Token | and |
Token ID | 290 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
cowboy | +0.069 |
stronger | +0.069 |
generous | +0.069 |
helpful | +0.067 |
se | +0.066 |
Neg logit contributions | |
icky | -0.089 |
jo | -0.089 |
anne | -0.087 |
Jamie | -0.087 |
ash | -0.083 |
Token | sorry |
Token ID | 7926 |
Feature activation | +2.99e+00 |
Pos logit contributions | |
stronger | +0.060 |
cowboy | +0.059 |
Days | +0.057 |
vet | +0.054 |
Thank | +0.054 |
Neg logit contributions | |
Jamie | -0.113 |
jo | -0.113 |
anne | -0.112 |
icky | -0.110 |
ash | -0.106 |
Token | . |
Token ID | 13 |
Feature activation | +8.92e-01 |
Pos logit contributions | |
stronger | +0.111 |
generous | +0.107 |
helpful | +0.105 |
cowboy | +0.104 |
Thank | +0.103 |
Neg logit contributions | |
Jamie | -0.111 |
icky | -0.111 |
jo | -0.111 |
anne | -0.110 |
ash | -0.106 |
Token |
|
Token ID | 198 |
Feature activation | +3.99e+00 |
Pos logit contributions | |
Thank | +0.034 |
stronger | +0.033 |
vet | +0.033 |
Can | +0.033 |
generous | +0.033 |
Neg logit contributions | |
jo | -0.031 |
ash | -0.031 |
anne | -0.031 |
icky | -0.030 |
Jamie | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | +4.33e+00 |
Pos logit contributions | |
stronger | +0.138 |
cowboy | +0.132 |
generous | +0.131 |
Thank | +0.130 |
helpful | +0.129 |
Neg logit contributions | |
jo | -0.153 |
Jamie | -0.151 |
anne | -0.151 |
icky | -0.150 |
ash | -0.145 |
Token | " |
Token ID | 1 |
Feature activation | +1.06e+01 |
Pos logit contributions | |
stronger | +0.099 |
cowboy | +0.097 |
fuller | +0.088 |
generous | +0.085 |
helpful | +0.085 |
Neg logit contributions | |
jo | -0.219 |
icky | -0.219 |
Jamie | -0.217 |
anne | -0.217 |
umble | -0.205 |
Token | Yes |
Token ID | 5297 |
Feature activation | -3.82e+00 |
Pos logit contributions | |
brand | +0.293 |
stronger | +0.285 |
cowboy | +0.280 |
fuller | +0.256 |
helpful | +0.256 |
Neg logit contributions | |
Jamie | -0.448 |
anne | -0.445 |
icky | -0.425 |
Megan | -0.420 |
jo | -0.418 |
Token | , |
Token ID | 11 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
jo | +0.165 |
icky | +0.159 |
ash | +0.158 |
umble | +0.155 |
anne | +0.155 |
Neg logit contributions | |
stronger | -0.124 |
generous | -0.119 |
cowboy | -0.119 |
helpful | -0.118 |
nicer | -0.114 |
Token | he |
Token ID | 339 |
Feature activation | +4.09e+00 |
Pos logit contributions | |
stronger | +0.121 |
Days | +0.120 |
please | +0.119 |
Thank | +0.117 |
fuller | +0.117 |
Neg logit contributions | |
Jamie | -0.106 |
icky | -0.096 |
anne | -0.095 |
jo | -0.095 |
Megan | -0.091 |
Token | did |
Token ID | 750 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
stronger | +0.176 |
cowboy | +0.171 |
generous | +0.170 |
helpful | +0.168 |
Thank | +0.166 |
Neg logit contributions | |
jo | -0.124 |
Jamie | -0.121 |
icky | -0.120 |
anne | -0.120 |
ash | -0.113 |
Token | ," |
Token ID | 553 |
Feature activation | +3.31e+00 |
Pos logit contributions | |
stronger | +0.056 |
helpful | +0.054 |
generous | +0.052 |
cowboy | +0.050 |
nicer | +0.050 |
Neg logit contributions | |
ash | -0.063 |
jo | -0.061 |
anne | -0.060 |
icky | -0.059 |
umble | -0.058 |
Token | is |
Token ID | 318 |
Feature activation | +7.23e-01 |
Pos logit contributions | |
stronger | +0.111 |
cowboy | +0.107 |
generous | +0.105 |
helpful | +0.103 |
Thank | +0.102 |
Neg logit contributions | |
icky | -0.155 |
jo | -0.155 |
Jamie | -0.152 |
anne | -0.152 |
ash | -0.147 |
Token | sorry |
Token ID | 7926 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
cowboy | +0.019 |
stronger | +0.019 |
Days | +0.018 |
Thank | +0.018 |
generous | +0.018 |
Neg logit contributions | |
Jamie | -0.033 |
icky | -0.032 |
jo | -0.032 |
anne | -0.032 |
ash | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +5.14e-01 |
Pos logit contributions | |
stronger | +0.087 |
generous | +0.081 |
helpful | +0.080 |
cowboy | +0.077 |
nicer | +0.077 |
Neg logit contributions | |
anne | -0.096 |
jo | -0.093 |
loe | -0.093 |
Jamie | -0.092 |
ash | -0.092 |
Token |
|
Token ID | 198 |
Feature activation | +3.83e+00 |
Pos logit contributions | |
Thank | +0.017 |
stronger | +0.017 |
Can | +0.017 |
generous | +0.017 |
vet | +0.016 |
Neg logit contributions | |
ash | -0.020 |
jo | -0.020 |
anne | -0.020 |
icky | -0.020 |
ared | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | +4.20e+00 |
Pos logit contributions | |
stronger | +0.130 |
generous | +0.123 |
Thank | +0.122 |
helpful | +0.121 |
cowboy | +0.121 |
Neg logit contributions | |
jo | -0.149 |
anne | -0.148 |
Jamie | -0.147 |
icky | -0.145 |
ash | -0.143 |
Token | " |
Token ID | 1 |
Feature activation | +1.06e+01 |
Pos logit contributions | |
stronger | +0.083 |
cowboy | +0.081 |
generous | +0.074 |
helpful | +0.073 |
fuller | +0.071 |
Neg logit contributions | |
jo | -0.225 |
icky | -0.223 |
Jamie | -0.222 |
anne | -0.222 |
ash | -0.212 |
Token | I |
Token ID | 40 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
stronger | +0.281 |
cowboy | +0.280 |
brand | +0.270 |
fuller | +0.254 |
Days | +0.250 |
Neg logit contributions | |
Jamie | -0.461 |
anne | -0.457 |
icky | -0.443 |
jo | -0.441 |
Megan | -0.428 |
Token | 'm |
Token ID | 1101 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
stronger | +0.083 |
generous | +0.081 |
helpful | +0.080 |
cowboy | +0.080 |
Thank | +0.079 |
Neg logit contributions | |
jo | -0.082 |
Jamie | -0.081 |
anne | -0.080 |
icky | -0.080 |
ash | -0.076 |
Token | sorry |
Token ID | 7926 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
stronger | +0.006 |
cowboy | +0.006 |
generous | +0.006 |
helpful | +0.006 |
Thank | +0.006 |
Neg logit contributions | |
jo | -0.013 |
anne | -0.012 |
icky | -0.012 |
Jamie | -0.012 |
ash | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
stronger | +0.071 |
helpful | +0.070 |
generous | +0.070 |
kind | +0.065 |
wise | +0.062 |
Neg logit contributions | |
anne | -0.108 |
jo | -0.106 |
ash | -0.106 |
loe | -0.100 |
icky | -0.099 |
Token | Mia |
Token ID | 32189 |
Feature activation | -3.65e+00 |
Pos logit contributions | |
stronger | +0.109 |
Days | +0.107 |
Thank | +0.106 |
cowboy | +0.106 |
please | +0.105 |
Neg logit contributions | |
Jamie | -0.068 |
icky | -0.065 |
jo | -0.064 |
anne | -0.062 |
ash | -0.057 |
Token | for |
Token ID | 329 |
Feature activation | -8.72e-01 |
Pos logit contributions | |
stronger | +0.127 |
cowboy | +0.121 |
generous | +0.120 |
Thank | +0.117 |
helpful | +0.117 |
Neg logit contributions | |
Jamie | -0.131 |
jo | -0.130 |
anne | -0.128 |
icky | -0.128 |
ash | -0.122 |
Token | him |
Token ID | 683 |
Feature activation | +8.65e-01 |
Pos logit contributions | |
Jamie | +0.032 |
icky | +0.032 |
jo | +0.032 |
anne | +0.031 |
ash | +0.030 |
Neg logit contributions | |
stronger | -0.030 |
cowboy | -0.030 |
generous | -0.029 |
Thank | -0.029 |
Days | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | +6.77e-01 |
Pos logit contributions | |
stronger | +0.035 |
generous | +0.033 |
helpful | +0.033 |
cowboy | +0.032 |
Thank | +0.031 |
Neg logit contributions | |
jo | -0.030 |
Jamie | -0.030 |
anne | -0.029 |
icky | -0.029 |
ash | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +3.87e+00 |
Pos logit contributions | |
Thank | +0.026 |
stronger | +0.025 |
vet | +0.025 |
generous | +0.025 |
Can | +0.025 |
Neg logit contributions | |
jo | -0.023 |
anne | -0.023 |
ash | -0.023 |
icky | -0.023 |
Jamie | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | +4.23e+00 |
Pos logit contributions | |
stronger | +0.130 |
generous | +0.124 |
Thank | +0.123 |
cowboy | +0.123 |
helpful | +0.122 |
Neg logit contributions | |
jo | -0.151 |
Jamie | -0.149 |
anne | -0.149 |
icky | -0.148 |
ash | -0.144 |
Token | " |
Token ID | 1 |
Feature activation | +1.06e+01 |
Pos logit contributions | |
stronger | +0.093 |
cowboy | +0.090 |
fuller | +0.082 |
generous | +0.081 |
helpful | +0.080 |
Neg logit contributions | |
jo | -0.219 |
icky | -0.217 |
Jamie | -0.216 |
anne | -0.216 |
umble | -0.204 |
Token | Okay |
Token ID | 16454 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
stronger | +0.292 |
brand | +0.289 |
cowboy | +0.268 |
helpful | +0.261 |
fuller | +0.261 |
Neg logit contributions | |
Jamie | -0.455 |
anne | -0.433 |
jo | -0.423 |
icky | -0.422 |
Megan | -0.416 |
Token | , |
Token ID | 11 |
Feature activation | +3.43e+00 |
Pos logit contributions | |
jo | +0.094 |
icky | +0.093 |
anne | +0.091 |
ash | +0.090 |
umble | +0.089 |
Neg logit contributions | |
stronger | -0.089 |
generous | -0.086 |
helpful | -0.085 |
cowboy | -0.084 |
Thank | -0.081 |
Token | Tom |
Token ID | 4186 |
Feature activation | +3.86e-01 |
Pos logit contributions | |
stronger | +0.145 |
Days | +0.141 |
Thank | +0.139 |
please | +0.139 |
cowboy | +0.139 |
Neg logit contributions | |
Jamie | -0.104 |
jo | -0.096 |
icky | -0.095 |
anne | -0.094 |
Megan | -0.088 |
Token | . |
Token ID | 13 |
Feature activation | +4.62e-01 |
Pos logit contributions | |
stronger | +0.015 |
generous | +0.014 |
helpful | +0.014 |
cowboy | +0.014 |
Thank | +0.014 |
Neg logit contributions | |
icky | -0.014 |
jo | -0.014 |
Jamie | -0.013 |
anne | -0.013 |
ash | -0.013 |
Token | You |
Token ID | 921 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
stronger | +0.017 |
Thank | +0.017 |
generous | +0.017 |
cowboy | +0.017 |
vet | +0.017 |
Neg logit contributions | |
jo | -0.017 |
anne | -0.016 |
icky | -0.016 |
ash | -0.016 |
Jamie | -0.016 |
Token | say |
Token ID | 910 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
stronger | +0.088 |
cowboy | +0.085 |
generous | +0.083 |
Days | +0.083 |
helpful | +0.082 |
Neg logit contributions | |
jo | -0.123 |
Jamie | -0.122 |
anne | -0.121 |
icky | -0.120 |
ash | -0.114 |
Token | sorry |
Token ID | 7926 |
Feature activation | +4.26e+00 |
Pos logit contributions | |
stronger | +0.055 |
cowboy | +0.054 |
generous | +0.053 |
helpful | +0.052 |
Thank | +0.050 |
Neg logit contributions | |
jo | -0.062 |
anne | -0.061 |
icky | -0.061 |
Jamie | -0.060 |
ash | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
stronger | +0.155 |
generous | +0.146 |
helpful | +0.144 |
nicer | +0.142 |
cowboy | +0.140 |
Neg logit contributions | |
Jamie | -0.158 |
anne | -0.158 |
loe | -0.156 |
jo | -0.154 |
icky | -0.152 |
Token |
|
Token ID | 198 |
Feature activation | +4.08e+00 |
Pos logit contributions | |
stronger | +0.047 |
Thank | +0.046 |
generous | +0.045 |
vet | +0.045 |
cowboy | +0.044 |
Neg logit contributions | |
jo | -0.041 |
ash | -0.041 |
anne | -0.040 |
icky | -0.040 |
Jamie | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | +4.34e+00 |
Pos logit contributions | |
stronger | +0.130 |
generous | +0.124 |
Thank | +0.124 |
vet | +0.122 |
helpful | +0.121 |
Neg logit contributions | |
jo | -0.164 |
anne | -0.163 |
Jamie | -0.162 |
ash | -0.160 |
icky | -0.158 |
Token | " |
Token ID | 1 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
stronger | +0.097 |
cowboy | +0.096 |
fuller | +0.086 |
generous | +0.086 |
helpful | +0.086 |
Neg logit contributions | |
jo | -0.221 |
icky | -0.220 |
Jamie | -0.219 |
anne | -0.218 |
umble | -0.207 |
Token | We |
Token ID | 1135 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
brand | +0.306 |
cowboy | +0.276 |
stronger | +0.274 |
fuller | +0.254 |
helpful | +0.250 |
Neg logit contributions | |
Jamie | -0.423 |
anne | -0.422 |
icky | -0.405 |
Megan | -0.405 |
jo | -0.389 |
Token | were |
Token ID | 547 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
stronger | +0.059 |
cowboy | +0.058 |
generous | +0.057 |
helpful | +0.056 |
Thank | +0.056 |
Neg logit contributions | |
jo | -0.059 |
Jamie | -0.058 |
anne | -0.058 |
icky | -0.057 |
ash | -0.054 |
Token | naughty |
Token ID | 45128 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
stronger | +0.007 |
cowboy | +0.006 |
Thank | +0.006 |
Days | +0.006 |
generous | +0.006 |
Neg logit contributions | |
icky | -0.011 |
Jamie | -0.011 |
jo | -0.011 |
anne | -0.011 |
ash | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
stronger | +0.005 |
cowboy | +0.005 |
generous | +0.005 |
helpful | +0.005 |
se | +0.005 |
Neg logit contributions | |
anne | -0.006 |
jo | -0.006 |
icky | -0.006 |
Jamie | -0.006 |
ash | -0.006 |
Token | We |
Token ID | 775 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
jo | +0.009 |
anne | +0.009 |
icky | +0.009 |
Jamie | +0.008 |
ash | +0.008 |
Neg logit contributions | |
Thank | -0.010 |
generous | -0.010 |
stronger | -0.010 |
helpful | -0.010 |
vet | -0.010 |
Token | she |
Token ID | 673 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
jo | +0.085 |
anne | +0.079 |
ash | +0.078 |
ert | +0.075 |
Jamie | +0.072 |
Neg logit contributions | |
vet | -0.062 |
Thank | -0.062 |
Can | -0.058 |
There | -0.056 |
stronger | -0.056 |
Token | says |
Token ID | 1139 |
Feature activation | +5.77e-01 |
Pos logit contributions | |
stronger | +0.037 |
cowboy | +0.035 |
generous | +0.035 |
helpful | +0.034 |
Thank | +0.034 |
Neg logit contributions | |
jo | -0.057 |
icky | -0.056 |
anne | -0.055 |
Jamie | -0.055 |
ash | -0.054 |
Token | . |
Token ID | 13 |
Feature activation | +8.04e-01 |
Pos logit contributions | |
stronger | +0.020 |
cowboy | +0.020 |
generous | +0.019 |
helpful | +0.019 |
Thank | +0.018 |
Neg logit contributions | |
jo | -0.022 |
icky | -0.022 |
ash | -0.021 |
anne | -0.021 |
umble | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | +3.98e+00 |
Pos logit contributions | |
stronger | +0.025 |
Thank | +0.025 |
vet | +0.025 |
Can | +0.025 |
generous | +0.024 |
Neg logit contributions | |
jo | -0.032 |
ash | -0.032 |
anne | -0.032 |
icky | -0.031 |
ared | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | +4.23e+00 |
Pos logit contributions | |
stronger | +0.119 |
vet | +0.112 |
generous | +0.112 |
Thank | +0.111 |
helpful | +0.109 |
Neg logit contributions | |
anne | -0.166 |
jo | -0.166 |
ash | -0.164 |
Jamie | -0.161 |
icky | -0.158 |
Token | " |
Token ID | 1 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
stronger | +0.091 |
cowboy | +0.088 |
generous | +0.082 |
helpful | +0.080 |
fuller | +0.079 |
Neg logit contributions | |
jo | -0.219 |
icky | -0.218 |
Jamie | -0.218 |
anne | -0.216 |
ash | -0.206 |
Token | Okay |
Token ID | 16454 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
stronger | +0.300 |
brand | +0.298 |
helpful | +0.284 |
cowboy | +0.281 |
fuller | +0.275 |
Neg logit contributions | |
Jamie | -0.439 |
anne | -0.434 |
jo | -0.424 |
Megan | -0.420 |
icky | -0.420 |
Token | , |
Token ID | 11 |
Feature activation | +3.38e+00 |
Pos logit contributions | |
jo | +0.091 |
icky | +0.090 |
anne | +0.087 |
ash | +0.087 |
umble | +0.085 |
Neg logit contributions | |
stronger | -0.085 |
generous | -0.082 |
helpful | -0.080 |
cowboy | -0.078 |
Thank | -0.077 |
Token | mom |
Token ID | 1995 |
Feature activation | +7.25e-01 |
Pos logit contributions | |
stronger | +0.140 |
cowboy | +0.136 |
Days | +0.133 |
generous | +0.133 |
Thank | +0.133 |
Neg logit contributions | |
Jamie | -0.105 |
jo | -0.100 |
anne | -0.099 |
icky | -0.098 |
ash | -0.091 |
Token | , |
Token ID | 11 |
Feature activation | +3.41e+00 |
Pos logit contributions | |
stronger | +0.027 |
generous | +0.026 |
helpful | +0.026 |
Thank | +0.025 |
cowboy | +0.025 |
Neg logit contributions | |
icky | -0.026 |
jo | -0.025 |
anne | -0.025 |
Jamie | -0.025 |
ash | -0.024 |
Token | we |
Token ID | 356 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
stronger | +0.121 |
Days | +0.117 |
cowboy | +0.117 |
fuller | +0.116 |
please | +0.115 |
Neg logit contributions | |
Jamie | -0.127 |
jo | -0.114 |
icky | -0.113 |
anne | -0.112 |
Megan | -0.111 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
cowboy | +0.116 |
stronger | +0.111 |
fuller | +0.107 |
golf | +0.106 |
Days | +0.106 |
Neg logit contributions | |
Jamie | -0.098 |
jo | -0.097 |
anne | -0.096 |
icky | -0.091 |
bie | -0.087 |
Token | says |
Token ID | 1139 |
Feature activation | +8.34e-01 |
Pos logit contributions | |
icky | +0.071 |
jo | +0.069 |
anne | +0.068 |
Jamie | +0.068 |
ash | +0.066 |
Neg logit contributions | |
stronger | -0.046 |
Thank | -0.045 |
cowboy | -0.045 |
generous | -0.045 |
please | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | +7.19e-01 |
Pos logit contributions | |
stronger | +0.029 |
cowboy | +0.029 |
generous | +0.028 |
helpful | +0.028 |
Thank | +0.027 |
Neg logit contributions | |
jo | -0.032 |
icky | -0.032 |
anne | -0.031 |
ash | -0.030 |
umble | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | +3.88e+00 |
Pos logit contributions | |
stronger | +0.024 |
Thank | +0.023 |
Can | +0.023 |
vet | +0.023 |
generous | +0.023 |
Neg logit contributions | |
ash | -0.028 |
jo | -0.028 |
anne | -0.028 |
icky | -0.027 |
ared | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | +4.16e+00 |
Pos logit contributions | |
stronger | +0.106 |
Thank | +0.100 |
vet | +0.100 |
generous | +0.099 |
helpful | +0.098 |
Neg logit contributions | |
anne | -0.173 |
jo | -0.169 |
ash | -0.169 |
Jamie | -0.167 |
ared | -0.166 |
Token | " |
Token ID | 1 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
stronger | +0.085 |
cowboy | +0.082 |
generous | +0.076 |
helpful | +0.074 |
fuller | +0.074 |
Neg logit contributions | |
jo | -0.221 |
icky | -0.219 |
Jamie | -0.218 |
anne | -0.217 |
ash | -0.208 |
Token | I |
Token ID | 40 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
stronger | +0.308 |
brand | +0.302 |
fuller | +0.289 |
helpful | +0.286 |
cowboy | +0.283 |
Neg logit contributions | |
Jamie | -0.417 |
anne | -0.416 |
jo | -0.406 |
Today | -0.402 |
icky | -0.399 |
Token | love |
Token ID | 1842 |
Feature activation | +9.66e-01 |
Pos logit contributions | |
stronger | +0.072 |
cowboy | +0.071 |
generous | +0.069 |
helpful | +0.068 |
Thank | +0.068 |
Neg logit contributions | |
jo | -0.064 |
icky | -0.063 |
Jamie | -0.062 |
anne | -0.062 |
ash | -0.059 |
Token | you |
Token ID | 345 |
Feature activation | -3.83e+00 |
Pos logit contributions | |
stronger | +0.036 |
cowboy | +0.036 |
Days | +0.035 |
fuller | +0.034 |
generous | +0.034 |
Neg logit contributions | |
Jamie | -0.034 |
icky | -0.034 |
jo | -0.033 |
anne | -0.032 |
ash | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
jo | +0.143 |
anne | +0.140 |
Jamie | +0.140 |
icky | +0.139 |
ash | +0.132 |
Neg logit contributions | |
stronger | -0.141 |
cowboy | -0.136 |
generous | -0.134 |
helpful | -0.132 |
Thank | -0.130 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
stronger | +0.052 |
cowboy | +0.051 |
Thank | +0.050 |
generous | +0.050 |
helpful | +0.049 |
Neg logit contributions | |
Jamie | -0.037 |
jo | -0.037 |
icky | -0.037 |
anne | -0.036 |
ash | -0.034 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
stronger | +0.145 |
Thank | +0.144 |
please | +0.138 |
cowboy | +0.137 |
Did | +0.134 |
Neg logit contributions | |
Jamie | -0.193 |
jo | -0.192 |
icky | -0.192 |
anne | -0.187 |
bie | -0.180 |
Token | says |
Token ID | 1139 |
Feature activation | +8.34e-01 |
Pos logit contributions | |
icky | +0.071 |
jo | +0.069 |
anne | +0.068 |
Jamie | +0.068 |
ash | +0.066 |
Neg logit contributions | |
stronger | -0.046 |
Thank | -0.045 |
cowboy | -0.045 |
generous | -0.045 |
please | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | +7.19e-01 |
Pos logit contributions | |
stronger | +0.029 |
cowboy | +0.029 |
generous | +0.028 |
helpful | +0.028 |
Thank | +0.027 |
Neg logit contributions | |
jo | -0.032 |
icky | -0.032 |
anne | -0.031 |
ash | -0.030 |
umble | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | +3.88e+00 |
Pos logit contributions | |
stronger | +0.024 |
Thank | +0.023 |
Can | +0.023 |
vet | +0.023 |
generous | +0.023 |
Neg logit contributions | |
ash | -0.028 |
jo | -0.028 |
anne | -0.028 |
icky | -0.027 |
ared | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | +4.16e+00 |
Pos logit contributions | |
stronger | +0.106 |
Thank | +0.100 |
vet | +0.100 |
generous | +0.099 |
helpful | +0.098 |
Neg logit contributions | |
anne | -0.173 |
jo | -0.169 |
ash | -0.169 |
Jamie | -0.167 |
ared | -0.166 |
Token | " |
Token ID | 1 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
stronger | +0.085 |
cowboy | +0.082 |
generous | +0.076 |
helpful | +0.074 |
fuller | +0.074 |
Neg logit contributions | |
jo | -0.221 |
icky | -0.219 |
Jamie | -0.218 |
anne | -0.217 |
ash | -0.208 |
Token | Orange |
Token ID | 40141 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
stronger | +0.308 |
brand | +0.302 |
fuller | +0.289 |
helpful | +0.286 |
cowboy | +0.283 |
Neg logit contributions | |
Jamie | -0.417 |
anne | -0.416 |
jo | -0.406 |
Today | -0.402 |
icky | -0.399 |
Token | who |
Token ID | 508 |
Feature activation | -5.31e-01 |
Pos logit contributions | |
jo | +0.056 |
anne | +0.056 |
Jamie | +0.053 |
icky | +0.050 |
spine | +0.050 |
Neg logit contributions | |
stronger | -0.050 |
helpful | -0.050 |
generous | -0.050 |
kind | -0.047 |
cowboy | -0.047 |
Token | ?" |
Token ID | 1701 |
Feature activation | +7.00e-01 |
Pos logit contributions | |
jo | +0.020 |
anne | +0.019 |
icky | +0.019 |
Jamie | +0.019 |
ash | +0.018 |
Neg logit contributions | |
stronger | -0.019 |
cowboy | -0.019 |
generous | -0.018 |
helpful | -0.018 |
Thank | -0.018 |
Token | Max |
Token ID | 5436 |
Feature activation | +8.79e-02 |
Pos logit contributions | |
cowboy | +0.023 |
generous | +0.023 |
vet | +0.023 |
stronger | +0.023 |
Thank | +0.023 |
Neg logit contributions | |
ash | -0.029 |
icky | -0.028 |
jo | -0.028 |
anne | -0.027 |
ared | -0.027 |
Token | says |
Token ID | 1139 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
stronger | +0.004 |
cowboy | +0.004 |
generous | +0.004 |
se | +0.004 |
helpful | +0.004 |
Neg logit contributions | |
icky | -0.003 |
jo | -0.003 |
anne | -0.003 |
ash | -0.003 |
Jamie | -0.003 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
stronger | +0.004 |
helpful | +0.004 |
generous | +0.004 |
nicer | +0.004 |
nearby | +0.004 |
Neg logit contributions | |
anne | -0.005 |
jo | -0.005 |
Jamie | -0.005 |
raw | -0.004 |
loe | -0.004 |
Token | says |
Token ID | 1139 |
Feature activation | +8.34e-01 |
Pos logit contributions | |
icky | +0.071 |
jo | +0.069 |
anne | +0.068 |
Jamie | +0.068 |
ash | +0.066 |
Neg logit contributions | |
stronger | -0.046 |
Thank | -0.045 |
cowboy | -0.045 |
generous | -0.045 |
please | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | +7.19e-01 |
Pos logit contributions | |
stronger | +0.029 |
cowboy | +0.029 |
generous | +0.028 |
helpful | +0.028 |
Thank | +0.027 |
Neg logit contributions | |
jo | -0.032 |
icky | -0.032 |
anne | -0.031 |
ash | -0.030 |
umble | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | +3.88e+00 |
Pos logit contributions | |
stronger | +0.024 |
Thank | +0.023 |
Can | +0.023 |
vet | +0.023 |
generous | +0.023 |
Neg logit contributions | |
ash | -0.028 |
jo | -0.028 |
anne | -0.028 |
icky | -0.027 |
ared | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | +4.16e+00 |
Pos logit contributions | |
stronger | +0.106 |
Thank | +0.100 |
vet | +0.100 |
generous | +0.099 |
helpful | +0.098 |
Neg logit contributions | |
anne | -0.173 |
jo | -0.169 |
ash | -0.169 |
Jamie | -0.167 |
ared | -0.166 |
Token | " |
Token ID | 1 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
stronger | +0.085 |
cowboy | +0.082 |
generous | +0.076 |
helpful | +0.074 |
fuller | +0.074 |
Neg logit contributions | |
jo | -0.221 |
icky | -0.219 |
Jamie | -0.218 |
anne | -0.217 |
ash | -0.208 |
Token | That |
Token ID | 2504 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
stronger | +0.308 |
brand | +0.302 |
fuller | +0.289 |
helpful | +0.286 |
cowboy | +0.283 |
Neg logit contributions | |
Jamie | -0.417 |
anne | -0.416 |
jo | -0.406 |
Today | -0.402 |
icky | -0.399 |
Token | 's |
Token ID | 338 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
jo | +0.041 |
Jamie | +0.040 |
anne | +0.040 |
icky | +0.039 |
ash | +0.038 |
Neg logit contributions | |
stronger | -0.044 |
generous | -0.042 |
cowboy | -0.042 |
helpful | -0.042 |
vet | -0.041 |
Token | wonderful |
Token ID | 7932 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
stronger | +0.034 |
generous | +0.032 |
helpful | +0.031 |
cowboy | +0.030 |
Thank | +0.029 |
Neg logit contributions | |
jo | -0.086 |
anne | -0.085 |
Jamie | -0.083 |
icky | -0.081 |
ash | -0.079 |
Token | , |
Token ID | 11 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
jo | +0.016 |
icky | +0.016 |
anne | +0.015 |
ash | +0.015 |
Jamie | +0.015 |
Neg logit contributions | |
generous | -0.013 |
helpful | -0.013 |
stronger | -0.012 |
cowboy | -0.012 |
kind | -0.012 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
stronger | +0.058 |
cowboy | +0.057 |
generous | +0.056 |
helpful | +0.056 |
Thank | +0.055 |
Neg logit contributions | |
jo | -0.049 |
anne | -0.047 |
icky | -0.047 |
Jamie | -0.046 |
ash | -0.045 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.88e-01 |
Pos logit contributions | |
Thank | +0.018 |
vet | +0.018 |
Can | +0.017 |
cowboy | +0.017 |
generous | +0.017 |
Neg logit contributions | |
ash | -0.023 |
icky | -0.023 |
jo | -0.023 |
anne | -0.022 |
ert | -0.022 |
Token | says |
Token ID | 1139 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
icky | +0.013 |
jo | +0.013 |
anne | +0.013 |
Jamie | +0.013 |
ash | +0.012 |
Neg logit contributions | |
stronger | -0.008 |
generous | -0.008 |
Thank | -0.007 |
cowboy | -0.007 |
se | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
cowboy | +0.083 |
stronger | +0.083 |
Thank | +0.080 |
generous | +0.080 |
Days | +0.079 |
Neg logit contributions | |
Jamie | -0.050 |
jo | -0.049 |
icky | -0.049 |
anne | -0.049 |
ash | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | +3.87e+00 |
Pos logit contributions | |
stronger | +0.018 |
cowboy | +0.018 |
generous | +0.017 |
Thank | +0.017 |
helpful | +0.017 |
Neg logit contributions | |
jo | -0.019 |
anne | -0.019 |
icky | -0.019 |
Jamie | -0.019 |
ash | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +4.14e+00 |
Pos logit contributions | |
stronger | +0.117 |
vet | +0.110 |
Thank | +0.109 |
generous | +0.109 |
helpful | +0.108 |
Neg logit contributions | |
anne | -0.164 |
jo | -0.162 |
Jamie | -0.160 |
ash | -0.159 |
icky | -0.155 |
Token | " |
Token ID | 1 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
cowboy | +0.100 |
stronger | +0.100 |
fuller | +0.094 |
generous | +0.090 |
helpful | +0.088 |
Neg logit contributions | |
icky | -0.205 |
jo | -0.203 |
Jamie | -0.201 |
anne | -0.199 |
umble | -0.193 |
Token | No |
Token ID | 2949 |
Feature activation | -3.59e+00 |
Pos logit contributions | |
brand | +0.329 |
cowboy | +0.326 |
stronger | +0.323 |
fuller | +0.308 |
captain | +0.295 |
Neg logit contributions | |
anne | -0.390 |
icky | -0.379 |
umble | -0.378 |
Jamie | -0.377 |
Today | -0.374 |
Token | , |
Token ID | 11 |
Feature activation | +2.72e+00 |
Pos logit contributions | |
anne | +0.177 |
jo | +0.176 |
ash | +0.175 |
anny | +0.167 |
Jamie | +0.165 |
Neg logit contributions | |
cowboy | -0.098 |
stronger | -0.095 |
helpful | -0.093 |
generous | -0.090 |
vet | -0.089 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
stronger | +0.107 |
cowboy | +0.103 |
generous | +0.102 |
Thank | +0.101 |
Days | +0.100 |
Neg logit contributions | |
Jamie | -0.090 |
jo | -0.089 |
icky | -0.088 |
anne | -0.088 |
ash | -0.082 |
Token | . |
Token ID | 13 |
Feature activation | -4.30e-01 |
Pos logit contributions | |
jo | +0.048 |
anne | +0.047 |
icky | +0.047 |
Jamie | +0.047 |
ash | +0.046 |
Neg logit contributions | |
stronger | -0.037 |
helpful | -0.036 |
generous | -0.035 |
kind | -0.033 |
Thank | -0.033 |
Token | I |
Token ID | 314 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
jo | +0.017 |
anne | +0.017 |
Jamie | +0.015 |
ash | +0.015 |
icky | +0.015 |
Neg logit contributions | |
vet | -0.015 |
Thank | -0.015 |
Can | -0.015 |
generous | -0.014 |
stronger | -0.014 |
Token | Tom |
Token ID | 4186 |
Feature activation | +9.44e-01 |
Pos logit contributions | |
generous | +0.101 |
stronger | +0.101 |
cowboy | +0.100 |
Thank | +0.099 |
helpful | +0.098 |
Neg logit contributions | |
jo | -0.102 |
ash | -0.102 |
icky | -0.101 |
anne | -0.099 |
ared | -0.094 |
Token | says |
Token ID | 1139 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
stronger | +0.026 |
generous | +0.025 |
helpful | +0.023 |
kind | +0.022 |
cowboy | +0.022 |
Neg logit contributions | |
ash | -0.043 |
icky | -0.043 |
anne | -0.043 |
jo | -0.042 |
amp | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | +9.54e-01 |
Pos logit contributions | |
stronger | +0.080 |
generous | +0.078 |
cowboy | +0.077 |
helpful | +0.077 |
Thank | +0.075 |
Neg logit contributions | |
jo | -0.047 |
icky | -0.047 |
Jamie | -0.046 |
anne | -0.046 |
ash | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | +4.29e+00 |
Pos logit contributions | |
stronger | +0.033 |
cowboy | +0.032 |
generous | +0.031 |
Thank | +0.031 |
helpful | +0.031 |
Neg logit contributions | |
jo | -0.036 |
icky | -0.036 |
anne | -0.036 |
Jamie | -0.036 |
ash | -0.035 |
Token |
|
Token ID | 198 |
Feature activation | +4.49e+00 |
Pos logit contributions | |
stronger | +0.141 |
generous | +0.132 |
helpful | +0.130 |
vet | +0.130 |
Thank | +0.130 |
Neg logit contributions | |
jo | -0.170 |
anne | -0.170 |
Jamie | -0.169 |
ash | -0.167 |
icky | -0.166 |
Token | " |
Token ID | 1 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
cowboy | +0.096 |
stronger | +0.095 |
generous | +0.087 |
helpful | +0.085 |
fuller | +0.085 |
Neg logit contributions | |
jo | -0.234 |
anne | -0.231 |
icky | -0.231 |
Jamie | -0.231 |
umble | -0.216 |
Token | How |
Token ID | 2437 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
brand | +0.283 |
stronger | +0.282 |
fuller | +0.271 |
cowboy | +0.268 |
nicer | +0.256 |
Neg logit contributions | |
anne | -0.415 |
Jamie | -0.412 |
jo | -0.406 |
Megan | -0.405 |
icky | -0.389 |
Token | do |
Token ID | 466 |
Feature activation | +2.98e+00 |
Pos logit contributions | |
stronger | +0.098 |
generous | +0.097 |
helpful | +0.096 |
kind | +0.091 |
cowboy | +0.091 |
Neg logit contributions | |
jo | -0.089 |
anne | -0.088 |
ash | -0.086 |
ared | -0.083 |
Jamie | -0.082 |
Token | we |
Token ID | 356 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
stronger | +0.091 |
cowboy | +0.089 |
generous | +0.087 |
helpful | +0.087 |
vet | +0.081 |
Neg logit contributions | |
jo | -0.131 |
anne | -0.130 |
icky | -0.129 |
ash | -0.128 |
umble | -0.125 |
Token | do |
Token ID | 466 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
stronger | +0.049 |
cowboy | +0.049 |
Days | +0.047 |
Thank | +0.046 |
fuller | +0.045 |
Neg logit contributions | |
jo | -0.054 |
Jamie | -0.053 |
icky | -0.053 |
anne | -0.053 |
ash | -0.051 |
Token | that |
Token ID | 326 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
stronger | +0.077 |
cowboy | +0.075 |
Thank | +0.074 |
generous | +0.073 |
Days | +0.073 |
Neg logit contributions | |
Jamie | -0.085 |
jo | -0.085 |
icky | -0.085 |
anne | -0.083 |
ash | -0.079 |
Token | Ben |
Token ID | 3932 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
stronger | +0.116 |
generous | +0.116 |
cowboy | +0.116 |
Thank | +0.114 |
helpful | +0.113 |
Neg logit contributions | |
jo | -0.123 |
icky | -0.122 |
ash | -0.120 |
anne | -0.119 |
Jamie | -0.115 |
Token | says |
Token ID | 1139 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
stronger | +0.058 |
generous | +0.056 |
cowboy | +0.056 |
se | +0.054 |
helpful | +0.054 |
Neg logit contributions | |
jo | -0.057 |
icky | -0.057 |
anne | -0.055 |
ash | -0.054 |
obby | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
stronger | +0.095 |
cowboy | +0.092 |
generous | +0.091 |
helpful | +0.090 |
Thank | +0.089 |
Neg logit contributions | |
jo | -0.053 |
icky | -0.052 |
Jamie | -0.052 |
anne | -0.052 |
ash | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | +4.24e+00 |
Pos logit contributions | |
stronger | +0.034 |
cowboy | +0.033 |
generous | +0.032 |
helpful | +0.032 |
Thank | +0.032 |
Neg logit contributions | |
jo | -0.040 |
Jamie | -0.039 |
icky | -0.039 |
anne | -0.039 |
ash | -0.037 |
Token |
|
Token ID | 198 |
Feature activation | +4.42e+00 |
Pos logit contributions | |
stronger | +0.158 |
cowboy | +0.155 |
generous | +0.151 |
helpful | +0.149 |
Thank | +0.148 |
Neg logit contributions | |
jo | -0.151 |
icky | -0.149 |
Jamie | -0.149 |
anne | -0.148 |
ash | -0.140 |
Token | " |
Token ID | 1 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
cowboy | +0.112 |
stronger | +0.108 |
fuller | +0.101 |
generous | +0.098 |
helpful | +0.098 |
Neg logit contributions | |
icky | -0.213 |
Jamie | -0.212 |
jo | -0.212 |
anne | -0.209 |
umble | -0.195 |
Token | OK |
Token ID | 11380 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
brand | +0.298 |
stronger | +0.290 |
cowboy | +0.284 |
fuller | +0.280 |
helpful | +0.265 |
Neg logit contributions | |
Jamie | -0.422 |
anne | -0.420 |
jo | -0.416 |
icky | -0.411 |
Megan | -0.407 |
Token | . |
Token ID | 13 |
Feature activation | +6.99e-01 |
Pos logit contributions | |
jo | +0.072 |
icky | +0.070 |
anne | +0.070 |
Jamie | +0.068 |
ash | +0.067 |
Neg logit contributions | |
stronger | -0.078 |
generous | -0.074 |
cowboy | -0.074 |
helpful | -0.073 |
Thank | -0.071 |
Token | But |
Token ID | 887 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
stronger | +0.028 |
cowboy | +0.028 |
generous | +0.028 |
Thank | +0.027 |
helpful | +0.027 |
Neg logit contributions | |
jo | -0.023 |
icky | -0.022 |
anne | -0.022 |
Jamie | -0.022 |
ash | -0.021 |
Token | what |
Token ID | 644 |
Feature activation | -5.78e-01 |
Pos logit contributions | |
Jamie | +0.121 |
icky | +0.116 |
jo | +0.109 |
Today | +0.108 |
anne | +0.107 |
Neg logit contributions | |
cowboy | -0.121 |
stronger | -0.119 |
Days | -0.117 |
fuller | -0.116 |
please | -0.115 |
Token | is |
Token ID | 318 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
jo | +0.021 |
anne | +0.021 |
Jamie | +0.020 |
ash | +0.020 |
icky | +0.020 |
Neg logit contributions | |
stronger | -0.022 |
generous | -0.021 |
helpful | -0.021 |
cowboy | -0.020 |
kind | -0.020 |
Token | man |
Token ID | 582 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
jo | +0.028 |
anne | +0.027 |
icky | +0.027 |
Jamie | +0.027 |
ash | +0.025 |
Neg logit contributions | |
stronger | -0.053 |
cowboy | -0.051 |
generous | -0.051 |
Thank | -0.050 |
helpful | -0.050 |
Token | says |
Token ID | 1139 |
Feature activation | +9.10e-01 |
Pos logit contributions | |
stronger | +0.068 |
cowboy | +0.065 |
generous | +0.065 |
vet | +0.063 |
nearby | +0.062 |
Neg logit contributions | |
icky | -0.083 |
jo | -0.079 |
ash | -0.078 |
umble | -0.077 |
anne | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | +7.87e-01 |
Pos logit contributions | |
stronger | +0.034 |
cowboy | +0.033 |
generous | +0.033 |
helpful | +0.032 |
vet | +0.031 |
Neg logit contributions | |
icky | -0.032 |
jo | -0.032 |
anne | -0.031 |
ash | -0.031 |
umble | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | +4.13e+00 |
Pos logit contributions | |
stronger | +0.026 |
generous | +0.025 |
vet | +0.024 |
Thank | +0.024 |
cowboy | +0.024 |
Neg logit contributions | |
anne | -0.031 |
jo | -0.031 |
icky | -0.031 |
ash | -0.031 |
ared | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | +4.30e+00 |
Pos logit contributions | |
stronger | +0.125 |
vet | +0.117 |
generous | +0.116 |
Thank | +0.116 |
helpful | +0.113 |
Neg logit contributions | |
anne | -0.172 |
jo | -0.171 |
ash | -0.168 |
Jamie | -0.168 |
icky | -0.164 |
Token | " |
Token ID | 1 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
stronger | +0.090 |
cowboy | +0.087 |
generous | +0.082 |
helpful | +0.081 |
Thank | +0.079 |
Neg logit contributions | |
jo | -0.224 |
icky | -0.222 |
Jamie | -0.222 |
anne | -0.221 |
ash | -0.212 |
Token | I |
Token ID | 40 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
brand | +0.287 |
stronger | +0.277 |
cowboy | +0.268 |
helpful | +0.267 |
fuller | +0.262 |
Neg logit contributions | |
Jamie | -0.438 |
anne | -0.429 |
jo | -0.426 |
icky | -0.423 |
Megan | -0.420 |
Token | want |
Token ID | 765 |
Feature activation | +3.77e+00 |
Pos logit contributions | |
stronger | +0.075 |
cowboy | +0.073 |
generous | +0.072 |
helpful | +0.071 |
Thank | +0.071 |
Neg logit contributions | |
jo | -0.068 |
Jamie | -0.067 |
icky | -0.067 |
anne | -0.067 |
ash | -0.063 |
Token | chocolate |
Token ID | 11311 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
stronger | +0.125 |
cowboy | +0.118 |
generous | +0.117 |
helpful | +0.116 |
vet | +0.113 |
Neg logit contributions | |
jo | -0.157 |
anne | -0.155 |
icky | -0.151 |
Jamie | -0.151 |
ash | -0.146 |
Token | , |
Token ID | 11 |
Feature activation | +9.14e-01 |
Pos logit contributions | |
stronger | +0.059 |
cowboy | +0.056 |
generous | +0.055 |
helpful | +0.054 |
Thank | +0.054 |
Neg logit contributions | |
jo | -0.064 |
anne | -0.063 |
Jamie | -0.061 |
icky | -0.061 |
ash | -0.058 |
Token | it |
Token ID | 340 |
Feature activation | +7.02e-01 |
Pos logit contributions | |
stronger | +0.032 |
cowboy | +0.031 |
Thank | +0.031 |
Days | +0.031 |
generous | +0.030 |
Neg logit contributions | |
Jamie | -0.034 |
jo | -0.033 |
icky | -0.033 |
anne | -0.033 |
ash | -0.031 |
Token | is |
Token ID | 318 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
stronger | +0.128 |
cowboy | +0.124 |
generous | +0.121 |
helpful | +0.119 |
Thank | +0.117 |
Neg logit contributions | |
icky | -0.175 |
jo | -0.175 |
anne | -0.172 |
Jamie | -0.171 |
ash | -0.167 |
Token | sorry |
Token ID | 7926 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
cowboy | +0.032 |
stronger | +0.032 |
Days | +0.031 |
Thank | +0.031 |
fuller | +0.030 |
Neg logit contributions | |
Jamie | -0.056 |
icky | -0.055 |
jo | -0.054 |
anne | -0.054 |
ash | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | +4.00e-01 |
Pos logit contributions | |
stronger | +0.089 |
helpful | +0.082 |
generous | +0.082 |
nicer | +0.078 |
cowboy | +0.078 |
Neg logit contributions | |
anne | -0.113 |
ash | -0.109 |
loe | -0.109 |
jo | -0.109 |
icky | -0.108 |
Token |
|
Token ID | 198 |
Feature activation | +3.72e+00 |
Pos logit contributions | |
stronger | +0.014 |
Thank | +0.014 |
Can | +0.013 |
generous | +0.013 |
vet | +0.013 |
Neg logit contributions | |
anne | -0.015 |
jo | -0.015 |
ash | -0.015 |
icky | -0.015 |
ared | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +4.09e+00 |
Pos logit contributions | |
stronger | +0.124 |
generous | +0.118 |
Thank | +0.117 |
helpful | +0.116 |
cowboy | +0.115 |
Neg logit contributions | |
jo | -0.146 |
anne | -0.145 |
Jamie | -0.144 |
icky | -0.142 |
ash | -0.141 |
Token | " |
Token ID | 1 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
stronger | +0.081 |
cowboy | +0.079 |
generous | +0.072 |
helpful | +0.071 |
fuller | +0.070 |
Neg logit contributions | |
jo | -0.219 |
icky | -0.216 |
Jamie | -0.216 |
anne | -0.215 |
ash | -0.206 |
Token | Thank |
Token ID | 10449 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
stronger | +0.275 |
cowboy | +0.272 |
brand | +0.261 |
fuller | +0.250 |
generous | +0.245 |
Neg logit contributions | |
Jamie | -0.460 |
anne | -0.452 |
jo | -0.442 |
icky | -0.442 |
Megan | -0.426 |
Token | you |
Token ID | 345 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
stronger | +0.056 |
generous | +0.055 |
helpful | +0.054 |
Thank | +0.052 |
se | +0.049 |
Neg logit contributions | |
jo | -0.103 |
anne | -0.102 |
Jamie | -0.097 |
icky | -0.097 |
ash | -0.097 |
Token | , |
Token ID | 11 |
Feature activation | +2.53e+00 |
Pos logit contributions | |
jo | +0.062 |
anne | +0.061 |
ash | +0.060 |
icky | +0.060 |
umble | +0.058 |
Neg logit contributions | |
stronger | -0.056 |
generous | -0.053 |
helpful | -0.052 |
cowboy | -0.050 |
nicer | -0.049 |
Token | Sam |
Token ID | 3409 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
stronger | +0.108 |
cowboy | +0.104 |
Thank | +0.103 |
generous | +0.103 |
Days | +0.102 |
Neg logit contributions | |
Jamie | -0.076 |
jo | -0.074 |
icky | -0.073 |
anne | -0.072 |
ash | -0.067 |
Token | ," |
Token ID | 553 |
Feature activation | +3.48e+00 |
Pos logit contributions | |
icky | +0.039 |
Jamie | +0.038 |
jo | +0.038 |
ash | +0.037 |
anne | +0.037 |
Neg logit contributions | |
stronger | -0.041 |
generous | -0.039 |
helpful | -0.039 |
cowboy | -0.038 |
Thank | -0.037 |
Token | Ben |
Token ID | 3932 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
stronger | +0.107 |
cowboy | +0.106 |
generous | +0.105 |
Thank | +0.103 |
helpful | +0.103 |
Neg logit contributions | |
jo | -0.121 |
icky | -0.121 |
anne | -0.118 |
ash | -0.118 |
Jamie | -0.115 |
Token | said |
Token ID | 531 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
stronger | +0.046 |
cowboy | +0.044 |
generous | +0.044 |
helpful | +0.042 |
vet | +0.042 |
Neg logit contributions | |
icky | -0.051 |
jo | -0.051 |
anne | -0.050 |
ash | -0.050 |
Jamie | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | +8.66e-01 |
Pos logit contributions | |
stronger | +0.104 |
cowboy | +0.103 |
generous | +0.100 |
Thank | +0.100 |
helpful | +0.099 |
Neg logit contributions | |
jo | -0.055 |
Jamie | -0.054 |
icky | -0.054 |
anne | -0.054 |
ash | -0.049 |
Token |
|
Token ID | 198 |
Feature activation | +4.16e+00 |
Pos logit contributions | |
stronger | +0.030 |
cowboy | +0.029 |
generous | +0.028 |
Thank | +0.028 |
helpful | +0.028 |
Neg logit contributions | |
jo | -0.034 |
anne | -0.033 |
icky | -0.033 |
Jamie | -0.033 |
ash | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | +4.34e+00 |
Pos logit contributions | |
stronger | +0.144 |
cowboy | +0.137 |
generous | +0.136 |
Thank | +0.134 |
helpful | +0.134 |
Neg logit contributions | |
jo | -0.159 |
anne | -0.157 |
Jamie | -0.157 |
icky | -0.155 |
ash | -0.152 |
Token | " |
Token ID | 1 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
stronger | +0.100 |
cowboy | +0.099 |
fuller | +0.091 |
generous | +0.091 |
helpful | +0.091 |
Neg logit contributions | |
icky | -0.217 |
jo | -0.217 |
Jamie | -0.215 |
anne | -0.214 |
obby | -0.201 |
Token | Yes |
Token ID | 5297 |
Feature activation | -3.91e+00 |
Pos logit contributions | |
stronger | +0.250 |
brand | +0.244 |
cowboy | +0.242 |
fuller | +0.233 |
nicer | +0.227 |
Neg logit contributions | |
Jamie | -0.480 |
icky | -0.478 |
jo | -0.472 |
anne | -0.469 |
umble | -0.458 |
Token | , |
Token ID | 11 |
Feature activation | +2.92e+00 |
Pos logit contributions | |
jo | +0.168 |
ash | +0.164 |
anne | +0.164 |
icky | +0.162 |
umble | +0.157 |
Neg logit contributions | |
stronger | -0.126 |
cowboy | -0.125 |
generous | -0.118 |
helpful | -0.117 |
captain | -0.114 |
Token | thank |
Token ID | 5875 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
stronger | +0.108 |
Thank | +0.104 |
please | +0.103 |
Days | +0.103 |
fuller | +0.102 |
Neg logit contributions | |
Jamie | -0.105 |
jo | -0.099 |
icky | -0.099 |
anne | -0.098 |
ash | -0.091 |
Token | you |
Token ID | 345 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
stronger | +0.059 |
cowboy | +0.057 |
generous | +0.056 |
helpful | +0.056 |
Thank | +0.055 |
Neg logit contributions | |
jo | -0.071 |
anne | -0.070 |
icky | -0.070 |
Jamie | -0.069 |
ash | -0.067 |
Token | ," |
Token ID | 553 |
Feature activation | +3.26e+00 |
Pos logit contributions | |
jo | +0.061 |
anne | +0.061 |
icky | +0.060 |
Jamie | +0.059 |
ash | +0.057 |
Neg logit contributions | |
stronger | -0.067 |
generous | -0.064 |
cowboy | -0.063 |
helpful | -0.062 |
Thank | -0.060 |
Token | Lily |
Token ID | 20037 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
generous | +0.103 |
cowboy | +0.103 |
stronger | +0.102 |
Thank | +0.101 |
helpful | +0.100 |
Neg logit contributions | |
jo | -0.104 |
icky | -0.104 |
ash | -0.103 |
anne | -0.101 |
umble | -0.096 |
Token | says |
Token ID | 1139 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
stronger | +0.012 |
generous | +0.012 |
cowboy | +0.012 |
helpful | +0.012 |
kind | +0.011 |
Neg logit contributions | |
icky | -0.018 |
jo | -0.017 |
ash | -0.017 |
obby | -0.017 |
anne | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +6.88e-01 |
Pos logit contributions | |
stronger | +0.080 |
cowboy | +0.078 |
generous | +0.077 |
helpful | +0.076 |
Thank | +0.075 |
Neg logit contributions | |
jo | -0.043 |
icky | -0.042 |
Jamie | -0.042 |
anne | -0.042 |
ash | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | +4.02e+00 |
Pos logit contributions | |
stronger | +0.024 |
cowboy | +0.023 |
generous | +0.023 |
helpful | +0.023 |
Thank | +0.022 |
Neg logit contributions | |
jo | -0.026 |
Jamie | -0.026 |
icky | -0.026 |
anne | -0.026 |
ash | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | +4.25e+00 |
Pos logit contributions | |
stronger | +0.135 |
generous | +0.127 |
cowboy | +0.126 |
Thank | +0.126 |
helpful | +0.125 |
Neg logit contributions | |
jo | -0.157 |
anne | -0.156 |
Jamie | -0.155 |
icky | -0.153 |
ash | -0.151 |
Token | " |
Token ID | 1 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
stronger | +0.102 |
cowboy | +0.102 |
fuller | +0.095 |
generous | +0.092 |
helpful | +0.090 |
Neg logit contributions | |
jo | -0.210 |
icky | -0.209 |
Jamie | -0.207 |
anne | -0.206 |
umble | -0.196 |
Token | You |
Token ID | 1639 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
stronger | +0.295 |
brand | +0.292 |
fuller | +0.290 |
cowboy | +0.288 |
helpful | +0.269 |
Neg logit contributions | |
anne | -0.402 |
jo | -0.396 |
Jamie | -0.394 |
icky | -0.391 |
Today | -0.388 |
Token | 're |
Token ID | 821 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
Jamie | +0.064 |
icky | +0.064 |
jo | +0.063 |
anne | +0.057 |
Today | +0.056 |
Neg logit contributions | |
stronger | -0.131 |
Days | -0.130 |
please | -0.126 |
cowboy | -0.125 |
generous | -0.125 |
Token | welcome |
Token ID | 7062 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
Jamie | +0.008 |
icky | +0.008 |
jo | +0.008 |
anne | +0.008 |
ash | +0.008 |
Neg logit contributions | |
stronger | -0.006 |
cowboy | -0.005 |
generous | -0.005 |
Thank | -0.005 |
helpful | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
stronger | +0.072 |
generous | +0.070 |
cowboy | +0.070 |
helpful | +0.069 |
Thank | +0.067 |
Neg logit contributions | |
jo | -0.063 |
icky | -0.061 |
anne | -0.061 |
Jamie | -0.060 |
ash | -0.058 |
Token | kids |
Token ID | 3988 |
Feature activation | -8.49e-01 |
Pos logit contributions | |
stronger | +0.073 |
Days | +0.073 |
please | +0.072 |
Mail | +0.072 |
fuller | +0.072 |
Neg logit contributions | |
Jamie | -0.048 |
jo | -0.044 |
icky | -0.044 |
anne | -0.042 |
Jo | -0.042 |
Token | Ben |
Token ID | 3932 |
Feature activation | -9.30e-01 |
Pos logit contributions | |
anne | +0.219 |
jo | +0.216 |
umble | +0.201 |
Jamie | +0.193 |
Today | +0.189 |
Neg logit contributions | |
se | -0.174 |
cowboy | -0.174 |
stronger | -0.173 |
store | -0.171 |
vet | -0.167 |
Token | says |
Token ID | 1139 |
Feature activation | +7.26e-01 |
Pos logit contributions | |
icky | +0.037 |
jo | +0.037 |
anne | +0.037 |
Jamie | +0.036 |
obby | +0.035 |
Neg logit contributions | |
cowboy | -0.030 |
stronger | -0.030 |
generous | -0.028 |
se | -0.028 |
vet | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | +7.59e-01 |
Pos logit contributions | |
stronger | +0.027 |
cowboy | +0.026 |
generous | +0.026 |
helpful | +0.026 |
Thank | +0.025 |
Neg logit contributions | |
icky | -0.026 |
jo | -0.026 |
anne | -0.025 |
ash | -0.024 |
umble | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | +3.92e+00 |
Pos logit contributions | |
stronger | +0.025 |
Thank | +0.024 |
generous | +0.024 |
Can | +0.024 |
cowboy | +0.024 |
Neg logit contributions | |
anne | -0.029 |
jo | -0.029 |
ash | -0.029 |
icky | -0.029 |
ared | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +4.19e+00 |
Pos logit contributions | |
stronger | +0.114 |
vet | +0.106 |
generous | +0.105 |
Thank | +0.105 |
helpful | +0.103 |
Neg logit contributions | |
anne | -0.169 |
jo | -0.166 |
ash | -0.163 |
Jamie | -0.163 |
ared | -0.161 |
Token | " |
Token ID | 1 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
stronger | +0.088 |
cowboy | +0.085 |
generous | +0.079 |
helpful | +0.078 |
fuller | +0.076 |
Neg logit contributions | |
jo | -0.219 |
icky | -0.218 |
Jamie | -0.218 |
anne | -0.216 |
ash | -0.207 |
Token | Qu |
Token ID | 4507 |
Feature activation | +3.85e+00 |
Pos logit contributions | |
stronger | +0.291 |
brand | +0.285 |
helpful | +0.272 |
cowboy | +0.270 |
fuller | +0.270 |
Neg logit contributions | |
Jamie | -0.434 |
jo | -0.425 |
anne | -0.423 |
icky | -0.421 |
Today | -0.411 |
Token | ack |
Token ID | 441 |
Feature activation | +7.48e-02 |
Pos logit contributions | |
stronger | +0.147 |
cowboy | +0.142 |
generous | +0.141 |
helpful | +0.137 |
Thank | +0.137 |
Neg logit contributions | |
jo | -0.136 |
anne | -0.135 |
Jamie | -0.135 |
icky | -0.122 |
Jo | -0.122 |
Token | , |
Token ID | 11 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
stronger | +0.003 |
generous | +0.003 |
cowboy | +0.003 |
helpful | +0.002 |
se | +0.002 |
Neg logit contributions | |
jo | -0.003 |
anne | -0.003 |
Jamie | -0.003 |
icky | -0.003 |
obby | -0.003 |
Token | qu |
Token ID | 627 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
stronger | +0.065 |
Days | +0.062 |
cowboy | +0.061 |
generous | +0.061 |
Thank | +0.060 |
Neg logit contributions | |
Jamie | -0.089 |
icky | -0.088 |
jo | -0.087 |
anne | -0.086 |
ash | -0.084 |
Token | ack |
Token ID | 441 |
Feature activation | -8.15e-01 |
Pos logit contributions | |
stronger | +0.055 |
generous | +0.052 |
cowboy | +0.052 |
helpful | +0.051 |
Thank | +0.051 |
Neg logit contributions | |
Jamie | -0.049 |
jo | -0.049 |
anne | -0.048 |
icky | -0.045 |
obby | -0.043 |
Token | Ben |
Token ID | 3932 |
Feature activation | -9.30e-01 |
Pos logit contributions | |
jo | +0.037 |
icky | +0.037 |
anne | +0.036 |
Jamie | +0.036 |
ash | +0.034 |
Neg logit contributions | |
stronger | -0.057 |
cowboy | -0.055 |
generous | -0.055 |
helpful | -0.054 |
Thank | -0.053 |
Token | says |
Token ID | 1139 |
Feature activation | +7.26e-01 |
Pos logit contributions | |
icky | +0.037 |
jo | +0.037 |
anne | +0.037 |
Jamie | +0.036 |
obby | +0.035 |
Neg logit contributions | |
cowboy | -0.030 |
stronger | -0.030 |
generous | -0.028 |
se | -0.028 |
vet | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | +7.59e-01 |
Pos logit contributions | |
stronger | +0.027 |
cowboy | +0.026 |
generous | +0.026 |
helpful | +0.026 |
Thank | +0.025 |
Neg logit contributions | |
icky | -0.026 |
jo | -0.026 |
anne | -0.025 |
ash | -0.024 |
umble | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | +3.92e+00 |
Pos logit contributions | |
stronger | +0.025 |
Thank | +0.024 |
generous | +0.024 |
Can | +0.024 |
cowboy | +0.024 |
Neg logit contributions | |
anne | -0.029 |
jo | -0.029 |
ash | -0.029 |
icky | -0.029 |
ared | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +4.19e+00 |
Pos logit contributions | |
stronger | +0.114 |
vet | +0.106 |
generous | +0.105 |
Thank | +0.105 |
helpful | +0.103 |
Neg logit contributions | |
anne | -0.169 |
jo | -0.166 |
ash | -0.163 |
Jamie | -0.163 |
ared | -0.161 |
Token | " |
Token ID | 1 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
stronger | +0.088 |
cowboy | +0.085 |
generous | +0.079 |
helpful | +0.078 |
fuller | +0.076 |
Neg logit contributions | |
jo | -0.219 |
icky | -0.218 |
Jamie | -0.218 |
anne | -0.216 |
ash | -0.207 |
Token | That |
Token ID | 2504 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
stronger | +0.291 |
brand | +0.285 |
helpful | +0.272 |
cowboy | +0.270 |
fuller | +0.270 |
Neg logit contributions | |
Jamie | -0.434 |
jo | -0.425 |
anne | -0.423 |
icky | -0.421 |
Today | -0.411 |
Token | was |
Token ID | 373 |
Feature activation | -9.93e-01 |
Pos logit contributions | |
anne | +0.041 |
jo | +0.041 |
Jamie | +0.040 |
icky | +0.039 |
ash | +0.038 |
Neg logit contributions | |
stronger | -0.044 |
cowboy | -0.042 |
generous | -0.042 |
helpful | -0.042 |
vet | -0.041 |
Token | cool |
Token ID | 3608 |
Feature activation | +8.96e-01 |
Pos logit contributions | |
jo | +0.047 |
Jamie | +0.047 |
icky | +0.047 |
anne | +0.046 |
ash | +0.045 |
Neg logit contributions | |
stronger | -0.025 |
cowboy | -0.024 |
generous | -0.023 |
Thank | -0.023 |
helpful | -0.023 |
Token | ! |
Token ID | 0 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
stronger | +0.029 |
cowboy | +0.028 |
helpful | +0.028 |
generous | +0.028 |
se | +0.027 |
Neg logit contributions | |
anne | -0.036 |
jo | -0.036 |
Jamie | -0.035 |
icky | -0.035 |
loe | -0.034 |
Token | But |
Token ID | 887 |
Feature activation | -4.47e+00 |
Pos logit contributions | |
anne | +0.070 |
jo | +0.070 |
icky | +0.068 |
Jamie | +0.066 |
ash | +0.065 |
Neg logit contributions | |
stronger | -0.063 |
Thank | -0.063 |
cowboy | -0.062 |
Can | -0.062 |
generous | -0.062 |
Token | hugged |
Token ID | 41130 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
Jamie | +0.041 |
jo | +0.040 |
icky | +0.040 |
anne | +0.039 |
ash | +0.037 |
Neg logit contributions | |
stronger | -0.031 |
cowboy | -0.029 |
generous | -0.028 |
Days | -0.028 |
Thank | -0.028 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
stronger | +0.041 |
cowboy | +0.040 |
generous | +0.038 |
vet | +0.038 |
Thank | +0.037 |
Neg logit contributions | |
icky | -0.067 |
anne | -0.062 |
ash | -0.062 |
jo | -0.061 |
ared | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
Jamie | +0.059 |
icky | +0.059 |
anne | +0.059 |
jo | +0.058 |
ash | +0.054 |
Neg logit contributions | |
stronger | -0.045 |
generous | -0.042 |
helpful | -0.040 |
cowboy | -0.040 |
Thank | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | +3.43e+00 |
Pos logit contributions | |
jo | +0.007 |
anne | +0.007 |
icky | +0.007 |
Jamie | +0.007 |
ash | +0.007 |
Neg logit contributions | |
stronger | -0.008 |
cowboy | -0.008 |
Thank | -0.008 |
generous | -0.008 |
helpful | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +3.86e+00 |
Pos logit contributions | |
stronger | +0.112 |
generous | +0.107 |
Thank | +0.106 |
cowboy | +0.105 |
helpful | +0.105 |
Neg logit contributions | |
Jamie | -0.135 |
jo | -0.135 |
anne | -0.134 |
icky | -0.133 |
ash | -0.130 |
Token | " |
Token ID | 1 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
stronger | +0.094 |
cowboy | +0.093 |
fuller | +0.085 |
generous | +0.081 |
nicer | +0.081 |
Neg logit contributions | |
jo | -0.192 |
anne | -0.189 |
icky | -0.189 |
Jamie | -0.186 |
umble | -0.178 |
Token | Thank |
Token ID | 10449 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
stronger | +0.300 |
cowboy | +0.282 |
smarter | +0.272 |
fuller | +0.270 |
brand | +0.270 |
Neg logit contributions | |
anne | -0.418 |
Jamie | -0.406 |
jo | -0.395 |
Today | -0.389 |
Megan | -0.385 |
Token | you |
Token ID | 345 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
stronger | +0.064 |
generous | +0.062 |
helpful | +0.061 |
Thank | +0.061 |
cowboy | +0.059 |
Neg logit contributions | |
jo | -0.091 |
anne | -0.089 |
Jamie | -0.088 |
icky | -0.088 |
ash | -0.084 |
Token | , |
Token ID | 11 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
jo | +0.068 |
anne | +0.068 |
icky | +0.066 |
Jamie | +0.065 |
ash | +0.064 |
Neg logit contributions | |
stronger | -0.069 |
generous | -0.066 |
helpful | -0.065 |
cowboy | -0.064 |
nicer | -0.062 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
please | +0.109 |
stronger | +0.108 |
Days | +0.108 |
fuller | +0.106 |
Mail | +0.105 |
Neg logit contributions | |
Jamie | -0.062 |
icky | -0.056 |
jo | -0.056 |
anne | -0.054 |
Jo | -0.053 |
Token | ! |
Token ID | 0 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
icky | +0.037 |
Jamie | +0.037 |
jo | +0.036 |
anne | +0.036 |
ash | +0.035 |
Neg logit contributions | |
stronger | -0.046 |
generous | -0.044 |
helpful | -0.044 |
cowboy | -0.043 |
Thank | -0.043 |
Token | feel |
Token ID | 1254 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
stronger | +0.128 |
cowboy | +0.124 |
generous | +0.121 |
helpful | +0.119 |
Thank | +0.118 |
Neg logit contributions | |
jo | -0.177 |
icky | -0.175 |
Jamie | -0.175 |
anne | -0.174 |
ash | -0.167 |
Token | sorry |
Token ID | 7926 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
cowboy | +0.029 |
Offic | +0.028 |
Days | +0.027 |
please | +0.027 |
Mail | +0.026 |
Neg logit contributions | |
Jamie | -0.044 |
icky | -0.044 |
anne | -0.042 |
jo | -0.042 |
umble | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +4.37e-01 |
Pos logit contributions | |
stronger | +0.079 |
generous | +0.071 |
helpful | +0.070 |
nicer | +0.067 |
kind | +0.065 |
Neg logit contributions | |
anne | -0.123 |
Jamie | -0.119 |
loe | -0.118 |
ared | -0.115 |
jo | -0.115 |
Token |
|
Token ID | 198 |
Feature activation | +3.76e+00 |
Pos logit contributions | |
Can | +0.015 |
Thank | +0.015 |
vet | +0.015 |
stronger | +0.014 |
generous | +0.014 |
Neg logit contributions | |
ash | -0.017 |
jo | -0.017 |
anne | -0.017 |
icky | -0.016 |
ared | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | +4.13e+00 |
Pos logit contributions | |
stronger | +0.124 |
generous | +0.118 |
Thank | +0.118 |
helpful | +0.117 |
cowboy | +0.115 |
Neg logit contributions | |
jo | -0.148 |
anne | -0.147 |
Jamie | -0.147 |
icky | -0.144 |
ash | -0.143 |
Token | " |
Token ID | 1 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
cowboy | +0.093 |
stronger | +0.092 |
fuller | +0.081 |
generous | +0.080 |
helpful | +0.078 |
Neg logit contributions | |
jo | -0.211 |
icky | -0.210 |
Jamie | -0.209 |
anne | -0.209 |
umble | -0.197 |
Token | Sorry |
Token ID | 14385 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
cowboy | +0.285 |
stronger | +0.269 |
brand | +0.260 |
saddle | +0.243 |
fuller | +0.241 |
Neg logit contributions | |
anne | -0.455 |
Jamie | -0.449 |
icky | -0.444 |
Megan | -0.429 |
jo | -0.426 |
Token | , |
Token ID | 11 |
Feature activation | +3.10e+00 |
Pos logit contributions | |
stronger | +0.014 |
helpful | +0.014 |
generous | +0.014 |
kind | +0.013 |
nicer | +0.012 |
Neg logit contributions | |
jo | -0.022 |
anne | -0.022 |
ash | -0.021 |
icky | -0.021 |
umble | -0.020 |
Token | Rose |
Token ID | 8049 |
Feature activation | -2.91e+00 |
Pos logit contributions | |
stronger | +0.127 |
cowboy | +0.123 |
Thank | +0.121 |
generous | +0.120 |
Days | +0.120 |
Neg logit contributions | |
Jamie | -0.098 |
jo | -0.097 |
icky | -0.096 |
anne | -0.095 |
ash | -0.088 |
Token | ," |
Token ID | 553 |
Feature activation | +3.63e+00 |
Pos logit contributions | |
ash | +0.101 |
Jamie | +0.101 |
icky | +0.101 |
jo | +0.100 |
anne | +0.096 |
Neg logit contributions | |
stronger | -0.112 |
generous | -0.108 |
helpful | -0.108 |
cowboy | -0.106 |
vet | -0.102 |
Token | Ben |
Token ID | 3932 |
Feature activation | +9.28e-01 |
Pos logit contributions | |
stronger | +0.156 |
cowboy | +0.152 |
generous | +0.150 |
helpful | +0.148 |
Thank | +0.147 |
Neg logit contributions | |
jo | -0.111 |
icky | -0.110 |
anne | -0.109 |
Jamie | -0.106 |
ash | -0.105 |
Token | Tom |
Token ID | 4186 |
Feature activation | -3.94e-01 |
Pos logit contributions | |
cowboy | +0.040 |
stronger | +0.036 |
generous | +0.034 |
helpful | +0.032 |
captain | +0.031 |
Neg logit contributions | |
jo | -0.157 |
anne | -0.156 |
icky | -0.155 |
ash | -0.154 |
Jamie | -0.153 |
Token | says |
Token ID | 1139 |
Feature activation | +8.24e-01 |
Pos logit contributions | |
icky | +0.017 |
jo | +0.016 |
anne | +0.016 |
Jamie | +0.016 |
ash | +0.015 |
Neg logit contributions | |
stronger | -0.012 |
cowboy | -0.012 |
Thank | -0.011 |
generous | -0.011 |
se | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +7.64e-01 |
Pos logit contributions | |
stronger | +0.031 |
cowboy | +0.030 |
generous | +0.030 |
helpful | +0.029 |
Thank | +0.028 |
Neg logit contributions | |
jo | -0.030 |
icky | -0.029 |
anne | -0.029 |
ash | -0.028 |
umble | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | +3.94e+00 |
Pos logit contributions | |
stronger | +0.024 |
Thank | +0.024 |
Can | +0.023 |
vet | +0.023 |
generous | +0.023 |
Neg logit contributions | |
jo | -0.031 |
ash | -0.031 |
anne | -0.031 |
icky | -0.030 |
umble | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | +4.17e+00 |
Pos logit contributions | |
stronger | +0.110 |
vet | +0.102 |
Thank | +0.101 |
generous | +0.100 |
helpful | +0.099 |
Neg logit contributions | |
anne | -0.173 |
jo | -0.170 |
ash | -0.170 |
Jamie | -0.167 |
ared | -0.167 |
Token | " |
Token ID | 1 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
stronger | +0.083 |
cowboy | +0.080 |
generous | +0.076 |
helpful | +0.074 |
Thank | +0.072 |
Neg logit contributions | |
jo | -0.222 |
Jamie | -0.220 |
icky | -0.220 |
anne | -0.219 |
ash | -0.210 |
Token | OK |
Token ID | 11380 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
stronger | +0.292 |
brand | +0.289 |
helpful | +0.274 |
fuller | +0.273 |
cowboy | +0.272 |
Neg logit contributions | |
Jamie | -0.432 |
jo | -0.430 |
anne | -0.427 |
icky | -0.418 |
Today | -0.408 |
Token | , |
Token ID | 11 |
Feature activation | +3.42e+00 |
Pos logit contributions | |
jo | +0.079 |
icky | +0.079 |
anne | +0.077 |
ash | +0.075 |
umble | +0.074 |
Neg logit contributions | |
stronger | -0.078 |
generous | -0.075 |
cowboy | -0.074 |
helpful | -0.074 |
Thank | -0.071 |
Token | Tom |
Token ID | 4186 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
stronger | +0.154 |
cowboy | +0.149 |
Days | +0.149 |
Thank | +0.148 |
generous | +0.148 |
Neg logit contributions | |
Jamie | -0.095 |
jo | -0.088 |
anne | -0.087 |
icky | -0.087 |
ash | -0.080 |
Token | , |
Token ID | 11 |
Feature activation | +3.34e+00 |
Pos logit contributions | |
stronger | +0.015 |
generous | +0.014 |
helpful | +0.014 |
cowboy | +0.014 |
Thank | +0.014 |
Neg logit contributions | |
icky | -0.015 |
jo | -0.015 |
Jamie | -0.014 |
anne | -0.014 |
ash | -0.014 |
Token | but |
Token ID | 475 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
stronger | +0.113 |
fuller | +0.112 |
Days | +0.112 |
Thank | +0.111 |
please | +0.109 |
Neg logit contributions | |
Jamie | -0.127 |
icky | -0.118 |
jo | -0.116 |
anne | -0.113 |
ash | -0.111 |
Token | sh |
Token ID | 427 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
stronger | +0.055 |
cowboy | +0.054 |
Thank | +0.054 |
Offic | +0.052 |
please | +0.052 |
Neg logit contributions | |
icky | -0.070 |
jo | -0.068 |
umble | -0.067 |
Jamie | -0.066 |
anne | -0.066 |
Token | iver |
Token ID | 1428 |
Feature activation | -2.57e-01 |
Pos logit contributions | |
stronger | +0.035 |
cowboy | +0.033 |
generous | +0.033 |
helpful | +0.032 |
Thank | +0.032 |
Neg logit contributions | |
jo | -0.045 |
Jamie | -0.045 |
icky | -0.045 |
anne | -0.045 |
ash | -0.043 |
Token | go |
Token ID | 467 |
Feature activation | +3.27e+00 |
Pos logit contributions | |
Jamie | +0.010 |
anne | +0.009 |
jo | +0.009 |
icky | +0.009 |
obby | +0.009 |
Neg logit contributions | |
stronger | -0.010 |
cowboy | -0.009 |
helpful | -0.009 |
generous | -0.009 |
se | -0.009 |
Token | down |
Token ID | 866 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
cowboy | +0.121 |
stronger | +0.119 |
generous | +0.116 |
Thank | +0.116 |
Days | +0.114 |
Neg logit contributions | |
jo | -0.116 |
icky | -0.111 |
anne | -0.111 |
Jamie | -0.109 |
ash | -0.107 |
Token | his |
Token ID | 465 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
jo | +0.069 |
anne | +0.069 |
Jamie | +0.067 |
icky | +0.067 |
ash | +0.064 |
Neg logit contributions | |
stronger | -0.091 |
generous | -0.084 |
helpful | -0.084 |
Thank | -0.084 |
cowboy | -0.084 |
Token | spine |
Token ID | 19656 |
Feature activation | -1.08e+01 |
Pos logit contributions | |
Jamie | +0.096 |
anne | +0.094 |
icky | +0.093 |
ash | +0.093 |
raw | +0.090 |
Neg logit contributions | |
helpful | -0.092 |
stronger | -0.092 |
generous | -0.091 |
cowboy | -0.090 |
vet | -0.087 |
Token | . |
Token ID | 13 |
Feature activation | -2.86e+00 |
Pos logit contributions | |
loe | +0.445 |
jo | +0.434 |
ash | +0.430 |
Jamie | +0.429 |
anne | +0.427 |
Neg logit contributions | |
stronger | -0.359 |
helpful | -0.334 |
generous | -0.333 |
nearby | -0.326 |
se | -0.326 |
Token | He |
Token ID | 679 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
jo | +0.108 |
anne | +0.107 |
icky | +0.106 |
ash | +0.104 |
Jamie | +0.103 |
Neg logit contributions | |
stronger | -0.099 |
generous | -0.096 |
Thank | -0.096 |
cowboy | -0.096 |
Can | -0.095 |
Token | looked |
Token ID | 3114 |
Feature activation | +5.76e-01 |
Pos logit contributions | |
stronger | +0.067 |
cowboy | +0.065 |
generous | +0.063 |
helpful | +0.062 |
Thank | +0.061 |
Neg logit contributions | |
jo | -0.108 |
icky | -0.108 |
Jamie | -0.107 |
anne | -0.107 |
ash | -0.103 |
Token | around |
Token ID | 1088 |
Feature activation | -3.98e+00 |
Pos logit contributions | |
stronger | +0.020 |
generous | +0.019 |
helpful | +0.019 |
cowboy | +0.019 |
nearby | +0.018 |
Neg logit contributions | |
icky | -0.024 |
jo | -0.023 |
anne | -0.023 |
Jamie | -0.023 |
umble | -0.022 |
Token | but |
Token ID | 475 |
Feature activation | -5.02e+00 |
Pos logit contributions | |
jo | +0.168 |
anne | +0.168 |
icky | +0.165 |
Jamie | +0.161 |
ash | +0.160 |
Neg logit contributions | |
stronger | -0.126 |
cowboy | -0.120 |
generous | -0.120 |
helpful | -0.118 |
Thank | -0.116 |
Token | sh |
Token ID | 427 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
cowboy | +0.071 |
stronger | +0.070 |
Thank | +0.067 |
vet | +0.065 |
please | +0.065 |
Neg logit contributions | |
icky | -0.097 |
Jamie | -0.094 |
jo | -0.094 |
anne | -0.093 |
umble | -0.092 |
Token | iver |
Token ID | 1428 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
stronger | +0.058 |
cowboy | +0.057 |
generous | +0.055 |
Thank | +0.054 |
helpful | +0.054 |
Neg logit contributions | |
jo | -0.069 |
icky | -0.068 |
anne | -0.068 |
Jamie | -0.067 |
ash | -0.065 |
Token | go |
Token ID | 467 |
Feature activation | +3.59e+00 |
Pos logit contributions | |
stronger | +0.010 |
cowboy | +0.010 |
generous | +0.010 |
helpful | +0.010 |
Thank | +0.009 |
Neg logit contributions | |
jo | -0.012 |
icky | -0.012 |
Jamie | -0.012 |
anne | -0.012 |
ash | -0.012 |
Token | down |
Token ID | 866 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
cowboy | +0.122 |
stronger | +0.121 |
generous | +0.117 |
Thank | +0.117 |
Days | +0.116 |
Neg logit contributions | |
jo | -0.138 |
anne | -0.135 |
icky | -0.133 |
Jamie | -0.132 |
ash | -0.129 |
Token | her |
Token ID | 607 |
Feature activation | -3.33e+00 |
Pos logit contributions | |
jo | +0.068 |
anne | +0.067 |
Jamie | +0.066 |
icky | +0.066 |
ash | +0.063 |
Neg logit contributions | |
stronger | -0.074 |
generous | -0.068 |
helpful | -0.068 |
Thank | -0.067 |
cowboy | -0.067 |
Token | spine |
Token ID | 19656 |
Feature activation | -1.06e+01 |
Pos logit contributions | |
jo | +0.143 |
icky | +0.139 |
Jamie | +0.138 |
ash | +0.138 |
anne | +0.138 |
Neg logit contributions | |
generous | -0.107 |
stronger | -0.107 |
helpful | -0.104 |
cowboy | -0.098 |
vet | -0.097 |
Token | . |
Token ID | 13 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
loe | +0.444 |
jo | +0.439 |
ash | +0.433 |
obby | +0.426 |
Jamie | +0.421 |
Neg logit contributions | |
stronger | -0.346 |
generous | -0.318 |
helpful | -0.317 |
nearby | -0.314 |
se | -0.310 |
Token | It |
Token ID | 632 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
jo | +0.099 |
anne | +0.093 |
icky | +0.092 |
ash | +0.092 |
Jamie | +0.089 |
Neg logit contributions | |
generous | -0.104 |
vet | -0.103 |
stronger | -0.103 |
Thank | -0.102 |
Can | -0.101 |
Token | tasted |
Token ID | 29187 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
stronger | +0.073 |
cowboy | +0.070 |
generous | +0.069 |
helpful | +0.068 |
Thank | +0.068 |
Neg logit contributions | |
jo | -0.079 |
icky | -0.078 |
Jamie | -0.078 |
anne | -0.077 |
ash | -0.074 |
Token | different |
Token ID | 1180 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
jo | +0.008 |
anne | +0.007 |
Jamie | +0.007 |
icky | +0.007 |
ash | +0.007 |
Neg logit contributions | |
stronger | -0.004 |
generous | -0.004 |
helpful | -0.004 |
cowboy | -0.003 |
Thank | -0.003 |
Token | than |
Token ID | 621 |
Feature activation | -4.37e+00 |
Pos logit contributions | |
jo | +0.070 |
anne | +0.069 |
loe | +0.068 |
Jamie | +0.064 |
icky | +0.064 |
Neg logit contributions | |
stronger | -0.042 |
helpful | -0.041 |
generous | -0.040 |
nicer | -0.037 |
kind | -0.036 |
Token | found |
Token ID | 1043 |
Feature activation | +3.27e+00 |
Pos logit contributions | |
jo | +0.024 |
Jamie | +0.024 |
anne | +0.024 |
icky | +0.024 |
ash | +0.023 |
Neg logit contributions | |
stronger | -0.013 |
cowboy | -0.012 |
generous | -0.012 |
helpful | -0.011 |
Thank | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
stronger | +0.103 |
cowboy | +0.102 |
generous | +0.098 |
Thank | +0.096 |
helpful | +0.096 |
Neg logit contributions | |
Jamie | -0.133 |
icky | -0.133 |
jo | -0.130 |
anne | -0.129 |
ash | -0.125 |
Token | pretty |
Token ID | 2495 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
Days | +0.066 |
Thank | +0.066 |
stronger | +0.064 |
please | +0.061 |
Did | +0.061 |
Neg logit contributions | |
icky | -0.101 |
Jamie | -0.099 |
anne | -0.097 |
bie | -0.096 |
jo | -0.096 |
Token | cup |
Token ID | 6508 |
Feature activation | +5.34e-01 |
Pos logit contributions | |
stronger | +0.056 |
Days | +0.051 |
cowboy | +0.051 |
Thank | +0.050 |
generous | +0.050 |
Neg logit contributions | |
jo | -0.111 |
Jamie | -0.110 |
anne | -0.110 |
icky | -0.110 |
ash | -0.105 |
Token | hidden |
Token ID | 7104 |
Feature activation | +4.51e-01 |
Pos logit contributions | |
stronger | +0.021 |
generous | +0.021 |
cowboy | +0.021 |
Thank | +0.020 |
helpful | +0.020 |
Neg logit contributions | |
jo | -0.018 |
Jamie | -0.017 |
anne | -0.017 |
icky | -0.017 |
obby | -0.016 |
Token | amongst |
Token ID | 12077 |
Feature activation | -1.05e+01 |
Pos logit contributions | |
stronger | +0.015 |
cowboy | +0.015 |
generous | +0.015 |
helpful | +0.014 |
Thank | +0.014 |
Neg logit contributions | |
jo | -0.018 |
anne | -0.018 |
Jamie | -0.018 |
icky | -0.018 |
ash | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | -3.68e+00 |
Pos logit contributions | |
jo | +0.441 |
anne | +0.428 |
icky | +0.416 |
ash | +0.411 |
Jamie | +0.406 |
Neg logit contributions | |
stronger | -0.355 |
generous | -0.332 |
Thank | -0.330 |
helpful | -0.328 |
cowboy | -0.323 |
Token | trees |
Token ID | 7150 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
jo | +0.177 |
anne | +0.173 |
Jamie | +0.173 |
icky | +0.172 |
ash | +0.165 |
Neg logit contributions | |
stronger | -0.096 |
cowboy | -0.090 |
generous | -0.090 |
Thank | -0.089 |
helpful | -0.089 |
Token | . |
Token ID | 13 |
Feature activation | -2.91e+00 |
Pos logit contributions | |
jo | +0.088 |
Jamie | +0.086 |
anne | +0.086 |
icky | +0.085 |
ash | +0.079 |
Neg logit contributions | |
stronger | -0.151 |
cowboy | -0.145 |
generous | -0.145 |
helpful | -0.143 |
Thank | -0.142 |
Token | She |
Token ID | 1375 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
jo | +0.107 |
icky | +0.103 |
anne | +0.102 |
Jamie | +0.102 |
ash | +0.098 |
Neg logit contributions | |
stronger | -0.107 |
Thank | -0.106 |
generous | -0.104 |
helpful | -0.103 |
cowboy | -0.102 |
Token | was |
Token ID | 373 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
stronger | +0.055 |
cowboy | +0.053 |
fuller | +0.051 |
Thank | +0.050 |
nicer | +0.049 |
Neg logit contributions | |
Jamie | -0.059 |
anne | -0.055 |
jo | -0.055 |
icky | -0.054 |
bie | -0.053 |
Token | any |
Token ID | 597 |
Feature activation | -3.57e+00 |
Pos logit contributions | |
jo | +0.294 |
anne | +0.286 |
icky | +0.283 |
Jamie | +0.274 |
ash | +0.272 |
Neg logit contributions | |
stronger | -0.241 |
generous | -0.227 |
helpful | -0.224 |
cowboy | -0.224 |
Thank | -0.218 |
Token | conflict |
Token ID | 5358 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
icky | +0.130 |
jo | +0.128 |
Jamie | +0.128 |
anne | +0.127 |
ash | +0.122 |
Neg logit contributions | |
cowboy | -0.123 |
stronger | -0.120 |
Thank | -0.118 |
Days | -0.117 |
Mail | -0.115 |
Token | . |
Token ID | 13 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
jo | +0.065 |
anne | +0.063 |
icky | +0.062 |
Jamie | +0.060 |
umble | +0.059 |
Neg logit contributions | |
stronger | -0.079 |
generous | -0.074 |
helpful | -0.074 |
cowboy | -0.073 |
golf | -0.071 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
ash | +0.119 |
jo | +0.118 |
elle | +0.115 |
ared | +0.115 |
ol | +0.114 |
Neg logit contributions | |
Thank | -0.042 |
And | -0.039 |
Can | -0.035 |
Would | -0.034 |
Did | -0.034 |
Token | O |
Token ID | 46 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
anne | +0.076 |
jo | +0.075 |
umble | +0.073 |
ared | +0.073 |
ash | +0.073 |
Neg logit contributions | |
Can | -0.045 |
Thank | -0.045 |
nearby | -0.044 |
stronger | -0.043 |
There | -0.043 |
Token | liv |
Token ID | 16017 |
Feature activation | -1.05e+01 |
Pos logit contributions | |
Jamie | +0.090 |
raw | +0.088 |
fists | +0.088 |
Today | +0.086 |
ol | +0.085 |
Neg logit contributions | |
Days | -0.097 |
nearby | -0.094 |
Thank | -0.093 |
please | -0.091 |
generous | -0.090 |
Token | ia |
Token ID | 544 |
Feature activation | -4.89e+00 |
Pos logit contributions | |
jo | +0.301 |
ash | +0.299 |
Today | +0.297 |
umble | +0.294 |
obby | +0.273 |
Neg logit contributions | |
Days | -0.403 |
vet | -0.361 |
set | -0.349 |
kind | -0.348 |
se | -0.348 |
Token | was |
Token ID | 373 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
icky | +0.217 |
jo | +0.211 |
ash | +0.207 |
anne | +0.207 |
Jamie | +0.207 |
Neg logit contributions | |
stronger | -0.132 |
Thank | -0.131 |
generous | -0.131 |
cowboy | -0.130 |
helpful | -0.128 |
Token | feeling |
Token ID | 4203 |
Feature activation | -3.13e+00 |
Pos logit contributions | |
jo | +0.107 |
Jamie | +0.106 |
icky | +0.106 |
anne | +0.105 |
ash | +0.101 |
Neg logit contributions | |
stronger | -0.068 |
cowboy | -0.066 |
generous | -0.063 |
Thank | -0.062 |
helpful | -0.062 |
Token | very |
Token ID | 845 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
icky | +0.129 |
Jamie | +0.129 |
jo | +0.124 |
anne | +0.124 |
umble | +0.120 |
Neg logit contributions | |
cowboy | -0.094 |
Offic | -0.088 |
Days | -0.088 |
stronger | -0.087 |
Tweet | -0.086 |
Token | restless |
Token ID | 42537 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
jo | +0.116 |
anne | +0.113 |
icky | +0.112 |
Jamie | +0.112 |
ash | +0.109 |
Neg logit contributions | |
stronger | -0.048 |
generous | -0.047 |
helpful | -0.045 |
cowboy | -0.043 |
Thank | -0.042 |
Token | aroo |
Token ID | 38049 |
Feature activation | -5.15e-01 |
Pos logit contributions | |
jo | +0.098 |
anne | +0.095 |
icky | +0.093 |
Jamie | +0.093 |
obby | +0.090 |
Neg logit contributions | |
stronger | -0.056 |
nearby | -0.053 |
cowboy | -0.053 |
generous | -0.052 |
helpful | -0.052 |
Token | , |
Token ID | 11 |
Feature activation | -4.42e-01 |
Pos logit contributions | |
jo | +0.017 |
anne | +0.016 |
Jamie | +0.016 |
icky | +0.016 |
ash | +0.015 |
Neg logit contributions | |
stronger | -0.021 |
cowboy | -0.020 |
generous | -0.020 |
helpful | -0.020 |
Thank | -0.020 |
Token | standing |
Token ID | 5055 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
jo | +0.019 |
Jamie | +0.019 |
icky | +0.018 |
anne | +0.018 |
ash | +0.018 |
Neg logit contributions | |
stronger | -0.013 |
cowboy | -0.013 |
generous | -0.012 |
helpful | -0.012 |
Thank | -0.012 |
Token | very |
Token ID | 845 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
cowboy | +0.035 |
Days | +0.032 |
stronger | +0.032 |
vet | +0.031 |
generous | +0.031 |
Neg logit contributions | |
Jamie | -0.057 |
icky | -0.055 |
anne | -0.053 |
jo | -0.053 |
ash | -0.053 |
Token | still |
Token ID | 991 |
Feature activation | -3.35e+00 |
Pos logit contributions | |
icky | +0.077 |
Jamie | +0.076 |
jo | +0.075 |
anne | +0.075 |
umble | +0.072 |
Neg logit contributions | |
cowboy | -0.036 |
stronger | -0.034 |
generous | -0.032 |
Thank | -0.032 |
helpful | -0.031 |
Token | amongst |
Token ID | 12077 |
Feature activation | -1.04e+01 |
Pos logit contributions | |
jo | +0.119 |
icky | +0.116 |
anne | +0.116 |
Jamie | +0.114 |
ash | +0.110 |
Neg logit contributions | |
stronger | -0.130 |
generous | -0.122 |
cowboy | -0.122 |
helpful | -0.121 |
Thank | -0.118 |
Token | the |
Token ID | 262 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
jo | +0.470 |
anne | +0.466 |
ash | +0.451 |
icky | +0.442 |
umble | +0.438 |
Neg logit contributions | |
stronger | -0.316 |
generous | -0.291 |
helpful | -0.287 |
nearby | -0.279 |
Thank | -0.279 |
Token | grass |
Token ID | 8701 |
Feature activation | -3.35e+00 |
Pos logit contributions | |
jo | +0.172 |
Jamie | +0.170 |
anne | +0.169 |
icky | +0.169 |
ash | +0.162 |
Neg logit contributions | |
stronger | -0.090 |
generous | -0.085 |
cowboy | -0.084 |
helpful | -0.083 |
Thank | -0.082 |
Token | . |
Token ID | 13 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
jo | +0.104 |
anne | +0.101 |
Jamie | +0.098 |
icky | +0.095 |
ash | +0.093 |
Neg logit contributions | |
stronger | -0.143 |
generous | -0.134 |
nearby | -0.133 |
helpful | -0.132 |
cowboy | -0.132 |
Token | Sam |
Token ID | 3409 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
jo | +0.102 |
anne | +0.100 |
icky | +0.099 |
Jamie | +0.094 |
ash | +0.094 |
Neg logit contributions | |
stronger | -0.112 |
Thank | -0.108 |
generous | -0.106 |
golf | -0.105 |
cowboy | -0.105 |
Token | was |
Token ID | 373 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
jo | +0.108 |
Jamie | +0.106 |
anne | +0.106 |
icky | +0.105 |
ash | +0.099 |
Neg logit contributions | |
stronger | -0.096 |
cowboy | -0.094 |
generous | -0.090 |
Thank | -0.089 |
helpful | -0.089 |
Token | which |
Token ID | 543 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
Jamie | +0.009 |
jo | +0.009 |
anne | +0.008 |
Jo | +0.008 |
Mary | +0.008 |
Neg logit contributions | |
Days | -0.012 |
cowboy | -0.011 |
fuller | -0.011 |
Mic | -0.011 |
stronger | -0.011 |
Token | made |
Token ID | 925 |
Feature activation | +8.81e-01 |
Pos logit contributions | |
jo | +0.046 |
icky | +0.046 |
anne | +0.045 |
Jamie | +0.044 |
ash | +0.044 |
Neg logit contributions | |
stronger | -0.033 |
generous | -0.032 |
helpful | -0.031 |
cowboy | -0.031 |
Thank | -0.031 |
Token | her |
Token ID | 607 |
Feature activation | -5.11e-01 |
Pos logit contributions | |
stronger | +0.030 |
generous | +0.028 |
cowboy | +0.028 |
helpful | +0.028 |
vet | +0.027 |
Neg logit contributions | |
jo | -0.035 |
anne | -0.035 |
icky | -0.034 |
Jamie | -0.034 |
ash | -0.033 |
Token | stand |
Token ID | 1302 |
Feature activation | +2.72e+00 |
Pos logit contributions | |
icky | +0.024 |
jo | +0.024 |
anne | +0.024 |
ash | +0.024 |
Jamie | +0.024 |
Neg logit contributions | |
stronger | -0.014 |
generous | -0.013 |
helpful | -0.013 |
cowboy | -0.012 |
se | -0.011 |
Token | out |
Token ID | 503 |
Feature activation | -3.31e+00 |
Pos logit contributions | |
stronger | +0.090 |
cowboy | +0.085 |
generous | +0.084 |
helpful | +0.083 |
Thank | +0.081 |
Neg logit contributions | |
jo | -0.110 |
Jamie | -0.109 |
icky | -0.108 |
anne | -0.108 |
ash | -0.103 |
Token | amongst |
Token ID | 12077 |
Feature activation | -1.04e+01 |
Pos logit contributions | |
jo | +0.108 |
anne | +0.107 |
icky | +0.107 |
Jamie | +0.106 |
ash | +0.100 |
Neg logit contributions | |
stronger | -0.138 |
generous | -0.130 |
helpful | -0.130 |
cowboy | -0.129 |
Thank | -0.125 |
Token | the |
Token ID | 262 |
Feature activation | -3.46e+00 |
Pos logit contributions | |
jo | +0.461 |
anne | +0.451 |
ash | +0.436 |
icky | +0.435 |
umble | +0.424 |
Neg logit contributions | |
stronger | -0.322 |
generous | -0.304 |
helpful | -0.300 |
Thank | -0.288 |
nearby | -0.284 |
Token | group |
Token ID | 1448 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
jo | +0.156 |
icky | +0.154 |
anne | +0.154 |
Jamie | +0.154 |
ash | +0.147 |
Neg logit contributions | |
stronger | -0.098 |
cowboy | -0.094 |
generous | -0.092 |
helpful | -0.091 |
Thank | -0.089 |
Token | . |
Token ID | 13 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
jo | +0.069 |
icky | +0.069 |
anne | +0.067 |
Jamie | +0.066 |
bie | +0.064 |
Neg logit contributions | |
stronger | -0.085 |
generous | -0.083 |
helpful | -0.082 |
nearby | -0.080 |
cowboy | -0.079 |
Token | Her |
Token ID | 2332 |
Feature activation | -4.40e-01 |
Pos logit contributions | |
jo | +0.091 |
icky | +0.088 |
anne | +0.087 |
ash | +0.086 |
Jamie | +0.085 |
Neg logit contributions | |
Thank | -0.094 |
vet | -0.092 |
generous | -0.090 |
Can | -0.089 |
stronger | -0.088 |
Token | dress |
Token ID | 6576 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
icky | +0.019 |
Jamie | +0.018 |
anne | +0.018 |
jo | +0.018 |
umble | +0.017 |
Neg logit contributions | |
stronger | -0.013 |
Thank | -0.013 |
cowboy | -0.013 |
Did | -0.013 |
fuller | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | +2.57e+00 |
Pos logit contributions | |
Jamie | +0.042 |
icky | +0.041 |
jo | +0.041 |
anne | +0.040 |
umble | +0.038 |
Neg logit contributions | |
cowboy | -0.034 |
Days | -0.033 |
Thank | -0.032 |
stronger | -0.032 |
please | -0.032 |
Token | chill |
Token ID | 20493 |
Feature activation | -7.64e-01 |
Pos logit contributions | |
cowboy | +0.079 |
stronger | +0.076 |
please | +0.074 |
Thank | +0.073 |
Indians | +0.072 |
Neg logit contributions | |
jo | -0.102 |
Jamie | -0.101 |
icky | -0.101 |
anne | -0.101 |
umble | -0.098 |
Token | run |
Token ID | 1057 |
Feature activation | +2.26e+00 |
Pos logit contributions | |
jo | +0.030 |
icky | +0.029 |
Jamie | +0.029 |
anne | +0.029 |
ash | +0.028 |
Neg logit contributions | |
stronger | -0.026 |
cowboy | -0.025 |
generous | -0.025 |
helpful | -0.024 |
Thank | -0.024 |
Token | down |
Token ID | 866 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
stronger | +0.088 |
cowboy | +0.087 |
generous | +0.086 |
Thank | +0.085 |
please | +0.085 |
Neg logit contributions | |
jo | -0.075 |
anne | -0.072 |
icky | -0.071 |
Jamie | -0.071 |
ash | -0.069 |
Token | his |
Token ID | 465 |
Feature activation | -2.36e+00 |
Pos logit contributions | |
jo | +0.059 |
Jamie | +0.059 |
icky | +0.058 |
anne | +0.057 |
ash | +0.054 |
Neg logit contributions | |
stronger | -0.070 |
cowboy | -0.070 |
generous | -0.068 |
Thank | -0.067 |
helpful | -0.067 |
Token | spine |
Token ID | 19656 |
Feature activation | -1.04e+01 |
Pos logit contributions | |
icky | +0.069 |
Jamie | +0.067 |
anne | +0.066 |
jo | +0.066 |
ash | +0.065 |
Neg logit contributions | |
stronger | -0.106 |
generous | -0.104 |
cowboy | -0.103 |
helpful | -0.103 |
vet | -0.099 |
Token | . |
Token ID | 13 |
Feature activation | -2.69e+00 |
Pos logit contributions | |
ash | +0.410 |
jo | +0.403 |
loe | +0.403 |
icky | +0.397 |
anne | +0.392 |
Neg logit contributions | |
stronger | -0.353 |
generous | -0.329 |
helpful | -0.328 |
nearby | -0.320 |
se | -0.314 |
Token | He |
Token ID | 679 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
jo | +0.099 |
anne | +0.097 |
icky | +0.094 |
ash | +0.093 |
bie | +0.089 |
Neg logit contributions | |
stronger | -0.099 |
Thank | -0.095 |
cowboy | -0.094 |
vet | -0.094 |
golf | -0.093 |
Token | could |
Token ID | 714 |
Feature activation | +2.88e+00 |
Pos logit contributions | |
stronger | +0.075 |
cowboy | +0.074 |
generous | +0.072 |
Thank | +0.071 |
helpful | +0.070 |
Neg logit contributions | |
jo | -0.092 |
Jamie | -0.091 |
anne | -0.090 |
icky | -0.088 |
ash | -0.083 |
Token | not |
Token ID | 407 |
Feature activation | -9.10e-01 |
Pos logit contributions | |
stronger | +0.081 |
cowboy | +0.081 |
generous | +0.077 |
Thank | +0.077 |
fuller | +0.075 |
Neg logit contributions | |
Jamie | -0.128 |
jo | -0.128 |
anne | -0.126 |
icky | -0.125 |
umble | -0.118 |
Token | believe |
Token ID | 1975 |
Feature activation | -4.05e+00 |
Pos logit contributions | |
jo | +0.040 |
Jamie | +0.040 |
anne | +0.039 |
icky | +0.039 |
umble | +0.037 |
Neg logit contributions | |
stronger | -0.026 |
cowboy | -0.026 |
generous | -0.025 |
Thank | -0.025 |
helpful | -0.024 |
Token | saw |
Token ID | 2497 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
stronger | +0.018 |
cowboy | +0.018 |
Days | +0.018 |
Thank | +0.018 |
generous | +0.017 |
Neg logit contributions | |
jo | -0.023 |
Jamie | -0.022 |
anne | -0.022 |
icky | -0.022 |
bie | -0.021 |
Token | a |
Token ID | 257 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
stronger | +0.070 |
cowboy | +0.069 |
Thank | +0.067 |
generous | +0.066 |
Days | +0.065 |
Neg logit contributions | |
Jamie | -0.089 |
jo | -0.088 |
anne | -0.087 |
icky | -0.087 |
ash | -0.081 |
Token | big |
Token ID | 1263 |
Feature activation | +4.06e+00 |
Pos logit contributions | |
Thank | +0.077 |
Days | +0.077 |
please | +0.072 |
Mail | +0.072 |
stronger | +0.071 |
Neg logit contributions | |
jo | -0.114 |
Jamie | -0.113 |
anne | -0.112 |
icky | -0.111 |
bie | -0.109 |
Token | horse |
Token ID | 8223 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
Days | +0.137 |
stronger | +0.132 |
please | +0.130 |
Thank | +0.129 |
Mail | +0.126 |
Neg logit contributions | |
Jamie | -0.155 |
jo | -0.153 |
anne | -0.149 |
icky | -0.147 |
Today | -0.143 |
Token | standing |
Token ID | 5055 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
stronger | +0.058 |
cowboy | +0.057 |
Days | +0.056 |
generous | +0.056 |
Thank | +0.055 |
Neg logit contributions | |
jo | -0.065 |
Jamie | -0.064 |
anne | -0.063 |
icky | -0.063 |
ash | -0.059 |
Token | amongst |
Token ID | 12077 |
Feature activation | -1.04e+01 |
Pos logit contributions | |
cowboy | +0.039 |
Days | +0.037 |
stronger | +0.036 |
vet | +0.036 |
Thank | +0.035 |
Neg logit contributions | |
Jamie | -0.043 |
icky | -0.043 |
jo | -0.041 |
ash | -0.041 |
anne | -0.041 |
Token | the |
Token ID | 262 |
Feature activation | -3.54e+00 |
Pos logit contributions | |
jo | +0.448 |
anne | +0.439 |
ash | +0.436 |
icky | +0.432 |
umble | +0.417 |
Neg logit contributions | |
stronger | -0.327 |
generous | -0.303 |
Thank | -0.302 |
helpful | -0.299 |
vet | -0.292 |
Token | trees |
Token ID | 7150 |
Feature activation | -3.13e+00 |
Pos logit contributions | |
jo | +0.150 |
Jamie | +0.145 |
anne | +0.144 |
icky | +0.143 |
ash | +0.138 |
Neg logit contributions | |
stronger | -0.106 |
cowboy | -0.102 |
Thank | -0.099 |
generous | -0.099 |
helpful | -0.098 |
Token | . |
Token ID | 13 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
jo | +0.091 |
icky | +0.090 |
Jamie | +0.090 |
anne | +0.089 |
ash | +0.084 |
Neg logit contributions | |
stronger | -0.137 |
cowboy | -0.132 |
generous | -0.131 |
helpful | -0.129 |
Thank | -0.129 |
Token |
|
Token ID | 198 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
jo | +0.090 |
icky | +0.089 |
ash | +0.088 |
anne | +0.086 |
ared | +0.083 |
Neg logit contributions | |
stronger | -0.094 |
Thank | -0.094 |
vet | -0.093 |
generous | -0.092 |
cowboy | -0.090 |
Token |
|
Token ID | 198 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
stronger | +0.058 |
cowboy | +0.057 |
generous | +0.055 |
helpful | +0.054 |
Thank | +0.054 |
Neg logit contributions | |
jo | -0.049 |
icky | -0.049 |
Jamie | -0.049 |
anne | -0.048 |
umble | -0.044 |
Token | Mama |
Token ID | 40084 |
Feature activation | -2.88e+00 |
Pos logit contributions | |
stronger | +0.018 |
cowboy | +0.017 |
helpful | +0.017 |
se | +0.017 |
generous | +0.017 |
Neg logit contributions | |
jo | -0.026 |
anne | -0.025 |
ared | -0.024 |
ash | -0.024 |
icky | -0.023 |
Token | began |
Token ID | 2540 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
jo | +0.116 |
Jamie | +0.114 |
icky | +0.114 |
anne | +0.114 |
ash | +0.108 |
Neg logit contributions | |
stronger | -0.094 |
cowboy | -0.091 |
generous | -0.089 |
helpful | -0.087 |
Thank | -0.087 |
Token | to |
Token ID | 284 |
Feature activation | +7.39e-01 |
Pos logit contributions | |
stronger | +0.049 |
generous | +0.047 |
helpful | +0.047 |
se | +0.046 |
cowboy | +0.044 |
Neg logit contributions | |
jo | -0.073 |
anne | -0.072 |
ash | -0.071 |
icky | -0.070 |
Jamie | -0.069 |
Token | cry |
Token ID | 3960 |
Feature activation | -3.86e+00 |
Pos logit contributions | |
stronger | +0.024 |
cowboy | +0.023 |
generous | +0.022 |
Thank | +0.022 |
helpful | +0.022 |
Neg logit contributions | |
Jamie | -0.030 |
jo | -0.030 |
icky | -0.030 |
anne | -0.030 |
ash | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
anne | +0.135 |
jo | +0.134 |
Jamie | +0.130 |
icky | +0.130 |
ash | +0.127 |
Neg logit contributions | |
stronger | -0.155 |
helpful | -0.144 |
generous | -0.143 |
cowboy | -0.142 |
nearby | -0.134 |
Token | Tears |
Token ID | 40328 |
Feature activation | -1.03e+01 |
Pos logit contributions | |
jo | +0.095 |
ash | +0.093 |
anne | +0.090 |
ared | +0.089 |
icky | +0.088 |
Neg logit contributions | |
generous | -0.065 |
stronger | -0.064 |
helpful | -0.064 |
vet | -0.062 |
Can | -0.061 |
Token | started |
Token ID | 2067 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
jo | +0.412 |
icky | +0.408 |
anne | +0.406 |
Jamie | +0.405 |
ash | +0.385 |
Neg logit contributions | |
stronger | -0.332 |
cowboy | -0.325 |
generous | -0.315 |
helpful | -0.309 |
Thank | -0.308 |
Token | streaming |
Token ID | 11305 |
Feature activation | -6.89e+00 |
Pos logit contributions | |
jo | +0.054 |
anne | +0.054 |
ash | +0.053 |
icky | +0.052 |
Jamie | +0.052 |
Neg logit contributions | |
stronger | -0.039 |
helpful | -0.036 |
generous | -0.035 |
se | -0.034 |
cowboy | -0.032 |
Token | down |
Token ID | 866 |
Feature activation | -3.79e+00 |
Pos logit contributions | |
anne | +0.277 |
jo | +0.276 |
icky | +0.275 |
Jamie | +0.272 |
ash | +0.271 |
Neg logit contributions | |
stronger | -0.230 |
helpful | -0.216 |
generous | -0.212 |
cowboy | -0.210 |
strong | -0.202 |
Token | her |
Token ID | 607 |
Feature activation | -4.21e+00 |
Pos logit contributions | |
ash | +0.175 |
jo | +0.175 |
ared | +0.167 |
anne | +0.167 |
fault | +0.164 |
Neg logit contributions | |
helpful | -0.106 |
stronger | -0.104 |
generous | -0.102 |
se | -0.099 |
strong | -0.098 |
Token | cheeks |
Token ID | 25839 |
Feature activation | -5.76e+00 |
Pos logit contributions | |
ash | +0.205 |
anne | +0.195 |
jo | +0.193 |
icky | +0.191 |
amp | +0.188 |
Neg logit contributions | |
generous | -0.111 |
helpful | -0.106 |
stronger | -0.100 |
strong | -0.097 |
brave | -0.095 |
Token | no |
Token ID | 645 |
Feature activation | -2.41e+00 |
Pos logit contributions | |
stronger | +0.017 |
cowboy | +0.016 |
generous | +0.016 |
helpful | +0.016 |
Thank | +0.015 |
Neg logit contributions | |
jo | -0.021 |
anne | -0.020 |
Jamie | -0.020 |
icky | -0.020 |
ash | -0.019 |
Token | energy |
Token ID | 2568 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
jo | +0.094 |
anne | +0.094 |
Jamie | +0.092 |
icky | +0.091 |
ash | +0.087 |
Neg logit contributions | |
stronger | -0.086 |
cowboy | -0.082 |
generous | -0.082 |
helpful | -0.081 |
vet | -0.078 |
Token | to |
Token ID | 284 |
Feature activation | +5.55e-01 |
Pos logit contributions | |
jo | +0.050 |
Jamie | +0.049 |
icky | +0.049 |
anne | +0.049 |
ash | +0.046 |
Neg logit contributions | |
stronger | -0.049 |
cowboy | -0.048 |
generous | -0.047 |
helpful | -0.046 |
Thank | -0.046 |
Token | play |
Token ID | 711 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
stronger | +0.018 |
cowboy | +0.018 |
generous | +0.017 |
Thank | +0.017 |
helpful | +0.017 |
Neg logit contributions | |
jo | -0.022 |
icky | -0.022 |
Jamie | -0.022 |
anne | -0.022 |
ash | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
stronger | +0.105 |
Thank | +0.103 |
cowboy | +0.102 |
generous | +0.101 |
Days | +0.100 |
Neg logit contributions | |
icky | -0.113 |
Jamie | -0.111 |
jo | -0.110 |
anne | -0.105 |
ash | -0.105 |
Token | Tears |
Token ID | 40328 |
Feature activation | -1.03e+01 |
Pos logit contributions | |
ash | +0.096 |
jo | +0.093 |
anne | +0.092 |
iness | +0.088 |
ol | +0.088 |
Neg logit contributions | |
vet | -0.057 |
Thank | -0.053 |
generous | -0.052 |
Can | -0.050 |
stronger | -0.050 |
Token | rolled |
Token ID | 11686 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
jo | +0.388 |
icky | +0.386 |
anne | +0.382 |
Jamie | +0.381 |
ash | +0.361 |
Neg logit contributions | |
stronger | -0.351 |
cowboy | -0.345 |
generous | -0.336 |
helpful | -0.330 |
vet | -0.329 |
Token | down |
Token ID | 866 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
stronger | +0.038 |
cowboy | +0.035 |
helpful | +0.035 |
generous | +0.035 |
Thank | +0.034 |
Neg logit contributions | |
anne | -0.041 |
jo | -0.041 |
icky | -0.041 |
Jamie | -0.041 |
umble | -0.038 |
Token | his |
Token ID | 465 |
Feature activation | -3.02e+00 |
Pos logit contributions | |
jo | +0.130 |
anne | +0.129 |
icky | +0.124 |
ash | +0.123 |
Jamie | +0.120 |
Neg logit contributions | |
stronger | -0.102 |
helpful | -0.093 |
generous | -0.092 |
cowboy | -0.091 |
Thank | -0.089 |
Token | cheeks |
Token ID | 25839 |
Feature activation | -5.18e+00 |
Pos logit contributions | |
anne | +0.115 |
ash | +0.114 |
Jamie | +0.112 |
jo | +0.110 |
icky | +0.105 |
Neg logit contributions | |
stronger | -0.110 |
generous | -0.109 |
helpful | -0.107 |
cowboy | -0.106 |
strong | -0.102 |
Token | and |
Token ID | 290 |
Feature activation | -6.41e-01 |
Pos logit contributions | |
jo | +0.183 |
icky | +0.182 |
Jamie | +0.181 |
anne | +0.180 |
ash | +0.170 |
Neg logit contributions | |
stronger | -0.188 |
cowboy | -0.186 |
generous | -0.182 |
Thank | -0.180 |
helpful | -0.179 |
Token | to |
Token ID | 284 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
stronger | +0.014 |
cowboy | +0.014 |
generous | +0.014 |
helpful | +0.013 |
Thank | +0.013 |
Neg logit contributions | |
jo | -0.019 |
anne | -0.019 |
icky | -0.018 |
Jamie | -0.018 |
ash | -0.018 |
Token | go |
Token ID | 467 |
Feature activation | +3.99e+00 |
Pos logit contributions | |
stronger | +0.033 |
cowboy | +0.033 |
generous | +0.031 |
Thank | +0.031 |
helpful | +0.031 |
Neg logit contributions | |
Jamie | -0.045 |
jo | -0.045 |
icky | -0.045 |
anne | -0.045 |
ash | -0.042 |
Token | to |
Token ID | 284 |
Feature activation | +4.94e-01 |
Pos logit contributions | |
stronger | +0.145 |
cowboy | +0.142 |
generous | +0.139 |
Thank | +0.137 |
helpful | +0.137 |
Neg logit contributions | |
jo | -0.146 |
Jamie | -0.144 |
icky | -0.143 |
anne | -0.143 |
ash | -0.136 |
Token | work |
Token ID | 670 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
stronger | +0.016 |
vet | +0.015 |
cowboy | +0.014 |
helpful | +0.014 |
store | +0.014 |
Neg logit contributions | |
anne | -0.022 |
jo | -0.022 |
ash | -0.021 |
icky | -0.021 |
ared | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
stronger | +0.064 |
generous | +0.059 |
helpful | +0.059 |
cowboy | +0.058 |
nicer | +0.057 |
Neg logit contributions | |
anne | -0.055 |
jo | -0.055 |
icky | -0.054 |
Jamie | -0.053 |
umble | -0.051 |
Token | Tears |
Token ID | 40328 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
icky | +0.081 |
anne | +0.081 |
ash | +0.079 |
jo | +0.077 |
Jamie | +0.076 |
Neg logit contributions | |
vet | -0.078 |
stronger | -0.074 |
generous | -0.073 |
store | -0.073 |
Days | -0.072 |
Token | started |
Token ID | 2067 |
Feature activation | -9.97e-01 |
Pos logit contributions | |
jo | +0.394 |
icky | +0.389 |
anne | +0.387 |
Jamie | +0.383 |
ash | +0.364 |
Neg logit contributions | |
stronger | -0.342 |
cowboy | -0.338 |
generous | -0.328 |
vet | -0.323 |
helpful | -0.321 |
Token | flowing |
Token ID | 17609 |
Feature activation | -4.60e-01 |
Pos logit contributions | |
jo | +0.042 |
anne | +0.042 |
icky | +0.041 |
Jamie | +0.041 |
ash | +0.041 |
Neg logit contributions | |
stronger | -0.033 |
helpful | -0.030 |
generous | -0.030 |
cowboy | -0.029 |
se | -0.028 |
Token | down |
Token ID | 866 |
Feature activation | -2.90e+00 |
Pos logit contributions | |
jo | +0.016 |
icky | +0.016 |
Jamie | +0.016 |
anne | +0.016 |
ash | +0.015 |
Neg logit contributions | |
stronger | -0.018 |
cowboy | -0.017 |
generous | -0.017 |
helpful | -0.016 |
Thank | -0.016 |
Token | his |
Token ID | 465 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
jo | +0.121 |
anne | +0.121 |
icky | +0.119 |
ash | +0.117 |
umble | +0.113 |
Neg logit contributions | |
stronger | -0.095 |
helpful | -0.089 |
generous | -0.088 |
cowboy | -0.085 |
nearby | -0.084 |
Token | cheeks |
Token ID | 25839 |
Feature activation | -5.10e+00 |
Pos logit contributions | |
ash | +0.106 |
Jamie | +0.103 |
anne | +0.101 |
icky | +0.100 |
raw | +0.100 |
Neg logit contributions | |
generous | -0.097 |
helpful | -0.096 |
stronger | -0.091 |
strong | -0.090 |
cowboy | -0.089 |
Token | and |
Token ID | 290 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
jo | +0.088 |
icky | +0.088 |
Jamie | +0.087 |
anne | +0.087 |
ash | +0.083 |
Neg logit contributions | |
stronger | -0.076 |
cowboy | -0.074 |
generous | -0.072 |
helpful | -0.071 |
Thank | -0.070 |
Token | headed |
Token ID | 9153 |
Feature activation | +8.92e-01 |
Pos logit contributions | |
Jamie | +0.011 |
jo | +0.011 |
anne | +0.011 |
icky | +0.011 |
ash | +0.010 |
Neg logit contributions | |
stronger | -0.008 |
cowboy | -0.007 |
Days | -0.007 |
generous | -0.007 |
vet | -0.007 |
Token | off |
Token ID | 572 |
Feature activation | -5.27e+00 |
Pos logit contributions | |
stronger | +0.034 |
cowboy | +0.033 |
generous | +0.033 |
Thank | +0.033 |
Days | +0.033 |
Neg logit contributions | |
Jamie | -0.030 |
jo | -0.030 |
icky | -0.030 |
anne | -0.029 |
ash | -0.028 |
Token | singing |
Token ID | 13777 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
jo | +0.200 |
anne | +0.196 |
icky | +0.196 |
Jamie | +0.194 |
ash | +0.186 |
Neg logit contributions | |
stronger | -0.189 |
cowboy | -0.181 |
generous | -0.178 |
helpful | -0.176 |
Thank | -0.174 |
Token | " |
Token ID | 366 |
Feature activation | +4.97e+00 |
Pos logit contributions | |
jo | +0.034 |
Jamie | +0.034 |
icky | +0.033 |
anne | +0.033 |
ash | +0.031 |
Neg logit contributions | |
stronger | -0.038 |
cowboy | -0.037 |
Thank | -0.037 |
Days | -0.037 |
vet | -0.037 |
Token | Oh |
Token ID | 5812 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
stronger | +0.139 |
cowboy | +0.136 |
generous | +0.131 |
helpful | +0.129 |
vet | +0.127 |
Neg logit contributions | |
jo | -0.224 |
anne | -0.219 |
icky | -0.219 |
Jamie | -0.219 |
ash | -0.213 |
Token | , |
Token ID | 11 |
Feature activation | -8.10e-01 |
Pos logit contributions | |
jo | +0.403 |
icky | +0.397 |
anne | +0.393 |
umble | +0.385 |
ash | +0.382 |
Neg logit contributions | |
stronger | -0.341 |
generous | -0.340 |
cowboy | -0.338 |
helpful | -0.335 |
Thank | -0.308 |
Token | the |
Token ID | 262 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
jo | +0.027 |
icky | +0.026 |
Jamie | +0.026 |
anne | +0.026 |
ash | +0.025 |
Neg logit contributions | |
stronger | -0.032 |
cowboy | -0.032 |
generous | -0.031 |
helpful | -0.031 |
Thank | -0.030 |
Token | farmer |
Token ID | 18739 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
Jamie | +0.011 |
jo | +0.011 |
icky | +0.011 |
anne | +0.011 |
ash | +0.010 |
Neg logit contributions | |
stronger | -0.008 |
cowboy | -0.007 |
generous | -0.007 |
Days | -0.007 |
Thank | -0.007 |
Token | in |
Token ID | 287 |
Feature activation | -5.74e+00 |
Pos logit contributions | |
stronger | +0.065 |
cowboy | +0.063 |
generous | +0.063 |
helpful | +0.062 |
Thank | +0.062 |
Neg logit contributions | |
jo | -0.054 |
icky | -0.054 |
anne | -0.053 |
Jamie | -0.053 |
ash | -0.050 |
Token | the |
Token ID | 262 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
Jamie | +0.202 |
jo | +0.202 |
icky | +0.201 |
anne | +0.197 |
ash | +0.188 |
Neg logit contributions | |
stronger | -0.212 |
cowboy | -0.207 |
generous | -0.200 |
Thank | -0.200 |
helpful | -0.198 |
Token | he |
Token ID | 339 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
cowboy | +0.017 |
stronger | +0.017 |
Days | +0.017 |
generous | +0.016 |
vet | +0.016 |
Neg logit contributions | |
Jamie | -0.020 |
jo | -0.020 |
anne | -0.019 |
icky | -0.019 |
ash | -0.018 |
Token | seemed |
Token ID | 3947 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
cowboy | +0.044 |
stronger | +0.043 |
Days | +0.042 |
Thank | +0.042 |
vet | +0.042 |
Neg logit contributions | |
Jamie | -0.047 |
jo | -0.047 |
anne | -0.047 |
icky | -0.044 |
bie | -0.043 |
Token | like |
Token ID | 588 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
jo | +0.068 |
icky | +0.067 |
Jamie | +0.066 |
anne | +0.066 |
ash | +0.063 |
Neg logit contributions | |
stronger | -0.054 |
cowboy | -0.052 |
generous | -0.051 |
helpful | -0.050 |
Thank | -0.050 |
Token | he |
Token ID | 339 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
Jamie | +0.051 |
icky | +0.045 |
jo | +0.045 |
anne | +0.044 |
Jo | +0.044 |
Neg logit contributions | |
Days | -0.053 |
cowboy | -0.053 |
Tweet | -0.052 |
stronger | -0.052 |
please | -0.052 |
Token | was |
Token ID | 373 |
Feature activation | -2.63e+00 |
Pos logit contributions | |
stronger | +0.002 |
cowboy | +0.001 |
generous | +0.001 |
Thank | +0.001 |
helpful | +0.001 |
Neg logit contributions | |
jo | -0.002 |
Jamie | -0.002 |
anne | -0.002 |
icky | -0.002 |
ash | -0.001 |
Token | amongst |
Token ID | 12077 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
Jamie | +0.121 |
anne | +0.120 |
jo | +0.120 |
icky | +0.120 |
ash | +0.114 |
Neg logit contributions | |
cowboy | -0.067 |
stronger | -0.066 |
Thank | -0.064 |
Days | -0.063 |
generous | -0.063 |
Token | the |
Token ID | 262 |
Feature activation | -3.20e+00 |
Pos logit contributions | |
jo | +0.445 |
anne | +0.441 |
icky | +0.428 |
ash | +0.427 |
umble | +0.416 |
Neg logit contributions | |
stronger | -0.321 |
generous | -0.297 |
helpful | -0.295 |
cowboy | -0.286 |
nearby | -0.281 |
Token | stars |
Token ID | 5788 |
Feature activation | -5.29e+00 |
Pos logit contributions | |
jo | +0.145 |
Jamie | +0.143 |
icky | +0.143 |
anne | +0.143 |
ash | +0.136 |
Neg logit contributions | |
stronger | -0.087 |
cowboy | -0.085 |
generous | -0.081 |
Thank | -0.080 |
helpful | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
jo | +0.167 |
icky | +0.164 |
anne | +0.163 |
Jamie | +0.163 |
ash | +0.154 |
Neg logit contributions | |
stronger | -0.220 |
cowboy | -0.213 |
generous | -0.209 |
helpful | -0.207 |
Thank | -0.204 |
Token | But |
Token ID | 887 |
Feature activation | -5.25e+00 |
Pos logit contributions | |
icky | +0.096 |
anne | +0.095 |
jo | +0.094 |
ash | +0.093 |
ared | +0.092 |
Neg logit contributions | |
Thank | -0.084 |
stronger | -0.083 |
vet | -0.081 |
generous | -0.079 |
cowboy | -0.078 |
Token | , |
Token ID | 11 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
jo | +0.240 |
icky | +0.239 |
anne | +0.238 |
ash | +0.229 |
Jamie | +0.227 |
Neg logit contributions | |
stronger | -0.151 |
cowboy | -0.142 |
generous | -0.141 |
helpful | -0.138 |
Thank | -0.137 |
Token | bee |
Token ID | 20697 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
icky | +0.053 |
jo | +0.052 |
Jamie | +0.051 |
anne | +0.051 |
ash | +0.051 |
Neg logit contributions | |
stronger | -0.037 |
cowboy | -0.036 |
generous | -0.036 |
helpful | -0.035 |
Thank | -0.034 |
Token | could |
Token ID | 714 |
Feature activation | +3.51e+00 |
Pos logit contributions | |
stronger | +0.037 |
cowboy | +0.036 |
generous | +0.035 |
helpful | +0.034 |
Thank | +0.034 |
Neg logit contributions | |
jo | -0.048 |
anne | -0.047 |
Jamie | -0.047 |
icky | -0.046 |
ash | -0.044 |
Token | sense |
Token ID | 2565 |
Feature activation | -4.66e+00 |
Pos logit contributions | |
cowboy | +0.106 |
fuller | +0.101 |
stronger | +0.101 |
Thank | +0.100 |
Did | +0.099 |
Neg logit contributions | |
Jamie | -0.146 |
icky | -0.144 |
anne | -0.142 |
jo | -0.141 |
umble | -0.137 |
Token | the |
Token ID | 262 |
Feature activation | -6.15e-01 |
Pos logit contributions | |
jo | +0.177 |
icky | +0.175 |
anne | +0.172 |
Jamie | +0.170 |
ash | +0.168 |
Neg logit contributions | |
stronger | -0.167 |
generous | -0.157 |
helpful | -0.156 |
cowboy | -0.152 |
Thank | -0.150 |
Token | stress |
Token ID | 5503 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
jo | +0.028 |
Jamie | +0.027 |
icky | +0.027 |
anne | +0.027 |
ash | +0.026 |
Neg logit contributions | |
stronger | -0.017 |
cowboy | -0.017 |
generous | -0.016 |
Thank | -0.016 |
helpful | -0.016 |
Token | amongst |
Token ID | 12077 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
jo | +0.100 |
icky | +0.099 |
Jamie | +0.099 |
anne | +0.097 |
ash | +0.094 |
Neg logit contributions | |
stronger | -0.089 |
cowboy | -0.084 |
generous | -0.084 |
helpful | -0.083 |
Thank | -0.081 |
Token | her |
Token ID | 607 |
Feature activation | -4.52e+00 |
Pos logit contributions | |
jo | +0.481 |
ash | +0.472 |
obby | +0.461 |
icky | +0.459 |
anne | +0.453 |
Neg logit contributions | |
stronger | -0.273 |
nearby | -0.256 |
generous | -0.252 |
helpful | -0.250 |
Thank | -0.242 |
Token | hive |
Token ID | 35881 |
Feature activation | -5.12e-01 |
Pos logit contributions | |
jo | +0.168 |
icky | +0.166 |
Jamie | +0.165 |
anne | +0.163 |
ash | +0.162 |
Neg logit contributions | |
stronger | -0.159 |
generous | -0.159 |
helpful | -0.156 |
cowboy | -0.149 |
vet | -0.146 |
Token | . |
Token ID | 13 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
jo | +0.017 |
icky | +0.017 |
Jamie | +0.017 |
anne | +0.016 |
ash | +0.016 |
Neg logit contributions | |
stronger | -0.020 |
generous | -0.020 |
helpful | -0.019 |
cowboy | -0.019 |
Thank | -0.019 |
Token | She |
Token ID | 1375 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
jo | +0.069 |
icky | +0.068 |
ash | +0.067 |
anne | +0.066 |
ared | +0.064 |
Neg logit contributions | |
vet | -0.072 |
Thank | -0.071 |
stronger | -0.069 |
generous | -0.068 |
Can | -0.067 |
Token | tried |
Token ID | 3088 |
Feature activation | +3.61e+00 |
Pos logit contributions | |
stronger | +0.058 |
cowboy | +0.057 |
generous | +0.054 |
Thank | +0.054 |
helpful | +0.053 |
Neg logit contributions | |
jo | -0.075 |
anne | -0.074 |
Jamie | -0.074 |
icky | -0.074 |
ash | -0.070 |
Token | felt |
Token ID | 2936 |
Feature activation | -8.00e-01 |
Pos logit contributions | |
jo | +0.114 |
anne | +0.111 |
Jamie | +0.111 |
icky | +0.107 |
ash | +0.100 |
Neg logit contributions | |
stronger | -0.097 |
cowboy | -0.095 |
Thank | -0.094 |
generous | -0.093 |
helpful | -0.092 |
Token | a |
Token ID | 257 |
Feature activation | +2.53e+00 |
Pos logit contributions | |
Jamie | +0.030 |
icky | +0.030 |
jo | +0.030 |
anne | +0.030 |
ash | +0.029 |
Neg logit contributions | |
cowboy | -0.026 |
please | -0.024 |
Offic | -0.024 |
Thank | -0.024 |
Days | -0.024 |
Token | chill |
Token ID | 20493 |
Feature activation | -8.70e-01 |
Pos logit contributions | |
cowboy | +0.080 |
please | +0.079 |
stronger | +0.078 |
Thank | +0.077 |
Mail | +0.075 |
Neg logit contributions | |
icky | -0.096 |
jo | -0.095 |
anne | -0.095 |
Jamie | -0.093 |
umble | -0.090 |
Token | down |
Token ID | 866 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
jo | +0.033 |
Jamie | +0.033 |
icky | +0.032 |
anne | +0.032 |
ash | +0.031 |
Neg logit contributions | |
stronger | -0.030 |
cowboy | -0.030 |
generous | -0.029 |
Thank | -0.029 |
helpful | -0.028 |
Token | his |
Token ID | 465 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
jo | +0.053 |
icky | +0.052 |
Jamie | +0.052 |
anne | +0.052 |
ash | +0.049 |
Neg logit contributions | |
stronger | -0.058 |
cowboy | -0.056 |
generous | -0.055 |
helpful | -0.055 |
Thank | -0.054 |
Token | spine |
Token ID | 19656 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
icky | +0.064 |
Jamie | +0.063 |
jo | +0.063 |
anne | +0.063 |
ash | +0.060 |
Neg logit contributions | |
stronger | -0.099 |
generous | -0.096 |
helpful | -0.096 |
cowboy | -0.096 |
vet | -0.092 |
Token | . |
Token ID | 13 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
jo | +0.383 |
ash | +0.381 |
icky | +0.373 |
loe | +0.370 |
Jamie | +0.369 |
Neg logit contributions | |
stronger | -0.367 |
generous | -0.339 |
helpful | -0.338 |
nearby | -0.336 |
se | -0.332 |
Token | He |
Token ID | 679 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
jo | +0.092 |
anne | +0.091 |
ash | +0.089 |
icky | +0.088 |
Jamie | +0.087 |
Neg logit contributions | |
stronger | -0.075 |
Thank | -0.074 |
vet | -0.072 |
generous | -0.071 |
helpful | -0.070 |
Token | suddenly |
Token ID | 6451 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
cowboy | +0.087 |
stronger | +0.087 |
generous | +0.084 |
Thank | +0.083 |
helpful | +0.082 |
Neg logit contributions | |
jo | -0.091 |
Jamie | -0.089 |
anne | -0.089 |
icky | -0.086 |
bie | -0.081 |
Token | remembered |
Token ID | 12086 |
Feature activation | +8.72e-01 |
Pos logit contributions | |
jo | +0.088 |
icky | +0.087 |
Jamie | +0.087 |
anne | +0.086 |
ash | +0.083 |
Neg logit contributions | |
stronger | -0.055 |
cowboy | -0.053 |
generous | -0.052 |
helpful | -0.051 |
Thank | -0.051 |
Token | the |
Token ID | 262 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
stronger | +0.030 |
cowboy | +0.029 |
generous | +0.029 |
helpful | +0.028 |
Thank | +0.028 |
Neg logit contributions | |
jo | -0.034 |
icky | -0.033 |
anne | -0.033 |
Jamie | -0.033 |
ash | -0.032 |
Token | girl |
Token ID | 2576 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
stronger | +0.038 |
cowboy | +0.036 |
generous | +0.035 |
Thank | +0.035 |
helpful | +0.035 |
Neg logit contributions | |
jo | -0.050 |
Jamie | -0.049 |
icky | -0.049 |
anne | -0.049 |
ash | -0.046 |
Token | didn |
Token ID | 1422 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
jo | +0.065 |
icky | +0.064 |
Jamie | +0.064 |
anne | +0.063 |
ash | +0.061 |
Neg logit contributions | |
stronger | -0.052 |
cowboy | -0.050 |
generous | -0.049 |
helpful | -0.049 |
Thank | -0.047 |
Token | 't |
Token ID | 470 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
jo | +0.005 |
anne | +0.005 |
Jamie | +0.005 |
icky | +0.005 |
ash | +0.005 |
Neg logit contributions | |
stronger | -0.004 |
cowboy | -0.004 |
generous | -0.004 |
helpful | -0.004 |
Thank | -0.004 |
Token | understand |
Token ID | 1833 |
Feature activation | -7.32e-01 |
Pos logit contributions | |
Jamie | +0.018 |
jo | +0.018 |
icky | +0.018 |
anne | +0.018 |
ash | +0.017 |
Neg logit contributions | |
stronger | -0.013 |
generous | -0.012 |
helpful | -0.012 |
cowboy | -0.012 |
Thank | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
anne | +0.030 |
jo | +0.029 |
icky | +0.029 |
obby | +0.028 |
Jamie | +0.027 |
Neg logit contributions | |
stronger | -0.025 |
generous | -0.024 |
helpful | -0.023 |
nearby | -0.023 |
kind | -0.023 |
Token | Tears |
Token ID | 40328 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
jo | +0.072 |
ash | +0.071 |
anne | +0.069 |
icky | +0.067 |
raw | +0.062 |
Neg logit contributions | |
vet | -0.067 |
Thank | -0.065 |
generous | -0.064 |
stronger | -0.064 |
store | -0.063 |
Token | started |
Token ID | 2067 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
icky | +0.372 |
jo | +0.372 |
anne | +0.369 |
Jamie | +0.359 |
ash | +0.341 |
Neg logit contributions | |
cowboy | -0.353 |
stronger | -0.349 |
generous | -0.335 |
vet | -0.334 |
please | -0.332 |
Token | rolling |
Token ID | 10708 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
jo | +0.043 |
Jamie | +0.042 |
anne | +0.042 |
icky | +0.042 |
ash | +0.041 |
Neg logit contributions | |
stronger | -0.036 |
generous | -0.033 |
helpful | -0.033 |
cowboy | -0.033 |
Thank | -0.032 |
Token | down |
Token ID | 866 |
Feature activation | -3.13e+00 |
Pos logit contributions | |
stronger | +0.005 |
cowboy | +0.004 |
generous | +0.004 |
helpful | +0.004 |
Thank | +0.004 |
Neg logit contributions | |
jo | -0.005 |
icky | -0.005 |
Jamie | -0.005 |
anne | -0.005 |
ash | -0.004 |
Token | her |
Token ID | 607 |
Feature activation | -3.96e+00 |
Pos logit contributions | |
jo | +0.130 |
anne | +0.127 |
ash | +0.125 |
icky | +0.125 |
Jamie | +0.122 |
Neg logit contributions | |
stronger | -0.107 |
helpful | -0.099 |
generous | -0.099 |
Thank | -0.095 |
nearby | -0.094 |
Token | face |
Token ID | 1986 |
Feature activation | -4.31e+00 |
Pos logit contributions | |
ash | +0.163 |
icky | +0.153 |
Jamie | +0.148 |
anne | +0.148 |
jo | +0.148 |
Neg logit contributions | |
generous | -0.136 |
helpful | -0.133 |
stronger | -0.132 |
se | -0.126 |
strong | -0.126 |
Token | but |
Token ID | 475 |
Feature activation | -3.48e+00 |
Pos logit contributions | |
cowboy | +0.056 |
stronger | +0.055 |
generous | +0.053 |
Thank | +0.053 |
Days | +0.053 |
Neg logit contributions | |
Jamie | -0.055 |
jo | -0.052 |
anne | -0.051 |
icky | -0.051 |
umble | -0.049 |
Token | it |
Token ID | 340 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
jo | +0.142 |
anne | +0.139 |
icky | +0.139 |
Jamie | +0.137 |
ash | +0.132 |
Neg logit contributions | |
stronger | -0.115 |
cowboy | -0.109 |
generous | -0.108 |
helpful | -0.107 |
Thank | -0.106 |
Token | seemed |
Token ID | 3947 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
stronger | +0.007 |
cowboy | +0.007 |
generous | +0.007 |
helpful | +0.007 |
Thank | +0.006 |
Neg logit contributions | |
jo | -0.007 |
Jamie | -0.007 |
icky | -0.007 |
anne | -0.006 |
ash | -0.006 |
Token | impossible |
Token ID | 5340 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
jo | +0.086 |
anne | +0.084 |
icky | +0.084 |
Jamie | +0.083 |
ash | +0.080 |
Neg logit contributions | |
stronger | -0.062 |
generous | -0.057 |
helpful | -0.057 |
cowboy | -0.056 |
Thank | -0.054 |
Token | . |
Token ID | 13 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
anne | +0.056 |
jo | +0.056 |
icky | +0.055 |
ash | +0.054 |
umble | +0.053 |
Neg logit contributions | |
stronger | -0.060 |
helpful | -0.056 |
generous | -0.055 |
nicer | -0.053 |
smarter | -0.052 |
Token | Tears |
Token ID | 40328 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
jo | +0.084 |
anne | +0.083 |
ash | +0.079 |
icky | +0.078 |
ared | +0.076 |
Neg logit contributions | |
vet | -0.083 |
Thank | -0.083 |
stronger | -0.083 |
Can | -0.081 |
generous | -0.080 |
Token | started |
Token ID | 2067 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
jo | +0.409 |
icky | +0.406 |
Jamie | +0.405 |
anne | +0.405 |
ash | +0.384 |
Neg logit contributions | |
stronger | -0.321 |
cowboy | -0.316 |
generous | -0.306 |
helpful | -0.299 |
vet | -0.297 |
Token | to |
Token ID | 284 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
jo | +0.046 |
anne | +0.046 |
icky | +0.045 |
Jamie | +0.044 |
ash | +0.044 |
Neg logit contributions | |
stronger | -0.035 |
helpful | -0.032 |
generous | -0.032 |
se | -0.031 |
cowboy | -0.030 |
Token | come |
Token ID | 1282 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
jo | +0.057 |
icky | +0.056 |
anne | +0.056 |
Jamie | +0.054 |
ash | +0.054 |
Neg logit contributions | |
stronger | -0.036 |
cowboy | -0.033 |
generous | -0.033 |
helpful | -0.032 |
Thank | -0.032 |
Token | from |
Token ID | 422 |
Feature activation | -5.62e+00 |
Pos logit contributions | |
stronger | +0.050 |
cowboy | +0.047 |
generous | +0.047 |
helpful | +0.047 |
Thank | +0.045 |
Neg logit contributions | |
jo | -0.047 |
anne | -0.047 |
Jamie | -0.046 |
icky | -0.046 |
ash | -0.044 |
Token | her |
Token ID | 607 |
Feature activation | -3.02e+00 |
Pos logit contributions | |
jo | +0.249 |
icky | +0.244 |
anne | +0.243 |
ash | +0.239 |
anny | +0.230 |
Neg logit contributions | |
stronger | -0.176 |
cowboy | -0.164 |
golf | -0.161 |
Thank | -0.161 |
helpful | -0.161 |
Token | sh |
Token ID | 427 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
cowboy | +0.080 |
stronger | +0.079 |
Thank | +0.076 |
please | +0.074 |
Offic | +0.074 |
Neg logit contributions | |
icky | -0.099 |
jo | -0.097 |
anne | -0.096 |
Jamie | -0.095 |
umble | -0.095 |
Token | iver |
Token ID | 1428 |
Feature activation | +7.88e-01 |
Pos logit contributions | |
stronger | +0.069 |
cowboy | +0.067 |
generous | +0.065 |
Thank | +0.064 |
helpful | +0.064 |
Neg logit contributions | |
jo | -0.088 |
icky | -0.087 |
anne | -0.087 |
Jamie | -0.085 |
ash | -0.083 |
Token | go |
Token ID | 467 |
Feature activation | +3.92e+00 |
Pos logit contributions | |
stronger | +0.028 |
helpful | +0.025 |
generous | +0.025 |
cowboy | +0.025 |
nearby | +0.025 |
Neg logit contributions | |
Jamie | -0.032 |
jo | -0.031 |
icky | -0.031 |
anne | -0.031 |
obby | -0.030 |
Token | down |
Token ID | 866 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
cowboy | +0.147 |
Thank | +0.141 |
Days | +0.139 |
generous | +0.138 |
stronger | +0.138 |
Neg logit contributions | |
jo | -0.135 |
anne | -0.129 |
icky | -0.128 |
ash | -0.128 |
Today | -0.125 |
Token | her |
Token ID | 607 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
jo | +0.042 |
anne | +0.041 |
Jamie | +0.040 |
icky | +0.040 |
ash | +0.038 |
Neg logit contributions | |
stronger | -0.053 |
helpful | -0.049 |
generous | -0.049 |
Thank | -0.049 |
cowboy | -0.048 |
Token | spine |
Token ID | 19656 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
Jamie | +0.101 |
icky | +0.101 |
anne | +0.099 |
jo | +0.099 |
ash | +0.098 |
Neg logit contributions | |
stronger | -0.105 |
helpful | -0.104 |
generous | -0.104 |
vet | -0.097 |
cowboy | -0.097 |
Token | . |
Token ID | 13 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
loe | +0.430 |
jo | +0.429 |
ash | +0.421 |
Jamie | +0.418 |
obby | +0.414 |
Neg logit contributions | |
stronger | -0.323 |
helpful | -0.298 |
generous | -0.295 |
nearby | -0.289 |
strong | -0.284 |
Token | It |
Token ID | 632 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
jo | +0.101 |
icky | +0.100 |
ash | +0.097 |
anne | +0.097 |
Jamie | +0.096 |
Neg logit contributions | |
Thank | -0.082 |
Can | -0.079 |
stronger | -0.079 |
generous | -0.079 |
vet | -0.078 |
Token | was |
Token ID | 373 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
stronger | +0.094 |
cowboy | +0.092 |
generous | +0.089 |
helpful | +0.088 |
Thank | +0.088 |
Neg logit contributions | |
jo | -0.088 |
Jamie | -0.086 |
anne | -0.086 |
icky | -0.086 |
ash | -0.081 |
Token | very |
Token ID | 845 |
Feature activation | -8.55e-01 |
Pos logit contributions | |
jo | +0.059 |
Jamie | +0.058 |
icky | +0.058 |
anne | +0.058 |
ash | +0.055 |
Neg logit contributions | |
stronger | -0.047 |
cowboy | -0.045 |
generous | -0.044 |
helpful | -0.043 |
Thank | -0.043 |
Token | cold |
Token ID | 4692 |
Feature activation | -9.11e-01 |
Pos logit contributions | |
jo | +0.042 |
Jamie | +0.041 |
anne | +0.040 |
icky | +0.040 |
ash | +0.039 |
Neg logit contributions | |
stronger | -0.023 |
helpful | -0.023 |
generous | -0.022 |
Thank | -0.020 |
cowboy | -0.019 |
Token | happy |
Token ID | 3772 |
Feature activation | +5.76e-01 |
Pos logit contributions | |
Jamie | +0.141 |
icky | +0.140 |
jo | +0.139 |
anne | +0.139 |
ash | +0.133 |
Neg logit contributions | |
stronger | -0.064 |
cowboy | -0.063 |
vet | -0.058 |
generous | -0.057 |
Thank | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | -3.42e+00 |
Pos logit contributions | |
helpful | +0.016 |
generous | +0.016 |
se | +0.016 |
golf | +0.016 |
stronger | +0.016 |
Neg logit contributions | |
icky | -0.025 |
jo | -0.025 |
ash | -0.025 |
loe | -0.025 |
raw | -0.025 |
Token | She |
Token ID | 1375 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
jo | +0.149 |
icky | +0.147 |
ash | +0.146 |
ert | +0.146 |
ared | +0.139 |
Neg logit contributions | |
Thank | -0.104 |
Can | -0.101 |
generous | -0.089 |
store | -0.089 |
helpful | -0.084 |
Token | was |
Token ID | 373 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
stronger | +0.044 |
cowboy | +0.043 |
generous | +0.042 |
helpful | +0.041 |
Thank | +0.041 |
Neg logit contributions | |
jo | -0.062 |
icky | -0.061 |
anne | -0.061 |
Jamie | -0.061 |
ash | -0.059 |
Token | playing |
Token ID | 2712 |
Feature activation | +4.13e+00 |
Pos logit contributions | |
Jamie | +0.055 |
icky | +0.052 |
anne | +0.051 |
jo | +0.050 |
ash | +0.049 |
Neg logit contributions | |
cowboy | -0.039 |
Days | -0.039 |
stronger | -0.038 |
Tweet | -0.037 |
Mic | -0.037 |
Token | amongst |
Token ID | 12077 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
stronger | +0.153 |
fuller | +0.152 |
Days | +0.148 |
please | +0.145 |
Offic | +0.143 |
Neg logit contributions | |
Jamie | -0.142 |
icky | -0.140 |
ash | -0.132 |
jo | -0.128 |
Kate | -0.124 |
Token | the |
Token ID | 262 |
Feature activation | -3.30e+00 |
Pos logit contributions | |
jo | +0.381 |
anne | +0.373 |
icky | +0.370 |
Jamie | +0.359 |
ash | +0.356 |
Neg logit contributions | |
stronger | -0.363 |
generous | -0.349 |
cowboy | -0.346 |
helpful | -0.344 |
Thank | -0.340 |
Token | trees |
Token ID | 7150 |
Feature activation | -3.08e+00 |
Pos logit contributions | |
icky | +0.145 |
jo | +0.145 |
Jamie | +0.145 |
anne | +0.141 |
ash | +0.138 |
Neg logit contributions | |
stronger | -0.094 |
cowboy | -0.090 |
Thank | -0.087 |
generous | -0.086 |
Days | -0.085 |
Token | and |
Token ID | 290 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
jo | +0.103 |
Jamie | +0.102 |
icky | +0.102 |
anne | +0.101 |
ash | +0.096 |
Neg logit contributions | |
cowboy | -0.120 |
stronger | -0.120 |
generous | -0.115 |
Thank | -0.113 |
helpful | -0.113 |
Token | the |
Token ID | 262 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
Jamie | +0.045 |
jo | +0.043 |
icky | +0.042 |
anne | +0.042 |
ash | +0.040 |
Neg logit contributions | |
stronger | -0.038 |
cowboy | -0.037 |
Days | -0.036 |
generous | -0.036 |
vet | -0.035 |
Token | birds |
Token ID | 10087 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
jo | +0.097 |
Jamie | +0.096 |
icky | +0.096 |
anne | +0.095 |
ash | +0.092 |
Neg logit contributions | |
stronger | -0.053 |
cowboy | -0.051 |
generous | -0.049 |
Thank | -0.048 |
helpful | -0.048 |
Token | a |
Token ID | 257 |
Feature activation | +2.98e+00 |
Pos logit contributions | |
Jamie | +0.020 |
icky | +0.020 |
jo | +0.019 |
anne | +0.019 |
umble | +0.018 |
Neg logit contributions | |
cowboy | -0.015 |
Days | -0.014 |
stronger | -0.014 |
please | -0.014 |
Thank | -0.014 |
Token | sh |
Token ID | 427 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
stronger | +0.089 |
cowboy | +0.086 |
Thank | +0.083 |
generous | +0.082 |
please | +0.082 |
Neg logit contributions | |
jo | -0.125 |
icky | -0.125 |
Jamie | -0.124 |
anne | -0.123 |
umble | -0.116 |
Token | iver |
Token ID | 1428 |
Feature activation | +7.46e-01 |
Pos logit contributions | |
stronger | +0.068 |
cowboy | +0.065 |
generous | +0.064 |
helpful | +0.063 |
Thank | +0.063 |
Neg logit contributions | |
jo | -0.079 |
Jamie | -0.078 |
icky | -0.078 |
anne | -0.077 |
ash | -0.074 |
Token | in |
Token ID | 287 |
Feature activation | -5.44e+00 |
Pos logit contributions | |
stronger | +0.027 |
cowboy | +0.026 |
generous | +0.026 |
helpful | +0.025 |
Thank | +0.025 |
Neg logit contributions | |
jo | -0.028 |
icky | -0.028 |
anne | -0.027 |
Jamie | -0.027 |
ash | -0.026 |
Token | her |
Token ID | 607 |
Feature activation | -3.93e+00 |
Pos logit contributions | |
jo | +0.199 |
anne | +0.195 |
icky | +0.193 |
ash | +0.189 |
Jamie | +0.188 |
Neg logit contributions | |
stronger | -0.205 |
generous | -0.196 |
helpful | -0.195 |
cowboy | -0.194 |
Thank | -0.193 |
Token | spine |
Token ID | 19656 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
icky | +0.151 |
ash | +0.149 |
anne | +0.149 |
jo | +0.148 |
obby | +0.147 |
Neg logit contributions | |
generous | -0.140 |
stronger | -0.138 |
helpful | -0.136 |
cowboy | -0.135 |
vet | -0.129 |
Token | . |
Token ID | 13 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
obby | +0.414 |
ash | +0.412 |
jo | +0.406 |
loe | +0.402 |
anne | +0.394 |
Neg logit contributions | |
stronger | -0.327 |
generous | -0.308 |
helpful | -0.306 |
nearby | -0.303 |
strong | -0.291 |
Token |
|
Token ID | 198 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
jo | +0.085 |
anne | +0.084 |
ash | +0.083 |
icky | +0.081 |
Jamie | +0.079 |
Neg logit contributions | |
stronger | -0.072 |
Thank | -0.072 |
vet | -0.070 |
Can | -0.069 |
generous | -0.069 |
Token |
|
Token ID | 198 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
stronger | +0.067 |
generous | +0.062 |
helpful | +0.062 |
Thank | +0.062 |
vet | +0.061 |
Neg logit contributions | |
jo | -0.086 |
anne | -0.086 |
ash | -0.085 |
Jamie | -0.084 |
icky | -0.084 |
Token | She |
Token ID | 3347 |
Feature activation | +2.54e+00 |
Pos logit contributions | |
cowboy | +0.071 |
stronger | +0.066 |
fuller | +0.061 |
captain | +0.060 |
nicer | +0.059 |
Neg logit contributions | |
Jamie | -0.122 |
icky | -0.122 |
jo | -0.122 |
anne | -0.121 |
umble | -0.118 |
Token | wants |
Token ID | 3382 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
stronger | +0.103 |
cowboy | +0.100 |
generous | +0.098 |
helpful | +0.096 |
vet | +0.096 |
Neg logit contributions | |
jo | -0.083 |
icky | -0.082 |
anne | -0.082 |
Jamie | -0.080 |
ash | -0.078 |
Token | feather |
Token ID | 17972 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
Thank | +0.029 |
cowboy | +0.029 |
stronger | +0.029 |
Days | +0.028 |
please | +0.028 |
Neg logit contributions | |
jo | -0.053 |
Jamie | -0.052 |
anne | -0.052 |
icky | -0.052 |
Today | -0.047 |
Token | � |
Token ID | 6184 |
Feature activation | -3.95e+00 |
Pos logit contributions | |
stronger | +0.073 |
cowboy | +0.071 |
generous | +0.070 |
helpful | +0.069 |
Thank | +0.069 |
Neg logit contributions | |
jo | -0.053 |
Jamie | -0.052 |
icky | -0.052 |
anne | -0.052 |
ash | -0.049 |
Token | � |
Token ID | 95 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
jo | +0.146 |
anne | +0.142 |
Jamie | +0.141 |
icky | +0.141 |
ash | +0.135 |
Neg logit contributions | |
stronger | -0.143 |
cowboy | -0.141 |
generous | -0.137 |
helpful | -0.135 |
Thank | -0.135 |
Token | € |
Token ID | 26391 |
Feature activation | -5.16e+00 |
Pos logit contributions | |
jo | +0.047 |
icky | +0.047 |
Jamie | +0.046 |
anne | +0.046 |
bie | +0.044 |
Neg logit contributions | |
stronger | -0.049 |
Thank | -0.048 |
cowboy | -0.048 |
generous | -0.047 |
helpful | -0.046 |
Token | � |
Token ID | 447 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
jo | +0.189 |
Jamie | +0.186 |
anne | +0.185 |
icky | +0.185 |
ash | +0.171 |
Neg logit contributions | |
stronger | -0.191 |
cowboy | -0.185 |
generous | -0.180 |
helpful | -0.177 |
Thank | -0.176 |
Token | � |
Token ID | 250 |
Feature activation | +6.50e+00 |
Pos logit contributions | |
cowboy | +0.117 |
vet | +0.108 |
Days | +0.107 |
captain | +0.106 |
stronger | +0.106 |
Neg logit contributions | |
icky | -0.076 |
Jamie | -0.074 |
Today | -0.071 |
jo | -0.069 |
anne | -0.069 |
Token | the |
Token ID | 262 |
Feature activation | +4.25e-01 |
Pos logit contributions | |
stronger | +0.165 |
cowboy | +0.164 |
generous | +0.156 |
helpful | +0.154 |
vet | +0.147 |
Neg logit contributions | |
jo | -0.306 |
Jamie | -0.303 |
icky | -0.303 |
anne | -0.300 |
ash | -0.295 |
Token | same |
Token ID | 976 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
stronger | +0.012 |
cowboy | +0.011 |
Thank | +0.011 |
Days | +0.011 |
generous | +0.011 |
Neg logit contributions | |
jo | -0.019 |
Jamie | -0.019 |
icky | -0.019 |
anne | -0.018 |
ash | -0.017 |
Token | kind |
Token ID | 1611 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
jo | +0.016 |
icky | +0.016 |
Jamie | +0.016 |
anne | +0.016 |
ash | +0.016 |
Neg logit contributions | |
cowboy | -0.010 |
stronger | -0.010 |
generous | -0.009 |
helpful | -0.009 |
vet | -0.008 |
Token | of |
Token ID | 286 |
Feature activation | -6.39e-01 |
Pos logit contributions | |
generous | +0.065 |
captain | +0.065 |
helpful | +0.064 |
cowboy | +0.064 |
kind | +0.061 |
Neg logit contributions | |
ash | -0.114 |
icky | -0.114 |
jo | -0.114 |
anne | -0.113 |
Jamie | -0.111 |
Token | feather |
Token ID | 17972 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
jo | +0.027 |
icky | +0.026 |
Jamie | +0.026 |
anne | +0.026 |
ash | +0.025 |
Neg logit contributions | |
stronger | -0.020 |
cowboy | -0.019 |
generous | -0.019 |
helpful | -0.019 |
Thank | -0.018 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
Offic | +0.055 |
please | +0.054 |
Did | +0.054 |
fuller | +0.053 |
Days | +0.052 |
Neg logit contributions | |
icky | -0.046 |
Today | -0.043 |
fault | -0.042 |
Jamie | -0.039 |
ash | -0.037 |
Token | came |
Token ID | 1625 |
Feature activation | +3.89e+00 |
Pos logit contributions | |
Jamie | +0.005 |
jo | +0.005 |
anne | +0.005 |
icky | +0.005 |
ash | +0.005 |
Neg logit contributions | |
stronger | -0.004 |
Days | -0.004 |
cowboy | -0.004 |
generous | -0.004 |
fuller | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
stronger | +0.137 |
generous | +0.128 |
cowboy | +0.128 |
helpful | +0.127 |
Thank | +0.122 |
Neg logit contributions | |
jo | -0.153 |
anne | -0.152 |
Jamie | -0.149 |
icky | -0.147 |
ash | -0.140 |
Token | said |
Token ID | 531 |
Feature activation | +7.19e-01 |
Pos logit contributions | |
Days | +0.077 |
cowboy | +0.071 |
stronger | +0.070 |
vet | +0.069 |
fuller | +0.068 |
Neg logit contributions | |
Jamie | -0.072 |
bie | -0.066 |
anne | -0.066 |
jo | -0.064 |
Jo | -0.064 |
Token | , |
Token ID | 11 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
stronger | +0.028 |
cowboy | +0.027 |
generous | +0.027 |
helpful | +0.027 |
Thank | +0.026 |
Neg logit contributions | |
jo | -0.026 |
anne | -0.025 |
icky | -0.024 |
Jamie | -0.024 |
ash | -0.024 |
Token | " |
Token ID | 366 |
Feature activation | +8.38e+00 |
Pos logit contributions | |
stronger | +0.100 |
fuller | +0.100 |
Days | +0.099 |
cowboy | +0.098 |
nicer | +0.095 |
Neg logit contributions | |
Jamie | -0.097 |
icky | -0.093 |
jo | -0.091 |
anne | -0.089 |
Jo | -0.087 |
Token | It |
Token ID | 1026 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
stronger | +0.221 |
cowboy | +0.205 |
generous | +0.203 |
helpful | +0.201 |
captain | +0.195 |
Neg logit contributions | |
jo | -0.398 |
anne | -0.385 |
Jamie | -0.383 |
icky | -0.370 |
ash | -0.370 |
Token | 's |
Token ID | 338 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
stronger | +0.061 |
cowboy | +0.060 |
generous | +0.058 |
Thank | +0.058 |
helpful | +0.058 |
Neg logit contributions | |
jo | -0.042 |
icky | -0.041 |
Jamie | -0.041 |
anne | -0.041 |
ash | -0.038 |
Token | time |
Token ID | 640 |
Feature activation | -4.29e+00 |
Pos logit contributions | |
stronger | +0.005 |
cowboy | +0.005 |
generous | +0.005 |
helpful | +0.005 |
Thank | +0.004 |
Neg logit contributions | |
jo | -0.005 |
anne | -0.005 |
Jamie | -0.005 |
icky | -0.004 |
umble | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
anne | +0.220 |
jo | +0.216 |
loe | +0.209 |
icky | +0.208 |
umble | +0.204 |
Neg logit contributions | |
helpful | -0.102 |
generous | -0.099 |
stronger | -0.098 |
kind | -0.087 |
healthy | -0.086 |
Token | go |
Token ID | 467 |
Feature activation | +3.82e+00 |
Pos logit contributions | |
Did | +0.014 |
Mail | +0.013 |
cowboy | +0.012 |
fuller | +0.012 |
stronger | +0.012 |
Neg logit contributions | |
icky | -0.011 |
raw | -0.011 |
upid | -0.011 |
ash | -0.011 |
loe | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
stronger | +0.060 |
helpful | +0.050 |
generous | +0.049 |
golf | +0.048 |
nearby | +0.048 |
Neg logit contributions | |
anne | -0.096 |
jo | -0.095 |
bie | -0.088 |
ash | -0.087 |
obby | -0.086 |
Token | shouted |
Token ID | 17293 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
cowboy | +0.042 |
stronger | +0.041 |
generous | +0.040 |
helpful | +0.040 |
Days | +0.039 |
Neg logit contributions | |
Jamie | -0.056 |
jo | -0.056 |
icky | -0.055 |
anne | -0.055 |
ash | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
jo | +0.053 |
anne | +0.052 |
Jamie | +0.051 |
icky | +0.050 |
ash | +0.048 |
Neg logit contributions | |
stronger | -0.072 |
cowboy | -0.069 |
generous | -0.068 |
helpful | -0.067 |
Thank | -0.066 |
Token |
|
Token ID | 198 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
jo | +0.059 |
ash | +0.058 |
anne | +0.057 |
ared | +0.054 |
bie | +0.054 |
Neg logit contributions | |
stronger | -0.048 |
Thank | -0.047 |
Can | -0.046 |
vet | -0.046 |
Did | -0.045 |
Token |
|
Token ID | 198 |
Feature activation | +2.91e+00 |
Pos logit contributions | |
stronger | +0.075 |
helpful | +0.069 |
generous | +0.068 |
vet | +0.068 |
Days | +0.067 |
Neg logit contributions | |
jo | -0.109 |
anne | -0.109 |
ash | -0.107 |
Jamie | -0.105 |
ared | -0.102 |
Token | " |
Token ID | 1 |
Feature activation | +8.66e+00 |
Pos logit contributions | |
stronger | +0.053 |
cowboy | +0.050 |
generous | +0.048 |
helpful | +0.046 |
Thank | +0.046 |
Neg logit contributions | |
jo | -0.160 |
Jamie | -0.158 |
anne | -0.158 |
icky | -0.158 |
ash | -0.153 |
Token | Give |
Token ID | 23318 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
stronger | +0.271 |
cowboy | +0.265 |
generous | +0.260 |
helpful | +0.255 |
fuller | +0.244 |
Neg logit contributions | |
icky | -0.358 |
jo | -0.356 |
Jamie | -0.356 |
anne | -0.353 |
ash | -0.337 |
Token | me |
Token ID | 502 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
stronger | +0.038 |
cowboy | +0.036 |
generous | +0.034 |
helpful | +0.034 |
nicer | +0.033 |
Neg logit contributions | |
jo | -0.070 |
anne | -0.068 |
icky | -0.066 |
Jamie | -0.066 |
umble | -0.065 |
Token | the |
Token ID | 262 |
Feature activation | +6.64e-01 |
Pos logit contributions | |
jo | +0.026 |
anne | +0.025 |
Jamie | +0.024 |
icky | +0.024 |
umble | +0.023 |
Neg logit contributions | |
stronger | -0.017 |
generous | -0.016 |
cowboy | -0.016 |
helpful | -0.016 |
nicer | -0.015 |
Token | ring |
Token ID | 5858 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
stronger | +0.013 |
cowboy | +0.012 |
Days | +0.012 |
generous | +0.012 |
Thank | +0.012 |
Neg logit contributions | |
jo | -0.034 |
Jamie | -0.034 |
anne | -0.034 |
icky | -0.034 |
ash | -0.033 |
Token | , |
Token ID | 11 |
Feature activation | +6.74e-01 |
Pos logit contributions | |
stronger | +0.055 |
generous | +0.050 |
helpful | +0.049 |
nicer | +0.048 |
cowboy | +0.048 |
Neg logit contributions | |
jo | -0.055 |
anne | -0.054 |
Jamie | -0.053 |
icky | -0.053 |
loe | -0.050 |
Token | a |
Token ID | 257 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
jo | +0.004 |
Jamie | +0.004 |
anne | +0.003 |
icky | +0.003 |
ash | +0.003 |
Neg logit contributions | |
stronger | -0.002 |
cowboy | -0.002 |
generous | -0.002 |
Days | -0.002 |
Thank | -0.002 |
Token | smile |
Token ID | 8212 |
Feature activation | +8.95e-01 |
Pos logit contributions | |
Days | +0.065 |
Thank | +0.062 |
stronger | +0.060 |
please | +0.059 |
fuller | +0.056 |
Neg logit contributions | |
jo | -0.108 |
Jamie | -0.108 |
icky | -0.107 |
anne | -0.107 |
bie | -0.106 |
Token | . |
Token ID | 13 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
cowboy | +0.039 |
stronger | +0.038 |
generous | +0.037 |
helpful | +0.036 |
Thank | +0.036 |
Neg logit contributions | |
Jamie | -0.026 |
anne | -0.026 |
jo | -0.026 |
icky | -0.026 |
ash | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
jo | +0.114 |
anne | +0.109 |
ash | +0.109 |
ert | +0.107 |
umble | +0.107 |
Neg logit contributions | |
Can | -0.061 |
Thank | -0.057 |
stronger | -0.055 |
vet | -0.055 |
generous | -0.054 |
Token |
|
Token ID | 198 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
stronger | +0.059 |
generous | +0.057 |
cowboy | +0.056 |
Thank | +0.056 |
helpful | +0.055 |
Neg logit contributions | |
jo | -0.073 |
anne | -0.072 |
Jamie | -0.071 |
ash | -0.070 |
icky | -0.070 |
Token | " |
Token ID | 1 |
Feature activation | +8.42e+00 |
Pos logit contributions | |
stronger | +0.050 |
cowboy | +0.048 |
generous | +0.045 |
helpful | +0.045 |
fuller | +0.043 |
Neg logit contributions | |
icky | -0.116 |
Jamie | -0.116 |
jo | -0.115 |
anne | -0.115 |
ash | -0.110 |
Token | Hello |
Token ID | 15496 |
Feature activation | +4.78e-01 |
Pos logit contributions | |
stronger | +0.271 |
helpful | +0.260 |
generous | +0.255 |
cowboy | +0.249 |
smarter | +0.247 |
Neg logit contributions | |
Jamie | -0.348 |
icky | -0.331 |
jo | -0.328 |
anne | -0.323 |
Today | -0.310 |
Token | , |
Token ID | 11 |
Feature activation | +9.04e-01 |
Pos logit contributions | |
stronger | +0.018 |
cowboy | +0.017 |
generous | +0.017 |
helpful | +0.017 |
nicer | +0.016 |
Neg logit contributions | |
jo | -0.017 |
anne | -0.017 |
ash | -0.016 |
icky | -0.016 |
umble | -0.016 |
Token | Mr |
Token ID | 1770 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
Days | +0.030 |
stronger | +0.030 |
Thank | +0.029 |
please | +0.029 |
cowboy | +0.028 |
Neg logit contributions | |
Jamie | -0.036 |
icky | -0.035 |
jo | -0.035 |
anne | -0.033 |
ash | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
anne | +0.038 |
ash | +0.038 |
jo | +0.038 |
loe | +0.037 |
Jamie | +0.037 |
Neg logit contributions | |
stronger | -0.042 |
vet | -0.041 |
cowboy | -0.040 |
generous | -0.040 |
helpful | -0.039 |
Token | Snow |
Token ID | 7967 |
Feature activation | +5.84e-01 |
Pos logit contributions | |
Jamie | +0.040 |
jo | +0.040 |
icky | +0.040 |
anne | +0.038 |
obby | +0.036 |
Neg logit contributions | |
stronger | -0.041 |
generous | -0.040 |
cowboy | -0.039 |
Days | -0.039 |
helpful | -0.039 |
Token | angry |
Token ID | 7954 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
jo | +0.053 |
anne | +0.052 |
icky | +0.052 |
Jamie | +0.052 |
ash | +0.050 |
Neg logit contributions | |
stronger | -0.032 |
generous | -0.030 |
helpful | -0.029 |
cowboy | -0.029 |
Thank | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -9.76e-01 |
Pos logit contributions | |
stronger | +0.080 |
generous | +0.076 |
helpful | +0.074 |
nicer | +0.072 |
Thank | +0.072 |
Neg logit contributions | |
anne | -0.067 |
jo | -0.066 |
Jamie | -0.063 |
icky | -0.061 |
loe | -0.060 |
Token | He |
Token ID | 679 |
Feature activation | +3.72e+00 |
Pos logit contributions | |
ash | +0.042 |
jo | +0.042 |
anne | +0.042 |
ert | +0.042 |
bie | +0.041 |
Neg logit contributions | |
stronger | -0.031 |
Thank | -0.030 |
generous | -0.030 |
Can | -0.029 |
vet | -0.028 |
Token | says |
Token ID | 1139 |
Feature activation | +7.61e-01 |
Pos logit contributions | |
stronger | +0.167 |
cowboy | +0.164 |
generous | +0.160 |
Thank | +0.159 |
helpful | +0.159 |
Neg logit contributions | |
jo | -0.103 |
Jamie | -0.101 |
anne | -0.100 |
icky | -0.100 |
ash | -0.093 |
Token | : |
Token ID | 25 |
Feature activation | +3.10e+00 |
Pos logit contributions | |
stronger | +0.041 |
cowboy | +0.040 |
Thank | +0.039 |
generous | +0.039 |
helpful | +0.039 |
Neg logit contributions | |
icky | -0.014 |
jo | -0.014 |
Jamie | -0.014 |
anne | -0.014 |
ash | -0.012 |
Token | " |
Token ID | 366 |
Feature activation | +8.81e+00 |
Pos logit contributions | |
cowboy | +0.104 |
stronger | +0.102 |
Days | +0.101 |
fuller | +0.098 |
helpful | +0.097 |
Neg logit contributions | |
icky | -0.121 |
Jamie | -0.119 |
jo | -0.115 |
anne | -0.114 |
Today | -0.111 |
Token | You |
Token ID | 1639 |
Feature activation | -4.10e+00 |
Pos logit contributions | |
stronger | +0.202 |
cowboy | +0.193 |
generous | +0.186 |
helpful | +0.182 |
Thank | +0.180 |
Neg logit contributions | |
jo | -0.443 |
Jamie | -0.438 |
icky | -0.438 |
anne | -0.437 |
ash | -0.420 |
Token | should |
Token ID | 815 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
jo | +0.145 |
anne | +0.143 |
icky | +0.139 |
Jamie | +0.138 |
ash | +0.134 |
Neg logit contributions | |
stronger | -0.159 |
cowboy | -0.158 |
generous | -0.155 |
helpful | -0.155 |
Thank | -0.151 |
Token | not |
Token ID | 407 |
Feature activation | +6.38e-03 |
Pos logit contributions | |
stronger | +0.037 |
cowboy | +0.035 |
generous | +0.035 |
helpful | +0.034 |
Thank | +0.033 |
Neg logit contributions | |
jo | -0.070 |
Jamie | -0.068 |
anne | -0.068 |
icky | -0.068 |
ash | -0.066 |
Token | be |
Token ID | 307 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
stronger | +0.000 |
helpful | +0.000 |
generous | +0.000 |
Thank | +0.000 |
vet | +0.000 |
Neg logit contributions | |
Jamie | -0.000 |
jo | -0.000 |
anne | -0.000 |
icky | -0.000 |
ash | -0.000 |
Token | here |
Token ID | 994 |
Feature activation | +3.01e+00 |
Pos logit contributions | |
stronger | +0.044 |
generous | +0.041 |
helpful | +0.041 |
cowboy | +0.039 |
Thank | +0.037 |
Neg logit contributions | |
jo | -0.108 |
Jamie | -0.106 |
anne | -0.106 |
icky | -0.106 |
ash | -0.101 |
Token | hungry |
Token ID | 14720 |
Feature activation | +3.75e+00 |
Pos logit contributions | |
Jamie | +0.029 |
icky | +0.028 |
anne | +0.027 |
jo | +0.027 |
bie | +0.026 |
Neg logit contributions | |
cowboy | -0.014 |
Days | -0.014 |
Thank | -0.013 |
Tweet | -0.013 |
Mic | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
Tweet | +0.173 |
Days | +0.171 |
Mic | +0.171 |
cowboy | +0.170 |
please | +0.169 |
Neg logit contributions | |
Jamie | -0.099 |
icky | -0.093 |
Today | -0.089 |
jo | -0.089 |
anne | -0.082 |
Token | excited |
Token ID | 6568 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
Days | +0.050 |
cowboy | +0.048 |
Mic | +0.046 |
vet | +0.046 |
Tweet | +0.045 |
Neg logit contributions | |
Jamie | -0.078 |
anne | -0.075 |
jo | -0.074 |
bie | -0.071 |
icky | -0.071 |
Token | . |
Token ID | 13 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
stronger | +0.087 |
generous | +0.084 |
helpful | +0.082 |
nicer | +0.081 |
se | +0.077 |
Neg logit contributions | |
loe | -0.098 |
icky | -0.097 |
jo | -0.096 |
anne | -0.096 |
raw | -0.093 |
Token | He |
Token ID | 679 |
Feature activation | +3.77e+00 |
Pos logit contributions | |
ash | +0.051 |
anne | +0.051 |
jo | +0.050 |
ared | +0.049 |
icky | +0.048 |
Neg logit contributions | |
Thank | -0.034 |
Can | -0.034 |
generous | -0.034 |
vet | -0.033 |
stronger | -0.033 |
Token | wondered |
Token ID | 14028 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
stronger | +0.131 |
Days | +0.125 |
cowboy | +0.124 |
fuller | +0.123 |
Thank | +0.121 |
Neg logit contributions | |
Jamie | -0.145 |
jo | -0.142 |
anne | -0.140 |
icky | -0.134 |
bie | -0.128 |
Token | what |
Token ID | 644 |
Feature activation | -8.90e-01 |
Pos logit contributions | |
stronger | +0.038 |
cowboy | +0.037 |
Thank | +0.036 |
generous | +0.036 |
Days | +0.036 |
Neg logit contributions | |
Jamie | -0.038 |
jo | -0.038 |
icky | -0.037 |
anne | -0.037 |
ash | -0.035 |
Token | he |
Token ID | 339 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
jo | +0.036 |
anne | +0.036 |
icky | +0.036 |
Jamie | +0.035 |
ash | +0.034 |
Neg logit contributions | |
stronger | -0.029 |
cowboy | -0.028 |
generous | -0.028 |
helpful | -0.027 |
Thank | -0.027 |
Token | would |
Token ID | 561 |
Feature activation | +3.60e+00 |
Pos logit contributions | |
stronger | +0.070 |
cowboy | +0.069 |
Days | +0.067 |
generous | +0.066 |
vet | +0.065 |
Neg logit contributions | |
jo | -0.076 |
Jamie | -0.074 |
anne | -0.073 |
icky | -0.071 |
ash | -0.068 |
Token | dream |
Token ID | 4320 |
Feature activation | +9.28e-01 |
Pos logit contributions | |
stronger | +0.122 |
cowboy | +0.120 |
fuller | +0.114 |
Days | +0.114 |
generous | +0.113 |
Neg logit contributions | |
Jamie | -0.136 |
jo | -0.134 |
icky | -0.134 |
anne | -0.131 |
Today | -0.126 |
Token | about |
Token ID | 546 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
stronger | +0.040 |
cowboy | +0.038 |
generous | +0.038 |
helpful | +0.038 |
Thank | +0.037 |
Neg logit contributions | |
jo | -0.028 |
Jamie | -0.028 |
anne | -0.028 |
icky | -0.028 |
ash | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | +4.65e-01 |
Pos logit contributions | |
stronger | +0.115 |
cowboy | +0.113 |
generous | +0.110 |
Thank | +0.109 |
helpful | +0.109 |
Neg logit contributions | |
jo | -0.083 |
Jamie | -0.081 |
icky | -0.081 |
anne | -0.081 |
ash | -0.076 |
Token | she |
Token ID | 673 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
stronger | +0.015 |
cowboy | +0.015 |
Days | +0.014 |
Thank | +0.014 |
generous | +0.014 |
Neg logit contributions | |
Jamie | -0.019 |
jo | -0.019 |
anne | -0.018 |
icky | -0.018 |
ash | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
jo | +0.008 |
Jamie | +0.008 |
icky | +0.008 |
anne | +0.008 |
ash | +0.007 |
Neg logit contributions | |
stronger | -0.006 |
cowboy | -0.005 |
generous | -0.005 |
helpful | -0.005 |
Thank | -0.005 |
Token | walking |
Token ID | 6155 |
Feature activation | +2.57e+00 |
Pos logit contributions | |
jo | +0.091 |
Jamie | +0.090 |
icky | +0.090 |
anne | +0.090 |
ash | +0.086 |
Neg logit contributions | |
stronger | -0.050 |
cowboy | -0.048 |
generous | -0.046 |
helpful | -0.046 |
Thank | -0.045 |
Token | along |
Token ID | 1863 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
stronger | +0.085 |
generous | +0.081 |
cowboy | +0.080 |
Thank | +0.079 |
helpful | +0.079 |
Neg logit contributions | |
icky | -0.107 |
jo | -0.106 |
anne | -0.106 |
Jamie | -0.103 |
umble | -0.100 |
Token | a |
Token ID | 257 |
Feature activation | +6.16e-01 |
Pos logit contributions | |
jo | +0.103 |
anne | +0.100 |
icky | +0.099 |
ash | +0.098 |
umble | +0.095 |
Neg logit contributions | |
stronger | -0.073 |
generous | -0.070 |
helpful | -0.068 |
Thank | -0.067 |
nearby | -0.066 |
Token | street |
Token ID | 4675 |
Feature activation | +7.12e-01 |
Pos logit contributions | |
stronger | +0.017 |
Thank | +0.017 |
cowboy | +0.017 |
fuller | +0.016 |
please | +0.016 |
Neg logit contributions | |
jo | -0.027 |
anne | -0.027 |
icky | -0.027 |
Jamie | -0.027 |
ash | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
stronger | +0.028 |
generous | +0.027 |
Thank | +0.026 |
helpful | +0.026 |
cowboy | +0.026 |
Neg logit contributions | |
icky | -0.025 |
jo | -0.025 |
anne | -0.024 |
Jamie | -0.024 |
ash | -0.022 |
Token | She |
Token ID | 1375 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
jo | +0.111 |
icky | +0.110 |
ash | +0.109 |
anny | +0.105 |
ert | +0.103 |
Neg logit contributions | |
Thank | -0.105 |
Can | -0.104 |
stronger | -0.100 |
vet | -0.098 |
generous | -0.098 |
Token | saw |
Token ID | 2497 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
stronger | +0.047 |
cowboy | +0.045 |
generous | +0.044 |
Thank | +0.043 |
helpful | +0.043 |
Neg logit contributions | |
Jamie | -0.076 |
jo | -0.076 |
anne | -0.074 |
icky | -0.072 |
ash | -0.070 |
Token | a |
Token ID | 257 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
stronger | +0.074 |
cowboy | +0.073 |
Thank | +0.072 |
please | +0.071 |
Days | +0.070 |
Neg logit contributions | |
Jamie | -0.075 |
jo | -0.069 |
icky | -0.069 |
Jo | -0.068 |
anne | -0.068 |
Token | so |
Token ID | 523 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
jo | +0.156 |
anne | +0.152 |
icky | +0.152 |
Jamie | +0.151 |
ash | +0.146 |
Neg logit contributions | |
stronger | -0.092 |
generous | -0.087 |
helpful | -0.086 |
cowboy | -0.082 |
nicer | -0.080 |
Token | wonderful |
Token ID | 7932 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
jo | +0.137 |
Jamie | +0.137 |
anne | +0.136 |
icky | +0.136 |
ash | +0.130 |
Neg logit contributions | |
stronger | -0.063 |
cowboy | -0.063 |
generous | -0.057 |
Thank | -0.057 |
vet | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
jo | +0.064 |
icky | +0.063 |
anne | +0.062 |
Jamie | +0.060 |
ash | +0.060 |
Neg logit contributions | |
stronger | -0.074 |
helpful | -0.072 |
generous | -0.072 |
cowboy | -0.071 |
Thank | -0.069 |
Token | |
Token ID | 220 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
anne | +0.077 |
ared | +0.077 |
ash | +0.076 |
jo | +0.076 |
icky | +0.075 |
Neg logit contributions | |
Thank | -0.057 |
generous | -0.053 |
There | -0.053 |
Did | -0.052 |
stronger | -0.052 |
Token |
|
Token ID | 198 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
jo | +0.127 |
anne | +0.125 |
ash | +0.118 |
Jamie | +0.117 |
ared | +0.115 |
Neg logit contributions | |
stronger | -0.079 |
generous | -0.071 |
helpful | -0.071 |
Thank | -0.070 |
vet | -0.070 |
Token |
|
Token ID | 198 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
cowboy | +0.096 |
stronger | +0.094 |
generous | +0.091 |
nicer | +0.090 |
fuller | +0.089 |
Neg logit contributions | |
icky | -0.071 |
jo | -0.070 |
Jamie | -0.069 |
anne | -0.066 |
umble | -0.062 |
Token | Everyone |
Token ID | 16190 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
stronger | +0.083 |
cowboy | +0.083 |
nicer | +0.083 |
fuller | +0.079 |
smarter | +0.076 |
Neg logit contributions | |
jo | -0.109 |
icky | -0.107 |
Jamie | -0.104 |
anne | -0.101 |
umble | -0.098 |
Token | in |
Token ID | 287 |
Feature activation | -5.14e+00 |
Pos logit contributions | |
icky | +0.081 |
jo | +0.080 |
anne | +0.077 |
Jamie | +0.077 |
ash | +0.077 |
Neg logit contributions | |
generous | -0.054 |
stronger | -0.054 |
helpful | -0.053 |
se | -0.051 |
Thank | -0.051 |
Token | the |
Token ID | 262 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
jo | +0.197 |
icky | +0.193 |
anne | +0.193 |
ash | +0.190 |
Jamie | +0.188 |
Neg logit contributions | |
stronger | -0.179 |
cowboy | -0.174 |
generous | -0.173 |
Thank | -0.171 |
helpful | -0.171 |
Token | garden |
Token ID | 11376 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
icky | +0.072 |
anne | +0.072 |
Jamie | +0.069 |
jo | +0.068 |
Today | +0.068 |
Neg logit contributions | |
cowboy | -0.062 |
stronger | -0.062 |
please | -0.059 |
fuller | -0.055 |
Did | -0.055 |
Token | was |
Token ID | 373 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
cowboy | +0.049 |
stronger | +0.049 |
Days | +0.046 |
generous | +0.046 |
helpful | +0.045 |
Neg logit contributions | |
Jamie | -0.051 |
jo | -0.050 |
anne | -0.050 |
icky | -0.050 |
ash | -0.046 |
Token | driver |
Token ID | 4639 |
Feature activation | +4.28e+00 |
Pos logit contributions | |
cowboy | +0.160 |
stronger | +0.154 |
Days | +0.151 |
please | +0.150 |
fuller | +0.147 |
Neg logit contributions | |
jo | -0.261 |
anne | -0.256 |
Jamie | -0.253 |
icky | -0.242 |
Jo | -0.231 |
Token | smiled |
Token ID | 13541 |
Feature activation | +6.96e-01 |
Pos logit contributions | |
stronger | +0.143 |
cowboy | +0.143 |
Days | +0.137 |
Thank | +0.136 |
generous | +0.136 |
Neg logit contributions | |
jo | -0.168 |
Jamie | -0.165 |
anne | -0.164 |
icky | -0.161 |
ash | -0.153 |
Token | and |
Token ID | 290 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
Days | +0.030 |
cowboy | +0.029 |
please | +0.029 |
Offic | +0.029 |
Tweet | +0.029 |
Neg logit contributions | |
jo | -0.019 |
Jamie | -0.019 |
icky | -0.019 |
Today | -0.018 |
anne | -0.018 |
Token | replied |
Token ID | 8712 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
Days | +0.063 |
stronger | +0.059 |
cowboy | +0.058 |
fuller | +0.057 |
generous | +0.057 |
Neg logit contributions | |
Jamie | -0.051 |
anne | -0.047 |
jo | -0.047 |
Jo | -0.045 |
Megan | -0.043 |
Token | " |
Token ID | 366 |
Feature activation | +8.14e+00 |
Pos logit contributions | |
icky | +0.007 |
Jamie | +0.007 |
jo | +0.007 |
anne | +0.007 |
ash | +0.006 |
Neg logit contributions | |
stronger | -0.006 |
cowboy | -0.006 |
Days | -0.006 |
Thank | -0.006 |
fuller | -0.006 |
Token | Hi |
Token ID | 17250 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
cowboy | +0.192 |
stronger | +0.187 |
generous | +0.171 |
helpful | +0.171 |
vet | +0.170 |
Neg logit contributions | |
jo | -0.403 |
anne | -0.396 |
Jamie | -0.391 |
icky | -0.382 |
umble | -0.381 |
Token | , |
Token ID | 11 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
stronger | +0.004 |
cowboy | +0.004 |
generous | +0.004 |
Thank | +0.004 |
helpful | +0.004 |
Neg logit contributions | |
jo | -0.003 |
Jamie | -0.003 |
icky | -0.003 |
anne | -0.003 |
ash | -0.003 |
Token | my |
Token ID | 616 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
please | +0.038 |
stronger | +0.037 |
Days | +0.037 |
cowboy | +0.037 |
Thank | +0.036 |
Neg logit contributions | |
Jamie | -0.035 |
jo | -0.033 |
icky | -0.032 |
anne | -0.032 |
Jo | -0.032 |
Token | name |
Token ID | 1438 |
Feature activation | +8.48e-01 |
Pos logit contributions | |
jo | +0.005 |
icky | +0.005 |
Jamie | +0.005 |
anne | +0.005 |
ash | +0.005 |
Neg logit contributions | |
stronger | -0.003 |
cowboy | -0.003 |
generous | -0.003 |
helpful | -0.003 |
Thank | -0.003 |
Token | is |
Token ID | 318 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
stronger | +0.039 |
generous | +0.037 |
cowboy | +0.037 |
helpful | +0.037 |
Thank | +0.037 |
Neg logit contributions | |
jo | -0.024 |
anne | -0.023 |
icky | -0.023 |
Jamie | -0.023 |
ash | -0.021 |
Token | Paul |
Token ID | 3362 |
Feature activation | -3.22e+00 |
Pos logit contributions | |
jo | +0.035 |
anne | +0.035 |
icky | +0.034 |
ared | +0.034 |
ash | +0.034 |
Neg logit contributions | |
cowboy | -0.043 |
stronger | -0.043 |
generous | -0.042 |
helpful | -0.042 |
Thank | -0.039 |
Token | curious |
Token ID | 11040 |
Feature activation | -4.33e-01 |
Pos logit contributions | |
Jamie | +0.073 |
icky | +0.072 |
anne | +0.071 |
jo | +0.069 |
umble | +0.069 |
Neg logit contributions | |
cowboy | -0.035 |
Days | -0.034 |
vet | -0.034 |
Did | -0.033 |
Offic | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
jo | +0.013 |
anne | +0.012 |
icky | +0.012 |
Jamie | +0.012 |
ash | +0.012 |
Neg logit contributions | |
stronger | -0.019 |
generous | -0.019 |
cowboy | -0.019 |
helpful | -0.018 |
nicer | -0.018 |
Token | He |
Token ID | 679 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
anne | +0.123 |
jo | +0.116 |
ash | +0.114 |
ared | +0.114 |
icky | +0.114 |
Neg logit contributions | |
stronger | -0.078 |
Thank | -0.077 |
Can | -0.077 |
vet | -0.076 |
generous | -0.074 |
Token | decided |
Token ID | 3066 |
Feature activation | +4.26e+00 |
Pos logit contributions | |
stronger | +0.092 |
fuller | +0.088 |
cowboy | +0.086 |
Thank | +0.086 |
generous | +0.085 |
Neg logit contributions | |
jo | -0.104 |
anne | -0.099 |
Jamie | -0.098 |
icky | -0.091 |
bie | -0.090 |
Token | to |
Token ID | 284 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
fuller | +0.194 |
Days | +0.194 |
Thank | +0.194 |
Tweet | +0.191 |
cowboy | +0.190 |
Neg logit contributions | |
Jamie | -0.106 |
jo | -0.097 |
anne | -0.093 |
Jo | -0.088 |
Mary | -0.085 |
Token | go |
Token ID | 467 |
Feature activation | +4.53e+00 |
Pos logit contributions | |
stronger | +0.067 |
cowboy | +0.064 |
Thank | +0.062 |
fuller | +0.062 |
generous | +0.061 |
Neg logit contributions | |
Jamie | -0.081 |
jo | -0.079 |
icky | -0.079 |
anne | -0.078 |
umble | -0.074 |
Token | looking |
Token ID | 2045 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
Days | +0.192 |
Thank | +0.188 |
cowboy | +0.185 |
Did | +0.184 |
fuller | +0.183 |
Neg logit contributions | |
jo | -0.131 |
ash | -0.127 |
Jamie | -0.126 |
icky | -0.126 |
Today | -0.125 |
Token | for |
Token ID | 329 |
Feature activation | -3.29e+00 |
Pos logit contributions | |
stronger | +0.040 |
generous | +0.039 |
helpful | +0.038 |
cowboy | +0.037 |
strong | +0.035 |
Neg logit contributions | |
anne | -0.040 |
icky | -0.039 |
jo | -0.039 |
Jamie | -0.038 |
loe | -0.038 |
Token | one |
Token ID | 530 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
Jamie | +0.112 |
jo | +0.101 |
icky | +0.099 |
Jo | +0.098 |
anne | +0.095 |
Neg logit contributions | |
Thank | -0.128 |
Tweet | -0.122 |
please | -0.121 |
Days | -0.121 |
cowboy | -0.120 |
Token | . |
Token ID | 13 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
jo | +0.069 |
anne | +0.068 |
Jamie | +0.067 |
icky | +0.067 |
ash | +0.062 |
Neg logit contributions | |
stronger | -0.076 |
cowboy | -0.072 |
generous | -0.072 |
helpful | -0.071 |
vet | -0.069 |
Token | After |
Token ID | 2293 |
Feature activation | -3.83e+00 |
Pos logit contributions | |
anne | +0.116 |
jo | +0.111 |
icky | +0.110 |
ash | +0.110 |
umble | +0.106 |
Neg logit contributions | |
stronger | -0.066 |
Thank | -0.064 |
vet | -0.064 |
Can | -0.062 |
generous | -0.062 |
Token | , |
Token ID | 11 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
stronger | +0.157 |
Days | +0.153 |
cowboy | +0.150 |
please | +0.149 |
Thank | +0.147 |
Neg logit contributions | |
Jamie | -0.186 |
jo | -0.181 |
anne | -0.178 |
icky | -0.178 |
Today | -0.169 |
Token | furry |
Token ID | 40857 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
Days | +0.039 |
please | +0.038 |
cowboy | +0.036 |
vet | +0.035 |
stronger | +0.035 |
Neg logit contributions | |
Jamie | -0.057 |
anne | -0.055 |
jo | -0.053 |
Today | -0.052 |
Megan | -0.051 |
Token | , |
Token ID | 11 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
stronger | +0.066 |
cowboy | +0.066 |
generous | +0.063 |
helpful | +0.062 |
Thank | +0.061 |
Neg logit contributions | |
jo | -0.108 |
icky | -0.107 |
anne | -0.107 |
Jamie | -0.106 |
ash | -0.103 |
Token | brown |
Token ID | 7586 |
Feature activation | +3.65e+00 |
Pos logit contributions | |
Days | +0.047 |
please | +0.045 |
cowboy | +0.044 |
stronger | +0.042 |
vet | +0.042 |
Neg logit contributions | |
Jamie | -0.059 |
anne | -0.056 |
jo | -0.055 |
Today | -0.053 |
Megan | -0.053 |
Token | dog |
Token ID | 3290 |
Feature activation | +5.37e-01 |
Pos logit contributions | |
stronger | +0.102 |
cowboy | +0.101 |
generous | +0.097 |
helpful | +0.095 |
Thank | +0.094 |
Neg logit contributions | |
jo | -0.164 |
icky | -0.162 |
anne | -0.162 |
Jamie | -0.161 |
ash | -0.155 |
Token | . |
Token ID | 13 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
stronger | +0.023 |
cowboy | +0.022 |
generous | +0.022 |
helpful | +0.022 |
vet | +0.021 |
Neg logit contributions | |
jo | -0.016 |
icky | -0.016 |
anne | -0.016 |
Jamie | -0.016 |
ash | -0.015 |
Token | He |
Token ID | 679 |
Feature activation | +3.10e+00 |
Pos logit contributions | |
jo | +0.039 |
anne | +0.038 |
icky | +0.038 |
ash | +0.037 |
Jamie | +0.036 |
Neg logit contributions | |
stronger | -0.048 |
vet | -0.048 |
Thank | -0.048 |
cowboy | -0.047 |
generous | -0.047 |
Token | had |
Token ID | 550 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
stronger | +0.102 |
cowboy | +0.099 |
generous | +0.095 |
Days | +0.094 |
fuller | +0.094 |
Neg logit contributions | |
Jamie | -0.125 |
jo | -0.125 |
anne | -0.123 |
icky | -0.121 |
ash | -0.113 |
Token | a |
Token ID | 257 |
Feature activation | +3.22e+00 |
Pos logit contributions | |
stronger | +0.041 |
cowboy | +0.040 |
Thank | +0.039 |
generous | +0.039 |
fuller | +0.038 |
Neg logit contributions | |
Jamie | -0.059 |
jo | -0.057 |
anne | -0.057 |
icky | -0.056 |
ash | -0.052 |
Token | collar |
Token ID | 19908 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
stronger | +0.094 |
Thank | +0.090 |
Days | +0.090 |
please | +0.088 |
fuller | +0.088 |
Neg logit contributions | |
Jamie | -0.135 |
jo | -0.134 |
icky | -0.134 |
anne | -0.134 |
Today | -0.131 |
Token | and |
Token ID | 290 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
stronger | +0.075 |
cowboy | +0.074 |
please | +0.073 |
Days | +0.073 |
Thank | +0.073 |
Neg logit contributions | |
jo | -0.063 |
Jamie | -0.062 |
anne | -0.061 |
icky | -0.061 |
ash | -0.057 |
Token | up |
Token ID | 510 |
Feature activation | -9.41e-01 |
Pos logit contributions | |
stronger | +0.119 |
cowboy | +0.116 |
generous | +0.113 |
Thank | +0.111 |
helpful | +0.111 |
Neg logit contributions | |
jo | -0.152 |
icky | -0.151 |
Jamie | -0.151 |
anne | -0.149 |
ash | -0.143 |
Token | an |
Token ID | 281 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
jo | +0.042 |
anne | +0.041 |
icky | +0.040 |
Jamie | +0.040 |
ash | +0.038 |
Neg logit contributions | |
stronger | -0.030 |
generous | -0.027 |
helpful | -0.027 |
cowboy | -0.027 |
vet | -0.026 |
Token | onion |
Token ID | 21670 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
jo | +0.086 |
icky | +0.085 |
Jamie | +0.085 |
anne | +0.084 |
ash | +0.082 |
Neg logit contributions | |
stronger | -0.064 |
cowboy | -0.063 |
Thank | -0.061 |
generous | -0.060 |
helpful | -0.059 |
Token | for |
Token ID | 329 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
stronger | +0.044 |
cowboy | +0.043 |
generous | +0.042 |
helpful | +0.042 |
Thank | +0.042 |
Neg logit contributions | |
jo | -0.033 |
Jamie | -0.032 |
anne | -0.032 |
icky | -0.032 |
ash | -0.030 |
Token | dinner |
Token ID | 8073 |
Feature activation | -5.09e-01 |
Pos logit contributions | |
icky | +0.123 |
Jamie | +0.123 |
jo | +0.120 |
anne | +0.119 |
ash | +0.113 |
Neg logit contributions | |
stronger | -0.113 |
cowboy | -0.111 |
Thank | -0.107 |
generous | -0.107 |
helpful | -0.104 |
Token | . |
Token ID | 13 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
jo | +0.014 |
anne | +0.013 |
Jamie | +0.013 |
icky | +0.013 |
ash | +0.012 |
Neg logit contributions | |
stronger | -0.023 |
cowboy | -0.023 |
generous | -0.023 |
helpful | -0.022 |
Thank | -0.022 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
jo | +0.069 |
icky | +0.067 |
anne | +0.067 |
Jamie | +0.067 |
ash | +0.064 |
Neg logit contributions | |
stronger | -0.076 |
cowboy | -0.073 |
Thank | -0.073 |
generous | -0.072 |
helpful | -0.071 |
Token | didn |
Token ID | 1422 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
icky | +0.079 |
anne | +0.078 |
Jamie | +0.078 |
jo | +0.078 |
bie | +0.073 |
Neg logit contributions | |
stronger | -0.080 |
cowboy | -0.079 |
fuller | -0.074 |
nicer | -0.073 |
Thank | -0.072 |
Token | 't |
Token ID | 470 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
jo | +0.012 |
anne | +0.012 |
Jamie | +0.012 |
ash | +0.011 |
icky | +0.011 |
Neg logit contributions | |
stronger | -0.009 |
cowboy | -0.008 |
generous | -0.008 |
helpful | -0.008 |
vet | -0.008 |
Token | like |
Token ID | 588 |
Feature activation | -7.30e-01 |
Pos logit contributions | |
jo | +0.007 |
Jamie | +0.006 |
anne | +0.006 |
icky | +0.006 |
ash | +0.006 |
Neg logit contributions | |
stronger | -0.006 |
cowboy | -0.005 |
generous | -0.005 |
helpful | -0.005 |
Thank | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | -7.18e-01 |
Pos logit contributions | |
jo | +0.029 |
Jamie | +0.029 |
icky | +0.029 |
anne | +0.029 |
ash | +0.027 |
Neg logit contributions | |
stronger | -0.024 |
cowboy | -0.024 |
generous | -0.023 |
helpful | -0.023 |
Thank | -0.023 |
Token | old |
Token ID | 727 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
jo | +0.018 |
anne | +0.017 |
Jamie | +0.017 |
icky | +0.017 |
ash | +0.015 |
Neg logit contributions | |
stronger | -0.041 |
cowboy | -0.041 |
Thank | -0.039 |
generous | -0.039 |
helpful | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | -3.36e+00 |
Pos logit contributions | |
generous | +0.042 |
cowboy | +0.042 |
helpful | +0.041 |
stronger | +0.041 |
vet | +0.039 |
Neg logit contributions | |
icky | -0.065 |
anne | -0.060 |
ash | -0.060 |
raw | -0.059 |
umble | -0.059 |
Token | She |
Token ID | 1375 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
ash | +0.137 |
ert | +0.134 |
jo | +0.133 |
icky | +0.130 |
anne | +0.128 |
Neg logit contributions | |
Thank | -0.105 |
store | -0.104 |
Can | -0.103 |
generous | -0.097 |
vet | -0.097 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
stronger | +0.042 |
cowboy | +0.041 |
generous | +0.040 |
helpful | +0.040 |
Thank | +0.039 |
Neg logit contributions | |
icky | -0.057 |
jo | -0.057 |
Jamie | -0.056 |
anne | -0.056 |
ash | -0.055 |
Token | her |
Token ID | 607 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
stronger | +0.058 |
cowboy | +0.056 |
generous | +0.054 |
helpful | +0.053 |
Thank | +0.053 |
Neg logit contributions | |
jo | -0.088 |
icky | -0.086 |
anne | -0.086 |
Jamie | -0.086 |
ash | -0.083 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
jo | +0.067 |
anne | +0.067 |
icky | +0.066 |
Jamie | +0.065 |
ash | +0.065 |
Neg logit contributions | |
generous | -0.044 |
stronger | -0.044 |
helpful | -0.044 |
cowboy | -0.043 |
vet | -0.042 |
Token | looking |
Token ID | 2045 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
jo | +0.041 |
Jamie | +0.040 |
anne | +0.040 |
icky | +0.039 |
ash | +0.037 |
Neg logit contributions | |
cowboy | -0.046 |
stronger | -0.046 |
Days | -0.044 |
Thank | -0.044 |
generous | -0.043 |
Token | at |
Token ID | 379 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
stronger | +0.044 |
generous | +0.044 |
helpful | +0.043 |
cowboy | +0.040 |
nearby | +0.040 |
Neg logit contributions | |
anne | -0.067 |
jo | -0.067 |
icky | -0.066 |
Jamie | -0.065 |
loe | -0.062 |
Token | a |
Token ID | 257 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
jo | +0.120 |
anne | +0.120 |
icky | +0.120 |
umble | +0.114 |
ash | +0.114 |
Neg logit contributions | |
stronger | -0.080 |
generous | -0.077 |
helpful | -0.077 |
store | -0.076 |
nearby | -0.075 |
Token | huge |
Token ID | 3236 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
stronger | +0.052 |
Thank | +0.052 |
Days | +0.051 |
cowboy | +0.051 |
please | +0.049 |
Neg logit contributions | |
Jamie | -0.077 |
jo | -0.077 |
anne | -0.077 |
icky | -0.077 |
bie | -0.073 |
Token | file |
Token ID | 2393 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
stronger | +0.071 |
Days | +0.070 |
Thank | +0.068 |
cowboy | +0.067 |
please | +0.066 |
Neg logit contributions | |
Jamie | -0.096 |
jo | -0.095 |
anne | -0.094 |
icky | -0.092 |
ash | -0.089 |
Token | . |
Token ID | 13 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
cowboy | +0.223 |
Days | +0.219 |
Thank | +0.213 |
Offic | +0.213 |
stronger | +0.212 |
Neg logit contributions | |
icky | -0.117 |
Jamie | -0.110 |
ash | -0.106 |
jo | -0.102 |
anne | -0.101 |
Token | She |
Token ID | 1375 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
jo | +0.073 |
anne | +0.070 |
ash | +0.067 |
icky | +0.066 |
Jamie | +0.064 |
Neg logit contributions | |
vet | -0.060 |
Thank | -0.058 |
stronger | -0.058 |
generous | -0.058 |
Can | -0.058 |
Token | asks |
Token ID | 7893 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
stronger | +0.102 |
cowboy | +0.099 |
generous | +0.098 |
helpful | +0.096 |
Thank | +0.096 |
Neg logit contributions | |
jo | -0.073 |
Jamie | -0.071 |
icky | -0.071 |
anne | -0.071 |
ash | -0.066 |
Token | her |
Token ID | 607 |
Feature activation | +4.43e-01 |
Pos logit contributions | |
cowboy | +0.106 |
stronger | +0.105 |
Thank | +0.102 |
Days | +0.102 |
generous | +0.100 |
Neg logit contributions | |
Jamie | -0.069 |
icky | -0.067 |
jo | -0.065 |
anne | -0.064 |
ash | -0.061 |
Token | mom |
Token ID | 1995 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
cowboy | +0.017 |
Thank | +0.017 |
stronger | +0.017 |
Days | +0.017 |
Tweet | +0.016 |
Neg logit contributions | |
Jamie | -0.015 |
icky | -0.015 |
jo | -0.014 |
anne | -0.014 |
bie | -0.014 |
Token | if |
Token ID | 611 |
Feature activation | -7.89e-01 |
Pos logit contributions | |
stronger | +0.041 |
cowboy | +0.040 |
generous | +0.039 |
Thank | +0.038 |
helpful | +0.038 |
Neg logit contributions | |
jo | -0.042 |
Jamie | -0.042 |
icky | -0.042 |
anne | -0.041 |
ash | -0.039 |
Token | they |
Token ID | 484 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
Jamie | +0.027 |
icky | +0.025 |
anne | +0.025 |
jo | +0.025 |
Megan | +0.023 |
Neg logit contributions | |
stronger | -0.030 |
cowboy | -0.030 |
Days | -0.029 |
Thank | -0.028 |
generous | -0.028 |
Token | can |
Token ID | 460 |
Feature activation | +3.81e+00 |
Pos logit contributions | |
stronger | +0.063 |
cowboy | +0.062 |
generous | +0.059 |
helpful | +0.058 |
Thank | +0.058 |
Neg logit contributions | |
icky | -0.083 |
Jamie | -0.082 |
jo | -0.082 |
anne | -0.081 |
ash | -0.078 |
Token | buy |
Token ID | 2822 |
Feature activation | +3.89e+00 |
Pos logit contributions | |
cowboy | +0.126 |
stronger | +0.124 |
Days | +0.118 |
generous | +0.117 |
Thank | +0.116 |
Neg logit contributions | |
icky | -0.153 |
Jamie | -0.147 |
jo | -0.145 |
anne | -0.144 |
ash | -0.142 |
Token | one |
Token ID | 530 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
cowboy | +0.147 |
Days | +0.147 |
Offic | +0.137 |
Mic | +0.135 |
stronger | +0.135 |
Neg logit contributions | |
icky | -0.128 |
Jamie | -0.127 |
ash | -0.124 |
raw | -0.120 |
anne | -0.115 |
Token | . |
Token ID | 13 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
jo | +0.043 |
anne | +0.041 |
Jamie | +0.040 |
icky | +0.038 |
loe | +0.038 |
Neg logit contributions | |
stronger | -0.052 |
generous | -0.049 |
helpful | -0.049 |
nicer | -0.047 |
vet | -0.047 |
Token | face |
Token ID | 1986 |
Feature activation | -3.47e+00 |
Pos logit contributions | |
jo | +0.066 |
icky | +0.066 |
anne | +0.065 |
Jamie | +0.065 |
ash | +0.064 |
Neg logit contributions | |
stronger | -0.047 |
generous | -0.046 |
cowboy | -0.045 |
helpful | -0.045 |
se | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | -7.33e-01 |
Pos logit contributions | |
jo | +0.096 |
icky | +0.096 |
anne | +0.094 |
Jamie | +0.094 |
ash | +0.089 |
Neg logit contributions | |
stronger | -0.157 |
cowboy | -0.151 |
generous | -0.150 |
helpful | -0.149 |
Thank | -0.147 |
Token | thanked |
Token ID | 26280 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
jo | +0.031 |
Jamie | +0.031 |
icky | +0.031 |
anne | +0.031 |
ash | +0.029 |
Neg logit contributions | |
stronger | -0.022 |
cowboy | -0.021 |
generous | -0.021 |
helpful | -0.021 |
Thank | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
stronger | +0.046 |
cowboy | +0.044 |
generous | +0.044 |
helpful | +0.044 |
Thank | +0.043 |
Neg logit contributions | |
icky | -0.047 |
jo | -0.047 |
anne | -0.046 |
Jamie | -0.045 |
ash | -0.045 |
Token | girl |
Token ID | 2576 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
cowboy | +0.029 |
stronger | +0.029 |
Days | +0.029 |
Tweet | +0.027 |
Thank | +0.027 |
Neg logit contributions | |
Jamie | -0.043 |
jo | -0.042 |
anne | -0.042 |
icky | -0.042 |
umble | -0.040 |
Token | for |
Token ID | 329 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
anne | +0.055 |
icky | +0.055 |
jo | +0.054 |
Jamie | +0.053 |
loe | +0.052 |
Neg logit contributions | |
stronger | -0.063 |
generous | -0.061 |
helpful | -0.058 |
cowboy | -0.056 |
nearby | -0.055 |
Token | being |
Token ID | 852 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
Jamie | +0.091 |
Jo | +0.082 |
Annie | +0.076 |
Kate | +0.076 |
Mary | +0.074 |
Neg logit contributions | |
Tweet | -0.103 |
Days | -0.102 |
Mic | -0.099 |
ju | -0.099 |
cowboy | -0.098 |
Token | so |
Token ID | 523 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
Days | +0.049 |
cowboy | +0.048 |
vet | +0.047 |
Tweet | +0.046 |
Offic | +0.046 |
Neg logit contributions | |
Jamie | -0.055 |
icky | -0.051 |
anne | -0.050 |
jo | -0.050 |
ash | -0.050 |
Token | playful |
Token ID | 34264 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
Jamie | +0.142 |
icky | +0.137 |
Jo | +0.134 |
anne | +0.133 |
jo | +0.131 |
Neg logit contributions | |
cowboy | -0.056 |
Mic | -0.053 |
vet | -0.052 |
please | -0.047 |
Did | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
anne | +0.030 |
icky | +0.030 |
jo | +0.030 |
Jamie | +0.029 |
ash | +0.027 |
Neg logit contributions | |
stronger | -0.044 |
cowboy | -0.043 |
generous | -0.043 |
helpful | -0.042 |
Thank | -0.041 |
Token | They |
Token ID | 1119 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
ash | +0.081 |
icky | +0.080 |
anne | +0.080 |
jo | +0.079 |
ared | +0.078 |
Neg logit contributions | |
Thank | -0.076 |
vet | -0.075 |
stronger | -0.074 |
generous | -0.073 |
Can | -0.071 |
Token | She |
Token ID | 1375 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
jo | +0.092 |
anne | +0.091 |
Jamie | +0.090 |
icky | +0.086 |
ash | +0.085 |
Neg logit contributions | |
stronger | -0.106 |
Thank | -0.105 |
generous | -0.103 |
helpful | -0.101 |
Can | -0.100 |
Token | loved |
Token ID | 6151 |
Feature activation | +5.85e-01 |
Pos logit contributions | |
stronger | +0.056 |
cowboy | +0.053 |
generous | +0.051 |
helpful | +0.050 |
Thank | +0.050 |
Neg logit contributions | |
jo | -0.078 |
icky | -0.077 |
anne | -0.076 |
Jamie | -0.076 |
ash | -0.072 |
Token | to |
Token ID | 284 |
Feature activation | +4.27e-01 |
Pos logit contributions | |
stronger | +0.024 |
Days | +0.023 |
fuller | +0.023 |
cowboy | +0.023 |
Thank | +0.023 |
Neg logit contributions | |
Jamie | -0.018 |
icky | -0.018 |
jo | -0.017 |
anne | -0.017 |
bie | -0.016 |
Token | go |
Token ID | 467 |
Feature activation | +3.91e+00 |
Pos logit contributions | |
stronger | +0.013 |
fuller | +0.013 |
Days | +0.013 |
Did | +0.013 |
Thank | +0.012 |
Neg logit contributions | |
Jamie | -0.018 |
icky | -0.017 |
jo | -0.017 |
anne | -0.017 |
Today | -0.017 |
Token | camping |
Token ID | 22498 |
Feature activation | +2.69e+00 |
Pos logit contributions | |
stronger | +0.147 |
Days | +0.146 |
Thank | +0.145 |
cowboy | +0.145 |
fuller | +0.141 |
Neg logit contributions | |
jo | -0.131 |
icky | -0.129 |
Jamie | -0.127 |
ash | -0.124 |
anne | -0.123 |
Token | with |
Token ID | 351 |
Feature activation | -6.25e+00 |
Pos logit contributions | |
stronger | +0.086 |
cowboy | +0.084 |
Thank | +0.083 |
Days | +0.082 |
generous | +0.082 |
Neg logit contributions | |
jo | -0.107 |
Jamie | -0.107 |
icky | -0.106 |
anne | -0.102 |
ash | -0.099 |
Token | her |
Token ID | 607 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
Jamie | +0.184 |
jo | +0.182 |
icky | +0.182 |
anne | +0.176 |
ash | +0.165 |
Neg logit contributions | |
stronger | -0.276 |
cowboy | -0.268 |
Thank | -0.263 |
generous | -0.261 |
fuller | -0.258 |
Token | family |
Token ID | 1641 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
jo | +0.127 |
icky | +0.126 |
Jamie | +0.125 |
anne | +0.123 |
ash | +0.117 |
Neg logit contributions | |
stronger | -0.089 |
cowboy | -0.084 |
Thank | -0.084 |
generous | -0.082 |
fuller | -0.081 |
Token | . |
Token ID | 13 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
anne | +0.043 |
jo | +0.042 |
Jamie | +0.041 |
icky | +0.040 |
umble | +0.038 |
Neg logit contributions | |
stronger | -0.056 |
generous | -0.056 |
helpful | -0.055 |
cowboy | -0.055 |
Thank | -0.052 |
Token | They |
Token ID | 1119 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
anne | +0.092 |
jo | +0.091 |
ared | +0.090 |
icky | +0.088 |
Jamie | +0.086 |
Neg logit contributions | |
Can | -0.092 |
store | -0.090 |
vet | -0.088 |
Thank | -0.088 |
cowboy | -0.086 |
Token | would |
Token ID | 561 |
Feature activation | +3.76e+00 |
Pos logit contributions | |
stronger | +0.061 |
vet | +0.059 |
nicer | +0.057 |
Thank | +0.057 |
cowboy | +0.056 |
Neg logit contributions | |
jo | -0.070 |
Jamie | -0.068 |
icky | -0.066 |
bie | -0.066 |
anne | -0.065 |
Token | the |
Token ID | 262 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
stronger | +0.070 |
cowboy | +0.069 |
generous | +0.067 |
helpful | +0.066 |
Thank | +0.064 |
Neg logit contributions | |
jo | -0.083 |
icky | -0.082 |
ash | -0.081 |
anne | -0.081 |
umble | -0.078 |
Token | sun |
Token ID | 4252 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
stronger | +0.055 |
Days | +0.054 |
cowboy | +0.053 |
Thank | +0.053 |
please | +0.053 |
Neg logit contributions | |
Jamie | -0.085 |
jo | -0.085 |
icky | -0.084 |
anne | -0.084 |
ash | -0.078 |
Token | came |
Token ID | 1625 |
Feature activation | +3.36e+00 |
Pos logit contributions | |
stronger | +0.050 |
cowboy | +0.048 |
generous | +0.047 |
helpful | +0.046 |
Thank | +0.046 |
Neg logit contributions | |
jo | -0.067 |
anne | -0.066 |
Jamie | -0.066 |
icky | -0.066 |
ash | -0.063 |
Token | out |
Token ID | 503 |
Feature activation | -2.97e+00 |
Pos logit contributions | |
cowboy | +0.130 |
stronger | +0.128 |
generous | +0.125 |
Thank | +0.125 |
helpful | +0.123 |
Neg logit contributions | |
icky | -0.113 |
jo | -0.112 |
Jamie | -0.111 |
anne | -0.110 |
ash | -0.104 |
Token | from |
Token ID | 422 |
Feature activation | -5.15e+00 |
Pos logit contributions | |
jo | +0.114 |
anne | +0.111 |
icky | +0.111 |
Jamie | +0.110 |
ash | +0.107 |
Neg logit contributions | |
stronger | -0.110 |
generous | -0.101 |
helpful | -0.101 |
Thank | -0.097 |
fuller | -0.096 |
Token | behind |
Token ID | 2157 |
Feature activation | -6.72e+00 |
Pos logit contributions | |
jo | +0.203 |
icky | +0.200 |
anne | +0.198 |
Jamie | +0.192 |
ash | +0.189 |
Neg logit contributions | |
stronger | -0.181 |
cowboy | -0.171 |
Thank | -0.170 |
generous | -0.169 |
helpful | -0.168 |
Token | a |
Token ID | 257 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
jo | +0.308 |
umble | +0.300 |
ash | +0.300 |
icky | +0.298 |
anne | +0.295 |
Neg logit contributions | |
stronger | -0.183 |
nearby | -0.178 |
helpful | -0.173 |
generous | -0.173 |
Thank | -0.173 |
Token | cloud |
Token ID | 6279 |
Feature activation | -5.14e-01 |
Pos logit contributions | |
cowboy | +0.036 |
Thank | +0.035 |
please | +0.035 |
stronger | +0.034 |
Days | +0.034 |
Neg logit contributions | |
anne | -0.063 |
jo | -0.062 |
Jamie | -0.062 |
icky | -0.062 |
Today | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
icky | +0.018 |
Jamie | +0.017 |
jo | +0.017 |
anne | +0.016 |
obby | +0.016 |
Neg logit contributions | |
stronger | -0.019 |
nearby | -0.019 |
se | -0.019 |
cowboy | -0.018 |
generous | -0.018 |
Token | It |
Token ID | 632 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
jo | +0.058 |
ash | +0.056 |
anne | +0.056 |
icky | +0.056 |
ared | +0.056 |
Neg logit contributions | |
vet | -0.064 |
stronger | -0.063 |
generous | -0.062 |
Thank | -0.062 |
cowboy | -0.060 |
Token | was |
Token ID | 373 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
stronger | +0.105 |
cowboy | +0.102 |
generous | +0.100 |
helpful | +0.099 |
Thank | +0.098 |
Neg logit contributions | |
jo | -0.119 |
Jamie | -0.118 |
icky | -0.118 |
anne | -0.117 |
ash | -0.112 |
Token | . |
Token ID | 13 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
stronger | +0.002 |
cowboy | +0.002 |
generous | +0.002 |
Thank | +0.002 |
helpful | +0.002 |
Neg logit contributions | |
jo | -0.001 |
icky | -0.001 |
anne | -0.001 |
Jamie | -0.001 |
ash | -0.001 |
Token | Kitty |
Token ID | 21168 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
jo | +0.068 |
Jamie | +0.068 |
icky | +0.067 |
anne | +0.067 |
ash | +0.065 |
Neg logit contributions | |
stronger | -0.088 |
cowboy | -0.085 |
Thank | -0.085 |
generous | -0.085 |
helpful | -0.085 |
Token | liked |
Token ID | 8288 |
Feature activation | +4.87e-01 |
Pos logit contributions | |
jo | +0.108 |
Jamie | +0.107 |
anne | +0.106 |
icky | +0.106 |
ash | +0.103 |
Neg logit contributions | |
stronger | -0.079 |
generous | -0.076 |
cowboy | -0.075 |
helpful | -0.075 |
Thank | -0.075 |
Token | to |
Token ID | 284 |
Feature activation | +9.34e-01 |
Pos logit contributions | |
stronger | +0.020 |
fuller | +0.020 |
Thank | +0.020 |
cowboy | +0.019 |
Days | +0.019 |
Neg logit contributions | |
icky | -0.015 |
Jamie | -0.015 |
jo | -0.014 |
anne | -0.014 |
umble | -0.013 |
Token | slide |
Token ID | 10649 |
Feature activation | +4.71e-01 |
Pos logit contributions | |
stronger | +0.028 |
fuller | +0.028 |
Did | +0.027 |
Days | +0.027 |
cowboy | +0.027 |
Neg logit contributions | |
jo | -0.038 |
Jamie | -0.037 |
icky | -0.037 |
anne | -0.037 |
umble | -0.036 |
Token | on |
Token ID | 319 |
Feature activation | -5.86e+00 |
Pos logit contributions | |
stronger | +0.014 |
nearby | +0.012 |
golf | +0.012 |
helpful | +0.012 |
Thank | +0.012 |
Neg logit contributions | |
anne | -0.023 |
jo | -0.023 |
icky | -0.022 |
Jamie | -0.022 |
obby | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
jo | +0.213 |
Jamie | +0.209 |
icky | +0.209 |
anne | +0.209 |
ash | +0.198 |
Neg logit contributions | |
stronger | -0.217 |
cowboy | -0.210 |
generous | -0.206 |
helpful | -0.203 |
Thank | -0.202 |
Token | floor |
Token ID | 4314 |
Feature activation | -3.18e+00 |
Pos logit contributions | |
jo | +0.065 |
anne | +0.064 |
Jamie | +0.064 |
icky | +0.063 |
ash | +0.060 |
Neg logit contributions | |
stronger | -0.053 |
cowboy | -0.051 |
generous | -0.050 |
helpful | -0.050 |
Thank | -0.049 |
Token | with |
Token ID | 351 |
Feature activation | -6.37e+00 |
Pos logit contributions | |
icky | +0.094 |
Jamie | +0.093 |
jo | +0.093 |
anne | +0.091 |
ash | +0.087 |
Neg logit contributions | |
cowboy | -0.137 |
stronger | -0.136 |
generous | -0.130 |
Thank | -0.130 |
helpful | -0.129 |
Token | the |
Token ID | 262 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
Jamie | +0.193 |
icky | +0.192 |
jo | +0.188 |
anne | +0.184 |
ash | +0.173 |
Neg logit contributions | |
stronger | -0.273 |
cowboy | -0.270 |
generous | -0.260 |
Thank | -0.259 |
fuller | -0.256 |
Token | er |
Token ID | 1931 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
anne | +0.057 |
jo | +0.056 |
Jamie | +0.055 |
icky | +0.053 |
ash | +0.052 |
Neg logit contributions | |
stronger | -0.043 |
generous | -0.041 |
cowboy | -0.041 |
helpful | -0.041 |
Thank | -0.039 |
Token | She |
Token ID | 1375 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
jo | +0.074 |
anne | +0.073 |
icky | +0.069 |
Jamie | +0.069 |
ash | +0.069 |
Neg logit contributions | |
stronger | -0.078 |
Thank | -0.075 |
vet | -0.074 |
cowboy | -0.074 |
generous | -0.074 |
Token | felt |
Token ID | 2936 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
stronger | +0.061 |
cowboy | +0.058 |
generous | +0.057 |
helpful | +0.056 |
se | +0.056 |
Neg logit contributions | |
jo | -0.091 |
icky | -0.089 |
anne | -0.089 |
Jamie | -0.087 |
ash | -0.086 |
Token | something |
Token ID | 1223 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
jo | +0.015 |
Jamie | +0.015 |
icky | +0.015 |
anne | +0.015 |
ash | +0.014 |
Neg logit contributions | |
stronger | -0.011 |
cowboy | -0.011 |
generous | -0.010 |
helpful | -0.010 |
Thank | -0.010 |
Token | slim |
Token ID | 18862 |
Feature activation | +5.04e-01 |
Pos logit contributions | |
jo | +0.102 |
Jamie | +0.099 |
anne | +0.099 |
icky | +0.098 |
ash | +0.095 |
Neg logit contributions | |
stronger | -0.052 |
helpful | -0.048 |
generous | -0.048 |
cowboy | -0.046 |
se | -0.045 |
Token | y |
Token ID | 88 |
Feature activation | -7.23e-01 |
Pos logit contributions | |
stronger | +0.023 |
nearby | +0.022 |
Thank | +0.021 |
Can | +0.021 |
cowboy | +0.021 |
Neg logit contributions | |
jo | -0.014 |
anne | -0.014 |
Jamie | -0.014 |
loe | -0.013 |
icky | -0.013 |
Token | on |
Token ID | 319 |
Feature activation | -5.41e+00 |
Pos logit contributions | |
jo | +0.026 |
anne | +0.026 |
Jamie | +0.025 |
icky | +0.025 |
ash | +0.023 |
Neg logit contributions | |
stronger | -0.028 |
cowboy | -0.026 |
generous | -0.026 |
helpful | -0.026 |
Thank | -0.025 |
Token | her |
Token ID | 607 |
Feature activation | -2.71e+00 |
Pos logit contributions | |
jo | +0.250 |
anne | +0.239 |
ash | +0.232 |
obby | +0.231 |
Jamie | +0.220 |
Neg logit contributions | |
stronger | -0.168 |
cowboy | -0.159 |
vet | -0.151 |
helpful | -0.151 |
generous | -0.150 |
Token | fingers |
Token ID | 9353 |
Feature activation | -4.70e+00 |
Pos logit contributions | |
jo | +0.120 |
anne | +0.119 |
ash | +0.114 |
Jamie | +0.112 |
obby | +0.111 |
Neg logit contributions | |
stronger | -0.083 |
cowboy | -0.079 |
generous | -0.078 |
helpful | -0.077 |
se | -0.076 |
Token | . |
Token ID | 13 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
jo | +0.146 |
anne | +0.143 |
Jamie | +0.139 |
icky | +0.136 |
ash | +0.131 |
Neg logit contributions | |
stronger | -0.204 |
generous | -0.188 |
helpful | -0.187 |
cowboy | -0.186 |
Thank | -0.183 |
Token | She |
Token ID | 1375 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
jo | +0.083 |
anne | +0.081 |
ash | +0.077 |
Jamie | +0.077 |
icky | +0.076 |
Neg logit contributions | |
stronger | -0.091 |
Thank | -0.089 |
vet | -0.087 |
generous | -0.086 |
Can | -0.086 |
Token | pulled |
Token ID | 5954 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
stronger | +0.053 |
cowboy | +0.051 |
generous | +0.050 |
helpful | +0.049 |
Thank | +0.049 |
Neg logit contributions | |
jo | -0.075 |
icky | -0.074 |
Jamie | -0.074 |
anne | -0.074 |
ash | -0.070 |
Token | not |
Token ID | 407 |
Feature activation | +3.00e-02 |
Pos logit contributions | |
cowboy | +0.028 |
stronger | +0.027 |
generous | +0.026 |
Thank | +0.026 |
helpful | +0.026 |
Neg logit contributions | |
Jamie | -0.047 |
icky | -0.047 |
jo | -0.047 |
anne | -0.046 |
ash | -0.044 |
Token | care |
Token ID | 1337 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
stronger | +0.001 |
cowboy | +0.001 |
generous | +0.001 |
helpful | +0.001 |
Thank | +0.001 |
Neg logit contributions | |
jo | -0.001 |
Jamie | -0.001 |
icky | -0.001 |
anne | -0.001 |
ash | -0.001 |
Token | about |
Token ID | 546 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
anne | +0.052 |
Jamie | +0.051 |
jo | +0.049 |
bie | +0.047 |
ash | +0.046 |
Neg logit contributions | |
stronger | -0.065 |
helpful | -0.062 |
generous | -0.061 |
cowboy | -0.058 |
nearby | -0.057 |
Token | the |
Token ID | 262 |
Feature activation | -4.36e-01 |
Pos logit contributions | |
jo | +0.044 |
anne | +0.043 |
icky | +0.043 |
Jamie | +0.042 |
ash | +0.041 |
Neg logit contributions | |
stronger | -0.045 |
cowboy | -0.043 |
generous | -0.043 |
helpful | -0.042 |
Thank | -0.042 |
Token | ring |
Token ID | 5858 |
Feature activation | +9.86e-01 |
Pos logit contributions | |
jo | +0.017 |
anne | +0.017 |
icky | +0.017 |
Today | +0.017 |
Jamie | +0.016 |
Neg logit contributions | |
cowboy | -0.013 |
stronger | -0.012 |
fuller | -0.012 |
Days | -0.012 |
please | -0.012 |
Token | anymore |
Token ID | 7471 |
Feature activation | -6.26e+00 |
Pos logit contributions | |
stronger | +0.042 |
cowboy | +0.039 |
generous | +0.039 |
helpful | +0.039 |
vet | +0.038 |
Neg logit contributions | |
jo | -0.031 |
icky | -0.031 |
Jamie | -0.031 |
anne | -0.031 |
ash | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
anne | +0.239 |
raw | +0.232 |
jo | +0.232 |
ash | +0.232 |
Jamie | +0.224 |
Neg logit contributions | |
stronger | -0.224 |
helpful | -0.210 |
generous | -0.210 |
cowboy | -0.206 |
nearby | -0.206 |
Token | She |
Token ID | 1375 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
anne | +0.072 |
ash | +0.072 |
jo | +0.071 |
Jamie | +0.068 |
icky | +0.067 |
Neg logit contributions | |
Thank | -0.058 |
vet | -0.055 |
stronger | -0.054 |
Can | -0.053 |
generous | -0.053 |
Token | realized |
Token ID | 6939 |
Feature activation | +8.27e-01 |
Pos logit contributions | |
stronger | +0.080 |
cowboy | +0.079 |
fuller | +0.076 |
generous | +0.076 |
Days | +0.075 |
Neg logit contributions | |
jo | -0.078 |
Jamie | -0.077 |
anne | -0.076 |
icky | -0.072 |
bie | -0.069 |
Token | it |
Token ID | 340 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
stronger | +0.023 |
cowboy | +0.021 |
generous | +0.021 |
helpful | +0.021 |
vet | +0.020 |
Neg logit contributions | |
jo | -0.039 |
anne | -0.039 |
icky | -0.039 |
Jamie | -0.038 |
ash | -0.038 |
Token | was |
Token ID | 373 |
Feature activation | -9.85e-01 |
Pos logit contributions | |
stronger | +0.043 |
helpful | +0.040 |
generous | +0.038 |
nearby | +0.037 |
Can | +0.036 |
Neg logit contributions | |
Jamie | -0.074 |
raw | -0.073 |
anne | -0.072 |
jo | -0.071 |
ash | -0.071 |