Token | ice |
Token ID | 4771 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
protective | +0.022 |
sneaky | +0.022 |
fast | +0.021 |
brave | +0.020 |
ggy | +0.019 |
Neg logit contributions | |
aman | -0.033 |
umm | -0.031 |
en | -0.030 |
berry | -0.030 |
rap | -0.029 |
Token | cream |
Token ID | 8566 |
Feature activation | +2.57e-03 |
Pos logit contributions | |
aman | +0.058 |
berry | +0.056 |
umm | +0.055 |
en | +0.055 |
rap | +0.053 |
Neg logit contributions | |
protective | -0.035 |
sneaky | -0.034 |
fast | -0.031 |
dodged | -0.030 |
villains | -0.030 |
Token | taste |
Token ID | 6938 |
Feature activation | +4.60e-02 |
Pos logit contributions | |
aman | +0.006 |
en | +0.005 |
rap | +0.005 |
berry | +0.005 |
umm | +0.005 |
Neg logit contributions | |
fast | -0.004 |
protective | -0.004 |
sneaky | -0.004 |
brave | -0.003 |
strong | -0.003 |
Token | even |
Token ID | 772 |
Feature activation | +5.27e-02 |
Pos logit contributions | |
aman | +0.101 |
berry | +0.095 |
umm | +0.093 |
en | +0.092 |
rap | +0.092 |
Neg logit contributions | |
protective | -0.067 |
sneaky | -0.065 |
fast | -0.063 |
brave | -0.059 |
dodged | -0.057 |
Token | y |
Token ID | 331 |
Feature activation | +9.89e-02 |
Pos logit contributions | |
aman | +0.113 |
berry | +0.108 |
umm | +0.106 |
en | +0.104 |
rap | +0.103 |
Neg logit contributions | |
protective | -0.079 |
sneaky | -0.078 |
fast | -0.074 |
brave | -0.070 |
dodged | -0.069 |
Token | umm |
Token ID | 13929 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
aman | +0.096 |
airport | +0.082 |
destination | +0.077 |
date | +0.073 |
neath | +0.072 |
Neg logit contributions | |
protective | -0.246 |
fast | -0.241 |
sneaky | -0.240 |
ages | -0.233 |
hog | -0.229 |
Token | ier |
Token ID | 959 |
Feature activation | +5.59e-02 |
Pos logit contributions | |
aman | +0.476 |
airport | +0.452 |
neath | +0.450 |
destination | +0.446 |
amaz | +0.437 |
Neg logit contributions | |
ages | -0.316 |
protective | -0.305 |
fast | -0.291 |
ars | -0.286 |
sneaky | -0.286 |
Token | ! |
Token ID | 0 |
Feature activation | +7.17e-03 |
Pos logit contributions | |
aman | +0.123 |
berry | +0.115 |
umm | +0.113 |
rap | +0.112 |
en | +0.111 |
Neg logit contributions | |
protective | -0.079 |
sneaky | -0.075 |
fast | -0.074 |
brave | -0.068 |
ages | -0.067 |
Token | And |
Token ID | 843 |
Feature activation | -2.81e-02 |
Pos logit contributions | |
en | +0.015 |
umm | +0.015 |
aman | +0.015 |
rap | +0.015 |
berry | +0.014 |
Neg logit contributions | |
fast | -0.009 |
protective | -0.009 |
hog | -0.009 |
sneaky | -0.009 |
ggy | -0.008 |
Token | every |
Token ID | 790 |
Feature activation | +4.52e-02 |
Pos logit contributions | |
protective | +0.042 |
sneaky | +0.041 |
fast | +0.038 |
dodged | +0.037 |
brave | +0.036 |
Neg logit contributions | |
aman | -0.060 |
berry | -0.057 |
umm | -0.056 |
en | -0.056 |
rap | -0.055 |
Token | time |
Token ID | 640 |
Feature activation | +6.85e-03 |
Pos logit contributions | |
aman | +0.073 |
umm | +0.071 |
rap | +0.069 |
en | +0.068 |
berry | +0.067 |
Neg logit contributions | |
protective | -0.091 |
sneaky | -0.090 |
fast | -0.087 |
brave | -0.084 |
ggy | -0.083 |
Token | the |
Token ID | 262 |
Feature activation | -2.90e-03 |
Pos logit contributions | |
aman | +0.039 |
berry | +0.037 |
umm | +0.037 |
en | +0.036 |
rap | +0.035 |
Neg logit contributions | |
protective | -0.029 |
sneaky | -0.028 |
fast | -0.027 |
dodged | -0.026 |
villains | -0.025 |
Token | biggest |
Token ID | 4094 |
Feature activation | +4.35e-02 |
Pos logit contributions | |
protective | +0.004 |
sneaky | +0.004 |
fast | +0.004 |
brave | +0.004 |
sharp | +0.004 |
Neg logit contributions | |
aman | -0.006 |
en | -0.006 |
rap | -0.006 |
berry | -0.006 |
umm | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +4.82e-02 |
Pos logit contributions | |
aman | +0.090 |
rap | +0.087 |
umm | +0.084 |
en | +0.082 |
icer | +0.081 |
Neg logit contributions | |
protective | -0.068 |
sneaky | -0.065 |
fast | -0.064 |
brave | -0.061 |
sharp | -0.058 |
Token | brightest |
Token ID | 33871 |
Feature activation | +6.60e-02 |
Pos logit contributions | |
aman | +0.088 |
berry | +0.087 |
umm | +0.087 |
en | +0.083 |
el | +0.077 |
Neg logit contributions | |
villains | -0.079 |
protective | -0.078 |
ggy | -0.078 |
ars | -0.078 |
dodged | -0.076 |
Token | sun |
Token ID | 4252 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
aman | +0.134 |
rap | +0.128 |
umm | +0.123 |
icer | +0.118 |
en | +0.118 |
Neg logit contributions | |
protective | -0.106 |
sneaky | -0.100 |
fast | -0.100 |
brave | -0.094 |
sharp | -0.092 |
Token | fl |
Token ID | 2704 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
aman | +0.228 |
rap | +0.214 |
icer | +0.204 |
umm | +0.203 |
amaz | +0.192 |
Neg logit contributions | |
fast | -0.266 |
protective | -0.257 |
ages | -0.245 |
sneaky | -0.245 |
hog | -0.242 |
Token | owers |
Token ID | 3618 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
aman | +0.362 |
rap | +0.347 |
reality | +0.328 |
destination | +0.328 |
en | +0.324 |
Neg logit contributions | |
ars | -0.400 |
ages | -0.394 |
hog | -0.385 |
aro | -0.381 |
ards | -0.379 |
Token | she |
Token ID | 673 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
aman | +0.250 |
umm | +0.234 |
berry | +0.233 |
en | +0.229 |
rap | +0.228 |
Neg logit contributions | |
protective | -0.193 |
sneaky | -0.187 |
fast | -0.181 |
dodged | -0.171 |
brave | -0.170 |
Token | could |
Token ID | 714 |
Feature activation | +8.06e-02 |
Pos logit contributions | |
aman | +0.283 |
berry | +0.270 |
umm | +0.267 |
en | +0.265 |
rap | +0.259 |
Neg logit contributions | |
protective | -0.160 |
sneaky | -0.156 |
fast | -0.146 |
dodged | -0.140 |
brave | -0.137 |
Token | find |
Token ID | 1064 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
aman | +0.181 |
umm | +0.167 |
icer | +0.161 |
berry | +0.155 |
en | +0.153 |
Neg logit contributions | |
fast | -0.119 |
protective | -0.118 |
sneaky | -0.115 |
brave | -0.113 |
strong | -0.107 |
Token | . |
Token ID | 13 |
Feature activation | +4.75e-03 |
Pos logit contributions | |
aman | +0.025 |
umm | +0.024 |
rap | +0.023 |
berry | +0.023 |
en | +0.023 |
Neg logit contributions | |
protective | -0.017 |
sneaky | -0.016 |
fast | -0.016 |
brave | -0.015 |
dodged | -0.014 |
Token | made |
Token ID | 925 |
Feature activation | -3.91e-02 |
Pos logit contributions | |
aman | +0.050 |
berry | +0.048 |
umm | +0.048 |
en | +0.047 |
rap | +0.046 |
Neg logit contributions | |
protective | -0.031 |
sneaky | -0.030 |
fast | -0.028 |
dodged | -0.027 |
brave | -0.026 |
Token | him |
Token ID | 683 |
Feature activation | -6.73e-02 |
Pos logit contributions | |
protective | +0.063 |
sneaky | +0.061 |
fast | +0.058 |
dodged | +0.056 |
villains | +0.055 |
Neg logit contributions | |
aman | -0.078 |
berry | -0.075 |
umm | -0.073 |
en | -0.073 |
rap | -0.071 |
Token | an |
Token ID | 281 |
Feature activation | -4.68e-02 |
Pos logit contributions | |
protective | +0.100 |
sneaky | +0.097 |
dodged | +0.090 |
villains | +0.090 |
ars | +0.089 |
Neg logit contributions | |
aman | -0.135 |
berry | -0.135 |
en | -0.131 |
rap | -0.124 |
umm | -0.124 |
Token | even |
Token ID | 772 |
Feature activation | +2.07e-02 |
Pos logit contributions | |
protective | +0.067 |
sneaky | +0.065 |
fast | +0.060 |
dodged | +0.060 |
villains | +0.058 |
Neg logit contributions | |
aman | -0.102 |
berry | -0.098 |
umm | -0.097 |
en | -0.096 |
rap | -0.093 |
Token | y |
Token ID | 331 |
Feature activation | +7.92e-02 |
Pos logit contributions | |
en | +0.038 |
aman | +0.037 |
berry | +0.036 |
umm | +0.035 |
rap | +0.034 |
Neg logit contributions | |
protective | -0.030 |
villains | -0.030 |
aro | -0.029 |
ggy | -0.029 |
Rex | -0.029 |
Token | umm |
Token ID | 13929 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
aman | +0.066 |
berry | +0.050 |
neath | +0.049 |
date | +0.048 |
rap | +0.047 |
Neg logit contributions | |
protective | -0.208 |
sneaky | -0.203 |
fast | -0.201 |
ggy | -0.195 |
hog | -0.194 |
Token | ier |
Token ID | 959 |
Feature activation | +4.14e-02 |
Pos logit contributions | |
aman | +0.509 |
neath | +0.472 |
destination | +0.471 |
rap | +0.468 |
berry | +0.464 |
Neg logit contributions | |
protective | -0.254 |
fast | -0.238 |
sneaky | -0.235 |
brave | -0.227 |
ggy | -0.223 |
Token | snack |
Token ID | 26906 |
Feature activation | +5.69e-02 |
Pos logit contributions | |
aman | +0.082 |
berry | +0.077 |
umm | +0.076 |
rap | +0.076 |
en | +0.075 |
Neg logit contributions | |
protective | -0.069 |
sneaky | -0.065 |
fast | -0.063 |
brave | -0.060 |
ggy | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
aman | +0.108 |
berry | +0.103 |
umm | +0.101 |
en | +0.100 |
rap | +0.097 |
Neg logit contributions | |
protective | -0.097 |
sneaky | -0.095 |
fast | -0.090 |
dodged | -0.088 |
brave | -0.086 |
Token | they |
Token ID | 484 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
aman | +0.049 |
berry | +0.047 |
umm | +0.046 |
en | +0.046 |
rap | +0.045 |
Neg logit contributions | |
protective | -0.037 |
sneaky | -0.036 |
fast | -0.034 |
dodged | -0.033 |
brave | -0.032 |
Token | both |
Token ID | 1111 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
aman | +0.063 |
berry | +0.060 |
umm | +0.059 |
en | +0.059 |
rap | +0.057 |
Neg logit contributions | |
protective | -0.050 |
sneaky | -0.048 |
fast | -0.046 |
dodged | -0.045 |
brave | -0.043 |
Token | oil |
Token ID | 3056 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
aman | +0.059 |
umm | +0.056 |
berry | +0.055 |
en | +0.054 |
rap | +0.053 |
Neg logit contributions | |
protective | -0.054 |
sneaky | -0.053 |
fast | -0.050 |
dodged | -0.049 |
brave | -0.048 |
Token | , |
Token ID | 11 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
aman | +0.036 |
berry | +0.034 |
umm | +0.034 |
en | +0.033 |
rap | +0.032 |
Neg logit contributions | |
protective | -0.032 |
sneaky | -0.032 |
fast | -0.030 |
dodged | -0.029 |
brave | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
aman | +0.066 |
berry | +0.062 |
umm | +0.062 |
en | +0.060 |
rap | +0.059 |
Neg logit contributions | |
protective | -0.061 |
sneaky | -0.060 |
fast | -0.057 |
dodged | -0.055 |
brave | -0.054 |
Token | the |
Token ID | 262 |
Feature activation | +4.40e-02 |
Pos logit contributions | |
aman | +0.045 |
berry | +0.042 |
umm | +0.042 |
en | +0.041 |
rap | +0.040 |
Neg logit contributions | |
protective | -0.041 |
sneaky | -0.040 |
fast | -0.038 |
dodged | -0.037 |
brave | -0.037 |
Token | y |
Token ID | 331 |
Feature activation | +9.33e-02 |
Pos logit contributions | |
aman | +0.084 |
en | +0.078 |
rap | +0.077 |
berry | +0.077 |
umm | +0.075 |
Neg logit contributions | |
protective | -0.077 |
sneaky | -0.077 |
fast | -0.072 |
brave | -0.070 |
dodged | -0.069 |
Token | umm |
Token ID | 13929 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
aman | +0.081 |
airport | +0.057 |
neath | +0.054 |
?' | +0.051 |
rap | +0.051 |
Neg logit contributions | |
protective | -0.245 |
sneaky | -0.239 |
fast | -0.236 |
ages | -0.234 |
sharp | -0.231 |
Token | iest |
Token ID | 6386 |
Feature activation | +6.40e-02 |
Pos logit contributions | |
aman | +0.509 |
destination | +0.476 |
rap | +0.475 |
airport | +0.449 |
berry | +0.449 |
Neg logit contributions | |
protective | -0.253 |
ages | -0.237 |
ars | -0.234 |
sneaky | -0.232 |
fast | -0.226 |
Token | spaghetti |
Token ID | 48241 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
aman | +0.124 |
rap | +0.113 |
en | +0.106 |
umm | +0.105 |
berry | +0.104 |
Neg logit contributions | |
protective | -0.112 |
sneaky | -0.111 |
fast | -0.105 |
brave | -0.103 |
sharp | -0.098 |
Token | ever |
Token ID | 1683 |
Feature activation | +7.21e-02 |
Pos logit contributions | |
aman | +0.231 |
berry | +0.215 |
en | +0.213 |
umm | +0.212 |
rap | +0.208 |
Neg logit contributions | |
protective | -0.169 |
sneaky | -0.165 |
fast | -0.157 |
dodged | -0.150 |
brave | -0.148 |
Token | ! |
Token ID | 0 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
aman | +0.146 |
umm | +0.137 |
berry | +0.137 |
rap | +0.130 |
en | +0.129 |
Neg logit contributions | |
protective | -0.117 |
sneaky | -0.114 |
fast | -0.110 |
dodged | -0.105 |
brave | -0.104 |
Token |
|
Token ID | 198 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
aman | +0.039 |
en | +0.037 |
berry | +0.037 |
umm | +0.037 |
rap | +0.036 |
Neg logit contributions | |
protective | -0.032 |
sneaky | -0.031 |
fast | -0.030 |
dodged | -0.029 |
brave | -0.028 |
Token | amongst |
Token ID | 12077 |
Feature activation | -2.53e-03 |
Pos logit contributions | |
aman | +0.095 |
berry | +0.091 |
en | +0.088 |
umm | +0.088 |
rap | +0.086 |
Neg logit contributions | |
protective | -0.097 |
sneaky | -0.093 |
villains | -0.087 |
dodged | -0.087 |
fast | -0.087 |
Token | the |
Token ID | 262 |
Feature activation | +4.85e-02 |
Pos logit contributions | |
protective | +0.004 |
sneaky | +0.004 |
fast | +0.004 |
dodged | +0.004 |
villains | +0.004 |
Neg logit contributions | |
aman | -0.005 |
berry | -0.005 |
en | -0.005 |
umm | -0.005 |
rap | -0.004 |
Token | grass |
Token ID | 8701 |
Feature activation | +4.23e-02 |
Pos logit contributions | |
aman | +0.100 |
berry | +0.096 |
en | +0.094 |
umm | +0.094 |
rap | +0.090 |
Neg logit contributions | |
protective | -0.075 |
sneaky | -0.073 |
fast | -0.068 |
dodged | -0.067 |
villains | -0.065 |
Token | and |
Token ID | 290 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
aman | +0.086 |
umm | +0.082 |
berry | +0.081 |
en | +0.080 |
rap | +0.079 |
Neg logit contributions | |
protective | -0.065 |
sneaky | -0.064 |
fast | -0.062 |
dodged | -0.059 |
brave | -0.057 |
Token | wild |
Token ID | 4295 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
aman | +0.053 |
berry | +0.050 |
umm | +0.049 |
en | +0.048 |
rap | +0.047 |
Neg logit contributions | |
protective | -0.047 |
sneaky | -0.046 |
fast | -0.044 |
dodged | -0.043 |
brave | -0.042 |
Token | fl |
Token ID | 2704 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
aman | +0.063 |
berry | +0.059 |
umm | +0.059 |
en | +0.057 |
rap | +0.057 |
Neg logit contributions | |
protective | -0.054 |
sneaky | -0.052 |
fast | -0.050 |
dodged | -0.048 |
brave | -0.047 |
Token | owers |
Token ID | 3618 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
aman | +0.324 |
rap | +0.294 |
en | +0.284 |
umm | +0.278 |
destination | +0.275 |
Neg logit contributions | |
ars | -0.434 |
ages | -0.418 |
protective | -0.411 |
ggy | -0.406 |
dodged | -0.406 |
Token | . |
Token ID | 13 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
aman | +0.199 |
berry | +0.191 |
en | +0.187 |
umm | +0.186 |
rap | +0.179 |
Neg logit contributions | |
protective | -0.181 |
sneaky | -0.176 |
fast | -0.166 |
dodged | -0.163 |
villains | -0.160 |
Token | Anna |
Token ID | 11735 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
aman | +0.022 |
umm | +0.021 |
berry | +0.020 |
en | +0.020 |
rap | +0.020 |
Neg logit contributions | |
protective | -0.017 |
sneaky | -0.017 |
fast | -0.016 |
dodged | -0.015 |
brave | -0.015 |
Token | was |
Token ID | 373 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
protective | +0.045 |
sneaky | +0.044 |
fast | +0.041 |
villains | +0.041 |
dodged | +0.039 |
Neg logit contributions | |
aman | -0.056 |
berry | -0.054 |
en | -0.054 |
rap | -0.050 |
el | -0.049 |
Token | so |
Token ID | 523 |
Feature activation | -5.28e-02 |
Pos logit contributions | |
aman | +0.045 |
berry | +0.044 |
en | +0.044 |
rap | +0.042 |
umm | +0.042 |
Neg logit contributions | |
protective | -0.023 |
ars | -0.021 |
villains | -0.021 |
sneaky | -0.021 |
dodged | -0.020 |
Token | stretch |
Token ID | 7539 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
aman | +0.162 |
umm | +0.153 |
berry | +0.153 |
en | +0.148 |
rap | +0.145 |
Neg logit contributions | |
protective | -0.148 |
sneaky | -0.144 |
fast | -0.138 |
dodged | -0.133 |
brave | -0.131 |
Token | out |
Token ID | 503 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
aman | +0.032 |
berry | +0.031 |
umm | +0.031 |
en | +0.030 |
rap | +0.029 |
Neg logit contributions | |
protective | -0.029 |
sneaky | -0.028 |
fast | -0.027 |
dodged | -0.026 |
brave | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
protective | +0.023 |
sneaky | +0.022 |
fast | +0.022 |
brave | +0.021 |
dodged | +0.020 |
Neg logit contributions | |
aman | -0.029 |
umm | -0.027 |
berry | -0.027 |
rap | -0.026 |
en | -0.026 |
Token | enormous |
Token ID | 9812 |
Feature activation | +6.37e-02 |
Pos logit contributions | |
aman | +0.048 |
berry | +0.045 |
umm | +0.045 |
en | +0.044 |
rap | +0.043 |
Neg logit contributions | |
protective | -0.039 |
sneaky | -0.038 |
fast | -0.036 |
dodged | -0.035 |
brave | -0.034 |
Token | sun |
Token ID | 4252 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
aman | +0.136 |
umm | +0.124 |
rap | +0.124 |
berry | +0.123 |
icer | +0.120 |
Neg logit contributions | |
protective | -0.099 |
sneaky | -0.095 |
fast | -0.091 |
brave | -0.087 |
dodged | -0.084 |
Token | fl |
Token ID | 2704 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
aman | +0.241 |
umm | +0.223 |
rap | +0.218 |
berry | +0.213 |
icer | +0.213 |
Neg logit contributions | |
protective | -0.240 |
fast | -0.233 |
sneaky | -0.229 |
ages | -0.218 |
brave | -0.217 |
Token | owers |
Token ID | 3618 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
aman | +0.340 |
rap | +0.318 |
en | +0.300 |
umm | +0.293 |
reating | +0.283 |
Neg logit contributions | |
ars | -0.417 |
ages | -0.408 |
protective | -0.393 |
hog | -0.391 |
dodged | -0.387 |
Token | would |
Token ID | 561 |
Feature activation | +9.88e-02 |
Pos logit contributions | |
aman | +0.219 |
berry | +0.205 |
umm | +0.205 |
en | +0.201 |
rap | +0.197 |
Neg logit contributions | |
protective | -0.185 |
sneaky | -0.180 |
fast | -0.172 |
dodged | -0.166 |
brave | -0.163 |
Token | show |
Token ID | 905 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
aman | +0.191 |
umm | +0.181 |
berry | +0.180 |
en | +0.174 |
rap | +0.171 |
Neg logit contributions | |
protective | -0.168 |
sneaky | -0.163 |
fast | -0.159 |
dodged | -0.150 |
brave | -0.149 |
Token | their |
Token ID | 511 |
Feature activation | -4.47e-02 |
Pos logit contributions | |
protective | +0.030 |
sneaky | +0.029 |
fast | +0.028 |
dodged | +0.028 |
villains | +0.027 |
Neg logit contributions | |
aman | -0.026 |
berry | -0.025 |
en | -0.025 |
umm | -0.024 |
rap | -0.023 |
Token | bright |
Token ID | 6016 |
Feature activation | +4.06e-03 |
Pos logit contributions | |
protective | +0.071 |
sneaky | +0.068 |
fast | +0.065 |
brave | +0.063 |
cunning | +0.059 |
Neg logit contributions | |
aman | -0.093 |
berry | -0.088 |
umm | -0.087 |
en | -0.086 |
rap | -0.084 |
Token | . |
Token ID | 13 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
aman | +0.079 |
umm | +0.072 |
berry | +0.070 |
rap | +0.069 |
en | +0.069 |
Neg logit contributions | |
protective | -0.055 |
sneaky | -0.053 |
fast | -0.052 |
dodged | -0.050 |
brave | -0.048 |
Token | It |
Token ID | 632 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
aman | +0.022 |
berry | +0.021 |
en | +0.021 |
umm | +0.021 |
rap | +0.020 |
Neg logit contributions | |
protective | -0.016 |
sneaky | -0.016 |
fast | -0.015 |
dodged | -0.015 |
brave | -0.014 |
Token | looked |
Token ID | 3114 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
protective | +0.048 |
sneaky | +0.047 |
fast | +0.044 |
dodged | +0.043 |
villains | +0.042 |
Neg logit contributions | |
aman | -0.060 |
berry | -0.057 |
umm | -0.056 |
en | -0.056 |
rap | -0.054 |
Token | even |
Token ID | 772 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
protective | +0.019 |
sneaky | +0.019 |
fast | +0.017 |
dodged | +0.017 |
brave | +0.016 |
Neg logit contributions | |
aman | -0.044 |
berry | -0.042 |
umm | -0.041 |
en | -0.041 |
rap | -0.040 |
Token | y |
Token ID | 331 |
Feature activation | +7.46e-02 |
Pos logit contributions | |
aman | +0.054 |
berry | +0.052 |
umm | +0.051 |
rap | +0.050 |
en | +0.050 |
Neg logit contributions | |
protective | -0.034 |
sneaky | -0.034 |
fast | -0.032 |
brave | -0.029 |
dodged | -0.029 |
Token | umm |
Token ID | 13929 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
aman | +0.059 |
berry | +0.047 |
en | +0.042 |
rap | +0.041 |
destination | +0.040 |
Neg logit contributions | |
protective | -0.206 |
sneaky | -0.203 |
fast | -0.201 |
brave | -0.191 |
ars | -0.190 |
Token | ier |
Token ID | 959 |
Feature activation | +3.91e-02 |
Pos logit contributions | |
aman | +0.457 |
destination | +0.424 |
neath | +0.420 |
airport | +0.410 |
berry | +0.403 |
Neg logit contributions | |
ages | -0.296 |
protective | -0.285 |
fast | -0.281 |
sneaky | -0.278 |
ars | -0.276 |
Token | . |
Token ID | 13 |
Feature activation | +4.73e-03 |
Pos logit contributions | |
aman | +0.091 |
berry | +0.088 |
umm | +0.086 |
en | +0.083 |
rap | +0.083 |
Neg logit contributions | |
protective | -0.051 |
sneaky | -0.048 |
fast | -0.047 |
brave | -0.043 |
dodged | -0.042 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.12e-03 |
Pos logit contributions | |
aman | +0.010 |
en | +0.010 |
berry | +0.010 |
umm | +0.010 |
rap | +0.009 |
Neg logit contributions | |
protective | -0.007 |
sneaky | -0.007 |
fast | -0.006 |
dodged | -0.006 |
brave | -0.006 |
Token | wanted |
Token ID | 2227 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
aman | +0.006 |
berry | +0.006 |
en | +0.006 |
rap | +0.005 |
umm | +0.005 |
Neg logit contributions | |
protective | -0.005 |
sneaky | -0.005 |
fast | -0.005 |
villains | -0.005 |
aro | -0.005 |
Token | to |
Token ID | 284 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
aman | +0.061 |
umm | +0.058 |
berry | +0.058 |
en | +0.057 |
rap | +0.055 |
Neg logit contributions | |
protective | -0.052 |
sneaky | -0.051 |
fast | -0.049 |
dodged | -0.046 |
brave | -0.046 |
Token | with |
Token ID | 351 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
protective | +0.024 |
sneaky | +0.024 |
villains | +0.022 |
dodged | +0.022 |
fast | +0.022 |
Neg logit contributions | |
aman | -0.024 |
berry | -0.023 |
en | -0.023 |
umm | -0.022 |
rap | -0.021 |
Token | lots |
Token ID | 6041 |
Feature activation | -4.47e-03 |
Pos logit contributions | |
protective | +0.049 |
sneaky | +0.047 |
dodged | +0.044 |
ars | +0.043 |
fast | +0.043 |
Neg logit contributions | |
aman | -0.069 |
berry | -0.067 |
en | -0.066 |
umm | -0.065 |
rap | -0.062 |
Token | of |
Token ID | 286 |
Feature activation | +4.86e-02 |
Pos logit contributions | |
protective | +0.008 |
sneaky | +0.008 |
dodged | +0.008 |
fast | +0.007 |
villains | +0.007 |
Neg logit contributions | |
aman | -0.008 |
berry | -0.008 |
en | -0.007 |
umm | -0.007 |
rap | -0.007 |
Token | lovely |
Token ID | 14081 |
Feature activation | +7.09e-02 |
Pos logit contributions | |
aman | +0.101 |
berry | +0.099 |
en | +0.096 |
umm | +0.096 |
rap | +0.090 |
Neg logit contributions | |
protective | -0.072 |
sneaky | -0.069 |
dodged | -0.064 |
fast | -0.063 |
ggy | -0.063 |
Token | wild |
Token ID | 4295 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
aman | +0.152 |
umm | +0.144 |
rap | +0.141 |
berry | +0.140 |
en | +0.137 |
Neg logit contributions | |
protective | -0.106 |
sneaky | -0.102 |
fast | -0.099 |
brave | -0.094 |
dodged | -0.092 |
Token | fl |
Token ID | 2704 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
aman | +0.052 |
berry | +0.050 |
umm | +0.049 |
en | +0.049 |
rap | +0.047 |
Neg logit contributions | |
protective | -0.042 |
sneaky | -0.041 |
fast | -0.039 |
dodged | -0.038 |
villains | -0.037 |
Token | owers |
Token ID | 3618 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
aman | +0.317 |
rap | +0.285 |
en | +0.274 |
destination | +0.264 |
?' | +0.264 |
Neg logit contributions | |
ars | -0.434 |
protective | -0.426 |
ages | -0.420 |
hog | -0.420 |
dodged | -0.419 |
Token | growing |
Token ID | 3957 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
aman | +0.201 |
berry | +0.190 |
umm | +0.188 |
en | +0.186 |
rap | +0.181 |
Neg logit contributions | |
protective | -0.181 |
sneaky | -0.176 |
fast | -0.168 |
dodged | -0.163 |
brave | -0.159 |
Token | on |
Token ID | 319 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
aman | +0.202 |
berry | +0.192 |
umm | +0.189 |
en | +0.187 |
rap | +0.182 |
Neg logit contributions | |
protective | -0.178 |
sneaky | -0.174 |
fast | -0.165 |
dodged | -0.160 |
brave | -0.156 |
Token | it |
Token ID | 340 |
Feature activation | +4.93e-02 |
Pos logit contributions | |
aman | +0.042 |
umm | +0.041 |
berry | +0.039 |
rap | +0.038 |
en | +0.038 |
Neg logit contributions | |
fast | -0.031 |
protective | -0.030 |
sneaky | -0.030 |
brave | -0.027 |
sharp | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
aman | +0.094 |
berry | +0.090 |
umm | +0.088 |
en | +0.088 |
rap | +0.085 |
Neg logit contributions | |
protective | -0.084 |
sneaky | -0.081 |
fast | -0.077 |
dodged | -0.075 |
villains | -0.074 |
Token | lots |
Token ID | 6041 |
Feature activation | -2.76e-02 |
Pos logit contributions | |
aman | +0.023 |
berry | +0.023 |
en | +0.022 |
umm | +0.021 |
el | +0.020 |
Neg logit contributions | |
protective | -0.022 |
ars | -0.021 |
sneaky | -0.020 |
dodged | -0.020 |
cunning | -0.019 |
Token | of |
Token ID | 286 |
Feature activation | +3.60e-02 |
Pos logit contributions | |
protective | +0.059 |
sneaky | +0.057 |
dodged | +0.057 |
ars | +0.055 |
ggy | +0.055 |
Neg logit contributions | |
berry | -0.040 |
en | -0.039 |
umm | -0.036 |
aman | -0.036 |
el | -0.031 |
Token | trees |
Token ID | 7150 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
aman | +0.075 |
berry | +0.074 |
en | +0.073 |
umm | +0.071 |
rap | +0.067 |
Neg logit contributions | |
protective | -0.051 |
sneaky | -0.048 |
dodged | -0.047 |
ggy | -0.045 |
fast | -0.045 |
Token | and |
Token ID | 290 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
aman | +0.035 |
berry | +0.034 |
en | +0.033 |
umm | +0.033 |
rap | +0.032 |
Neg logit contributions | |
protective | -0.033 |
sneaky | -0.032 |
fast | -0.030 |
dodged | -0.030 |
villains | -0.029 |
Token | wild |
Token ID | 4295 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
berry | +0.049 |
en | +0.049 |
aman | +0.048 |
umm | +0.046 |
rap | +0.042 |
Neg logit contributions | |
protective | -0.048 |
sneaky | -0.045 |
ggy | -0.044 |
ars | -0.044 |
dodged | -0.043 |
Token | fl |
Token ID | 2704 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
aman | +0.048 |
berry | +0.046 |
umm | +0.045 |
en | +0.045 |
rap | +0.043 |
Neg logit contributions | |
protective | -0.041 |
sneaky | -0.040 |
fast | -0.038 |
dodged | -0.037 |
villains | -0.036 |
Token | owers |
Token ID | 3618 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
aman | +0.296 |
rap | +0.262 |
en | +0.261 |
berry | +0.254 |
umm | +0.250 |
Neg logit contributions | |
ars | -0.448 |
protective | -0.445 |
sneaky | -0.442 |
dodged | -0.437 |
ages | -0.437 |
Token | . |
Token ID | 13 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
aman | +0.190 |
berry | +0.183 |
en | +0.179 |
umm | +0.177 |
rap | +0.169 |
Neg logit contributions | |
protective | -0.178 |
sneaky | -0.173 |
fast | -0.163 |
dodged | -0.161 |
villains | -0.158 |
Token | As |
Token ID | 1081 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
aman | +0.023 |
berry | +0.023 |
umm | +0.022 |
en | +0.021 |
icer | +0.020 |
Neg logit contributions | |
protective | -0.026 |
sneaky | -0.024 |
ggy | -0.023 |
fast | -0.023 |
dodged | -0.023 |
Token | he |
Token ID | 339 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
protective | +0.033 |
sneaky | +0.030 |
ars | +0.029 |
ages | +0.028 |
dodged | +0.028 |
Neg logit contributions | |
aman | -0.053 |
berry | -0.051 |
en | -0.051 |
umm | -0.049 |
rap | -0.048 |
Token | stepped |
Token ID | 10764 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
aman | +0.057 |
berry | +0.057 |
en | +0.055 |
umm | +0.051 |
el | +0.051 |
Neg logit contributions | |
protective | -0.059 |
sneaky | -0.056 |
fast | -0.053 |
villains | -0.053 |
dodged | -0.050 |
Token | into |
Token ID | 656 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
protective | +0.016 |
sneaky | +0.016 |
fast | +0.016 |
brave | +0.015 |
dodged | +0.014 |
Neg logit contributions | |
aman | -0.024 |
umm | -0.022 |
berry | -0.022 |
en | -0.021 |
rap | -0.021 |
Token | big |
Token ID | 1263 |
Feature activation | +4.80e-02 |
Pos logit contributions | |
sneaky | +0.022 |
protective | +0.021 |
fast | +0.021 |
brave | +0.020 |
sharp | +0.019 |
Neg logit contributions | |
aman | -0.041 |
umm | -0.038 |
berry | -0.038 |
rap | -0.036 |
en | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
aman | +0.114 |
rap | +0.108 |
umm | +0.100 |
omm | +0.098 |
icer | +0.097 |
Neg logit contributions | |
sneaky | -0.066 |
protective | -0.065 |
fast | -0.062 |
brave | -0.062 |
sharp | -0.057 |
Token | bright |
Token ID | 6016 |
Feature activation | +5.93e-02 |
Pos logit contributions | |
berry | +0.032 |
umm | +0.032 |
aman | +0.032 |
en | +0.031 |
?' | +0.030 |
Neg logit contributions | |
ggy | -0.020 |
protective | -0.020 |
villains | -0.020 |
ars | -0.019 |
Rex | -0.019 |
Token | sun |
Token ID | 4252 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
aman | +0.142 |
rap | +0.132 |
omm | +0.121 |
icer | +0.121 |
umm | +0.120 |
Neg logit contributions | |
protective | -0.080 |
fast | -0.077 |
sneaky | -0.076 |
brave | -0.073 |
strong | -0.069 |
Token | fl |
Token ID | 2704 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
aman | +0.243 |
rap | +0.219 |
umm | +0.217 |
icer | +0.212 |
en | +0.208 |
Neg logit contributions | |
fast | -0.167 |
protective | -0.166 |
sneaky | -0.161 |
ages | -0.157 |
brave | -0.153 |
Token | owers |
Token ID | 3618 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
aman | +0.381 |
rap | +0.360 |
reating | +0.346 |
amaz | +0.343 |
en | +0.337 |
Neg logit contributions | |
ars | -0.356 |
ages | -0.348 |
sneaky | -0.320 |
fast | -0.319 |
ards | -0.319 |
Token | . |
Token ID | 13 |
Feature activation | +2.87e-03 |
Pos logit contributions | |
aman | +0.215 |
umm | +0.200 |
berry | +0.199 |
en | +0.197 |
rap | +0.194 |
Neg logit contributions | |
protective | -0.163 |
sneaky | -0.159 |
fast | -0.153 |
brave | -0.145 |
dodged | -0.144 |
Token |
|
Token ID | 198 |
Feature activation | +5.55e-03 |
Pos logit contributions | |
aman | +0.006 |
umm | +0.006 |
en | +0.006 |
rap | +0.006 |
berry | +0.006 |
Neg logit contributions | |
fast | -0.004 |
protective | -0.004 |
sneaky | -0.004 |
dodged | -0.004 |
Rex | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -4.51e-03 |
Pos logit contributions | |
umm | +0.011 |
aman | +0.011 |
en | +0.011 |
berry | +0.011 |
rap | +0.010 |
Neg logit contributions | |
fast | -0.008 |
protective | -0.008 |
sneaky | -0.008 |
ggy | -0.008 |
dodged | -0.008 |
Token | One |
Token ID | 3198 |
Feature activation | +1.93e-02 |
Pos logit contributions | |
protective | +0.009 |
sneaky | +0.009 |
fast | +0.008 |
dodged | +0.008 |
villains | +0.008 |
Neg logit contributions | |
aman | -0.007 |
berry | -0.007 |
umm | -0.007 |
en | -0.007 |
rap | -0.006 |
Token | shiny |
Token ID | 22441 |
Feature activation | +6.56e-02 |
Pos logit contributions | |
aman | +0.107 |
berry | +0.104 |
umm | +0.101 |
en | +0.101 |
rap | +0.097 |
Neg logit contributions | |
protective | -0.080 |
sneaky | -0.077 |
fast | -0.072 |
dodged | -0.071 |
ars | -0.069 |
Token | rocks |
Token ID | 12586 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
aman | +0.139 |
umm | +0.131 |
berry | +0.130 |
en | +0.127 |
rap | +0.126 |
Neg logit contributions | |
protective | -0.099 |
sneaky | -0.096 |
fast | -0.092 |
dodged | -0.088 |
brave | -0.086 |
Token | , |
Token ID | 11 |
Feature activation | +3.82e-02 |
Pos logit contributions | |
berry | +0.087 |
aman | +0.086 |
en | +0.085 |
umm | +0.076 |
el | +0.075 |
Neg logit contributions | |
protective | -0.118 |
sneaky | -0.116 |
villains | -0.111 |
dodged | -0.109 |
ggy | -0.108 |
Token | fluffy |
Token ID | 39145 |
Feature activation | +5.71e-02 |
Pos logit contributions | |
berry | +0.063 |
en | +0.063 |
aman | +0.059 |
umm | +0.054 |
el | +0.053 |
Neg logit contributions | |
protective | -0.070 |
sneaky | -0.066 |
ggy | -0.065 |
villains | -0.063 |
hog | -0.063 |
Token | snow |
Token ID | 6729 |
Feature activation | +9.48e-02 |
Pos logit contributions | |
aman | +0.120 |
umm | +0.112 |
berry | +0.111 |
en | +0.109 |
rap | +0.108 |
Neg logit contributions | |
protective | -0.088 |
sneaky | -0.085 |
fast | -0.082 |
dodged | -0.078 |
brave | -0.077 |
Token | fl |
Token ID | 2704 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
aman | +0.183 |
berry | +0.173 |
umm | +0.171 |
en | +0.169 |
rap | +0.165 |
Neg logit contributions | |
protective | -0.159 |
sneaky | -0.155 |
fast | -0.148 |
dodged | -0.144 |
brave | -0.140 |
Token | akes |
Token ID | 1124 |
Feature activation | +2.86e-02 |
Pos logit contributions | |
aman | +0.292 |
rap | +0.255 |
en | +0.250 |
reality | +0.240 |
amaz | +0.238 |
Neg logit contributions | |
ars | -0.450 |
ages | -0.439 |
dodged | -0.436 |
fast | -0.432 |
sneaky | -0.428 |
Token | and |
Token ID | 290 |
Feature activation | +4.89e-02 |
Pos logit contributions | |
aman | +0.050 |
berry | +0.049 |
en | +0.048 |
umm | +0.047 |
rap | +0.045 |
Neg logit contributions | |
protective | -0.053 |
sneaky | -0.052 |
fast | -0.048 |
dodged | -0.048 |
villains | -0.047 |
Token | colorful |
Token ID | 20239 |
Feature activation | +9.81e-02 |
Pos logit contributions | |
berry | +0.090 |
en | +0.088 |
aman | +0.083 |
umm | +0.082 |
el | +0.078 |
Neg logit contributions | |
protective | -0.076 |
sneaky | -0.072 |
ars | -0.069 |
ggy | -0.068 |
hog | -0.067 |
Token | fish |
Token ID | 5916 |
Feature activation | +7.15e-02 |
Pos logit contributions | |
aman | +0.220 |
umm | +0.201 |
rap | +0.199 |
berry | +0.194 |
en | +0.192 |
Neg logit contributions | |
protective | -0.142 |
sneaky | -0.137 |
fast | -0.132 |
brave | -0.124 |
dodged | -0.124 |
Token | . |
Token ID | 13 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
berry | +0.124 |
aman | +0.120 |
en | +0.119 |
umm | +0.113 |
rap | +0.110 |
Neg logit contributions | |
protective | -0.131 |
sneaky | -0.128 |
fast | -0.119 |
dodged | -0.118 |
villains | -0.116 |
Token | bakery |
Token ID | 38164 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
protective | +0.010 |
sneaky | +0.010 |
fast | +0.010 |
brave | +0.009 |
villains | +0.009 |
Neg logit contributions | |
aman | -0.009 |
en | -0.008 |
berry | -0.008 |
umm | -0.008 |
rap | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
umm | +0.031 |
aman | +0.031 |
en | +0.031 |
rap | +0.030 |
berry | +0.030 |
Neg logit contributions | |
fast | -0.018 |
sneaky | -0.018 |
protective | -0.017 |
brave | -0.016 |
ranger | -0.016 |
Token | buy |
Token ID | 2822 |
Feature activation | +9.89e-03 |
Pos logit contributions | |
aman | +0.058 |
umm | +0.057 |
berry | +0.056 |
en | +0.055 |
rap | +0.052 |
Neg logit contributions | |
sneaky | -0.052 |
protective | -0.051 |
fast | -0.050 |
brave | -0.048 |
dodged | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
aman | +0.020 |
berry | +0.019 |
umm | +0.019 |
en | +0.019 |
rap | +0.019 |
Neg logit contributions | |
protective | -0.015 |
sneaky | -0.015 |
fast | -0.014 |
brave | -0.013 |
dodged | -0.013 |
Token | y |
Token ID | 331 |
Feature activation | +6.97e-02 |
Pos logit contributions | |
sneaky | +0.039 |
protective | +0.039 |
fast | +0.037 |
brave | +0.035 |
dodged | +0.033 |
Neg logit contributions | |
aman | -0.050 |
en | -0.049 |
rap | -0.047 |
berry | -0.047 |
umm | -0.047 |
Token | umm |
Token ID | 13929 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
aman | +0.063 |
neath | +0.055 |
rap | +0.054 |
?' | +0.053 |
en | +0.052 |
Neg logit contributions | |
protective | -0.172 |
sneaky | -0.166 |
hog | -0.164 |
fast | -0.162 |
ages | -0.160 |
Token | iest |
Token ID | 6386 |
Feature activation | +4.53e-02 |
Pos logit contributions | |
aman | +0.483 |
rap | +0.453 |
destination | +0.453 |
en | +0.446 |
amaz | +0.444 |
Neg logit contributions | |
protective | -0.229 |
ages | -0.223 |
ars | -0.215 |
sneaky | -0.209 |
fast | -0.209 |
Token | pastry |
Token ID | 41738 |
Feature activation | +6.77e-02 |
Pos logit contributions | |
aman | +0.093 |
en | +0.089 |
rap | +0.089 |
umm | +0.085 |
neath | +0.082 |
Neg logit contributions | |
protective | -0.072 |
sneaky | -0.070 |
fast | -0.067 |
brave | -0.066 |
sharp | -0.062 |
Token | he |
Token ID | 339 |
Feature activation | +9.96e-02 |
Pos logit contributions | |
aman | +0.142 |
en | +0.133 |
berry | +0.132 |
umm | +0.131 |
rap | +0.129 |
Neg logit contributions | |
protective | -0.103 |
sneaky | -0.101 |
fast | -0.096 |
dodged | -0.091 |
brave | -0.091 |
Token | could |
Token ID | 714 |
Feature activation | +9.24e-02 |
Pos logit contributions | |
aman | +0.233 |
berry | +0.222 |
umm | +0.220 |
en | +0.218 |
rap | +0.212 |
Neg logit contributions | |
protective | -0.131 |
sneaky | -0.130 |
fast | -0.120 |
dodged | -0.115 |
brave | -0.115 |
Token | find |
Token ID | 1064 |
Feature activation | +1.45e-02 |
Pos logit contributions | |
aman | +0.203 |
umm | +0.193 |
icer | +0.181 |
en | +0.180 |
berry | +0.179 |
Neg logit contributions | |
fast | -0.136 |
protective | -0.135 |
sneaky | -0.133 |
brave | -0.128 |
ggy | -0.124 |
Token | in |
Token ID | 287 |
Feature activation | -8.45e-03 |
Pos logit contributions | |
aman | +0.018 |
umm | +0.017 |
berry | +0.016 |
en | +0.016 |
rap | +0.016 |
Neg logit contributions | |
protective | -0.013 |
fast | -0.013 |
sneaky | -0.013 |
brave | -0.012 |
dodged | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
protective | +0.012 |
sneaky | +0.012 |
fast | +0.012 |
brave | +0.011 |
dodged | +0.011 |
Neg logit contributions | |
aman | -0.019 |
umm | -0.017 |
berry | -0.017 |
en | -0.017 |
rap | -0.017 |
Token | cold |
Token ID | 4692 |
Feature activation | +8.26e-03 |
Pos logit contributions | |
rap | +0.059 |
aman | +0.058 |
umm | +0.056 |
en | +0.055 |
berry | +0.053 |
Neg logit contributions | |
fast | -0.048 |
protective | -0.046 |
brave | -0.046 |
sneaky | -0.046 |
sharp | -0.045 |
Token | white |
Token ID | 2330 |
Feature activation | +6.05e-02 |
Pos logit contributions | |
aman | +0.014 |
berry | +0.013 |
en | +0.013 |
umm | +0.013 |
rap | +0.012 |
Neg logit contributions | |
protective | -0.016 |
sneaky | -0.016 |
dodged | -0.015 |
fast | -0.015 |
villains | -0.014 |
Token | snow |
Token ID | 6729 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
aman | +0.119 |
rap | +0.113 |
umm | +0.110 |
berry | +0.105 |
en | +0.104 |
Neg logit contributions | |
protective | -0.097 |
fast | -0.095 |
sneaky | -0.093 |
brave | -0.088 |
ars | -0.087 |
Token | fl |
Token ID | 2704 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
aman | +0.167 |
berry | +0.166 |
en | +0.165 |
umm | +0.163 |
rap | +0.151 |
Neg logit contributions | |
protective | -0.193 |
sneaky | -0.188 |
dodged | -0.174 |
villains | -0.171 |
cunning | -0.170 |
Token | akes |
Token ID | 1124 |
Feature activation | +3.70e-02 |
Pos logit contributions | |
aman | +0.271 |
rap | +0.253 |
en | +0.240 |
reality | +0.239 |
amaz | +0.229 |
Neg logit contributions | |
ars | -0.453 |
ages | -0.440 |
fast | -0.427 |
ggy | -0.426 |
dodged | -0.420 |
Token | and |
Token ID | 290 |
Feature activation | +4.25e-02 |
Pos logit contributions | |
aman | +0.065 |
berry | +0.063 |
en | +0.063 |
umm | +0.061 |
rap | +0.058 |
Neg logit contributions | |
protective | -0.068 |
sneaky | -0.066 |
dodged | -0.061 |
villains | -0.061 |
fast | -0.060 |
Token | sh |
Token ID | 427 |
Feature activation | +8.79e-02 |
Pos logit contributions | |
aman | +0.082 |
berry | +0.080 |
en | +0.080 |
umm | +0.077 |
rap | +0.074 |
Neg logit contributions | |
protective | -0.069 |
sneaky | -0.067 |
ggy | -0.061 |
dodged | -0.061 |
villains | -0.061 |
Token | iver |
Token ID | 1428 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
aman | +0.124 |
berry | +0.116 |
en | +0.113 |
umm | +0.113 |
rap | +0.107 |
Neg logit contributions | |
protective | -0.195 |
sneaky | -0.192 |
fast | -0.183 |
dodged | -0.180 |
villains | -0.177 |
Token | when |
Token ID | 618 |
Feature activation | +3.82e-02 |
Pos logit contributions | |
aman | +0.020 |
en | +0.020 |
berry | +0.019 |
umm | +0.019 |
rap | +0.018 |
Neg logit contributions | |
protective | -0.020 |
sneaky | -0.020 |
dodged | -0.019 |
cunning | -0.018 |
ars | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | +8.10e-03 |
Pos logit contributions | |
fast | +0.022 |
sneaky | +0.021 |
protective | +0.021 |
brave | +0.020 |
dodged | +0.019 |
Neg logit contributions | |
aman | -0.033 |
en | -0.033 |
rap | -0.032 |
umm | -0.032 |
berry | -0.031 |
Token | it |
Token ID | 340 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
aman | +0.017 |
berry | +0.016 |
umm | +0.016 |
en | +0.016 |
rap | +0.015 |
Neg logit contributions | |
protective | -0.012 |
sneaky | -0.012 |
fast | -0.011 |
dodged | -0.011 |
villains | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | +4.78e-02 |
Pos logit contributions | |
aman | +0.065 |
berry | +0.062 |
umm | +0.061 |
en | +0.061 |
rap | +0.058 |
Neg logit contributions | |
protective | -0.056 |
sneaky | -0.054 |
fast | -0.051 |
dodged | -0.050 |
villains | -0.049 |
Token | the |
Token ID | 262 |
Feature activation | +8.69e-03 |
Pos logit contributions | |
aman | +0.125 |
umm | +0.124 |
berry | +0.122 |
en | +0.119 |
rap | +0.115 |
Neg logit contributions | |
fast | -0.053 |
sneaky | -0.050 |
protective | -0.050 |
brave | -0.046 |
strong | -0.045 |
Token | y |
Token ID | 331 |
Feature activation | +7.09e-02 |
Pos logit contributions | |
en | +0.019 |
aman | +0.018 |
rap | +0.018 |
berry | +0.018 |
umm | +0.017 |
Neg logit contributions | |
sneaky | -0.013 |
fast | -0.013 |
protective | -0.013 |
brave | -0.012 |
strong | -0.011 |
Token | umm |
Token ID | 13929 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
aman | +0.044 |
rap | +0.032 |
?' | +0.031 |
berry | +0.030 |
en | +0.030 |
Neg logit contributions | |
protective | -0.203 |
sneaky | -0.199 |
fast | -0.198 |
hog | -0.189 |
ars | -0.189 |
Token | iest |
Token ID | 6386 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
aman | +0.457 |
rap | +0.436 |
destination | +0.433 |
neath | +0.420 |
en | +0.420 |
Neg logit contributions | |
protective | -0.257 |
fast | -0.251 |
sneaky | -0.245 |
ggy | -0.244 |
ars | -0.241 |
Token | ice |
Token ID | 4771 |
Feature activation | +8.54e-02 |
Pos logit contributions | |
rap | +0.101 |
aman | +0.098 |
neath | +0.096 |
en | +0.095 |
?' | +0.091 |
Neg logit contributions | |
fast | -0.076 |
sneaky | -0.075 |
protective | -0.075 |
brave | -0.071 |
ggy | -0.069 |
Token | cream |
Token ID | 8566 |
Feature activation | +5.55e-02 |
Pos logit contributions | |
aman | +0.199 |
berry | +0.192 |
umm | +0.190 |
en | +0.187 |
rap | +0.182 |
Neg logit contributions | |
protective | -0.110 |
sneaky | -0.106 |
fast | -0.097 |
dodged | -0.096 |
villains | -0.094 |
Token | she |
Token ID | 673 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
aman | +0.114 |
en | +0.110 |
berry | +0.108 |
rap | +0.107 |
umm | +0.105 |
Neg logit contributions | |
protective | -0.083 |
sneaky | -0.082 |
fast | -0.081 |
dodged | -0.075 |
brave | -0.075 |
Token | ever |
Token ID | 1683 |
Feature activation | +4.58e-02 |
Pos logit contributions | |
aman | +0.276 |
berry | +0.261 |
umm | +0.258 |
en | +0.257 |
rap | +0.251 |
Neg logit contributions | |
protective | -0.131 |
sneaky | -0.131 |
fast | -0.121 |
dodged | -0.116 |
brave | -0.115 |
Token | the |
Token ID | 262 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
aman | +0.091 |
berry | +0.087 |
umm | +0.085 |
en | +0.085 |
rap | +0.082 |
Neg logit contributions | |
protective | -0.086 |
sneaky | -0.084 |
fast | -0.080 |
dodged | -0.078 |
villains | -0.076 |
Token | music |
Token ID | 2647 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
aman | +0.050 |
en | +0.048 |
berry | +0.048 |
umm | +0.047 |
date | +0.046 |
Neg logit contributions | |
protective | -0.041 |
sneaky | -0.039 |
dodged | -0.037 |
hog | -0.036 |
ages | -0.035 |
Token | was |
Token ID | 373 |
Feature activation | +5.37e-02 |
Pos logit contributions | |
protective | +0.049 |
sneaky | +0.048 |
dodged | +0.044 |
fast | +0.044 |
ars | +0.044 |
Neg logit contributions | |
aman | -0.042 |
berry | -0.041 |
en | -0.041 |
umm | -0.040 |
date | -0.038 |
Token | even |
Token ID | 772 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
aman | +0.105 |
en | +0.104 |
berry | +0.098 |
rap | +0.097 |
umm | +0.095 |
Neg logit contributions | |
protective | -0.082 |
sneaky | -0.078 |
ars | -0.078 |
villains | -0.075 |
dodged | -0.075 |
Token | y |
Token ID | 331 |
Feature activation | +7.86e-02 |
Pos logit contributions | |
en | +0.046 |
aman | +0.045 |
umm | +0.041 |
berry | +0.041 |
rap | +0.039 |
Neg logit contributions | |
cious | -0.057 |
protective | -0.056 |
villains | -0.055 |
ggy | -0.055 |
aro | -0.055 |
Token | umm |
Token ID | 13929 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
aman | +0.056 |
berry | +0.046 |
en | +0.042 |
rap | +0.040 |
umm | +0.039 |
Neg logit contributions | |
protective | -0.227 |
sneaky | -0.224 |
fast | -0.219 |
dodged | -0.212 |
brave | -0.212 |
Token | ier |
Token ID | 959 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
aman | +0.411 |
destination | +0.372 |
rap | +0.367 |
neath | +0.360 |
berry | +0.355 |
Neg logit contributions | |
protective | -0.331 |
sneaky | -0.324 |
fast | -0.317 |
ggy | -0.312 |
brave | -0.309 |
Token | than |
Token ID | 621 |
Feature activation | +4.29e-02 |
Pos logit contributions | |
aman | +0.079 |
berry | +0.076 |
umm | +0.074 |
en | +0.074 |
rap | +0.072 |
Neg logit contributions | |
protective | -0.063 |
sneaky | -0.062 |
fast | -0.058 |
dodged | -0.057 |
brave | -0.055 |
Token | before |
Token ID | 878 |
Feature activation | +4.76e-02 |
Pos logit contributions | |
aman | +0.075 |
berry | +0.072 |
en | +0.071 |
umm | +0.070 |
rap | +0.067 |
Neg logit contributions | |
protective | -0.079 |
sneaky | -0.076 |
dodged | -0.071 |
fast | -0.071 |
ars | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
aman | +0.105 |
umm | +0.102 |
berry | +0.100 |
rap | +0.099 |
en | +0.097 |
Neg logit contributions | |
fast | -0.066 |
sneaky | -0.066 |
protective | -0.065 |
brave | -0.060 |
ggy | -0.057 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
aman | +0.024 |
berry | +0.023 |
umm | +0.022 |
en | +0.022 |
rap | +0.022 |
Neg logit contributions | |
protective | -0.020 |
sneaky | -0.019 |
fast | -0.019 |
dodged | -0.018 |
brave | -0.018 |
Token | to |
Token ID | 284 |
Feature activation | +4.55e-02 |
Pos logit contributions | |
aman | +0.035 |
berry | +0.034 |
en | +0.033 |
umm | +0.033 |
rap | +0.031 |
Neg logit contributions | |
protective | -0.036 |
sneaky | -0.035 |
dodged | -0.033 |
fast | -0.033 |
villains | -0.032 |
Token | make |
Token ID | 787 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
aman | +0.087 |
berry | +0.083 |
umm | +0.082 |
en | +0.081 |
rap | +0.078 |
Neg logit contributions | |
protective | -0.078 |
sneaky | -0.076 |
fast | -0.072 |
dodged | -0.070 |
brave | -0.068 |
Token | food |
Token ID | 2057 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
protective | +0.023 |
sneaky | +0.023 |
fast | +0.022 |
brave | +0.021 |
ggy | +0.021 |
Neg logit contributions | |
aman | -0.031 |
umm | -0.029 |
berry | -0.029 |
en | -0.027 |
rap | -0.027 |
Token | look |
Token ID | 804 |
Feature activation | -2.97e-03 |
Pos logit contributions | |
protective | +0.017 |
sneaky | +0.016 |
fast | +0.016 |
brave | +0.015 |
dodged | +0.015 |
Neg logit contributions | |
aman | -0.021 |
berry | -0.020 |
umm | -0.019 |
en | -0.019 |
rap | -0.019 |
Token | y |
Token ID | 331 |
Feature activation | +8.15e-02 |
Pos logit contributions | |
protective | +0.004 |
sneaky | +0.004 |
fast | +0.004 |
dodged | +0.003 |
brave | +0.003 |
Neg logit contributions | |
aman | -0.007 |
berry | -0.007 |
umm | -0.006 |
en | -0.006 |
rap | -0.006 |
Token | umm |
Token ID | 13929 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
aman | +0.059 |
berry | +0.045 |
en | +0.040 |
rap | +0.039 |
date | +0.035 |
Neg logit contributions | |
protective | -0.233 |
sneaky | -0.229 |
fast | -0.226 |
ars | -0.217 |
brave | -0.216 |
Token | ier |
Token ID | 959 |
Feature activation | +4.24e-02 |
Pos logit contributions | |
aman | +0.419 |
berry | +0.376 |
destination | +0.367 |
amaz | +0.367 |
neath | +0.365 |
Neg logit contributions | |
ages | -0.321 |
protective | -0.313 |
fast | -0.303 |
ars | -0.302 |
sneaky | -0.299 |
Token | than |
Token ID | 621 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
aman | +0.091 |
berry | +0.086 |
umm | +0.084 |
rap | +0.082 |
en | +0.082 |
Neg logit contributions | |
protective | -0.063 |
sneaky | -0.060 |
fast | -0.059 |
brave | -0.055 |
dodged | -0.054 |
Token | before |
Token ID | 878 |
Feature activation | +4.70e-02 |
Pos logit contributions | |
aman | +0.088 |
umm | +0.082 |
berry | +0.082 |
en | +0.080 |
rap | +0.079 |
Neg logit contributions | |
protective | -0.077 |
sneaky | -0.077 |
fast | -0.075 |
brave | -0.070 |
dodged | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | +2.43e-03 |
Pos logit contributions | |
aman | +0.106 |
umm | +0.104 |
berry | +0.100 |
rap | +0.099 |
en | +0.098 |
Neg logit contributions | |
fast | -0.064 |
sneaky | -0.063 |
protective | -0.062 |
brave | -0.057 |
ggy | -0.054 |
Token | She |
Token ID | 1375 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
rap | +0.005 |
en | +0.005 |
umm | +0.005 |
aman | +0.005 |
berry | +0.005 |
Neg logit contributions | |
fast | -0.003 |
protective | -0.003 |
hog | -0.003 |
sneaky | -0.003 |
brave | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
protective | +0.054 |
sneaky | +0.052 |
fast | +0.049 |
dodged | +0.048 |
villains | +0.047 |
Neg logit contributions | |
aman | -0.078 |
berry | -0.075 |
en | -0.073 |
umm | -0.073 |
rap | -0.071 |
Token | garden |
Token ID | 11376 |
Feature activation | +6.38e-02 |
Pos logit contributions | |
aman | +0.043 |
berry | +0.042 |
umm | +0.041 |
en | +0.041 |
rap | +0.040 |
Neg logit contributions | |
protective | -0.028 |
sneaky | -0.027 |
fast | -0.025 |
dodged | -0.024 |
brave | -0.024 |
Token | full |
Token ID | 1336 |
Feature activation | +8.69e-02 |
Pos logit contributions | |
aman | +0.129 |
berry | +0.122 |
umm | +0.121 |
en | +0.121 |
rap | +0.118 |
Neg logit contributions | |
protective | -0.101 |
sneaky | -0.098 |
fast | -0.094 |
dodged | -0.089 |
brave | -0.088 |
Token | of |
Token ID | 286 |
Feature activation | +4.07e-02 |
Pos logit contributions | |
aman | +0.175 |
en | +0.165 |
berry | +0.163 |
umm | +0.162 |
rap | +0.161 |
Neg logit contributions | |
protective | -0.139 |
sneaky | -0.137 |
fast | -0.130 |
brave | -0.124 |
dodged | -0.121 |
Token | wild |
Token ID | 4295 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
aman | +0.089 |
berry | +0.084 |
umm | +0.084 |
en | +0.083 |
rap | +0.081 |
Neg logit contributions | |
protective | -0.059 |
sneaky | -0.058 |
fast | -0.054 |
brave | -0.051 |
dodged | -0.051 |
Token | fl |
Token ID | 2704 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
aman | +0.026 |
umm | +0.024 |
berry | +0.024 |
rap | +0.024 |
en | +0.024 |
Neg logit contributions | |
protective | -0.017 |
sneaky | -0.017 |
fast | -0.016 |
brave | -0.015 |
villains | -0.015 |
Token | owers |
Token ID | 3618 |
Feature activation | +9.41e-02 |
Pos logit contributions | |
aman | +0.336 |
rap | +0.307 |
en | +0.306 |
destination | +0.294 |
amaz | +0.290 |
Neg logit contributions | |
ars | -0.388 |
ages | -0.385 |
ards | -0.366 |
protective | -0.363 |
hog | -0.362 |
Token | . |
Token ID | 13 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
aman | +0.180 |
berry | +0.172 |
umm | +0.169 |
en | +0.168 |
rap | +0.162 |
Neg logit contributions | |
protective | -0.159 |
sneaky | -0.155 |
fast | -0.146 |
dodged | -0.144 |
villains | -0.140 |
Token | He |
Token ID | 679 |
Feature activation | +8.63e-03 |
Pos logit contributions | |
aman | +0.033 |
berry | +0.031 |
en | +0.031 |
umm | +0.031 |
rap | +0.030 |
Neg logit contributions | |
protective | -0.026 |
sneaky | -0.026 |
fast | -0.024 |
dodged | -0.024 |
brave | -0.023 |
Token | decided |
Token ID | 3066 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
aman | +0.017 |
berry | +0.017 |
en | +0.016 |
umm | +0.016 |
rap | +0.015 |
Neg logit contributions | |
protective | -0.014 |
sneaky | -0.013 |
fast | -0.012 |
dodged | -0.012 |
villains | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
aman | +0.038 |
en | +0.035 |
berry | +0.035 |
rap | +0.034 |
umm | +0.033 |
Neg logit contributions | |
protective | -0.044 |
sneaky | -0.043 |
dodged | -0.041 |
ars | -0.040 |
villains | -0.040 |
Token | me |
Token ID | 502 |
Feature activation | +9.64e-02 |
Pos logit contributions | |
aman | +0.101 |
umm | +0.099 |
berry | +0.096 |
rap | +0.094 |
icer | +0.090 |
Neg logit contributions | |
protective | -0.095 |
fast | -0.094 |
sneaky | -0.092 |
brave | -0.088 |
strong | -0.083 |
Token | adow |
Token ID | 4584 |
Feature activation | -4.69e-03 |
Pos logit contributions | |
aman | +0.156 |
berry | +0.147 |
umm | +0.145 |
en | +0.143 |
rap | +0.138 |
Neg logit contributions | |
protective | -0.192 |
sneaky | -0.188 |
fast | -0.180 |
dodged | -0.176 |
brave | -0.172 |
Token | full |
Token ID | 1336 |
Feature activation | +8.71e-02 |
Pos logit contributions | |
protective | +0.008 |
sneaky | +0.008 |
fast | +0.008 |
dodged | +0.007 |
villains | +0.007 |
Neg logit contributions | |
aman | -0.009 |
berry | -0.008 |
umm | -0.008 |
en | -0.008 |
rap | -0.008 |
Token | of |
Token ID | 286 |
Feature activation | +5.06e-02 |
Pos logit contributions | |
aman | +0.169 |
berry | +0.160 |
umm | +0.158 |
en | +0.157 |
rap | +0.154 |
Neg logit contributions | |
protective | -0.146 |
sneaky | -0.142 |
fast | -0.136 |
brave | -0.129 |
dodged | -0.129 |
Token | sun |
Token ID | 4252 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
aman | +0.110 |
berry | +0.105 |
en | +0.103 |
umm | +0.103 |
rap | +0.099 |
Neg logit contributions | |
protective | -0.072 |
sneaky | -0.070 |
fast | -0.065 |
dodged | -0.065 |
villains | -0.062 |
Token | fl |
Token ID | 2704 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
aman | +0.177 |
umm | +0.166 |
rap | +0.166 |
icer | +0.159 |
berry | +0.158 |
Neg logit contributions | |
protective | -0.236 |
fast | -0.232 |
sneaky | -0.225 |
brave | -0.219 |
ages | -0.219 |
Token | owers |
Token ID | 3618 |
Feature activation | +9.76e-02 |
Pos logit contributions | |
aman | +0.304 |
rap | +0.290 |
en | +0.275 |
berry | +0.262 |
umm | +0.258 |
Neg logit contributions | |
ars | -0.401 |
protective | -0.392 |
fast | -0.386 |
sneaky | -0.386 |
hog | -0.383 |
Token | and |
Token ID | 290 |
Feature activation | +5.20e-02 |
Pos logit contributions | |
aman | +0.186 |
berry | +0.177 |
umm | +0.174 |
en | +0.173 |
rap | +0.168 |
Neg logit contributions | |
protective | -0.166 |
sneaky | -0.161 |
fast | -0.153 |
dodged | -0.149 |
brave | -0.146 |
Token | tall |
Token ID | 7331 |
Feature activation | +5.14e-02 |
Pos logit contributions | |
aman | +0.101 |
berry | +0.096 |
en | +0.095 |
umm | +0.094 |
rap | +0.090 |
Neg logit contributions | |
protective | -0.085 |
sneaky | -0.082 |
dodged | -0.077 |
fast | -0.076 |
ggy | -0.076 |
Token | grass |
Token ID | 8701 |
Feature activation | +7.69e-02 |
Pos logit contributions | |
aman | +0.103 |
berry | +0.098 |
umm | +0.097 |
en | +0.096 |
rap | +0.093 |
Neg logit contributions | |
protective | -0.082 |
sneaky | -0.080 |
fast | -0.076 |
dodged | -0.074 |
brave | -0.072 |
Token | . |
Token ID | 13 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
aman | +0.148 |
berry | +0.141 |
umm | +0.138 |
en | +0.137 |
rap | +0.133 |
Neg logit contributions | |
protective | -0.130 |
sneaky | -0.126 |
fast | -0.120 |
dodged | -0.117 |
villains | -0.113 |
Token | magic |
Token ID | 5536 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
aman | +0.031 |
en | +0.029 |
berry | +0.029 |
rap | +0.028 |
umm | +0.028 |
Neg logit contributions | |
sneaky | -0.021 |
protective | -0.021 |
fast | -0.019 |
brave | -0.019 |
cunning | -0.018 |
Token | apple |
Token ID | 17180 |
Feature activation | +7.28e-02 |
Pos logit contributions | |
aman | +0.043 |
berry | +0.041 |
en | +0.040 |
umm | +0.040 |
rap | +0.039 |
Neg logit contributions | |
protective | -0.035 |
sneaky | -0.034 |
fast | -0.032 |
dodged | -0.031 |
brave | -0.031 |
Token | was |
Token ID | 373 |
Feature activation | +4.86e-02 |
Pos logit contributions | |
aman | +0.129 |
berry | +0.126 |
en | +0.123 |
umm | +0.122 |
rap | +0.116 |
Neg logit contributions | |
protective | -0.129 |
sneaky | -0.125 |
fast | -0.118 |
dodged | -0.115 |
villains | -0.114 |
Token | the |
Token ID | 262 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
aman | +0.105 |
berry | +0.101 |
umm | +0.100 |
en | +0.098 |
rap | +0.095 |
Neg logit contributions | |
protective | -0.072 |
sneaky | -0.070 |
fast | -0.066 |
dodged | -0.063 |
brave | -0.063 |
Token | y |
Token ID | 331 |
Feature activation | +7.06e-02 |
Pos logit contributions | |
aman | +0.028 |
berry | +0.027 |
en | +0.027 |
rap | +0.026 |
umm | +0.025 |
Neg logit contributions | |
sneaky | -0.020 |
protective | -0.020 |
fast | -0.019 |
brave | -0.019 |
strong | -0.017 |
Token | umm |
Token ID | 13929 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
aman | +0.050 |
rap | +0.034 |
berry | +0.033 |
?' | +0.032 |
en | +0.030 |
Neg logit contributions | |
protective | -0.196 |
sneaky | -0.194 |
fast | -0.190 |
hog | -0.186 |
ars | -0.183 |
Token | iest |
Token ID | 6386 |
Feature activation | +6.08e-02 |
Pos logit contributions | |
aman | +0.461 |
rap | +0.426 |
destination | +0.420 |
icer | +0.411 |
berry | +0.409 |
Neg logit contributions | |
protective | -0.252 |
sneaky | -0.238 |
fast | -0.237 |
ars | -0.235 |
ggy | -0.232 |
Token | thing |
Token ID | 1517 |
Feature activation | +5.12e-02 |
Pos logit contributions | |
aman | +0.127 |
rap | +0.120 |
icer | +0.117 |
en | +0.115 |
neath | +0.113 |
Neg logit contributions | |
sneaky | -0.093 |
protective | -0.092 |
brave | -0.089 |
fast | -0.088 |
sharp | -0.082 |
Token | they |
Token ID | 484 |
Feature activation | +5.99e-02 |
Pos logit contributions | |
aman | +0.109 |
berry | +0.105 |
en | +0.103 |
umm | +0.102 |
rap | +0.099 |
Neg logit contributions | |
protective | -0.075 |
sneaky | -0.073 |
fast | -0.067 |
dodged | -0.066 |
villains | -0.065 |
Token | ever |
Token ID | 1683 |
Feature activation | +5.35e-02 |
Pos logit contributions | |
aman | +0.132 |
berry | +0.127 |
en | +0.125 |
umm | +0.125 |
rap | +0.121 |
Neg logit contributions | |
protective | -0.083 |
sneaky | -0.080 |
fast | -0.074 |
dodged | -0.072 |
ggy | -0.070 |
Token | tasted |
Token ID | 29187 |
Feature activation | +8.21e-02 |
Pos logit contributions | |
aman | +0.122 |
umm | +0.114 |
berry | +0.114 |
rap | +0.110 |
en | +0.109 |
Neg logit contributions | |
protective | -0.074 |
sneaky | -0.073 |
fast | -0.069 |
dodged | -0.068 |
brave | -0.065 |
Token | sweet |
Token ID | 6029 |
Feature activation | +6.86e-02 |
Pos logit contributions | |
aman | +0.067 |
berry | +0.064 |
umm | +0.064 |
en | +0.063 |
rap | +0.061 |
Neg logit contributions | |
protective | -0.042 |
sneaky | -0.041 |
fast | -0.038 |
dodged | -0.037 |
villains | -0.037 |
Token | smell |
Token ID | 8508 |
Feature activation | +8.48e-03 |
Pos logit contributions | |
aman | +0.139 |
rap | +0.129 |
icer | +0.126 |
umm | +0.124 |
berry | +0.123 |
Neg logit contributions | |
protective | -0.113 |
sneaky | -0.107 |
fast | -0.103 |
brave | -0.099 |
sharp | -0.093 |
Token | of |
Token ID | 286 |
Feature activation | +6.09e-02 |
Pos logit contributions | |
aman | +0.016 |
berry | +0.015 |
en | +0.015 |
umm | +0.015 |
rap | +0.014 |
Neg logit contributions | |
protective | -0.015 |
sneaky | -0.014 |
fast | -0.013 |
dodged | -0.013 |
villains | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | +4.39e-02 |
Pos logit contributions | |
aman | +0.127 |
berry | +0.121 |
umm | +0.120 |
en | +0.119 |
rap | +0.116 |
Neg logit contributions | |
protective | -0.093 |
sneaky | -0.091 |
fast | -0.086 |
dodged | -0.082 |
brave | -0.081 |
Token | wild |
Token ID | 4295 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
aman | +0.087 |
berry | +0.085 |
umm | +0.083 |
en | +0.082 |
rap | +0.078 |
Neg logit contributions | |
protective | -0.070 |
sneaky | -0.068 |
dodged | -0.064 |
fast | -0.064 |
villains | -0.063 |
Token | fl |
Token ID | 2704 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
aman | +0.022 |
berry | +0.021 |
umm | +0.020 |
en | +0.020 |
rap | +0.020 |
Neg logit contributions | |
protective | -0.017 |
sneaky | -0.017 |
fast | -0.016 |
dodged | -0.015 |
villains | -0.015 |
Token | owers |
Token ID | 3618 |
Feature activation | +8.65e-02 |
Pos logit contributions | |
aman | +0.284 |
rap | +0.259 |
en | +0.247 |
berry | +0.236 |
destination | +0.228 |
Neg logit contributions | |
protective | -0.440 |
sneaky | -0.426 |
ars | -0.425 |
hog | -0.411 |
ages | -0.411 |
Token | along |
Token ID | 1863 |
Feature activation | +1.95e-03 |
Pos logit contributions | |
aman | +0.164 |
berry | +0.156 |
umm | +0.154 |
en | +0.152 |
rap | +0.148 |
Neg logit contributions | |
protective | -0.148 |
sneaky | -0.144 |
fast | -0.137 |
dodged | -0.133 |
brave | -0.130 |
Token | the |
Token ID | 262 |
Feature activation | +5.22e-02 |
Pos logit contributions | |
aman | +0.004 |
en | +0.004 |
berry | +0.004 |
umm | +0.003 |
rap | +0.003 |
Neg logit contributions | |
protective | -0.003 |
sneaky | -0.003 |
dodged | -0.003 |
ars | -0.003 |
villains | -0.003 |
Token | creek |
Token ID | 39459 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
aman | +0.096 |
en | +0.093 |
berry | +0.092 |
umm | +0.091 |
destination | +0.085 |
Neg logit contributions | |
protective | -0.088 |
sneaky | -0.087 |
dodged | -0.085 |
ggy | -0.080 |
villains | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
aman | +0.022 |
berry | +0.021 |
en | +0.021 |
umm | +0.020 |
rap | +0.019 |
Neg logit contributions | |
protective | -0.022 |
sneaky | -0.022 |
dodged | -0.020 |
villains | -0.020 |
fast | -0.020 |
Token | is |
Token ID | 318 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
protective | +0.080 |
villains | +0.078 |
sneaky | +0.078 |
fast | +0.074 |
cious | +0.074 |
Neg logit contributions | |
berry | -0.051 |
en | -0.051 |
aman | -0.048 |
el | -0.047 |
rap | -0.045 |
Token | stronger |
Token ID | 7387 |
Feature activation | -9.73e-03 |
Pos logit contributions | |
en | +0.042 |
rap | +0.041 |
asleep | +0.041 |
once | +0.040 |
el | +0.040 |
Neg logit contributions | |
ages | -0.042 |
ars | -0.042 |
cious | -0.041 |
villains | -0.040 |
hog | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | -7.07e-04 |
Pos logit contributions | |
protective | +0.019 |
sneaky | +0.018 |
dodged | +0.017 |
cunning | +0.017 |
villains | +0.016 |
Neg logit contributions | |
aman | -0.016 |
en | -0.016 |
berry | -0.015 |
umm | -0.014 |
rap | -0.014 |
Token | He |
Token ID | 679 |
Feature activation | -4.91e-03 |
Pos logit contributions | |
protective | +0.001 |
sneaky | +0.001 |
fast | +0.001 |
dodged | +0.001 |
villains | +0.001 |
Neg logit contributions | |
aman | -0.001 |
berry | -0.001 |
umm | -0.001 |
en | -0.001 |
rap | -0.001 |
Token | dr |
Token ID | 1553 |
Feature activation | +1.20e-03 |
Pos logit contributions | |
protective | +0.008 |
sneaky | +0.008 |
fast | +0.007 |
villains | +0.007 |
ggy | +0.007 |
Neg logit contributions | |
aman | -0.009 |
berry | -0.009 |
en | -0.009 |
rap | -0.008 |
umm | -0.008 |
Token | ags |
Token ID | 3775 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
aman | +0.002 |
umm | +0.002 |
berry | +0.002 |
en | +0.002 |
rap | +0.002 |
Neg logit contributions | |
protective | -0.002 |
fast | -0.002 |
sneaky | -0.002 |
ars | -0.002 |
dodged | -0.002 |
Token | Lily |
Token ID | 20037 |
Feature activation | -7.12e-02 |
Pos logit contributions | |
protective | +0.364 |
sneaky | +0.357 |
dodged | +0.329 |
villains | +0.328 |
ars | +0.326 |
Neg logit contributions | |
aman | -0.277 |
en | -0.271 |
berry | -0.264 |
rap | -0.250 |
el | -0.245 |
Token | into |
Token ID | 656 |
Feature activation | -9.19e-02 |
Pos logit contributions | |
protective | +0.132 |
sneaky | +0.131 |
villains | +0.121 |
dodged | +0.121 |
cunning | +0.119 |
Neg logit contributions | |
aman | -0.121 |
en | -0.116 |
berry | -0.115 |
rap | -0.111 |
umm | -0.105 |
Token | the |
Token ID | 262 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
protective | +0.171 |
sneaky | +0.167 |
fast | +0.159 |
dodged | +0.156 |
villains | +0.152 |
Neg logit contributions | |
aman | -0.162 |
berry | -0.154 |
umm | -0.150 |
en | -0.150 |
rap | -0.144 |
Token | pond |
Token ID | 16723 |
Feature activation | -9.16e-02 |
Pos logit contributions | |
protective | +0.050 |
sneaky | +0.049 |
dodged | +0.047 |
fast | +0.046 |
villains | +0.045 |
Neg logit contributions | |
aman | -0.049 |
berry | -0.048 |
umm | -0.046 |
en | -0.046 |
rap | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | -5.34e-03 |
Pos logit contributions | |
protective | +0.183 |
sneaky | +0.179 |
dodged | +0.173 |
villains | +0.172 |
ggy | +0.167 |
Neg logit contributions | |
aman | -0.141 |
berry | -0.139 |
en | -0.137 |
rap | -0.127 |
umm | -0.124 |
Token | opens |
Token ID | 9808 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
protective | +0.054 |
sneaky | +0.053 |
ggy | +0.052 |
villains | +0.051 |
ars | +0.051 |
Neg logit contributions | |
aman | -0.047 |
en | -0.045 |
berry | -0.043 |
neath | -0.041 |
rap | -0.041 |
Token | it |
Token ID | 340 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
umm | +0.032 |
aman | +0.031 |
berry | +0.030 |
en | +0.028 |
rap | +0.028 |
Neg logit contributions | |
fast | -0.020 |
protective | -0.019 |
sneaky | -0.018 |
brave | -0.017 |
ggy | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -7.33e-03 |
Pos logit contributions | |
protective | +0.021 |
fast | +0.021 |
sneaky | +0.020 |
brave | +0.019 |
dodged | +0.018 |
Neg logit contributions | |
aman | -0.034 |
umm | -0.032 |
berry | -0.032 |
en | -0.031 |
rap | -0.030 |
Token | He |
Token ID | 679 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
protective | +0.011 |
sneaky | +0.011 |
fast | +0.010 |
dodged | +0.010 |
villains | +0.010 |
Neg logit contributions | |
aman | -0.015 |
berry | -0.014 |
umm | -0.014 |
en | -0.014 |
rap | -0.014 |
Token | dr |
Token ID | 1553 |
Feature activation | +2.18e-03 |
Pos logit contributions | |
protective | +0.028 |
sneaky | +0.027 |
fast | +0.026 |
dodged | +0.025 |
brave | +0.025 |
Neg logit contributions | |
aman | -0.028 |
berry | -0.027 |
umm | -0.027 |
en | -0.026 |
rap | -0.025 |
Token | ags |
Token ID | 3775 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
aman | +0.005 |
umm | +0.004 |
berry | +0.004 |
en | +0.004 |
rap | +0.004 |
Neg logit contributions | |
fast | -0.003 |
protective | -0.003 |
sneaky | -0.003 |
ars | -0.003 |
brave | -0.003 |
Token | Sue |
Token ID | 26089 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
protective | +0.322 |
sneaky | +0.315 |
dodged | +0.291 |
fast | +0.287 |
villains | +0.285 |
Neg logit contributions | |
aman | -0.320 |
berry | -0.304 |
en | -0.302 |
rap | -0.288 |
umm | -0.286 |
Token | to |
Token ID | 284 |
Feature activation | -5.00e-02 |
Pos logit contributions | |
protective | +0.178 |
sneaky | +0.172 |
fast | +0.168 |
dodged | +0.160 |
brave | +0.158 |
Neg logit contributions | |
aman | -0.194 |
umm | -0.186 |
berry | -0.184 |
en | -0.178 |
rap | -0.172 |
Token | the |
Token ID | 262 |
Feature activation | -5.23e-02 |
Pos logit contributions | |
protective | +0.074 |
sneaky | +0.073 |
fast | +0.072 |
brave | +0.067 |
dodged | +0.066 |
Neg logit contributions | |
aman | -0.107 |
umm | -0.104 |
berry | -0.102 |
en | -0.100 |
rap | -0.096 |
Token | street |
Token ID | 4675 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
protective | +0.105 |
sneaky | +0.102 |
dodged | +0.099 |
fast | +0.096 |
ars | +0.095 |
Neg logit contributions | |
aman | -0.082 |
berry | -0.076 |
umm | -0.075 |
en | -0.074 |
rap | -0.072 |
Token | . |
Token ID | 13 |
Feature activation | -8.76e-03 |
Pos logit contributions | |
protective | +0.224 |
sneaky | +0.218 |
fast | +0.205 |
dodged | +0.202 |
villains | +0.196 |
Neg logit contributions | |
aman | -0.250 |
berry | -0.238 |
umm | -0.234 |
en | -0.233 |
rap | -0.225 |
Token | the |
Token ID | 262 |
Feature activation | -9.95e-03 |
Pos logit contributions | |
protective | +0.110 |
sneaky | +0.108 |
ars | +0.103 |
dodged | +0.102 |
villains | +0.099 |
Neg logit contributions | |
aman | -0.103 |
en | -0.100 |
berry | -0.094 |
rap | -0.093 |
umm | -0.089 |
Token | arms |
Token ID | 5101 |
Feature activation | -6.93e-02 |
Pos logit contributions | |
protective | +0.019 |
sneaky | +0.019 |
fast | +0.018 |
dodged | +0.018 |
villains | +0.017 |
Neg logit contributions | |
aman | -0.017 |
berry | -0.016 |
umm | -0.015 |
en | -0.015 |
rap | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +6.92e-03 |
Pos logit contributions | |
protective | +0.118 |
sneaky | +0.115 |
fast | +0.107 |
dodged | +0.106 |
villains | +0.103 |
Neg logit contributions | |
aman | -0.132 |
berry | -0.126 |
en | -0.124 |
umm | -0.123 |
rap | -0.119 |
Token | She |
Token ID | 1375 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
aman | +0.010 |
berry | +0.009 |
umm | +0.009 |
en | +0.009 |
rap | +0.008 |
Neg logit contributions | |
protective | -0.015 |
sneaky | -0.015 |
dodged | -0.014 |
ggy | -0.014 |
fast | -0.014 |
Token | dr |
Token ID | 1553 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
aman | +0.029 |
berry | +0.027 |
en | +0.027 |
umm | +0.027 |
rap | +0.026 |
Neg logit contributions | |
protective | -0.026 |
sneaky | -0.025 |
fast | -0.024 |
dodged | -0.023 |
villains | -0.023 |
Token | ags |
Token ID | 3775 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
aman | +0.026 |
umm | +0.025 |
berry | +0.024 |
en | +0.022 |
?' | +0.021 |
Neg logit contributions | |
fast | -0.018 |
protective | -0.017 |
sneaky | -0.017 |
ars | -0.016 |
brave | -0.016 |
Token | them |
Token ID | 606 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
protective | +0.329 |
sneaky | +0.320 |
dodged | +0.297 |
villains | +0.292 |
ars | +0.290 |
Neg logit contributions | |
aman | -0.314 |
en | -0.299 |
berry | -0.298 |
rap | -0.283 |
umm | -0.276 |
Token | to |
Token ID | 284 |
Feature activation | -4.10e-02 |
Pos logit contributions | |
protective | +0.255 |
sneaky | +0.250 |
villains | +0.233 |
dodged | +0.232 |
cunning | +0.228 |
Neg logit contributions | |
aman | -0.216 |
en | -0.205 |
berry | -0.204 |
rap | -0.197 |
umm | -0.190 |
Token | the |
Token ID | 262 |
Feature activation | -4.80e-02 |
Pos logit contributions | |
protective | +0.060 |
sneaky | +0.059 |
fast | +0.057 |
dodged | +0.053 |
brave | +0.053 |
Neg logit contributions | |
aman | -0.088 |
umm | -0.085 |
berry | -0.085 |
en | -0.082 |
rap | -0.079 |
Token | office |
Token ID | 2607 |
Feature activation | -6.98e-02 |
Pos logit contributions | |
protective | +0.096 |
sneaky | +0.094 |
dodged | +0.092 |
ars | +0.088 |
fast | +0.088 |
Neg logit contributions | |
aman | -0.073 |
berry | -0.069 |
umm | -0.068 |
en | -0.067 |
rap | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | -7.90e-03 |
Pos logit contributions | |
protective | +0.131 |
sneaky | +0.128 |
dodged | +0.121 |
villains | +0.117 |
cunning | +0.117 |
Neg logit contributions | |
aman | -0.119 |
en | -0.111 |
berry | -0.110 |
rap | -0.107 |
umm | -0.106 |
Token | . |
Token ID | 13 |
Feature activation | -4.71e-03 |
Pos logit contributions | |
protective | +0.098 |
sneaky | +0.097 |
dodged | +0.090 |
villains | +0.088 |
cunning | +0.086 |
Neg logit contributions | |
aman | -0.088 |
en | -0.087 |
berry | -0.085 |
rap | -0.082 |
umm | -0.081 |
Token |
|
Token ID | 198 |
Feature activation | -3.12e-04 |
Pos logit contributions | |
protective | +0.010 |
sneaky | +0.009 |
dodged | +0.009 |
fast | +0.009 |
ggy | +0.009 |
Neg logit contributions | |
aman | -0.007 |
berry | -0.007 |
umm | -0.007 |
en | -0.007 |
rap | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -8.88e-03 |
Pos logit contributions | |
protective | +0.001 |
sneaky | +0.001 |
fast | +0.001 |
dodged | +0.001 |
villains | +0.001 |
Neg logit contributions | |
aman | -0.001 |
berry | -0.001 |
en | -0.001 |
umm | -0.000 |
rap | -0.000 |
Token | " |
Token ID | 1 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
protective | +0.016 |
sneaky | +0.016 |
dodged | +0.015 |
villains | +0.015 |
cunning | +0.015 |
Neg logit contributions | |
aman | -0.016 |
berry | -0.014 |
rap | -0.013 |
en | -0.013 |
?' | -0.013 |
Token | K |
Token ID | 42 |
Feature activation | -6.03e-02 |
Pos logit contributions | |
protective | +0.056 |
sneaky | +0.055 |
ggy | +0.051 |
fast | +0.051 |
dodged | +0.051 |
Neg logit contributions | |
aman | -0.050 |
berry | -0.047 |
umm | -0.044 |
en | -0.044 |
date | -0.042 |
Token | ang |
Token ID | 648 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
protective | +0.112 |
sneaky | +0.109 |
fast | +0.105 |
dodged | +0.102 |
ars | +0.101 |
Neg logit contributions | |
aman | -0.104 |
umm | -0.100 |
berry | -0.096 |
en | -0.096 |
rap | -0.093 |
Token | aro |
Token ID | 12022 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
protective | +0.193 |
sneaky | +0.192 |
villains | +0.177 |
hind | +0.160 |
fast | +0.159 |
Neg logit contributions | |
berry | -0.418 |
el | -0.401 |
aman | -0.398 |
en | -0.395 |
?' | -0.383 |
Token | os |
Token ID | 418 |
Feature activation | -8.87e-02 |
Pos logit contributions | |
protective | +0.299 |
sneaky | +0.292 |
fast | +0.277 |
villains | +0.269 |
dodged | +0.269 |
Neg logit contributions | |
aman | -0.223 |
berry | -0.215 |
en | -0.209 |
umm | -0.201 |
rap | -0.197 |
Token | are |
Token ID | 389 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
protective | +0.184 |
sneaky | +0.181 |
villains | +0.169 |
fast | +0.164 |
ggy | +0.164 |
Neg logit contributions | |
en | -0.129 |
aman | -0.129 |
berry | -0.128 |
rap | -0.117 |
el | -0.116 |
Token | wild |
Token ID | 4295 |
Feature activation | -4.17e-02 |
Pos logit contributions | |
rap | +0.029 |
en | +0.029 |
once | +0.029 |
twenty | +0.028 |
enough | +0.028 |
Neg logit contributions | |
hog | -0.024 |
cious | -0.023 |
ike | -0.022 |
suit | -0.022 |
villains | -0.021 |
Token | animals |
Token ID | 4695 |
Feature activation | -9.70e-03 |
Pos logit contributions | |
protective | +0.078 |
sneaky | +0.076 |
dodged | +0.073 |
fast | +0.070 |
cunning | +0.069 |
Neg logit contributions | |
aman | -0.069 |
berry | -0.069 |
en | -0.069 |
umm | -0.066 |
rap | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
aman | +0.060 |
berry | +0.058 |
en | +0.056 |
rap | +0.053 |
umm | +0.052 |
Neg logit contributions | |
protective | -0.075 |
sneaky | -0.074 |
villains | -0.069 |
dodged | -0.069 |
hog | -0.067 |
Token |
|
Token ID | 198 |
Feature activation | -2.74e-03 |
Pos logit contributions | |
protective | +0.025 |
sneaky | +0.024 |
fast | +0.023 |
dodged | +0.022 |
villains | +0.022 |
Neg logit contributions | |
aman | -0.029 |
berry | -0.027 |
umm | -0.027 |
en | -0.027 |
rap | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | -8.90e-03 |
Pos logit contributions | |
protective | +0.005 |
sneaky | +0.005 |
dodged | +0.005 |
fast | +0.005 |
villains | +0.005 |
Neg logit contributions | |
aman | -0.005 |
berry | -0.004 |
en | -0.004 |
umm | -0.004 |
rap | -0.004 |
Token | " |
Token ID | 1 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
protective | +0.015 |
sneaky | +0.015 |
dodged | +0.014 |
cunning | +0.013 |
hog | +0.013 |
Neg logit contributions | |
aman | -0.017 |
berry | -0.016 |
en | -0.015 |
rap | -0.015 |
date | -0.014 |
Token | By |
Token ID | 3886 |
Feature activation | +2.19e-02 |
Pos logit contributions | |
protective | +0.058 |
sneaky | +0.056 |
fast | +0.052 |
dodged | +0.051 |
hog | +0.051 |
Neg logit contributions | |
aman | -0.053 |
berry | -0.049 |
umm | -0.048 |
date | -0.045 |
en | -0.045 |
Token | e |
Token ID | 68 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
aman | +0.047 |
umm | +0.045 |
berry | +0.044 |
en | +0.042 |
rap | +0.040 |
Neg logit contributions | |
protective | -0.033 |
sneaky | -0.032 |
fast | -0.032 |
brave | -0.031 |
ggy | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
protective | +0.289 |
sneaky | +0.283 |
fast | +0.269 |
dodged | +0.263 |
ggy | +0.258 |
Neg logit contributions | |
aman | -0.329 |
umm | -0.312 |
berry | -0.311 |
en | -0.305 |
rap | -0.295 |
Token | doctor |
Token ID | 6253 |
Feature activation | -3.61e-02 |
Pos logit contributions | |
protective | +0.057 |
sneaky | +0.054 |
dodged | +0.051 |
ars | +0.050 |
fast | +0.050 |
Neg logit contributions | |
aman | -0.071 |
berry | -0.067 |
en | -0.065 |
umm | -0.065 |
rap | -0.062 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.96e-03 |
Pos logit contributions | |
protective | +0.061 |
sneaky | +0.060 |
fast | +0.057 |
dodged | +0.056 |
brave | +0.055 |
Neg logit contributions | |
aman | -0.069 |
berry | -0.066 |
umm | -0.065 |
en | -0.064 |
rap | -0.062 |
Token | . |
Token ID | 13 |
Feature activation | -8.26e-03 |
Pos logit contributions | |
fast | +0.003 |
protective | +0.003 |
sneaky | +0.003 |
brave | +0.003 |
ggy | +0.003 |
Neg logit contributions | |
aman | -0.004 |
umm | -0.004 |
berry | -0.004 |
en | -0.004 |
rap | -0.004 |
Token | I |
Token ID | 314 |
Feature activation | +3.55e-02 |
Pos logit contributions | |
protective | +0.014 |
sneaky | +0.013 |
fast | +0.013 |
villains | +0.012 |
brave | +0.012 |
Neg logit contributions | |
aman | -0.016 |
en | -0.015 |
berry | -0.015 |
rap | -0.015 |
umm | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | +2.26e-03 |
Pos logit contributions | |
protective | +0.082 |
sneaky | +0.080 |
fast | +0.074 |
dodged | +0.074 |
villains | +0.072 |
Neg logit contributions | |
aman | -0.087 |
berry | -0.084 |
en | -0.082 |
umm | -0.081 |
rap | -0.079 |
Token |
|
Token ID | 198 |
Feature activation | +3.90e-03 |
Pos logit contributions | |
aman | +0.004 |
berry | +0.004 |
umm | +0.004 |
en | +0.004 |
rap | +0.004 |
Neg logit contributions | |
protective | -0.004 |
sneaky | -0.004 |
fast | -0.004 |
dodged | -0.004 |
villains | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -4.71e-03 |
Pos logit contributions | |
aman | +0.007 |
berry | +0.006 |
en | +0.006 |
umm | +0.006 |
rap | +0.006 |
Neg logit contributions | |
protective | -0.007 |
sneaky | -0.007 |
fast | -0.007 |
dodged | -0.007 |
villains | -0.006 |
Token | " |
Token ID | 1 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
protective | +0.009 |
sneaky | +0.009 |
villains | +0.008 |
cunning | +0.008 |
dodged | +0.008 |
Neg logit contributions | |
aman | -0.008 |
berry | -0.008 |
rap | -0.007 |
en | -0.007 |
date | -0.007 |
Token | By |
Token ID | 3886 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
protective | +0.060 |
sneaky | +0.058 |
fast | +0.055 |
dodged | +0.054 |
ggy | +0.053 |
Neg logit contributions | |
aman | -0.049 |
berry | -0.047 |
umm | -0.045 |
en | -0.045 |
rap | -0.042 |
Token | e |
Token ID | 68 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
aman | +0.038 |
umm | +0.035 |
berry | +0.035 |
en | +0.034 |
rap | +0.032 |
Neg logit contributions | |
protective | -0.027 |
sneaky | -0.027 |
fast | -0.026 |
brave | -0.025 |
dodged | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | -3.14e-02 |
Pos logit contributions | |
protective | +0.301 |
sneaky | +0.294 |
fast | +0.278 |
dodged | +0.272 |
villains | +0.266 |
Neg logit contributions | |
aman | -0.315 |
berry | -0.301 |
umm | -0.293 |
en | -0.293 |
rap | -0.283 |
Token | bye |
Token ID | 33847 |
Feature activation | -3.19e-02 |
Pos logit contributions | |
protective | +0.045 |
ages | +0.044 |
sneaky | +0.042 |
dodged | +0.040 |
hog | +0.039 |
Neg logit contributions | |
berry | -0.065 |
aman | -0.065 |
en | -0.060 |
umm | -0.059 |
destination | -0.059 |
Token | , |
Token ID | 11 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
fast | +0.047 |
sneaky | +0.046 |
protective | +0.045 |
brave | +0.043 |
ggy | +0.041 |
Neg logit contributions | |
aman | -0.069 |
umm | -0.064 |
en | -0.062 |
berry | -0.061 |
rap | -0.061 |
Token | bye |
Token ID | 33847 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
protective | +0.017 |
sneaky | +0.016 |
ages | +0.016 |
dodged | +0.015 |
ars | +0.015 |
Neg logit contributions | |
aman | -0.030 |
berry | -0.030 |
en | -0.028 |
umm | -0.028 |
rap | -0.027 |
Token | !" |
Token ID | 2474 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
fast | +0.024 |
sneaky | +0.023 |
protective | +0.023 |
brave | +0.022 |
ggy | +0.021 |
Neg logit contributions | |
aman | -0.035 |
umm | -0.033 |
berry | -0.032 |
en | -0.032 |
rap | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | -7.84e-03 |
Pos logit contributions | |
protective | +0.163 |
ars | +0.159 |
sneaky | +0.157 |
dodged | +0.154 |
Fin | +0.152 |
Neg logit contributions | |
en | -0.098 |
berry | -0.091 |
rap | -0.089 |
aman | -0.087 |
again | -0.083 |
Token |
|
Token ID | 198 |
Feature activation | -4.74e-03 |
Pos logit contributions | |
protective | +0.016 |
sneaky | +0.016 |
dodged | +0.014 |
cunning | +0.014 |
hog | +0.014 |
Neg logit contributions | |
aman | -0.012 |
berry | -0.011 |
umm | -0.011 |
rap | -0.011 |
en | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
protective | +0.010 |
sneaky | +0.009 |
dodged | +0.009 |
villains | +0.009 |
cunning | +0.009 |
Neg logit contributions | |
aman | -0.007 |
berry | -0.007 |
rap | -0.007 |
en | -0.007 |
umm | -0.006 |
Token | " |
Token ID | 1 |
Feature activation | -3.40e-02 |
Pos logit contributions | |
protective | +0.022 |
sneaky | +0.021 |
cunning | +0.020 |
villains | +0.019 |
hog | +0.019 |
Neg logit contributions | |
aman | -0.023 |
rap | -0.021 |
berry | -0.020 |
en | -0.019 |
date | -0.019 |
Token | By |
Token ID | 3886 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
protective | +0.061 |
sneaky | +0.059 |
fast | +0.056 |
dodged | +0.055 |
villains | +0.054 |
Neg logit contributions | |
aman | -0.063 |
berry | -0.060 |
umm | -0.059 |
en | -0.058 |
rap | -0.056 |
Token | e |
Token ID | 68 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
aman | +0.035 |
umm | +0.033 |
berry | +0.033 |
en | +0.031 |
rap | +0.030 |
Neg logit contributions | |
protective | -0.024 |
sneaky | -0.024 |
fast | -0.023 |
brave | -0.022 |
dodged | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | -3.59e-02 |
Pos logit contributions | |
protective | +0.277 |
sneaky | +0.271 |
fast | +0.257 |
dodged | +0.253 |
ggy | +0.252 |
Neg logit contributions | |
aman | -0.334 |
umm | -0.318 |
berry | -0.316 |
en | -0.313 |
rap | -0.299 |
Token | goat |
Token ID | 26917 |
Feature activation | -7.25e-02 |
Pos logit contributions | |
protective | +0.049 |
sneaky | +0.046 |
dodged | +0.045 |
villains | +0.045 |
ars | +0.044 |
Neg logit contributions | |
aman | -0.082 |
berry | -0.074 |
umm | -0.072 |
rap | -0.071 |
en | -0.070 |
Token | !" |
Token ID | 2474 |
Feature activation | -4.11e-02 |
Pos logit contributions | |
protective | +0.146 |
sneaky | +0.142 |
fast | +0.134 |
dodged | +0.132 |
villains | +0.130 |
Neg logit contributions | |
aman | -0.115 |
berry | -0.110 |
en | -0.108 |
umm | -0.105 |
rap | -0.103 |
Token | Tom |
Token ID | 4186 |
Feature activation | -5.72e-02 |
Pos logit contributions | |
protective | +0.084 |
sneaky | +0.076 |
cunning | +0.076 |
cious | +0.075 |
ages | +0.074 |
Neg logit contributions | |
aman | -0.061 |
berry | -0.056 |
en | -0.052 |
el | -0.051 |
airport | -0.051 |
Token | and |
Token ID | 290 |
Feature activation | -4.33e-02 |
Pos logit contributions | |
protective | +0.113 |
sneaky | +0.098 |
cunning | +0.096 |
villains | +0.095 |
aro | +0.095 |
Neg logit contributions | |
berry | -0.088 |
en | -0.086 |
aman | -0.084 |
once | -0.081 |
again | -0.078 |
Token | . |
Token ID | 13 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
protective | +0.029 |
sneaky | +0.028 |
fast | +0.026 |
dodged | +0.026 |
brave | +0.025 |
Neg logit contributions | |
aman | -0.036 |
umm | -0.034 |
berry | -0.034 |
en | -0.034 |
rap | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | +6.87e-03 |
Pos logit contributions | |
aman | +0.030 |
umm | +0.028 |
en | +0.028 |
berry | +0.028 |
rap | +0.028 |
Neg logit contributions | |
protective | -0.023 |
sneaky | -0.022 |
fast | -0.021 |
dodged | -0.020 |
brave | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -4.86e-03 |
Pos logit contributions | |
aman | +0.013 |
umm | +0.013 |
berry | +0.013 |
en | +0.013 |
rap | +0.012 |
Neg logit contributions | |
protective | -0.011 |
sneaky | -0.011 |
fast | -0.011 |
dodged | -0.010 |
ggy | -0.010 |
Token | " |
Token ID | 1 |
Feature activation | -3.64e-02 |
Pos logit contributions | |
protective | +0.008 |
sneaky | +0.007 |
fast | +0.007 |
dodged | +0.007 |
villains | +0.007 |
Neg logit contributions | |
aman | -0.010 |
berry | -0.009 |
en | -0.009 |
umm | -0.009 |
rap | -0.009 |
Token | By |
Token ID | 3886 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
protective | +0.070 |
sneaky | +0.068 |
fast | +0.065 |
dodged | +0.063 |
villains | +0.062 |
Neg logit contributions | |
aman | -0.063 |
berry | -0.059 |
umm | -0.058 |
en | -0.057 |
rap | -0.055 |
Token | e |
Token ID | 68 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
aman | +0.034 |
berry | +0.032 |
en | +0.032 |
umm | +0.032 |
rap | +0.031 |
Neg logit contributions | |
protective | -0.031 |
sneaky | -0.030 |
fast | -0.028 |
dodged | -0.027 |
villains | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | -3.40e-02 |
Pos logit contributions | |
protective | +0.286 |
sneaky | +0.279 |
fast | +0.264 |
dodged | +0.259 |
brave | +0.251 |
Neg logit contributions | |
aman | -0.326 |
berry | -0.310 |
umm | -0.307 |
en | -0.303 |
rap | -0.293 |
Token | bird |
Token ID | 6512 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
protective | +0.056 |
sneaky | +0.055 |
fast | +0.051 |
dodged | +0.051 |
villains | +0.050 |
Neg logit contributions | |
aman | -0.067 |
berry | -0.063 |
umm | -0.062 |
en | -0.061 |
rap | -0.059 |
Token | , |
Token ID | 11 |
Feature activation | -8.40e-03 |
Pos logit contributions | |
protective | +0.068 |
sneaky | +0.066 |
fast | +0.063 |
dodged | +0.062 |
villains | +0.062 |
Neg logit contributions | |
aman | -0.052 |
berry | -0.050 |
en | -0.049 |
umm | -0.045 |
rap | -0.045 |
Token | we |
Token ID | 356 |
Feature activation | +6.54e-03 |
Pos logit contributions | |
protective | +0.012 |
sneaky | +0.012 |
fast | +0.012 |
brave | +0.011 |
dodged | +0.011 |
Neg logit contributions | |
aman | -0.018 |
berry | -0.017 |
umm | -0.017 |
en | -0.017 |
rap | -0.017 |
Token | hope |
Token ID | 2911 |
Feature activation | -2.33e-03 |
Pos logit contributions | |
aman | +0.012 |
umm | +0.011 |
berry | +0.011 |
en | +0.011 |
rap | +0.011 |
Neg logit contributions | |
protective | -0.012 |
sneaky | -0.011 |
fast | -0.011 |
dodged | -0.011 |
brave | -0.011 |
Token | away |
Token ID | 1497 |
Feature activation | -8.81e-02 |
Pos logit contributions | |
protective | +0.050 |
sneaky | +0.049 |
dodged | +0.046 |
ars | +0.046 |
villains | +0.046 |
Neg logit contributions | |
aman | -0.040 |
en | -0.039 |
berry | -0.037 |
rap | -0.035 |
umm | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
protective | +0.177 |
ars | +0.177 |
sneaky | +0.174 |
dodged | +0.172 |
villains | +0.170 |
Neg logit contributions | |
aman | -0.125 |
en | -0.124 |
berry | -0.118 |
rap | -0.108 |
destination | -0.107 |
Token |
|
Token ID | 198 |
Feature activation | -1.87e-03 |
Pos logit contributions | |
protective | +0.020 |
sneaky | +0.020 |
fast | +0.019 |
dodged | +0.019 |
villains | +0.018 |
Neg logit contributions | |
aman | -0.019 |
berry | -0.017 |
en | -0.017 |
umm | -0.017 |
rap | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -9.14e-03 |
Pos logit contributions | |
protective | +0.004 |
sneaky | +0.004 |
dodged | +0.003 |
fast | +0.003 |
villains | +0.003 |
Neg logit contributions | |
aman | -0.003 |
berry | -0.003 |
en | -0.003 |
rap | -0.003 |
umm | -0.003 |
Token | Jul |
Token ID | 16980 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
protective | +0.016 |
sneaky | +0.016 |
fast | +0.015 |
dodged | +0.015 |
villains | +0.015 |
Neg logit contributions | |
aman | -0.017 |
en | -0.015 |
berry | -0.015 |
rap | -0.015 |
umm | -0.015 |
Token | iet |
Token ID | 1155 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
protective | +0.068 |
sneaky | +0.065 |
fast | +0.063 |
brave | +0.061 |
dodged | +0.061 |
Neg logit contributions | |
aman | -0.059 |
berry | -0.057 |
umm | -0.057 |
rap | -0.052 |
en | -0.052 |
Token | was |
Token ID | 373 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
protective | +0.292 |
sneaky | +0.282 |
fast | +0.262 |
villains | +0.260 |
dodged | +0.255 |
Neg logit contributions | |
aman | -0.314 |
en | -0.309 |
berry | -0.302 |
umm | -0.281 |
el | -0.279 |
Token | safe |
Token ID | 3338 |
Feature activation | -7.58e-02 |
Pos logit contributions | |
protective | +0.012 |
sneaky | +0.012 |
ars | +0.011 |
villains | +0.011 |
dodged | +0.011 |
Neg logit contributions | |
aman | -0.026 |
en | -0.025 |
berry | -0.025 |
rap | -0.024 |
umm | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -8.42e-03 |
Pos logit contributions | |
protective | +0.120 |
sneaky | +0.118 |
fast | +0.117 |
brave | +0.110 |
dodged | +0.105 |
Neg logit contributions | |
aman | -0.153 |
berry | -0.146 |
umm | -0.146 |
rap | -0.143 |
en | -0.140 |
Token | She |
Token ID | 1375 |
Feature activation | -6.87e-04 |
Pos logit contributions | |
protective | +0.014 |
sneaky | +0.014 |
fast | +0.013 |
brave | +0.012 |
dodged | +0.012 |
Neg logit contributions | |
aman | -0.016 |
berry | -0.016 |
umm | -0.015 |
en | -0.015 |
rap | -0.015 |
Token | smiled |
Token ID | 13541 |
Feature activation | +4.04e-03 |
Pos logit contributions | |
protective | +0.001 |
sneaky | +0.001 |
villains | +0.001 |
ars | +0.001 |
fast | +0.001 |
Neg logit contributions | |
aman | -0.001 |
en | -0.001 |
berry | -0.001 |
el | -0.001 |
rap | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -2.94e-03 |
Pos logit contributions | |
protective | +0.123 |
ars | +0.121 |
sneaky | +0.120 |
villains | +0.119 |
dodged | +0.118 |
Neg logit contributions | |
en | -0.082 |
berry | -0.078 |
aman | -0.076 |
rap | -0.072 |
umm | -0.071 |
Token |
|
Token ID | 198 |
Feature activation | +5.48e-04 |
Pos logit contributions | |
protective | +0.006 |
sneaky | +0.006 |
cunning | +0.006 |
villains | +0.006 |
brave | +0.005 |
Neg logit contributions | |
aman | -0.004 |
berry | -0.004 |
umm | -0.004 |
rap | -0.004 |
en | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -7.10e-03 |
Pos logit contributions | |
aman | +0.001 |
berry | +0.001 |
rap | +0.001 |
en | +0.001 |
umm | +0.001 |
Neg logit contributions | |
protective | -0.001 |
sneaky | -0.001 |
villains | -0.001 |
dodged | -0.001 |
cunning | -0.001 |
Token | " |
Token ID | 1 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
protective | +0.014 |
sneaky | +0.014 |
cunning | +0.013 |
villains | +0.013 |
hind | +0.013 |
Neg logit contributions | |
aman | -0.011 |
rap | -0.010 |
berry | -0.009 |
date | -0.008 |
?' | -0.008 |
Token | By |
Token ID | 3886 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
protective | +0.052 |
sneaky | +0.050 |
fast | +0.047 |
dodged | +0.046 |
villains | +0.045 |
Neg logit contributions | |
aman | -0.055 |
berry | -0.053 |
umm | -0.052 |
en | -0.051 |
rap | -0.049 |
Token | e |
Token ID | 68 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
aman | +0.035 |
umm | +0.033 |
berry | +0.032 |
en | +0.031 |
rap | +0.030 |
Neg logit contributions | |
protective | -0.024 |
sneaky | -0.024 |
fast | -0.023 |
brave | -0.022 |
dodged | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
protective | +0.280 |
sneaky | +0.274 |
fast | +0.260 |
dodged | +0.255 |
ggy | +0.253 |
Neg logit contributions | |
aman | -0.326 |
umm | -0.309 |
berry | -0.308 |
en | -0.304 |
rap | -0.292 |
Token | frog |
Token ID | 21264 |
Feature activation | -4.85e-02 |
Pos logit contributions | |
ages | +0.052 |
protective | +0.052 |
ars | +0.048 |
cious | +0.048 |
sneaky | +0.046 |
Neg logit contributions | |
berry | -0.065 |
aman | -0.060 |
umm | -0.058 |
date | -0.055 |
en | -0.055 |
Token | !" |
Token ID | 2474 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
protective | +0.097 |
sneaky | +0.094 |
fast | +0.089 |
villains | +0.088 |
dodged | +0.088 |
Neg logit contributions | |
aman | -0.076 |
berry | -0.075 |
en | -0.071 |
umm | -0.069 |
rap | -0.068 |
Token | Tom |
Token ID | 4186 |
Feature activation | -5.54e-02 |
Pos logit contributions | |
protective | +0.091 |
cunning | +0.083 |
sneaky | +0.082 |
cious | +0.081 |
ages | +0.080 |
Neg logit contributions | |
aman | -0.046 |
berry | -0.039 |
el | -0.038 |
en | -0.037 |
airport | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
protective | +0.129 |
sneaky | +0.118 |
villains | +0.116 |
cunning | +0.115 |
aro | +0.114 |
Neg logit contributions | |
en | -0.065 |
berry | -0.063 |
once | -0.057 |
aman | -0.057 |
again | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
en | +0.043 |
berry | +0.041 |
again | +0.040 |
once | +0.040 |
aman | +0.039 |
Neg logit contributions | |
protective | -0.083 |
sneaky | -0.080 |
Fin | -0.077 |
dodged | -0.077 |
villains | -0.076 |
Token |
|
Token ID | 198 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
protective | +0.052 |
sneaky | +0.049 |
dodged | +0.046 |
ars | +0.045 |
ggy | +0.045 |
Neg logit contributions | |
berry | -0.030 |
aman | -0.029 |
date | -0.027 |
en | -0.026 |
umm | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
protective | +0.029 |
sneaky | +0.028 |
cunning | +0.026 |
dodged | +0.026 |
villains | +0.026 |
Neg logit contributions | |
aman | -0.014 |
berry | -0.013 |
rap | -0.013 |
en | -0.013 |
?' | -0.012 |
Token | " |
Token ID | 1 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
protective | +0.031 |
sneaky | +0.030 |
aro | +0.029 |
hind | +0.029 |
cunning | +0.029 |
Neg logit contributions | |
aman | -0.018 |
ara | -0.017 |
rap | -0.017 |
berry | -0.017 |
en | -0.017 |
Token | By |
Token ID | 3886 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
protective | +0.061 |
sneaky | +0.057 |
ages | +0.055 |
hind | +0.054 |
ars | +0.053 |
Neg logit contributions | |
aman | -0.038 |
berry | -0.038 |
umm | -0.035 |
date | -0.034 |
en | -0.032 |
Token | e |
Token ID | 68 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
aman | +0.039 |
umm | +0.037 |
berry | +0.037 |
en | +0.035 |
rap | +0.034 |
Neg logit contributions | |
protective | -0.030 |
sneaky | -0.029 |
fast | -0.028 |
dodged | -0.027 |
brave | -0.027 |
Token | - |
Token ID | 12 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
protective | +0.301 |
sneaky | +0.293 |
fast | +0.277 |
dodged | +0.271 |
villains | +0.266 |
Neg logit contributions | |
aman | -0.309 |
berry | -0.295 |
en | -0.287 |
umm | -0.286 |
rap | -0.277 |
Token | bye |
Token ID | 16390 |
Feature activation | -2.66e-02 |
Pos logit contributions | |
protective | +0.267 |
sneaky | +0.255 |
hind | +0.243 |
slim | +0.238 |
cunning | +0.238 |
Neg logit contributions | |
aman | -0.160 |
berry | -0.154 |
en | -0.147 |
rap | -0.143 |
umm | -0.139 |
Token | , |
Token ID | 11 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
protective | +0.048 |
sneaky | +0.046 |
fast | +0.043 |
dodged | +0.043 |
villains | +0.042 |
Neg logit contributions | |
aman | -0.048 |
berry | -0.047 |
en | -0.045 |
umm | -0.045 |
rap | -0.043 |
Token | hoop |
Token ID | 44792 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
ages | +0.026 |
cious | +0.023 |
ars | +0.023 |
protective | +0.023 |
aro | +0.021 |
Neg logit contributions | |
aman | -0.028 |
berry | -0.025 |
date | -0.023 |
airport | -0.023 |
sunrise | -0.022 |
Token | !" |
Token ID | 2474 |
Feature activation | -4.04e-02 |
Pos logit contributions | |
protective | +0.042 |
sneaky | +0.041 |
fast | +0.038 |
dodged | +0.038 |
cunning | +0.037 |
Neg logit contributions | |
aman | -0.034 |
berry | -0.033 |
en | -0.032 |
umm | -0.031 |
rap | -0.030 |
Token | !" |
Token ID | 2474 |
Feature activation | -4.27e-02 |
Pos logit contributions | |
protective | +0.274 |
sneaky | +0.267 |
fast | +0.254 |
dodged | +0.248 |
brave | +0.242 |
Neg logit contributions | |
aman | -0.298 |
berry | -0.282 |
umm | -0.280 |
en | -0.277 |
rap | -0.267 |
Token |
|
Token ID | 198 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
protective | +0.070 |
sneaky | +0.068 |
fast | +0.067 |
dodged | +0.063 |
brave | +0.063 |
Neg logit contributions | |
aman | -0.086 |
umm | -0.082 |
berry | -0.081 |
en | -0.081 |
rap | -0.077 |
Token |
|
Token ID | 198 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
protective | +0.032 |
sneaky | +0.031 |
fast | +0.029 |
dodged | +0.029 |
villains | +0.028 |
Neg logit contributions | |
aman | -0.031 |
berry | -0.029 |
umm | -0.028 |
en | -0.028 |
rap | -0.027 |
Token | " |
Token ID | 1 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
protective | +0.037 |
sneaky | +0.036 |
dodged | +0.034 |
fast | +0.034 |
villains | +0.033 |
Neg logit contributions | |
aman | -0.030 |
berry | -0.029 |
umm | -0.027 |
en | -0.027 |
rap | -0.027 |
Token | By |
Token ID | 3886 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
protective | +0.054 |
sneaky | +0.052 |
fast | +0.048 |
dodged | +0.047 |
ggy | +0.047 |
Neg logit contributions | |
berry | -0.050 |
aman | -0.049 |
umm | -0.048 |
en | -0.045 |
date | -0.045 |
Token | e |
Token ID | 68 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
aman | +0.038 |
umm | +0.036 |
berry | +0.035 |
en | +0.034 |
rap | +0.033 |
Neg logit contributions | |
protective | -0.028 |
sneaky | -0.027 |
fast | -0.027 |
brave | -0.025 |
dodged | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
protective | +0.291 |
sneaky | +0.284 |
fast | +0.270 |
dodged | +0.264 |
brave | +0.257 |
Neg logit contributions | |
aman | -0.316 |
berry | -0.300 |
umm | -0.297 |
en | -0.294 |
rap | -0.283 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
protective | +0.042 |
sneaky | +0.041 |
fast | +0.038 |
dodged | +0.037 |
villains | +0.036 |
Neg logit contributions | |
aman | -0.064 |
berry | -0.061 |
umm | -0.060 |
en | -0.060 |
rap | -0.058 |
Token | !" |
Token ID | 2474 |
Feature activation | -4.00e-02 |
Pos logit contributions | |
protective | +0.026 |
fast | +0.025 |
sneaky | +0.025 |
brave | +0.024 |
ggy | +0.023 |
Neg logit contributions | |
aman | -0.033 |
umm | -0.031 |
berry | -0.031 |
en | -0.029 |
rap | -0.029 |
Token | Mom |
Token ID | 11254 |
Feature activation | +5.76e-02 |
Pos logit contributions | |
protective | +0.077 |
sneaky | +0.076 |
fast | +0.074 |
dodged | +0.071 |
brave | +0.070 |
Neg logit contributions | |
aman | -0.069 |
umm | -0.065 |
berry | -0.065 |
en | -0.064 |
rap | -0.062 |
Token | my |
Token ID | 1820 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
umm | +0.075 |
aman | +0.074 |
rap | +0.063 |
destination | +0.062 |
date | +0.061 |
Neg logit contributions | |
fast | -0.133 |
protective | -0.131 |
sneaky | -0.129 |
brave | -0.128 |
dodged | -0.128 |
Token | . |
Token ID | 13 |
Feature activation | +3.58e-03 |
Pos logit contributions | |
aman | +0.050 |
berry | +0.048 |
umm | +0.047 |
en | +0.046 |
rap | +0.045 |
Neg logit contributions | |
protective | -0.040 |
sneaky | -0.039 |
fast | -0.038 |
dodged | -0.036 |
brave | -0.036 |
Token | |
Token ID | 220 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
aman | +0.008 |
berry | +0.008 |
en | +0.008 |
umm | +0.007 |
rap | +0.007 |
Neg logit contributions | |
protective | -0.005 |
sneaky | -0.005 |
fast | -0.005 |
dodged | -0.004 |
brave | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +8.54e-03 |
Pos logit contributions | |
aman | +0.041 |
umm | +0.040 |
en | +0.039 |
berry | +0.039 |
rap | +0.036 |
Neg logit contributions | |
fast | -0.024 |
sneaky | -0.023 |
protective | -0.023 |
dodged | -0.022 |
brave | -0.021 |
Token | " |
Token ID | 1 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
aman | +0.015 |
berry | +0.014 |
en | +0.013 |
umm | +0.013 |
rap | +0.013 |
Neg logit contributions | |
protective | -0.016 |
sneaky | -0.016 |
dodged | -0.015 |
fast | -0.015 |
villains | -0.014 |
Token | By |
Token ID | 3886 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
protective | +0.043 |
sneaky | +0.043 |
fast | +0.041 |
dodged | +0.040 |
brave | +0.039 |
Neg logit contributions | |
aman | -0.050 |
en | -0.048 |
berry | -0.048 |
umm | -0.048 |
rap | -0.048 |
Token | e |
Token ID | 68 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
aman | +0.041 |
umm | +0.038 |
berry | +0.038 |
en | +0.036 |
rap | +0.034 |
Neg logit contributions | |
protective | -0.028 |
sneaky | -0.027 |
fast | -0.027 |
brave | -0.027 |
ggy | -0.025 |
Token | little |
Token ID | 1310 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
protective | +0.273 |
sneaky | +0.266 |
fast | +0.253 |
ggy | +0.249 |
dodged | +0.248 |
Neg logit contributions | |
aman | -0.330 |
umm | -0.315 |
berry | -0.309 |
en | -0.306 |
rap | -0.296 |
Token | crab |
Token ID | 32202 |
Feature activation | -4.00e-02 |
Pos logit contributions | |
sneaky | +0.027 |
protective | +0.027 |
fast | +0.025 |
brave | +0.024 |
dodged | +0.023 |
Neg logit contributions | |
aman | -0.060 |
rap | -0.056 |
berry | -0.055 |
umm | -0.055 |
en | -0.055 |
Token | !" |
Token ID | 2474 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
protective | +0.077 |
sneaky | +0.074 |
fast | +0.070 |
dodged | +0.070 |
villains | +0.068 |
Neg logit contributions | |
aman | -0.067 |
berry | -0.065 |
en | -0.063 |
umm | -0.061 |
rap | -0.060 |
Token | he |
Token ID | 339 |
Feature activation | -8.02e-03 |
Pos logit contributions | |
protective | +0.056 |
sneaky | +0.053 |
villains | +0.051 |
hog | +0.050 |
dodged | +0.049 |
Neg logit contributions | |
aman | -0.062 |
berry | -0.059 |
en | -0.056 |
?' | -0.055 |
el | -0.054 |
Token | said |
Token ID | 531 |
Feature activation | +3.37e-02 |
Pos logit contributions | |
protective | +0.014 |
sneaky | +0.013 |
villains | +0.012 |
ars | +0.012 |
cowardly | +0.012 |
Neg logit contributions | |
aman | -0.013 |
berry | -0.013 |
en | -0.013 |
once | -0.012 |
el | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +9.57e-03 |
Pos logit contributions | |
protective | +0.091 |
sneaky | +0.088 |
fast | +0.085 |
dodged | +0.082 |
brave | +0.080 |
Neg logit contributions | |
aman | -0.097 |
berry | -0.091 |
umm | -0.091 |
en | -0.088 |
rap | -0.086 |
Token |
|
Token ID | 198 |
Feature activation | -3.25e-03 |
Pos logit contributions | |
aman | +0.020 |
umm | +0.019 |
en | +0.019 |
rap | +0.018 |
berry | +0.018 |
Neg logit contributions | |
protective | -0.014 |
sneaky | -0.014 |
fast | -0.014 |
dodged | -0.013 |
hog | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
protective | +0.005 |
sneaky | +0.005 |
fast | +0.005 |
dodged | +0.005 |
ggy | +0.005 |
Neg logit contributions | |
aman | -0.006 |
umm | -0.006 |
berry | -0.006 |
en | -0.006 |
rap | -0.006 |
Token | " |
Token ID | 1 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
protective | +0.017 |
sneaky | +0.017 |
fast | +0.015 |
dodged | +0.015 |
villains | +0.015 |
Neg logit contributions | |
aman | -0.022 |
berry | -0.021 |
en | -0.020 |
umm | -0.020 |
rap | -0.020 |
Token | By |
Token ID | 3886 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
protective | +0.074 |
sneaky | +0.073 |
fast | +0.069 |
dodged | +0.067 |
villains | +0.066 |
Neg logit contributions | |
aman | -0.066 |
berry | -0.062 |
umm | -0.060 |
en | -0.059 |
rap | -0.056 |
Token | e |
Token ID | 68 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
aman | +0.025 |
berry | +0.024 |
umm | +0.023 |
en | +0.023 |
rap | +0.022 |
Neg logit contributions | |
protective | -0.021 |
sneaky | -0.021 |
fast | -0.020 |
dodged | -0.019 |
brave | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | -3.26e-02 |
Pos logit contributions | |
protective | +0.290 |
sneaky | +0.283 |
fast | +0.267 |
dodged | +0.262 |
villains | +0.256 |
Neg logit contributions | |
aman | -0.314 |
berry | -0.300 |
en | -0.293 |
umm | -0.292 |
rap | -0.282 |
Token | bunny |
Token ID | 44915 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
protective | +0.052 |
sneaky | +0.049 |
dodged | +0.046 |
fast | +0.045 |
ars | +0.045 |
Neg logit contributions | |
aman | -0.066 |
berry | -0.063 |
umm | -0.061 |
en | -0.061 |
rap | -0.059 |
Token | !" |
Token ID | 2474 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
protective | +0.072 |
sneaky | +0.070 |
dodged | +0.065 |
fast | +0.065 |
villains | +0.064 |
Neg logit contributions | |
aman | -0.059 |
berry | -0.058 |
en | -0.057 |
umm | -0.052 |
rap | -0.052 |
Token | Ben |
Token ID | 3932 |
Feature activation | -4.68e-02 |
Pos logit contributions | |
protective | +0.058 |
sneaky | +0.055 |
dodged | +0.050 |
cunning | +0.050 |
fast | +0.049 |
Neg logit contributions | |
aman | -0.064 |
berry | -0.060 |
en | -0.058 |
?' | -0.055 |
el | -0.054 |
Token | says |
Token ID | 1139 |
Feature activation | +6.69e-03 |
Pos logit contributions | |
protective | +0.079 |
sneaky | +0.076 |
fast | +0.070 |
villains | +0.069 |
dodged | +0.069 |
Neg logit contributions | |
aman | -0.087 |
berry | -0.086 |
en | -0.086 |
rap | -0.076 |
?' | -0.075 |
Token | . |
Token ID | 13 |
Feature activation | +9.57e-03 |
Pos logit contributions | |
protective | +0.134 |
sneaky | +0.131 |
fast | +0.128 |
brave | +0.122 |
dodged | +0.118 |
Neg logit contributions | |
aman | -0.169 |
berry | -0.160 |
umm | -0.160 |
en | -0.158 |
rap | -0.153 |
Token |
|
Token ID | 198 |
Feature activation | -3.25e-03 |
Pos logit contributions | |
aman | +0.020 |
umm | +0.019 |
en | +0.019 |
rap | +0.018 |
berry | +0.018 |
Neg logit contributions | |
protective | -0.014 |
sneaky | -0.014 |
fast | -0.014 |
dodged | -0.013 |
hog | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
protective | +0.005 |
sneaky | +0.005 |
fast | +0.005 |
dodged | +0.005 |
ggy | +0.005 |
Neg logit contributions | |
aman | -0.006 |
umm | -0.006 |
berry | -0.006 |
en | -0.006 |
rap | -0.006 |
Token | " |
Token ID | 1 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
protective | +0.017 |
sneaky | +0.017 |
fast | +0.015 |
dodged | +0.015 |
villains | +0.015 |
Neg logit contributions | |
aman | -0.022 |
berry | -0.021 |
en | -0.020 |
umm | -0.020 |
rap | -0.020 |
Token | By |
Token ID | 3886 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
protective | +0.074 |
sneaky | +0.073 |
fast | +0.069 |
dodged | +0.067 |
villains | +0.066 |
Neg logit contributions | |
aman | -0.066 |
berry | -0.062 |
umm | -0.060 |
en | -0.059 |
rap | -0.056 |
Token | e |
Token ID | 68 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
aman | +0.025 |
berry | +0.024 |
umm | +0.023 |
en | +0.023 |
rap | +0.022 |
Neg logit contributions | |
protective | -0.021 |
sneaky | -0.021 |
fast | -0.020 |
dodged | -0.019 |
brave | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | -3.26e-02 |
Pos logit contributions | |
protective | +0.290 |
sneaky | +0.283 |
fast | +0.267 |
dodged | +0.262 |
villains | +0.256 |
Neg logit contributions | |
aman | -0.314 |
berry | -0.300 |
en | -0.293 |
umm | -0.292 |
rap | -0.282 |
Token | oct |
Token ID | 19318 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
protective | +0.052 |
sneaky | +0.049 |
dodged | +0.046 |
fast | +0.045 |
ars | +0.045 |
Neg logit contributions | |
aman | -0.066 |
berry | -0.063 |
umm | -0.061 |
en | -0.061 |
rap | -0.059 |
Token | opus |
Token ID | 25790 |
Feature activation | -5.18e-02 |
Pos logit contributions | |
protective | +0.070 |
sneaky | +0.066 |
fast | +0.064 |
dodged | +0.060 |
cunning | +0.059 |
Neg logit contributions | |
berry | -0.075 |
aman | -0.074 |
umm | -0.071 |
rap | -0.069 |
en | -0.069 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.97e-02 |
Pos logit contributions | |
protective | +0.101 |
sneaky | +0.099 |
fast | +0.092 |
dodged | +0.092 |
villains | +0.091 |
Neg logit contributions | |
aman | -0.083 |
berry | -0.081 |
en | -0.080 |
umm | -0.074 |
rap | -0.073 |
Token | Ben |
Token ID | 3932 |
Feature activation | -4.39e-02 |
Pos logit contributions | |
protective | +0.051 |
sneaky | +0.048 |
dodged | +0.045 |
cunning | +0.044 |
fast | +0.044 |
Neg logit contributions | |
aman | -0.056 |
berry | -0.053 |
en | -0.052 |
?' | -0.049 |
rap | -0.048 |
Token | â |
Token ID | 22940 |
Feature activation | +2.84e-02 |
Pos logit contributions | |
aman | +0.015 |
berry | +0.015 |
umm | +0.014 |
en | +0.014 |
rap | +0.014 |
Neg logit contributions | |
protective | -0.009 |
sneaky | -0.009 |
fast | -0.008 |
dodged | -0.008 |
villains | -0.008 |
Token | € |
Token ID | 26391 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
aman | +0.050 |
en | +0.046 |
umm | +0.046 |
berry | +0.044 |
rap | +0.043 |
Neg logit contributions | |
sneaky | -0.054 |
protective | -0.054 |
villains | -0.050 |
fast | -0.049 |
dodged | -0.048 |
Token | � |
Token ID | 129 |
Feature activation | -3.14e-02 |
Pos logit contributions | |
protective | +0.028 |
sneaky | +0.027 |
fast | +0.025 |
dodged | +0.025 |
villains | +0.024 |
Neg logit contributions | |
aman | -0.029 |
berry | -0.028 |
umm | -0.027 |
en | -0.027 |
rap | -0.026 |
Token | � |
Token ID | 241 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
sneaky | +0.054 |
protective | +0.054 |
dodged | +0.050 |
villains | +0.050 |
fast | +0.049 |
Neg logit contributions | |
aman | -0.059 |
en | -0.055 |
rap | -0.054 |
berry | -0.050 |
umm | -0.050 |
Token | By |
Token ID | 3886 |
Feature activation | +3.03e-02 |
Pos logit contributions | |
aman | +0.021 |
berry | +0.020 |
umm | +0.019 |
en | +0.019 |
rap | +0.019 |
Neg logit contributions | |
protective | -0.017 |
sneaky | -0.017 |
fast | -0.016 |
dodged | -0.015 |
villains | -0.015 |
Token | e |
Token ID | 68 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
aman | +0.070 |
umm | +0.067 |
berry | +0.063 |
en | +0.060 |
icer | +0.058 |
Neg logit contributions | |
fast | -0.041 |
protective | -0.040 |
brave | -0.040 |
sneaky | -0.039 |
ggy | -0.038 |
Token | bye |
Token ID | 33847 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
protective | +0.285 |
sneaky | +0.278 |
fast | +0.263 |
dodged | +0.257 |
villains | +0.251 |
Neg logit contributions | |
aman | -0.318 |
berry | -0.304 |
umm | -0.296 |
en | -0.296 |
rap | -0.285 |
Token | , |
Token ID | 11 |
Feature activation | +3.88e-03 |
Pos logit contributions | |
protective | +0.045 |
sneaky | +0.044 |
fast | +0.044 |
brave | +0.041 |
ggy | +0.040 |
Neg logit contributions | |
aman | -0.060 |
umm | -0.056 |
berry | -0.056 |
rap | -0.055 |
en | -0.055 |
Token | Tommy |
Token ID | 19919 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
aman | +0.007 |
berry | +0.007 |
umm | +0.007 |
en | +0.007 |
rap | +0.007 |
Neg logit contributions | |
protective | -0.007 |
sneaky | -0.007 |
fast | -0.006 |
dodged | -0.006 |
villains | -0.006 |
Token | ! |
Token ID | 0 |
Feature activation | +6.37e-04 |
Pos logit contributions | |
aman | +0.040 |
berry | +0.037 |
umm | +0.037 |
en | +0.036 |
rap | +0.035 |
Neg logit contributions | |
protective | -0.037 |
sneaky | -0.037 |
fast | -0.035 |
dodged | -0.034 |
brave | -0.034 |
Token | See |
Token ID | 4091 |
Feature activation | +7.69e-02 |
Pos logit contributions | |
aman | +0.001 |
en | +0.001 |
rap | +0.001 |
umm | +0.001 |
berry | +0.001 |
Neg logit contributions | |
fast | -0.001 |
protective | -0.001 |
sneaky | -0.001 |
dodged | -0.001 |
Rex | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +6.99e-03 |
Pos logit contributions | |
aman | +0.033 |
berry | +0.033 |
en | +0.032 |
umm | +0.031 |
rap | +0.030 |
Neg logit contributions | |
protective | -0.039 |
sneaky | -0.038 |
dodged | -0.036 |
ars | -0.035 |
cunning | -0.035 |
Token |
|
Token ID | 198 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
aman | +0.012 |
berry | +0.012 |
umm | +0.012 |
en | +0.012 |
rap | +0.011 |
Neg logit contributions | |
protective | -0.013 |
sneaky | -0.012 |
dodged | -0.011 |
fast | -0.011 |
ggy | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | -1.49e-04 |
Pos logit contributions | |
aman | +0.016 |
berry | +0.016 |
rap | +0.015 |
en | +0.015 |
umm | +0.014 |
Neg logit contributions | |
protective | -0.021 |
sneaky | -0.020 |
dodged | -0.019 |
villains | -0.018 |
fast | -0.018 |
Token | " |
Token ID | 1 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
protective | +0.000 |
sneaky | +0.000 |
villains | +0.000 |
dodged | +0.000 |
hog | +0.000 |
Neg logit contributions | |
aman | -0.000 |
berry | -0.000 |
rap | -0.000 |
en | -0.000 |
?' | -0.000 |
Token | By |
Token ID | 3886 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
protective | +0.044 |
sneaky | +0.043 |
fast | +0.041 |
dodged | +0.041 |
brave | +0.040 |
Neg logit contributions | |
aman | -0.047 |
berry | -0.043 |
umm | -0.043 |
en | -0.043 |
rap | -0.043 |
Token | e |
Token ID | 68 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
aman | +0.040 |
umm | +0.038 |
berry | +0.038 |
en | +0.036 |
rap | +0.035 |
Neg logit contributions | |
protective | -0.031 |
sneaky | -0.030 |
fast | -0.030 |
brave | -0.028 |
dodged | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
protective | +0.291 |
sneaky | +0.284 |
fast | +0.269 |
dodged | +0.263 |
villains | +0.256 |
Neg logit contributions | |
aman | -0.311 |
berry | -0.296 |
umm | -0.291 |
en | -0.289 |
rap | -0.279 |
Token | dolphin |
Token ID | 44136 |
Feature activation | -7.24e-02 |
Pos logit contributions | |
ages | +0.059 |
protective | +0.057 |
ars | +0.056 |
dodged | +0.055 |
ggy | +0.055 |
Neg logit contributions | |
berry | -0.059 |
aman | -0.058 |
en | -0.055 |
umm | -0.051 |
airport | -0.051 |
Token | !" |
Token ID | 2474 |
Feature activation | -3.73e-02 |
Pos logit contributions | |
protective | +0.142 |
sneaky | +0.139 |
dodged | +0.131 |
villains | +0.130 |
fast | +0.129 |
Neg logit contributions | |
aman | -0.113 |
berry | -0.112 |
en | -0.109 |
umm | -0.101 |
rap | -0.100 |
Token | Ben |
Token ID | 3932 |
Feature activation | -3.71e-02 |
Pos logit contributions | |
protective | +0.075 |
sneaky | +0.070 |
ages | +0.068 |
dodged | +0.065 |
cunning | +0.065 |
Neg logit contributions | |
berry | -0.058 |
aman | -0.056 |
en | -0.052 |
?' | -0.051 |
el | -0.049 |
Token | and |
Token ID | 290 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
protective | +0.082 |
sneaky | +0.078 |
villains | +0.075 |
aro | +0.073 |
ggy | +0.073 |
Neg logit contributions | |
en | -0.049 |
berry | -0.048 |
aman | -0.045 |
once | -0.041 |
el | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
aman | +0.065 |
umm | +0.063 |
berry | +0.061 |
en | +0.060 |
rap | +0.059 |
Neg logit contributions | |
protective | -0.049 |
sneaky | -0.047 |
fast | -0.045 |
dodged | -0.044 |
brave | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
aman | +0.025 |
berry | +0.024 |
umm | +0.024 |
en | +0.024 |
rap | +0.023 |
Neg logit contributions | |
protective | -0.019 |
sneaky | -0.019 |
fast | -0.017 |
dodged | -0.017 |
villains | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | +7.54e-05 |
Pos logit contributions | |
aman | +0.026 |
umm | +0.026 |
berry | +0.025 |
en | +0.025 |
rap | +0.023 |
Neg logit contributions | |
protective | -0.019 |
fast | -0.019 |
sneaky | -0.019 |
ggy | -0.017 |
dodged | -0.017 |
Token | " |
Token ID | 1 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
aman | +0.000 |
berry | +0.000 |
en | +0.000 |
rap | +0.000 |
umm | +0.000 |
Neg logit contributions | |
protective | -0.000 |
sneaky | -0.000 |
dodged | -0.000 |
fast | -0.000 |
villains | -0.000 |
Token | By |
Token ID | 3886 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
protective | +0.066 |
sneaky | +0.065 |
fast | +0.062 |
dodged | +0.060 |
villains | +0.059 |
Neg logit contributions | |
aman | -0.048 |
berry | -0.046 |
umm | -0.044 |
en | -0.044 |
rap | -0.041 |
Token | e |
Token ID | 68 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
aman | +0.034 |
berry | +0.033 |
en | +0.032 |
umm | +0.032 |
rap | +0.031 |
Neg logit contributions | |
protective | -0.032 |
sneaky | -0.031 |
fast | -0.029 |
dodged | -0.029 |
villains | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
protective | +0.309 |
sneaky | +0.300 |
fast | +0.283 |
dodged | +0.277 |
villains | +0.276 |
Neg logit contributions | |
aman | -0.293 |
berry | -0.284 |
en | -0.275 |
umm | -0.267 |
rap | -0.264 |
Token | Rex |
Token ID | 17853 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
protective | +0.058 |
sneaky | +0.054 |
ages | +0.054 |
ars | +0.053 |
dodged | +0.050 |
Neg logit contributions | |
aman | -0.066 |
berry | -0.064 |
en | -0.061 |
umm | -0.061 |
date | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
protective | +0.211 |
sneaky | +0.204 |
fast | +0.191 |
villains | +0.190 |
dodged | +0.189 |
Neg logit contributions | |
aman | -0.171 |
berry | -0.167 |
en | -0.165 |
rap | -0.155 |
umm | -0.154 |
Token | We |
Token ID | 775 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
protective | +0.025 |
sneaky | +0.025 |
fast | +0.024 |
dodged | +0.023 |
villains | +0.023 |
Neg logit contributions | |
aman | -0.024 |
berry | -0.022 |
en | -0.022 |
umm | -0.022 |
rap | -0.021 |
Token | will |
Token ID | 481 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
protective | +0.020 |
sneaky | +0.020 |
fast | +0.019 |
dodged | +0.018 |
brave | +0.018 |
Neg logit contributions | |
aman | -0.019 |
berry | -0.018 |
umm | -0.018 |
en | -0.018 |
rap | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +3.71e-04 |
Pos logit contributions | |
aman | +0.004 |
umm | +0.004 |
berry | +0.004 |
en | +0.004 |
rap | +0.004 |
Neg logit contributions | |
protective | -0.003 |
sneaky | -0.003 |
fast | -0.003 |
brave | -0.003 |
dodged | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +4.71e-03 |
Pos logit contributions | |
aman | +0.001 |
berry | +0.001 |
umm | +0.001 |
en | +0.001 |
rap | +0.001 |
Neg logit contributions | |
protective | -0.001 |
sneaky | -0.001 |
dodged | -0.001 |
fast | -0.001 |
ggy | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -4.24e-03 |
Pos logit contributions | |
aman | +0.008 |
berry | +0.008 |
en | +0.008 |
umm | +0.007 |
rap | +0.007 |
Neg logit contributions | |
protective | -0.009 |
sneaky | -0.009 |
fast | -0.008 |
dodged | -0.008 |
villains | -0.008 |
Token | " |
Token ID | 1 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
protective | +0.007 |
sneaky | +0.007 |
cunning | +0.007 |
hog | +0.006 |
hind | +0.006 |
Neg logit contributions | |
aman | -0.008 |
rap | -0.007 |
berry | -0.007 |
en | -0.007 |
date | -0.007 |
Token | By |
Token ID | 3886 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
protective | +0.047 |
sneaky | +0.047 |
fast | +0.045 |
dodged | +0.045 |
villains | +0.043 |
Neg logit contributions | |
aman | -0.044 |
berry | -0.040 |
umm | -0.040 |
en | -0.040 |
rap | -0.039 |
Token | e |
Token ID | 68 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
aman | +0.045 |
umm | +0.042 |
berry | +0.041 |
en | +0.040 |
rap | +0.038 |
Neg logit contributions | |
protective | -0.031 |
sneaky | -0.030 |
fast | -0.029 |
brave | -0.028 |
dodged | -0.028 |
Token | - |
Token ID | 12 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
protective | +0.277 |
sneaky | +0.270 |
fast | +0.257 |
dodged | +0.252 |
ggy | +0.249 |
Neg logit contributions | |
aman | -0.323 |
umm | -0.305 |
berry | -0.303 |
en | -0.300 |
rap | -0.288 |
Token | bye |
Token ID | 16390 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
protective | +0.243 |
sneaky | +0.234 |
hind | +0.214 |
cunning | +0.213 |
dodged | +0.212 |
Neg logit contributions | |
berry | -0.186 |
aman | -0.184 |
en | -0.176 |
rap | -0.173 |
umm | -0.170 |
Token | , |
Token ID | 11 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
protective | +0.043 |
sneaky | +0.042 |
fast | +0.040 |
dodged | +0.039 |
brave | +0.038 |
Neg logit contributions | |
aman | -0.050 |
berry | -0.047 |
umm | -0.047 |
en | -0.046 |
rap | -0.045 |
Token | monkey |
Token ID | 21657 |
Feature activation | -6.15e-02 |
Pos logit contributions | |
protective | +0.021 |
ages | +0.019 |
ars | +0.018 |
cious | +0.018 |
sneaky | +0.018 |
Neg logit contributions | |
berry | -0.028 |
aman | -0.026 |
umm | -0.025 |
rap | -0.025 |
en | -0.024 |
Token | !" |
Token ID | 2474 |
Feature activation | -3.29e-02 |
Pos logit contributions | |
protective | +0.117 |
sneaky | +0.114 |
fast | +0.107 |
dodged | +0.105 |
villains | +0.104 |
Neg logit contributions | |
aman | -0.104 |
berry | -0.100 |
en | -0.097 |
umm | -0.096 |
rap | -0.094 |
Token | . |
Token ID | 13 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
aman | +0.004 |
umm | +0.004 |
berry | +0.004 |
rap | +0.004 |
en | +0.004 |
Neg logit contributions | |
protective | -0.003 |
sneaky | -0.003 |
fast | -0.003 |
dodged | -0.003 |
brave | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +5.35e-03 |
Pos logit contributions | |
aman | +0.022 |
umm | +0.021 |
berry | +0.021 |
en | +0.020 |
rap | +0.020 |
Neg logit contributions | |
protective | -0.017 |
sneaky | -0.017 |
fast | -0.016 |
dodged | -0.016 |
villains | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | -5.45e-03 |
Pos logit contributions | |
aman | +0.009 |
berry | +0.009 |
en | +0.009 |
umm | +0.009 |
rap | +0.008 |
Neg logit contributions | |
protective | -0.010 |
sneaky | -0.010 |
fast | -0.009 |
dodged | -0.009 |
villains | -0.009 |
Token | " |
Token ID | 1 |
Feature activation | -3.41e-02 |
Pos logit contributions | |
protective | +0.009 |
sneaky | +0.009 |
dodged | +0.009 |
cunning | +0.008 |
fast | +0.008 |
Neg logit contributions | |
aman | -0.010 |
berry | -0.009 |
rap | -0.009 |
?' | -0.009 |
en | -0.009 |
Token | Cl |
Token ID | 2601 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
protective | +0.064 |
sneaky | +0.063 |
fast | +0.059 |
dodged | +0.058 |
villains | +0.057 |
Neg logit contributions | |
aman | -0.060 |
berry | -0.057 |
umm | -0.055 |
en | -0.055 |
rap | -0.052 |
Token | ang |
Token ID | 648 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
protective | +0.062 |
ars | +0.062 |
sneaky | +0.061 |
fast | +0.059 |
ggy | +0.059 |
Neg logit contributions | |
aman | -0.037 |
umm | -0.033 |
en | -0.033 |
rap | -0.031 |
berry | -0.031 |
Token | ! |
Token ID | 0 |
Feature activation | -4.32e-02 |
Pos logit contributions | |
protective | +0.175 |
sneaky | +0.169 |
villains | +0.150 |
fast | +0.147 |
dodged | +0.136 |
Neg logit contributions | |
berry | -0.414 |
aman | -0.411 |
en | -0.406 |
el | -0.397 |
?' | -0.389 |
Token | Cl |
Token ID | 1012 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
protective | +0.070 |
sneaky | +0.068 |
fast | +0.068 |
brave | +0.064 |
dodged | +0.064 |
Neg logit contributions | |
aman | -0.083 |
umm | -0.080 |
en | -0.079 |
rap | -0.076 |
berry | -0.076 |
Token | ang |
Token ID | 648 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
aman | +0.017 |
umm | +0.016 |
berry | +0.016 |
en | +0.015 |
rap | +0.015 |
Neg logit contributions | |
protective | -0.027 |
sneaky | -0.027 |
fast | -0.026 |
dodged | -0.026 |
ars | -0.026 |
Token | ! |
Token ID | 0 |
Feature activation | -2.81e-02 |
Pos logit contributions | |
protective | +0.151 |
sneaky | +0.149 |
villains | +0.138 |
fast | +0.121 |
hind | +0.118 |
Neg logit contributions | |
berry | -0.379 |
el | -0.372 |
en | -0.370 |
aman | -0.368 |
?' | -0.362 |
Token | Cl |
Token ID | 1012 |
Feature activation | +2.42e-02 |
Pos logit contributions | |
protective | +0.051 |
sneaky | +0.049 |
fast | +0.048 |
dodged | +0.046 |
brave | +0.045 |
Neg logit contributions | |
aman | -0.050 |
umm | -0.047 |
berry | -0.047 |
en | -0.047 |
rap | -0.045 |
Token | Amy |
Token ID | 14235 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
protective | +0.019 |
sneaky | +0.018 |
dodged | +0.017 |
ars | +0.017 |
fast | +0.017 |
Neg logit contributions | |
aman | -0.018 |
berry | -0.018 |
en | -0.017 |
rap | -0.016 |
umm | -0.016 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
aman | +0.044 |
umm | +0.042 |
berry | +0.041 |
en | +0.041 |
rap | +0.040 |
Neg logit contributions | |
protective | -0.032 |
sneaky | -0.031 |
fast | -0.029 |
dodged | -0.029 |
brave | -0.028 |
Token | a |
Token ID | 257 |
Feature activation | -6.85e-03 |
Pos logit contributions | |
aman | +0.032 |
berry | +0.030 |
en | +0.029 |
umm | +0.029 |
rap | +0.028 |
Neg logit contributions | |
protective | -0.026 |
sneaky | -0.024 |
dodged | -0.023 |
ars | -0.023 |
villains | -0.022 |
Token | big |
Token ID | 1263 |
Feature activation | +4.39e-02 |
Pos logit contributions | |
protective | +0.008 |
sneaky | +0.008 |
fast | +0.008 |
dodged | +0.007 |
villains | +0.007 |
Neg logit contributions | |
aman | -0.016 |
berry | -0.016 |
umm | -0.015 |
en | -0.015 |
rap | -0.015 |
Token | t |
Token ID | 256 |
Feature activation | +8.09e-02 |
Pos logit contributions | |
aman | +0.087 |
umm | +0.082 |
en | +0.081 |
rap | +0.080 |
berry | +0.079 |
Neg logit contributions | |
protective | -0.073 |
sneaky | -0.071 |
fast | -0.070 |
brave | -0.065 |
dodged | -0.062 |
Token | eddy |
Token ID | 21874 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
aman | +0.066 |
en | +0.056 |
umm | +0.053 |
berry | +0.052 |
rap | +0.048 |
Neg logit contributions | |
protective | -0.219 |
fast | -0.215 |
sneaky | -0.214 |
hog | -0.208 |
ggy | -0.206 |
Token | bear |
Token ID | 6842 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
aman | +0.326 |
umm | +0.317 |
rap | +0.296 |
en | +0.294 |
berry | +0.287 |
Neg logit contributions | |
protective | -0.173 |
sneaky | -0.170 |
fast | -0.167 |
brave | -0.159 |
ggy | -0.150 |
Token | . |
Token ID | 13 |
Feature activation | +1.45e-02 |
Pos logit contributions | |
aman | +0.036 |
umm | +0.035 |
en | +0.033 |
berry | +0.033 |
rap | +0.032 |
Neg logit contributions | |
protective | -0.023 |
sneaky | -0.022 |
fast | -0.022 |
ggy | -0.021 |
brave | -0.021 |
Token | She |
Token ID | 1375 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
aman | +0.025 |
berry | +0.024 |
umm | +0.023 |
en | +0.023 |
rap | +0.022 |
Neg logit contributions | |
protective | -0.027 |
sneaky | -0.026 |
dodged | -0.025 |
fast | -0.024 |
ars | -0.024 |
Token | said |
Token ID | 531 |
Feature activation | +3.76e-02 |
Pos logit contributions | |
aman | +0.046 |
berry | +0.045 |
en | +0.044 |
rap | +0.042 |
el | +0.041 |
Neg logit contributions | |
protective | -0.039 |
sneaky | -0.037 |
villains | -0.034 |
ars | -0.034 |
fast | -0.034 |
Token | , |
Token ID | 11 |
Feature activation | -4.59e-02 |
Pos logit contributions | |
aman | +0.069 |
berry | +0.066 |
umm | +0.064 |
en | +0.064 |
rap | +0.062 |
Neg logit contributions | |
protective | -0.067 |
sneaky | -0.065 |
fast | -0.062 |
dodged | -0.060 |
villains | -0.059 |
Token | about |
Token ID | 546 |
Feature activation | +6.25e-02 |
Pos logit contributions | |
protective | +0.001 |
sneaky | +0.001 |
dodged | +0.001 |
fast | +0.001 |
ars | +0.001 |
Neg logit contributions | |
aman | -0.001 |
berry | -0.001 |
en | -0.001 |
umm | -0.001 |
rap | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +3.35e-02 |
Pos logit contributions | |
aman | +0.117 |
umm | +0.111 |
berry | +0.111 |
en | +0.107 |
rap | +0.105 |
Neg logit contributions | |
protective | -0.110 |
sneaky | -0.107 |
fast | -0.103 |
dodged | -0.097 |
brave | -0.097 |
Token | different |
Token ID | 1180 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
aman | +0.073 |
en | +0.068 |
berry | +0.067 |
rap | +0.067 |
umm | +0.065 |
Neg logit contributions | |
protective | -0.052 |
sneaky | -0.052 |
fast | -0.049 |
brave | -0.047 |
sharp | -0.044 |
Token | fruits |
Token ID | 15921 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
aman | +0.037 |
rap | +0.034 |
umm | +0.032 |
berry | +0.032 |
en | +0.032 |
Neg logit contributions | |
protective | -0.025 |
sneaky | -0.024 |
fast | -0.022 |
brave | -0.022 |
dodged | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | +5.27e-02 |
Pos logit contributions | |
aman | +0.207 |
berry | +0.199 |
umm | +0.195 |
en | +0.194 |
rap | +0.186 |
Neg logit contributions | |
protective | -0.175 |
sneaky | -0.170 |
fast | -0.159 |
dodged | -0.157 |
villains | -0.154 |
Token | vegetables |
Token ID | 13701 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
aman | +0.094 |
berry | +0.089 |
umm | +0.088 |
en | +0.086 |
rap | +0.084 |
Neg logit contributions | |
protective | -0.097 |
sneaky | -0.095 |
fast | -0.090 |
dodged | -0.088 |
brave | -0.086 |
Token | she |
Token ID | 673 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
aman | +0.286 |
berry | +0.269 |
umm | +0.269 |
en | +0.264 |
rap | +0.258 |
Neg logit contributions | |
protective | -0.225 |
sneaky | -0.219 |
fast | -0.209 |
dodged | -0.200 |
brave | -0.198 |
Token | could |
Token ID | 714 |
Feature activation | +8.01e-02 |
Pos logit contributions | |
aman | +0.251 |
berry | +0.246 |
en | +0.242 |
umm | +0.242 |
rap | +0.234 |
Neg logit contributions | |
protective | -0.159 |
sneaky | -0.151 |
fast | -0.142 |
dodged | -0.138 |
villains | -0.136 |
Token | eat |
Token ID | 4483 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
aman | +0.173 |
umm | +0.162 |
berry | +0.158 |
en | +0.157 |
icer | +0.151 |
Neg logit contributions | |
protective | -0.120 |
sneaky | -0.119 |
fast | -0.117 |
brave | -0.111 |
ggy | -0.107 |
Token | . |
Token ID | 13 |
Feature activation | +1.56e-04 |
Pos logit contributions | |
aman | +0.069 |
umm | +0.065 |
berry | +0.064 |
en | +0.063 |
rap | +0.062 |
Neg logit contributions | |
protective | -0.049 |
sneaky | -0.049 |
fast | -0.048 |
brave | -0.044 |
dodged | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
aman | +0.000 |
en | +0.000 |
umm | +0.000 |
berry | +0.000 |
rap | +0.000 |
Neg logit contributions | |
protective | -0.000 |
sneaky | -0.000 |
fast | -0.000 |
dodged | -0.000 |
hog | -0.000 |
Token | cakes |
Token ID | 30849 |
Feature activation | +7.10e-03 |
Pos logit contributions | |
aman | +0.074 |
berry | +0.070 |
umm | +0.069 |
en | +0.069 |
rap | +0.068 |
Neg logit contributions | |
protective | -0.050 |
sneaky | -0.049 |
fast | -0.046 |
dodged | -0.045 |
brave | -0.044 |
Token | and |
Token ID | 290 |
Feature activation | -7.93e-03 |
Pos logit contributions | |
aman | +0.015 |
umm | +0.014 |
en | +0.014 |
berry | +0.014 |
rap | +0.014 |
Neg logit contributions | |
protective | -0.010 |
sneaky | -0.010 |
fast | -0.010 |
dodged | -0.009 |
brave | -0.009 |
Token | cookies |
Token ID | 14746 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
protective | +0.014 |
sneaky | +0.014 |
fast | +0.013 |
dodged | +0.013 |
villains | +0.013 |
Neg logit contributions | |
aman | -0.014 |
berry | -0.013 |
umm | -0.013 |
en | -0.013 |
rap | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
aman | +0.070 |
umm | +0.066 |
berry | +0.066 |
en | +0.065 |
rap | +0.063 |
Neg logit contributions | |
protective | -0.055 |
sneaky | -0.054 |
fast | -0.052 |
dodged | -0.050 |
brave | -0.049 |
Token | ch |
Token ID | 442 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
aman | +0.030 |
berry | +0.029 |
umm | +0.028 |
en | +0.028 |
rap | +0.027 |
Neg logit contributions | |
protective | -0.028 |
sneaky | -0.027 |
fast | -0.026 |
dodged | -0.025 |
villains | -0.024 |
Token | ocol |
Token ID | 4668 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
aman | +0.032 |
berry | +0.027 |
umm | +0.027 |
en | +0.025 |
rap | +0.024 |
Neg logit contributions | |
protective | -0.068 |
sneaky | -0.067 |
fast | -0.064 |
dodged | -0.064 |
ars | -0.063 |
Token | ates |
Token ID | 689 |
Feature activation | +7.21e-03 |
Pos logit contributions | |
aman | +0.281 |
umm | +0.251 |
berry | +0.248 |
en | +0.243 |
destination | +0.240 |
Neg logit contributions | |
protective | -0.273 |
ages | -0.269 |
ars | -0.269 |
sneaky | -0.267 |
dodged | -0.261 |
Token | and |
Token ID | 290 |
Feature activation | +3.80e-02 |
Pos logit contributions | |
aman | +0.014 |
berry | +0.013 |
umm | +0.013 |
en | +0.013 |
rap | +0.013 |
Neg logit contributions | |
protective | -0.012 |
sneaky | -0.012 |
fast | -0.011 |
dodged | -0.011 |
brave | -0.010 |
Token | ice |
Token ID | 4771 |
Feature activation | +8.69e-02 |
Pos logit contributions | |
aman | +0.072 |
berry | +0.070 |
umm | +0.068 |
en | +0.068 |
rap | +0.065 |
Neg logit contributions | |
protective | -0.064 |
sneaky | -0.062 |
fast | -0.058 |
dodged | -0.057 |
villains | -0.056 |
Token | cream |
Token ID | 8566 |
Feature activation | +6.40e-02 |
Pos logit contributions | |
aman | +0.206 |
berry | +0.195 |
umm | +0.194 |
en | +0.192 |
rap | +0.188 |
Neg logit contributions | |
protective | -0.110 |
sneaky | -0.106 |
fast | -0.100 |
dodged | -0.095 |
brave | -0.092 |
Token | . |
Token ID | 13 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
aman | +0.139 |
umm | +0.129 |
en | +0.128 |
rap | +0.128 |
berry | +0.127 |
Neg logit contributions | |
protective | -0.091 |
sneaky | -0.090 |
fast | -0.088 |
dodged | -0.081 |
brave | -0.080 |
Token | has |
Token ID | 468 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
protective | +0.060 |
sneaky | +0.060 |
fast | +0.058 |
dodged | +0.056 |
brave | +0.055 |
Neg logit contributions | |
aman | -0.080 |
umm | -0.072 |
berry | -0.069 |
rap | -0.068 |
en | -0.067 |
Token | lettuce |
Token ID | 39406 |
Feature activation | +4.93e-02 |
Pos logit contributions | |
aman | +0.036 |
berry | +0.034 |
umm | +0.034 |
en | +0.033 |
rap | +0.033 |
Neg logit contributions | |
sneaky | -0.018 |
protective | -0.018 |
fast | -0.017 |
sharp | -0.016 |
brave | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | -4.02e-03 |
Pos logit contributions | |
aman | +0.108 |
umm | +0.100 |
berry | +0.095 |
en | +0.095 |
rap | +0.094 |
Neg logit contributions | |
protective | -0.074 |
sneaky | -0.073 |
fast | -0.070 |
dodged | -0.066 |
brave | -0.065 |
Token | tomatoes |
Token ID | 23972 |
Feature activation | +9.42e-02 |
Pos logit contributions | |
protective | +0.007 |
sneaky | +0.006 |
fast | +0.006 |
dodged | +0.006 |
brave | +0.006 |
Neg logit contributions | |
aman | -0.008 |
berry | -0.008 |
umm | -0.008 |
en | -0.007 |
rap | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
aman | +0.221 |
umm | +0.205 |
berry | +0.196 |
en | +0.194 |
rap | +0.193 |
Neg logit contributions | |
protective | -0.127 |
sneaky | -0.123 |
fast | -0.123 |
dodged | -0.112 |
brave | -0.110 |
Token | cuc |
Token ID | 38421 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
aman | +0.028 |
berry | +0.026 |
umm | +0.026 |
en | +0.026 |
rap | +0.025 |
Neg logit contributions | |
protective | -0.024 |
sneaky | -0.024 |
fast | -0.022 |
dodged | -0.022 |
villains | -0.021 |
Token | umbers |
Token ID | 17024 |
Feature activation | +6.12e-02 |
Pos logit contributions | |
aman | +0.334 |
destination | +0.282 |
airport | +0.281 |
sunrise | +0.280 |
en | +0.277 |
Neg logit contributions | |
ars | -0.274 |
ages | -0.252 |
dodged | -0.239 |
sneaky | -0.230 |
protective | -0.229 |
Token | . |
Token ID | 13 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
aman | +0.141 |
umm | +0.134 |
berry | +0.127 |
en | +0.123 |
rap | +0.121 |
Neg logit contributions | |
protective | -0.084 |
sneaky | -0.081 |
fast | -0.081 |
dodged | -0.074 |
brave | -0.073 |
Token | And |
Token ID | 843 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
aman | +0.042 |
en | +0.040 |
umm | +0.040 |
berry | +0.039 |
rap | +0.039 |
Neg logit contributions | |
fast | -0.020 |
sneaky | -0.020 |
protective | -0.020 |
dodged | -0.018 |
villains | -0.017 |
Token | it |
Token ID | 340 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
protective | +0.054 |
sneaky | +0.052 |
dodged | +0.048 |
fast | +0.048 |
villains | +0.047 |
Neg logit contributions | |
aman | -0.080 |
berry | -0.078 |
en | -0.077 |
umm | -0.077 |
rap | -0.075 |
Token | has |
Token ID | 468 |
Feature activation | +3.22e-02 |
Pos logit contributions | |
protective | +0.016 |
sneaky | +0.016 |
dodged | +0.015 |
fast | +0.015 |
brave | +0.015 |
Neg logit contributions | |
aman | -0.024 |
umm | -0.021 |
berry | -0.020 |
rap | -0.019 |
date | -0.019 |
Token | are |
Token ID | 389 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
protective | +0.070 |
sneaky | +0.068 |
fast | +0.065 |
dodged | +0.064 |
brave | +0.062 |
Neg logit contributions | |
aman | -0.080 |
berry | -0.076 |
umm | -0.076 |
en | -0.074 |
rap | -0.072 |
Token | happy |
Token ID | 3772 |
Feature activation | -3.42e-02 |
Pos logit contributions | |
aman | +0.058 |
berry | +0.057 |
en | +0.057 |
rap | +0.056 |
umm | +0.055 |
Neg logit contributions | |
protective | -0.033 |
villains | -0.031 |
sneaky | -0.031 |
ars | -0.031 |
hog | -0.030 |
Token | with |
Token ID | 351 |
Feature activation | -6.09e-02 |
Pos logit contributions | |
protective | +0.047 |
fast | +0.047 |
sneaky | +0.047 |
brave | +0.043 |
villains | +0.041 |
Neg logit contributions | |
aman | -0.080 |
umm | -0.074 |
rap | -0.072 |
berry | -0.072 |
icer | -0.071 |
Token | their |
Token ID | 511 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
ars | +0.128 |
sneaky | +0.122 |
dodged | +0.122 |
villains | +0.122 |
ages | +0.121 |
Neg logit contributions | |
aman | -0.088 |
en | -0.084 |
el | -0.080 |
berry | -0.078 |
rap | -0.078 |
Token | snow |
Token ID | 6729 |
Feature activation | +4.31e-02 |
Pos logit contributions | |
protective | +0.048 |
sneaky | +0.046 |
fast | +0.046 |
brave | +0.045 |
sharp | +0.044 |
Neg logit contributions | |
aman | -0.051 |
umm | -0.051 |
en | -0.047 |
?' | -0.046 |
berry | -0.046 |
Token | fl |
Token ID | 2704 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
aman | +0.079 |
berry | +0.072 |
umm | +0.072 |
en | +0.071 |
rap | +0.069 |
Neg logit contributions | |
protective | -0.078 |
sneaky | -0.076 |
fast | -0.074 |
dodged | -0.071 |
ggy | -0.071 |
Token | akes |
Token ID | 1124 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
aman | +0.244 |
en | +0.210 |
reality | +0.210 |
airport | +0.199 |
reating | +0.197 |
Neg logit contributions | |
ars | -0.341 |
ages | -0.326 |
ggy | -0.321 |
dodged | -0.304 |
ards | -0.300 |
Token | . |
Token ID | 13 |
Feature activation | +9.29e-03 |
Pos logit contributions | |
protective | +0.023 |
sneaky | +0.023 |
fast | +0.021 |
dodged | +0.021 |
villains | +0.020 |
Neg logit contributions | |
aman | -0.024 |
berry | -0.023 |
en | -0.023 |
umm | -0.022 |
rap | -0.022 |
Token | They |
Token ID | 1119 |
Feature activation | -3.89e-02 |
Pos logit contributions | |
aman | +0.021 |
berry | +0.020 |
umm | +0.020 |
en | +0.020 |
rap | +0.019 |
Neg logit contributions | |
protective | -0.013 |
sneaky | -0.012 |
fast | -0.012 |
dodged | -0.011 |
villains | -0.011 |
Token | glue |
Token ID | 22749 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
protective | +0.066 |
sneaky | +0.064 |
fast | +0.061 |
dodged | +0.060 |
brave | +0.059 |
Neg logit contributions | |
aman | -0.075 |
berry | -0.071 |
umm | -0.071 |
en | -0.069 |
rap | -0.067 |
Token | them |
Token ID | 606 |
Feature activation | -3.29e-02 |
Pos logit contributions | |
aman | +0.030 |
umm | +0.030 |
berry | +0.027 |
?' | +0.026 |
icer | +0.026 |
Neg logit contributions | |
fast | -0.018 |
protective | -0.018 |
sneaky | -0.017 |
dodged | -0.016 |
brave | -0.016 |
Token | er |
Token ID | 1931 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
protective | +0.048 |
sneaky | +0.047 |
fast | +0.045 |
brave | +0.043 |
sharp | +0.043 |
Neg logit contributions | |
aman | -0.064 |
berry | -0.062 |
umm | -0.060 |
en | -0.056 |
neath | -0.055 |
Token | aser |
Token ID | 6005 |
Feature activation | -9.51e-03 |
Pos logit contributions | |
aman | +0.032 |
berry | +0.029 |
umm | +0.029 |
en | +0.029 |
rap | +0.027 |
Neg logit contributions | |
protective | -0.026 |
sneaky | -0.025 |
fast | -0.024 |
dodged | -0.023 |
brave | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
protective | +0.015 |
sneaky | +0.015 |
fast | +0.015 |
dodged | +0.014 |
brave | +0.014 |
Neg logit contributions | |
aman | -0.019 |
umm | -0.018 |
berry | -0.018 |
en | -0.018 |
rap | -0.017 |
Token | but |
Token ID | 475 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
aman | +0.042 |
berry | +0.042 |
en | +0.041 |
umm | +0.039 |
el | +0.038 |
Neg logit contributions | |
protective | -0.033 |
sneaky | -0.031 |
ars | -0.030 |
dodged | -0.029 |
ages | -0.028 |
Token | he |
Token ID | 339 |
Feature activation | +4.92e-02 |
Pos logit contributions | |
aman | +0.031 |
en | +0.030 |
berry | +0.030 |
umm | +0.029 |
rap | +0.028 |
Neg logit contributions | |
protective | -0.025 |
ars | -0.024 |
sneaky | -0.024 |
ages | -0.023 |
dodged | -0.023 |
Token | wanted |
Token ID | 2227 |
Feature activation | +4.25e-02 |
Pos logit contributions | |
aman | +0.097 |
berry | +0.096 |
en | +0.095 |
rap | +0.092 |
umm | +0.090 |
Neg logit contributions | |
protective | -0.075 |
sneaky | -0.072 |
fast | -0.067 |
villains | -0.065 |
ars | -0.064 |
Token | one |
Token ID | 530 |
Feature activation | +9.01e-04 |
Pos logit contributions | |
aman | +0.081 |
berry | +0.077 |
umm | +0.076 |
en | +0.075 |
rap | +0.073 |
Neg logit contributions | |
protective | -0.073 |
sneaky | -0.071 |
fast | -0.068 |
dodged | -0.066 |
brave | -0.064 |
Token | . |
Token ID | 13 |
Feature activation | -2.60e-04 |
Pos logit contributions | |
aman | +0.002 |
berry | +0.002 |
umm | +0.002 |
en | +0.002 |
rap | +0.002 |
Neg logit contributions | |
protective | -0.001 |
sneaky | -0.001 |
fast | -0.001 |
dodged | -0.001 |
brave | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +8.79e-03 |
Pos logit contributions | |
protective | +0.000 |
sneaky | +0.000 |
fast | +0.000 |
dodged | +0.000 |
villains | +0.000 |
Neg logit contributions | |
aman | -0.001 |
berry | -0.001 |
umm | -0.001 |
en | -0.001 |
rap | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | +4.14e-04 |
Pos logit contributions | |
aman | +0.016 |
berry | +0.016 |
umm | +0.015 |
en | +0.015 |
rap | +0.015 |
Neg logit contributions | |
protective | -0.015 |
sneaky | -0.015 |
fast | -0.014 |
dodged | -0.014 |
brave | -0.014 |
Token | " |
Token ID | 1 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
aman | +0.001 |
berry | +0.001 |
en | +0.001 |
rap | +0.001 |
umm | +0.001 |
Neg logit contributions | |
protective | -0.001 |
sneaky | -0.001 |
dodged | -0.001 |
fast | -0.001 |
villains | -0.001 |
Token | your |
Token ID | 534 |
Feature activation | +5.78e-03 |
Pos logit contributions | |
aman | +0.116 |
berry | +0.110 |
umm | +0.108 |
en | +0.107 |
rap | +0.104 |
Neg logit contributions | |
protective | -0.107 |
sneaky | -0.105 |
fast | -0.100 |
dodged | -0.097 |
brave | -0.095 |
Token | friends |
Token ID | 2460 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
aman | +0.011 |
berry | +0.011 |
umm | +0.011 |
en | +0.011 |
rap | +0.010 |
Neg logit contributions | |
protective | -0.009 |
sneaky | -0.009 |
fast | -0.009 |
dodged | -0.009 |
villains | -0.008 |
Token | will |
Token ID | 481 |
Feature activation | +6.78e-02 |
Pos logit contributions | |
protective | +0.028 |
sneaky | +0.027 |
fast | +0.026 |
dodged | +0.025 |
brave | +0.025 |
Neg logit contributions | |
aman | -0.029 |
umm | -0.027 |
berry | -0.027 |
en | -0.026 |
rap | -0.026 |
Token | believe |
Token ID | 1975 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
aman | +0.135 |
berry | +0.128 |
umm | +0.128 |
en | +0.124 |
rap | +0.120 |
Neg logit contributions | |
protective | -0.111 |
sneaky | -0.108 |
fast | -0.104 |
dodged | -0.100 |
brave | -0.098 |
Token | that |
Token ID | 326 |
Feature activation | +5.99e-02 |
Pos logit contributions | |
aman | +0.067 |
berry | +0.063 |
umm | +0.062 |
en | +0.060 |
rap | +0.059 |
Neg logit contributions | |
protective | -0.055 |
sneaky | -0.053 |
fast | -0.052 |
brave | -0.049 |
dodged | -0.049 |
Token | the |
Token ID | 262 |
Feature activation | +4.15e-02 |
Pos logit contributions | |
aman | +0.115 |
berry | +0.111 |
en | +0.110 |
umm | +0.108 |
rap | +0.105 |
Neg logit contributions | |
protective | -0.099 |
sneaky | -0.096 |
fast | -0.088 |
dodged | -0.088 |
ggy | -0.085 |
Token | ink |
Token ID | 16882 |
Feature activation | +6.52e-02 |
Pos logit contributions | |
aman | +0.084 |
berry | +0.080 |
en | +0.078 |
umm | +0.078 |
rap | +0.076 |
Neg logit contributions | |
protective | -0.067 |
sneaky | -0.066 |
fast | -0.062 |
dodged | -0.060 |
brave | -0.059 |
Token | won |
Token ID | 1839 |
Feature activation | +3.11e-02 |
Pos logit contributions | |
aman | +0.108 |
berry | +0.102 |
umm | +0.100 |
en | +0.099 |
rap | +0.096 |
Neg logit contributions | |
protective | -0.127 |
sneaky | -0.124 |
fast | -0.118 |
dodged | -0.116 |
villains | -0.113 |
Token | 't |
Token ID | 470 |
Feature activation | -5.19e-02 |
Pos logit contributions | |
aman | +0.063 |
umm | +0.060 |
berry | +0.060 |
en | +0.057 |
rap | +0.056 |
Neg logit contributions | |
protective | -0.049 |
sneaky | -0.048 |
fast | -0.047 |
dodged | -0.044 |
brave | -0.044 |
Token | hurt |
Token ID | 5938 |
Feature activation | -1.94e-03 |
Pos logit contributions | |
protective | +0.090 |
sneaky | +0.087 |
fast | +0.081 |
dodged | +0.081 |
ggy | +0.079 |
Neg logit contributions | |
aman | -0.097 |
berry | -0.092 |
en | -0.091 |
umm | -0.090 |
rap | -0.087 |
Token | them |
Token ID | 606 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
protective | +0.003 |
sneaky | +0.003 |
fast | +0.003 |
dodged | +0.003 |
villains | +0.003 |
Neg logit contributions | |
aman | -0.004 |
berry | -0.003 |
umm | -0.003 |
en | -0.003 |
rap | -0.003 |
Token | Tom |
Token ID | 13787 |
Feature activation | -3.78e-02 |
Pos logit contributions | |
protective | +0.013 |
sneaky | +0.013 |
fast | +0.011 |
villains | +0.011 |
dodged | +0.011 |
Neg logit contributions | |
aman | -0.020 |
berry | -0.019 |
rap | -0.018 |
en | -0.018 |
umm | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | -3.94e-02 |
Pos logit contributions | |
protective | +0.063 |
sneaky | +0.060 |
villains | +0.058 |
aro | +0.057 |
fast | +0.055 |
Neg logit contributions | |
berry | -0.072 |
en | -0.072 |
aman | -0.071 |
rap | -0.065 |
umm | -0.062 |
Token | Jen |
Token ID | 13364 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
protective | +0.058 |
sneaky | +0.056 |
fast | +0.051 |
villains | +0.050 |
dodged | +0.050 |
Neg logit contributions | |
aman | -0.085 |
berry | -0.082 |
en | -0.080 |
umm | -0.078 |
rap | -0.077 |
Token | hugged |
Token ID | 41130 |
Feature activation | -6.84e-03 |
Pos logit contributions | |
protective | +0.047 |
sneaky | +0.045 |
fast | +0.041 |
villains | +0.040 |
ggy | +0.040 |
Neg logit contributions | |
aman | -0.071 |
berry | -0.070 |
en | -0.068 |
umm | -0.066 |
rap | -0.064 |
Token | and |
Token ID | 290 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
protective | +0.012 |
sneaky | +0.012 |
fast | +0.011 |
dodged | +0.011 |
villains | +0.011 |
Neg logit contributions | |
aman | -0.013 |
berry | -0.012 |
en | -0.012 |
umm | -0.012 |
rap | -0.012 |
Token | said |
Token ID | 531 |
Feature activation | +6.44e-02 |
Pos logit contributions | |
protective | +0.038 |
sneaky | +0.035 |
villains | +0.034 |
aro | +0.034 |
hog | +0.034 |
Neg logit contributions | |
en | -0.036 |
aman | -0.036 |
berry | -0.035 |
amaz | -0.033 |
sets | -0.032 |
Token | , |
Token ID | 11 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
aman | +0.105 |
berry | +0.104 |
en | +0.100 |
umm | +0.098 |
rap | +0.096 |
Neg logit contributions | |
protective | -0.122 |
sneaky | -0.119 |
fast | -0.110 |
dodged | -0.110 |
villains | -0.109 |
Token | " |
Token ID | 366 |
Feature activation | -7.11e-03 |
Pos logit contributions | |
protective | +0.031 |
sneaky | +0.029 |
villains | +0.027 |
ggy | +0.027 |
dodged | +0.027 |
Neg logit contributions | |
aman | -0.032 |
berry | -0.031 |
en | -0.029 |
umm | -0.029 |
rap | -0.028 |
Token | We |
Token ID | 1135 |
Feature activation | -1.54e-03 |
Pos logit contributions | |
protective | +0.015 |
sneaky | +0.015 |
villains | +0.013 |
ggy | +0.013 |
fast | +0.013 |
Neg logit contributions | |
umm | -0.011 |
aman | -0.010 |
berry | -0.010 |
en | -0.010 |
date | -0.009 |
Token | are |
Token ID | 389 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
protective | +0.003 |
sneaky | +0.003 |
brave | +0.002 |
dodged | +0.002 |
fast | +0.002 |
Neg logit contributions | |
aman | -0.003 |
umm | -0.003 |
berry | -0.003 |
en | -0.003 |
rap | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
aman | +0.047 |
berry | +0.045 |
umm | +0.044 |
en | +0.044 |
rap | +0.043 |
Neg logit contributions | |
protective | -0.023 |
sneaky | -0.022 |
fast | -0.021 |
dodged | -0.020 |
villains | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -7.46e-03 |
Pos logit contributions | |
aman | +0.028 |
umm | +0.028 |
berry | +0.026 |
rap | +0.025 |
en | +0.025 |
Neg logit contributions | |
protective | -0.018 |
sneaky | -0.018 |
fast | -0.017 |
brave | -0.016 |
dodged | -0.016 |
Token | |
Token ID | 220 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
protective | +0.013 |
sneaky | +0.012 |
fast | +0.012 |
dodged | +0.011 |
villains | +0.011 |
Neg logit contributions | |
aman | -0.014 |
berry | -0.014 |
umm | -0.014 |
en | -0.013 |
rap | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | +1.54e-03 |
Pos logit contributions | |
aman | +0.039 |
umm | +0.037 |
berry | +0.037 |
en | +0.037 |
rap | +0.035 |
Neg logit contributions | |
protective | -0.033 |
sneaky | -0.033 |
fast | -0.032 |
dodged | -0.031 |
brave | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | -6.11e-03 |
Pos logit contributions | |
aman | +0.003 |
berry | +0.003 |
umm | +0.003 |
en | +0.003 |
rap | +0.002 |
Neg logit contributions | |
protective | -0.003 |
sneaky | -0.003 |
fast | -0.003 |
dodged | -0.003 |
villains | -0.002 |
Token | " |
Token ID | 1 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
protective | +0.009 |
sneaky | +0.009 |
dodged | +0.008 |
fast | +0.008 |
ggy | +0.008 |
Neg logit contributions | |
aman | -0.012 |
berry | -0.012 |
en | -0.012 |
umm | -0.012 |
rap | -0.012 |
Token | Hi |
Token ID | 17250 |
Feature activation | +7.07e-02 |
Pos logit contributions | |
protective | +0.053 |
sneaky | +0.053 |
fast | +0.051 |
dodged | +0.050 |
brave | +0.048 |
Neg logit contributions | |
aman | -0.052 |
umm | -0.048 |
en | -0.047 |
rap | -0.047 |
berry | -0.047 |
Token | Sp |
Token ID | 1338 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
aman | +0.177 |
umm | +0.165 |
en | +0.151 |
date | +0.150 |
rap | +0.147 |
Neg logit contributions | |
fast | -0.091 |
brave | -0.090 |
sneaky | -0.088 |
protective | -0.080 |
smart | -0.077 |
Token | ud |
Token ID | 463 |
Feature activation | -9.19e-02 |
Pos logit contributions | |
protective | +0.046 |
sneaky | +0.045 |
ggy | +0.043 |
fast | +0.043 |
dodged | +0.042 |
Neg logit contributions | |
aman | -0.041 |
berry | -0.037 |
umm | -0.037 |
en | -0.037 |
destination | -0.035 |
Token | !" |
Token ID | 2474 |
Feature activation | -4.08e-02 |
Pos logit contributions | |
protective | +0.171 |
sneaky | +0.167 |
fast | +0.157 |
dodged | +0.154 |
villains | +0.153 |
Neg logit contributions | |
aman | -0.160 |
berry | -0.156 |
en | -0.151 |
umm | -0.147 |
rap | -0.144 |
Token | said |
Token ID | 531 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
protective | +0.067 |
sneaky | +0.062 |
villains | +0.062 |
ggy | +0.060 |
cunning | +0.059 |
Neg logit contributions | |
berry | -0.075 |
en | -0.069 |
el | -0.067 |
aman | -0.065 |
umm | -0.064 |
Token | Lily |
Token ID | 20037 |
Feature activation | -4.38e-02 |
Pos logit contributions | |
aman | +0.048 |
berry | +0.046 |
en | +0.045 |
umm | +0.045 |
rap | +0.043 |
Neg logit contributions | |
protective | -0.035 |
sneaky | -0.034 |
fast | -0.031 |
dodged | -0.031 |
villains | -0.030 |
Token | had |
Token ID | 550 |
Feature activation | +9.52e-03 |
Pos logit contributions | |
protective | +0.014 |
sneaky | +0.014 |
fast | +0.013 |
dodged | +0.012 |
villains | +0.012 |
Neg logit contributions | |
aman | -0.020 |
berry | -0.019 |
en | -0.019 |
umm | -0.019 |
rap | -0.018 |
Token | seen |
Token ID | 1775 |
Feature activation | -4.55e-02 |
Pos logit contributions | |
aman | +0.018 |
berry | +0.017 |
en | +0.017 |
umm | +0.016 |
rap | +0.016 |
Neg logit contributions | |
protective | -0.016 |
sneaky | -0.015 |
fast | -0.014 |
dodged | -0.014 |
villains | -0.014 |
Token | many |
Token ID | 867 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
protective | +0.074 |
sneaky | +0.072 |
fast | +0.067 |
dodged | +0.067 |
ars | +0.065 |
Neg logit contributions | |
aman | -0.089 |
berry | -0.085 |
en | -0.085 |
umm | -0.083 |
rap | -0.080 |
Token | things |
Token ID | 1243 |
Feature activation | -3.00e-03 |
Pos logit contributions | |
protective | +0.047 |
sneaky | +0.046 |
dodged | +0.045 |
ggy | +0.043 |
ages | +0.043 |
Neg logit contributions | |
aman | -0.065 |
en | -0.064 |
berry | -0.064 |
umm | -0.061 |
rap | -0.058 |
Token | and |
Token ID | 290 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
protective | +0.005 |
sneaky | +0.005 |
fast | +0.004 |
dodged | +0.004 |
villains | +0.004 |
Neg logit contributions | |
aman | -0.006 |
berry | -0.006 |
en | -0.006 |
umm | -0.005 |
rap | -0.005 |
Token | was |
Token ID | 373 |
Feature activation | +5.37e-02 |
Pos logit contributions | |
en | +0.048 |
aman | +0.048 |
berry | +0.046 |
umm | +0.045 |
rap | +0.043 |
Neg logit contributions | |
protective | -0.042 |
sneaky | -0.040 |
dodged | -0.038 |
ars | -0.038 |
fast | -0.037 |
Token | even |
Token ID | 772 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
aman | +0.125 |
berry | +0.121 |
umm | +0.120 |
en | +0.118 |
rap | +0.115 |
Neg logit contributions | |
protective | -0.070 |
sneaky | -0.068 |
fast | -0.064 |
dodged | -0.060 |
brave | -0.060 |
Token | brave |
Token ID | 14802 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
aman | +0.050 |
en | +0.049 |
berry | +0.048 |
umm | +0.047 |
rap | +0.044 |
Neg logit contributions | |
protective | -0.044 |
sneaky | -0.042 |
ggy | -0.041 |
ars | -0.040 |
dodged | -0.040 |
Token | enough |
Token ID | 1576 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
protective | +0.019 |
sneaky | +0.018 |
fast | +0.018 |
brave | +0.017 |
dodged | +0.017 |
Neg logit contributions | |
aman | -0.021 |
berry | -0.020 |
rap | -0.019 |
umm | -0.019 |
icer | -0.018 |
Token | to |
Token ID | 284 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
aman | +0.064 |
berry | +0.060 |
en | +0.059 |
umm | +0.059 |
rap | +0.056 |
Neg logit contributions | |
protective | -0.068 |
sneaky | -0.066 |
dodged | -0.063 |
fast | -0.062 |
villains | -0.062 |
Token | stand |
Token ID | 1302 |
Feature activation | -7.18e-04 |
Pos logit contributions | |
aman | +0.106 |
berry | +0.101 |
umm | +0.099 |
en | +0.098 |
rap | +0.094 |
Neg logit contributions | |
protective | -0.104 |
sneaky | -0.102 |
fast | -0.097 |
dodged | -0.095 |
brave | -0.092 |
Token | - |
Token ID | 12 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
aman | +0.109 |
en | +0.108 |
berry | +0.107 |
el | +0.101 |
rap | +0.099 |
Neg logit contributions | |
protective | -0.125 |
sneaky | -0.123 |
dodged | -0.112 |
cunning | -0.111 |
villains | -0.111 |
Token | round |
Token ID | 744 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
sneaky | +0.037 |
protective | +0.036 |
slim | +0.033 |
dodged | +0.032 |
cunning | +0.032 |
Neg logit contributions | |
aman | -0.043 |
rap | -0.039 |
berry | -0.039 |
en | -0.039 |
umm | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | +5.99e-03 |
Pos logit contributions | |
protective | +0.075 |
sneaky | +0.073 |
villains | +0.067 |
dodged | +0.066 |
cunning | +0.065 |
Neg logit contributions | |
aman | -0.071 |
berry | -0.068 |
en | -0.068 |
umm | -0.063 |
rap | -0.062 |
Token | They |
Token ID | 1119 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
aman | +0.011 |
berry | +0.011 |
umm | +0.010 |
en | +0.010 |
rap | +0.010 |
Neg logit contributions | |
protective | -0.011 |
sneaky | -0.010 |
fast | -0.010 |
dodged | -0.009 |
villains | -0.009 |
Token | wondered |
Token ID | 14028 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
protective | +0.075 |
sneaky | +0.069 |
ggy | +0.068 |
hog | +0.066 |
ages | +0.066 |
Neg logit contributions | |
aman | -0.076 |
berry | -0.072 |
en | -0.072 |
el | -0.066 |
destination | -0.066 |
Token | what |
Token ID | 644 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
aman | +0.071 |
berry | +0.066 |
en | +0.065 |
rap | +0.063 |
umm | +0.062 |
Neg logit contributions | |
protective | -0.078 |
sneaky | -0.076 |
dodged | -0.071 |
fast | -0.070 |
ars | -0.070 |
Token | was |
Token ID | 373 |
Feature activation | +4.71e-02 |
Pos logit contributions | |
berry | +0.019 |
aman | +0.019 |
en | +0.018 |
date | +0.017 |
?' | +0.017 |
Neg logit contributions | |
protective | -0.022 |
sneaky | -0.021 |
ars | -0.021 |
hind | -0.020 |
dodged | -0.020 |
Token | wrong |
Token ID | 2642 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
aman | +0.072 |
en | +0.069 |
berry | +0.065 |
rap | +0.065 |
umm | +0.062 |
Neg logit contributions | |
protective | -0.095 |
sneaky | -0.092 |
dodged | -0.089 |
villains | -0.086 |
ars | -0.085 |
Token | with |
Token ID | 351 |
Feature activation | -5.36e-02 |
Pos logit contributions | |
aman | +0.042 |
berry | +0.040 |
umm | +0.040 |
en | +0.038 |
rap | +0.037 |
Neg logit contributions | |
protective | -0.034 |
sneaky | -0.033 |
fast | -0.032 |
dodged | -0.030 |
brave | -0.030 |
Token | it |
Token ID | 340 |
Feature activation | +6.58e-03 |
Pos logit contributions | |
protective | +0.103 |
sneaky | +0.101 |
hog | +0.100 |
dodged | +0.096 |
villains | +0.095 |
Neg logit contributions | |
aman | -0.085 |
en | -0.083 |
el | -0.078 |
rap | -0.076 |
berry | -0.076 |
Token | . |
Token ID | 13 |
Feature activation | -7.42e-03 |
Pos logit contributions | |
aman | +0.012 |
berry | +0.011 |
en | +0.011 |
umm | +0.011 |
rap | +0.011 |
Neg logit contributions | |
protective | -0.012 |
sneaky | -0.011 |
dodged | -0.011 |
fast | -0.011 |
villains | -0.011 |
Token | but |
Token ID | 475 |
Feature activation | +6.61e-03 |
Pos logit contributions | |
aman | +0.023 |
berry | +0.022 |
en | +0.021 |
umm | +0.021 |
rap | +0.020 |
Neg logit contributions | |
protective | -0.014 |
sneaky | -0.014 |
dodged | -0.013 |
fast | -0.012 |
ars | -0.012 |
Token | I |
Token ID | 314 |
Feature activation | +5.47e-02 |
Pos logit contributions | |
aman | +0.013 |
berry | +0.013 |
umm | +0.012 |
en | +0.012 |
rap | +0.012 |
Neg logit contributions | |
protective | -0.011 |
sneaky | -0.010 |
fast | -0.010 |
dodged | -0.009 |
villains | -0.009 |
Token | 'm |
Token ID | 1101 |
Feature activation | +4.50e-02 |
Pos logit contributions | |
aman | +0.092 |
berry | +0.088 |
en | +0.085 |
umm | +0.083 |
rap | +0.081 |
Neg logit contributions | |
protective | -0.105 |
sneaky | -0.102 |
fast | -0.097 |
dodged | -0.095 |
villains | -0.094 |
Token | not |
Token ID | 407 |
Feature activation | -4.10e-02 |
Pos logit contributions | |
aman | +0.088 |
en | +0.081 |
berry | +0.080 |
?' | +0.079 |
rap | +0.078 |
Neg logit contributions | |
protective | -0.066 |
ars | -0.064 |
ages | -0.064 |
villains | -0.064 |
dodged | -0.062 |
Token | dangerous |
Token ID | 4923 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
protective | +0.050 |
ars | +0.046 |
sneaky | +0.046 |
hog | +0.045 |
dodged | +0.045 |
Neg logit contributions | |
aman | -0.092 |
en | -0.090 |
berry | -0.088 |
rap | -0.085 |
?' | -0.084 |
Token | . |
Token ID | 13 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
protective | +0.028 |
sneaky | +0.028 |
fast | +0.027 |
brave | +0.025 |
dodged | +0.025 |
Neg logit contributions | |
aman | -0.033 |
berry | -0.031 |
umm | -0.031 |
en | -0.030 |
rap | -0.030 |
Token | Let |
Token ID | 3914 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
aman | +0.019 |
berry | +0.019 |
umm | +0.019 |
en | +0.018 |
rap | +0.017 |
Neg logit contributions | |
protective | -0.022 |
sneaky | -0.022 |
dodged | -0.021 |
fast | -0.020 |
ggy | -0.020 |
Token | me |
Token ID | 502 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
aman | +0.042 |
berry | +0.040 |
umm | +0.040 |
en | +0.039 |
rap | +0.038 |
Neg logit contributions | |
protective | -0.037 |
sneaky | -0.036 |
fast | -0.034 |
dodged | -0.033 |
brave | -0.033 |
Token | show |
Token ID | 905 |
Feature activation | -5.05e-02 |
Pos logit contributions | |
aman | +0.027 |
berry | +0.026 |
en | +0.026 |
umm | +0.026 |
rap | +0.025 |
Neg logit contributions | |
protective | -0.027 |
sneaky | -0.026 |
fast | -0.024 |
dodged | -0.024 |
villains | -0.023 |
Token | you |
Token ID | 345 |
Feature activation | -2.33e-02 |
Pos logit contributions | |
protective | +0.090 |
sneaky | +0.088 |
dodged | +0.081 |
fast | +0.080 |
villains | +0.079 |
Neg logit contributions | |
aman | -0.091 |
en | -0.087 |
berry | -0.086 |
rap | -0.083 |
umm | -0.082 |
Token | ." |
Token ID | 526 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
protective | +0.042 |
sneaky | +0.042 |
dodged | +0.040 |
villains | +0.039 |
fast | +0.038 |
Neg logit contributions | |
aman | -0.040 |
en | -0.039 |
berry | -0.039 |
rap | -0.037 |
?' | -0.035 |
Token | shout |
Token ID | 9191 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
protective | +0.069 |
sneaky | +0.067 |
dodged | +0.062 |
fast | +0.060 |
villains | +0.058 |
Neg logit contributions | |
aman | -0.133 |
en | -0.130 |
berry | -0.130 |
umm | -0.125 |
rap | -0.123 |
Token | coming |
Token ID | 2406 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
protective | +0.019 |
sneaky | +0.019 |
fast | +0.018 |
brave | +0.017 |
dodged | +0.017 |
Neg logit contributions | |
aman | -0.021 |
umm | -0.020 |
berry | -0.020 |
en | -0.019 |
rap | -0.019 |
Token | from |
Token ID | 422 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
protective | +0.078 |
sneaky | +0.077 |
fast | +0.074 |
dodged | +0.070 |
brave | +0.070 |
Neg logit contributions | |
aman | -0.084 |
berry | -0.079 |
umm | -0.078 |
en | -0.077 |
rap | -0.074 |
Token | the |
Token ID | 262 |
Feature activation | -8.07e-04 |
Pos logit contributions | |
protective | +0.042 |
sneaky | +0.041 |
fast | +0.041 |
brave | +0.037 |
dodged | +0.036 |
Neg logit contributions | |
aman | -0.069 |
umm | -0.066 |
berry | -0.066 |
en | -0.062 |
rap | -0.062 |
Token | beach |
Token ID | 10481 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
protective | +0.002 |
sneaky | +0.002 |
fast | +0.001 |
dodged | +0.001 |
villains | +0.001 |
Neg logit contributions | |
aman | -0.001 |
berry | -0.001 |
umm | -0.001 |
en | -0.001 |
rap | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +8.27e-03 |
Pos logit contributions | |
aman | +0.029 |
berry | +0.028 |
en | +0.028 |
umm | +0.027 |
rap | +0.026 |
Neg logit contributions | |
protective | -0.029 |
sneaky | -0.028 |
dodged | -0.026 |
fast | -0.026 |
villains | -0.026 |
Token | It |
Token ID | 632 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
aman | +0.014 |
berry | +0.013 |
umm | +0.013 |
en | +0.012 |
?' | +0.012 |
Neg logit contributions | |
protective | -0.016 |
sneaky | -0.015 |
dodged | -0.014 |
ars | -0.014 |
fast | -0.014 |
Token | was |
Token ID | 373 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
protective | +0.085 |
sneaky | +0.081 |
ages | +0.080 |
ars | +0.080 |
villains | +0.079 |
Neg logit contributions | |
en | -0.057 |
aman | -0.054 |
berry | -0.051 |
el | -0.047 |
neath | -0.046 |
Token | a |
Token ID | 257 |
Feature activation | -7.31e-03 |
Pos logit contributions | |
aman | +0.034 |
berry | +0.032 |
umm | +0.032 |
en | +0.032 |
rap | +0.031 |
Neg logit contributions | |
protective | -0.023 |
sneaky | -0.022 |
fast | -0.021 |
dodged | -0.020 |
brave | -0.020 |
Token | little |
Token ID | 1310 |
Feature activation | +2.31e-02 |
Pos logit contributions | |
protective | +0.009 |
ages | +0.009 |
ars | +0.009 |
dodged | +0.009 |
aro | +0.008 |
Neg logit contributions | |
aman | -0.016 |
berry | -0.016 |
umm | -0.015 |
date | -0.015 |
destination | -0.014 |
Token | girl |
Token ID | 2576 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
berry | +0.042 |
aman | +0.041 |
umm | +0.040 |
en | +0.039 |
el | +0.035 |
Neg logit contributions | |
protective | -0.038 |
ars | -0.037 |
dodged | -0.037 |
ggy | -0.037 |
sneaky | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
aman | +0.067 |
berry | +0.063 |
umm | +0.062 |
en | +0.061 |
rap | +0.061 |
Neg logit contributions | |
protective | -0.052 |
sneaky | -0.051 |
fast | -0.050 |
brave | -0.047 |
dodged | -0.046 |
Token | And |
Token ID | 843 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
aman | +0.022 |
berry | +0.021 |
umm | +0.021 |
en | +0.020 |
rap | +0.020 |
Neg logit contributions | |
protective | -0.022 |
sneaky | -0.021 |
dodged | -0.020 |
fast | -0.020 |
villains | -0.019 |
Token | from |
Token ID | 422 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
protective | +0.054 |
sneaky | +0.051 |
dodged | +0.047 |
ars | +0.046 |
fast | +0.046 |
Neg logit contributions | |
aman | -0.081 |
berry | -0.077 |
umm | -0.076 |
en | -0.076 |
rap | -0.074 |
Token | that |
Token ID | 326 |
Feature activation | +3.01e-02 |
Pos logit contributions | |
protective | +0.022 |
sneaky | +0.021 |
dodged | +0.020 |
ars | +0.019 |
villains | +0.019 |
Neg logit contributions | |
aman | -0.033 |
en | -0.032 |
berry | -0.031 |
rap | -0.030 |
umm | -0.030 |
Token | day |
Token ID | 1110 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
aman | +0.054 |
umm | +0.052 |
berry | +0.051 |
en | +0.049 |
rap | +0.049 |
Neg logit contributions | |
protective | -0.056 |
sneaky | -0.055 |
fast | -0.054 |
brave | -0.052 |
dodged | -0.049 |
Token | on |
Token ID | 319 |
Feature activation | -2.66e-02 |
Pos logit contributions | |
aman | +0.046 |
berry | +0.044 |
umm | +0.043 |
en | +0.043 |
rap | +0.041 |
Neg logit contributions | |
protective | -0.040 |
sneaky | -0.039 |
fast | -0.038 |
dodged | -0.036 |
brave | -0.036 |
Token | , |
Token ID | 11 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
protective | +0.045 |
sneaky | +0.044 |
fast | +0.042 |
dodged | +0.040 |
ggy | +0.040 |
Neg logit contributions | |
aman | -0.052 |
berry | -0.049 |
umm | -0.049 |
en | -0.048 |
rap | -0.047 |
Token | Tom |
Token ID | 4186 |
Feature activation | -3.97e-03 |
Pos logit contributions | |
aman | +0.021 |
berry | +0.020 |
umm | +0.020 |
en | +0.020 |
rap | +0.019 |
Neg logit contributions | |
protective | -0.016 |
sneaky | -0.015 |
fast | -0.014 |
dodged | -0.014 |
villains | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | -1.74e-03 |
Pos logit contributions | |
protective | +0.006 |
sneaky | +0.006 |
fast | +0.006 |
dodged | +0.006 |
brave | +0.006 |
Neg logit contributions | |
aman | -0.008 |
umm | -0.008 |
berry | -0.008 |
en | -0.008 |
rap | -0.007 |
Token | his |
Token ID | 465 |
Feature activation | +2.44e-02 |
Pos logit contributions | |
protective | +0.003 |
sneaky | +0.003 |
fast | +0.003 |
dodged | +0.003 |
ars | +0.002 |
Neg logit contributions | |
aman | -0.003 |
berry | -0.003 |
en | -0.003 |
umm | -0.003 |
rap | -0.003 |
Token | mom |
Token ID | 1995 |
Feature activation | +8.56e-02 |
Pos logit contributions | |
aman | +0.038 |
berry | +0.037 |
umm | +0.035 |
en | +0.034 |
destination | +0.033 |
Neg logit contributions | |
protective | -0.048 |
sneaky | -0.046 |
dodged | -0.046 |
hog | -0.044 |
ars | -0.044 |
Token | down |
Token ID | 866 |
Feature activation | +4.05e-02 |
Pos logit contributions | |
aman | +0.031 |
berry | +0.030 |
en | +0.029 |
umm | +0.029 |
rap | +0.027 |
Neg logit contributions | |
protective | -0.036 |
sneaky | -0.036 |
dodged | -0.033 |
fast | -0.033 |
villains | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | +2.45e-03 |
Pos logit contributions | |
aman | +0.083 |
berry | +0.078 |
umm | +0.078 |
en | +0.076 |
rap | +0.075 |
Neg logit contributions | |
protective | -0.064 |
sneaky | -0.062 |
fast | -0.060 |
dodged | -0.057 |
brave | -0.056 |
Token |
|
Token ID | 198 |
Feature activation | +8.45e-03 |
Pos logit contributions | |
aman | +0.005 |
berry | +0.005 |
umm | +0.005 |
en | +0.005 |
rap | +0.005 |
Neg logit contributions | |
protective | -0.004 |
sneaky | -0.004 |
fast | -0.004 |
dodged | -0.003 |
villains | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -8.60e-04 |
Pos logit contributions | |
aman | +0.016 |
umm | +0.015 |
berry | +0.015 |
en | +0.015 |
rap | +0.014 |
Neg logit contributions | |
protective | -0.015 |
sneaky | -0.014 |
fast | -0.014 |
dodged | -0.013 |
ggy | -0.013 |
Token | " |
Token ID | 1 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
protective | +0.001 |
sneaky | +0.001 |
fast | +0.001 |
dodged | +0.001 |
cunning | +0.001 |
Neg logit contributions | |
aman | -0.002 |
berry | -0.002 |
en | -0.002 |
rap | -0.002 |
umm | -0.002 |
Token | Can |
Token ID | 6090 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
protective | +0.053 |
sneaky | +0.052 |
fast | +0.049 |
dodged | +0.048 |
villains | +0.047 |
Neg logit contributions | |
aman | -0.044 |
berry | -0.042 |
umm | -0.041 |
en | -0.041 |
rap | -0.039 |
Token | I |
Token ID | 314 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
berry | +0.017 |
aman | +0.017 |
el | +0.015 |
umm | +0.015 |
?' | +0.015 |
Neg logit contributions | |
protective | -0.020 |
sneaky | -0.019 |
ages | -0.017 |
ars | -0.017 |
dodged | -0.017 |
Token | go |
Token ID | 467 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
protective | +0.031 |
sneaky | +0.030 |
fast | +0.030 |
brave | +0.028 |
dodged | +0.028 |
Neg logit contributions | |
aman | -0.039 |
umm | -0.037 |
en | -0.036 |
berry | -0.035 |
rap | -0.034 |
Token | on |
Token ID | 319 |
Feature activation | -8.46e-02 |
Pos logit contributions | |
protective | +0.043 |
fast | +0.043 |
sneaky | +0.043 |
brave | +0.040 |
dodged | +0.040 |
Neg logit contributions | |
aman | -0.051 |
umm | -0.049 |
berry | -0.046 |
en | -0.046 |
rap | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
protective | +0.127 |
sneaky | +0.124 |
fast | +0.120 |
dodged | +0.113 |
brave | +0.112 |
Neg logit contributions | |
aman | -0.178 |
umm | -0.170 |
berry | -0.170 |
en | -0.167 |
rap | -0.162 |
Token | log |
Token ID | 2604 |
Feature activation | -7.37e-02 |
Pos logit contributions | |
protective | +0.044 |
sneaky | +0.043 |
dodged | +0.042 |
ages | +0.040 |
ars | +0.040 |
Neg logit contributions | |
aman | -0.048 |
berry | -0.046 |
en | -0.044 |
umm | -0.043 |
destination | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | +3.19e-03 |
Pos logit contributions | |
protective | +0.006 |
sneaky | +0.006 |
dodged | +0.005 |
ggy | +0.005 |
villains | +0.005 |
Neg logit contributions | |
aman | -0.005 |
berry | -0.005 |
en | -0.004 |
umm | -0.004 |
rap | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -3.52e-03 |
Pos logit contributions | |
aman | +0.006 |
berry | +0.005 |
en | +0.005 |
umm | +0.005 |
rap | +0.005 |
Neg logit contributions | |
protective | -0.006 |
sneaky | -0.006 |
fast | -0.005 |
dodged | -0.005 |
villains | -0.005 |
Token | They |
Token ID | 2990 |
Feature activation | -4.52e-02 |
Pos logit contributions | |
protective | +0.006 |
sneaky | +0.006 |
aro | +0.005 |
dodged | +0.005 |
ggy | +0.005 |
Neg logit contributions | |
aman | -0.006 |
berry | -0.006 |
en | -0.006 |
rap | -0.006 |
date | -0.005 |
Token | give |
Token ID | 1577 |
Feature activation | -5.89e-02 |
Pos logit contributions | |
protective | +0.072 |
sneaky | +0.071 |
fast | +0.069 |
dodged | +0.068 |
brave | +0.067 |
Neg logit contributions | |
aman | -0.090 |
umm | -0.089 |
berry | -0.086 |
en | -0.084 |
rap | -0.081 |
Token | the |
Token ID | 262 |
Feature activation | -7.17e-02 |
Pos logit contributions | |
protective | +0.094 |
sneaky | +0.092 |
fast | +0.091 |
brave | +0.085 |
dodged | +0.085 |
Neg logit contributions | |
aman | -0.120 |
umm | -0.115 |
berry | -0.111 |
en | -0.109 |
rap | -0.106 |
Token | book |
Token ID | 1492 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
protective | +0.128 |
sneaky | +0.123 |
dodged | +0.117 |
fast | +0.116 |
villains | +0.114 |
Neg logit contributions | |
aman | -0.130 |
berry | -0.125 |
umm | -0.120 |
en | -0.118 |
rap | -0.117 |
Token | back |
Token ID | 736 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
protective | +0.232 |
sneaky | +0.226 |
dodged | +0.218 |
villains | +0.216 |
cunning | +0.210 |
Neg logit contributions | |
aman | -0.159 |
berry | -0.154 |
en | -0.146 |
rap | -0.144 |
destination | -0.139 |
Token | to |
Token ID | 284 |
Feature activation | +5.91e-03 |
Pos logit contributions | |
protective | +0.037 |
sneaky | +0.036 |
dodged | +0.034 |
villains | +0.034 |
cunning | +0.033 |
Neg logit contributions | |
aman | -0.029 |
berry | -0.028 |
en | -0.027 |
rap | -0.025 |
el | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
aman | +0.011 |
berry | +0.010 |
en | +0.010 |
rap | +0.010 |
umm | +0.010 |
Neg logit contributions | |
protective | -0.011 |
sneaky | -0.010 |
dodged | -0.010 |
ars | -0.009 |
villains | -0.009 |
Token | l |
Token ID | 300 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
protective | +0.023 |
sneaky | +0.023 |
fast | +0.021 |
dodged | +0.020 |
brave | +0.020 |
Neg logit contributions | |
aman | -0.041 |
berry | -0.039 |
en | -0.039 |
umm | -0.039 |
rap | -0.037 |
Token | ibrarian |
Token ID | 35808 |
Feature activation | -3.04e-04 |
Pos logit contributions | |
protective | +0.015 |
sneaky | +0.015 |
fast | +0.014 |
dodged | +0.014 |
brave | +0.013 |
Neg logit contributions | |
aman | -0.021 |
berry | -0.020 |
umm | -0.020 |
rap | -0.020 |
en | -0.019 |
Token | The |
Token ID | 464 |
Feature activation | -4.78e-02 |
Pos logit contributions | |
aman | +0.002 |
rap | +0.002 |
en | +0.001 |
berry | +0.001 |
?' | +0.001 |
Neg logit contributions | |
protective | -0.001 |
sneaky | -0.001 |
aro | -0.001 |
dodged | -0.001 |
villains | -0.001 |
Token | prince |
Token ID | 19716 |
Feature activation | -5.37e-02 |
Pos logit contributions | |
protective | +0.095 |
dodged | +0.089 |
sneaky | +0.089 |
ggy | +0.087 |
fast | +0.086 |
Neg logit contributions | |
berry | -0.072 |
aman | -0.072 |
umm | -0.070 |
en | -0.068 |
rap | -0.064 |
Token | brave |
Token ID | 14802 |
Feature activation | -4.89e-02 |
Pos logit contributions | |
protective | +0.092 |
sneaky | +0.087 |
fast | +0.083 |
villains | +0.082 |
ars | +0.079 |
Neg logit contributions | |
en | -0.098 |
aman | -0.097 |
berry | -0.095 |
el | -0.088 |
rap | -0.087 |
Token | ly |
Token ID | 306 |
Feature activation | +3.27e-04 |
Pos logit contributions | |
protective | +0.058 |
sneaky | +0.056 |
fast | +0.052 |
dodged | +0.049 |
brave | +0.049 |
Neg logit contributions | |
aman | -0.120 |
umm | -0.113 |
berry | -0.113 |
rap | -0.110 |
en | -0.109 |
Token | stepped |
Token ID | 10764 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
aman | +0.001 |
en | +0.001 |
berry | +0.001 |
el | +0.001 |
rap | +0.001 |
Neg logit contributions | |
protective | -0.001 |
sneaky | -0.001 |
fast | -0.000 |
villains | -0.000 |
ars | -0.000 |
Token | forward |
Token ID | 2651 |
Feature activation | -8.50e-02 |
Pos logit contributions | |
aman | +0.028 |
berry | +0.027 |
en | +0.026 |
umm | +0.026 |
rap | +0.025 |
Neg logit contributions | |
protective | -0.029 |
sneaky | -0.028 |
dodged | -0.026 |
fast | -0.026 |
villains | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | -6.98e-04 |
Pos logit contributions | |
protective | +0.149 |
sneaky | +0.146 |
fast | +0.139 |
dodged | +0.135 |
brave | +0.132 |
Neg logit contributions | |
aman | -0.158 |
berry | -0.150 |
umm | -0.148 |
en | -0.146 |
rap | -0.141 |
Token | said |
Token ID | 531 |
Feature activation | +6.60e-02 |
Pos logit contributions | |
ars | +0.001 |
villains | +0.001 |
protective | +0.001 |
ages | +0.001 |
ggy | +0.001 |
Neg logit contributions | |
en | -0.001 |
aman | -0.001 |
neath | -0.001 |
destination | -0.001 |
again | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
aman | +0.107 |
berry | +0.103 |
en | +0.100 |
rap | +0.097 |
umm | +0.097 |
Neg logit contributions | |
protective | -0.124 |
sneaky | -0.120 |
dodged | -0.115 |
villains | -0.113 |
fast | -0.113 |
Token | " |
Token ID | 366 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
protective | +0.027 |
sneaky | +0.026 |
dodged | +0.024 |
villains | +0.024 |
fast | +0.024 |
Neg logit contributions | |
aman | -0.032 |
berry | -0.030 |
en | -0.029 |
umm | -0.029 |
rap | -0.028 |
Token | Dragon |
Token ID | 17808 |
Feature activation | -4.80e-02 |
Pos logit contributions | |
protective | +0.025 |
sneaky | +0.024 |
fast | +0.023 |
dodged | +0.023 |
brave | +0.022 |
Neg logit contributions | |
aman | -0.028 |
berry | -0.027 |
umm | -0.027 |
rap | -0.027 |
en | -0.026 |
Token | |
Token ID | 220 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
aman | +0.001 |
berry | +0.001 |
en | +0.001 |
umm | +0.001 |
rap | +0.001 |
Neg logit contributions | |
protective | -0.001 |
sneaky | -0.001 |
fast | -0.001 |
dodged | -0.001 |
villains | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
aman | +0.031 |
umm | +0.031 |
en | +0.030 |
berry | +0.030 |
rap | +0.028 |
Neg logit contributions | |
protective | -0.022 |
fast | -0.022 |
sneaky | -0.021 |
dodged | -0.020 |
ggy | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | +4.55e-03 |
Pos logit contributions | |
aman | +0.019 |
berry | +0.018 |
en | +0.016 |
rap | +0.016 |
umm | +0.016 |
Neg logit contributions | |
protective | -0.024 |
sneaky | -0.024 |
dodged | -0.022 |
villains | -0.022 |
fast | -0.021 |
Token | Andy |
Token ID | 35314 |
Feature activation | -6.79e-02 |
Pos logit contributions | |
aman | +0.009 |
berry | +0.009 |
en | +0.009 |
rap | +0.009 |
umm | +0.008 |
Neg logit contributions | |
protective | -0.007 |
sneaky | -0.007 |
fast | -0.006 |
dodged | -0.006 |
afraid | -0.006 |
Token | was |
Token ID | 373 |
Feature activation | +7.43e-06 |
Pos logit contributions | |
protective | +0.100 |
sneaky | +0.097 |
fast | +0.091 |
dodged | +0.089 |
brave | +0.086 |
Neg logit contributions | |
aman | -0.145 |
berry | -0.139 |
umm | -0.137 |
en | -0.136 |
rap | -0.132 |
Token | very |
Token ID | 845 |
Feature activation | -1.07e-01 |
Token | grateful |
Token ID | 14066 |
Feature activation | -6.84e-03 |
Pos logit contributions | |
villains | +0.156 |
Rex | +0.151 |
aro | +0.146 |
ars | +0.143 |
protective | +0.140 |
Neg logit contributions | |
aman | -0.197 |
berry | -0.195 |
rap | -0.184 |
en | -0.182 |
umm | -0.180 |
Token | and |
Token ID | 290 |
Feature activation | -1.82e-03 |
Pos logit contributions | |
protective | +0.013 |
sneaky | +0.013 |
dodged | +0.012 |
fast | +0.012 |
villains | +0.012 |
Neg logit contributions | |
aman | -0.012 |
berry | -0.011 |
en | -0.011 |
umm | -0.011 |
rap | -0.010 |
Token | said |
Token ID | 531 |
Feature activation | +6.33e-02 |
Pos logit contributions | |
protective | +0.003 |
ars | +0.003 |
villains | +0.003 |
sneaky | +0.003 |
Rex | +0.003 |
Neg logit contributions | |
aman | -0.003 |
en | -0.003 |
berry | -0.003 |
umm | -0.003 |
once | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
berry | +0.096 |
aman | +0.096 |
umm | +0.091 |
en | +0.090 |
rap | +0.087 |
Neg logit contributions | |
protective | -0.123 |
sneaky | -0.118 |
ars | -0.112 |
dodged | -0.112 |
villains | -0.111 |
Token | " |
Token ID | 366 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
protective | +0.031 |
sneaky | +0.030 |
dodged | +0.027 |
villains | +0.027 |
fast | +0.027 |
Neg logit contributions | |
aman | -0.036 |
berry | -0.034 |
umm | -0.033 |
en | -0.033 |
rap | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | +3.09e-02 |
Pos logit contributions | |
fast | +0.028 |
protective | +0.028 |
sneaky | +0.028 |
brave | +0.026 |
sharp | +0.024 |
Neg logit contributions | |
aman | -0.036 |
umm | -0.036 |
berry | -0.033 |
rap | -0.033 |
en | -0.033 |
Token | park |
Token ID | 3952 |
Feature activation | -6.86e-03 |
Pos logit contributions | |
aman | +0.052 |
umm | +0.050 |
rap | +0.049 |
en | +0.048 |
berry | +0.048 |
Neg logit contributions | |
protective | -0.059 |
fast | -0.058 |
sneaky | -0.057 |
brave | -0.054 |
sharp | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | -1.05e-03 |
Pos logit contributions | |
protective | +0.012 |
sneaky | +0.012 |
dodged | +0.011 |
villains | +0.011 |
fast | +0.011 |
Neg logit contributions | |
aman | -0.012 |
berry | -0.012 |
en | -0.012 |
umm | -0.011 |
rap | -0.011 |
Token | She |
Token ID | 1375 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
protective | +0.002 |
sneaky | +0.002 |
fast | +0.002 |
brave | +0.002 |
dodged | +0.002 |
Neg logit contributions | |
aman | -0.002 |
en | -0.002 |
umm | -0.002 |
berry | -0.002 |
rap | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | -2.56e-03 |
Pos logit contributions | |
aman | +0.025 |
en | +0.024 |
berry | +0.024 |
rap | +0.023 |
umm | +0.022 |
Neg logit contributions | |
protective | -0.026 |
sneaky | -0.025 |
villains | -0.023 |
dodged | -0.022 |
hog | -0.022 |
Token | very |
Token ID | 845 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
protective | +0.004 |
ars | +0.004 |
villains | +0.004 |
hog | +0.004 |
sneaky | +0.004 |
Neg logit contributions | |
aman | -0.004 |
berry | -0.004 |
en | -0.004 |
rap | -0.004 |
date | -0.004 |
Token | alert |
Token ID | 7995 |
Feature activation | -4.72e-02 |
Pos logit contributions | |
Rex | +0.143 |
villains | +0.139 |
hog | +0.135 |
ages | +0.132 |
dodged | +0.132 |
Neg logit contributions | |
rap | -0.190 |
umm | -0.185 |
aman | -0.184 |
berry | -0.184 |
date | -0.182 |
Token | , |
Token ID | 11 |
Feature activation | -3.02e-03 |
Pos logit contributions | |
protective | +0.083 |
sneaky | +0.081 |
fast | +0.077 |
dodged | +0.075 |
villains | +0.073 |
Neg logit contributions | |
aman | -0.087 |
berry | -0.083 |
umm | -0.082 |
en | -0.081 |
rap | -0.078 |
Token | and |
Token ID | 290 |
Feature activation | -9.31e-03 |
Pos logit contributions | |
protective | +0.005 |
sneaky | +0.005 |
ars | +0.005 |
villains | +0.005 |
ggy | +0.005 |
Neg logit contributions | |
aman | -0.006 |
berry | -0.005 |
en | -0.005 |
umm | -0.005 |
rap | -0.005 |
Token | watched |
Token ID | 7342 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
protective | +0.015 |
sneaky | +0.014 |
villains | +0.013 |
dodged | +0.013 |
ars | +0.013 |
Neg logit contributions | |
aman | -0.018 |
berry | -0.017 |
en | -0.017 |
umm | -0.017 |
rap | -0.016 |
Token | everything |
Token ID | 2279 |
Feature activation | -5.54e-03 |
Pos logit contributions | |
aman | +0.073 |
umm | +0.071 |
berry | +0.068 |
icer | +0.068 |
destination | +0.064 |
Neg logit contributions | |
fast | -0.040 |
sneaky | -0.035 |
brave | -0.035 |
protective | -0.035 |
strong | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | +3.15e-03 |
Pos logit contributions | |
aman | +0.040 |
umm | +0.039 |
berry | +0.038 |
en | +0.038 |
rap | +0.036 |
Neg logit contributions | |
protective | -0.031 |
sneaky | -0.030 |
fast | -0.030 |
dodged | -0.028 |
ggy | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | -5.30e-03 |
Pos logit contributions | |
aman | +0.006 |
umm | +0.006 |
berry | +0.006 |
en | +0.006 |
rap | +0.006 |
Neg logit contributions | |
protective | -0.005 |
sneaky | -0.005 |
fast | -0.005 |
dodged | -0.005 |
brave | -0.005 |
Token | L |
Token ID | 43 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
protective | +0.009 |
sneaky | +0.009 |
aro | +0.008 |
dodged | +0.008 |
villains | +0.008 |
Neg logit contributions | |
aman | -0.010 |
berry | -0.009 |
en | -0.009 |
rap | -0.009 |
date | -0.009 |
Token | ily |
Token ID | 813 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
protective | +0.026 |
sneaky | +0.026 |
fast | +0.023 |
ggy | +0.023 |
brave | +0.022 |
Neg logit contributions | |
aman | -0.078 |
umm | -0.076 |
berry | -0.074 |
destination | -0.073 |
en | -0.072 |
Token | felt |
Token ID | 2936 |
Feature activation | -4.84e-02 |
Pos logit contributions | |
protective | +0.055 |
sneaky | +0.053 |
aro | +0.052 |
villains | +0.050 |
fast | +0.050 |
Neg logit contributions | |
aman | -0.052 |
en | -0.051 |
berry | -0.047 |
el | -0.045 |
rap | -0.044 |
Token | sad |
Token ID | 6507 |
Feature activation | -8.87e-02 |
Pos logit contributions | |
cious | +0.066 |
aro | +0.065 |
ike | +0.065 |
hog | +0.064 |
Rex | +0.063 |
Neg logit contributions | |
en | -0.090 |
once | -0.089 |
?' | -0.089 |
aman | -0.088 |
el | -0.087 |
Token | because |
Token ID | 780 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
protective | +0.134 |
sneaky | +0.131 |
fast | +0.127 |
brave | +0.118 |
dodged | +0.118 |
Neg logit contributions | |
aman | -0.188 |
berry | -0.180 |
umm | -0.177 |
en | -0.173 |
rap | -0.170 |
Token | she |
Token ID | 673 |
Feature activation | +7.14e-02 |
Pos logit contributions | |
protective | +0.040 |
sneaky | +0.040 |
fast | +0.039 |
brave | +0.036 |
dodged | +0.035 |
Neg logit contributions | |
aman | -0.062 |
berry | -0.059 |
umm | -0.059 |
en | -0.058 |
rap | -0.057 |
Token | didn |
Token ID | 1422 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
en | +0.125 |
aman | +0.124 |
berry | +0.119 |
rap | +0.116 |
el | +0.114 |
Neg logit contributions | |
protective | -0.116 |
aro | -0.116 |
sneaky | -0.108 |
ars | -0.105 |
hog | -0.105 |
Token | 't |
Token ID | 470 |
Feature activation | -7.30e-02 |
Pos logit contributions | |
protective | +0.020 |
sneaky | +0.020 |
fast | +0.020 |
ggy | +0.018 |
dodged | +0.018 |
Neg logit contributions | |
aman | -0.028 |
berry | -0.027 |
umm | -0.027 |
en | -0.026 |
rap | -0.025 |
Token | mean |
Token ID | 1612 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
protective | +0.126 |
sneaky | +0.123 |
hog | +0.116 |
dodged | +0.114 |
aro | +0.113 |
Neg logit contributions | |
en | -0.129 |
aman | -0.125 |
berry | -0.124 |
rap | -0.121 |
umm | -0.119 |