0.019 |
lessly | +0.016 |
On | +0.016 |
In | +0.015 |
hard | +0.015 |
Neg logit contributions | |
ados | -0.013 |
ishment | -0.013 |
Majesty | -0.013 |
ans | -0.013 |
ci | -0.013 |
Token | was |
Token ID | 373 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
much | +0.099 |
arest | +0.087 |
lessly | +0.085 |
wonderful | +0.081 |
vere | +0.080 |
Neg logit contributions | |
gy | -0.100 |
Harry | -0.091 |
stick | -0.091 |
Cr | -0.091 |
lee | -0.091 |
Token | the |
Token ID | 262 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
Cr | +0.013 |
ados | +0.013 |
Majesty | +0.013 |
ishment | +0.012 |
amb | +0.012 |
Neg logit contributions | |
much | -0.012 |
wonderful | -0.010 |
hard | -0.010 |
lessly | -0.010 |
On | -0.009 |
Token | hole |
Token ID | 7604 |
Feature activation | +4.31e-02 |
Pos logit contributions | |
much | +0.036 |
br | +0.035 |
lessly | +0.034 |
On | +0.034 |
lic | +0.033 |
Neg logit contributions | |
ados | -0.031 |
Cr | -0.029 |
ishment | -0.028 |
Majesty | -0.027 |
Harry | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -9.95e-03 |
Pos logit contributions | |
much | +0.068 |
wonderful | +0.062 |
br | +0.061 |
In | +0.060 |
On | +0.058 |
Neg logit contributions | |
ados | -0.081 |
Cr | -0.079 |
ishment | -0.078 |
lee | -0.076 |
Majesty | -0.075 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
Cr | +0.017 |
ados | +0.017 |
ishment | +0.016 |
Majesty | +0.016 |
ans | +0.015 |
Neg logit contributions | |
much | -0.021 |
On | -0.017 |
In | -0.017 |
lessly | -0.017 |
wonderful | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | +4.34e-02 |
Pos logit contributions | |
much | +0.076 |
wonderful | +0.065 |
lessly | +0.064 |
On | +0.062 |
br | +0.062 |
Neg logit contributions | |
ados | -0.077 |
Cr | -0.076 |
Majesty | -0.074 |
ishment | -0.074 |
lee | -0.069 |
Token | something |
Token ID | 1223 |
Feature activation | +9.29e-02 |
Pos logit contributions | |
much | +0.081 |
br | +0.075 |
lessly | +0.074 |
wonderful | +0.072 |
ching | +0.071 |
Neg logit contributions | |
ados | -0.071 |
Cr | -0.067 |
ishment | -0.063 |
Harry | -0.061 |
Majesty | -0.061 |
Token | popped |
Token ID | 22928 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
much | +0.113 |
ris | +0.112 |
ifying | +0.108 |
Today | +0.106 |
million | +0.106 |
Neg logit contributions | |
Harry | -0.161 |
Fro | -0.159 |
Freddy | -0.157 |
Sammy | -0.154 |
Cr | -0.154 |
Token | out |
Token ID | 503 |
Feature activation | +3.03e-02 |
Pos logit contributions | |
much | +0.032 |
wonderful | +0.028 |
lessly | +0.027 |
On | +0.026 |
br | +0.026 |
Neg logit contributions | |
ados | -0.036 |
Cr | -0.035 |
ishment | -0.034 |
gy | -0.033 |
lee | -0.033 |
Token | of |
Token ID | 286 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
much | +0.045 |
wonderful | +0.040 |
arest | +0.040 |
br | +0.039 |
In | +0.039 |
Neg logit contributions | |
ados | -0.057 |
lee | -0.056 |
Cr | -0.055 |
Harry | -0.054 |
ishment | -0.052 |
Token | the |
Token ID | 262 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
much | +0.026 |
lessly | +0.023 |
br | +0.023 |
wonderful | +0.023 |
On | +0.022 |
Neg logit contributions | |
ados | -0.015 |
Cr | -0.015 |
ishment | -0.014 |
Majesty | -0.013 |
lee | -0.013 |
Token | hole |
Token ID | 7604 |
Feature activation | +5.16e-02 |
Pos logit contributions | |
much | +0.044 |
lessly | +0.040 |
On | +0.040 |
br | +0.040 |
In | +0.039 |
Neg logit contributions | |
ados | -0.032 |
ishment | -0.030 |
Cr | -0.030 |
Majesty | -0.029 |
gy | -0.028 |
Token | big |
Token ID | 1263 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
ados | +0.044 |
Cr | +0.044 |
ishment | +0.041 |
Majesty | +0.040 |
icans | +0.039 |
Neg logit contributions | |
much | -0.045 |
wonderful | -0.039 |
On | -0.037 |
hard | -0.037 |
In | -0.036 |
Token | garden |
Token ID | 11376 |
Feature activation | +6.87e-03 |
Pos logit contributions | |
Cr | +0.047 |
ados | +0.046 |
cess | +0.045 |
amb | +0.044 |
Majesty | +0.044 |
Neg logit contributions | |
much | -0.038 |
wonderful | -0.034 |
big | -0.033 |
hard | -0.032 |
bad | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
much | +0.009 |
On | +0.007 |
wonderful | +0.006 |
lessly | +0.006 |
In | +0.006 |
Neg logit contributions | |
ados | -0.017 |
Cr | -0.017 |
Majesty | -0.017 |
ishment | -0.016 |
lee | -0.016 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | +4.13e-02 |
Pos logit contributions | |
ishment | +0.024 |
ados | +0.024 |
ans | +0.024 |
Cr | +0.024 |
Majesty | +0.024 |
Neg logit contributions | |
much | -0.023 |
On | -0.019 |
In | -0.019 |
lessly | -0.018 |
hard | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
much | +0.072 |
wonderful | +0.060 |
br | +0.058 |
lessly | +0.058 |
On | +0.056 |
Neg logit contributions | |
Cr | -0.081 |
ados | -0.080 |
ishment | -0.080 |
Majesty | -0.078 |
lee | -0.073 |
Token | everything |
Token ID | 2279 |
Feature activation | +9.29e-02 |
Pos logit contributions | |
much | +0.068 |
wonderful | +0.057 |
lessly | +0.057 |
On | +0.055 |
br | +0.054 |
Neg logit contributions | |
ados | -0.077 |
Cr | -0.076 |
Majesty | -0.074 |
ishment | -0.074 |
lee | -0.069 |
Token | started |
Token ID | 2067 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
much | +0.158 |
br | +0.140 |
lessly | +0.140 |
On | +0.138 |
arest | +0.138 |
Neg logit contributions | |
Cr | -0.145 |
knees | -0.144 |
lee | -0.143 |
ados | -0.142 |
ishment | -0.135 |
Token | to |
Token ID | 284 |
Feature activation | +7.60e-03 |
Pos logit contributions | |
much | +0.043 |
wonderful | +0.037 |
lessly | +0.036 |
On | +0.035 |
br | +0.035 |
Neg logit contributions | |
ados | -0.039 |
Cr | -0.039 |
Majesty | -0.038 |
ishment | -0.037 |
lee | -0.035 |
Token | shake |
Token ID | 13279 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
much | +0.016 |
lessly | +0.014 |
wonderful | +0.014 |
br | +0.014 |
On | +0.013 |
Neg logit contributions | |
Cr | -0.011 |
ados | -0.011 |
ishment | -0.011 |
Majesty | -0.011 |
Fro | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -4.93e-04 |
Pos logit contributions | |
much | +0.035 |
In | +0.032 |
br | +0.032 |
wonderful | +0.031 |
lessly | +0.031 |
Neg logit contributions | |
ados | -0.041 |
ishment | -0.040 |
Cr | -0.039 |
gy | -0.038 |
lee | -0.037 |
Token | The |
Token ID | 383 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
ishment | +0.001 |
Majesty | +0.001 |
Cr | +0.001 |
ados | +0.001 |
ans | +0.001 |
Neg logit contributions | |
much | -0.001 |
In | -0.001 |
On | -0.001 |
lessly | -0.001 |
hard | -0.001 |
Token | when |
Token ID | 618 |
Feature activation | +5.62e-02 |
Pos logit contributions | |
much | +0.070 |
wonderful | +0.060 |
lessly | +0.060 |
br | +0.058 |
On | +0.058 |
Neg logit contributions | |
ados | -0.071 |
Cr | -0.071 |
ishment | -0.068 |
Majesty | -0.068 |
lee | -0.063 |
Token | he |
Token ID | 339 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
much | +0.095 |
lessly | +0.080 |
wonderful | +0.080 |
On | +0.077 |
br | +0.077 |
Neg logit contributions | |
ados | -0.110 |
Cr | -0.109 |
Majesty | -0.106 |
ishment | -0.106 |
lee | -0.099 |
Token | was |
Token ID | 373 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
much | +0.035 |
lessly | +0.033 |
wonderful | +0.029 |
br | +0.029 |
hard | +0.027 |
Neg logit contributions | |
Cr | -0.036 |
Fro | -0.034 |
ishment | -0.034 |
fro | -0.033 |
ados | -0.033 |
Token | gone |
Token ID | 3750 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
br | +0.049 |
much | +0.048 |
lessly | +0.048 |
In | +0.045 |
wonderful | +0.045 |
Neg logit contributions | |
Majesty | -0.058 |
lee | -0.058 |
Fro | -0.057 |
Cr | -0.056 |
ishment | -0.055 |
Token | , |
Token ID | 11 |
Feature activation | +3.99e-02 |
Pos logit contributions | |
much | +0.044 |
lessly | +0.039 |
wonderful | +0.038 |
In | +0.038 |
br | +0.037 |
Neg logit contributions | |
ados | -0.037 |
Cr | -0.036 |
ishment | -0.035 |
Majesty | -0.035 |
lee | -0.033 |
Token | something |
Token ID | 1223 |
Feature activation | +9.29e-02 |
Pos logit contributions | |
much | +0.070 |
lessly | +0.061 |
wonderful | +0.060 |
br | +0.059 |
On | +0.059 |
Neg logit contributions | |
ados | -0.074 |
Cr | -0.074 |
Majesty | -0.070 |
ishment | -0.070 |
Fro | -0.066 |
Token | magical |
Token ID | 10883 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
much | +0.101 |
On | +0.099 |
lessly | +0.098 |
Finally | +0.097 |
In | +0.092 |
Neg logit contributions | |
Cr | -0.190 |
Fro | -0.179 |
lee | -0.178 |
Nem | -0.177 |
ot | -0.172 |
Token | happened |
Token ID | 3022 |
Feature activation | +3.90e-04 |
Pos logit contributions | |
much | +0.056 |
lessly | +0.046 |
On | +0.045 |
In | +0.044 |
wonderful | +0.043 |
Neg logit contributions | |
Cr | -0.071 |
ados | -0.071 |
ishment | -0.068 |
lee | -0.068 |
ican | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | +2.34e-03 |
Pos logit contributions | |
much | +0.001 |
wonderful | +0.001 |
lessly | +0.001 |
br | +0.001 |
In | +0.001 |
Neg logit contributions | |
Cr | -0.001 |
ados | -0.001 |
ishment | -0.001 |
Majesty | -0.001 |
lee | -0.001 |
Token | A |
Token ID | 317 |
Feature activation | +5.70e-02 |
Pos logit contributions | |
much | +0.005 |
In | +0.004 |
On | +0.004 |
lessly | +0.004 |
wonderful | +0.004 |
Neg logit contributions | |
Majesty | -0.004 |
ishment | -0.004 |
ados | -0.004 |
ans | -0.004 |
ishes | -0.004 |
Token | great |
Token ID | 1049 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
lessly | +0.077 |
much | +0.071 |
On | +0.067 |
In | +0.067 |
hest | +0.061 |
Neg logit contributions | |
Majesty | -0.116 |
Cr | -0.114 |
Fro | -0.108 |
fro | -0.108 |
ados | -0.106 |
Token | to |
Token ID | 284 |
Feature activation | -3.40e-02 |
Pos logit contributions | |
Majesty | +0.003 |
ados | +0.003 |
Cr | +0.003 |
ishment | +0.003 |
gy | +0.003 |
Neg logit contributions | |
much | -0.003 |
wonderful | -0.002 |
On | -0.002 |
hard | -0.002 |
lessly | -0.002 |
Token | join |
Token ID | 4654 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
ados | +0.066 |
Majesty | +0.063 |
ishment | +0.061 |
Cr | +0.058 |
gy | +0.057 |
Neg logit contributions | |
much | -0.064 |
wonderful | -0.056 |
On | -0.052 |
lot | -0.052 |
hard | -0.052 |
Token | them |
Token ID | 606 |
Feature activation | +8.21e-03 |
Pos logit contributions | |
Majesty | +0.027 |
ados | +0.025 |
ishment | +0.025 |
Cr | +0.024 |
gy | +0.023 |
Neg logit contributions | |
much | -0.022 |
wonderful | -0.018 |
on | -0.018 |
hard | -0.017 |
On | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -1.48e-03 |
Pos logit contributions | |
much | +0.013 |
wonderful | +0.010 |
lessly | +0.010 |
On | +0.010 |
hard | +0.009 |
Neg logit contributions | |
ados | -0.018 |
Cr | -0.018 |
Majesty | -0.018 |
ishment | -0.017 |
lee | -0.016 |
Token | but |
Token ID | 475 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
ados | +0.003 |
Cr | +0.003 |
Majesty | +0.003 |
ishment | +0.003 |
lee | +0.003 |
Neg logit contributions | |
much | -0.003 |
wonderful | -0.002 |
lessly | -0.002 |
br | -0.002 |
On | -0.002 |
Token | something |
Token ID | 1223 |
Feature activation | +9.28e-02 |
Pos logit contributions | |
much | +0.056 |
lessly | +0.049 |
wonderful | +0.048 |
br | +0.047 |
On | +0.047 |
Neg logit contributions | |
Cr | -0.052 |
ados | -0.051 |
ishment | -0.050 |
Majesty | -0.049 |
Fro | -0.046 |
Token | was |
Token ID | 373 |
Feature activation | +5.69e-02 |
Pos logit contributions | |
Finally | +0.117 |
ifying | +0.111 |
arest | +0.108 |
On | +0.108 |
much | +0.103 |
Neg logit contributions | |
lee | -0.167 |
Cr | -0.165 |
Fro | -0.155 |
D | -0.154 |
ot | -0.153 |
Token | different |
Token ID | 1180 |
Feature activation | +3.83e-02 |
Pos logit contributions | |
arest | +0.080 |
Finally | +0.077 |
br | +0.073 |
lessly | +0.071 |
much | +0.071 |
Neg logit contributions | |
Cr | -0.109 |
Fro | -0.105 |
ishment | -0.098 |
Harry | -0.098 |
gy | -0.097 |
Token | . |
Token ID | 13 |
Feature activation | -2.55e-03 |
Pos logit contributions | |
much | +0.068 |
Finally | +0.064 |
In | +0.062 |
arest | +0.061 |
br | +0.061 |
Neg logit contributions | |
ados | -0.065 |
Cr | -0.062 |
ishment | -0.061 |
lee | -0.059 |
keeping | -0.059 |
Token | Her |
Token ID | 2332 |
Feature activation | +8.55e-03 |
Pos logit contributions | |
ishment | +0.005 |
Majesty | +0.005 |
ados | +0.004 |
ans | +0.004 |
keeping | +0.004 |
Neg logit contributions | |
much | -0.005 |
In | -0.004 |
On | -0.004 |
lessly | -0.004 |
lot | -0.004 |
Token | friends |
Token ID | 2460 |
Feature activation | +4.32e-02 |
Pos logit contributions | |
much | +0.016 |
wonderful | +0.014 |
hard | +0.014 |
bad | +0.013 |
big | +0.013 |
Neg logit contributions | |
Cr | -0.016 |
ados | -0.016 |
ishment | -0.015 |
Majesty | -0.014 |
Rib | -0.014 |
Token | she |
Token ID | 673 |
Feature activation | +5.23e-03 |
Pos logit contributions | |
much | +0.010 |
wonderful | +0.009 |
lessly | +0.008 |
On | +0.008 |
hard | +0.008 |
Neg logit contributions | |
Cr | -0.009 |
ados | -0.009 |
Majesty | -0.009 |
ishment | -0.009 |
lee | -0.008 |
Token | worked |
Token ID | 3111 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
much | +0.007 |
wonderful | +0.006 |
lessly | +0.006 |
On | +0.006 |
br | +0.006 |
Neg logit contributions | |
ados | -0.012 |
Cr | -0.012 |
ishment | -0.011 |
Majesty | -0.011 |
lee | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | +1.70e-03 |
Pos logit contributions | |
Over | +0.025 |
br | +0.025 |
In | +0.025 |
arest | +0.025 |
much | +0.024 |
Neg logit contributions | |
ados | -0.033 |
ishment | -0.030 |
Cr | -0.029 |
lee | -0.029 |
gy | -0.028 |
Token | but |
Token ID | 475 |
Feature activation | +3.62e-02 |
Pos logit contributions | |
much | +0.003 |
lessly | +0.002 |
wonderful | +0.002 |
On | +0.002 |
hard | +0.002 |
Neg logit contributions | |
ados | -0.003 |
Cr | -0.003 |
Majesty | -0.003 |
ishment | -0.003 |
lee | -0.003 |
Token | then |
Token ID | 788 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
much | +0.067 |
lessly | +0.058 |
wonderful | +0.057 |
br | +0.056 |
On | +0.055 |
Neg logit contributions | |
ados | -0.065 |
Cr | -0.064 |
ishment | -0.062 |
Majesty | -0.062 |
lee | -0.058 |
Token | something |
Token ID | 1223 |
Feature activation | +9.28e-02 |
Pos logit contributions | |
much | +0.044 |
wonderful | +0.036 |
lessly | +0.035 |
On | +0.035 |
br | +0.034 |
Neg logit contributions | |
ados | -0.054 |
Cr | -0.054 |
ishment | -0.053 |
Majesty | -0.053 |
lee | -0.049 |
Token | bad |
Token ID | 2089 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
ris | +0.100 |
arest | +0.100 |
On | +0.098 |
ifying | +0.098 |
much | +0.096 |
Neg logit contributions | |
lee | -0.166 |
Cr | -0.165 |
Freddy | -0.163 |
Fro | -0.161 |
Nem | -0.160 |
Token | happened |
Token ID | 3022 |
Feature activation | +4.77e-06 |
Pos logit contributions | |
much | +0.025 |
wonderful | +0.021 |
lessly | +0.020 |
On | +0.020 |
In | +0.019 |
Neg logit contributions | |
ados | -0.037 |
Cr | -0.035 |
lee | -0.034 |
ishment | -0.034 |
Harry | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | +4.25e-03 |
Token | The |
Token ID | 383 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
much | +0.009 |
On | +0.008 |
In | +0.008 |
lessly | +0.007 |
hard | +0.007 |
Neg logit contributions | |
Majesty | -0.007 |
ishment | -0.007 |
ans | -0.007 |
Cr | -0.007 |
ados | -0.007 |
Token | apple |
Token ID | 17180 |
Feature activation | +4.89e-02 |
Pos logit contributions | |
much | +0.074 |
lessly | +0.066 |
wonderful | +0.064 |
br | +0.064 |
On | +0.063 |
Neg logit contributions | |
ados | -0.053 |
Majesty | -0.051 |
Cr | -0.051 |
ishment | -0.050 |
lee | -0.046 |
Token | they |
Token ID | 484 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
much | +0.132 |
lessly | +0.127 |
br | +0.125 |
ching | +0.121 |
wonderful | +0.118 |
Neg logit contributions | |
Cr | -0.092 |
ados | -0.087 |
ishment | -0.083 |
gy | -0.081 |
Majesty | -0.081 |
Token | reached |
Token ID | 4251 |
Feature activation | -6.24e-04 |
Pos logit contributions | |
much | +0.062 |
br | +0.059 |
lessly | +0.058 |
wonderful | +0.056 |
ching | +0.054 |
Neg logit contributions | |
Cr | -0.055 |
Fro | -0.048 |
lee | -0.046 |
fro | -0.046 |
ishment | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
Cr | +0.001 |
ados | +0.001 |
ishment | +0.001 |
Majesty | +0.001 |
Fro | +0.001 |
Neg logit contributions | |
much | -0.001 |
lessly | -0.001 |
br | -0.001 |
wonderful | -0.001 |
In | -0.001 |
Token | stand |
Token ID | 1302 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
lessly | +0.031 |
much | +0.031 |
ching | +0.028 |
lic | +0.028 |
br | +0.027 |
Neg logit contributions | |
Majesty | -0.025 |
ishment | -0.022 |
Cr | -0.022 |
exp | -0.022 |
fro | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | +3.72e-02 |
Pos logit contributions | |
much | +0.072 |
wonderful | +0.062 |
lessly | +0.062 |
On | +0.061 |
br | +0.060 |
Neg logit contributions | |
ados | -0.067 |
Cr | -0.066 |
Majesty | -0.064 |
ishment | -0.064 |
lee | -0.060 |
Token | things |
Token ID | 1243 |
Feature activation | +9.27e-02 |
Pos logit contributions | |
much | +0.087 |
br | +0.082 |
lessly | +0.081 |
ching | +0.078 |
wonderful | +0.078 |
Neg logit contributions | |
Cr | -0.045 |
ados | -0.043 |
ishment | -0.040 |
Harry | -0.038 |
Majesty | -0.037 |
Token | weren |
Token ID | 6304 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
much | +0.150 |
br | +0.130 |
Finally | +0.130 |
On | +0.128 |
One | +0.127 |
Neg logit contributions | |
Cr | -0.153 |
Harry | -0.149 |
ados | -0.148 |
Sammy | -0.147 |
lee | -0.146 |
Token | 't |
Token ID | 470 |
Feature activation | +6.13e-02 |
Pos logit contributions | |
wonderful | +0.067 |
much | +0.066 |
lessly | +0.064 |
br | +0.062 |
importantly | +0.061 |
Neg logit contributions | |
Cr | -0.045 |
ados | -0.038 |
Harry | -0.036 |
ican | -0.036 |
Twe | -0.035 |
Token | as |
Token ID | 355 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
Finally | +0.093 |
In | +0.089 |
Over | +0.089 |
arest | +0.088 |
br | +0.086 |
Neg logit contributions | |
Harry | -0.101 |
Cr | -0.098 |
ishment | -0.094 |
Nem | -0.093 |
ados | -0.092 |
Token | expected |
Token ID | 2938 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
br | +0.063 |
ching | +0.061 |
In | +0.058 |
arest | +0.057 |
On | +0.056 |
Neg logit contributions | |
Cr | -0.125 |
Harry | -0.120 |
Nem | -0.115 |
Sammy | -0.114 |
Joey | -0.113 |
Token | . |
Token ID | 13 |
Feature activation | +5.15e-03 |
Pos logit contributions | |
much | +0.035 |
wonderful | +0.031 |
On | +0.031 |
lessly | +0.031 |
In | +0.031 |
Neg logit contributions | |
Cr | -0.038 |
ados | -0.037 |
lee | -0.035 |
ishment | -0.035 |
Harry | -0.034 |
Token | they |
Token ID | 484 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
much | +0.094 |
lessly | +0.081 |
wonderful | +0.079 |
br | +0.077 |
On | +0.077 |
Neg logit contributions | |
ados | -0.108 |
Cr | -0.108 |
Majesty | -0.104 |
ishment | -0.103 |
lee | -0.097 |
Token | reached |
Token ID | 4251 |
Feature activation | +1.38e-03 |
Pos logit contributions | |
much | +0.046 |
lessly | +0.045 |
wonderful | +0.044 |
valuable | +0.044 |
br | +0.044 |
Neg logit contributions | |
Cr | -0.041 |
Fro | -0.039 |
fro | -0.037 |
lee | -0.035 |
Rib | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
much | +0.003 |
lessly | +0.003 |
wonderful | +0.003 |
br | +0.003 |
On | +0.003 |
Neg logit contributions | |
Cr | -0.002 |
ados | -0.002 |
ishment | -0.002 |
Majesty | -0.002 |
Fro | -0.002 |
Token | bottom |
Token ID | 4220 |
Feature activation | +4.43e-02 |
Pos logit contributions | |
lessly | +0.041 |
lic | +0.039 |
br | +0.038 |
ching | +0.038 |
importantly | +0.038 |
Neg logit contributions | |
Majesty | -0.021 |
ASH | -0.021 |
fro | -0.020 |
ican | -0.020 |
Cr | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
much | +0.084 |
lessly | +0.076 |
wonderful | +0.076 |
On | +0.073 |
In | +0.073 |
Neg logit contributions | |
ados | -0.073 |
Cr | -0.073 |
ishment | -0.068 |
Majesty | -0.066 |
Fro | -0.066 |
Token | something |
Token ID | 1223 |
Feature activation | +9.27e-02 |
Pos logit contributions | |
much | +0.058 |
lessly | +0.049 |
wonderful | +0.049 |
br | +0.048 |
On | +0.048 |
Neg logit contributions | |
ados | -0.064 |
Cr | -0.064 |
ishment | -0.062 |
Majesty | -0.062 |
lee | -0.058 |
Token | bad |
Token ID | 2089 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
lessly | +0.122 |
ifying | +0.122 |
Finally | +0.119 |
On | +0.115 |
arest | +0.115 |
Neg logit contributions | |
Cr | -0.159 |
Fro | -0.152 |
Nem | -0.151 |
ot | -0.147 |
Freddy | -0.145 |
Token | happened |
Token ID | 3022 |
Feature activation | +2.35e-03 |
Pos logit contributions | |
much | +0.031 |
lessly | +0.025 |
wonderful | +0.025 |
On | +0.025 |
In | +0.024 |
Neg logit contributions | |
Cr | -0.042 |
ados | -0.041 |
ishment | -0.039 |
Fro | -0.039 |
Harry | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | +7.37e-03 |
Pos logit contributions | |
much | +0.004 |
wonderful | +0.004 |
lessly | +0.004 |
br | +0.003 |
In | +0.003 |
Neg logit contributions | |
Cr | -0.004 |
ados | -0.004 |
ishment | -0.004 |
Majesty | -0.004 |
gy | -0.004 |
Token | Tim |
Token ID | 5045 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
much | +0.019 |
lessly | +0.016 |
In | +0.016 |
On | +0.016 |
wonderful | +0.016 |
Neg logit contributions | |
Majesty | -0.009 |
ados | -0.009 |
ishment | -0.009 |
Cr | -0.008 |
lee | -0.007 |
Token | 's |
Token ID | 338 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
much | +0.026 |
wonderful | +0.022 |
lessly | +0.022 |
br | +0.021 |
On | +0.021 |
Neg logit contributions | |
ados | -0.030 |
Cr | -0.029 |
ishment | -0.029 |
Majesty | -0.029 |
lee | -0.027 |
Token | give |
Token ID | 1577 |
Feature activation | -5.56e-02 |
Pos logit contributions | |
ados | +0.158 |
Cr | +0.154 |
ishment | +0.153 |
Majesty | +0.151 |
lee | +0.143 |
Neg logit contributions | |
much | -0.160 |
wonderful | -0.136 |
On | -0.130 |
hard | -0.128 |
lessly | -0.127 |
Token | to |
Token ID | 284 |
Feature activation | -6.70e-02 |
Pos logit contributions | |
Majesty | +0.102 |
ishment | +0.100 |
Cr | +0.098 |
ados | +0.097 |
cess | +0.090 |
Neg logit contributions | |
much | -0.105 |
wonderful | -0.088 |
hard | -0.085 |
lessly | -0.084 |
lot | -0.080 |
Token | the |
Token ID | 262 |
Feature activation | -4.15e-02 |
Pos logit contributions | |
ishment | +0.130 |
ados | +0.124 |
amb | +0.123 |
Cr | +0.121 |
Majesty | +0.120 |
Neg logit contributions | |
much | -0.127 |
wonderful | -0.106 |
lot | -0.098 |
hard | -0.097 |
bad | -0.093 |
Token | rh |
Token ID | 9529 |
Feature activation | -7.28e-02 |
Pos logit contributions | |
ados | +0.066 |
Cr | +0.066 |
Majesty | +0.064 |
ishment | +0.063 |
lee | +0.058 |
Neg logit contributions | |
much | -0.085 |
wonderful | -0.073 |
lessly | -0.073 |
On | -0.071 |
br | -0.071 |
Token | in |
Token ID | 259 |
Feature activation | -5.42e-02 |
Pos logit contributions | |
ados | +0.161 |
Cr | +0.159 |
Majesty | +0.155 |
ishment | +0.154 |
Fro | +0.145 |
Neg logit contributions | |
much | -0.104 |
wonderful | -0.086 |
lessly | -0.085 |
On | -0.082 |
br | -0.080 |
Token | oc |
Token ID | 420 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
ishment | +0.095 |
ados | +0.091 |
Majesty | +0.089 |
Cr | +0.088 |
lee | +0.087 |
Neg logit contributions | |
much | -0.111 |
wonderful | -0.093 |
br | -0.090 |
hard | -0.089 |
On | -0.088 |
Token | eros |
Token ID | 27498 |
Feature activation | -5.96e-02 |
Pos logit contributions | |
Jenna | +0.144 |
ishment | +0.135 |
lee | +0.133 |
Cr | +0.132 |
Rachel | +0.132 |
Neg logit contributions | |
much | -0.326 |
br | -0.299 |
hard | -0.298 |
ier | -0.288 |
lessly | -0.288 |
Token | . |
Token ID | 13 |
Feature activation | -5.43e-02 |
Pos logit contributions | |
ishment | +0.131 |
ados | +0.129 |
Cr | +0.127 |
Majesty | +0.126 |
lee | +0.120 |
Neg logit contributions | |
much | -0.092 |
wonderful | -0.073 |
lessly | -0.068 |
hard | -0.067 |
On | -0.066 |
Token | Lily |
Token ID | 20037 |
Feature activation | -3.83e-02 |
Pos logit contributions | |
ados | +0.108 |
ishment | +0.107 |
Majesty | +0.105 |
ishes | +0.102 |
urg | +0.102 |
Neg logit contributions | |
much | -0.095 |
On | -0.079 |
He | -0.075 |
In | -0.074 |
lessly | -0.069 |
Token | walked |
Token ID | 6807 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
ados | +0.103 |
Cr | +0.101 |
ishment | +0.098 |
Majesty | +0.094 |
ican | +0.094 |
Neg logit contributions | |
much | -0.040 |
wonderful | -0.031 |
hard | -0.030 |
had | -0.030 |
also | -0.029 |
Token | to |
Token ID | 284 |
Feature activation | -5.40e-02 |
Pos logit contributions | |
ados | +0.062 |
Cr | +0.061 |
Majesty | +0.060 |
ishment | +0.060 |
lee | +0.056 |
Neg logit contributions | |
much | -0.051 |
lessly | -0.042 |
wonderful | -0.042 |
On | -0.041 |
In | -0.040 |
Token | stick |
Token ID | 4859 |
Feature activation | -2.96e-02 |
Pos logit contributions | |
ishment | +0.174 |
ados | +0.174 |
Cr | +0.173 |
Majesty | +0.170 |
ican | +0.162 |
Neg logit contributions | |
much | -0.142 |
wonderful | -0.116 |
On | -0.115 |
hard | -0.113 |
ching | -0.109 |
Token | the |
Token ID | 262 |
Feature activation | -4.98e-02 |
Pos logit contributions | |
Majesty | +0.059 |
ishment | +0.057 |
Cr | +0.055 |
ados | +0.055 |
cess | +0.054 |
Neg logit contributions | |
much | -0.053 |
hard | -0.045 |
wonderful | -0.043 |
lessly | -0.042 |
br | -0.040 |
Token | mustache |
Token ID | 49303 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
ados | +0.087 |
Cr | +0.086 |
ishment | +0.084 |
Majesty | +0.083 |
lee | +0.078 |
Neg logit contributions | |
much | -0.095 |
wonderful | -0.081 |
lessly | -0.080 |
br | -0.078 |
On | -0.078 |
Token | on |
Token ID | 319 |
Feature activation | -4.63e-02 |
Pos logit contributions | |
Cr | +0.071 |
ados | +0.071 |
ishment | +0.070 |
Majesty | +0.069 |
lee | +0.066 |
Neg logit contributions | |
much | -0.055 |
lessly | -0.045 |
wonderful | -0.044 |
br | -0.043 |
On | -0.043 |
Token | her |
Token ID | 607 |
Feature activation | -7.95e-02 |
Pos logit contributions | |
Majesty | +0.084 |
Cr | +0.083 |
ishment | +0.082 |
ados | +0.082 |
lee | +0.077 |
Neg logit contributions | |
much | -0.089 |
wonderful | -0.074 |
hard | -0.072 |
On | -0.071 |
lot | -0.070 |
Token | upper |
Token ID | 6727 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
Cr | +0.138 |
ishment | +0.127 |
ados | +0.123 |
ican | +0.117 |
icans | +0.115 |
Neg logit contributions | |
much | -0.163 |
hard | -0.147 |
wonderful | -0.146 |
big | -0.140 |
bad | -0.138 |
Token | lip |
Token ID | 10645 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
ishment | +0.224 |
Cr | +0.223 |
icans | +0.200 |
ados | +0.200 |
ican | +0.198 |
Neg logit contributions | |
much | -0.293 |
He | -0.244 |
hard | -0.237 |
big | -0.234 |
ching | -0.234 |
Token | with |
Token ID | 351 |
Feature activation | -5.80e-03 |
Pos logit contributions | |
Cr | +0.101 |
ados | +0.100 |
ishment | +0.099 |
Majesty | +0.098 |
lee | +0.093 |
Neg logit contributions | |
much | -0.065 |
lessly | -0.051 |
wonderful | -0.050 |
br | -0.049 |
On | -0.048 |
Token | some |
Token ID | 617 |
Feature activation | +4.09e-03 |
Pos logit contributions | |
ados | +0.011 |
Cr | +0.011 |
Majesty | +0.011 |
ishment | +0.011 |
lee | +0.010 |
Neg logit contributions | |
much | -0.010 |
lessly | -0.008 |
On | -0.008 |
wonderful | -0.008 |
br | -0.008 |
Token | tape |
Token ID | 9154 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
much | +0.007 |
On | +0.006 |
lessly | +0.006 |
In | +0.006 |
wonderful | +0.006 |
Neg logit contributions | |
ados | -0.007 |
Majesty | -0.007 |
Cr | -0.007 |
ishment | -0.007 |
lee | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -4.75e-02 |
Pos logit contributions | |
Cr | +0.028 |
ados | +0.028 |
Majesty | +0.027 |
ishment | +0.027 |
lee | +0.026 |
Neg logit contributions | |
much | -0.018 |
lessly | -0.014 |
wonderful | -0.014 |
br | -0.013 |
On | -0.013 |
Token | 't |
Token ID | 470 |
Feature activation | -5.32e-02 |
Pos logit contributions | |
much | +0.048 |
wonderful | +0.039 |
br | +0.038 |
hard | +0.037 |
lessly | +0.036 |
Neg logit contributions | |
ishment | -0.058 |
ados | -0.055 |
Majesty | -0.055 |
Cr | -0.054 |
gy | -0.052 |
Token | know |
Token ID | 760 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
Cr | +0.099 |
ados | +0.098 |
Majesty | +0.096 |
ishment | +0.093 |
lee | +0.088 |
Neg logit contributions | |
much | -0.093 |
lessly | -0.080 |
wonderful | -0.079 |
On | -0.078 |
br | -0.078 |
Token | the |
Token ID | 262 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
ados | +0.057 |
Cr | +0.056 |
Majesty | +0.055 |
ishment | +0.054 |
lee | +0.051 |
Neg logit contributions | |
much | -0.045 |
lessly | -0.038 |
wonderful | -0.038 |
On | -0.037 |
br | -0.036 |
Token | rh |
Token ID | 9529 |
Feature activation | -6.14e-02 |
Pos logit contributions | |
ados | +0.048 |
Majesty | +0.047 |
Cr | +0.047 |
ishment | +0.045 |
Fro | +0.043 |
Neg logit contributions | |
much | -0.046 |
lessly | -0.040 |
On | -0.040 |
In | -0.039 |
br | -0.039 |
Token | in |
Token ID | 259 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
ados | +0.125 |
Cr | +0.120 |
ishment | +0.115 |
Majesty | +0.114 |
ishes | +0.112 |
Neg logit contributions | |
much | -0.099 |
wonderful | -0.085 |
lessly | -0.084 |
On | -0.084 |
In | -0.081 |
Token | oc |
Token ID | 420 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
Cr | +0.062 |
ados | +0.061 |
Majesty | +0.059 |
ishment | +0.057 |
gy | +0.054 |
Neg logit contributions | |
much | -0.104 |
lessly | -0.093 |
wonderful | -0.091 |
On | -0.090 |
In | -0.089 |
Token | eros |
Token ID | 27498 |
Feature activation | -3.81e-02 |
Pos logit contributions | |
Cr | +0.120 |
ishment | +0.120 |
lee | +0.118 |
Jenna | +0.118 |
Fro | +0.116 |
Neg logit contributions | |
much | -0.337 |
lessly | -0.306 |
br | -0.306 |
hard | -0.303 |
ier | -0.300 |
Token | was |
Token ID | 373 |
Feature activation | +1.12e-04 |
Pos logit contributions | |
ados | +0.091 |
Cr | +0.091 |
Majesty | +0.088 |
ishment | +0.087 |
lee | +0.083 |
Neg logit contributions | |
much | -0.047 |
lessly | -0.038 |
wonderful | -0.037 |
On | -0.035 |
br | -0.035 |
Token | wild |
Token ID | 4295 |
Feature activation | -5.12e-02 |
Pos logit contributions | |
much | +0.000 |
lessly | +0.000 |
br | +0.000 |
In | +0.000 |
On | +0.000 |
Neg logit contributions | |
Cr | -0.000 |
ados | -0.000 |
Majesty | -0.000 |
ishment | -0.000 |
gy | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | -5.01e-02 |
Pos logit contributions | |
Cr | +0.115 |
ados | +0.113 |
Majesty | +0.113 |
ishment | +0.112 |
lee | +0.106 |
Neg logit contributions | |
much | -0.073 |
wonderful | -0.059 |
lessly | -0.058 |
hard | -0.055 |
br | -0.054 |
Token | Can |
Token ID | 1680 |
Feature activation | -6.77e-02 |
Pos logit contributions | |
ados | +0.094 |
Majesty | +0.092 |
ishment | +0.092 |
ishes | +0.090 |
ans | +0.088 |
Neg logit contributions | |
much | -0.096 |
On | -0.078 |
In | -0.077 |
He | -0.074 |
lessly | -0.072 |
Token | he |
Token ID | 339 |
Feature activation | -4.77e-02 |
Pos logit contributions | |
ados | +0.118 |
ishment | +0.117 |
Cr | +0.116 |
Majesty | +0.115 |
lee | +0.109 |
Neg logit contributions | |
much | -0.079 |
wonderful | -0.063 |
hard | -0.058 |
On | -0.057 |
br | -0.057 |
Token | would |
Token ID | 561 |
Feature activation | -7.18e-02 |
Pos logit contributions | |
ados | +0.112 |
Cr | +0.110 |
ishment | +0.108 |
Majesty | +0.107 |
lee | +0.102 |
Neg logit contributions | |
much | -0.064 |
wonderful | -0.050 |
lessly | -0.049 |
On | -0.048 |
br | -0.047 |
Token | take |
Token ID | 1011 |
Feature activation | -6.30e-02 |
Pos logit contributions | |
ados | +0.131 |
Cr | +0.129 |
ishment | +0.125 |
Majesty | +0.125 |
lee | +0.117 |
Neg logit contributions | |
much | -0.132 |
wonderful | -0.112 |
lessly | -0.111 |
On | -0.108 |
br | -0.107 |
Token | the |
Token ID | 262 |
Feature activation | -4.10e-02 |
Pos logit contributions | |
ados | +0.115 |
Cr | +0.114 |
ishment | +0.112 |
Majesty | +0.111 |
lee | +0.103 |
Neg logit contributions | |
much | -0.116 |
wonderful | -0.099 |
lessly | -0.097 |
br | -0.095 |
On | -0.093 |
Token | av |
Token ID | 1196 |
Feature activation | -9.68e-02 |
Pos logit contributions | |
ados | +0.071 |
Cr | +0.070 |
Majesty | +0.068 |
ishment | +0.068 |
lee | +0.062 |
Neg logit contributions | |
much | -0.078 |
lessly | -0.067 |
wonderful | -0.066 |
On | -0.065 |
In | -0.065 |
Token | oc |
Token ID | 420 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
Cr | +0.156 |
Majesty | +0.154 |
ishment | +0.151 |
ados | +0.148 |
lee | +0.143 |
Neg logit contributions | |
much | -0.197 |
br | -0.171 |
On | -0.167 |
wonderful | -0.166 |
lessly | -0.164 |
Token | ados |
Token ID | 22484 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
Fro | +0.092 |
Cr | +0.088 |
Rib | +0.080 |
fro | +0.074 |
ishment | +0.073 |
Neg logit contributions | |
much | -0.378 |
lessly | -0.348 |
br | -0.348 |
hard | -0.343 |
wonderful | -0.334 |
Token | and |
Token ID | 290 |
Feature activation | -6.40e-02 |
Pos logit contributions | |
ados | +0.041 |
Cr | +0.040 |
ishment | +0.038 |
Majesty | +0.037 |
gy | +0.035 |
Neg logit contributions | |
much | -0.042 |
wonderful | -0.037 |
lessly | -0.036 |
br | -0.035 |
In | -0.035 |
Token | scatter |
Token ID | 41058 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
Cr | +0.109 |
ados | +0.109 |
ishment | +0.107 |
Majesty | +0.106 |
lee | +0.099 |
Neg logit contributions | |
much | -0.124 |
wonderful | -0.107 |
lessly | -0.104 |
On | -0.102 |
br | -0.101 |
Token | them |
Token ID | 606 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
ados | +0.031 |
Cr | +0.030 |
ishment | +0.028 |
Majesty | +0.027 |
lee | +0.026 |
Neg logit contributions | |
much | -0.039 |
lessly | -0.034 |
wonderful | -0.034 |
On | -0.034 |
In | -0.033 |
Token | around |
Token ID | 1088 |
Feature activation | -4.96e-02 |
Pos logit contributions | |
ados | +0.033 |
Cr | +0.032 |
ishment | +0.032 |
Majesty | +0.031 |
gy | +0.030 |
Neg logit contributions | |
much | -0.022 |
lessly | -0.018 |
wonderful | -0.018 |
br | -0.018 |
In | -0.018 |
Token | She |
Token ID | 1375 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
ans | +0.065 |
ishment | +0.065 |
ados | +0.065 |
Majesty | +0.062 |
ishes | +0.062 |
Neg logit contributions | |
much | -0.065 |
On | -0.054 |
In | -0.052 |
He | -0.050 |
lessly | -0.050 |
Token | loved |
Token ID | 6151 |
Feature activation | -5.50e-02 |
Pos logit contributions | |
Cr | +0.038 |
ados | +0.038 |
Majesty | +0.037 |
ishment | +0.037 |
lee | +0.036 |
Neg logit contributions | |
much | -0.021 |
lessly | -0.017 |
wonderful | -0.016 |
br | -0.016 |
On | -0.015 |
Token | to |
Token ID | 284 |
Feature activation | -5.85e-02 |
Pos logit contributions | |
ishment | +0.135 |
Majesty | +0.125 |
cess | +0.119 |
ASH | +0.119 |
Cr | +0.119 |
Neg logit contributions | |
much | -0.084 |
wonderful | -0.070 |
hard | -0.068 |
big | -0.063 |
lot | -0.063 |
Token | eat |
Token ID | 4483 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
ishment | +0.111 |
ados | +0.110 |
Cr | +0.106 |
Majesty | +0.101 |
ASH | +0.097 |
Neg logit contributions | |
much | -0.114 |
wonderful | -0.100 |
hard | -0.094 |
On | -0.093 |
lot | -0.092 |
Token | av |
Token ID | 1196 |
Feature activation | -8.64e-02 |
Pos logit contributions | |
Majesty | +0.074 |
ishment | +0.073 |
Cr | +0.073 |
ados | +0.071 |
gy | +0.068 |
Neg logit contributions | |
much | -0.075 |
wonderful | -0.064 |
hard | -0.060 |
br | -0.060 |
On | -0.060 |
Token | oc |
Token ID | 420 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
Cr | +0.132 |
Majesty | +0.130 |
ishment | +0.128 |
ados | +0.128 |
lee | +0.119 |
Neg logit contributions | |
much | -0.186 |
wonderful | -0.159 |
br | -0.158 |
On | -0.156 |
In | -0.153 |
Token | ados |
Token ID | 22484 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
Fro | +0.089 |
ishment | +0.089 |
Cr | +0.082 |
gy | +0.080 |
Rib | +0.080 |
Neg logit contributions | |
much | -0.365 |
br | -0.339 |
hard | -0.328 |
wonderful | -0.325 |
ier | -0.312 |
Token | . |
Token ID | 13 |
Feature activation | -3.64e-02 |
Pos logit contributions | |
ados | +0.031 |
Cr | +0.029 |
ishment | +0.028 |
Majesty | +0.027 |
Fro | +0.026 |
Neg logit contributions | |
much | -0.027 |
lessly | -0.023 |
wonderful | -0.023 |
br | -0.022 |
In | -0.022 |
Token | One |
Token ID | 1881 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
keeping | +0.085 |
ados | +0.085 |
Majesty | +0.084 |
urg | +0.081 |
ishment | +0.080 |
Neg logit contributions | |
much | -0.051 |
On | -0.042 |
In | -0.039 |
One | -0.038 |
wonderful | -0.037 |
Token | day |
Token ID | 1110 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
much | +0.041 |
lessly | +0.034 |
wonderful | +0.034 |
br | +0.032 |
On | +0.032 |
Neg logit contributions | |
ados | -0.059 |
Cr | -0.058 |
Majesty | -0.056 |
ishment | -0.056 |
lee | -0.053 |
Token | , |
Token ID | 11 |
Feature activation | -5.29e-03 |
Pos logit contributions | |
much | +0.027 |
wonderful | +0.023 |
lessly | +0.022 |
br | +0.022 |
On | +0.022 |
Neg logit contributions | |
ados | -0.028 |
Cr | -0.028 |
ishment | -0.028 |
Majesty | -0.028 |
lee | -0.026 |
Token | am |
Token ID | 716 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
Cr | +0.081 |
Majesty | +0.079 |
ados | +0.075 |
ishment | +0.075 |
Fro | +0.075 |
Neg logit contributions | |
much | -0.055 |
lessly | -0.049 |
br | -0.046 |
wonderful | -0.045 |
On | -0.044 |
Token | taking |
Token ID | 2263 |
Feature activation | -6.92e-02 |
Pos logit contributions | |
ados | +0.028 |
Cr | +0.028 |
Majesty | +0.028 |
ishment | +0.027 |
lee | +0.026 |
Neg logit contributions | |
much | -0.018 |
lessly | -0.015 |
wonderful | -0.014 |
On | -0.014 |
br | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | -4.49e-02 |
Pos logit contributions | |
ados | +0.128 |
Cr | +0.127 |
ishment | +0.125 |
Majesty | +0.124 |
lee | +0.116 |
Neg logit contributions | |
much | -0.127 |
wonderful | -0.108 |
lessly | -0.104 |
br | -0.102 |
hard | -0.101 |
Token | rh |
Token ID | 9529 |
Feature activation | -6.28e-02 |
Pos logit contributions | |
ados | +0.077 |
Cr | +0.076 |
Majesty | +0.074 |
ishment | +0.073 |
Fro | +0.068 |
Neg logit contributions | |
much | -0.086 |
lessly | -0.074 |
wonderful | -0.073 |
On | -0.072 |
In | -0.071 |
Token | in |
Token ID | 259 |
Feature activation | -5.58e-02 |
Pos logit contributions | |
ados | +0.136 |
Cr | +0.133 |
Majesty | +0.128 |
ishment | +0.128 |
icans | +0.121 |
Neg logit contributions | |
much | -0.093 |
wonderful | -0.078 |
lessly | -0.077 |
On | -0.075 |
In | -0.073 |
Token | oc |
Token ID | 420 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
ados | +0.082 |
Cr | +0.081 |
ishment | +0.079 |
Majesty | +0.078 |
lee | +0.072 |
Neg logit contributions | |
much | -0.123 |
wonderful | -0.107 |
lessly | -0.105 |
br | -0.104 |
On | -0.103 |
Token | eros |
Token ID | 27498 |
Feature activation | -5.83e-02 |
Pos logit contributions | |
Cr | +0.124 |
Jenna | +0.121 |
ishment | +0.121 |
Fro | +0.117 |
Po | +0.115 |
Neg logit contributions | |
much | -0.325 |
br | -0.297 |
hard | -0.293 |
lessly | -0.292 |
ier | -0.286 |
Token | to |
Token ID | 284 |
Feature activation | -5.80e-02 |
Pos logit contributions | |
ados | +0.114 |
Cr | +0.113 |
ishment | +0.111 |
Majesty | +0.110 |
lee | +0.104 |
Neg logit contributions | |
much | -0.100 |
wonderful | -0.083 |
lessly | -0.081 |
br | -0.079 |
On | -0.079 |
Token | a |
Token ID | 257 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
ados | +0.106 |
Cr | +0.102 |
ishment | +0.102 |
Majesty | +0.099 |
lee | +0.096 |
Neg logit contributions | |
much | -0.110 |
wonderful | -0.093 |
lessly | -0.089 |
On | -0.088 |
br | -0.087 |
Token | new |
Token ID | 649 |
Feature activation | -6.07e-02 |
Pos logit contributions | |
Majesty | +0.060 |
Cr | +0.059 |
ishment | +0.057 |
ados | +0.057 |
Fro | +0.055 |
Neg logit contributions | |
much | -0.036 |
lessly | -0.036 |
In | -0.031 |
br | -0.030 |
On | -0.030 |
Token | home |
Token ID | 1363 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
ados | +0.100 |
Cr | +0.098 |
ishment | +0.094 |
Majesty | +0.093 |
lee | +0.088 |
Neg logit contributions | |
much | -0.124 |
wonderful | -0.108 |
lessly | -0.104 |
On | -0.104 |
br | -0.102 |
Token | 't |
Token ID | 470 |
Feature activation | -4.99e-02 |
Pos logit contributions | |
ados | +0.143 |
Cr | +0.142 |
Majesty | +0.138 |
ishment | +0.137 |
lee | +0.130 |
Neg logit contributions | |
much | -0.104 |
wonderful | -0.085 |
lessly | -0.085 |
On | -0.082 |
br | -0.081 |
Token | cry |
Token ID | 3960 |
Feature activation | -8.03e-02 |
Pos logit contributions | |
ados | +0.084 |
Cr | +0.082 |
ishment | +0.081 |
Majesty | +0.081 |
lee | +0.074 |
Neg logit contributions | |
much | -0.100 |
wonderful | -0.086 |
lessly | -0.085 |
On | -0.083 |
br | -0.082 |
Token | , |
Token ID | 11 |
Feature activation | -4.93e-02 |
Pos logit contributions | |
ados | +0.174 |
ishment | +0.172 |
Cr | +0.171 |
Majesty | +0.168 |
icans | +0.160 |
Neg logit contributions | |
much | -0.126 |
lessly | -0.099 |
hard | -0.096 |
wonderful | -0.094 |
bad | -0.089 |
Token | boys |
Token ID | 6510 |
Feature activation | -5.96e-02 |
Pos logit contributions | |
ishment | +0.089 |
ados | +0.088 |
Majesty | +0.086 |
lee | +0.080 |
gy | +0.079 |
Neg logit contributions | |
much | -0.099 |
wonderful | -0.085 |
br | -0.078 |
hard | -0.078 |
lessly | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | -5.44e-02 |
Pos logit contributions | |
ados | +0.132 |
Cr | +0.131 |
Majesty | +0.128 |
ishment | +0.128 |
lee | +0.120 |
Neg logit contributions | |
much | -0.087 |
wonderful | -0.070 |
lessly | -0.069 |
br | -0.065 |
On | -0.065 |
Token | Acc |
Token ID | 6366 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
ishment | +0.103 |
ados | +0.103 |
Majesty | +0.097 |
icans | +0.094 |
keeping | +0.094 |
Neg logit contributions | |
much | -0.103 |
In | -0.083 |
He | -0.080 |
wonderful | -0.080 |
It | -0.079 |
Token | idents |
Token ID | 3231 |
Feature activation | -5.54e-04 |
Pos logit contributions | |
ados | +0.147 |
keeping | +0.147 |
Fro | +0.146 |
alk | +0.146 |
Cr | +0.144 |
Neg logit contributions | |
much | -0.275 |
hard | -0.248 |
wrong | -0.242 |
bad | -0.241 |
better | -0.237 |
Token | happen |
Token ID | 1645 |
Feature activation | -3.39e-02 |
Pos logit contributions | |
Cr | +0.001 |
disl | +0.001 |
ados | +0.001 |
ors | +0.001 |
Twe | +0.001 |
Neg logit contributions | |
much | -0.001 |
On | -0.001 |
In | -0.001 |
br | -0.001 |
security | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -5.16e-02 |
Pos logit contributions | |
ados | +0.070 |
Cr | +0.069 |
Majesty | +0.067 |
ishment | +0.067 |
lee | +0.063 |
Neg logit contributions | |
much | -0.053 |
lessly | -0.045 |
wonderful | -0.044 |
On | -0.044 |
br | -0.043 |
Token | I |
Token ID | 314 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
ados | +0.094 |
ishment | +0.093 |
ican | +0.089 |
icans | +0.088 |
ans | +0.087 |
Neg logit contributions | |
much | -0.102 |
lessly | -0.081 |
In | -0.081 |
He | -0.080 |
wonderful | -0.080 |
Token | have |
Token ID | 423 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
Cr | +0.047 |
gy | +0.045 |
Majesty | +0.045 |
Fro | +0.045 |
disl | +0.043 |
Neg logit contributions | |
much | -0.040 |
br | -0.036 |
lessly | -0.035 |
On | -0.034 |
In | -0.033 |
Token | She |
Token ID | 1375 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
ados | +0.057 |
Majesty | +0.057 |
ans | +0.056 |
ishment | +0.056 |
ishes | +0.053 |
Neg logit contributions | |
much | -0.057 |
On | -0.047 |
In | -0.045 |
lessly | -0.043 |
wonderful | -0.043 |
Token | loved |
Token ID | 6151 |
Feature activation | -4.89e-02 |
Pos logit contributions | |
ados | +0.027 |
Cr | +0.027 |
Majesty | +0.026 |
ishment | +0.026 |
lee | +0.025 |
Neg logit contributions | |
much | -0.014 |
lessly | -0.011 |
wonderful | -0.011 |
br | -0.010 |
On | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | -5.48e-02 |
Pos logit contributions | |
ishment | +0.122 |
Majesty | +0.115 |
ASH | +0.108 |
Cr | +0.108 |
cess | +0.108 |
Neg logit contributions | |
much | -0.073 |
wonderful | -0.061 |
hard | -0.058 |
big | -0.055 |
lot | -0.053 |
Token | eat |
Token ID | 4483 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
ishment | +0.108 |
ados | +0.105 |
Cr | +0.101 |
Majesty | +0.097 |
ASH | +0.094 |
Neg logit contributions | |
much | -0.105 |
wonderful | -0.091 |
hard | -0.086 |
On | -0.085 |
lot | -0.084 |
Token | av |
Token ID | 1196 |
Feature activation | -8.22e-02 |
Pos logit contributions | |
Majesty | +0.067 |
ishment | +0.066 |
Cr | +0.066 |
ados | +0.065 |
gy | +0.062 |
Neg logit contributions | |
much | -0.065 |
wonderful | -0.055 |
hard | -0.052 |
br | -0.051 |
On | -0.051 |
Token | oc |
Token ID | 420 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
Cr | +0.124 |
Majesty | +0.123 |
ados | +0.121 |
ishment | +0.121 |
lee | +0.113 |
Neg logit contributions | |
much | -0.177 |
wonderful | -0.152 |
br | -0.151 |
On | -0.149 |
In | -0.147 |
Token | ados |
Token ID | 22484 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
ishment | +0.090 |
Fro | +0.089 |
gy | +0.083 |
Rib | +0.082 |
Cr | +0.082 |
Neg logit contributions | |
much | -0.352 |
br | -0.327 |
hard | -0.318 |
wonderful | -0.315 |
ier | -0.305 |
Token | because |
Token ID | 780 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
ados | +0.023 |
Cr | +0.022 |
ishment | +0.021 |
Majesty | +0.021 |
Fro | +0.020 |
Neg logit contributions | |
much | -0.021 |
lessly | -0.018 |
wonderful | -0.018 |
br | -0.017 |
In | -0.017 |
Token | they |
Token ID | 484 |
Feature activation | +1.11e-03 |
Pos logit contributions | |
much | +0.032 |
lessly | +0.028 |
wonderful | +0.028 |
br | +0.027 |
On | +0.027 |
Neg logit contributions | |
ados | -0.034 |
Cr | -0.033 |
ishment | -0.030 |
Majesty | -0.030 |
lee | -0.029 |
Token | were |
Token ID | 547 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
much | +0.001 |
lessly | +0.001 |
br | +0.001 |
ching | +0.001 |
On | +0.001 |
Neg logit contributions | |
Cr | -0.003 |
ados | -0.003 |
Majesty | -0.002 |
ishment | -0.002 |
gy | -0.002 |
Token | y |
Token ID | 331 |
Feature activation | -6.34e-02 |
Pos logit contributions | |
much | +0.017 |
wonderful | +0.013 |
lessly | +0.012 |
On | +0.012 |
br | +0.012 |
Neg logit contributions | |
ados | -0.038 |
Cr | -0.038 |
ishment | -0.037 |
Majesty | -0.037 |
lee | -0.035 |
Token | it |
Token ID | 340 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
Majesty | +0.098 |
ishment | +0.097 |
Cr | +0.097 |
ados | +0.097 |
ican | +0.089 |
Neg logit contributions | |
much | -0.124 |
wonderful | -0.107 |
lessly | -0.102 |
hard | -0.102 |
lot | -0.096 |
Token | to |
Token ID | 284 |
Feature activation | -6.23e-02 |
Pos logit contributions | |
ishment | +0.024 |
ados | +0.024 |
Majesty | +0.023 |
Cr | +0.023 |
ican | +0.022 |
Neg logit contributions | |
much | -0.021 |
wonderful | -0.017 |
hard | -0.016 |
lessly | -0.015 |
bad | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | -5.14e-02 |
Pos logit contributions | |
ados | +0.101 |
ishment | +0.099 |
Majesty | +0.097 |
Cr | +0.096 |
amb | +0.090 |
Neg logit contributions | |
much | -0.130 |
wonderful | -0.112 |
hard | -0.105 |
lessly | -0.104 |
lot | -0.104 |
Token | rh |
Token ID | 9529 |
Feature activation | -5.86e-02 |
Pos logit contributions | |
ados | +0.087 |
Cr | +0.086 |
Majesty | +0.084 |
ishment | +0.083 |
lee | +0.076 |
Neg logit contributions | |
much | -0.100 |
lessly | -0.087 |
wonderful | -0.086 |
On | -0.084 |
br | -0.084 |
Token | in |
Token ID | 259 |
Feature activation | -6.13e-02 |
Pos logit contributions | |
ados | +0.123 |
Cr | +0.121 |
Majesty | +0.117 |
ishment | +0.117 |
gy | +0.110 |
Neg logit contributions | |
much | -0.088 |
wonderful | -0.075 |
lessly | -0.075 |
On | -0.073 |
In | -0.071 |
Token | oc |
Token ID | 420 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
ados | +0.103 |
Cr | +0.099 |
ishment | +0.098 |
Majesty | +0.095 |
lee | +0.092 |
Neg logit contributions | |
much | -0.127 |
wonderful | -0.107 |
On | -0.103 |
lessly | -0.103 |
hard | -0.102 |
Token | eros |
Token ID | 27498 |
Feature activation | -5.52e-02 |
Pos logit contributions | |
Jenna | +0.123 |
ishment | +0.120 |
Cr | +0.119 |
lee | +0.114 |
Po | +0.114 |
Neg logit contributions | |
much | -0.307 |
hard | -0.280 |
br | -0.273 |
lessly | -0.272 |
wonderful | -0.269 |
Token | . |
Token ID | 13 |
Feature activation | -4.47e-02 |
Pos logit contributions | |
ados | +0.130 |
ishment | +0.127 |
Cr | +0.126 |
Majesty | +0.124 |
lee | +0.118 |
Neg logit contributions | |
much | -0.077 |
wonderful | -0.059 |
lessly | -0.056 |
On | -0.054 |
hard | -0.054 |
Token | When |
Token ID | 1649 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
ishment | +0.086 |
ados | +0.084 |
Majesty | +0.083 |
ishes | +0.081 |
ans | +0.080 |
Neg logit contributions | |
much | -0.082 |
On | -0.067 |
In | -0.065 |
He | -0.064 |
lessly | -0.061 |
Token | he |
Token ID | 339 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
much | +0.020 |
wonderful | +0.016 |
hard | +0.015 |
lot | +0.014 |
br | +0.014 |
Neg logit contributions | |
ishment | -0.029 |
Majesty | -0.027 |
ados | -0.027 |
ans | -0.027 |
amb | -0.026 |
Token | tasted |
Token ID | 29187 |
Feature activation | -8.48e-03 |
Pos logit contributions | |
Cr | +0.028 |
ados | +0.027 |
Majesty | +0.027 |
ishment | +0.026 |
Fro | +0.026 |
Neg logit contributions | |
much | -0.020 |
lessly | -0.017 |
wonderful | -0.016 |
br | -0.015 |
On | -0.015 |
Token | ," |
Token ID | 553 |
Feature activation | -4.93e-02 |
Pos logit contributions | |
ados | +0.070 |
Cr | +0.067 |
Majesty | +0.064 |
ishment | +0.064 |
Fro | +0.058 |
Neg logit contributions | |
much | -0.077 |
lessly | -0.071 |
In | -0.068 |
On | -0.068 |
wonderful | -0.068 |
Token | said |
Token ID | 531 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
Majesty | +0.089 |
ados | +0.087 |
ishment | +0.084 |
lee | +0.080 |
ans | +0.080 |
Neg logit contributions | |
much | -0.096 |
wonderful | -0.079 |
hard | -0.078 |
bad | -0.077 |
On | -0.076 |
Token | the |
Token ID | 262 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
Majesty | +0.048 |
ishment | +0.047 |
ados | +0.046 |
ans | +0.043 |
ishes | +0.043 |
Neg logit contributions | |
much | -0.061 |
hard | -0.050 |
wonderful | -0.050 |
bad | -0.049 |
br | -0.049 |
Token | rh |
Token ID | 9529 |
Feature activation | -4.25e-02 |
Pos logit contributions | |
ados | +0.021 |
Cr | +0.021 |
ishment | +0.020 |
Majesty | +0.020 |
lee | +0.019 |
Neg logit contributions | |
much | -0.023 |
wonderful | -0.020 |
lessly | -0.020 |
br | -0.019 |
On | -0.019 |
Token | in |
Token ID | 259 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
ados | +0.087 |
Cr | +0.086 |
Majesty | +0.082 |
ishment | +0.081 |
Rib | +0.078 |
Neg logit contributions | |
much | -0.066 |
wonderful | -0.057 |
lessly | -0.056 |
On | -0.055 |
In | -0.054 |
Token | oc |
Token ID | 420 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
ados | +0.062 |
Cr | +0.062 |
Majesty | +0.060 |
ishment | +0.060 |
lee | +0.056 |
Neg logit contributions | |
much | -0.060 |
wonderful | -0.051 |
lessly | -0.051 |
On | -0.049 |
br | -0.049 |
Token | eros |
Token ID | 27498 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
Jenna | +0.146 |
ishment | +0.134 |
Rachel | +0.131 |
lee | +0.131 |
Harry | +0.128 |
Neg logit contributions | |
much | -0.280 |
hard | -0.263 |
br | -0.263 |
ier | -0.260 |
lessly | -0.252 |
Token | . |
Token ID | 13 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
ados | +0.047 |
Cr | +0.047 |
Majesty | +0.046 |
ishment | +0.045 |
lee | +0.043 |
Neg logit contributions | |
much | -0.035 |
wonderful | -0.029 |
lessly | -0.029 |
On | -0.028 |
br | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
ados | +0.096 |
ishment | +0.094 |
Majesty | +0.094 |
Cr | +0.092 |
lee | +0.087 |
Neg logit contributions | |
much | -0.073 |
lessly | -0.059 |
On | -0.058 |
wonderful | -0.057 |
br | -0.056 |
Token |
|
Token ID | 198 |
Feature activation | +5.37e-03 |
Pos logit contributions | |
much | +0.015 |
wonderful | +0.012 |
lessly | +0.012 |
br | +0.012 |
On | +0.012 |
Neg logit contributions | |
ados | -0.025 |
Cr | -0.025 |
Majesty | -0.024 |
ishment | -0.024 |
lee | -0.023 |
Token | The |
Token ID | 464 |
Feature activation | -1.94e-03 |
Pos logit contributions | |
much | +0.007 |
wonderful | +0.005 |
lessly | +0.005 |
On | +0.005 |
br | +0.005 |
Neg logit contributions | |
ados | -0.013 |
Cr | -0.013 |
Majesty | -0.013 |
ishment | -0.013 |
lee | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
ados | +0.031 |
Cr | +0.031 |
Majesty | +0.030 |
ishment | +0.030 |
lee | +0.028 |
Neg logit contributions | |
much | -0.018 |
wonderful | -0.014 |
lessly | -0.014 |
On | -0.014 |
br | -0.014 |
Token | " |
Token ID | 1 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
ados | +0.032 |
Cr | +0.032 |
ishment | +0.031 |
Majesty | +0.031 |
lee | +0.029 |
Neg logit contributions | |
much | -0.015 |
wonderful | -0.012 |
lessly | -0.011 |
On | -0.011 |
In | -0.011 |
Token | The |
Token ID | 464 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
Cr | +0.046 |
ados | +0.044 |
ishment | +0.043 |
Majesty | +0.042 |
Rib | +0.041 |
Neg logit contributions | |
much | -0.036 |
lessly | -0.029 |
wonderful | -0.029 |
hard | -0.028 |
br | -0.027 |
Token | rh |
Token ID | 9529 |
Feature activation | -4.36e-02 |
Pos logit contributions | |
Majesty | +0.027 |
ados | +0.027 |
Cr | +0.027 |
ishment | +0.025 |
Fro | +0.024 |
Neg logit contributions | |
much | -0.033 |
lessly | -0.029 |
On | -0.028 |
In | -0.028 |
br | -0.028 |
Token | in |
Token ID | 259 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
ados | +0.091 |
Cr | +0.088 |
Majesty | +0.085 |
ishment | +0.083 |
ican | +0.080 |
Neg logit contributions | |
much | -0.069 |
lessly | -0.058 |
wonderful | -0.058 |
On | -0.057 |
In | -0.055 |
Token | oc |
Token ID | 420 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
ados | +0.052 |
Cr | +0.051 |
Majesty | +0.050 |
ishment | +0.049 |
lee | +0.045 |
Neg logit contributions | |
much | -0.078 |
lessly | -0.068 |
wonderful | -0.068 |
On | -0.067 |
br | -0.066 |
Token | eros |
Token ID | 27498 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
Jenna | +0.132 |
ishment | +0.130 |
lee | +0.128 |
Rachel | +0.118 |
Cr | +0.115 |
Neg logit contributions | |
much | -0.280 |
br | -0.264 |
hard | -0.262 |
ier | -0.256 |
lessly | -0.252 |
Token | is |
Token ID | 318 |
Feature activation | +4.48e-03 |
Pos logit contributions | |
ados | +0.069 |
Cr | +0.069 |
Majesty | +0.067 |
ishment | +0.067 |
lee | +0.063 |
Neg logit contributions | |
much | -0.045 |
lessly | -0.036 |
wonderful | -0.036 |
On | -0.035 |
br | -0.035 |
Token | in |
Token ID | 287 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
much | +0.006 |
lessly | +0.006 |
In | +0.006 |
On | +0.005 |
br | +0.005 |
Neg logit contributions | |
Cr | -0.009 |
Majesty | -0.009 |
ados | -0.009 |
ishment | -0.009 |
Fro | -0.009 |
Token | there |
Token ID | 612 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
ados | +0.041 |
Cr | +0.040 |
ishment | +0.038 |
Majesty | +0.038 |
Fro | +0.035 |
Neg logit contributions | |
much | -0.055 |
lessly | -0.051 |
br | -0.048 |
wonderful | -0.048 |
In | -0.048 |
Token | , |
Token ID | 11 |
Feature activation | -3.91e-02 |
Pos logit contributions | |
ados | +0.027 |
Cr | +0.027 |
Majesty | +0.026 |
ishment | +0.026 |
Fro | +0.024 |
Neg logit contributions | |
much | -0.028 |
lessly | -0.025 |
wonderful | -0.024 |
br | -0.024 |
On | -0.023 |
Token | 't |
Token ID | 470 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
much | +0.003 |
lessly | +0.002 |
wonderful | +0.002 |
On | +0.002 |
br | +0.002 |
Neg logit contributions | |
Cr | -0.003 |
ados | -0.003 |
Fro | -0.002 |
ican | -0.002 |
gy | -0.002 |
Token | touch |
Token ID | 3638 |
Feature activation | -4.25e-02 |
Pos logit contributions | |
ados | +0.039 |
Cr | +0.038 |
Majesty | +0.037 |
ishment | +0.037 |
lee | +0.034 |
Neg logit contributions | |
much | -0.045 |
wonderful | -0.038 |
lessly | -0.038 |
On | -0.037 |
br | -0.037 |
Token | the |
Token ID | 262 |
Feature activation | -3.25e-02 |
Pos logit contributions | |
Cr | +0.075 |
Majesty | +0.074 |
ados | +0.074 |
ishment | +0.074 |
lee | +0.068 |
Neg logit contributions | |
much | -0.081 |
wonderful | -0.068 |
lessly | -0.066 |
br | -0.065 |
hard | -0.064 |
Token | rh |
Token ID | 9529 |
Feature activation | -4.58e-02 |
Pos logit contributions | |
ados | +0.052 |
Cr | +0.050 |
Majesty | +0.049 |
ishment | +0.049 |
Fro | +0.045 |
Neg logit contributions | |
much | -0.067 |
lessly | -0.058 |
wonderful | -0.057 |
On | -0.057 |
In | -0.056 |
Token | in |
Token ID | 259 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
ados | +0.098 |
Cr | +0.094 |
Majesty | +0.090 |
ishment | +0.090 |
ican | +0.086 |
Neg logit contributions | |
much | -0.071 |
wonderful | -0.060 |
lessly | -0.060 |
On | -0.058 |
In | -0.056 |
Token | oc |
Token ID | 420 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
ados | +0.056 |
Cr | +0.055 |
Majesty | +0.053 |
ishment | +0.053 |
lee | +0.048 |
Neg logit contributions | |
much | -0.088 |
wonderful | -0.077 |
lessly | -0.077 |
On | -0.075 |
br | -0.075 |
Token | eros |
Token ID | 27498 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
Jenna | +0.124 |
ishment | +0.117 |
lee | +0.116 |
Rachel | +0.115 |
Cr | +0.114 |
Neg logit contributions | |
much | -0.286 |
br | -0.268 |
hard | -0.265 |
ier | -0.262 |
ching | -0.255 |
Token | ! |
Token ID | 0 |
Feature activation | -5.01e-02 |
Pos logit contributions | |
Cr | +0.074 |
ados | +0.074 |
ishment | +0.072 |
Majesty | +0.072 |
lee | +0.068 |
Neg logit contributions | |
much | -0.063 |
wonderful | -0.052 |
lessly | -0.051 |
br | -0.050 |
On | -0.049 |
Token | It |
Token ID | 632 |
Feature activation | +9.99e-03 |
Pos logit contributions | |
ishment | +0.087 |
ados | +0.086 |
Majesty | +0.082 |
lee | +0.078 |
ishes | +0.077 |
Neg logit contributions | |
much | -0.102 |
On | -0.083 |
In | -0.083 |
lessly | -0.082 |
He | -0.081 |
Token | might |
Token ID | 1244 |
Feature activation | -6.28e-03 |
Pos logit contributions | |
much | +0.016 |
lessly | +0.013 |
wonderful | +0.013 |
On | +0.012 |
br | +0.012 |
Neg logit contributions | |
ados | -0.021 |
Cr | -0.021 |
Majesty | -0.020 |
ishment | -0.020 |
lee | -0.019 |
Token | bite |
Token ID | 13197 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
Cr | +0.011 |
ados | +0.011 |
Majesty | +0.010 |
Fro | +0.010 |
gy | +0.010 |
Neg logit contributions | |
much | -0.011 |
lessly | -0.010 |
On | -0.010 |
wonderful | -0.009 |
In | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -4.44e-02 |
Pos logit contributions | |
ados | +0.061 |
Cr | +0.061 |
Majesty | +0.059 |
ishment | +0.059 |
lee | +0.055 |
Neg logit contributions | |
much | -0.053 |
wonderful | -0.044 |
lessly | -0.044 |
On | -0.042 |
br | -0.042 |
Token | And |
Token ID | 843 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
ados | +0.076 |
ishment | +0.073 |
lee | +0.071 |
Cr | +0.070 |
Majesty | +0.069 |
Neg logit contributions | |
much | -0.089 |
On | -0.074 |
In | -0.074 |
wonderful | -0.073 |
br | -0.072 |
Token | put |
Token ID | 1234 |
Feature activation | -4.09e-02 |
Pos logit contributions | |
Cr | +0.044 |
ados | +0.044 |
Majesty | +0.044 |
ishment | +0.042 |
Fro | +0.039 |
Neg logit contributions | |
much | -0.053 |
lessly | -0.048 |
br | -0.046 |
wonderful | -0.045 |
ching | -0.045 |
Token | on |
Token ID | 319 |
Feature activation | -4.94e-02 |
Pos logit contributions | |
ados | +0.066 |
Cr | +0.064 |
Majesty | +0.063 |
ishment | +0.062 |
lee | +0.057 |
Neg logit contributions | |
much | -0.084 |
lessly | -0.072 |
wonderful | -0.072 |
On | -0.071 |
br | -0.070 |
Token | your |
Token ID | 534 |
Feature activation | -8.03e-02 |
Pos logit contributions | |
Cr | +0.078 |
ishment | +0.076 |
ados | +0.074 |
Majesty | +0.072 |
ican | +0.071 |
Neg logit contributions | |
much | -0.107 |
wonderful | -0.094 |
hard | -0.092 |
lot | -0.086 |
lessly | -0.086 |
Token | swe |
Token ID | 3490 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
Cr | +0.125 |
ados | +0.123 |
ishment | +0.118 |
Majesty | +0.114 |
lee | +0.107 |
Neg logit contributions | |
much | -0.171 |
wonderful | -0.153 |
hard | -0.145 |
br | -0.144 |
lessly | -0.144 |
Token | aters |
Token ID | 8605 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
Cr | +0.195 |
ican | +0.182 |
Rib | +0.171 |
ados | +0.169 |
icans | +0.167 |
Neg logit contributions | |
much | -0.244 |
lot | -0.208 |
On | -0.207 |
wonderful | -0.207 |
In | -0.198 |
Token | , |
Token ID | 11 |
Feature activation | -5.42e-02 |
Pos logit contributions | |
ados | +0.063 |
Cr | +0.062 |
Majesty | +0.061 |
ishment | +0.061 |
lee | +0.057 |
Neg logit contributions | |
much | -0.059 |
wonderful | -0.050 |
lessly | -0.050 |
On | -0.048 |
br | -0.048 |
Token | or |
Token ID | 393 |
Feature activation | -6.92e-02 |
Pos logit contributions | |
ados | +0.100 |
Cr | +0.100 |
Majesty | +0.097 |
ishment | +0.097 |
lee | +0.090 |
Neg logit contributions | |
much | -0.100 |
wonderful | -0.084 |
lessly | -0.082 |
On | -0.080 |
hard | -0.079 |
Token | you |
Token ID | 345 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
ados | +0.122 |
Cr | +0.120 |
Majesty | +0.117 |
ishment | +0.116 |
lee | +0.108 |
Neg logit contributions | |
much | -0.130 |
lessly | -0.112 |
wonderful | -0.111 |
On | -0.108 |
br | -0.107 |
Token | will |
Token ID | 481 |
Feature activation | -4.39e-02 |
Pos logit contributions | |
Cr | +0.026 |
Majesty | +0.026 |
ados | +0.026 |
gy | +0.026 |
ishment | +0.025 |
Neg logit contributions | |
much | -0.018 |
lessly | -0.016 |
wonderful | -0.015 |
On | -0.015 |
br | -0.014 |
Token | Did |
Token ID | 7731 |
Feature activation | -3.83e-02 |
Pos logit contributions | |
ishment | +0.075 |
ados | +0.072 |
ans | +0.070 |
icans | +0.067 |
Majesty | +0.067 |
Neg logit contributions | |
much | -0.076 |
In | -0.065 |
br | -0.061 |
lessly | -0.060 |
It | -0.059 |
Token | you |
Token ID | 345 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
ados | +0.059 |
ishment | +0.058 |
Cr | +0.057 |
Majesty | +0.057 |
keeping | +0.053 |
Neg logit contributions | |
much | -0.082 |
wonderful | -0.072 |
hard | -0.068 |
br | -0.067 |
lessly | -0.066 |
Token | know |
Token ID | 760 |
Feature activation | -6.31e-03 |
Pos logit contributions | |
ados | +0.074 |
Cr | +0.074 |
Majesty | +0.072 |
ishment | +0.071 |
lee | +0.067 |
Neg logit contributions | |
much | -0.069 |
wonderful | -0.058 |
lessly | -0.058 |
On | -0.056 |
br | -0.056 |
Token | that |
Token ID | 326 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
ados | +0.011 |
Cr | +0.011 |
Majesty | +0.010 |
ishment | +0.010 |
lee | +0.009 |
Neg logit contributions | |
much | -0.011 |
lessly | -0.011 |
On | -0.010 |
In | -0.010 |
br | -0.010 |
Token | av |
Token ID | 1196 |
Feature activation | -7.54e-02 |
Pos logit contributions | |
ados | +0.020 |
Cr | +0.020 |
Majesty | +0.019 |
ishment | +0.019 |
lee | +0.018 |
Neg logit contributions | |
much | -0.023 |
wonderful | -0.020 |
lessly | -0.020 |
On | -0.019 |
br | -0.019 |
Token | oc |
Token ID | 420 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
ados | +0.106 |
Cr | +0.105 |
Majesty | +0.101 |
ishment | +0.100 |
lee | +0.091 |
Neg logit contributions | |
much | -0.170 |
wonderful | -0.149 |
lessly | -0.148 |
On | -0.145 |
br | -0.144 |
Token | ados |
Token ID | 22484 |
Feature activation | +8.05e-05 |
Pos logit contributions | |
ishment | +0.064 |
Fro | +0.059 |
Cr | +0.052 |
Rib | +0.049 |
fro | +0.045 |
Neg logit contributions | |
much | -0.359 |
br | -0.335 |
hard | -0.325 |
lessly | -0.322 |
wonderful | -0.322 |
Token | are |
Token ID | 389 |
Feature activation | +6.97e-03 |
Pos logit contributions | |
much | +0.000 |
lessly | +0.000 |
On | +0.000 |
wonderful | +0.000 |
In | +0.000 |
Neg logit contributions | |
ados | -0.000 |
Cr | -0.000 |
Majesty | -0.000 |
ishment | -0.000 |
ishes | -0.000 |
Token | very |
Token ID | 845 |
Feature activation | -3.02e-03 |
Pos logit contributions | |
lessly | +0.009 |
much | +0.009 |
On | +0.008 |
Over | +0.008 |
ching | +0.008 |
Neg logit contributions | |
ados | -0.015 |
Cr | -0.015 |
Majesty | -0.014 |
ishment | -0.013 |
Harry | -0.013 |
Token | healthy |
Token ID | 5448 |
Feature activation | -6.41e-02 |
Pos logit contributions | |
Cr | +0.008 |
ados | +0.008 |
Majesty | +0.007 |
ishment | +0.007 |
lee | +0.007 |
Neg logit contributions | |
much | -0.003 |
lessly | -0.002 |
br | -0.002 |
On | -0.002 |
ching | -0.002 |
Token | for |
Token ID | 329 |
Feature activation | -3.33e-02 |
Pos logit contributions | |
ados | +0.123 |
Cr | +0.122 |
Majesty | +0.119 |
ishment | +0.118 |
lee | +0.111 |
Neg logit contributions | |
much | -0.111 |
wonderful | -0.093 |
lessly | -0.093 |
On | -0.090 |
br | -0.089 |
Token | s |
Token ID | 82 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
Cr | +0.072 |
ados | +0.067 |
Fro | +0.067 |
lee | +0.065 |
gy | +0.064 |
Neg logit contributions | |
much | -0.042 |
wonderful | -0.038 |
br | -0.036 |
lessly | -0.035 |
On | -0.033 |
Token | wrong |
Token ID | 2642 |
Feature activation | -3.66e-02 |
Pos logit contributions | |
ados | +0.046 |
Cr | +0.046 |
Majesty | +0.045 |
ishment | +0.044 |
lee | +0.042 |
Neg logit contributions | |
much | -0.033 |
lessly | -0.028 |
br | -0.027 |
On | -0.027 |
wonderful | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
ados | +0.066 |
Cr | +0.066 |
Majesty | +0.064 |
ishment | +0.064 |
lee | +0.059 |
Neg logit contributions | |
much | -0.066 |
lessly | -0.058 |
wonderful | -0.056 |
br | -0.055 |
On | -0.054 |
Token | rh |
Token ID | 9529 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
Cr | +0.027 |
ados | +0.027 |
Majesty | +0.026 |
ishment | +0.026 |
Fro | +0.024 |
Neg logit contributions | |
much | -0.039 |
lessly | -0.035 |
br | -0.034 |
On | -0.034 |
wonderful | -0.034 |
Token | in |
Token ID | 259 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
ados | +0.079 |
Cr | +0.075 |
Majesty | +0.073 |
ican | +0.072 |
Fro | +0.071 |
Neg logit contributions | |
much | -0.068 |
lessly | -0.059 |
On | -0.057 |
wonderful | -0.057 |
In | -0.055 |
Token | oc |
Token ID | 420 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
Cr | +0.050 |
Majesty | +0.048 |
ados | +0.048 |
ishment | +0.047 |
Fro | +0.045 |
Neg logit contributions | |
much | -0.057 |
lessly | -0.052 |
wonderful | -0.049 |
On | -0.048 |
br | -0.048 |
Token | eros |
Token ID | 27498 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
ishment | +0.126 |
Jenna | +0.106 |
Cr | +0.104 |
Harry | +0.098 |
lee | +0.096 |
Neg logit contributions | |
much | -0.284 |
br | -0.277 |
hard | -0.264 |
ier | -0.259 |
lessly | -0.257 |
Token | ? |
Token ID | 30 |
Feature activation | -3.40e-02 |
Pos logit contributions | |
ados | +0.051 |
Cr | +0.051 |
Majesty | +0.049 |
ishment | +0.048 |
lee | +0.045 |
Neg logit contributions | |
much | -0.058 |
lessly | -0.051 |
wonderful | -0.050 |
On | -0.049 |
br | -0.048 |
Token | â |
Token ID | 22940 |
Feature activation | -6.43e-02 |
Pos logit contributions | |
ados | +0.055 |
ishment | +0.053 |
Majesty | +0.053 |
Cr | +0.051 |
lee | +0.049 |
Neg logit contributions | |
much | -0.070 |
lessly | -0.059 |
On | -0.059 |
wonderful | -0.059 |
In | -0.058 |
Token | € |
Token ID | 26391 |
Feature activation | -6.00e-02 |
Pos logit contributions | |
ados | +0.092 |
Cr | +0.089 |
Majesty | +0.087 |
ishment | +0.085 |
Fro | +0.078 |
Neg logit contributions | |
much | -0.142 |
lessly | -0.126 |
wonderful | -0.125 |
In | -0.122 |
On | -0.122 |
Token | said |
Token ID | 531 |
Feature activation | -2.16e-03 |
Pos logit contributions | |
ados | +0.129 |
ishment | +0.126 |
Cr | +0.125 |
Majesty | +0.120 |
lee | +0.114 |
Neg logit contributions | |
much | -0.096 |
On | -0.075 |
In | -0.075 |
wonderful | -0.075 |
br | -0.074 |
Token | that |
Token ID | 326 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
ados | +0.126 |
Cr | +0.124 |
Majesty | +0.122 |
ishment | +0.121 |
lee | +0.114 |
Neg logit contributions | |
much | -0.092 |
wonderful | -0.075 |
lessly | -0.075 |
On | -0.071 |
br | -0.071 |
Token | the |
Token ID | 262 |
Feature activation | -9.06e-03 |
Pos logit contributions | |
Cr | +0.019 |
ados | +0.019 |
Majesty | +0.018 |
ishment | +0.018 |
lee | +0.017 |
Neg logit contributions | |
much | -0.019 |
lessly | -0.017 |
br | -0.016 |
On | -0.016 |
wonderful | -0.016 |
Token | bad |
Token ID | 2089 |
Feature activation | -4.25e-02 |
Pos logit contributions | |
ados | +0.016 |
Cr | +0.016 |
Majesty | +0.016 |
ishment | +0.016 |
lee | +0.015 |
Neg logit contributions | |
much | -0.016 |
lessly | -0.014 |
On | -0.014 |
br | -0.014 |
In | -0.014 |
Token | rh |
Token ID | 9529 |
Feature activation | -3.92e-02 |
Pos logit contributions | |
ados | +0.084 |
Cr | +0.083 |
ishment | +0.081 |
Majesty | +0.080 |
lee | +0.075 |
Neg logit contributions | |
much | -0.073 |
wonderful | -0.062 |
lessly | -0.059 |
hard | -0.058 |
br | -0.057 |
Token | in |
Token ID | 259 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
ados | +0.082 |
Cr | +0.081 |
Majesty | +0.077 |
ishment | +0.075 |
ican | +0.073 |
Neg logit contributions | |
much | -0.059 |
wonderful | -0.051 |
On | -0.050 |
lessly | -0.049 |
In | -0.048 |
Token | oc |
Token ID | 420 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
Cr | +0.041 |
ados | +0.040 |
Majesty | +0.039 |
ishment | +0.038 |
gy | +0.036 |
Neg logit contributions | |
much | -0.062 |
lessly | -0.055 |
wonderful | -0.055 |
On | -0.054 |
br | -0.053 |
Token | eros |
Token ID | 27498 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
ishment | +0.109 |
Cr | +0.102 |
lee | +0.099 |
Fro | +0.098 |
Jenna | +0.096 |
Neg logit contributions | |
much | -0.301 |
lessly | -0.276 |
br | -0.273 |
hard | -0.270 |
ier | -0.269 |
Token | was |
Token ID | 373 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
ados | +0.049 |
Cr | +0.049 |
Majesty | +0.048 |
ishment | +0.048 |
lee | +0.046 |
Neg logit contributions | |
much | -0.025 |
lessly | -0.020 |
wonderful | -0.019 |
On | -0.019 |
br | -0.019 |
Token | now |
Token ID | 783 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
much | +0.014 |
br | +0.013 |
On | +0.013 |
lessly | +0.013 |
In | +0.012 |
Neg logit contributions | |
Cr | -0.021 |
Majesty | -0.020 |
ados | -0.019 |
lee | -0.019 |
ishment | -0.019 |
Token | nice |
Token ID | 3621 |
Feature activation | -3.78e-02 |
Pos logit contributions | |
ados | +0.037 |
Cr | +0.037 |
Majesty | +0.036 |
ishment | +0.035 |
lee | +0.034 |
Neg logit contributions | |
much | -0.027 |
br | -0.026 |
On | -0.025 |
In | -0.025 |
lessly | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
ados | +0.073 |
Cr | +0.071 |
Majesty | +0.069 |
ishment | +0.069 |
lee | +0.065 |
Neg logit contributions | |
much | -0.065 |
lessly | -0.055 |
wonderful | -0.055 |
On | -0.054 |
br | -0.053 |
Token | to |
Token ID | 284 |
Feature activation | -5.29e-02 |
Pos logit contributions | |
ados | +0.064 |
Cr | +0.063 |
ishment | +0.062 |
Majesty | +0.062 |
lee | +0.057 |
Neg logit contributions | |
much | -0.059 |
wonderful | -0.049 |
lessly | -0.049 |
On | -0.048 |
br | -0.047 |
Token | see |
Token ID | 766 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
ados | +0.135 |
Majesty | +0.124 |
ishment | +0.122 |
amb | +0.118 |
Cr | +0.116 |
Neg logit contributions | |
much | -0.077 |
wonderful | -0.068 |
On | -0.064 |
big | -0.060 |
hard | -0.059 |
Token | the |
Token ID | 262 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
Cr | +0.037 |
ados | +0.036 |
ishment | +0.035 |
Majesty | +0.035 |
Fro | +0.033 |
Neg logit contributions | |
much | -0.031 |
lessly | -0.029 |
br | -0.028 |
In | -0.027 |
On | -0.027 |
Token | rh |
Token ID | 9529 |
Feature activation | -3.77e-02 |
Pos logit contributions | |
ados | +0.020 |
Cr | +0.020 |
Majesty | +0.020 |
ishment | +0.019 |
icans | +0.019 |
Neg logit contributions | |
much | -0.020 |
lessly | -0.018 |
br | -0.017 |
In | -0.017 |
On | -0.017 |
Token | in |
Token ID | 259 |
Feature activation | -2.60e-02 |
Pos logit contributions | |
ados | +0.075 |
Cr | +0.072 |
ishment | +0.070 |
Majesty | +0.069 |
icans | +0.068 |
Neg logit contributions | |
much | -0.061 |
lessly | -0.053 |
wonderful | -0.053 |
On | -0.049 |
br | -0.048 |
Token | oc |
Token ID | 420 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
Cr | +0.028 |
ados | +0.027 |
ishment | +0.025 |
Majesty | +0.024 |
gy | +0.024 |
Neg logit contributions | |
much | -0.065 |
lessly | -0.062 |
wonderful | -0.059 |
br | -0.058 |
In | -0.058 |
Token | eros |
Token ID | 27498 |
Feature activation | -3.42e-02 |
Pos logit contributions | |
Jenna | +0.134 |
Rachel | +0.123 |
lee | +0.119 |
Po | +0.118 |
ishment | +0.117 |
Neg logit contributions | |
much | -0.265 |
br | -0.245 |
hard | -0.244 |
ier | -0.237 |
security | -0.234 |
Token | . |
Token ID | 13 |
Feature activation | -3.59e-02 |
Pos logit contributions | |
ados | +0.073 |
Cr | +0.073 |
Majesty | +0.071 |
ishment | +0.071 |
lee | +0.067 |
Neg logit contributions | |
much | -0.052 |
wonderful | -0.042 |
lessly | -0.042 |
br | -0.040 |
On | -0.040 |
Token | Tim |
Token ID | 5045 |
Feature activation | -3.63e-02 |
Pos logit contributions | |
Majesty | +0.063 |
ados | +0.062 |
ishment | +0.057 |
ishes | +0.054 |
ans | +0.053 |
Neg logit contributions | |
much | -0.075 |
On | -0.067 |
In | -0.064 |
lot | -0.059 |
wonderful | -0.059 |
Token | my |
Token ID | 1820 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
ados | +0.073 |
Cr | +0.072 |
Majesty | +0.070 |
ishment | +0.069 |
lee | +0.064 |
Neg logit contributions | |
much | -0.061 |
wonderful | -0.051 |
On | -0.050 |
lessly | -0.050 |
In | -0.049 |
Token | was |
Token ID | 373 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
ados | +0.061 |
Cr | +0.059 |
Majesty | +0.058 |
ishment | +0.058 |
lee | +0.054 |
Neg logit contributions | |
much | -0.038 |
wonderful | -0.031 |
On | -0.030 |
lessly | -0.029 |
In | -0.029 |
Token | come |
Token ID | 1282 |
Feature activation | -9.57e-02 |
Pos logit contributions | |
much | +0.094 |
lessly | +0.090 |
br | +0.090 |
ching | +0.088 |
In | +0.087 |
Neg logit contributions | |
ados | -0.057 |
Cr | -0.055 |
ishment | -0.052 |
Majesty | -0.051 |
Harry | -0.049 |
Token | in |
Token ID | 287 |
Feature activation | -6.92e-02 |
Pos logit contributions | |
ados | +0.225 |
ishment | +0.210 |
Cr | +0.203 |
Majesty | +0.203 |
roc | +0.194 |
Neg logit contributions | |
much | -0.146 |
wonderful | -0.116 |
on | -0.110 |
On | -0.106 |
hard | -0.105 |
Token | ," |
Token ID | 553 |
Feature activation | -6.51e-02 |
Pos logit contributions | |
ados | +0.143 |
Cr | +0.142 |
Majesty | +0.138 |
ishment | +0.138 |
lee | +0.130 |
Neg logit contributions | |
much | -0.113 |
wonderful | -0.093 |
lessly | -0.090 |
On | -0.089 |
br | -0.086 |
Token | said |
Token ID | 531 |
Feature activation | -3.97e-02 |
Pos logit contributions | |
ados | +0.121 |
Majesty | +0.121 |
ishment | +0.120 |
Cr | +0.111 |
ans | +0.110 |
Neg logit contributions | |
much | -0.125 |
wonderful | -0.102 |
On | -0.099 |
hard | -0.099 |
In | -0.097 |
Token | his |
Token ID | 465 |
Feature activation | -5.49e-02 |
Pos logit contributions | |
ishment | +0.070 |
ados | +0.070 |
Majesty | +0.068 |
Cr | +0.065 |
ans | +0.064 |
Neg logit contributions | |
much | -0.080 |
wonderful | -0.066 |
hard | -0.064 |
lessly | -0.064 |
br | -0.061 |
Token | grand |
Token ID | 4490 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
ados | +0.081 |
Cr | +0.080 |
ishment | +0.077 |
Majesty | +0.076 |
lee | +0.070 |
Neg logit contributions | |
much | -0.121 |
wonderful | -0.106 |
lessly | -0.104 |
On | -0.102 |
br | -0.101 |
Token | pa |
Token ID | 8957 |
Feature activation | -4.18e-02 |
Pos logit contributions | |
Cr | +0.167 |
ados | +0.150 |
Rib | +0.146 |
Fro | +0.142 |
ican | +0.130 |
Neg logit contributions | |
much | -0.283 |
wonderful | -0.261 |
hard | -0.254 |
bad | -0.237 |
lessly | -0.237 |
Token | . |
Token ID | 13 |
Feature activation | -3.79e-02 |
Pos logit contributions | |
ados | +0.085 |
Cr | +0.084 |
Majesty | +0.082 |
ishment | +0.082 |
lee | +0.076 |
Neg logit contributions | |
much | -0.068 |
wonderful | -0.056 |
lessly | -0.056 |
On | -0.054 |
br | -0.054 |
Token | They |
Token ID | 1119 |
Feature activation | -5.92e-03 |
Pos logit contributions | |
ados | +0.088 |
ishment | +0.085 |
Majesty | +0.085 |
Cr | +0.084 |
ishes | +0.080 |
Neg logit contributions | |
much | -0.053 |
On | -0.042 |
lessly | -0.041 |
In | -0.040 |
wonderful | -0.040 |
Token | all |
Token ID | 477 |
Feature activation | +4.89e-03 |
Pos logit contributions | |
Cr | +0.011 |
ados | +0.011 |
ishment | +0.010 |
Majesty | +0.010 |
lee | +0.010 |
Neg logit contributions | |
much | -0.010 |
br | -0.009 |
lessly | -0.009 |
wonderful | -0.009 |
ching | -0.008 |
Token | turned |
Token ID | 2900 |
Feature activation | -6.19e-03 |
Pos logit contributions | |
much | +0.008 |
br | +0.007 |
wonderful | +0.007 |
lessly | +0.007 |
On | +0.007 |
Neg logit contributions | |
Cr | -0.009 |
ados | -0.009 |
lee | -0.009 |
ishment | -0.009 |
Majesty | -0.008 |
Token | Unicorn |
Token ID | 39481 |
Feature activation | -3.25e-02 |
Pos logit contributions | |
ados | +0.045 |
ishment | +0.042 |
Cr | +0.041 |
Majesty | +0.040 |
gy | +0.038 |
Neg logit contributions | |
much | -0.078 |
wonderful | -0.070 |
lessly | -0.066 |
hard | -0.065 |
On | -0.064 |
Token | !" |
Token ID | 2474 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
ishment | +0.070 |
ados | +0.070 |
Cr | +0.070 |
Majesty | +0.066 |
icans | +0.065 |
Neg logit contributions | |
much | -0.051 |
br | -0.041 |
wonderful | -0.041 |
lessly | -0.040 |
hard | -0.038 |
Token | The |
Token ID | 383 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
ados | +0.078 |
Majesty | +0.078 |
ishment | +0.076 |
ieties | +0.072 |
ishes | +0.072 |
Neg logit contributions | |
much | -0.061 |
On | -0.047 |
wonderful | -0.046 |
lessly | -0.045 |
In | -0.044 |
Token | rh |
Token ID | 9529 |
Feature activation | -3.68e-02 |
Pos logit contributions | |
ados | +0.020 |
Cr | +0.019 |
ishment | +0.019 |
Majesty | +0.019 |
lee | +0.017 |
Neg logit contributions | |
much | -0.024 |
wonderful | -0.021 |
lessly | -0.020 |
br | -0.020 |
On | -0.020 |
Token | in |
Token ID | 259 |
Feature activation | -3.17e-02 |
Pos logit contributions | |
ados | +0.077 |
Cr | +0.077 |
Majesty | +0.074 |
ishment | +0.073 |
Rib | +0.070 |
Neg logit contributions | |
much | -0.056 |
lessly | -0.047 |
wonderful | -0.047 |
On | -0.045 |
In | -0.043 |
Token | oc |
Token ID | 420 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
ados | +0.051 |
Cr | +0.049 |
ishment | +0.048 |
Majesty | +0.047 |
lee | +0.044 |
Neg logit contributions | |
much | -0.066 |
wonderful | -0.057 |
lessly | -0.056 |
On | -0.056 |
br | -0.056 |
Token | eros |
Token ID | 27498 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
Jenna | +0.152 |
ishment | +0.141 |
Rachel | +0.140 |
Harry | +0.137 |
lee | +0.131 |
Neg logit contributions | |
much | -0.236 |
ier | -0.231 |
br | -0.228 |
hard | -0.226 |
security | -0.218 |
Token | said |
Token ID | 531 |
Feature activation | -7.70e-03 |
Pos logit contributions | |
ados | +0.058 |
Cr | +0.056 |
ishment | +0.056 |
Majesty | +0.055 |
keeping | +0.052 |
Neg logit contributions | |
much | -0.040 |
wonderful | -0.032 |
lessly | -0.031 |
On | -0.031 |
hard | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | -4.95e-03 |
Pos logit contributions | |
ishment | +0.018 |
ados | +0.018 |
Majesty | +0.018 |
Cr | +0.017 |
icans | +0.017 |
Neg logit contributions | |
much | -0.011 |
lessly | -0.008 |
wonderful | -0.008 |
hard | -0.008 |
br | -0.008 |
Token | feeling |
Token ID | 4203 |
Feature activation | +3.82e-02 |
Pos logit contributions | |
ados | +0.010 |
ishment | +0.010 |
Cr | +0.010 |
Majesty | +0.010 |
ishes | +0.009 |
Neg logit contributions | |
much | -0.008 |
On | -0.007 |
lessly | -0.007 |
hard | -0.006 |
wonderful | -0.006 |
Token | much |
Token ID | 881 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
On | +0.043 |
much | +0.042 |
br | +0.042 |
In | +0.042 |
lessly | +0.041 |
Neg logit contributions | |
Cr | -0.085 |
ados | -0.080 |
Majesty | -0.080 |
Fro | -0.078 |
lee | -0.077 |
Token | there |
Token ID | 612 |
Feature activation | -9.90e-03 |
Pos logit contributions | |
Cr | +0.025 |
ados | +0.024 |
Majesty | +0.023 |
ishment | +0.023 |
Fro | +0.022 |
Neg logit contributions | |
much | -0.024 |
wonderful | -0.021 |
lessly | -0.020 |
br | -0.020 |
In | -0.020 |
Token | was |
Token ID | 373 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
ados | +0.025 |
Cr | +0.024 |
Majesty | +0.024 |
ishment | +0.024 |
lee | +0.023 |
Neg logit contributions | |
much | -0.012 |
wonderful | -0.009 |
lessly | -0.009 |
On | -0.008 |
In | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | +1.44e-03 |
Pos logit contributions | |
Cr | +0.032 |
ados | +0.031 |
Majesty | +0.031 |
ishment | +0.030 |
Fro | +0.029 |
Neg logit contributions | |
much | -0.022 |
lessly | -0.019 |
br | -0.018 |
wonderful | -0.018 |
On | -0.017 |
Token | rh |
Token ID | 9529 |
Feature activation | -4.39e-02 |
Pos logit contributions | |
much | +0.002 |
lessly | +0.002 |
br | +0.002 |
In | +0.002 |
On | +0.002 |
Neg logit contributions | |
Majesty | -0.003 |
Cr | -0.003 |
ados | -0.003 |
ishment | -0.003 |
Fro | -0.003 |
Token | in |
Token ID | 259 |
Feature activation | -3.82e-02 |
Pos logit contributions | |
ados | +0.091 |
Cr | +0.088 |
ishment | +0.084 |
Majesty | +0.083 |
ican | +0.082 |
Neg logit contributions | |
much | -0.068 |
wonderful | -0.059 |
lessly | -0.057 |
On | -0.055 |
In | -0.054 |
Token | oc |
Token ID | 420 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
Cr | +0.048 |
ados | +0.048 |
Majesty | +0.046 |
ishment | +0.045 |
gy | +0.041 |
Neg logit contributions | |
much | -0.090 |
lessly | -0.081 |
wonderful | -0.080 |
On | -0.078 |
br | -0.078 |
Token | eros |
Token ID | 27498 |
Feature activation | -3.66e-02 |
Pos logit contributions | |
ishment | +0.119 |
Jenna | +0.114 |
Cr | +0.111 |
lee | +0.109 |
Po | +0.106 |
Neg logit contributions | |
much | -0.277 |
br | -0.261 |
ier | -0.257 |
hard | -0.252 |
lessly | -0.251 |
Token | who |
Token ID | 508 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
ados | +0.076 |
Cr | +0.075 |
Majesty | +0.073 |
ishment | +0.073 |
lee | +0.069 |
Neg logit contributions | |
much | -0.058 |
wonderful | -0.048 |
lessly | -0.048 |
On | -0.046 |
br | -0.046 |
Token | was |
Token ID | 373 |
Feature activation | +7.55e-03 |
Pos logit contributions | |
Cr | +0.023 |
ados | +0.022 |
gy | +0.022 |
Majesty | +0.022 |
ishment | +0.021 |
Neg logit contributions | |
much | -0.015 |
lessly | -0.013 |
br | -0.012 |
wonderful | -0.012 |
On | -0.012 |
Token | always |
Token ID | 1464 |
Feature activation | -3.35e-03 |
Pos logit contributions | |
much | +0.007 |
lessly | +0.007 |
br | +0.006 |
wonderful | +0.006 |
On | +0.006 |
Neg logit contributions | |
Cr | -0.019 |
ados | -0.019 |
Majesty | -0.019 |
ishment | -0.018 |
gy | -0.018 |
Token | naughty |
Token ID | 45128 |
Feature activation | -4.30e-02 |
Pos logit contributions | |
Cr | +0.006 |
ados | +0.006 |
ishment | +0.006 |
Majesty | +0.006 |
gy | +0.006 |
Neg logit contributions | |
much | -0.005 |
br | -0.005 |
lessly | -0.005 |
In | -0.005 |
On | -0.005 |
Token | going |
Token ID | 1016 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
much | +0.043 |
lessly | +0.038 |
On | +0.037 |
wonderful | +0.036 |
br | +0.036 |
Neg logit contributions | |
Cr | -0.041 |
ados | -0.040 |
Majesty | -0.039 |
ishment | -0.038 |
gy | -0.036 |
Token | on |
Token ID | 319 |
Feature activation | -2.03e-03 |
Pos logit contributions | |
Cr | +0.026 |
ados | +0.026 |
Majesty | +0.025 |
ishment | +0.025 |
gy | +0.023 |
Neg logit contributions | |
much | -0.025 |
lessly | -0.021 |
wonderful | -0.021 |
br | -0.020 |
On | -0.020 |
Token | a |
Token ID | 257 |
Feature activation | +1.67e-03 |
Pos logit contributions | |
ados | +0.005 |
Cr | +0.004 |
Majesty | +0.004 |
ishment | +0.004 |
lee | +0.004 |
Neg logit contributions | |
much | -0.003 |
wonderful | -0.002 |
On | -0.002 |
lessly | -0.002 |
hard | -0.002 |
Token | trip |
Token ID | 5296 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
much | +0.003 |
lessly | +0.002 |
wonderful | +0.002 |
In | +0.002 |
On | +0.002 |
Neg logit contributions | |
Cr | -0.003 |
Majesty | -0.003 |
ados | -0.003 |
ishment | -0.003 |
gy | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
much | +0.026 |
wonderful | +0.024 |
lessly | +0.023 |
br | +0.023 |
In | +0.022 |
Neg logit contributions | |
Cr | -0.027 |
ados | -0.027 |
ishment | -0.026 |
gy | -0.026 |
keeping | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +4.84e-02 |
Pos logit contributions | |
ados | +0.025 |
Majesty | +0.024 |
Cr | +0.024 |
ishment | +0.024 |
lee | +0.023 |
Neg logit contributions | |
much | -0.013 |
On | -0.010 |
wonderful | -0.010 |
In | -0.010 |
lessly | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +4.23e-02 |
Pos logit contributions | |
lessly | +0.084 |
much | +0.082 |
wonderful | +0.079 |
br | +0.078 |
terrible | +0.076 |
Neg logit contributions | |
Cr | -0.073 |
Fro | -0.072 |
Rib | -0.070 |
Av | -0.065 |
Harry | -0.064 |
Token | " |
Token ID | 1 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
much | +0.048 |
lessly | +0.045 |
wonderful | +0.042 |
terrible | +0.039 |
br | +0.038 |
Neg logit contributions | |
Cr | -0.090 |
ishment | -0.087 |
Rib | -0.086 |
ados | -0.086 |
Majesty | -0.084 |
Token | Let |
Token ID | 5756 |
Feature activation | -4.39e-03 |
Pos logit contributions | |
much | +0.027 |
lessly | +0.024 |
wonderful | +0.024 |
hard | +0.023 |
bad | +0.023 |
Neg logit contributions | |
Cr | -0.023 |
Rib | -0.021 |
ASH | -0.021 |
ados | -0.021 |
ishment | -0.020 |
Token | 's |
Token ID | 338 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
ados | +0.010 |
ishment | +0.010 |
Cr | +0.010 |
Majesty | +0.010 |
keeping | +0.009 |
Neg logit contributions | |
much | -0.007 |
wonderful | -0.006 |
bad | -0.006 |
hard | -0.005 |
big | -0.005 |
Token | load |
Token ID | 3440 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
ados | +0.085 |
Cr | +0.082 |
ishment | +0.080 |
Majesty | +0.079 |
lee | +0.076 |
Neg logit contributions | |
much | -0.076 |
wonderful | -0.064 |
On | -0.062 |
In | -0.060 |
br | -0.060 |
Token | needed |
Token ID | 2622 |
Feature activation | +3.59e-02 |
Pos logit contributions | |
ados | +0.015 |
Cr | +0.015 |
ishment | +0.015 |
Majesty | +0.015 |
icans | +0.014 |
Neg logit contributions | |
much | -0.011 |
wonderful | -0.009 |
lessly | -0.009 |
On | -0.008 |
hard | -0.008 |
Token | it |
Token ID | 340 |
Feature activation | +3.28e-02 |
Pos logit contributions | |
much | +0.064 |
lessly | +0.058 |
wonderful | +0.056 |
On | +0.055 |
br | +0.055 |
Neg logit contributions | |
ados | -0.064 |
Cr | -0.064 |
Majesty | -0.060 |
ishment | -0.059 |
Fro | -0.057 |
Token | very |
Token ID | 845 |
Feature activation | +2.48e-02 |
Pos logit contributions | |
much | +0.056 |
br | +0.052 |
lessly | +0.051 |
On | +0.050 |
wonderful | +0.049 |
Neg logit contributions | |
ados | -0.056 |
Cr | -0.055 |
Majesty | -0.053 |
ishment | -0.052 |
gy | -0.052 |
Token | much |
Token ID | 881 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
much | +0.019 |
br | +0.016 |
lessly | +0.016 |
On | +0.015 |
In | +0.014 |
Neg logit contributions | |
Cr | -0.066 |
Majesty | -0.065 |
ados | -0.065 |
ishment | -0.064 |
lee | -0.062 |
Token | . |
Token ID | 13 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
In | +0.048 |
much | +0.047 |
lessly | +0.046 |
On | +0.046 |
Over | +0.045 |
Neg logit contributions | |
ados | -0.051 |
ishment | -0.050 |
Cr | -0.049 |
Majesty | -0.047 |
keeping | -0.044 |
Token | After |
Token ID | 2293 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
Majesty | +0.024 |
ishment | +0.023 |
ans | +0.022 |
icans | +0.022 |
ican | +0.022 |
Neg logit contributions | |
much | -0.023 |
In | -0.019 |
On | -0.018 |
It | -0.018 |
lessly | -0.018 |
Token | that |
Token ID | 326 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
much | +0.060 |
lessly | +0.052 |
wonderful | +0.051 |
br | +0.050 |
On | +0.049 |
Neg logit contributions | |
Cr | -0.080 |
ados | -0.080 |
Majesty | -0.075 |
ishment | -0.075 |
lee | -0.072 |
Token | , |
Token ID | 11 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
much | +0.045 |
lessly | +0.039 |
wonderful | +0.039 |
On | +0.038 |
In | +0.038 |
Neg logit contributions | |
Cr | -0.042 |
ados | -0.042 |
ishment | -0.040 |
Majesty | -0.039 |
Fro | -0.037 |
Token | the |
Token ID | 262 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
much | +0.052 |
wonderful | +0.042 |
hard | +0.041 |
On | +0.039 |
lot | +0.039 |
Neg logit contributions | |
Majesty | -0.061 |
ishment | -0.060 |
ados | -0.060 |
Cr | -0.058 |
ans | -0.058 |
Token | bunny |
Token ID | 44915 |
Feature activation | +4.53e-03 |
Pos logit contributions | |
much | +0.070 |
lessly | +0.062 |
br | +0.058 |
On | +0.058 |
In | +0.057 |
Neg logit contributions | |
Cr | -0.062 |
Majesty | -0.062 |
ados | -0.062 |
ishment | -0.061 |
Fro | -0.057 |
Token | sat |
Token ID | 3332 |
Feature activation | -1.24e-03 |
Pos logit contributions | |
much | +0.006 |
wonderful | +0.005 |
On | +0.004 |
In | +0.004 |
hard | +0.004 |
Neg logit contributions | |
ados | -0.011 |
Cr | -0.011 |
ishment | -0.011 |
Majesty | -0.010 |
icans | -0.010 |
Token | of |
Token ID | 286 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
much | +0.023 |
wonderful | +0.020 |
On | +0.020 |
lessly | +0.020 |
In | +0.019 |
Neg logit contributions | |
ados | -0.037 |
Cr | -0.036 |
ishment | -0.035 |
Majesty | -0.034 |
lee | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | +3.58e-03 |
Pos logit contributions | |
Cr | +0.049 |
ados | +0.049 |
Majesty | +0.049 |
ishment | +0.048 |
lee | +0.045 |
Neg logit contributions | |
much | -0.057 |
wonderful | -0.049 |
lessly | -0.046 |
hard | -0.046 |
On | -0.046 |
Token | window |
Token ID | 4324 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
much | +0.007 |
lessly | +0.006 |
wonderful | +0.006 |
On | +0.006 |
br | +0.006 |
Neg logit contributions | |
ados | -0.006 |
Cr | -0.006 |
Majesty | -0.006 |
ishment | -0.006 |
lee | -0.005 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
much | +0.029 |
wonderful | +0.024 |
lessly | +0.024 |
br | +0.023 |
On | +0.023 |
Neg logit contributions | |
ados | -0.031 |
Cr | -0.031 |
ishment | -0.030 |
Majesty | -0.030 |
lee | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
ados | +0.043 |
Cr | +0.043 |
ishment | +0.042 |
Majesty | +0.042 |
gy | +0.040 |
Neg logit contributions | |
much | -0.025 |
wonderful | -0.020 |
lessly | -0.019 |
On | -0.019 |
In | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +3.63e-02 |
Pos logit contributions | |
much | +0.053 |
wonderful | +0.043 |
lessly | +0.043 |
br | +0.041 |
On | +0.041 |
Neg logit contributions | |
ados | -0.088 |
Cr | -0.087 |
Majesty | -0.085 |
ishment | -0.085 |
lee | -0.081 |
Token | " |
Token ID | 1 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
much | +0.047 |
wonderful | +0.036 |
On | +0.036 |
lessly | +0.035 |
In | +0.035 |
Neg logit contributions | |
ados | -0.088 |
Cr | -0.086 |
ishment | -0.086 |
Majesty | -0.085 |
lee | -0.080 |
Token | We |
Token ID | 1135 |
Feature activation | +1.00e-02 |
Pos logit contributions | |
much | +0.048 |
wonderful | +0.038 |
lessly | +0.038 |
hard | +0.035 |
br | +0.034 |
Neg logit contributions | |
Cr | -0.078 |
ados | -0.073 |
Rib | -0.071 |
ican | -0.071 |
icans | -0.070 |
Token | had |
Token ID | 550 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
much | +0.016 |
lessly | +0.013 |
br | +0.013 |
wonderful | +0.013 |
On | +0.013 |
Neg logit contributions | |
Cr | -0.020 |
ados | -0.020 |
Majesty | -0.020 |
ishment | -0.020 |
lee | -0.019 |
Token | a |
Token ID | 257 |
Feature activation | +3.67e-02 |
Pos logit contributions | |
much | +0.023 |
wonderful | +0.019 |
hard | +0.017 |
br | +0.017 |
lessly | +0.017 |
Neg logit contributions | |
ados | -0.035 |
Cr | -0.034 |
ishment | -0.034 |
Majesty | -0.034 |
gy | -0.032 |
Token | plan |
Token ID | 1410 |
Feature activation | +3.02e-02 |
Pos logit contributions | |
lessly | +0.052 |
much | +0.051 |
In | +0.044 |
On | +0.044 |
br | +0.042 |
Neg logit contributions | |
Majesty | -0.077 |
Cr | -0.076 |
ados | -0.072 |
Fro | -0.071 |
gy | -0.071 |
Token | why |
Token ID | 1521 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
much | +0.015 |
br | +0.013 |
lessly | +0.013 |
wonderful | +0.013 |
On | +0.013 |
Neg logit contributions | |
ados | -0.016 |
Cr | -0.016 |
Majesty | -0.015 |
ishment | -0.015 |
lee | -0.014 |
Token | his |
Token ID | 465 |
Feature activation | +2.53e-03 |
Pos logit contributions | |
much | +0.075 |
wonderful | +0.063 |
lessly | +0.063 |
On | +0.061 |
br | +0.060 |
Neg logit contributions | |
ados | -0.081 |
Cr | -0.080 |
Majesty | -0.078 |
ishment | -0.078 |
lee | -0.073 |
Token | friend |
Token ID | 1545 |
Feature activation | +7.08e-03 |
Pos logit contributions | |
much | +0.005 |
wonderful | +0.004 |
hard | +0.004 |
lot | +0.004 |
bad | +0.004 |
Neg logit contributions | |
ados | -0.005 |
Cr | -0.005 |
ishment | -0.004 |
Majesty | -0.004 |
icans | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +3.14e-02 |
Pos logit contributions | |
much | +0.009 |
lessly | +0.008 |
wonderful | +0.007 |
On | +0.007 |
br | +0.007 |
Neg logit contributions | |
ados | -0.016 |
Cr | -0.016 |
Majesty | -0.016 |
ishment | -0.016 |
lee | -0.015 |
Token | frog |
Token ID | 21264 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
much | +0.066 |
wonderful | +0.058 |
lessly | +0.056 |
br | +0.055 |
On | +0.055 |
Neg logit contributions | |
ados | -0.050 |
Cr | -0.049 |
ishment | -0.048 |
Majesty | -0.047 |
lee | -0.044 |
Token | was |
Token ID | 373 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
much | +0.020 |
lessly | +0.015 |
wonderful | +0.015 |
On | +0.015 |
br | +0.015 |
Neg logit contributions | |
ados | -0.037 |
Cr | -0.037 |
Majesty | -0.036 |
ishment | -0.036 |
lee | -0.034 |
Token | so |
Token ID | 523 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
much | +0.055 |
lessly | +0.048 |
br | +0.047 |
On | +0.047 |
In | +0.046 |
Neg logit contributions | |
Cr | -0.068 |
ados | -0.067 |
Majesty | -0.066 |
ishment | -0.065 |
lee | -0.063 |
Token | encouraging |
Token ID | 12577 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
much | +0.043 |
br | +0.040 |
On | +0.038 |
lessly | +0.038 |
In | +0.037 |
Neg logit contributions | |
Cr | -0.079 |
Majesty | -0.074 |
ados | -0.073 |
lee | -0.073 |
ishment | -0.072 |
Token | . |
Token ID | 13 |
Feature activation | -3.05e-03 |
Pos logit contributions | |
Majesty | +0.013 |
ados | +0.013 |
Cr | +0.013 |
ishment | +0.013 |
gy | +0.011 |
Neg logit contributions | |
much | -0.024 |
wonderful | -0.021 |
lessly | -0.021 |
hard | -0.021 |
On | -0.020 |
Token | Joe |
Token ID | 5689 |
Feature activation | +8.12e-03 |
Pos logit contributions | |
ados | +0.006 |
ishment | +0.006 |
Majesty | +0.006 |
Cr | +0.006 |
icans | +0.005 |
Neg logit contributions | |
much | -0.005 |
lessly | -0.004 |
On | -0.004 |
wonderful | -0.004 |
In | -0.004 |
Token | had |
Token ID | 550 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
much | +0.011 |
wonderful | +0.009 |
lessly | +0.009 |
On | +0.008 |
In | +0.008 |
Neg logit contributions | |
ados | -0.019 |
Cr | -0.018 |
ishment | -0.018 |
Majesty | -0.018 |
icans | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
br | +0.019 |
much | +0.018 |
In | +0.018 |
wonderful | +0.018 |
arest | +0.018 |
Neg logit contributions | |
Cr | -0.017 |
ishment | -0.017 |
ados | -0.017 |
lee | -0.016 |
gy | -0.016 |
Token | They |
Token ID | 1119 |
Feature activation | +6.31e-03 |
Pos logit contributions | |
ados | +0.032 |
Cr | +0.031 |
Majesty | +0.031 |
ishment | +0.030 |
lee | +0.028 |
Neg logit contributions | |
much | -0.050 |
wonderful | -0.043 |
lessly | -0.043 |
On | -0.042 |
In | -0.042 |
Token | knew |
Token ID | 2993 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
br | +0.012 |
much | +0.011 |
lessly | +0.010 |
arest | +0.010 |
ching | +0.010 |
Neg logit contributions | |
Cr | -0.010 |
lee | -0.010 |
ishment | -0.010 |
gy | -0.009 |
Fro | -0.009 |
Token | that |
Token ID | 326 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
much | +0.055 |
br | +0.052 |
lessly | +0.051 |
On | +0.050 |
In | +0.049 |
Neg logit contributions | |
Cr | -0.048 |
ados | -0.045 |
Fro | -0.042 |
ishment | -0.041 |
Rib | -0.041 |
Token | Max |
Token ID | 5436 |
Feature activation | +8.12e-03 |
Pos logit contributions | |
much | +0.046 |
wonderful | +0.039 |
lessly | +0.039 |
On | +0.038 |
br | +0.038 |
Neg logit contributions | |
ados | -0.040 |
Cr | -0.039 |
Majesty | -0.038 |
ishment | -0.038 |
lee | -0.035 |
Token | was |
Token ID | 373 |
Feature activation | +3.69e-02 |
Pos logit contributions | |
much | +0.010 |
lessly | +0.009 |
br | +0.008 |
On | +0.008 |
wonderful | +0.008 |
Neg logit contributions | |
Cr | -0.019 |
ados | -0.019 |
ishment | -0.018 |
gy | -0.018 |
Majesty | -0.018 |
Token | still |
Token ID | 991 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
much | +0.045 |
lessly | +0.040 |
br | +0.039 |
On | +0.039 |
In | +0.036 |
Neg logit contributions | |
Cr | -0.088 |
ados | -0.085 |
ishment | -0.081 |
Majesty | -0.080 |
Fro | -0.080 |
Token | very |
Token ID | 845 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
much | +0.025 |
wonderful | +0.020 |
lessly | +0.020 |
On | +0.019 |
br | +0.019 |
Neg logit contributions | |
ados | -0.034 |
Cr | -0.033 |
Majesty | -0.033 |
ishment | -0.033 |
lee | -0.031 |
Token | little |
Token ID | 1310 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
much | +0.021 |
br | +0.020 |
On | +0.020 |
In | +0.019 |
lessly | +0.019 |
Neg logit contributions | |
Cr | -0.054 |
ados | -0.052 |
Majesty | -0.051 |
ishment | -0.051 |
lee | -0.050 |
Token | , |
Token ID | 11 |
Feature activation | -2.97e-03 |
Pos logit contributions | |
ados | +0.026 |
Cr | +0.026 |
Majesty | +0.025 |
ishment | +0.025 |
lee | +0.023 |
Neg logit contributions | |
much | -0.024 |
wonderful | -0.020 |
lessly | -0.020 |
On | -0.019 |
br | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | -6.12e-03 |
Pos logit contributions | |
ados | +0.006 |
Cr | +0.006 |
Majesty | +0.006 |
ishment | +0.006 |
lee | +0.006 |
Neg logit contributions | |
much | -0.005 |
wonderful | -0.004 |
lessly | -0.004 |
On | -0.004 |
br | -0.003 |
Token | very |
Token ID | 845 |
Feature activation | +4.76e-02 |
Pos logit contributions | |
much | +0.048 |
br | +0.044 |
lessly | +0.043 |
On | +0.041 |
In | +0.040 |
Neg logit contributions | |
Cr | -0.109 |
ados | -0.108 |
Majesty | -0.106 |
ishment | -0.103 |
lee | -0.103 |
Token | happy |
Token ID | 3772 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
much | +0.059 |
br | +0.055 |
lessly | +0.054 |
On | +0.053 |
ching | +0.052 |
Neg logit contributions | |
Cr | -0.103 |
Majesty | -0.101 |
ados | -0.101 |
ishment | -0.097 |
lee | -0.097 |
Token | . |
Token ID | 13 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
Cr | +0.047 |
ados | +0.046 |
Majesty | +0.046 |
ishment | +0.046 |
lee | +0.043 |
Neg logit contributions | |
much | -0.028 |
wonderful | -0.022 |
lessly | -0.021 |
On | -0.021 |
hard | -0.021 |
Token | She |
Token ID | 1375 |
Feature activation | +8.65e-03 |
Pos logit contributions | |
ados | +0.028 |
Cr | +0.028 |
ishment | +0.027 |
Majesty | +0.027 |
lee | +0.025 |
Neg logit contributions | |
much | -0.025 |
wonderful | -0.020 |
lessly | -0.020 |
On | -0.020 |
br | -0.020 |
Token | said |
Token ID | 531 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
much | +0.014 |
wonderful | +0.012 |
lessly | +0.012 |
On | +0.011 |
br | +0.011 |
Neg logit contributions | |
ados | -0.018 |
Cr | -0.017 |
ishment | -0.017 |
Majesty | -0.017 |
Fro | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | +6.86e-03 |
Pos logit contributions | |
much | +0.036 |
hard | +0.028 |
br | +0.027 |
wonderful | +0.027 |
bad | +0.025 |
Neg logit contributions | |
ishment | -0.054 |
gy | -0.050 |
keeping | -0.049 |
ans | -0.049 |
ci | -0.048 |
Token | " |
Token ID | 366 |
Feature activation | -5.76e-03 |
Pos logit contributions | |
much | +0.011 |
lessly | +0.009 |
wonderful | +0.009 |
On | +0.009 |
br | +0.009 |
Neg logit contributions | |
ados | -0.014 |
ishment | -0.014 |
Majesty | -0.014 |
Cr | -0.014 |
gy | -0.013 |
Token | Look |
Token ID | 8567 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
ados | +0.012 |
Cr | +0.012 |
Majesty | +0.012 |
ishment | +0.012 |
lee | +0.011 |
Neg logit contributions | |
much | -0.009 |
lessly | -0.007 |
wonderful | -0.007 |
br | -0.007 |
On | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | -2.14e-03 |
Pos logit contributions | |
keeping | +0.041 |
Cr | +0.041 |
ishment | +0.040 |
ados | +0.039 |
Majesty | +0.039 |
Neg logit contributions | |
much | -0.028 |
wonderful | -0.024 |
hard | -0.021 |
else | -0.020 |
br | -0.020 |
Token | Spot |
Token ID | 15899 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
ados | +0.003 |
ishment | +0.003 |
Cr | +0.003 |
Majesty | +0.003 |
lee | +0.003 |
Neg logit contributions | |
much | -0.005 |
wonderful | -0.004 |
br | -0.004 |
lessly | -0.004 |
On | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
Cr | +0.043 |
ados | +0.043 |
ishment | +0.043 |
Majesty | +0.042 |
lee | +0.041 |
Neg logit contributions | |
much | -0.026 |
wonderful | -0.021 |
hard | -0.020 |
br | -0.020 |
ching | -0.018 |
Token | cloud |
Token ID | 6279 |
Feature activation | +3.79e-02 |
Pos logit contributions | |
lessly | +0.057 |
much | +0.056 |
In | +0.052 |
On | +0.051 |
br | +0.047 |
Neg logit contributions | |
Majesty | -0.075 |
Cr | -0.071 |
gy | -0.069 |
ados | -0.069 |
Fro | -0.068 |
Token | , |
Token ID | 11 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
much | +0.067 |
wonderful | +0.056 |
lessly | +0.056 |
On | +0.055 |
br | +0.054 |
Neg logit contributions | |
ados | -0.071 |
Cr | -0.071 |
Majesty | -0.068 |
ishment | -0.068 |
lee | -0.064 |
Token | Lily |
Token ID | 20037 |
Feature activation | -9.59e-04 |
Pos logit contributions | |
much | +0.024 |
wonderful | +0.021 |
lessly | +0.021 |
On | +0.021 |
br | +0.021 |
Neg logit contributions | |
ados | -0.015 |
Cr | -0.014 |
Majesty | -0.014 |
ishment | -0.014 |
lee | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -9.45e-03 |
Pos logit contributions | |
Cr | +0.002 |
ados | +0.002 |
ishment | +0.002 |
Majesty | +0.002 |
icans | +0.002 |
Neg logit contributions | |
much | -0.001 |
wonderful | -0.001 |
lessly | -0.001 |
hard | -0.001 |
br | -0.001 |
Token | It |
Token ID | 632 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
ados | +0.018 |
keeping | +0.017 |
ans | +0.016 |
ishment | +0.016 |
Majesty | +0.016 |
Neg logit contributions | |
much | -0.017 |
In | -0.014 |
lessly | -0.014 |
It | -0.014 |
wonderful | -0.013 |
Token | 's |
Token ID | 338 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
much | +0.073 |
wonderful | +0.063 |
lessly | +0.060 |
br | +0.060 |
hard | +0.059 |
Neg logit contributions | |
ados | -0.069 |
Cr | -0.066 |
Majesty | -0.065 |
ishment | -0.065 |
lee | -0.060 |
Token | fog |
Token ID | 19558 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
much | +0.033 |
lessly | +0.030 |
br | +0.029 |
On | +0.028 |
In | +0.027 |
Neg logit contributions | |
Cr | -0.053 |
ishment | -0.051 |
ados | -0.050 |
Fro | -0.050 |
Majesty | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
much | +0.041 |
wonderful | +0.037 |
lessly | +0.037 |
br | +0.036 |
On | +0.036 |
Neg logit contributions | |
ados | -0.013 |
Cr | -0.013 |
ishment | -0.012 |
Majesty | -0.012 |
lee | -0.010 |
Token | It |
Token ID | 632 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
ados | +0.025 |
Majesty | +0.023 |
ishment | +0.023 |
keeping | +0.022 |
Cr | +0.022 |
Neg logit contributions | |
much | -0.029 |
lessly | -0.024 |
In | -0.024 |
wonderful | -0.023 |
On | -0.023 |
Token | makes |
Token ID | 1838 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
much | +0.064 |
wonderful | +0.055 |
lessly | +0.054 |
br | +0.053 |
On | +0.052 |
Neg logit contributions | |
ados | -0.068 |
Cr | -0.066 |
Majesty | -0.065 |
ishment | -0.065 |
lee | -0.060 |
Token | it |
Token ID | 340 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
much | +0.023 |
wonderful | +0.019 |
lessly | +0.018 |
hard | +0.018 |
br | +0.018 |
Neg logit contributions | |
ados | -0.023 |
Cr | -0.023 |
ishment | -0.022 |
Majesty | -0.022 |
gy | -0.020 |
Token | Finally |
Token ID | 9461 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
ados | +0.048 |
Majesty | +0.046 |
ishment | +0.045 |
Cr | +0.045 |
ishes | +0.044 |
Neg logit contributions | |
much | -0.046 |
lessly | -0.038 |
On | -0.037 |
In | -0.036 |
wonderful | -0.035 |
Token | , |
Token ID | 11 |
Feature activation | +1.45e-02 |
Pos logit contributions | |
much | +0.062 |
lessly | +0.056 |
wonderful | +0.055 |
On | +0.054 |
In | +0.053 |
Neg logit contributions | |
ados | -0.055 |
Cr | -0.054 |
Majesty | -0.051 |
ishment | -0.050 |
Fro | -0.048 |
Token | he |
Token ID | 339 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
much | +0.024 |
wonderful | +0.020 |
On | +0.019 |
lessly | +0.019 |
hard | +0.019 |
Neg logit contributions | |
ados | -0.030 |
Majesty | -0.029 |
ishment | -0.029 |
Cr | -0.029 |
lee | -0.028 |
Token | went |
Token ID | 1816 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
much | +0.020 |
lessly | +0.018 |
wonderful | +0.017 |
br | +0.016 |
On | +0.016 |
Neg logit contributions | |
Cr | -0.024 |
Fro | -0.023 |
lee | -0.022 |
gy | -0.022 |
Majesty | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | -4.45e-02 |
Pos logit contributions | |
ados | +0.021 |
Cr | +0.021 |
ishment | +0.020 |
Majesty | +0.020 |
Fro | +0.019 |
Neg logit contributions | |
much | -0.029 |
wonderful | -0.026 |
lessly | -0.026 |
br | -0.026 |
In | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
ados | +0.094 |
ishment | +0.088 |
Majesty | +0.088 |
lee | +0.085 |
ishes | +0.085 |
Neg logit contributions | |
much | -0.078 |
wonderful | -0.065 |
hard | -0.062 |
bad | -0.062 |
On | -0.061 |
Token | doctor |
Token ID | 6253 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
Majesty | +0.041 |
ados | +0.041 |
Cr | +0.040 |
ishment | +0.038 |
Fro | +0.036 |
Neg logit contributions | |
much | -0.056 |
lessly | -0.051 |
br | -0.048 |
wonderful | -0.048 |
On | -0.047 |
Token | even |
Token ID | 772 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
Cr | +0.096 |
ados | +0.096 |
Majesty | +0.092 |
ishment | +0.091 |
lee | +0.086 |
Neg logit contributions | |
much | -0.056 |
On | -0.042 |
wonderful | -0.041 |
lessly | -0.041 |
hard | -0.040 |
Token | though |
Token ID | 996 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
Cr | +0.036 |
ados | +0.036 |
ishment | +0.036 |
Majesty | +0.034 |
Fro | +0.033 |
Neg logit contributions | |
much | -0.030 |
lessly | -0.026 |
br | -0.025 |
wonderful | -0.025 |
In | -0.024 |
Token | his |
Token ID | 465 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
much | +0.025 |
lessly | +0.022 |
wonderful | +0.021 |
br | +0.021 |
On | +0.021 |
Neg logit contributions | |
ados | -0.029 |
Cr | -0.029 |
Majesty | -0.028 |
ishment | -0.028 |
Fro | -0.026 |
Token | wife |
Token ID | 3656 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
Cr | +0.021 |
ados | +0.021 |
ishment | +0.019 |
Majesty | +0.019 |
lee | +0.017 |
Neg logit contributions | |
much | -0.046 |
wonderful | -0.041 |
lessly | -0.040 |
hard | -0.039 |
On | -0.039 |
Token | sun |
Token ID | 4252 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
ados | +0.012 |
Cr | +0.012 |
ishment | +0.011 |
Majesty | +0.011 |
lee | +0.011 |
Neg logit contributions | |
much | -0.013 |
lessly | -0.012 |
wonderful | -0.012 |
On | -0.011 |
br | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -1.89e-03 |
Pos logit contributions | |
much | +0.017 |
lessly | +0.013 |
wonderful | +0.013 |
br | +0.013 |
On | +0.013 |
Neg logit contributions | |
Cr | -0.031 |
ados | -0.030 |
Majesty | -0.030 |
ishment | -0.030 |
lee | -0.028 |
Token | When |
Token ID | 1649 |
Feature activation | +5.31e-02 |
Pos logit contributions | |
ishment | +0.003 |
Majesty | +0.003 |
ados | +0.003 |
Cr | +0.003 |
ans | +0.003 |
Neg logit contributions | |
much | -0.004 |
lessly | -0.003 |
On | -0.003 |
In | -0.003 |
wonderful | -0.003 |
Token | he |
Token ID | 339 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
much | +0.088 |
lessly | +0.080 |
wonderful | +0.077 |
br | +0.076 |
On | +0.075 |
Neg logit contributions | |
Cr | -0.102 |
ados | -0.101 |
Majesty | -0.094 |
ishment | -0.094 |
Fro | -0.090 |
Token | woke |
Token ID | 19092 |
Feature activation | +2.72e-03 |
Pos logit contributions | |
much | +0.045 |
lessly | +0.042 |
wonderful | +0.039 |
br | +0.038 |
On | +0.037 |
Neg logit contributions | |
Cr | -0.036 |
ados | -0.034 |
Fro | -0.034 |
ishment | -0.033 |
Majesty | -0.033 |
Token | up |
Token ID | 510 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
much | +0.005 |
wonderful | +0.004 |
hard | +0.004 |
lessly | +0.004 |
bad | +0.004 |
Neg logit contributions | |
Cr | -0.005 |
ados | -0.005 |
Majesty | -0.005 |
ishment | -0.005 |
keeping | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
much | +0.046 |
wonderful | +0.039 |
lessly | +0.039 |
On | +0.038 |
br | +0.038 |
Neg logit contributions | |
ados | -0.047 |
Cr | -0.046 |
Majesty | -0.045 |
ishment | -0.045 |
lee | -0.042 |
Token | all |
Token ID | 477 |
Feature activation | +4.30e-02 |
Pos logit contributions | |
much | +0.061 |
lessly | +0.055 |
br | +0.054 |
wonderful | +0.053 |
On | +0.051 |
Neg logit contributions | |
Cr | -0.063 |
ados | -0.062 |
ishment | -0.057 |
Majesty | -0.057 |
Fro | -0.055 |
Token | his |
Token ID | 465 |
Feature activation | +9.93e-03 |
Pos logit contributions | |
much | +0.076 |
wonderful | +0.063 |
lessly | +0.063 |
On | +0.060 |
br | +0.059 |
Neg logit contributions | |
ados | -0.084 |
Cr | -0.083 |
Majesty | -0.082 |
ishment | -0.081 |
icans | -0.075 |
Token | candy |
Token ID | 18550 |
Feature activation | +5.83e-02 |
Pos logit contributions | |
much | +0.021 |
wonderful | +0.019 |
lessly | +0.018 |
On | +0.018 |
br | +0.018 |
Neg logit contributions | |
ados | -0.015 |
Cr | -0.015 |
ishment | -0.014 |
Majesty | -0.014 |
lee | -0.013 |
Token | was |
Token ID | 373 |
Feature activation | +5.73e-02 |
Pos logit contributions | |
much | +0.096 |
lessly | +0.087 |
wonderful | +0.083 |
On | +0.081 |
In | +0.079 |
Neg logit contributions | |
ados | -0.111 |
Cr | -0.099 |
Majesty | -0.098 |
knees | -0.098 |
ishment | -0.098 |
Token | them |
Token ID | 606 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
ados | +0.023 |
Cr | +0.023 |
ishment | +0.022 |
Majesty | +0.022 |
lee | +0.020 |
Neg logit contributions | |
much | -0.021 |
lessly | -0.018 |
wonderful | -0.018 |
On | -0.018 |
br | -0.018 |
Token | when |
Token ID | 618 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
much | +0.018 |
br | +0.016 |
lessly | +0.016 |
In | +0.015 |
wonderful | +0.015 |
Neg logit contributions | |
ados | -0.023 |
Cr | -0.022 |
ishment | -0.022 |
Majesty | -0.022 |
lee | -0.020 |
Token | they |
Token ID | 484 |
Feature activation | +1.23e-03 |
Pos logit contributions | |
much | +0.020 |
wonderful | +0.017 |
lessly | +0.016 |
On | +0.016 |
br | +0.016 |
Neg logit contributions | |
ados | -0.021 |
Cr | -0.021 |
ishment | -0.020 |
Majesty | -0.020 |
lee | -0.019 |
Token | were |
Token ID | 547 |
Feature activation | +2.15e-02 |
Pos logit contributions | |
much | +0.002 |
lessly | +0.002 |
br | +0.002 |
wonderful | +0.001 |
On | +0.001 |
Neg logit contributions | |
Cr | -0.003 |
ados | -0.003 |
Majesty | -0.002 |
ishment | -0.002 |
lee | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
much | +0.028 |
br | +0.026 |
On | +0.025 |
lessly | +0.025 |
In | +0.025 |
Neg logit contributions | |
ados | -0.045 |
Majesty | -0.044 |
Cr | -0.044 |
ishment | -0.044 |
lee | -0.043 |
Token | right |
Token ID | 826 |
Feature activation | -3.69e-03 |
Pos logit contributions | |
much | +0.029 |
lessly | +0.026 |
br | +0.025 |
On | +0.025 |
In | +0.023 |
Neg logit contributions | |
Majesty | -0.054 |
ados | -0.053 |
ishment | -0.052 |
Cr | -0.051 |
Fro | -0.048 |
Token | temperature |
Token ID | 5951 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
ados | +0.007 |
ishment | +0.007 |
Majesty | +0.007 |
Cr | +0.006 |
gy | +0.006 |
Neg logit contributions | |
much | -0.007 |
br | -0.006 |
lessly | -0.006 |
On | -0.006 |
wonderful | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
much | +0.031 |
wonderful | +0.027 |
lessly | +0.027 |
On | +0.026 |
br | +0.026 |
Neg logit contributions | |
ados | -0.042 |
Cr | -0.041 |
ishment | -0.040 |
Majesty | -0.040 |
lee | -0.038 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +9.61e-03 |
Pos logit contributions | |
ados | +0.037 |
ishment | +0.037 |
Majesty | +0.037 |
Cr | +0.036 |
ans | +0.035 |
Neg logit contributions | |
much | -0.020 |
In | -0.014 |
On | -0.014 |
lessly | -0.014 |
wonderful | -0.014 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
much | +0.016 |
lessly | +0.014 |
wonderful | +0.014 |
br | +0.013 |
On | +0.013 |
Neg logit contributions | |
Cr | -0.018 |
ados | -0.018 |
Fro | -0.017 |
Rib | -0.017 |
ishment | -0.017 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.62e-03 |
Pos logit contributions | |
much | +0.017 |
lessly | +0.014 |
wonderful | +0.014 |
br | +0.014 |
On | +0.013 |
Neg logit contributions | |
Cr | -0.022 |
ados | -0.022 |
Majesty | -0.022 |
ishment | -0.022 |
lee | -0.021 |
Token | away |
Token ID | 1497 |
Feature activation | +5.78e-03 |
Pos logit contributions | |
ados | +0.057 |
Cr | +0.057 |
Majesty | +0.056 |
ishment | +0.056 |
lee | +0.052 |
Neg logit contributions | |
much | -0.051 |
lessly | -0.043 |
wonderful | -0.043 |
On | -0.041 |
br | -0.041 |
Token | and |
Token ID | 290 |
Feature activation | -3.30e-02 |
Pos logit contributions | |
much | +0.011 |
wonderful | +0.010 |
In | +0.010 |
br | +0.010 |
arest | +0.009 |
Neg logit contributions | |
ados | -0.009 |
Cr | -0.009 |
ishment | -0.009 |
gy | -0.008 |
Fro | -0.008 |
Token | ran |
Token ID | 4966 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
ados | +0.065 |
Cr | +0.065 |
Majesty | +0.063 |
ishment | +0.063 |
lee | +0.059 |
Neg logit contributions | |
much | -0.055 |
wonderful | -0.046 |
lessly | -0.046 |
On | -0.044 |
br | -0.044 |
Token | back |
Token ID | 736 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
ados | +0.047 |
Cr | +0.046 |
Majesty | +0.044 |
ishment | +0.044 |
gy | +0.043 |
Neg logit contributions | |
much | -0.050 |
wonderful | -0.045 |
In | -0.043 |
br | -0.043 |
On | -0.043 |
Token | to |
Token ID | 284 |
Feature activation | -4.38e-02 |
Pos logit contributions | |
ados | +0.066 |
Cr | +0.065 |
ishment | +0.064 |
Majesty | +0.064 |
lee | +0.059 |
Neg logit contributions | |
much | -0.070 |
wonderful | -0.059 |
lessly | -0.059 |
On | -0.058 |
hard | -0.057 |
Token | his |
Token ID | 465 |
Feature activation | -4.38e-02 |
Pos logit contributions | |
ados | +0.091 |
ishment | +0.086 |
ishes | +0.083 |
amb | +0.081 |
Majesty | +0.080 |
Neg logit contributions | |
much | -0.080 |
wonderful | -0.066 |
hard | -0.063 |
bad | -0.060 |
lot | -0.060 |
Token | castle |
Token ID | 16669 |
Feature activation | +4.32e-03 |
Pos logit contributions | |
Cr | +0.075 |
ados | +0.072 |
ishment | +0.068 |
icans | +0.065 |
amb | +0.064 |
Neg logit contributions | |
much | -0.090 |
wonderful | -0.080 |
hard | -0.077 |
big | -0.074 |
bad | -0.073 |
Token | where |
Token ID | 810 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
much | +0.007 |
wonderful | +0.006 |
lessly | +0.006 |
br | +0.005 |
In | +0.005 |
Neg logit contributions | |
ados | -0.009 |
Majesty | -0.009 |
Cr | -0.008 |
ishment | -0.008 |
gy | -0.008 |
Token | he |
Token ID | 339 |
Feature activation | -5.89e-03 |
Pos logit contributions | |
much | +0.040 |
wonderful | +0.030 |
lessly | +0.029 |
hard | +0.029 |
On | +0.027 |
Neg logit contributions | |
ados | -0.062 |
ishment | -0.062 |
Majesty | -0.060 |
Cr | -0.059 |
amb | -0.057 |
Token | was |
Token ID | 373 |
Feature activation | +7.44e-03 |
Pos logit contributions | |
Cr | +0.013 |
Majesty | +0.013 |
ados | +0.012 |
ishment | +0.012 |
gy | +0.012 |
Neg logit contributions | |
much | -0.008 |
lessly | -0.007 |
wonderful | -0.007 |
On | -0.006 |
br | -0.006 |
Token | safe |
Token ID | 3338 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
much | +0.012 |
lessly | +0.010 |
wonderful | +0.010 |
On | +0.010 |
In | +0.010 |
Neg logit contributions | |
ados | -0.015 |
Cr | -0.015 |
Majesty | -0.014 |
ishment | -0.014 |
gy | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -3.09e-02 |
Pos logit contributions | |
ados | +0.040 |
Cr | +0.040 |
Majesty | +0.040 |
ishment | +0.039 |
gy | +0.037 |
Neg logit contributions | |
much | -0.027 |
br | -0.023 |
lessly | -0.023 |
wonderful | -0.023 |
On | -0.022 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
ados | +0.070 |
ishment | +0.068 |
Majesty | +0.068 |
Cr | +0.067 |
lee | +0.064 |
Neg logit contributions | |
much | -0.046 |
On | -0.035 |
wonderful | -0.035 |
lessly | -0.035 |
In | -0.034 |
Token | L |
Token ID | 43 |
Feature activation | -5.62e-02 |
Pos logit contributions | |
Cr | +0.023 |
ados | +0.023 |
Fro | +0.021 |
ishment | +0.021 |
Rib | +0.021 |
Neg logit contributions | |
much | -0.019 |
lessly | -0.017 |
wonderful | -0.017 |
br | -0.016 |
On | -0.016 |
Token | ily |
Token ID | 813 |
Feature activation | -4.80e-02 |
Pos logit contributions | |
Cr | +0.112 |
ados | +0.111 |
Majesty | +0.108 |
ishment | +0.108 |
ican | +0.101 |
Neg logit contributions | |
much | -0.101 |
wonderful | -0.083 |
lessly | -0.081 |
On | -0.080 |
In | -0.079 |
Token | and |
Token ID | 290 |
Feature activation | -6.28e-02 |
Pos logit contributions | |
ados | +0.116 |
Cr | +0.114 |
ishment | +0.112 |
Majesty | +0.111 |
ican | +0.103 |
Neg logit contributions | |
much | -0.064 |
wonderful | -0.051 |
In | -0.049 |
On | -0.047 |
hard | -0.047 |
Token | Ben |
Token ID | 3932 |
Feature activation | -3.67e-02 |
Pos logit contributions | |
Majesty | +0.139 |
ishment | +0.132 |
ados | +0.124 |
ans | +0.124 |
ieties | +0.123 |
Neg logit contributions | |
much | -0.116 |
wonderful | -0.092 |
big | -0.090 |
bad | -0.089 |
on | -0.088 |
Token | were |
Token ID | 547 |
Feature activation | -5.03e-03 |
Pos logit contributions | |
ados | +0.089 |
Cr | +0.088 |
Majesty | +0.087 |
ishment | +0.087 |
lee | +0.082 |
Neg logit contributions | |
much | -0.046 |
wonderful | -0.036 |
lessly | -0.034 |
On | -0.033 |
br | -0.033 |
Token | playing |
Token ID | 2712 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
ados | +0.011 |
Cr | +0.011 |
ishment | +0.011 |
Majesty | +0.011 |
ican | +0.011 |
Neg logit contributions | |
much | -0.008 |
wonderful | -0.006 |
hard | -0.006 |
lot | -0.006 |
In | -0.005 |
Token | in |
Token ID | 287 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
lee | +0.037 |
Majesty | +0.037 |
ishment | +0.036 |
Cr | +0.036 |
ados | +0.036 |
Neg logit contributions | |
much | -0.012 |
wonderful | -0.009 |
on | -0.008 |
On | -0.008 |
hard | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
Majesty | +0.041 |
ados | +0.039 |
ishment | +0.038 |
Cr | +0.038 |
amb | +0.037 |
Neg logit contributions | |
much | -0.029 |
wonderful | -0.024 |
On | -0.024 |
hard | -0.023 |
In | -0.022 |
Token | park |
Token ID | 3952 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
ados | +0.092 |
Cr | +0.091 |
ishment | +0.089 |
Majesty | +0.088 |
cess | +0.085 |
Neg logit contributions | |
much | -0.075 |
wonderful | -0.066 |
On | -0.061 |
In | -0.061 |
hard | -0.060 |
Token | want |
Token ID | 765 |
Feature activation | -4.10e-02 |
Pos logit contributions | |
Majesty | +0.094 |
Cr | +0.093 |
ados | +0.090 |
ishment | +0.089 |
gy | +0.087 |
Neg logit contributions | |
much | -0.085 |
lessly | -0.076 |
br | -0.073 |
On | -0.070 |
wonderful | -0.070 |
Token | a |
Token ID | 257 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
ados | +0.073 |
Cr | +0.072 |
ishment | +0.072 |
Majesty | +0.071 |
lee | +0.066 |
Neg logit contributions | |
much | -0.078 |
wonderful | -0.066 |
lessly | -0.064 |
br | -0.063 |
On | -0.063 |
Token | hug |
Token ID | 16225 |
Feature activation | -5.93e-02 |
Pos logit contributions | |
Cr | +0.049 |
Majesty | +0.048 |
ados | +0.047 |
ishment | +0.045 |
Fro | +0.045 |
Neg logit contributions | |
much | -0.041 |
lessly | -0.039 |
In | -0.035 |
On | -0.035 |
br | -0.033 |
Token | after |
Token ID | 706 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
ados | +0.121 |
Cr | +0.121 |
ishment | +0.119 |
Majesty | +0.118 |
lee | +0.112 |
Neg logit contributions | |
much | -0.097 |
wonderful | -0.079 |
lessly | -0.078 |
br | -0.076 |
On | -0.075 |
Token | playing |
Token ID | 2712 |
Feature activation | -5.57e-02 |
Pos logit contributions | |
Cr | +0.066 |
ishment | +0.066 |
Majesty | +0.065 |
ados | +0.064 |
gy | +0.060 |
Neg logit contributions | |
much | -0.083 |
wonderful | -0.070 |
hard | -0.067 |
lessly | -0.067 |
On | -0.066 |
Token | in |
Token ID | 287 |
Feature activation | -5.01e-02 |
Pos logit contributions | |
ishment | +0.123 |
Cr | +0.123 |
Majesty | +0.123 |
ados | +0.122 |
lee | +0.120 |
Neg logit contributions | |
much | -0.083 |
wonderful | -0.067 |
hard | -0.062 |
br | -0.061 |
On | -0.059 |
Token | the |
Token ID | 262 |
Feature activation | -5.60e-02 |
Pos logit contributions | |
ados | +0.079 |
Cr | +0.078 |
Majesty | +0.076 |
ishment | +0.075 |
lee | +0.070 |
Neg logit contributions | |
much | -0.106 |
wonderful | -0.092 |
lessly | -0.089 |
On | -0.088 |
br | -0.087 |
Token | rain |
Token ID | 6290 |
Feature activation | -4.12e-02 |
Pos logit contributions | |
ados | +0.095 |
Cr | +0.094 |
Majesty | +0.091 |
ishment | +0.091 |
lee | +0.084 |
Neg logit contributions | |
much | -0.110 |
lessly | -0.094 |
wonderful | -0.094 |
On | -0.091 |
br | -0.091 |
Token | ." |
Token ID | 526 |
Feature activation | -5.60e-02 |
Pos logit contributions | |
ados | +0.085 |
Cr | +0.084 |
ishment | +0.081 |
Majesty | +0.081 |
lee | +0.077 |
Neg logit contributions | |
much | -0.067 |
wonderful | -0.055 |
lessly | -0.054 |
br | -0.053 |
On | -0.052 |
Token | Lily |
Token ID | 20037 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
Majesty | +0.148 |
ishment | +0.147 |
ados | +0.145 |
ishes | +0.140 |
urg | +0.138 |
Neg logit contributions | |
much | -0.062 |
On | -0.044 |
He | -0.044 |
In | -0.042 |
It | -0.037 |
Token | smiled |
Token ID | 13541 |
Feature activation | -3.29e-02 |
Pos logit contributions | |
ados | +0.087 |
Cr | +0.080 |
Majesty | +0.079 |
ishment | +0.079 |
keeping | +0.076 |
Neg logit contributions | |
much | -0.038 |
had | -0.031 |
wonderful | -0.030 |
also | -0.030 |
On | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
ados | +0.036 |
Cr | +0.036 |
Majesty | +0.035 |
ishment | +0.035 |
lee | +0.032 |
Neg logit contributions | |
much | -0.043 |
wonderful | -0.037 |
lessly | -0.037 |
br | -0.036 |
On | -0.035 |
Token | rabbit |
Token ID | 22746 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
lessly | +0.021 |
much | +0.021 |
br | +0.020 |
On | +0.020 |
In | +0.020 |
Neg logit contributions | |
Cr | -0.014 |
ados | -0.013 |
Majesty | -0.013 |
ishment | -0.012 |
Fro | -0.012 |
Token | home |
Token ID | 1363 |
Feature activation | +8.28e-03 |
Pos logit contributions | |
ados | +0.034 |
ishment | +0.034 |
Cr | +0.033 |
Majesty | +0.033 |
lee | +0.031 |
Neg logit contributions | |
much | -0.031 |
wonderful | -0.026 |
lessly | -0.024 |
hard | -0.024 |
On | -0.024 |
Token | to |
Token ID | 284 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
much | +0.012 |
wonderful | +0.010 |
lessly | +0.010 |
br | +0.010 |
On | +0.010 |
Neg logit contributions | |
ados | -0.018 |
Cr | -0.017 |
Majesty | -0.017 |
ishment | -0.017 |
lee | -0.016 |
Token | her |
Token ID | 607 |
Feature activation | -3.72e-02 |
Pos logit contributions | |
ados | +0.063 |
Cr | +0.060 |
ishment | +0.060 |
Majesty | +0.059 |
icans | +0.054 |
Neg logit contributions | |
much | -0.083 |
wonderful | -0.072 |
lessly | -0.068 |
On | -0.067 |
hard | -0.067 |
Token | Mum |
Token ID | 38367 |
Feature activation | -4.76e-02 |
Pos logit contributions | |
ados | +0.060 |
Cr | +0.060 |
ishment | +0.057 |
Majesty | +0.055 |
lee | +0.052 |
Neg logit contributions | |
much | -0.078 |
wonderful | -0.068 |
On | -0.064 |
lessly | -0.064 |
hard | -0.064 |
Token | . |
Token ID | 13 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
ados | +0.113 |
Cr | +0.113 |
Majesty | +0.113 |
ishment | +0.110 |
ican | +0.103 |
Neg logit contributions | |
much | -0.061 |
wonderful | -0.048 |
On | -0.045 |
lessly | -0.045 |
hard | -0.043 |
Token | But |
Token ID | 887 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
ishment | +0.040 |
ados | +0.040 |
Majesty | +0.039 |
Cr | +0.038 |
icans | +0.037 |
Neg logit contributions | |
much | -0.030 |
On | -0.023 |
lessly | -0.023 |
In | -0.023 |
wonderful | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
much | +0.043 |
wonderful | +0.035 |
lessly | +0.034 |
On | +0.034 |
hard | +0.033 |
Neg logit contributions | |
ados | -0.050 |
Majesty | -0.048 |
Cr | -0.048 |
ishment | -0.048 |
lee | -0.045 |
Token | before |
Token ID | 878 |
Feature activation | +3.77e-02 |
Pos logit contributions | |
much | +0.047 |
wonderful | +0.039 |
lessly | +0.039 |
On | +0.038 |
br | +0.037 |
Neg logit contributions | |
ados | -0.050 |
Cr | -0.050 |
Majesty | -0.049 |
ishment | -0.049 |
lee | -0.045 |
Token | she |
Token ID | 673 |
Feature activation | +2.75e-03 |
Pos logit contributions | |
much | +0.068 |
lessly | +0.063 |
wonderful | +0.062 |
br | +0.060 |
In | +0.059 |
Neg logit contributions | |
ados | -0.064 |
Cr | -0.064 |
Fro | -0.058 |
lee | -0.058 |
ishment | -0.058 |
Token | , |
Token ID | 11 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
ados | +0.009 |
Cr | +0.009 |
Majesty | +0.009 |
ishment | +0.008 |
Fro | +0.008 |
Neg logit contributions | |
much | -0.009 |
lessly | -0.008 |
wonderful | -0.008 |
On | -0.008 |
br | -0.008 |
Token | there |
Token ID | 612 |
Feature activation | +8.71e-03 |
Pos logit contributions | |
Cr | +0.032 |
ados | +0.032 |
gy | +0.032 |
Majesty | +0.031 |
ishment | +0.031 |
Neg logit contributions | |
much | -0.021 |
lessly | -0.018 |
br | -0.018 |
wonderful | -0.017 |
On | -0.016 |
Token | was |
Token ID | 373 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
much | +0.012 |
lessly | +0.011 |
br | +0.010 |
wonderful | +0.010 |
On | +0.009 |
Neg logit contributions | |
Cr | -0.018 |
Fro | -0.018 |
gy | -0.018 |
ados | -0.017 |
Majesty | -0.017 |
Token | a |
Token ID | 257 |
Feature activation | -3.01e-03 |
Pos logit contributions | |
ishment | +0.029 |
ados | +0.029 |
Majesty | +0.029 |
Cr | +0.028 |
ishes | +0.027 |
Neg logit contributions | |
much | -0.013 |
wonderful | -0.009 |
On | -0.008 |
In | -0.008 |
hard | -0.008 |
Token | little |
Token ID | 1310 |
Feature activation | -4.81e-02 |
Pos logit contributions | |
ados | +0.007 |
Cr | +0.007 |
ishment | +0.007 |
Majesty | +0.007 |
lee | +0.006 |
Neg logit contributions | |
much | -0.004 |
wonderful | -0.003 |
lessly | -0.003 |
On | -0.003 |
br | -0.003 |
Token | boy |
Token ID | 2933 |
Feature activation | -4.76e-02 |
Pos logit contributions | |
ados | +0.098 |
ishment | +0.096 |
Majesty | +0.096 |
Cr | +0.095 |
keeping | +0.094 |
Neg logit contributions | |
much | -0.081 |
wonderful | -0.070 |
hard | -0.068 |
lot | -0.066 |
big | -0.065 |
Token | named |
Token ID | 3706 |
Feature activation | -6.90e-02 |
Pos logit contributions | |
ados | +0.094 |
Majesty | +0.092 |
ishment | +0.090 |
Cr | +0.088 |
icans | +0.081 |
Neg logit contributions | |
much | -0.090 |
wonderful | -0.071 |
On | -0.069 |
In | -0.069 |
hard | -0.069 |
Token | Tim |
Token ID | 5045 |
Feature activation | -4.79e-02 |
Pos logit contributions | |
ishment | +0.100 |
ados | +0.097 |
Majesty | +0.092 |
ieties | +0.087 |
ican | +0.084 |
Neg logit contributions | |
much | -0.167 |
wonderful | -0.140 |
hard | -0.139 |
In | -0.136 |
On | -0.135 |
Token | my |
Token ID | 1820 |
Feature activation | -3.86e-02 |
Pos logit contributions | |
Cr | +0.072 |
ados | +0.072 |
ishment | +0.070 |
gy | +0.070 |
Majesty | +0.068 |
Neg logit contributions | |
much | -0.096 |
br | -0.087 |
lessly | -0.086 |
wonderful | -0.084 |
On | -0.081 |
Token | . |
Token ID | 13 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
Cr | +0.076 |
ados | +0.076 |
gy | +0.074 |
ishment | +0.073 |
Majesty | +0.073 |
Neg logit contributions | |
much | -0.060 |
lessly | -0.053 |
br | -0.052 |
wonderful | -0.051 |
On | -0.049 |
Token | Tim |
Token ID | 5045 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
Majesty | +0.016 |
ados | +0.015 |
ishment | +0.015 |
Cr | +0.014 |
lee | +0.013 |
Neg logit contributions | |
much | -0.032 |
On | -0.028 |
wonderful | -0.028 |
lessly | -0.028 |
In | -0.027 |
Token | turned |
Token ID | 2900 |
Feature activation | -6.35e-03 |
Pos logit contributions | |
much | +0.031 |
wonderful | +0.026 |
lessly | +0.025 |
br | +0.025 |
On | +0.024 |
Neg logit contributions | |
Cr | -0.050 |
ados | -0.050 |
Majesty | -0.048 |
ishment | -0.048 |
lee | -0.046 |
Token | out |
Token ID | 503 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
ados | +0.012 |
Cr | +0.011 |
ishment | +0.011 |
Majesty | +0.011 |
lee | +0.010 |
Neg logit contributions | |
much | -0.011 |
wonderful | -0.010 |
On | -0.010 |
lessly | -0.010 |
In | -0.010 |
Token | that |
Token ID | 326 |
Feature activation | +1.25e-03 |
Pos logit contributions | |
ados | +0.018 |
Cr | +0.017 |
ishment | +0.017 |
lee | +0.016 |
Majesty | +0.016 |
Neg logit contributions | |
much | -0.018 |
br | -0.016 |
In | -0.016 |
wonderful | -0.016 |
On | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | -6.19e-03 |
Pos logit contributions | |
much | +0.002 |
lessly | +0.002 |
wonderful | +0.002 |
On | +0.002 |
br | +0.002 |
Neg logit contributions | |
ados | -0.002 |
Cr | -0.002 |
ishment | -0.002 |
Majesty | -0.002 |
lee | -0.002 |
Token | av |
Token ID | 1196 |
Feature activation | -6.37e-02 |
Pos logit contributions | |
ados | +0.009 |
Cr | +0.009 |
Majesty | +0.009 |
ishment | +0.009 |
Fro | +0.008 |
Neg logit contributions | |
much | -0.013 |
lessly | -0.012 |
On | -0.011 |
br | -0.011 |
wonderful | -0.011 |
Token | oc |
Token ID | 420 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
Cr | +0.101 |
ados | +0.098 |
Majesty | +0.097 |
ishment | +0.096 |
lee | +0.090 |
Neg logit contributions | |
much | -0.134 |
br | -0.114 |
wonderful | -0.114 |
lessly | -0.113 |
On | -0.111 |
Token | ados |
Token ID | 22484 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
Fro | +0.076 |
Cr | +0.070 |
Rib | +0.066 |
fro | +0.064 |
ishment | +0.064 |
Neg logit contributions | |
much | -0.316 |
br | -0.291 |
lessly | -0.288 |
ier | -0.288 |
hard | -0.286 |
Token | were |
Token ID | 547 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
much | +0.031 |
wonderful | +0.026 |
lessly | +0.025 |
br | +0.024 |
On | +0.024 |
Neg logit contributions | |
ados | -0.048 |
Cr | -0.047 |
ishment | -0.045 |
Majesty | -0.044 |
gy | -0.044 |
Token | bad |
Token ID | 2089 |
Feature activation | -3.63e-02 |
Pos logit contributions | |
br | +0.026 |
much | +0.026 |
On | +0.025 |
In | +0.025 |
arest | +0.024 |
Neg logit contributions | |
ados | -0.045 |
Cr | -0.043 |
lee | -0.041 |
ishment | -0.041 |
Majesty | -0.041 |
Token | and |
Token ID | 290 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
Cr | +0.081 |
ados | +0.078 |
ishment | +0.078 |
Majesty | +0.077 |
ican | +0.073 |
Neg logit contributions | |
much | -0.055 |
wonderful | -0.044 |
lessly | -0.043 |
hard | -0.042 |
br | -0.040 |
Token | made |
Token ID | 925 |
Feature activation | +5.87e-03 |
Pos logit contributions | |
ados | +0.027 |
Cr | +0.027 |
Majesty | +0.026 |
ishment | +0.026 |
lee | +0.025 |
Neg logit contributions | |
much | -0.020 |
wonderful | -0.016 |
lessly | -0.016 |
On | -0.015 |
br | -0.015 |
Token | there |
Token ID | 612 |
Feature activation | -9.90e-03 |
Pos logit contributions | |
Cr | +0.025 |
ados | +0.024 |
Majesty | +0.023 |
ishment | +0.023 |
Fro | +0.022 |
Neg logit contributions | |
much | -0.024 |
wonderful | -0.021 |
lessly | -0.020 |
br | -0.020 |
In | -0.020 |
Token | was |
Token ID | 373 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
ados | +0.025 |
Cr | +0.024 |
Majesty | +0.024 |
ishment | +0.024 |
lee | +0.023 |
Neg logit contributions | |
much | -0.012 |
wonderful | -0.009 |
lessly | -0.009 |
On | -0.008 |
In | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | +1.44e-03 |
Pos logit contributions | |
Cr | +0.032 |
ados | +0.031 |
Majesty | +0.031 |
ishment | +0.030 |
Fro | +0.029 |
Neg logit contributions | |
much | -0.022 |
lessly | -0.019 |
br | -0.018 |
wonderful | -0.018 |
On | -0.017 |
Token | rh |
Token ID | 9529 |
Feature activation | -4.39e-02 |
Pos logit contributions | |
much | +0.002 |
lessly | +0.002 |
br | +0.002 |
In | +0.002 |
On | +0.002 |
Neg logit contributions | |
Majesty | -0.003 |
Cr | -0.003 |
ados | -0.003 |
ishment | -0.003 |
Fro | -0.003 |
Token | in |
Token ID | 259 |
Feature activation | -3.82e-02 |
Pos logit contributions | |
ados | +0.091 |
Cr | +0.088 |
ishment | +0.084 |
Majesty | +0.083 |
ican | +0.082 |
Neg logit contributions | |
much | -0.068 |
wonderful | -0.059 |
lessly | -0.057 |
On | -0.055 |
In | -0.054 |
Token | oc |
Token ID | 420 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
Cr | +0.048 |
ados | +0.048 |
Majesty | +0.046 |
ishment | +0.045 |
gy | +0.041 |
Neg logit contributions | |
much | -0.090 |
lessly | -0.081 |
wonderful | -0.080 |
On | -0.078 |
br | -0.078 |
Token | eros |
Token ID | 27498 |
Feature activation | -3.66e-02 |
Pos logit contributions | |
ishment | +0.119 |
Jenna | +0.114 |
Cr | +0.111 |
lee | +0.109 |
Po | +0.106 |
Neg logit contributions | |
much | -0.277 |
br | -0.261 |
ier | -0.257 |
hard | -0.252 |
lessly | -0.251 |
Token | who |
Token ID | 508 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
ados | +0.076 |
Cr | +0.075 |
Majesty | +0.073 |
ishment | +0.073 |
lee | +0.069 |
Neg logit contributions | |
much | -0.058 |
wonderful | -0.048 |
lessly | -0.048 |
On | -0.046 |
br | -0.046 |
Token | was |
Token ID | 373 |
Feature activation | +7.55e-03 |
Pos logit contributions | |
Cr | +0.023 |
ados | +0.022 |
gy | +0.022 |
Majesty | +0.022 |
ishment | +0.021 |
Neg logit contributions | |
much | -0.015 |
lessly | -0.013 |
br | -0.012 |
wonderful | -0.012 |
On | -0.012 |
Token | always |
Token ID | 1464 |
Feature activation | -3.35e-03 |
Pos logit contributions | |
much | +0.007 |
lessly | +0.007 |
br | +0.006 |
wonderful | +0.006 |
On | +0.006 |
Neg logit contributions | |
Cr | -0.019 |
ados | -0.019 |
Majesty | -0.019 |
ishment | -0.018 |
gy | -0.018 |
Token | naughty |
Token ID | 45128 |
Feature activation | -4.30e-02 |
Pos logit contributions | |
Cr | +0.006 |
ados | +0.006 |
ishment | +0.006 |
Majesty | +0.006 |
gy | +0.006 |
Neg logit contributions | |
much | -0.005 |
br | -0.005 |
lessly | -0.005 |
In | -0.005 |
On | -0.005 |
Token | big |
Token ID | 1263 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
Cr | +0.008 |
Majesty | +0.008 |
ados | +0.008 |
fro | +0.007 |
Fro | +0.007 |
Neg logit contributions | |
lessly | -0.008 |
much | -0.008 |
In | -0.007 |
On | -0.007 |
br | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
ados | +0.057 |
Cr | +0.056 |
Majesty | +0.055 |
ishment | +0.055 |
lee | +0.050 |
Neg logit contributions | |
much | -0.066 |
wonderful | -0.056 |
lessly | -0.056 |
On | -0.055 |
br | -0.054 |
Token | beautiful |
Token ID | 4950 |
Feature activation | -4.27e-02 |
Pos logit contributions | |
Cr | +0.045 |
ados | +0.044 |
ishment | +0.042 |
Harry | +0.040 |
lee | +0.040 |
Neg logit contributions | |
much | -0.033 |
lessly | -0.030 |
br | -0.029 |
In | -0.029 |
t | -0.028 |
Token | rainbow |
Token ID | 27223 |
Feature activation | -1.63e-03 |
Pos logit contributions | |
ados | +0.069 |
Cr | +0.069 |
ishment | +0.067 |
Majesty | +0.066 |
lee | +0.061 |
Neg logit contributions | |
much | -0.087 |
wonderful | -0.076 |
lessly | -0.073 |
br | -0.072 |
On | -0.072 |
Token | - |
Token ID | 12 |
Feature activation | -6.09e-02 |
Pos logit contributions | |
ados | +0.003 |
Cr | +0.003 |
ishment | +0.003 |
Majesty | +0.003 |
lee | +0.003 |
Neg logit contributions | |
much | -0.002 |
wonderful | -0.002 |
lessly | -0.002 |
On | -0.002 |
In | -0.002 |
Token | col |
Token ID | 4033 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
ados | +0.102 |
Cr | +0.100 |
Majesty | +0.098 |
ishment | +0.098 |
lee | +0.090 |
Neg logit contributions | |
much | -0.122 |
wonderful | -0.104 |
lessly | -0.104 |
br | -0.101 |
On | -0.101 |
Token | oured |
Token ID | 8167 |
Feature activation | -5.70e-02 |
Pos logit contributions | |
Cr | +0.171 |
Majesty | +0.165 |
Fro | +0.165 |
Harry | +0.159 |
ados | +0.157 |
Neg logit contributions | |
much | -0.195 |
lessly | -0.185 |
br | -0.184 |
wonderful | -0.172 |
lic | -0.170 |
Token | butterfly |
Token ID | 35113 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
ishment | +0.100 |
Cr | +0.100 |
ados | +0.100 |
Majesty | +0.100 |
lee | +0.091 |
Neg logit contributions | |
much | -0.108 |
wonderful | -0.092 |
lessly | -0.090 |
br | -0.089 |
On | -0.086 |
Token | ! |
Token ID | 0 |
Feature activation | -4.25e-02 |
Pos logit contributions | |
Cr | +0.074 |
ishment | +0.073 |
ados | +0.073 |
Majesty | +0.072 |
Fro | +0.067 |
Neg logit contributions | |
much | -0.045 |
br | -0.036 |
wonderful | -0.035 |
lessly | -0.035 |
On | -0.033 |
Token | The |
Token ID | 383 |
Feature activation | +6.27e-04 |
Pos logit contributions | |
ishment | +0.078 |
Majesty | +0.078 |
ados | +0.075 |
ans | +0.073 |
ishes | +0.072 |
Neg logit contributions | |
much | -0.084 |
On | -0.070 |
lessly | -0.067 |
In | -0.067 |
It | -0.064 |
Token | little |
Token ID | 1310 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
much | +0.001 |
lessly | +0.001 |
On | +0.001 |
wonderful | +0.001 |
br | +0.001 |
Neg logit contributions | |
ados | -0.001 |
Cr | -0.001 |
Majesty | -0.001 |
ishment | -0.001 |
lee | -0.001 |
Token | 't |
Token ID | 470 |
Feature activation | -4.99e-02 |
Pos logit contributions | |
ados | +0.143 |
Cr | +0.142 |
Majesty | +0.138 |
ishment | +0.137 |
lee | +0.130 |
Neg logit contributions | |
much | -0.104 |
wonderful | -0.085 |
lessly | -0.085 |
On | -0.082 |
br | -0.081 |
Token | cry |
Token ID | 3960 |
Feature activation | -8.03e-02 |
Pos logit contributions | |
ados | +0.084 |
Cr | +0.082 |
ishment | +0.081 |
Majesty | +0.081 |
lee | +0.074 |
Neg logit contributions | |
much | -0.100 |
wonderful | -0.086 |
lessly | -0.085 |
On | -0.083 |
br | -0.082 |
Token | , |
Token ID | 11 |
Feature activation | -4.93e-02 |
Pos logit contributions | |
ados | +0.174 |
ishment | +0.172 |
Cr | +0.171 |
Majesty | +0.168 |
icans | +0.160 |
Neg logit contributions | |
much | -0.126 |
lessly | -0.099 |
hard | -0.096 |
wonderful | -0.094 |
bad | -0.089 |
Token | boys |
Token ID | 6510 |
Feature activation | -5.96e-02 |
Pos logit contributions | |
ishment | +0.089 |
ados | +0.088 |
Majesty | +0.086 |
lee | +0.080 |
gy | +0.079 |
Neg logit contributions | |
much | -0.099 |
wonderful | -0.085 |
br | -0.078 |
hard | -0.078 |
lessly | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | -5.44e-02 |
Pos logit contributions | |
ados | +0.132 |
Cr | +0.131 |
Majesty | +0.128 |
ishment | +0.128 |
lee | +0.120 |
Neg logit contributions | |
much | -0.087 |
wonderful | -0.070 |
lessly | -0.069 |
br | -0.065 |
On | -0.065 |
Token | Acc |
Token ID | 6366 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
ishment | +0.103 |
ados | +0.103 |
Majesty | +0.097 |
icans | +0.094 |
keeping | +0.094 |
Neg logit contributions | |
much | -0.103 |
In | -0.083 |
He | -0.080 |
wonderful | -0.080 |
It | -0.079 |
Token | idents |
Token ID | 3231 |
Feature activation | -5.54e-04 |
Pos logit contributions | |
ados | +0.147 |
keeping | +0.147 |
Fro | +0.146 |
alk | +0.146 |
Cr | +0.144 |
Neg logit contributions | |
much | -0.275 |
hard | -0.248 |
wrong | -0.242 |
bad | -0.241 |
better | -0.237 |
Token | happen |
Token ID | 1645 |
Feature activation | -3.39e-02 |
Pos logit contributions | |
Cr | +0.001 |
disl | +0.001 |
ados | +0.001 |
ors | +0.001 |
Twe | +0.001 |
Neg logit contributions | |
much | -0.001 |
On | -0.001 |
In | -0.001 |
br | -0.001 |
security | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -5.16e-02 |
Pos logit contributions | |
ados | +0.070 |
Cr | +0.069 |
Majesty | +0.067 |
ishment | +0.067 |
lee | +0.063 |
Neg logit contributions | |
much | -0.053 |
lessly | -0.045 |
wonderful | -0.044 |
On | -0.044 |
br | -0.043 |
Token | I |
Token ID | 314 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
ados | +0.094 |
ishment | +0.093 |
ican | +0.089 |
icans | +0.088 |
ans | +0.087 |
Neg logit contributions | |
much | -0.102 |
lessly | -0.081 |
In | -0.081 |
He | -0.080 |
wonderful | -0.080 |
Token | have |
Token ID | 423 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
Cr | +0.047 |
gy | +0.045 |
Majesty | +0.045 |
Fro | +0.045 |
disl | +0.043 |
Neg logit contributions | |
much | -0.040 |
br | -0.036 |
lessly | -0.035 |
On | -0.034 |
In | -0.033 |
Token | K |
Token ID | 42 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
much | +0.044 |
wonderful | +0.036 |
lessly | +0.034 |
On | +0.033 |
br | +0.033 |
Neg logit contributions | |
ados | -0.059 |
Cr | -0.057 |
Majesty | -0.056 |
ishment | -0.056 |
lee | -0.053 |
Token | itty |
Token ID | 9760 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
ados | +0.188 |
Cr | +0.184 |
ishment | +0.176 |
Majesty | +0.172 |
lee | +0.159 |
Neg logit contributions | |
much | -0.223 |
wonderful | -0.194 |
lessly | -0.193 |
In | -0.189 |
On | -0.188 |
Token | wanted |
Token ID | 2227 |
Feature activation | -4.84e-02 |
Pos logit contributions | |
ados | +0.259 |
ishment | +0.257 |
Cr | +0.256 |
Majesty | +0.247 |
lee | +0.233 |
Neg logit contributions | |
much | -0.113 |
wonderful | -0.084 |
br | -0.081 |
In | -0.080 |
lessly | -0.079 |
Token | to |
Token ID | 284 |
Feature activation | -7.50e-02 |
Pos logit contributions | |
ishment | +0.098 |
Majesty | +0.097 |
ados | +0.097 |
Cr | +0.094 |
lee | +0.089 |
Neg logit contributions | |
much | -0.085 |
wonderful | -0.071 |
hard | -0.067 |
On | -0.064 |
bad | -0.063 |
Token | introduce |
Token ID | 10400 |
Feature activation | -7.44e-02 |
Pos logit contributions | |
ados | +0.161 |
ishment | +0.160 |
Cr | +0.149 |
Majesty | +0.148 |
lee | +0.146 |
Neg logit contributions | |
much | -0.123 |
wonderful | -0.105 |
On | -0.095 |
bad | -0.094 |
big | -0.092 |
Token | Dog |
Token ID | 8532 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
Majesty | +0.116 |
ishment | +0.113 |
ados | +0.107 |
Cr | +0.106 |
lee | +0.099 |
Neg logit contributions | |
much | -0.164 |
wonderful | -0.139 |
lessly | -0.138 |
hard | -0.138 |
In | -0.132 |
Token | gy |
Token ID | 1360 |
Feature activation | -9.94e-02 |
Pos logit contributions | |
ados | +0.167 |
ishment | +0.162 |
Majesty | +0.160 |
Cr | +0.156 |
lee | +0.147 |
Neg logit contributions | |
much | -0.226 |
wonderful | -0.195 |
On | -0.188 |
ching | -0.188 |
hard | -0.185 |
Token | to |
Token ID | 284 |
Feature activation | -6.56e-02 |
Pos logit contributions | |
Cr | +0.190 |
ados | +0.190 |
ishment | +0.188 |
Majesty | +0.186 |
lee | +0.176 |
Neg logit contributions | |
much | -0.175 |
wonderful | -0.146 |
On | -0.141 |
lessly | -0.139 |
hard | -0.139 |
Token | her |
Token ID | 607 |
Feature activation | -5.73e-02 |
Pos logit contributions | |
ados | +0.111 |
Cr | +0.108 |
ishment | +0.106 |
Majesty | +0.106 |
lee | +0.098 |
Neg logit contributions | |
much | -0.131 |
wonderful | -0.113 |
lessly | -0.111 |
On | -0.109 |
br | -0.107 |
Token | ball |
Token ID | 2613 |
Feature activation | -4.73e-02 |
Pos logit contributions | |
ados | +0.115 |
Cr | +0.114 |
Majesty | +0.112 |
ishment | +0.111 |
lee | +0.104 |
Neg logit contributions | |
much | -0.093 |
lessly | -0.078 |
wonderful | -0.078 |
br | -0.075 |
On | -0.075 |
Token | . |
Token ID | 13 |
Feature activation | -4.89e-02 |
Pos logit contributions | |
ados | +0.106 |
Majesty | +0.106 |
ishment | +0.104 |
Cr | +0.103 |
lee | +0.099 |
Neg logit contributions | |
much | -0.069 |
lessly | -0.054 |
On | -0.053 |
wonderful | -0.053 |
br | -0.051 |