Token | He |
Token ID | 1544 |
Feature activation | +9.06e-02 |
Pos logit contributions | |
ck | +0.113 |
snatch | +0.111 |
unw | +0.109 |
rily | +0.108 |
uzz | +0.108 |
Neg logit contributions | |
End | -0.102 |
sky | -0.084 |
end | -0.082 |
happiest | -0.071 |
problem | -0.071 |
Token | smiled |
Token ID | 13541 |
Feature activation | -4.53e-02 |
Pos logit contributions | |
zip | +0.180 |
Flo | +0.176 |
snatch | +0.176 |
ck | +0.172 |
uzz | +0.169 |
Neg logit contributions | |
End | -0.118 |
sky | -0.110 |
problem | -0.110 |
end | -0.101 |
best | -0.095 |
Token | from |
Token ID | 422 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
End | +0.070 |
problem | +0.069 |
sky | +0.064 |
end | +0.059 |
happiest | +0.059 |
Neg logit contributions | |
snatch | -0.077 |
zip | -0.077 |
Flo | -0.076 |
ck | -0.076 |
unw | -0.075 |
Token | ear |
Token ID | 1027 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
zip | +0.300 |
Flo | +0.297 |
zzy | +0.286 |
snatch | +0.285 |
ck | +0.281 |
Neg logit contributions | |
End | -0.240 |
problem | -0.215 |
sky | -0.210 |
end | -0.204 |
gloomy | -0.195 |
Token | - |
Token ID | 12 |
Feature activation | +9.59e-02 |
Pos logit contributions | |
End | +0.043 |
problem | +0.041 |
sky | +0.040 |
end | +0.037 |
happiest | +0.037 |
Neg logit contributions | |
snatch | -0.043 |
zip | -0.043 |
ck | -0.042 |
Flo | -0.042 |
zzy | -0.041 |
Token | to |
Token ID | 1462 |
Feature activation | +3.44e-01 |
Pos logit contributions | |
snatch | +0.148 |
zip | +0.144 |
Flo | +0.141 |
ck | +0.140 |
unw | +0.138 |
Neg logit contributions | |
End | -0.167 |
sky | -0.149 |
day | -0.142 |
end | -0.141 |
problem | -0.140 |
Token | - |
Token ID | 12 |
Feature activation | +9.61e-02 |
Pos logit contributions | |
zip | +0.721 |
Flo | +0.720 |
snatch | +0.719 |
ck | +0.701 |
zzy | +0.686 |
Neg logit contributions | |
End | -0.405 |
problem | -0.381 |
sky | -0.354 |
end | -0.332 |
best | -0.320 |
Token | ear |
Token ID | 451 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
snatch | +0.136 |
zip | +0.131 |
Flo | +0.129 |
ck | +0.126 |
unw | +0.125 |
Neg logit contributions | |
End | -0.180 |
sky | -0.162 |
day | -0.155 |
end | -0.154 |
problem | -0.152 |
Token | with |
Token ID | 351 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
zip | +0.236 |
snatch | +0.234 |
Flo | +0.230 |
ck | +0.228 |
zzy | +0.223 |
Neg logit contributions | |
End | -0.163 |
problem | -0.152 |
sky | -0.152 |
end | -0.139 |
best | -0.135 |
Token | joy |
Token ID | 8716 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
zip | +0.346 |
Flo | +0.344 |
snatch | +0.340 |
ck | +0.338 |
zzy | +0.327 |
Neg logit contributions | |
End | -0.297 |
problem | -0.281 |
sky | -0.273 |
end | -0.259 |
best | -0.253 |
Token | . |
Token ID | 13 |
Feature activation | -4.35e-02 |
Pos logit contributions | |
End | +0.021 |
problem | +0.020 |
sky | +0.019 |
happiest | +0.018 |
end | +0.017 |
Neg logit contributions | |
ck | -0.027 |
snatch | -0.027 |
zip | -0.027 |
unw | -0.027 |
Flo | -0.027 |
Token | on |
Token ID | 319 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
snatch | +0.086 |
Flo | +0.086 |
zip | +0.085 |
ck | +0.084 |
unw | +0.081 |
Neg logit contributions | |
End | -0.082 |
problem | -0.079 |
sky | -0.076 |
end | -0.072 |
best | -0.072 |
Token | her |
Token ID | 607 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
snatch | +0.226 |
zip | +0.226 |
ck | +0.224 |
Flo | +0.223 |
zzy | +0.215 |
Neg logit contributions | |
End | -0.185 |
problem | -0.178 |
sky | -0.171 |
end | -0.158 |
happiest | -0.156 |
Token | t |
Token ID | 256 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
End | +0.045 |
problem | +0.041 |
sky | +0.039 |
gloomy | +0.036 |
happiest | +0.036 |
Neg logit contributions | |
snatch | -0.040 |
unw | -0.039 |
Flo | -0.039 |
zip | -0.038 |
ck | -0.038 |
Token | ippy |
Token ID | 41214 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
snatch | +0.037 |
ck | +0.036 |
Flo | +0.035 |
zip | +0.033 |
unw | +0.033 |
Neg logit contributions | |
problem | -0.051 |
End | -0.051 |
sky | -0.051 |
end | -0.047 |
best | -0.046 |
Token | - |
Token ID | 12 |
Feature activation | +9.58e-02 |
Pos logit contributions | |
Flo | +0.464 |
ck | +0.453 |
zip | +0.452 |
snatch | +0.450 |
unw | +0.441 |
Neg logit contributions | |
End | -0.433 |
problem | -0.406 |
sky | -0.379 |
happiest | -0.376 |
gloomy | -0.360 |
Token | to |
Token ID | 1462 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
snatch | +0.097 |
ck | +0.091 |
Flo | +0.088 |
unw | +0.084 |
zip | +0.083 |
Neg logit contributions | |
End | -0.214 |
sky | -0.205 |
problem | -0.199 |
end | -0.196 |
day | -0.191 |
Token | es |
Token ID | 274 |
Feature activation | +5.77e-02 |
Pos logit contributions | |
snatch | +0.556 |
zip | +0.550 |
Flo | +0.544 |
ck | +0.542 |
unw | +0.521 |
Neg logit contributions | |
End | -0.561 |
problem | -0.545 |
sky | -0.523 |
end | -0.493 |
best | -0.492 |
Token | . |
Token ID | 13 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
snatch | +0.127 |
Flo | +0.127 |
zip | +0.126 |
ck | +0.126 |
zzy | +0.119 |
Neg logit contributions | |
End | -0.059 |
problem | -0.058 |
sky | -0.057 |
end | -0.053 |
best | -0.049 |
Token | St |
Token ID | 520 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
sky | +0.029 |
End | +0.027 |
problem | +0.025 |
gloomy | +0.023 |
day | +0.023 |
Neg logit contributions | |
ck | -0.045 |
snatch | -0.043 |
zip | -0.041 |
Flo | -0.041 |
steer | -0.040 |
Token | ret |
Token ID | 1186 |
Feature activation | +5.93e-02 |
Pos logit contributions | |
snatch | +0.214 |
zip | +0.214 |
ck | +0.212 |
Flo | +0.211 |
zzy | +0.201 |
Neg logit contributions | |
End | -0.267 |
problem | -0.260 |
sky | -0.250 |
end | -0.236 |
happiest | -0.234 |
Token | ching |
Token ID | 10813 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
ck | +0.085 |
zip | +0.084 |
Flo | +0.084 |
snatch | +0.083 |
zzy | +0.079 |
Neg logit contributions | |
End | -0.111 |
problem | -0.108 |
sky | -0.103 |
happiest | -0.098 |
end | -0.098 |
Token | bedroom |
Token ID | 14043 |
Feature activation | -6.85e-02 |
Pos logit contributions | |
End | +0.029 |
problem | +0.026 |
sky | +0.024 |
happiest | +0.023 |
end | +0.023 |
Neg logit contributions | |
zip | -0.041 |
unw | -0.041 |
ck | -0.040 |
snatch | -0.040 |
Flo | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
End | +0.111 |
problem | +0.095 |
happiest | +0.089 |
sky | +0.086 |
end | +0.083 |
Neg logit contributions | |
unw | -0.119 |
zip | -0.112 |
snatch | -0.111 |
zzy | -0.107 |
ck | -0.106 |
Token | She |
Token ID | 1375 |
Feature activation | +5.76e-02 |
Pos logit contributions | |
sky | +0.026 |
End | +0.026 |
problem | +0.025 |
gloomy | +0.022 |
end | +0.022 |
Neg logit contributions | |
ck | -0.049 |
snatch | -0.047 |
Flo | -0.046 |
zip | -0.046 |
zzy | -0.045 |
Token | tip |
Token ID | 8171 |
Feature activation | +5.02e-02 |
Pos logit contributions | |
snatch | +0.114 |
Flo | +0.113 |
zip | +0.110 |
ck | +0.109 |
izer | +0.107 |
Neg logit contributions | |
problem | -0.075 |
End | -0.074 |
sky | -0.070 |
end | -0.065 |
gloomy | -0.064 |
Token | - |
Token ID | 12 |
Feature activation | +9.71e-02 |
Pos logit contributions | |
zip | +0.089 |
ck | +0.089 |
snatch | +0.088 |
Flo | +0.088 |
zzy | +0.085 |
Neg logit contributions | |
End | -0.076 |
problem | -0.073 |
sky | -0.069 |
happiest | -0.065 |
end | -0.065 |
Token | to |
Token ID | 1462 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
zip | +0.107 |
ck | +0.104 |
snatch | +0.103 |
Flo | +0.103 |
zzy | +0.097 |
Neg logit contributions | |
End | -0.214 |
problem | -0.210 |
sky | -0.201 |
happiest | -0.194 |
end | -0.193 |
Token | ed |
Token ID | 276 |
Feature activation | +9.88e-02 |
Pos logit contributions | |
zip | +0.636 |
ck | +0.631 |
snatch | +0.627 |
Flo | +0.622 |
zzy | +0.602 |
Neg logit contributions | |
End | -0.496 |
problem | -0.476 |
sky | -0.454 |
happiest | -0.420 |
end | -0.418 |
Token | to |
Token ID | 284 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
Flo | +0.193 |
snatch | +0.192 |
zzy | +0.187 |
ck | +0.187 |
onna | +0.186 |
Neg logit contributions | |
sky | -0.128 |
End | -0.122 |
problem | -0.122 |
end | -0.116 |
gloomy | -0.115 |
Token | the |
Token ID | 262 |
Feature activation | -9.39e-02 |
Pos logit contributions | |
Flo | +0.291 |
onna | +0.278 |
igi | +0.275 |
zzy | +0.274 |
uzz | +0.267 |
Neg logit contributions | |
end | -0.214 |
sky | -0.200 |
End | -0.198 |
gloomy | -0.188 |
problem | -0.187 |
Token | door |
Token ID | 3420 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
End | +0.115 |
problem | +0.108 |
sky | +0.095 |
happiest | +0.092 |
best | +0.089 |
Neg logit contributions | |
snatch | -0.189 |
zip | -0.189 |
Flo | -0.186 |
unw | -0.185 |
ck | -0.181 |
Token | , |
Token ID | 11 |
Feature activation | -7.71e-02 |
Pos logit contributions | |
End | +0.027 |
problem | +0.026 |
sky | +0.024 |
happiest | +0.023 |
best | +0.023 |
Neg logit contributions | |
zip | -0.032 |
snatch | -0.031 |
ck | -0.031 |
unw | -0.031 |
Flo | -0.030 |
Token | worked |
Token ID | 3111 |
Feature activation | -6.42e-02 |
Pos logit contributions | |
zip | +0.118 |
snatch | +0.118 |
ck | +0.117 |
Flo | +0.112 |
unw | +0.112 |
Neg logit contributions | |
End | -0.137 |
problem | -0.132 |
sky | -0.128 |
happiest | -0.120 |
end | -0.119 |
Token | , |
Token ID | 11 |
Feature activation | -8.50e-02 |
Pos logit contributions | |
End | +0.083 |
sky | +0.076 |
problem | +0.076 |
happiest | +0.069 |
end | +0.067 |
Neg logit contributions | |
zip | -0.127 |
ck | -0.126 |
snatch | -0.126 |
Flo | -0.125 |
unw | -0.122 |
Token | strugg |
Token ID | 3569 |
Feature activation | +7.22e-02 |
Pos logit contributions | |
End | +0.134 |
problem | +0.131 |
sky | +0.126 |
best | +0.118 |
happiest | +0.118 |
Neg logit contributions | |
zip | -0.144 |
Flo | -0.143 |
ck | -0.141 |
snatch | -0.141 |
zzy | -0.138 |
Token | lin |
Token ID | 2815 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
zip | +0.137 |
ck | +0.133 |
Flo | +0.131 |
snatch | +0.130 |
zzy | +0.129 |
Neg logit contributions | |
End | -0.104 |
problem | -0.098 |
sky | -0.094 |
end | -0.089 |
happiest | -0.088 |
Token | g |
Token ID | 308 |
Feature activation | +4.06e-03 |
Pos logit contributions | |
ck | +0.208 |
zip | +0.208 |
Flo | +0.203 |
unw | +0.202 |
steer | +0.202 |
Neg logit contributions | |
End | -0.158 |
problem | -0.127 |
sky | -0.124 |
end | -0.120 |
gloomy | -0.118 |
Token | to |
Token ID | 1462 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
ck | +0.006 |
zip | +0.006 |
snatch | +0.006 |
Flo | +0.006 |
uzz | +0.006 |
Neg logit contributions | |
End | -0.007 |
problem | -0.007 |
sky | -0.007 |
end | -0.006 |
happiest | -0.006 |
Token | be |
Token ID | 307 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
zip | +0.539 |
ck | +0.538 |
snatch | +0.531 |
Flo | +0.525 |
zzy | +0.501 |
Neg logit contributions | |
End | -0.595 |
problem | -0.566 |
sky | -0.563 |
happiest | -0.519 |
gloomy | -0.517 |
Token | the |
Token ID | 262 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
End | +0.020 |
problem | +0.019 |
sky | +0.019 |
end | +0.017 |
happiest | +0.017 |
Neg logit contributions | |
zip | -0.023 |
ck | -0.023 |
snatch | -0.023 |
Flo | -0.023 |
zzy | -0.022 |
Token | cle |
Token ID | 1190 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
End | +0.086 |
problem | +0.078 |
sky | +0.074 |
gloomy | +0.063 |
happiest | +0.059 |
Neg logit contributions | |
snatch | -0.244 |
zip | -0.240 |
Flo | -0.238 |
ck | -0.236 |
unw | -0.233 |
Token | ve |
Token ID | 303 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
End | +0.023 |
problem | +0.023 |
best | +0.021 |
end | +0.020 |
sky | +0.020 |
Neg logit contributions | |
zip | -0.022 |
Flo | -0.022 |
snatch | -0.022 |
ck | -0.021 |
unw | -0.021 |
Token | rest |
Token ID | 2118 |
Feature activation | +5.59e-02 |
Pos logit contributions | |
snatch | +0.282 |
zip | +0.282 |
ck | +0.282 |
Flo | +0.278 |
zzy | +0.267 |
Neg logit contributions | |
End | -0.316 |
problem | -0.307 |
sky | -0.296 |
happiest | -0.275 |
gloomy | -0.275 |
Token | cold |
Token ID | 4692 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
End | +0.062 |
problem | +0.059 |
sky | +0.053 |
end | +0.050 |
happiest | +0.048 |
Neg logit contributions | |
Flo | -0.113 |
zip | -0.112 |
ck | -0.112 |
snatch | -0.110 |
unw | -0.108 |
Token | . |
Token ID | 13 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
End | +0.201 |
problem | +0.177 |
happiest | +0.168 |
goals | +0.156 |
end | +0.150 |
Neg logit contributions | |
Flo | -0.153 |
zip | -0.151 |
unw | -0.145 |
snatch | -0.140 |
ck | -0.138 |
Token | Jane |
Token ID | 12091 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
End | +0.042 |
problem | +0.041 |
sky | +0.038 |
happiest | +0.037 |
end | +0.037 |
Neg logit contributions | |
zip | -0.032 |
snatch | -0.032 |
Flo | -0.031 |
ck | -0.031 |
zzy | -0.030 |
Token | tip |
Token ID | 8171 |
Feature activation | +5.06e-02 |
Pos logit contributions | |
ck | +0.337 |
Flo | +0.332 |
zip | +0.332 |
snatch | +0.327 |
zzy | +0.325 |
Neg logit contributions | |
End | -0.176 |
problem | -0.168 |
sky | -0.153 |
happiest | -0.139 |
end | -0.138 |
Token | - |
Token ID | 12 |
Feature activation | +9.68e-02 |
Pos logit contributions | |
zip | +0.091 |
ck | +0.091 |
Flo | +0.089 |
snatch | +0.088 |
zzy | +0.086 |
Neg logit contributions | |
End | -0.076 |
problem | -0.074 |
sky | -0.068 |
happiest | -0.064 |
end | -0.064 |
Token | to |
Token ID | 1462 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
zip | +0.133 |
snatch | +0.132 |
ck | +0.132 |
Flo | +0.131 |
zzy | +0.125 |
Neg logit contributions | |
End | -0.185 |
problem | -0.180 |
sky | -0.173 |
end | -0.164 |
happiest | -0.163 |
Token | ed |
Token ID | 276 |
Feature activation | +9.90e-02 |
Pos logit contributions | |
zip | +0.636 |
ck | +0.634 |
snatch | +0.632 |
Flo | +0.626 |
zzy | +0.605 |
Neg logit contributions | |
End | -0.489 |
problem | -0.472 |
sky | -0.449 |
end | -0.413 |
happiest | -0.413 |
Token | outside |
Token ID | 2354 |
Feature activation | +7.69e-02 |
Pos logit contributions | |
snatch | +0.183 |
zip | +0.181 |
ck | +0.180 |
Flo | +0.180 |
zzy | +0.178 |
Neg logit contributions | |
sky | -0.135 |
problem | -0.133 |
End | -0.133 |
gloomy | -0.125 |
best | -0.123 |
Token | and |
Token ID | 290 |
Feature activation | +8.72e-02 |
Pos logit contributions | |
snatch | +0.110 |
zip | +0.108 |
ck | +0.107 |
unw | +0.105 |
Flo | +0.104 |
Neg logit contributions | |
End | -0.140 |
problem | -0.138 |
sky | -0.127 |
happiest | -0.125 |
best | -0.123 |
Token | smiled |
Token ID | 13541 |
Feature activation | -4.02e-02 |
Pos logit contributions | |
zip | +0.151 |
ck | +0.150 |
snatch | +0.149 |
Flo | +0.146 |
unw | +0.145 |
Neg logit contributions | |
End | -0.137 |
problem | -0.133 |
sky | -0.122 |
end | -0.115 |
best | -0.114 |
Token | . |
Token ID | 13 |
Feature activation | -3.03e-02 |
Pos logit contributions | |
End | +0.071 |
problem | +0.068 |
sky | +0.060 |
end | +0.056 |
happiest | +0.055 |
Neg logit contributions | |
unw | -0.059 |
Flo | -0.059 |
zip | -0.058 |
zzy | -0.058 |
ck | -0.056 |
Token | j |
Token ID | 474 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
snatch | +0.015 |
zip | +0.015 |
ck | +0.015 |
Flo | +0.015 |
zzy | +0.014 |
Neg logit contributions | |
End | -0.010 |
problem | -0.009 |
sky | -0.009 |
end | -0.008 |
happiest | -0.008 |
Token | olly |
Token ID | 5098 |
Feature activation | +6.93e-02 |
Pos logit contributions | |
Flo | +0.026 |
zip | +0.026 |
ck | +0.025 |
snatch | +0.025 |
zzy | +0.025 |
Neg logit contributions | |
End | -0.022 |
problem | -0.021 |
sky | -0.020 |
happiest | -0.019 |
best | -0.019 |
Token | deer |
Token ID | 20096 |
Feature activation | +4.83e-02 |
Pos logit contributions | |
zip | +0.141 |
snatch | +0.139 |
ck | +0.137 |
Flo | +0.135 |
zzy | +0.131 |
Neg logit contributions | |
End | -0.088 |
problem | -0.086 |
sky | -0.082 |
end | -0.077 |
gloomy | -0.075 |
Token | named |
Token ID | 3706 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
snatch | +0.068 |
zip | +0.068 |
ck | +0.068 |
Flo | +0.067 |
zzy | +0.064 |
Neg logit contributions | |
End | -0.089 |
problem | -0.087 |
sky | -0.084 |
end | -0.079 |
happiest | -0.078 |
Token | B |
Token ID | 347 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
snatch | +0.180 |
ck | +0.173 |
har | +0.170 |
zip | +0.169 |
zzy | +0.168 |
Neg logit contributions | |
sky | -0.188 |
day | -0.177 |
best | -0.175 |
problem | -0.172 |
night | -0.171 |
Token | amb |
Token ID | 4131 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
snatch | +0.193 |
ck | +0.183 |
zip | +0.177 |
Flo | +0.169 |
unw | +0.168 |
Neg logit contributions | |
End | -0.247 |
sky | -0.243 |
problem | -0.236 |
end | -0.224 |
best | -0.223 |
Token | i |
Token ID | 72 |
Feature activation | +5.91e-02 |
Pos logit contributions | |
snatch | +0.465 |
zip | +0.465 |
ck | +0.464 |
Flo | +0.458 |
zzy | +0.436 |
Neg logit contributions | |
End | -0.651 |
problem | -0.634 |
sky | -0.615 |
end | -0.578 |
gloomy | -0.575 |
Token | . |
Token ID | 13 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
Flo | +0.110 |
ck | +0.109 |
zip | +0.107 |
uzz | +0.105 |
zzy | +0.104 |
Neg logit contributions | |
End | -0.085 |
problem | -0.085 |
sky | -0.076 |
happiest | -0.073 |
gloomy | -0.071 |
Token | B |
Token ID | 347 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
End | +0.032 |
problem | +0.031 |
sky | +0.031 |
gloomy | +0.029 |
best | +0.028 |
Neg logit contributions | |
ck | -0.021 |
snatch | -0.021 |
zip | -0.020 |
Flo | -0.020 |
zzy | -0.019 |
Token | amb |
Token ID | 4131 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
snatch | +0.121 |
ck | +0.109 |
zip | +0.106 |
unw | +0.098 |
Flo | +0.096 |
Neg logit contributions | |
End | -0.315 |
sky | -0.312 |
problem | -0.305 |
best | -0.294 |
happiest | -0.291 |
Token | i |
Token ID | 72 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
snatch | +0.523 |
zip | +0.519 |
Flo | +0.507 |
ck | +0.488 |
unw | +0.487 |
Neg logit contributions | |
End | -0.602 |
problem | -0.575 |
sky | -0.547 |
end | -0.523 |
best | -0.519 |
Token | could |
Token ID | 714 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
End | +0.167 |
problem | +0.160 |
sky | +0.152 |
end | +0.138 |
happiest | +0.136 |
Neg logit contributions | |
snatch | -0.294 |
zip | -0.294 |
ck | -0.292 |
Flo | -0.291 |
zzy | -0.281 |
Token | meet |
Token ID | 1826 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
End | +0.152 |
end | +0.151 |
sky | +0.144 |
problem | +0.143 |
best | +0.138 |
Neg logit contributions | |
zip | -0.176 |
Flo | -0.175 |
ck | -0.172 |
snatch | -0.170 |
igi | -0.166 |
Token | their |
Token ID | 511 |
Feature activation | -5.64e-02 |
Pos logit contributions | |
snatch | +0.022 |
zip | +0.022 |
ck | +0.021 |
Flo | +0.021 |
zzy | +0.020 |
Neg logit contributions | |
End | -0.019 |
problem | -0.019 |
sky | -0.018 |
end | -0.017 |
best | -0.017 |
Token | soon |
Token ID | 2582 |
Feature activation | -7.45e-02 |
Pos logit contributions | |
End | +0.069 |
problem | +0.063 |
sky | +0.061 |
gloomy | +0.055 |
end | +0.054 |
Neg logit contributions | |
snatch | -0.111 |
zip | -0.110 |
ck | -0.110 |
unw | -0.109 |
Flo | -0.108 |
Token | - |
Token ID | 12 |
Feature activation | +9.32e-02 |
Pos logit contributions | |
End | +0.109 |
problem | +0.098 |
sky | +0.094 |
happiest | +0.089 |
end | +0.087 |
Neg logit contributions | |
snatch | -0.134 |
ck | -0.133 |
unw | -0.131 |
zip | -0.131 |
Flo | -0.128 |
Token | to |
Token ID | 1462 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
snatch | +0.148 |
ck | +0.145 |
Flo | +0.144 |
zip | +0.144 |
zzy | +0.137 |
Neg logit contributions | |
End | -0.157 |
problem | -0.151 |
sky | -0.148 |
end | -0.138 |
happiest | -0.137 |
Token | - |
Token ID | 12 |
Feature activation | +9.35e-02 |
Pos logit contributions | |
zip | +0.543 |
snatch | +0.543 |
ck | +0.541 |
Flo | +0.533 |
zzy | +0.515 |
Neg logit contributions | |
End | -0.570 |
problem | -0.550 |
sky | -0.531 |
happiest | -0.494 |
end | -0.492 |
Token | be |
Token ID | 1350 |
Feature activation | +3.28e-02 |
Pos logit contributions | |
snatch | +0.148 |
ck | +0.145 |
Flo | +0.144 |
zip | +0.144 |
zzy | +0.137 |
Neg logit contributions | |
End | -0.159 |
problem | -0.152 |
sky | -0.149 |
end | -0.139 |
happiest | -0.138 |
Token | step |
Token ID | 2239 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
ck | +0.053 |
zip | +0.050 |
snatch | +0.049 |
zzy | +0.047 |
Flo | +0.046 |
Neg logit contributions | |
End | -0.054 |
sky | -0.053 |
problem | -0.051 |
happiest | -0.048 |
gloomy | -0.047 |
Token | dad |
Token ID | 9955 |
Feature activation | -6.40e-02 |
Pos logit contributions | |
End | +0.019 |
problem | +0.018 |
sky | +0.018 |
end | +0.016 |
best | +0.016 |
Neg logit contributions | |
snatch | -0.024 |
zip | -0.023 |
ck | -0.023 |
Flo | -0.023 |
zzy | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -3.59e-02 |
Pos logit contributions | |
End | +0.083 |
problem | +0.081 |
sky | +0.077 |
end | +0.073 |
best | +0.072 |
Neg logit contributions | |
zip | -0.124 |
Flo | -0.124 |
snatch | -0.123 |
ck | -0.122 |
zzy | -0.119 |
Token | B |
Token ID | 347 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
snatch | +0.180 |
ck | +0.173 |
har | +0.170 |
zip | +0.169 |
zzy | +0.168 |
Neg logit contributions | |
sky | -0.188 |
day | -0.177 |
best | -0.175 |
problem | -0.172 |
night | -0.171 |
Token | amb |
Token ID | 4131 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
snatch | +0.193 |
ck | +0.183 |
zip | +0.177 |
Flo | +0.169 |
unw | +0.168 |
Neg logit contributions | |
End | -0.247 |
sky | -0.243 |
problem | -0.236 |
end | -0.224 |
best | -0.223 |
Token | i |
Token ID | 72 |
Feature activation | +5.91e-02 |
Pos logit contributions | |
snatch | +0.465 |
zip | +0.465 |
ck | +0.464 |
Flo | +0.458 |
zzy | +0.436 |
Neg logit contributions | |
End | -0.651 |
problem | -0.634 |
sky | -0.615 |
end | -0.578 |
gloomy | -0.575 |
Token | . |
Token ID | 13 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
Flo | +0.110 |
ck | +0.109 |
zip | +0.107 |
uzz | +0.105 |
zzy | +0.104 |
Neg logit contributions | |
End | -0.085 |
problem | -0.085 |
sky | -0.076 |
happiest | -0.073 |
gloomy | -0.071 |
Token | B |
Token ID | 347 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
End | +0.032 |
problem | +0.031 |
sky | +0.031 |
gloomy | +0.029 |
best | +0.028 |
Neg logit contributions | |
ck | -0.021 |
snatch | -0.021 |
zip | -0.020 |
Flo | -0.020 |
zzy | -0.019 |
Token | amb |
Token ID | 4131 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
snatch | +0.121 |
ck | +0.109 |
zip | +0.106 |
unw | +0.098 |
Flo | +0.096 |
Neg logit contributions | |
End | -0.315 |
sky | -0.312 |
problem | -0.305 |
best | -0.294 |
happiest | -0.291 |
Token | i |
Token ID | 72 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
snatch | +0.523 |
zip | +0.519 |
Flo | +0.507 |
ck | +0.488 |
unw | +0.487 |
Neg logit contributions | |
End | -0.602 |
problem | -0.575 |
sky | -0.547 |
end | -0.523 |
best | -0.519 |
Token | loved |
Token ID | 6151 |
Feature activation | -7.77e-02 |
Pos logit contributions | |
Flo | +0.127 |
zip | +0.125 |
ck | +0.124 |
zzy | +0.120 |
uzz | +0.120 |
Neg logit contributions | |
End | -0.066 |
problem | -0.065 |
sky | -0.054 |
happiest | -0.053 |
end | -0.052 |
Token | to |
Token ID | 284 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
End | +0.140 |
problem | +0.133 |
sky | +0.127 |
end | +0.121 |
happiest | +0.121 |
Neg logit contributions | |
zip | -0.117 |
snatch | -0.116 |
Flo | -0.115 |
ck | -0.115 |
unw | -0.110 |
Token | play |
Token ID | 711 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
zip | +0.256 |
snatch | +0.253 |
ck | +0.252 |
Flo | +0.251 |
har | +0.249 |
Neg logit contributions | |
sky | -0.235 |
End | -0.232 |
end | -0.227 |
problem | -0.227 |
day | -0.219 |
Token | and |
Token ID | 290 |
Feature activation | +9.20e-02 |
Pos logit contributions | |
snatch | +0.040 |
onna | +0.039 |
har | +0.039 |
ge | +0.038 |
uzz | +0.038 |
Neg logit contributions | |
sky | -0.026 |
best | -0.025 |
end | -0.024 |
day | -0.024 |
days | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
zip | +0.184 |
Flo | +0.182 |
snatch | +0.181 |
ck | +0.177 |
zzy | +0.174 |
Neg logit contributions | |
End | -0.110 |
sky | -0.104 |
problem | -0.101 |
end | -0.097 |
day | -0.097 |
Token | kids |
Token ID | 3988 |
Feature activation | -5.65e-02 |
Pos logit contributions | |
End | +0.094 |
problem | +0.085 |
sky | +0.079 |
happiest | +0.073 |
end | +0.069 |
Neg logit contributions | |
snatch | -0.239 |
ck | -0.236 |
zip | -0.236 |
unw | -0.232 |
Flo | -0.231 |
Token | had |
Token ID | 550 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
End | +0.089 |
problem | +0.084 |
sky | +0.080 |
end | +0.075 |
gloomy | +0.074 |
Neg logit contributions | |
ck | -0.097 |
zip | -0.097 |
snatch | -0.096 |
Flo | -0.094 |
unw | -0.093 |
Token | their |
Token ID | 511 |
Feature activation | -5.80e-02 |
Pos logit contributions | |
End | +0.223 |
problem | +0.214 |
sky | +0.204 |
end | +0.189 |
happiest | +0.187 |
Neg logit contributions | |
zip | -0.286 |
snatch | -0.286 |
ck | -0.284 |
Flo | -0.282 |
zzy | -0.272 |
Token | t |
Token ID | 256 |
Feature activation | +2.64e-02 |
Pos logit contributions | |
End | +0.074 |
sky | +0.067 |
problem | +0.065 |
end | +0.059 |
gloomy | +0.058 |
Neg logit contributions | |
snatch | -0.111 |
ck | -0.110 |
unw | -0.109 |
zip | -0.108 |
Flo | -0.106 |
Token | amb |
Token ID | 4131 |
Feature activation | +3.40e-01 |
Pos logit contributions | |
snatch | +0.039 |
Flo | +0.038 |
ck | +0.037 |
zip | +0.036 |
zzy | +0.036 |
Neg logit contributions | |
End | -0.047 |
problem | -0.047 |
sky | -0.045 |
best | -0.043 |
end | -0.042 |
Token | our |
Token ID | 454 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
ck | +0.472 |
snatch | +0.440 |
zip | +0.439 |
Flo | +0.414 |
steer | +0.411 |
Neg logit contributions | |
End | -0.653 |
sky | -0.646 |
problem | -0.644 |
gloomy | -0.615 |
end | -0.587 |
Token | ines |
Token ID | 1127 |
Feature activation | -6.53e-02 |
Pos logit contributions | |
zip | +0.260 |
ck | +0.260 |
snatch | +0.259 |
Flo | +0.257 |
zzy | +0.250 |
Neg logit contributions | |
End | -0.234 |
problem | -0.223 |
sky | -0.211 |
happiest | -0.198 |
end | -0.197 |
Token | and |
Token ID | 290 |
Feature activation | +7.42e-02 |
Pos logit contributions | |
End | +0.093 |
problem | +0.089 |
sky | +0.081 |
happiest | +0.076 |
end | +0.075 |
Neg logit contributions | |
zip | -0.119 |
snatch | -0.118 |
Flo | -0.117 |
ck | -0.116 |
unw | -0.116 |
Token | drums |
Token ID | 24730 |
Feature activation | -6.83e-02 |
Pos logit contributions | |
snatch | +0.134 |
zip | +0.133 |
ck | +0.132 |
Flo | +0.130 |
unw | +0.128 |
Neg logit contributions | |
End | -0.109 |
problem | -0.104 |
sky | -0.098 |
end | -0.092 |
gloomy | -0.089 |
Token | , |
Token ID | 11 |
Feature activation | -8.81e-02 |
Pos logit contributions | |
End | +0.102 |
problem | +0.094 |
sky | +0.086 |
gloomy | +0.080 |
happiest | +0.079 |
Neg logit contributions | |
zip | -0.124 |
snatch | -0.123 |
Flo | -0.121 |
ck | -0.120 |
unw | -0.119 |
Token | stand |
Token ID | 1302 |
Feature activation | +4.38e-02 |
Pos logit contributions | |
Flo | +0.263 |
zip | +0.254 |
uzz | +0.247 |
ck | +0.243 |
snatch | +0.243 |
Neg logit contributions | |
end | -0.228 |
End | -0.223 |
problem | -0.207 |
sky | -0.206 |
best | -0.197 |
Token | on |
Token ID | 319 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
snatch | +0.071 |
zip | +0.071 |
ck | +0.070 |
Flo | +0.070 |
zzy | +0.067 |
Neg logit contributions | |
End | -0.072 |
problem | -0.070 |
sky | -0.067 |
end | -0.063 |
happiest | -0.062 |
Token | her |
Token ID | 607 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
zip | +0.201 |
snatch | +0.201 |
Flo | +0.198 |
ck | +0.197 |
unw | +0.196 |
Neg logit contributions | |
End | -0.203 |
problem | -0.197 |
sky | -0.181 |
happiest | -0.177 |
best | -0.170 |
Token | tip |
Token ID | 8171 |
Feature activation | +4.45e-02 |
Pos logit contributions | |
End | +0.037 |
problem | +0.034 |
sky | +0.032 |
end | +0.030 |
gloomy | +0.029 |
Neg logit contributions | |
unw | -0.039 |
Flo | -0.038 |
zip | -0.037 |
snatch | -0.037 |
ck | -0.036 |
Token | - |
Token ID | 12 |
Feature activation | +9.36e-02 |
Pos logit contributions | |
zip | +0.079 |
Flo | +0.077 |
snatch | +0.077 |
ck | +0.076 |
unw | +0.076 |
Neg logit contributions | |
End | -0.068 |
problem | -0.063 |
sky | -0.056 |
happiest | -0.056 |
end | -0.055 |
Token | to |
Token ID | 1462 |
Feature activation | +3.40e-01 |
Pos logit contributions | |
snatch | +0.092 |
ck | +0.090 |
zip | +0.090 |
Flo | +0.089 |
zzy | +0.082 |
Neg logit contributions | |
End | -0.214 |
problem | -0.208 |
sky | -0.204 |
end | -0.194 |
happiest | -0.193 |
Token | es |
Token ID | 274 |
Feature activation | +5.63e-02 |
Pos logit contributions | |
zip | +0.556 |
ck | +0.555 |
Flo | +0.544 |
snatch | +0.541 |
zzy | +0.523 |
Neg logit contributions | |
End | -0.572 |
problem | -0.550 |
sky | -0.529 |
happiest | -0.499 |
end | -0.487 |
Token | to |
Token ID | 284 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
snatch | +0.116 |
ck | +0.115 |
zip | +0.115 |
Flo | +0.115 |
zzy | +0.109 |
Neg logit contributions | |
End | -0.067 |
problem | -0.065 |
sky | -0.063 |
end | -0.058 |
best | -0.056 |
Token | reach |
Token ID | 3151 |
Feature activation | +5.28e-02 |
Pos logit contributions | |
Flo | +0.238 |
zip | +0.227 |
uzz | +0.226 |
ck | +0.225 |
snatch | +0.223 |
Neg logit contributions | |
end | -0.244 |
End | -0.241 |
sky | -0.235 |
problem | -0.223 |
day | -0.217 |
Token | the |
Token ID | 262 |
Feature activation | -9.75e-02 |
Pos logit contributions | |
snatch | +0.105 |
zip | +0.105 |
ck | +0.104 |
Flo | +0.103 |
zzy | +0.101 |
Neg logit contributions | |
End | -0.067 |
problem | -0.066 |
sky | -0.064 |
end | -0.060 |
best | -0.058 |
Token | top |
Token ID | 1353 |
Feature activation | -7.04e-02 |
Pos logit contributions | |
End | +0.115 |
problem | +0.105 |
sky | +0.092 |
happiest | +0.087 |
gloomy | +0.084 |
Neg logit contributions | |
zip | -0.200 |
snatch | -0.200 |
Flo | -0.200 |
unw | -0.196 |
ck | -0.192 |
Token | he |
Token ID | 339 |
Feature activation | -8.89e-02 |
Pos logit contributions | |
problem | +0.169 |
End | +0.167 |
sky | +0.150 |
happiest | +0.144 |
end | +0.140 |
Neg logit contributions | |
snatch | -0.130 |
steer | -0.129 |
ck | -0.124 |
Flo | -0.123 |
zip | -0.123 |
Token | smiled |
Token ID | 13541 |
Feature activation | -4.62e-02 |
Pos logit contributions | |
sky | +0.118 |
problem | +0.117 |
End | +0.116 |
gloomy | +0.107 |
happiest | +0.105 |
Neg logit contributions | |
snatch | -0.156 |
ck | -0.155 |
zip | -0.154 |
uzz | -0.154 |
Flo | -0.153 |
Token | from |
Token ID | 422 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
End | +0.076 |
problem | +0.071 |
sky | +0.066 |
end | +0.061 |
happiest | +0.061 |
Neg logit contributions | |
snatch | -0.074 |
zip | -0.074 |
Flo | -0.074 |
ck | -0.073 |
zzy | -0.073 |
Token | ear |
Token ID | 1027 |
Feature activation | -3.41e-02 |
Pos logit contributions | |
zip | +0.307 |
Flo | +0.306 |
zzy | +0.297 |
unw | +0.291 |
ck | +0.289 |
Neg logit contributions | |
End | -0.236 |
problem | -0.209 |
end | -0.204 |
sky | -0.203 |
gloomy | -0.192 |
Token | - |
Token ID | 12 |
Feature activation | +9.26e-02 |
Pos logit contributions | |
End | +0.052 |
problem | +0.050 |
sky | +0.048 |
end | +0.045 |
happiest | +0.044 |
Neg logit contributions | |
snatch | -0.059 |
zip | -0.059 |
ck | -0.059 |
Flo | -0.059 |
zzy | -0.056 |
Token | to |
Token ID | 1462 |
Feature activation | +3.39e-01 |
Pos logit contributions | |
snatch | +0.151 |
ck | +0.147 |
Flo | +0.146 |
zip | +0.144 |
unw | +0.144 |
Neg logit contributions | |
End | -0.150 |
sky | -0.138 |
problem | -0.132 |
end | -0.129 |
day | -0.128 |
Token | - |
Token ID | 12 |
Feature activation | +9.25e-02 |
Pos logit contributions | |
Flo | +0.708 |
snatch | +0.706 |
zip | +0.705 |
ck | +0.685 |
unw | +0.680 |
Neg logit contributions | |
End | -0.385 |
problem | -0.371 |
sky | -0.337 |
end | -0.319 |
best | -0.314 |
Token | ear |
Token ID | 451 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
snatch | +0.141 |
ck | +0.136 |
Flo | +0.135 |
unw | +0.134 |
zip | +0.133 |
Neg logit contributions | |
End | -0.160 |
sky | -0.148 |
problem | -0.141 |
end | -0.139 |
day | -0.138 |
Token | . |
Token ID | 13 |
Feature activation | -3.77e-02 |
Pos logit contributions | |
zip | +0.230 |
snatch | +0.230 |
Flo | +0.227 |
ck | +0.226 |
zzy | +0.219 |
Neg logit contributions | |
End | -0.152 |
problem | -0.144 |
sky | -0.142 |
end | -0.129 |
best | -0.127 |
Token | He |
Token ID | 679 |
Feature activation | +9.28e-02 |
Pos logit contributions | |
End | +0.060 |
sky | +0.058 |
problem | +0.057 |
best | +0.051 |
end | +0.051 |
Neg logit contributions | |
ck | -0.065 |
snatch | -0.064 |
zip | -0.063 |
Flo | -0.062 |
zzy | -0.060 |
Token | grabbed |
Token ID | 13176 |
Feature activation | +8.31e-02 |
Pos logit contributions | |
zip | +0.165 |
snatch | +0.165 |
Flo | +0.162 |
ck | +0.161 |
uzz | +0.159 |
Neg logit contributions | |
End | -0.127 |
sky | -0.127 |
problem | -0.126 |
happiest | -0.111 |
starts | -0.111 |
Token | investigate |
Token ID | 9161 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
igi | +0.262 |
Flo | +0.260 |
onna | +0.253 |
ck | +0.248 |
rily | +0.245 |
Neg logit contributions | |
end | -0.209 |
sky | -0.194 |
problem | -0.190 |
day | -0.178 |
night | -0.176 |
Token | . |
Token ID | 13 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
ck | +0.346 |
snatch | +0.340 |
zip | +0.333 |
Flo | +0.332 |
unw | +0.328 |
Neg logit contributions | |
End | -0.305 |
problem | -0.282 |
sky | -0.269 |
happiest | -0.261 |
best | -0.248 |
Token | She |
Token ID | 1375 |
Feature activation | +5.13e-02 |
Pos logit contributions | |
End | +0.042 |
sky | +0.040 |
problem | +0.039 |
gloomy | +0.035 |
end | +0.035 |
Neg logit contributions | |
ck | -0.059 |
snatch | -0.058 |
zip | -0.058 |
Flo | -0.057 |
zzy | -0.055 |
Token | tip |
Token ID | 8171 |
Feature activation | +4.65e-02 |
Pos logit contributions | |
Flo | +0.096 |
snatch | +0.095 |
zip | +0.093 |
ck | +0.091 |
izer | +0.090 |
Neg logit contributions | |
problem | -0.071 |
End | -0.069 |
sky | -0.068 |
end | -0.060 |
best | -0.060 |
Token | - |
Token ID | 12 |
Feature activation | +9.27e-02 |
Pos logit contributions | |
snatch | +0.088 |
zip | +0.086 |
Flo | +0.086 |
ck | +0.086 |
zzy | +0.082 |
Neg logit contributions | |
End | -0.064 |
problem | -0.062 |
sky | -0.060 |
end | -0.055 |
best | -0.055 |
Token | to |
Token ID | 1462 |
Feature activation | +3.39e-01 |
Pos logit contributions | |
snatch | +0.110 |
Flo | +0.105 |
ck | +0.104 |
zip | +0.102 |
unw | +0.099 |
Neg logit contributions | |
End | -0.192 |
sky | -0.185 |
problem | -0.184 |
end | -0.172 |
best | -0.169 |
Token | ed |
Token ID | 276 |
Feature activation | +9.64e-02 |
Pos logit contributions | |
snatch | +0.640 |
Flo | +0.631 |
zip | +0.625 |
ck | +0.614 |
unw | +0.607 |
Neg logit contributions | |
End | -0.457 |
problem | -0.443 |
sky | -0.419 |
best | -0.394 |
end | -0.393 |
Token | closer |
Token ID | 5699 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
Flo | +0.180 |
snatch | +0.178 |
ck | +0.176 |
zip | +0.176 |
onna | +0.173 |
Neg logit contributions | |
End | -0.129 |
sky | -0.129 |
problem | -0.127 |
best | -0.120 |
end | -0.120 |
Token | to |
Token ID | 284 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
snatch | +0.183 |
zip | +0.183 |
ck | +0.182 |
Flo | +0.180 |
zzy | +0.174 |
Neg logit contributions | |
End | -0.184 |
problem | -0.177 |
sky | -0.169 |
end | -0.159 |
happiest | -0.159 |
Token | the |
Token ID | 262 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
Flo | +0.302 |
zip | +0.289 |
zzy | +0.281 |
onna | +0.278 |
ck | +0.275 |
Neg logit contributions | |
end | -0.183 |
End | -0.179 |
sky | -0.171 |
problem | -0.160 |
happiest | -0.155 |
Token | jail |
Token ID | 7356 |
Feature activation | +5.75e-02 |
Pos logit contributions | |
End | +0.182 |
problem | +0.169 |
sky | +0.160 |
happiest | +0.160 |
best | +0.156 |
Neg logit contributions | |
snatch | -0.155 |
zip | -0.147 |
Flo | -0.142 |
unw | -0.141 |
ck | -0.138 |
Token | . |
Token ID | 13 |
Feature activation | -3.83e-02 |
Pos logit contributions | |
End | +0.030 |
problem | +0.029 |
sky | +0.026 |
best | +0.025 |
end | +0.024 |
Neg logit contributions | |
zip | -0.082 |
Flo | -0.081 |
snatch | -0.080 |
ck | -0.079 |
zzy | -0.077 |
Token |
|
Token ID | 198 |
Feature activation | +6.82e-02 |
Pos logit contributions | |
sky | +0.051 |
End | +0.048 |
problem | +0.045 |
best | +0.041 |
starts | +0.040 |
Neg logit contributions | |
ck | -0.079 |
zip | -0.078 |
snatch | -0.077 |
ge | -0.077 |
steer | -0.073 |
Token |
|
Token ID | 198 |
Feature activation | +6.71e-02 |
Pos logit contributions | |
snatch | +0.169 |
ge | +0.158 |
ck | +0.155 |
zip | +0.155 |
shave | +0.154 |
Neg logit contributions | |
End | -0.063 |
sky | -0.062 |
end | -0.054 |
day | -0.051 |
problem | -0.046 |
Token | " |
Token ID | 1 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
snatch | +0.098 |
zip | +0.095 |
ck | +0.092 |
unw | +0.091 |
shave | +0.090 |
Neg logit contributions | |
End | -0.129 |
end | -0.116 |
sky | -0.115 |
problem | -0.114 |
best | -0.105 |
Token | L |
Token ID | 43 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
End | +0.190 |
sky | +0.178 |
gloomy | +0.158 |
happiest | +0.158 |
end | +0.155 |
Neg logit contributions | |
ck | -0.183 |
snatch | -0.182 |
zip | -0.177 |
ge | -0.173 |
unw | -0.172 |
Token | amb |
Token ID | 4131 |
Feature activation | +3.39e-01 |
Pos logit contributions | |
zip | +0.349 |
snatch | +0.343 |
unw | +0.328 |
Flo | +0.320 |
steer | +0.317 |
Neg logit contributions | |
End | -0.272 |
problem | -0.250 |
end | -0.247 |
sky | -0.232 |
best | -0.222 |
Token | ? |
Token ID | 30 |
Feature activation | +3.06e-02 |
Pos logit contributions | |
zip | +0.498 |
snatch | +0.498 |
Flo | +0.495 |
ck | +0.490 |
zzy | +0.469 |
Neg logit contributions | |
End | -0.614 |
problem | -0.596 |
sky | -0.570 |
end | -0.541 |
happiest | -0.534 |
Token | Like |
Token ID | 4525 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
ck | +0.067 |
snatch | +0.064 |
zip | +0.063 |
zzy | +0.063 |
Flo | +0.061 |
Neg logit contributions | |
End | -0.035 |
sky | -0.034 |
problem | -0.031 |
gloomy | -0.028 |
happiest | -0.028 |
Token | sheep |
Token ID | 15900 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
zip | +0.254 |
snatch | +0.253 |
ck | +0.251 |
Flo | +0.250 |
zzy | +0.243 |
Neg logit contributions | |
End | -0.183 |
problem | -0.177 |
sky | -0.168 |
end | -0.155 |
best | -0.153 |
Token | ?" |
Token ID | 1701 |
Feature activation | +3.97e-02 |
Pos logit contributions | |
Flo | +0.024 |
snatch | +0.024 |
ck | +0.023 |
zip | +0.023 |
zzy | +0.023 |
Neg logit contributions | |
End | -0.017 |
problem | -0.015 |
sky | -0.015 |
happiest | -0.014 |
end | -0.013 |
Token | Mia |
Token ID | 32189 |
Feature activation | +8.98e-02 |
Pos logit contributions | |
zip | +0.062 |
snatch | +0.060 |
ck | +0.058 |
izer | +0.058 |
Flo | +0.058 |
Neg logit contributions | |
End | -0.071 |
sky | -0.069 |
problem | -0.068 |
end | -0.064 |
best | -0.063 |
Token | two |
Token ID | 734 |
Feature activation | +5.46e-03 |
Pos logit contributions | |
snatch | +0.211 |
ck | +0.201 |
zip | +0.196 |
Flo | +0.194 |
onna | +0.193 |
Neg logit contributions | |
sky | -0.172 |
End | -0.171 |
best | -0.163 |
problem | -0.160 |
end | -0.155 |
Token | of |
Token ID | 286 |
Feature activation | +9.24e-02 |
Pos logit contributions | |
unw | +0.008 |
Flo | +0.008 |
zip | +0.008 |
ck | +0.008 |
snatch | +0.007 |
Neg logit contributions | |
End | -0.010 |
problem | -0.009 |
sky | -0.009 |
ending | -0.008 |
happiest | -0.008 |
Token | them |
Token ID | 606 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
snatch | +0.170 |
zip | +0.169 |
ck | +0.167 |
Flo | +0.166 |
zzy | +0.164 |
Neg logit contributions | |
End | -0.131 |
problem | -0.127 |
sky | -0.124 |
end | -0.115 |
best | -0.113 |
Token | back |
Token ID | 736 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
snatch | +0.069 |
Flo | +0.067 |
zip | +0.067 |
ck | +0.067 |
zzy | +0.064 |
Neg logit contributions | |
End | -0.060 |
problem | -0.059 |
sky | -0.057 |
best | -0.054 |
end | -0.053 |
Token | - |
Token ID | 12 |
Feature activation | +9.32e-02 |
Pos logit contributions | |
zip | +0.050 |
ck | +0.049 |
snatch | +0.049 |
Flo | +0.049 |
unw | +0.047 |
Neg logit contributions | |
End | -0.055 |
problem | -0.052 |
sky | -0.050 |
happiest | -0.047 |
end | -0.047 |
Token | to |
Token ID | 1462 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
snatch | +0.141 |
ck | +0.137 |
zip | +0.136 |
Flo | +0.136 |
unw | +0.130 |
Neg logit contributions | |
End | -0.165 |
problem | -0.156 |
sky | -0.154 |
end | -0.145 |
best | -0.142 |
Token | - |
Token ID | 12 |
Feature activation | +9.36e-02 |
Pos logit contributions | |
snatch | +0.574 |
zip | +0.573 |
ck | +0.568 |
Flo | +0.567 |
zzy | +0.544 |
Neg logit contributions | |
End | -0.531 |
problem | -0.513 |
sky | -0.492 |
end | -0.458 |
happiest | -0.454 |
Token | back |
Token ID | 1891 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
snatch | +0.142 |
ck | +0.137 |
Flo | +0.136 |
zip | +0.136 |
unw | +0.130 |
Neg logit contributions | |
End | -0.165 |
problem | -0.157 |
sky | -0.155 |
end | -0.145 |
best | -0.142 |
Token | . |
Token ID | 13 |
Feature activation | -4.52e-02 |
Pos logit contributions | |
End | +0.031 |
problem | +0.030 |
sky | +0.028 |
end | +0.026 |
happiest | +0.026 |
Neg logit contributions | |
zip | -0.040 |
snatch | -0.040 |
Flo | -0.039 |
ck | -0.039 |
zzy | -0.038 |
Token | She |
Token ID | 1375 |
Feature activation | +4.17e-02 |
Pos logit contributions | |
End | +0.067 |
sky | +0.064 |
problem | +0.064 |
end | +0.057 |
best | +0.057 |
Neg logit contributions | |
ck | -0.082 |
snatch | -0.082 |
zip | -0.080 |
Flo | -0.079 |
zzy | -0.076 |
Token | draped |
Token ID | 38425 |
Feature activation | +6.07e-02 |
Pos logit contributions | |
snatch | +0.076 |
zip | +0.076 |
Flo | +0.075 |
ck | +0.075 |
zzy | +0.072 |
Neg logit contributions | |
End | -0.060 |
problem | -0.058 |
sky | -0.056 |
end | -0.051 |
best | -0.051 |
Token | wolf |
Token ID | 17481 |
Feature activation | +5.39e-02 |
Pos logit contributions | |
End | +0.168 |
problem | +0.149 |
happiest | +0.139 |
sky | +0.138 |
best | +0.135 |
Neg logit contributions | |
Flo | -0.164 |
snatch | -0.161 |
zip | -0.153 |
ck | -0.150 |
unw | -0.148 |
Token | . |
Token ID | 13 |
Feature activation | -3.48e-02 |
Pos logit contributions | |
snatch | +0.111 |
zip | +0.111 |
ck | +0.110 |
Flo | +0.110 |
zzy | +0.106 |
Neg logit contributions | |
End | -0.065 |
problem | -0.062 |
sky | -0.059 |
end | -0.053 |
happiest | -0.053 |
Token | She |
Token ID | 1375 |
Feature activation | +4.66e-02 |
Pos logit contributions | |
End | +0.049 |
problem | +0.047 |
sky | +0.046 |
end | +0.041 |
happiest | +0.041 |
Neg logit contributions | |
ck | -0.065 |
snatch | -0.065 |
zip | -0.064 |
Flo | -0.064 |
zzy | -0.062 |
Token | tip |
Token ID | 8171 |
Feature activation | +4.31e-02 |
Pos logit contributions | |
zip | +0.084 |
snatch | +0.084 |
Flo | +0.083 |
izer | +0.081 |
uzz | +0.079 |
Neg logit contributions | |
End | -0.063 |
problem | -0.062 |
sky | -0.061 |
became | -0.057 |
end | -0.056 |
Token | - |
Token ID | 12 |
Feature activation | +9.23e-02 |
Pos logit contributions | |
snatch | +0.088 |
zip | +0.086 |
Flo | +0.085 |
ck | +0.084 |
zzy | +0.084 |
Neg logit contributions | |
End | -0.051 |
sky | -0.049 |
problem | -0.048 |
best | -0.044 |
end | -0.043 |
Token | to |
Token ID | 1462 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
snatch | +0.155 |
ck | +0.149 |
Flo | +0.147 |
zip | +0.147 |
unw | +0.146 |
Neg logit contributions | |
End | -0.143 |
sky | -0.136 |
problem | -0.131 |
end | -0.122 |
day | -0.119 |
Token | ed |
Token ID | 276 |
Feature activation | +9.43e-02 |
Pos logit contributions | |
snatch | +0.705 |
Flo | +0.688 |
zip | +0.687 |
unw | +0.673 |
zzy | +0.671 |
Neg logit contributions | |
End | -0.377 |
problem | -0.354 |
sky | -0.331 |
days | -0.323 |
best | -0.319 |
Token | through |
Token ID | 832 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
onna | +0.188 |
zzy | +0.187 |
zip | +0.186 |
snatch | +0.186 |
Twe | +0.186 |
Neg logit contributions | |
sky | -0.113 |
End | -0.108 |
days | -0.103 |
problem | -0.103 |
night | -0.103 |
Token | the |
Token ID | 262 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
zip | +0.212 |
ck | +0.207 |
snatch | +0.206 |
Flo | +0.205 |
zzy | +0.204 |
Neg logit contributions | |
End | -0.140 |
sky | -0.136 |
problem | -0.135 |
end | -0.122 |
gloomy | -0.119 |
Token | trees |
Token ID | 7150 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
End | +0.134 |
happiest | +0.108 |
problem | +0.108 |
best | +0.102 |
sky | +0.096 |
Neg logit contributions | |
Flo | -0.199 |
snatch | -0.198 |
zip | -0.191 |
ck | -0.185 |
unw | -0.184 |
Token | and |
Token ID | 290 |
Feature activation | +7.59e-02 |
Pos logit contributions | |
End | +0.227 |
problem | +0.219 |
sky | +0.200 |
happiest | +0.190 |
end | +0.190 |
Neg logit contributions | |
snatch | -0.244 |
ck | -0.243 |
zip | -0.242 |
Flo | -0.240 |
unw | -0.237 |
Token | in |
Token ID | 287 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
snatch | +0.074 |
Flo | +0.074 |
ck | +0.073 |
zzy | +0.071 |
zip | +0.070 |
Neg logit contributions | |
problem | -0.052 |
End | -0.052 |
sky | -0.051 |
gloomy | -0.049 |
best | -0.049 |
Token | a |
Token ID | 257 |
Feature activation | +3.15e-03 |
Pos logit contributions | |
ck | +0.418 |
zip | +0.415 |
Flo | +0.410 |
zzy | +0.410 |
snatch | +0.404 |
Neg logit contributions | |
problem | -0.222 |
End | -0.218 |
sky | -0.217 |
end | -0.201 |
gloomy | -0.195 |
Token | trap |
Token ID | 12840 |
Feature activation | -2.69e-03 |
Pos logit contributions | |
Flo | +0.005 |
snatch | +0.005 |
zip | +0.005 |
unw | +0.005 |
ck | +0.005 |
Neg logit contributions | |
End | -0.005 |
happiest | -0.004 |
problem | -0.004 |
sky | -0.004 |
ending | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
End | +0.003 |
problem | +0.003 |
sky | +0.003 |
end | +0.003 |
gloomy | +0.002 |
Neg logit contributions | |
zip | -0.006 |
snatch | -0.006 |
ck | -0.006 |
Flo | -0.006 |
zzy | -0.005 |
Token | B |
Token ID | 347 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
sky | +0.031 |
End | +0.031 |
problem | +0.031 |
gloomy | +0.028 |
best | +0.027 |
Neg logit contributions | |
ck | -0.041 |
snatch | -0.040 |
zip | -0.039 |
zzy | -0.039 |
Flo | -0.038 |
Token | amb |
Token ID | 4131 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
snatch | +0.128 |
zip | +0.115 |
ck | +0.112 |
unw | +0.107 |
Flo | +0.101 |
Neg logit contributions | |
End | -0.286 |
sky | -0.281 |
problem | -0.279 |
best | -0.268 |
happiest | -0.266 |
Token | i |
Token ID | 72 |
Feature activation | +5.20e-02 |
Pos logit contributions | |
zip | +0.565 |
snatch | +0.562 |
Flo | +0.558 |
onna | +0.540 |
unw | +0.539 |
Neg logit contributions | |
End | -0.544 |
problem | -0.519 |
sky | -0.459 |
end | -0.459 |
happiest | -0.453 |
Token | tried |
Token ID | 3088 |
Feature activation | +6.09e-04 |
Pos logit contributions | |
ck | +0.099 |
snatch | +0.099 |
Flo | +0.099 |
zip | +0.099 |
zzy | +0.094 |
Neg logit contributions | |
End | -0.071 |
problem | -0.067 |
sky | -0.064 |
end | -0.060 |
happiest | -0.060 |
Token | to |
Token ID | 284 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
ck | +0.001 |
snatch | +0.001 |
zip | +0.001 |
Flo | +0.001 |
zzy | +0.001 |
Neg logit contributions | |
problem | -0.001 |
End | -0.001 |
sky | -0.001 |
best | -0.001 |
end | -0.001 |
Token | help |
Token ID | 1037 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
uzz | +0.351 |
uggets | +0.332 |
rily | +0.312 |
igi | +0.311 |
rost | +0.310 |
Neg logit contributions | |
end | -0.210 |
problem | -0.164 |
very | -0.163 |
End | -0.146 |
day | -0.143 |
Token | Th |
Token ID | 536 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
zzy | +0.058 |
ck | +0.054 |
zip | +0.054 |
snatch | +0.054 |
onna | +0.053 |
Neg logit contributions | |
problem | -0.039 |
end | -0.039 |
End | -0.038 |
sky | -0.036 |
best | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
zip | +0.012 |
Flo | +0.011 |
snatch | +0.011 |
ck | +0.011 |
unw | +0.011 |
Neg logit contributions | |
End | -0.009 |
problem | -0.009 |
sky | -0.009 |
happiest | -0.008 |
gloomy | -0.008 |
Token | One |
Token ID | 1881 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
End | +0.026 |
sky | +0.025 |
problem | +0.025 |
best | +0.023 |
end | +0.023 |
Neg logit contributions | |
ck | -0.034 |
snatch | -0.034 |
zip | -0.033 |
Flo | -0.032 |
zzy | -0.032 |
Token | day |
Token ID | 1110 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
Flo | +0.054 |
ck | +0.054 |
zzy | +0.054 |
zip | +0.053 |
ifi | +0.052 |
Neg logit contributions | |
End | -0.014 |
problem | -0.014 |
sky | -0.011 |
happiest | -0.010 |
goals | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | -7.64e-02 |
Pos logit contributions | |
End | +0.272 |
problem | +0.248 |
sky | +0.234 |
happiest | +0.229 |
end | +0.206 |
Neg logit contributions | |
Flo | -0.536 |
zip | -0.524 |
ck | -0.512 |
snatch | -0.508 |
unw | -0.499 |
Token | B |
Token ID | 347 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
End | +0.133 |
problem | +0.129 |
sky | +0.126 |
gloomy | +0.118 |
best | +0.117 |
Neg logit contributions | |
snatch | -0.118 |
ck | -0.116 |
zip | -0.116 |
Flo | -0.114 |
zzy | -0.111 |
Token | amb |
Token ID | 4131 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
snatch | +0.121 |
zip | +0.109 |
ck | +0.108 |
unw | +0.099 |
Flo | +0.097 |
Neg logit contributions | |
End | -0.301 |
sky | -0.297 |
problem | -0.293 |
best | -0.280 |
happiest | -0.277 |
Token | i |
Token ID | 72 |
Feature activation | +5.27e-02 |
Pos logit contributions | |
zip | +0.516 |
snatch | +0.516 |
Flo | +0.502 |
unw | +0.485 |
zzy | +0.485 |
Neg logit contributions | |
End | -0.598 |
problem | -0.573 |
sky | -0.539 |
end | -0.520 |
best | -0.516 |
Token | saw |
Token ID | 2497 |
Feature activation | -4.73e-03 |
Pos logit contributions | |
zip | +0.100 |
ck | +0.100 |
Flo | +0.100 |
snatch | +0.099 |
zzy | +0.096 |
Neg logit contributions | |
End | -0.072 |
problem | -0.068 |
sky | -0.065 |
end | -0.060 |
happiest | -0.060 |
Token | his |
Token ID | 465 |
Feature activation | +4.28e-02 |
Pos logit contributions | |
End | +0.008 |
problem | +0.008 |
sky | +0.007 |
happiest | +0.007 |
end | +0.007 |
Neg logit contributions | |
Flo | -0.008 |
ck | -0.008 |
zip | -0.008 |
snatch | -0.008 |
zzy | -0.007 |
Token | friend |
Token ID | 1545 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
zip | +0.084 |
snatch | +0.084 |
ck | +0.083 |
Flo | +0.080 |
zzy | +0.079 |
Neg logit contributions | |
problem | -0.061 |
best | -0.059 |
sky | -0.057 |
End | -0.056 |
happiest | -0.052 |
Token | Th |
Token ID | 536 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
zip | +0.322 |
snatch | +0.316 |
ck | +0.315 |
Flo | +0.307 |
zzy | +0.305 |
Neg logit contributions | |
problem | -0.159 |
best | -0.155 |
End | -0.155 |
sky | -0.152 |
end | -0.147 |
Token | started |
Token ID | 2067 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
End | +0.177 |
problem | +0.171 |
sky | +0.164 |
end | +0.149 |
happiest | +0.148 |
Neg logit contributions | |
snatch | -0.259 |
zip | -0.259 |
Flo | -0.258 |
ck | -0.257 |
zzy | -0.248 |
Token | hopping |
Token ID | 47153 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
End | +0.228 |
problem | +0.215 |
sky | +0.208 |
happiest | +0.197 |
end | +0.197 |
Neg logit contributions | |
snatch | -0.212 |
ck | -0.208 |
zip | -0.208 |
Flo | -0.202 |
unw | -0.201 |
Token | from |
Token ID | 422 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
zip | +0.035 |
snatch | +0.035 |
ck | +0.034 |
Flo | +0.034 |
zzy | +0.033 |
Neg logit contributions | |
End | -0.027 |
problem | -0.026 |
sky | -0.025 |
end | -0.023 |
best | -0.023 |
Token | foot |
Token ID | 2366 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
zip | +0.281 |
snatch | +0.281 |
ck | +0.279 |
Flo | +0.279 |
zzy | +0.268 |
Neg logit contributions | |
End | -0.231 |
problem | -0.223 |
sky | -0.214 |
end | -0.198 |
happiest | -0.195 |
Token | - |
Token ID | 12 |
Feature activation | +9.20e-02 |
Pos logit contributions | |
End | +0.048 |
problem | +0.046 |
sky | +0.043 |
happiest | +0.041 |
end | +0.040 |
Neg logit contributions | |
snatch | -0.060 |
zip | -0.060 |
ck | -0.060 |
Flo | -0.059 |
unw | -0.057 |
Token | to |
Token ID | 1462 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
snatch | +0.110 |
Flo | +0.107 |
ck | +0.107 |
zip | +0.103 |
unw | +0.099 |
Neg logit contributions | |
End | -0.188 |
sky | -0.180 |
problem | -0.179 |
end | -0.167 |
happiest | -0.165 |
Token | - |
Token ID | 12 |
Feature activation | +9.22e-02 |
Pos logit contributions | |
ck | +0.549 |
snatch | +0.544 |
zip | +0.542 |
Flo | +0.531 |
unw | +0.513 |
Neg logit contributions | |
End | -0.570 |
problem | -0.537 |
sky | -0.519 |
happiest | -0.492 |
end | -0.486 |
Token | foot |
Token ID | 5898 |
Feature activation | +8.63e-02 |
Pos logit contributions | |
snatch | +0.107 |
Flo | +0.104 |
ck | +0.103 |
zip | +0.099 |
unw | +0.096 |
Neg logit contributions | |
End | -0.191 |
sky | -0.184 |
problem | -0.182 |
end | -0.171 |
happiest | -0.168 |
Token | in |
Token ID | 287 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
snatch | +0.157 |
zip | +0.157 |
ck | +0.156 |
Flo | +0.155 |
zzy | +0.149 |
Neg logit contributions | |
End | -0.125 |
problem | -0.121 |
sky | -0.116 |
end | -0.107 |
happiest | -0.106 |
Token | impat |
Token ID | 28189 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
Flo | +0.346 |
ck | +0.345 |
zip | +0.343 |
snatch | +0.338 |
zzy | +0.326 |
Neg logit contributions | |
End | -0.269 |
problem | -0.258 |
sky | -0.253 |
end | -0.238 |
best | -0.227 |
Token | ience |
Token ID | 1240 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
ck | +0.358 |
snatch | +0.345 |
zip | +0.342 |
Flo | +0.338 |
unw | +0.335 |
Neg logit contributions | |
End | -0.208 |
sky | -0.195 |
problem | -0.178 |
happiest | -0.177 |
best | -0.172 |
Token | lesson |
Token ID | 11483 |
Feature activation | -8.50e-02 |
Pos logit contributions | |
End | +0.203 |
sky | +0.182 |
problem | +0.179 |
happiest | +0.169 |
gloomy | +0.159 |
Neg logit contributions | |
snatch | -0.311 |
Flo | -0.310 |
ck | -0.306 |
zip | -0.306 |
zzy | -0.303 |
Token | â |
Token ID | 22940 |
Feature activation | +4.62e-02 |
Pos logit contributions | |
End | +0.104 |
problem | +0.099 |
sky | +0.097 |
happiest | +0.091 |
gloomy | +0.087 |
Neg logit contributions | |
ck | -0.168 |
snatch | -0.166 |
zip | -0.165 |
zzy | -0.165 |
Flo | -0.163 |
Token | € |
Token ID | 26391 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
snatch | +0.070 |
zip | +0.069 |
Flo | +0.068 |
ck | +0.067 |
unw | +0.066 |
Neg logit contributions | |
End | -0.085 |
problem | -0.079 |
end | -0.075 |
sky | -0.075 |
day | -0.072 |
Token | � |
Token ID | 447 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
zip | +0.018 |
Flo | +0.018 |
zzy | +0.017 |
ck | +0.017 |
snatch | +0.017 |
Neg logit contributions | |
End | -0.021 |
problem | -0.021 |
sky | -0.020 |
end | -0.019 |
happiest | -0.019 |
Token | � |
Token ID | 251 |
Feature activation | -9.41e-03 |
Pos logit contributions | |
End | +0.217 |
problem | +0.217 |
sky | +0.204 |
happiest | +0.192 |
gloomy | +0.192 |
Neg logit contributions | |
zip | -0.213 |
Flo | -0.209 |
snatch | -0.208 |
ck | -0.206 |
zzy | -0.201 |
Token | to |
Token ID | 1462 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
problem | +0.015 |
End | +0.014 |
sky | +0.013 |
gloomy | +0.012 |
happiest | +0.012 |
Neg logit contributions | |
zip | -0.016 |
snatch | -0.016 |
Flo | -0.016 |
ck | -0.016 |
steer | -0.015 |
Token | be |
Token ID | 307 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
zip | +0.529 |
snatch | +0.529 |
Flo | +0.524 |
ck | +0.520 |
unw | +0.502 |
Neg logit contributions | |
End | -0.575 |
problem | -0.554 |
sky | -0.528 |
end | -0.506 |
best | -0.497 |
Token | more |
Token ID | 517 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
problem | +0.033 |
End | +0.032 |
sky | +0.031 |
end | +0.028 |
gloomy | +0.028 |
Neg logit contributions | |
zip | -0.036 |
ck | -0.036 |
snatch | -0.035 |
Flo | -0.035 |
zzy | -0.035 |
Token | careful |
Token ID | 8161 |
Feature activation | -5.69e-02 |
Pos logit contributions | |
End | +0.271 |
problem | +0.253 |
sky | +0.251 |
end | +0.237 |
happiest | +0.223 |
Neg logit contributions | |
ck | -0.254 |
snatch | -0.251 |
zip | -0.249 |
zzy | -0.246 |
Flo | -0.246 |
Token | when |
Token ID | 618 |
Feature activation | -5.17e-02 |
Pos logit contributions | |
End | +0.090 |
problem | +0.088 |
sky | +0.084 |
happiest | +0.077 |
gloomy | +0.077 |
Neg logit contributions | |
snatch | -0.094 |
ck | -0.093 |
zip | -0.092 |
Flo | -0.092 |
unw | -0.088 |
Token | flying |
Token ID | 7348 |
Feature activation | -4.66e-02 |
Pos logit contributions | |
End | +0.076 |
problem | +0.072 |
sky | +0.070 |
happiest | +0.066 |
end | +0.065 |
Neg logit contributions | |
snatch | -0.090 |
ck | -0.089 |
zip | -0.087 |
Flo | -0.087 |
steer | -0.085 |
Token | boy |
Token ID | 2933 |
Feature activation | +3.00e-02 |
Pos logit contributions | |
snatch | +0.139 |
zip | +0.138 |
Flo | +0.137 |
ck | +0.136 |
unw | +0.131 |
Neg logit contributions | |
End | -0.094 |
problem | -0.087 |
sky | -0.080 |
end | -0.075 |
happiest | -0.075 |
Token | grinned |
Token ID | 36268 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
zip | +0.056 |
ck | +0.056 |
Flo | +0.055 |
snatch | +0.055 |
unw | +0.053 |
Neg logit contributions | |
End | -0.043 |
problem | -0.041 |
sky | -0.038 |
gloomy | -0.036 |
end | -0.036 |
Token | from |
Token ID | 422 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
snatch | +0.048 |
zip | +0.048 |
ck | +0.048 |
Flo | +0.048 |
zzy | +0.046 |
Neg logit contributions | |
End | -0.041 |
problem | -0.040 |
sky | -0.038 |
happiest | -0.035 |
end | -0.035 |
Token | ear |
Token ID | 1027 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
snatch | +0.250 |
ck | +0.249 |
zip | +0.245 |
Flo | +0.241 |
unw | +0.230 |
Neg logit contributions | |
End | -0.256 |
problem | -0.253 |
sky | -0.244 |
happiest | -0.227 |
best | -0.227 |
Token | - |
Token ID | 12 |
Feature activation | +9.10e-02 |
Pos logit contributions | |
End | +0.059 |
problem | +0.056 |
sky | +0.053 |
end | +0.051 |
happiest | +0.050 |
Neg logit contributions | |
zip | -0.057 |
ck | -0.057 |
snatch | -0.056 |
Flo | -0.056 |
unw | -0.055 |
Token | to |
Token ID | 1462 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
snatch | +0.160 |
zip | +0.157 |
ck | +0.154 |
unw | +0.153 |
Flo | +0.153 |
Neg logit contributions | |
End | -0.140 |
sky | -0.122 |
problem | -0.121 |
end | -0.118 |
happiest | -0.116 |
Token | - |
Token ID | 12 |
Feature activation | +9.13e-02 |
Pos logit contributions | |
snatch | +0.621 |
zip | +0.620 |
Flo | +0.616 |
ck | +0.613 |
zzy | +0.592 |
Neg logit contributions | |
End | -0.481 |
problem | -0.464 |
sky | -0.438 |
end | -0.411 |
best | -0.407 |
Token | ear |
Token ID | 451 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
snatch | +0.152 |
zip | +0.149 |
ck | +0.147 |
unw | +0.146 |
Flo | +0.145 |
Neg logit contributions | |
End | -0.148 |
sky | -0.129 |
problem | -0.127 |
end | -0.126 |
day | -0.124 |
Token | . |
Token ID | 13 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
snatch | +0.219 |
zip | +0.219 |
ck | +0.217 |
Flo | +0.217 |
zzy | +0.209 |
Neg logit contributions | |
End | -0.153 |
problem | -0.146 |
sky | -0.140 |
end | -0.129 |
happiest | -0.127 |
Token | When |
Token ID | 1649 |
Feature activation | +4.07e-02 |
Pos logit contributions | |
End | +0.078 |
problem | +0.075 |
sky | +0.073 |
end | +0.068 |
happiest | +0.067 |
Neg logit contributions | |
snatch | -0.072 |
ck | -0.072 |
zip | -0.072 |
Flo | -0.071 |
zzy | -0.068 |
Token | they |
Token ID | 484 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
snatch | +0.066 |
zip | +0.066 |
Flo | +0.065 |
ck | +0.065 |
unw | +0.064 |
Neg logit contributions | |
End | -0.064 |
problem | -0.063 |
sky | -0.057 |
happiest | -0.057 |
best | -0.054 |
Token | it |
Token ID | 340 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
End | +0.153 |
problem | +0.140 |
sky | +0.133 |
happiest | +0.124 |
end | +0.116 |
Neg logit contributions | |
unw | -0.287 |
snatch | -0.286 |
zip | -0.286 |
Flo | -0.284 |
ck | -0.279 |
Token | ! |
Token ID | 0 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
End | +0.149 |
problem | +0.135 |
sky | +0.127 |
happiest | +0.121 |
gloomy | +0.117 |
Neg logit contributions | |
snatch | -0.188 |
ck | -0.184 |
zip | -0.184 |
Flo | -0.183 |
unw | -0.180 |
Token |
|
Token ID | 198 |
Feature activation | +5.13e-02 |
Pos logit contributions | |
End | +0.023 |
sky | +0.022 |
problem | +0.020 |
end | +0.018 |
day | +0.018 |
Neg logit contributions | |
ck | -0.036 |
snatch | -0.035 |
zip | -0.034 |
zzy | -0.033 |
Flo | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | +5.28e-02 |
Pos logit contributions | |
snatch | +0.111 |
zip | +0.105 |
ck | +0.105 |
unw | +0.101 |
steer | +0.101 |
Neg logit contributions | |
End | -0.069 |
sky | -0.060 |
end | -0.052 |
problem | -0.051 |
day | -0.051 |
Token | The |
Token ID | 464 |
Feature activation | +4.62e-03 |
Pos logit contributions | |
snatch | +0.088 |
ck | +0.085 |
rily | +0.081 |
ried | +0.081 |
onna | +0.081 |
Neg logit contributions | |
End | -0.093 |
sky | -0.077 |
end | -0.074 |
day | -0.070 |
problem | -0.068 |
Token | End |
Token ID | 5268 |
Feature activation | -4.06e-01 |
Pos logit contributions | |
snatch | +0.009 |
ck | +0.009 |
Flo | +0.009 |
zip | +0.009 |
unw | +0.008 |
Neg logit contributions | |
End | -0.006 |
problem | -0.006 |
sky | -0.005 |
happiest | -0.005 |
gloomy | -0.005 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.35e-02 |
Pos logit contributions | |
End | +0.534 |
problem | +0.509 |
sky | +0.483 |
happiest | +0.448 |
end | +0.445 |
Neg logit contributions | |
zip | -0.792 |
snatch | -0.791 |
ck | -0.785 |
Flo | -0.785 |
zzy | -0.765 |
Token | Once |
Token ID | 7454 |
Feature activation | +9.30e-02 |
Pos logit contributions | |
End | +0.050 |
sky | +0.048 |
problem | +0.045 |
end | +0.040 |
best | +0.038 |
Neg logit contributions | |
snatch | -0.095 |
ck | -0.095 |
zip | -0.092 |
Flo | -0.091 |
unw | -0.089 |
Token | there |
Token ID | 612 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
zip | +0.107 |
ck | +0.106 |
Flo | +0.106 |
zzy | +0.104 |
snatch | +0.102 |
Neg logit contributions | |
End | -0.188 |
problem | -0.182 |
sky | -0.177 |
happiest | -0.172 |
best | -0.167 |
Token | was |
Token ID | 373 |
Feature activation | -6.49e-02 |
Pos logit contributions | |
snatch | +0.237 |
zip | +0.237 |
ck | +0.235 |
Flo | +0.234 |
zzy | +0.228 |
Neg logit contributions | |
End | -0.100 |
problem | -0.094 |
sky | -0.088 |
end | -0.078 |
happiest | -0.077 |
Token | a |
Token ID | 257 |
Feature activation | -3.74e-03 |
Pos logit contributions | |
End | +0.072 |
problem | +0.067 |
sky | +0.066 |
gloomy | +0.060 |
end | +0.059 |
Neg logit contributions | |
snatch | -0.144 |
zip | -0.143 |
Flo | -0.139 |
ck | -0.139 |
zzy | -0.134 |
Token | !" |
Token ID | 2474 |
Feature activation | -9.67e-04 |
Pos logit contributions | |
problem | +0.078 |
End | +0.076 |
happiest | +0.069 |
gloomy | +0.068 |
sky | +0.068 |
Neg logit contributions | |
zip | -0.068 |
Flo | -0.064 |
unw | -0.064 |
snatch | -0.064 |
zzy | -0.061 |
Token | |
Token ID | 220 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
End | +0.001 |
sky | +0.001 |
problem | +0.001 |
end | +0.001 |
day | +0.001 |
Neg logit contributions | |
zip | -0.002 |
ck | -0.002 |
snatch | -0.002 |
ge | -0.002 |
zzy | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +5.30e-02 |
Pos logit contributions | |
End | +0.180 |
sky | +0.170 |
problem | +0.166 |
best | +0.155 |
end | +0.152 |
Neg logit contributions | |
snatch | -0.238 |
ck | -0.230 |
zip | -0.229 |
Flo | -0.221 |
unw | -0.220 |
Token |
|
Token ID | 198 |
Feature activation | +5.22e-02 |
Pos logit contributions | |
snatch | +0.124 |
ck | +0.119 |
ried | +0.118 |
ge | +0.116 |
steer | +0.114 |
Neg logit contributions | |
End | -0.057 |
sky | -0.044 |
day | -0.039 |
end | -0.038 |
living | -0.034 |
Token | The |
Token ID | 464 |
Feature activation | +2.16e-03 |
Pos logit contributions | |
snatch | +0.083 |
ck | +0.081 |
onna | +0.081 |
ried | +0.080 |
rily | +0.079 |
Neg logit contributions | |
End | -0.090 |
sky | -0.073 |
end | -0.072 |
day | -0.069 |
living | -0.065 |
Token | End |
Token ID | 5268 |
Feature activation | -4.01e-01 |
Pos logit contributions | |
snatch | +0.004 |
ck | +0.004 |
Flo | +0.004 |
zip | +0.004 |
unw | +0.004 |
Neg logit contributions | |
End | -0.003 |
problem | -0.003 |
sky | -0.002 |
happiest | -0.002 |
gloomy | -0.002 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
End | +0.554 |
problem | +0.527 |
sky | +0.502 |
happiest | +0.474 |
end | +0.464 |
Neg logit contributions | |
zip | -0.755 |
snatch | -0.753 |
Flo | -0.749 |
ck | -0.746 |
zzy | -0.732 |
Token | Once |
Token ID | 7454 |
Feature activation | +9.09e-02 |
Pos logit contributions | |
End | +0.053 |
sky | +0.051 |
problem | +0.048 |
end | +0.043 |
best | +0.041 |
Neg logit contributions | |
snatch | -0.099 |
ck | -0.099 |
zip | -0.096 |
Flo | -0.095 |
unw | -0.092 |
Token | upon |
Token ID | 2402 |
Feature activation | +9.64e-02 |
Pos logit contributions | |
zip | +0.096 |
ck | +0.096 |
Flo | +0.094 |
uzz | +0.094 |
zzy | +0.093 |
Neg logit contributions | |
End | -0.187 |
problem | -0.182 |
sky | -0.178 |
happiest | -0.176 |
best | -0.168 |
Token | a |
Token ID | 257 |
Feature activation | -8.00e-03 |
Pos logit contributions | |
snatch | +0.184 |
zip | +0.183 |
ck | +0.181 |
Flo | +0.180 |
zzy | +0.173 |
Neg logit contributions | |
End | -0.132 |
problem | -0.126 |
sky | -0.123 |
end | -0.112 |
gloomy | -0.111 |
Token | time |
Token ID | 640 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
End | +0.009 |
problem | +0.009 |
end | +0.008 |
sky | +0.008 |
gloomy | +0.007 |
Neg logit contributions | |
zip | -0.018 |
snatch | -0.018 |
ck | -0.018 |
Flo | -0.018 |
uzz | -0.017 |
Token | ! |
Token ID | 0 |
Feature activation | -3.98e-03 |
Pos logit contributions | |
End | +0.002 |
problem | +0.002 |
sky | +0.002 |
happiest | +0.001 |
end | +0.001 |
Neg logit contributions | |
snatch | -0.003 |
Flo | -0.003 |
ck | -0.003 |
zip | -0.003 |
zzy | -0.002 |
Token | |
Token ID | 220 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
End | +0.006 |
sky | +0.005 |
problem | +0.005 |
best | +0.005 |
gloomy | +0.004 |
Neg logit contributions | |
ck | -0.008 |
snatch | -0.008 |
zip | -0.008 |
zzy | -0.007 |
unw | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +5.54e-02 |
Pos logit contributions | |
End | +0.194 |
sky | +0.186 |
problem | +0.185 |
best | +0.169 |
end | +0.167 |
Neg logit contributions | |
snatch | -0.220 |
ck | -0.216 |
zip | -0.211 |
Flo | -0.208 |
unw | -0.204 |
Token |
|
Token ID | 198 |
Feature activation | +5.33e-02 |
Pos logit contributions | |
snatch | +0.123 |
ck | +0.119 |
zip | +0.114 |
unw | +0.113 |
steer | +0.112 |
Neg logit contributions | |
End | -0.070 |
sky | -0.061 |
day | -0.052 |
end | -0.051 |
problem | -0.049 |
Token | The |
Token ID | 464 |
Feature activation | +2.75e-03 |
Pos logit contributions | |
snatch | +0.086 |
ck | +0.085 |
unw | +0.080 |
onna | +0.080 |
zip | +0.080 |
Neg logit contributions | |
End | -0.096 |
sky | -0.081 |
end | -0.079 |
day | -0.075 |
problem | -0.074 |
Token | End |
Token ID | 5268 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
snatch | +0.005 |
ck | +0.005 |
Flo | +0.005 |
zip | +0.005 |
unw | +0.005 |
Neg logit contributions | |
End | -0.003 |
problem | -0.003 |
sky | -0.003 |
happiest | -0.003 |
gloomy | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -4.45e-02 |
Pos logit contributions | |
End | +0.527 |
problem | +0.504 |
sky | +0.479 |
end | +0.440 |
happiest | +0.440 |
Neg logit contributions | |
zip | -0.781 |
snatch | -0.781 |
ck | -0.775 |
Flo | -0.774 |
zzy | -0.750 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
End | +0.060 |
sky | +0.057 |
day | +0.049 |
end | +0.048 |
problem | +0.047 |
Neg logit contributions | |
ck | -0.096 |
snatch | -0.095 |
zip | -0.090 |
unw | -0.087 |
zzy | -0.086 |
Token | Once |
Token ID | 7454 |
Feature activation | +9.73e-02 |
Pos logit contributions | |
End | +0.054 |
sky | +0.050 |
problem | +0.050 |
end | +0.044 |
best | +0.043 |
Neg logit contributions | |
snatch | -0.091 |
ck | -0.091 |
zip | -0.090 |
Flo | -0.089 |
unw | -0.086 |
Token | there |
Token ID | 612 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
uzz | +0.094 |
ck | +0.092 |
zip | +0.091 |
zzy | +0.091 |
Flo | +0.090 |
Neg logit contributions | |
End | -0.206 |
problem | -0.198 |
sky | -0.198 |
happiest | -0.196 |
sounds | -0.186 |
Token | was |
Token ID | 373 |
Feature activation | -6.22e-02 |
Pos logit contributions | |
snatch | +0.225 |
zip | +0.225 |
Flo | +0.225 |
ck | +0.223 |
zzy | +0.217 |
Neg logit contributions | |
End | -0.116 |
problem | -0.113 |
sky | -0.105 |
gloomy | -0.093 |
happiest | -0.093 |
Token | place |
Token ID | 1295 |
Feature activation | -5.41e-02 |
Pos logit contributions | |
snatch | +0.039 |
Flo | +0.039 |
zip | +0.038 |
ck | +0.038 |
unw | +0.037 |
Neg logit contributions | |
End | -0.029 |
problem | -0.026 |
sky | -0.025 |
happiest | -0.024 |
gloomy | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -4.87e-02 |
Pos logit contributions | |
End | +0.084 |
problem | +0.078 |
sky | +0.074 |
happiest | +0.071 |
end | +0.069 |
Neg logit contributions | |
zip | -0.092 |
snatch | -0.092 |
Flo | -0.092 |
ck | -0.091 |
unw | -0.090 |
Token |
|
Token ID | 198 |
Feature activation | +5.90e-02 |
Pos logit contributions | |
End | +0.065 |
sky | +0.063 |
problem | +0.056 |
end | +0.053 |
day | +0.052 |
Neg logit contributions | |
ck | -0.101 |
snatch | -0.097 |
zip | -0.095 |
zzy | -0.093 |
unw | -0.090 |
Token |
|
Token ID | 198 |
Feature activation | +5.72e-02 |
Pos logit contributions | |
snatch | +0.125 |
ck | +0.117 |
zip | +0.117 |
unw | +0.115 |
ried | +0.111 |
Neg logit contributions | |
End | -0.083 |
sky | -0.072 |
end | -0.067 |
day | -0.067 |
problem | -0.065 |
Token | The |
Token ID | 464 |
Feature activation | +5.87e-03 |
Pos logit contributions | |
snatch | +0.103 |
ck | +0.097 |
unw | +0.094 |
ried | +0.094 |
rily | +0.094 |
Neg logit contributions | |
End | -0.097 |
end | -0.080 |
sky | -0.079 |
day | -0.076 |
problem | -0.074 |
Token | End |
Token ID | 5268 |
Feature activation | -3.99e-01 |
Pos logit contributions | |
snatch | +0.011 |
zip | +0.011 |
ck | +0.011 |
Flo | +0.011 |
unw | +0.011 |
Neg logit contributions | |
End | -0.008 |
problem | -0.007 |
sky | -0.007 |
happiest | -0.006 |
gloomy | -0.006 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.48e-02 |
Pos logit contributions | |
End | +0.544 |
problem | +0.516 |
sky | +0.490 |
happiest | +0.464 |
end | +0.454 |
Neg logit contributions | |
zip | -0.762 |
snatch | -0.757 |
Flo | -0.754 |
ck | -0.753 |
zzy | -0.737 |
Token | Once |
Token ID | 7454 |
Feature activation | +9.66e-02 |
Pos logit contributions | |
End | +0.051 |
sky | +0.050 |
problem | +0.046 |
end | +0.042 |
best | +0.040 |
Neg logit contributions | |
snatch | -0.098 |
ck | -0.097 |
zip | -0.094 |
Flo | -0.094 |
unw | -0.091 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
ck | +0.106 |
zip | +0.106 |
Flo | +0.105 |
zzy | +0.103 |
uzz | +0.102 |
Neg logit contributions | |
End | -0.197 |
problem | -0.190 |
sky | -0.186 |
happiest | -0.184 |
best | -0.175 |
Token | a |
Token ID | 257 |
Feature activation | -6.58e-03 |
Pos logit contributions | |
snatch | +0.196 |
zip | +0.195 |
ck | +0.193 |
Flo | +0.192 |
zzy | +0.185 |
Neg logit contributions | |
End | -0.137 |
problem | -0.129 |
sky | -0.127 |
end | -0.116 |
gloomy | -0.115 |
Token | time |
Token ID | 640 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
End | +0.007 |
problem | +0.007 |
sky | +0.006 |
end | +0.006 |
best | +0.006 |
Neg logit contributions | |
zip | -0.015 |
snatch | -0.015 |
ck | -0.014 |
Flo | -0.014 |
uzz | -0.014 |
Token | better |
Token ID | 1365 |
Feature activation | -7.15e-02 |
Pos logit contributions | |
ck | +0.036 |
snatch | +0.035 |
Flo | +0.035 |
unw | +0.033 |
zip | +0.033 |
Neg logit contributions | |
End | -0.024 |
problem | -0.023 |
sky | -0.022 |
happiest | -0.019 |
end | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -5.17e-02 |
Pos logit contributions | |
End | +0.106 |
problem | +0.099 |
sky | +0.099 |
happiest | +0.093 |
end | +0.088 |
Neg logit contributions | |
ck | -0.123 |
snatch | -0.121 |
Flo | -0.121 |
zip | -0.116 |
zzy | -0.115 |
Token |
|
Token ID | 198 |
Feature activation | +5.33e-02 |
Pos logit contributions | |
End | +0.069 |
sky | +0.067 |
problem | +0.059 |
day | +0.055 |
gloomy | +0.054 |
Neg logit contributions | |
ck | -0.107 |
snatch | -0.103 |
zip | -0.101 |
zzy | -0.098 |
unw | -0.095 |
Token |
|
Token ID | 198 |
Feature activation | +5.28e-02 |
Pos logit contributions | |
snatch | +0.112 |
zip | +0.105 |
ck | +0.104 |
unw | +0.103 |
ried | +0.100 |
Neg logit contributions | |
End | -0.076 |
sky | -0.064 |
end | -0.060 |
problem | -0.058 |
day | -0.058 |
Token | The |
Token ID | 464 |
Feature activation | +3.00e-03 |
Pos logit contributions | |
snatch | +0.091 |
rily | +0.086 |
ck | +0.085 |
unw | +0.085 |
onna | +0.084 |
Neg logit contributions | |
End | -0.092 |
end | -0.075 |
sky | -0.074 |
day | -0.070 |
problem | -0.070 |
Token | End |
Token ID | 5268 |
Feature activation | -3.98e-01 |
Pos logit contributions | |
snatch | +0.006 |
ck | +0.006 |
Flo | +0.006 |
zip | +0.006 |
unw | +0.006 |
Neg logit contributions | |
End | -0.004 |
problem | -0.004 |
sky | -0.003 |
happiest | -0.003 |
gloomy | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -4.47e-02 |
Pos logit contributions | |
End | +0.511 |
problem | +0.487 |
sky | +0.463 |
end | +0.424 |
happiest | +0.424 |
Neg logit contributions | |
zip | -0.790 |
snatch | -0.789 |
ck | -0.783 |
Flo | -0.782 |
zzy | -0.759 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.18e-02 |
Pos logit contributions | |
End | +0.060 |
sky | +0.057 |
day | +0.049 |
problem | +0.049 |
end | +0.048 |
Neg logit contributions | |
ck | -0.096 |
snatch | -0.095 |
zip | -0.089 |
unw | -0.086 |
Flo | -0.086 |
Token | Once |
Token ID | 7454 |
Feature activation | +9.80e-02 |
Pos logit contributions | |
End | +0.051 |
sky | +0.049 |
problem | +0.048 |
end | +0.042 |
best | +0.041 |
Neg logit contributions | |
snatch | -0.087 |
ck | -0.087 |
zip | -0.085 |
Flo | -0.085 |
unw | -0.082 |
Token | there |
Token ID | 612 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
uzz | +0.091 |
ck | +0.088 |
zip | +0.088 |
Flo | +0.086 |
zzy | +0.086 |
Neg logit contributions | |
End | -0.211 |
happiest | -0.203 |
problem | -0.201 |
sky | -0.200 |
sounds | -0.190 |
Token | was |
Token ID | 373 |
Feature activation | -6.08e-02 |
Pos logit contributions | |
zip | +0.223 |
snatch | +0.222 |
Flo | +0.222 |
ck | +0.219 |
unw | +0.214 |
Neg logit contributions | |
End | -0.122 |
problem | -0.119 |
sky | -0.110 |
happiest | -0.099 |
gloomy | -0.099 |
Token | . |
Token ID | 13 |
Feature activation | -5.61e-02 |
Pos logit contributions | |
End | +0.090 |
problem | +0.084 |
sky | +0.081 |
happiest | +0.075 |
gloomy | +0.075 |
Neg logit contributions | |
Flo | -0.092 |
snatch | -0.092 |
ck | -0.091 |
zip | -0.090 |
unw | -0.088 |
Token | |
Token ID | 220 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
End | +0.077 |
sky | +0.073 |
problem | +0.065 |
day | +0.062 |
end | +0.061 |
Neg logit contributions | |
ck | -0.113 |
snatch | -0.110 |
zip | -0.107 |
zzy | -0.103 |
steer | -0.102 |
Token |
|
Token ID | 198 |
Feature activation | +4.38e-02 |
Pos logit contributions | |
End | +0.183 |
sky | +0.174 |
problem | +0.163 |
best | +0.155 |
end | +0.150 |
Neg logit contributions | |
snatch | -0.253 |
ck | -0.247 |
zip | -0.238 |
steer | -0.232 |
unw | -0.232 |
Token |
|
Token ID | 198 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
snatch | +0.094 |
ck | +0.088 |
zip | +0.088 |
steer | +0.086 |
unw | +0.085 |
Neg logit contributions | |
End | -0.060 |
sky | -0.049 |
day | -0.046 |
end | -0.045 |
days | -0.040 |
Token | The |
Token ID | 464 |
Feature activation | +1.92e-04 |
Pos logit contributions | |
snatch | +0.072 |
ck | +0.069 |
zip | +0.066 |
steer | +0.066 |
shave | +0.066 |
Neg logit contributions | |
End | -0.076 |
sky | -0.062 |
end | -0.061 |
day | -0.059 |
days | -0.055 |
Token | End |
Token ID | 5268 |
Feature activation | -3.98e-01 |
Pos logit contributions | |
snatch | +0.000 |
ck | +0.000 |
Flo | +0.000 |
zip | +0.000 |
unw | +0.000 |
Neg logit contributions | |
problem | -0.000 |
End | -0.000 |
sky | -0.000 |
gloomy | -0.000 |
happiest | -0.000 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
End | +0.508 |
problem | +0.489 |
sky | +0.466 |
end | +0.424 |
best | +0.419 |
Neg logit contributions | |
snatch | -0.795 |
zip | -0.792 |
ck | -0.789 |
Flo | -0.782 |
zzy | -0.752 |
Token | Once |
Token ID | 7454 |
Feature activation | +9.29e-02 |
Pos logit contributions | |
End | +0.055 |
sky | +0.053 |
problem | +0.050 |
end | +0.045 |
best | +0.043 |
Neg logit contributions | |
snatch | -0.097 |
ck | -0.097 |
zip | -0.094 |
Flo | -0.093 |
unw | -0.091 |
Token | upon |
Token ID | 2402 |
Feature activation | +9.97e-02 |
Pos logit contributions | |
zip | +0.103 |
Flo | +0.102 |
ck | +0.102 |
zzy | +0.100 |
uzz | +0.099 |
Neg logit contributions | |
End | -0.189 |
problem | -0.183 |
sky | -0.178 |
happiest | -0.175 |
best | -0.169 |
Token | a |
Token ID | 257 |
Feature activation | -7.97e-03 |
Pos logit contributions | |
snatch | +0.195 |
zip | +0.193 |
ck | +0.192 |
Flo | +0.190 |
zzy | +0.183 |
Neg logit contributions | |
End | -0.133 |
problem | -0.125 |
sky | -0.123 |
end | -0.113 |
gloomy | -0.112 |
Token | time |
Token ID | 640 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
End | +0.009 |
problem | +0.008 |
end | +0.007 |
sky | +0.007 |
best | +0.007 |
Neg logit contributions | |
zip | -0.019 |
snatch | -0.018 |
ck | -0.018 |
Flo | -0.018 |
uzz | -0.017 |
Token | 't |
Token ID | 470 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
zip | +0.038 |
ck | +0.038 |
snatch | +0.037 |
zzy | +0.036 |
uzz | +0.036 |
Neg logit contributions | |
End | -0.036 |
sky | -0.036 |
problem | -0.035 |
end | -0.035 |
gloomy | -0.032 |
Token | vanish |
Token ID | 39572 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
zip | +0.205 |
uzz | +0.193 |
Flo | +0.192 |
zzy | +0.191 |
izer | +0.189 |
Neg logit contributions | |
problem | -0.175 |
end | -0.172 |
gloomy | -0.160 |
End | -0.160 |
sky | -0.158 |
Token | again |
Token ID | 757 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
zip | +0.185 |
snatch | +0.185 |
ck | +0.183 |
Flo | +0.183 |
zzy | +0.176 |
Neg logit contributions | |
End | -0.150 |
problem | -0.146 |
sky | -0.139 |
end | -0.129 |
best | -0.128 |
Token | ! |
Token ID | 0 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
snatch | +0.216 |
zip | +0.215 |
ck | +0.214 |
Flo | +0.213 |
zzy | +0.206 |
Neg logit contributions | |
End | -0.123 |
problem | -0.118 |
sky | -0.111 |
end | -0.101 |
happiest | -0.100 |
Token | The |
Token ID | 383 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
sky | +0.013 |
End | +0.013 |
day | +0.011 |
problem | +0.011 |
best | +0.010 |
Neg logit contributions | |
ck | -0.024 |
snatch | -0.023 |
zip | -0.022 |
steer | -0.021 |
ge | -0.021 |
Token | End |
Token ID | 5268 |
Feature activation | -3.96e-01 |
Pos logit contributions | |
snatch | +0.045 |
ck | +0.044 |
Flo | +0.044 |
zip | +0.044 |
unw | +0.042 |
Neg logit contributions | |
End | -0.031 |
problem | -0.030 |
sky | -0.028 |
happiest | -0.026 |
gloomy | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -4.66e-02 |
Pos logit contributions | |
problem | +0.406 |
End | +0.390 |
sky | +0.388 |
day | +0.338 |
best | +0.327 |
Neg logit contributions | |
snatch | -0.899 |
ck | -0.890 |
zip | -0.886 |
Flo | -0.869 |
unw | -0.824 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.39e-02 |
Pos logit contributions | |
End | +0.066 |
sky | +0.062 |
problem | +0.056 |
end | +0.054 |
day | +0.054 |
Neg logit contributions | |
ck | -0.095 |
snatch | -0.093 |
zip | -0.090 |
Flo | -0.086 |
steer | -0.086 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
End | +0.043 |
sky | +0.042 |
problem | +0.040 |
end | +0.036 |
best | +0.035 |
Neg logit contributions | |
ck | -0.071 |
snatch | -0.070 |
zip | -0.067 |
Flo | -0.067 |
unw | -0.065 |
Token | there |
Token ID | 612 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
uzz | +0.101 |
ck | +0.099 |
zip | +0.097 |
Flo | +0.096 |
zzy | +0.095 |
Neg logit contributions | |
End | -0.231 |
sky | -0.222 |
happiest | -0.220 |
problem | -0.219 |
sounds | -0.209 |
Token | was |
Token ID | 373 |
Feature activation | -5.76e-02 |
Pos logit contributions | |
Flo | +0.234 |
zip | +0.233 |
snatch | +0.233 |
ck | +0.229 |
unw | +0.225 |
Neg logit contributions | |
End | -0.132 |
problem | -0.128 |
sky | -0.118 |
happiest | -0.106 |
gloomy | -0.106 |
Token | . |
Token ID | 13 |
Feature activation | -4.20e-02 |
Pos logit contributions | |
problem | +0.016 |
End | +0.016 |
sky | +0.015 |
best | +0.014 |
gloomy | +0.014 |
Neg logit contributions | |
zip | -0.032 |
snatch | -0.032 |
Flo | -0.032 |
ck | -0.031 |
zzy | -0.030 |
Token | |
Token ID | 220 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
End | +0.058 |
sky | +0.057 |
problem | +0.050 |
day | +0.048 |
gloomy | +0.047 |
Neg logit contributions | |
ck | -0.086 |
snatch | -0.082 |
zip | -0.079 |
zzy | -0.076 |
steer | -0.075 |
Token |
|
Token ID | 198 |
Feature activation | +5.66e-02 |
Pos logit contributions | |
End | +0.165 |
sky | +0.156 |
problem | +0.145 |
best | +0.141 |
end | +0.135 |
Neg logit contributions | |
snatch | -0.248 |
ck | -0.241 |
zip | -0.231 |
unw | -0.226 |
steer | -0.226 |
Token |
|
Token ID | 198 |
Feature activation | +5.46e-02 |
Pos logit contributions | |
snatch | +0.129 |
ck | +0.125 |
ried | +0.119 |
extingu | +0.117 |
steer | +0.117 |
Neg logit contributions | |
End | -0.068 |
sky | -0.056 |
day | -0.052 |
end | -0.049 |
much | -0.043 |
Token | The |
Token ID | 464 |
Feature activation | +2.83e-03 |
Pos logit contributions | |
snatch | +0.091 |
ck | +0.090 |
onna | +0.085 |
rily | +0.085 |
ried | +0.084 |
Neg logit contributions | |
End | -0.095 |
sky | -0.078 |
end | -0.077 |
day | -0.075 |
problem | -0.069 |
Token | End |
Token ID | 5268 |
Feature activation | -3.96e-01 |
Pos logit contributions | |
snatch | +0.006 |
Flo | +0.006 |
ck | +0.006 |
zip | +0.006 |
unw | +0.005 |
Neg logit contributions | |
End | -0.003 |
problem | -0.003 |
sky | -0.003 |
happiest | -0.003 |
gloomy | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -4.59e-02 |
Pos logit contributions | |
End | +0.471 |
problem | +0.456 |
sky | +0.432 |
end | +0.389 |
best | +0.385 |
Neg logit contributions | |
snatch | -0.827 |
zip | -0.823 |
ck | -0.821 |
Flo | -0.813 |
unw | -0.782 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.24e-02 |
Pos logit contributions | |
End | +0.062 |
sky | +0.058 |
day | +0.050 |
end | +0.050 |
problem | +0.048 |
Neg logit contributions | |
ck | -0.100 |
snatch | -0.097 |
zip | -0.092 |
zzy | -0.090 |
Flo | -0.089 |
Token | Once |
Token ID | 7454 |
Feature activation | +9.60e-02 |
Pos logit contributions | |
End | +0.051 |
sky | +0.049 |
problem | +0.048 |
end | +0.042 |
best | +0.041 |
Neg logit contributions | |
ck | -0.089 |
snatch | -0.089 |
zip | -0.087 |
Flo | -0.086 |
unw | -0.084 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
uzz | +0.090 |
zip | +0.088 |
ck | +0.087 |
Flo | +0.087 |
zzy | +0.086 |
Neg logit contributions | |
End | -0.206 |
problem | -0.197 |
sky | -0.196 |
happiest | -0.195 |
sounds | -0.186 |
Token | a |
Token ID | 257 |
Feature activation | -2.97e-03 |
Pos logit contributions | |
snatch | +0.202 |
zip | +0.201 |
ck | +0.200 |
Flo | +0.198 |
zzy | +0.191 |
Neg logit contributions | |
End | -0.138 |
problem | -0.129 |
sky | -0.128 |
end | -0.117 |
gloomy | -0.116 |
Token | !" |
Token ID | 2474 |
Feature activation | +7.26e-03 |
Pos logit contributions | |
snatch | +0.185 |
zip | +0.184 |
ck | +0.183 |
Flo | +0.182 |
unw | +0.180 |
Neg logit contributions | |
End | -0.144 |
problem | -0.137 |
sky | -0.128 |
happiest | -0.120 |
gloomy | -0.116 |
Token | |
Token ID | 220 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
zip | +0.016 |
ck | +0.015 |
snatch | +0.015 |
izer | +0.014 |
steer | +0.014 |
Neg logit contributions | |
End | -0.009 |
sky | -0.008 |
day | -0.008 |
problem | -0.007 |
best | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +6.22e-02 |
Pos logit contributions | |
End | +0.161 |
sky | +0.151 |
problem | +0.141 |
best | +0.132 |
end | +0.131 |
Neg logit contributions | |
snatch | -0.254 |
ck | -0.245 |
zip | -0.242 |
steer | -0.234 |
Flo | -0.232 |
Token |
|
Token ID | 198 |
Feature activation | +5.90e-02 |
Pos logit contributions | |
snatch | +0.154 |
ried | +0.148 |
ck | +0.148 |
steer | +0.144 |
ge | +0.144 |
Neg logit contributions | |
End | -0.060 |
sky | -0.049 |
day | -0.043 |
end | -0.042 |
much | -0.036 |
Token | The |
Token ID | 464 |
Feature activation | +5.34e-03 |
Pos logit contributions | |
snatch | +0.096 |
ck | +0.094 |
ried | +0.092 |
onna | +0.089 |
rily | +0.088 |
Neg logit contributions | |
End | -0.102 |
end | -0.085 |
sky | -0.083 |
day | -0.079 |
living | -0.074 |
Token | End |
Token ID | 5268 |
Feature activation | -3.96e-01 |
Pos logit contributions | |
snatch | +0.010 |
ck | +0.010 |
Flo | +0.010 |
zip | +0.010 |
unw | +0.010 |
Neg logit contributions | |
End | -0.007 |
problem | -0.007 |
sky | -0.006 |
happiest | -0.006 |
gloomy | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
End | +0.499 |
problem | +0.481 |
sky | +0.457 |
end | +0.415 |
best | +0.409 |
Neg logit contributions | |
snatch | -0.796 |
zip | -0.795 |
ck | -0.790 |
Flo | -0.785 |
zzy | -0.754 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.19e-02 |
Pos logit contributions | |
End | +0.049 |
sky | +0.046 |
day | +0.039 |
end | +0.039 |
problem | +0.036 |
Neg logit contributions | |
ck | -0.089 |
snatch | -0.087 |
zip | -0.082 |
steer | -0.080 |
zzy | -0.079 |
Token | The |
Token ID | 464 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
End | +0.051 |
sky | +0.049 |
problem | +0.047 |
end | +0.042 |
best | +0.041 |
Neg logit contributions | |
ck | -0.088 |
snatch | -0.088 |
zip | -0.086 |
Flo | -0.085 |
unw | -0.082 |
Token | puppy |
Token ID | 26188 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
zip | +0.039 |
snatch | +0.039 |
ck | +0.038 |
Flo | +0.038 |
zzy | +0.037 |
Neg logit contributions | |
End | -0.019 |
problem | -0.018 |
sky | -0.017 |
end | -0.016 |
gloomy | -0.015 |
Token | was |
Token ID | 373 |
Feature activation | -6.15e-02 |
Pos logit contributions | |
zip | +0.258 |
snatch | +0.258 |
Flo | +0.257 |
ck | +0.256 |
unw | +0.248 |
Neg logit contributions | |
End | -0.174 |
problem | -0.166 |
sky | -0.156 |
end | -0.144 |
happiest | -0.143 |
Token | sharing |
Token ID | 7373 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
zip | +0.164 |
snatch | +0.163 |
Flo | +0.162 |
ck | +0.161 |
zzy | +0.159 |
Neg logit contributions | |
End | -0.108 |
problem | -0.102 |
sky | -0.096 |
end | -0.091 |
best | -0.089 |
Token | . |
Token ID | 13 |
Feature activation | -4.35e-02 |
Pos logit contributions | |
End | +0.018 |
problem | +0.018 |
sky | +0.017 |
happiest | +0.016 |
gloomy | +0.016 |
Neg logit contributions | |
zip | -0.018 |
snatch | -0.018 |
Flo | -0.018 |
ck | -0.018 |
unw | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | +5.77e-02 |
Pos logit contributions | |
End | +0.065 |
sky | +0.060 |
problem | +0.057 |
end | +0.053 |
best | +0.051 |
Neg logit contributions | |
ck | -0.082 |
snatch | -0.082 |
zip | -0.080 |
zzy | -0.077 |
Flo | -0.076 |
Token |
|
Token ID | 198 |
Feature activation | +5.53e-02 |
Pos logit contributions | |
snatch | +0.112 |
zip | +0.109 |
ck | +0.105 |
Flo | +0.105 |
unw | +0.104 |
Neg logit contributions | |
End | -0.087 |
sky | -0.076 |
problem | -0.073 |
end | -0.072 |
day | -0.068 |
Token | The |
Token ID | 464 |
Feature activation | +5.11e-03 |
Pos logit contributions | |
snatch | +0.097 |
zip | +0.090 |
unw | +0.089 |
rily | +0.089 |
ck | +0.089 |
Neg logit contributions | |
End | -0.099 |
end | -0.083 |
sky | -0.081 |
day | -0.078 |
problem | -0.074 |
Token | End |
Token ID | 5268 |
Feature activation | -3.95e-01 |
Pos logit contributions | |
snatch | +0.011 |
ck | +0.011 |
Flo | +0.011 |
zip | +0.011 |
unw | +0.010 |
Neg logit contributions | |
End | -0.005 |
problem | -0.005 |
sky | -0.005 |
happiest | -0.004 |
gloomy | -0.004 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.11e-02 |
Pos logit contributions | |
End | +0.536 |
problem | +0.513 |
sky | +0.488 |
happiest | +0.452 |
end | +0.450 |
Neg logit contributions | |
zip | -0.753 |
snatch | -0.751 |
ck | -0.746 |
Flo | -0.745 |
zzy | -0.724 |
Token | One |
Token ID | 3198 |
Feature activation | +2.41e-02 |
Pos logit contributions | |
End | +0.048 |
sky | +0.047 |
problem | +0.043 |
end | +0.040 |
best | +0.038 |
Neg logit contributions | |
snatch | -0.089 |
ck | -0.089 |
zip | -0.085 |
Flo | -0.085 |
unw | -0.083 |
Token | day |
Token ID | 1110 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
zzy | +0.059 |
ck | +0.059 |
zip | +0.058 |
snatch | +0.058 |
steer | +0.057 |
Neg logit contributions | |
problem | -0.015 |
End | -0.014 |
sky | -0.013 |
happiest | -0.011 |
starts | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -8.02e-02 |
Pos logit contributions | |
End | +0.194 |
problem | +0.183 |
sky | +0.168 |
end | +0.142 |
best | +0.140 |
Neg logit contributions | |
snatch | -0.626 |
zip | -0.625 |
ck | -0.622 |
Flo | -0.618 |
zzy | -0.600 |
Token | a |
Token ID | 257 |
Feature activation | -4.89e-03 |
Pos logit contributions | |
End | +0.098 |
sky | +0.090 |
gloomy | +0.086 |
problem | +0.082 |
day | +0.082 |
Neg logit contributions | |
snatch | -0.172 |
ck | -0.168 |
zip | -0.165 |
Flo | -0.160 |
onna | -0.158 |
Token | . |
Token ID | 13 |
Feature activation | -3.63e-02 |
Pos logit contributions | |
snatch | +0.078 |
Flo | +0.077 |
ck | +0.076 |
zip | +0.076 |
unw | +0.075 |
Neg logit contributions | |
End | -0.062 |
problem | -0.057 |
sky | -0.053 |
happiest | -0.051 |
gloomy | -0.050 |
Token | |
Token ID | 220 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
sky | +0.047 |
End | +0.046 |
day | +0.038 |
problem | +0.037 |
sun | +0.035 |
Neg logit contributions | |
ck | -0.079 |
snatch | -0.073 |
zip | -0.070 |
zzy | -0.068 |
ge | -0.066 |
Token |
|
Token ID | 198 |
Feature activation | +5.86e-02 |
Pos logit contributions | |
End | +0.153 |
sky | +0.147 |
problem | +0.130 |
best | +0.129 |
end | +0.122 |
Neg logit contributions | |
snatch | -0.257 |
ck | -0.251 |
zip | -0.237 |
unw | -0.236 |
Flo | -0.232 |
Token |
|
Token ID | 198 |
Feature activation | +5.63e-02 |
Pos logit contributions | |
snatch | +0.133 |
ck | +0.129 |
ried | +0.123 |
extingu | +0.121 |
unw | +0.121 |
Neg logit contributions | |
End | -0.069 |
sky | -0.057 |
day | -0.051 |
end | -0.047 |
much | -0.044 |
Token | The |
Token ID | 464 |
Feature activation | +5.21e-03 |
Pos logit contributions | |
snatch | +0.093 |
ck | +0.092 |
onna | +0.087 |
ried | +0.087 |
unw | +0.086 |
Neg logit contributions | |
End | -0.098 |
sky | -0.079 |
end | -0.078 |
day | -0.076 |
days | -0.070 |
Token | End |
Token ID | 5268 |
Feature activation | -3.94e-01 |
Pos logit contributions | |
snatch | +0.010 |
ck | +0.010 |
Flo | +0.010 |
zip | +0.010 |
unw | +0.010 |
Neg logit contributions | |
End | -0.006 |
problem | -0.006 |
sky | -0.006 |
happiest | -0.005 |
gloomy | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -4.33e-02 |
Pos logit contributions | |
End | +0.453 |
problem | +0.442 |
sky | +0.418 |
end | +0.374 |
best | +0.372 |
Neg logit contributions | |
snatch | -0.837 |
zip | -0.833 |
ck | -0.832 |
Flo | -0.821 |
unw | -0.789 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.25e-02 |
Pos logit contributions | |
End | +0.053 |
sky | +0.051 |
day | +0.043 |
end | +0.042 |
night | +0.039 |
Neg logit contributions | |
ck | -0.101 |
snatch | -0.097 |
zip | -0.092 |
zzy | -0.089 |
unw | -0.089 |
Token | Once |
Token ID | 7454 |
Feature activation | +9.81e-02 |
Pos logit contributions | |
End | +0.051 |
sky | +0.049 |
problem | +0.048 |
end | +0.042 |
best | +0.041 |
Neg logit contributions | |
ck | -0.089 |
snatch | -0.089 |
zip | -0.087 |
Flo | -0.086 |
unw | -0.084 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
zip | +0.097 |
uzz | +0.097 |
ck | +0.097 |
Flo | +0.096 |
zzy | +0.094 |
Neg logit contributions | |
End | -0.207 |
problem | -0.199 |
sky | -0.197 |
happiest | -0.194 |
end | -0.184 |
Token | a |
Token ID | 257 |
Feature activation | -3.06e-03 |
Pos logit contributions | |
snatch | +0.203 |
zip | +0.202 |
ck | +0.201 |
Flo | +0.199 |
zzy | +0.192 |
Neg logit contributions | |
End | -0.142 |
problem | -0.133 |
sky | -0.131 |
end | -0.120 |
gloomy | -0.119 |
Token | steal |
Token ID | 8711 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
zip | +0.230 |
Flo | +0.228 |
zzy | +0.225 |
ck | +0.219 |
uzz | +0.215 |
Neg logit contributions | |
End | -0.202 |
end | -0.201 |
problem | -0.200 |
sky | -0.195 |
day | -0.184 |
Token | from |
Token ID | 422 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
zip | +0.304 |
snatch | +0.302 |
ck | +0.300 |
Flo | +0.298 |
zzy | +0.291 |
Neg logit contributions | |
End | -0.237 |
problem | -0.232 |
sky | -0.222 |
end | -0.207 |
best | -0.203 |
Token | them |
Token ID | 606 |
Feature activation | +3.83e-02 |
Pos logit contributions | |
zip | +0.291 |
zzy | +0.284 |
Flo | +0.282 |
snatch | +0.278 |
ck | +0.277 |
Neg logit contributions | |
End | -0.228 |
problem | -0.225 |
sky | -0.214 |
end | -0.206 |
best | -0.204 |
Token | . |
Token ID | 13 |
Feature activation | -3.84e-02 |
Pos logit contributions | |
snatch | +0.076 |
zip | +0.076 |
ck | +0.076 |
Flo | +0.075 |
unw | +0.073 |
Neg logit contributions | |
End | -0.049 |
problem | -0.046 |
sky | -0.044 |
happiest | -0.040 |
end | -0.039 |
Token | The |
Token ID | 383 |
Feature activation | +3.33e-02 |
Pos logit contributions | |
End | +0.051 |
sky | +0.050 |
problem | +0.046 |
best | +0.041 |
end | +0.041 |
Neg logit contributions | |
ck | -0.079 |
snatch | -0.075 |
zip | -0.074 |
zzy | -0.071 |
unw | -0.070 |
Token | End |
Token ID | 5268 |
Feature activation | -3.94e-01 |
Pos logit contributions | |
snatch | +0.064 |
Flo | +0.063 |
ck | +0.063 |
zip | +0.063 |
unw | +0.060 |
Neg logit contributions | |
End | -0.043 |
problem | -0.042 |
sky | -0.039 |
happiest | -0.037 |
gloomy | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | -3.73e-02 |
Pos logit contributions | |
End | +0.444 |
problem | +0.442 |
sky | +0.421 |
end | +0.370 |
best | +0.367 |
Neg logit contributions | |
snatch | -0.845 |
zip | -0.838 |
ck | -0.837 |
Flo | -0.826 |
unw | -0.792 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.46e-02 |
Pos logit contributions | |
End | +0.052 |
sky | +0.049 |
problem | +0.044 |
end | +0.043 |
day | +0.043 |
Neg logit contributions | |
ck | -0.076 |
snatch | -0.075 |
zip | -0.073 |
Flo | -0.069 |
zzy | -0.069 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
End | +0.046 |
sky | +0.046 |
problem | +0.043 |
end | +0.039 |
day | +0.038 |
Neg logit contributions | |
ck | -0.070 |
snatch | -0.069 |
Flo | -0.066 |
zip | -0.066 |
unw | -0.064 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
uzz | +0.097 |
ck | +0.095 |
zip | +0.093 |
Flo | +0.091 |
zzy | +0.090 |
Neg logit contributions | |
End | -0.225 |
sky | -0.218 |
happiest | -0.217 |
problem | -0.215 |
sounds | -0.206 |
Token | a |
Token ID | 257 |
Feature activation | +2.04e-03 |
Pos logit contributions | |
snatch | +0.223 |
zip | +0.221 |
ck | +0.220 |
Flo | +0.218 |
zzy | +0.210 |
Neg logit contributions | |
End | -0.143 |
sky | -0.134 |
problem | -0.133 |
end | -0.121 |
gloomy | -0.121 |
Token | . |
Token ID | 13 |
Feature activation | -4.61e-02 |
Pos logit contributions | |
ck | +0.191 |
snatch | +0.187 |
Flo | +0.186 |
unw | +0.185 |
zip | +0.184 |
Neg logit contributions | |
End | -0.157 |
problem | -0.146 |
sky | -0.135 |
happiest | -0.131 |
end | -0.127 |
Token | |
Token ID | 220 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
End | +0.065 |
sky | +0.063 |
problem | +0.061 |
happiest | +0.055 |
best | +0.055 |
Neg logit contributions | |
ck | -0.088 |
snatch | -0.087 |
zip | -0.085 |
Flo | -0.084 |
zzy | -0.083 |
Token |
|
Token ID | 198 |
Feature activation | +5.39e-02 |
Pos logit contributions | |
sky | +0.167 |
End | +0.164 |
best | +0.150 |
problem | +0.149 |
happiest | +0.134 |
Neg logit contributions | |
snatch | -0.274 |
ck | -0.271 |
zip | -0.248 |
zzy | -0.247 |
Flo | -0.247 |
Token |
|
Token ID | 198 |
Feature activation | +5.32e-02 |
Pos logit contributions | |
snatch | +0.114 |
ck | +0.112 |
ried | +0.103 |
zip | +0.102 |
unw | +0.102 |
Neg logit contributions | |
End | -0.068 |
sky | -0.065 |
day | -0.053 |
end | -0.052 |
living | -0.049 |
Token | The |
Token ID | 464 |
Feature activation | +5.66e-03 |
Pos logit contributions | |
snatch | +0.077 |
ck | +0.076 |
zip | +0.074 |
Flo | +0.072 |
unw | +0.071 |
Neg logit contributions | |
End | -0.100 |
sky | -0.093 |
problem | -0.092 |
end | -0.088 |
happiest | -0.085 |
Token | End |
Token ID | 5268 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
snatch | +0.010 |
ck | +0.010 |
zip | +0.010 |
Flo | +0.010 |
unw | +0.010 |
Neg logit contributions | |
End | -0.008 |
problem | -0.008 |
sky | -0.007 |
happiest | -0.007 |
gloomy | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -4.80e-02 |
Pos logit contributions | |
End | +0.538 |
problem | +0.511 |
sky | +0.482 |
happiest | +0.462 |
end | +0.446 |
Neg logit contributions | |
zip | -0.750 |
ck | -0.743 |
Flo | -0.743 |
snatch | -0.742 |
zzy | -0.732 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.60e-02 |
Pos logit contributions | |
End | +0.068 |
sky | +0.066 |
problem | +0.057 |
end | +0.057 |
day | +0.055 |
Neg logit contributions | |
ck | -0.096 |
snatch | -0.096 |
zip | -0.092 |
zzy | -0.088 |
Flo | -0.088 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
End | +0.066 |
sky | +0.064 |
problem | +0.062 |
end | +0.056 |
best | +0.055 |
Neg logit contributions | |
ck | -0.086 |
snatch | -0.086 |
zip | -0.084 |
Flo | -0.083 |
zzy | -0.080 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
ck | +0.100 |
zip | +0.100 |
uzz | +0.098 |
zzy | +0.096 |
Flo | +0.095 |
Neg logit contributions | |
End | -0.213 |
sky | -0.201 |
problem | -0.200 |
happiest | -0.199 |
end | -0.187 |
Token | a |
Token ID | 257 |
Feature activation | -2.40e-04 |
Pos logit contributions | |
snatch | +0.187 |
zip | +0.187 |
ck | +0.185 |
Flo | +0.185 |
zzy | +0.178 |
Neg logit contributions | |
End | -0.145 |
problem | -0.139 |
sky | -0.133 |
end | -0.123 |
happiest | -0.122 |
Token | . |
Token ID | 13 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
zip | +0.189 |
snatch | +0.188 |
ck | +0.187 |
Flo | +0.185 |
zzy | +0.181 |
Neg logit contributions | |
End | -0.137 |
problem | -0.133 |
sky | -0.127 |
end | -0.118 |
best | -0.117 |
Token | |
Token ID | 220 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
End | +0.054 |
sky | +0.053 |
problem | +0.047 |
day | +0.044 |
best | +0.043 |
Neg logit contributions | |
ck | -0.079 |
snatch | -0.076 |
zip | -0.075 |
steer | -0.071 |
zzy | -0.070 |
Token |
|
Token ID | 198 |
Feature activation | +5.72e-02 |
Pos logit contributions | |
End | +0.167 |
sky | +0.159 |
problem | +0.152 |
best | +0.144 |
end | +0.138 |
Neg logit contributions | |
snatch | -0.234 |
ck | -0.229 |
zip | -0.221 |
unw | -0.215 |
Flo | -0.215 |
Token |
|
Token ID | 198 |
Feature activation | +5.42e-02 |
Pos logit contributions | |
snatch | +0.124 |
ck | +0.118 |
zip | +0.116 |
unw | +0.114 |
steer | +0.113 |
Neg logit contributions | |
End | -0.076 |
sky | -0.063 |
day | -0.058 |
end | -0.056 |
problem | -0.051 |
Token | The |
Token ID | 464 |
Feature activation | +7.53e-03 |
Pos logit contributions | |
snatch | +0.092 |
ck | +0.088 |
zip | +0.086 |
unw | +0.085 |
onna | +0.085 |
Neg logit contributions | |
End | -0.096 |
sky | -0.078 |
end | -0.076 |
day | -0.074 |
problem | -0.070 |
Token | End |
Token ID | 5268 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
snatch | +0.016 |
ck | +0.016 |
Flo | +0.016 |
zip | +0.016 |
unw | +0.015 |
Neg logit contributions | |
End | -0.008 |
problem | -0.007 |
sky | -0.007 |
happiest | -0.006 |
gloomy | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -4.37e-02 |
Pos logit contributions | |
End | +0.479 |
problem | +0.463 |
sky | +0.440 |
end | +0.397 |
best | +0.393 |
Neg logit contributions | |
snatch | -0.809 |
zip | -0.806 |
ck | -0.803 |
Flo | -0.796 |
unw | -0.765 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.33e-02 |
Pos logit contributions | |
End | +0.057 |
sky | +0.054 |
end | +0.046 |
day | +0.046 |
problem | +0.044 |
Neg logit contributions | |
ck | -0.096 |
snatch | -0.094 |
zip | -0.090 |
steer | -0.086 |
zzy | -0.086 |
Token | Once |
Token ID | 7454 |
Feature activation | +9.82e-02 |
Pos logit contributions | |
End | +0.054 |
sky | +0.051 |
problem | +0.050 |
end | +0.045 |
best | +0.044 |
Neg logit contributions | |
snatch | -0.089 |
ck | -0.089 |
zip | -0.087 |
Flo | -0.087 |
unw | -0.084 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
uzz | +0.093 |
ck | +0.090 |
zip | +0.089 |
Flo | +0.088 |
zzy | +0.088 |
Neg logit contributions | |
End | -0.211 |
problem | -0.202 |
sky | -0.200 |
happiest | -0.200 |
sounds | -0.190 |
Token | a |
Token ID | 257 |
Feature activation | -2.90e-03 |
Pos logit contributions | |
snatch | +0.203 |
zip | +0.202 |
ck | +0.200 |
Flo | +0.199 |
zzy | +0.191 |
Neg logit contributions | |
End | -0.141 |
problem | -0.133 |
sky | -0.131 |
end | -0.120 |
gloomy | -0.119 |
Token | . |
Token ID | 13 |
Feature activation | -4.73e-02 |
Pos logit contributions | |
snatch | +0.004 |
ck | +0.004 |
zip | +0.004 |
Flo | +0.004 |
unw | +0.004 |
Neg logit contributions | |
End | -0.003 |
problem | -0.003 |
sky | -0.003 |
happiest | -0.003 |
gloomy | -0.003 |
Token | |
Token ID | 220 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
End | +0.057 |
sky | +0.056 |
problem | +0.045 |
day | +0.043 |
best | +0.041 |
Neg logit contributions | |
ck | -0.106 |
snatch | -0.099 |
zip | -0.097 |
steer | -0.093 |
zzy | -0.091 |
Token |
|
Token ID | 198 |
Feature activation | +5.44e-02 |
Pos logit contributions | |
End | +0.175 |
sky | +0.166 |
problem | +0.160 |
best | +0.149 |
end | +0.142 |
Neg logit contributions | |
snatch | -0.245 |
ck | -0.242 |
zip | -0.231 |
Flo | -0.229 |
zzy | -0.226 |
Token |
|
Token ID | 198 |
Feature activation | +5.19e-02 |
Pos logit contributions | |
snatch | +0.121 |
ck | +0.116 |
ried | +0.115 |
rily | +0.114 |
unw | +0.113 |
Neg logit contributions | |
End | -0.067 |
sky | -0.052 |
day | -0.045 |
end | -0.044 |
much | -0.044 |
Token | The |
Token ID | 464 |
Feature activation | +2.31e-03 |
Pos logit contributions | |
snatch | +0.088 |
ried | +0.085 |
unw | +0.084 |
rily | +0.084 |
ck | +0.083 |
Neg logit contributions | |
End | -0.091 |
end | -0.071 |
sky | -0.070 |
day | -0.066 |
much | -0.065 |
Token | End |
Token ID | 5268 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
snatch | +0.005 |
ck | +0.005 |
Flo | +0.005 |
zip | +0.005 |
unw | +0.005 |
Neg logit contributions | |
End | -0.002 |
problem | -0.002 |
sky | -0.002 |
happiest | -0.002 |
gloomy | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
End | +0.483 |
problem | +0.470 |
sky | +0.444 |
end | +0.401 |
best | +0.398 |
Neg logit contributions | |
snatch | -0.807 |
zip | -0.805 |
ck | -0.801 |
Flo | -0.791 |
unw | -0.761 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
End | +0.050 |
sky | +0.046 |
day | +0.037 |
end | +0.036 |
best | +0.034 |
Neg logit contributions | |
ck | -0.107 |
snatch | -0.104 |
zip | -0.099 |
unw | -0.097 |
zzy | -0.096 |
Token | J |
Token ID | 41 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
End | +0.054 |
problem | +0.051 |
sky | +0.050 |
end | +0.045 |
best | +0.044 |
Neg logit contributions | |
snatch | -0.085 |
ck | -0.084 |
zip | -0.084 |
Flo | -0.083 |
zzy | -0.080 |
Token | ill |
Token ID | 359 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
snatch | +0.119 |
Flo | +0.116 |
zip | +0.116 |
ck | +0.116 |
zzy | +0.105 |
Neg logit contributions | |
End | -0.217 |
problem | -0.212 |
sky | -0.207 |
end | -0.196 |
best | -0.194 |
Token | walked |
Token ID | 6807 |
Feature activation | +9.38e-02 |
Pos logit contributions | |
ck | +0.411 |
zip | +0.411 |
snatch | +0.408 |
Flo | +0.404 |
zzy | +0.396 |
Neg logit contributions | |
End | -0.315 |
problem | -0.304 |
sky | -0.283 |
happiest | -0.266 |
gloomy | -0.264 |
Token | belonged |
Token ID | 19611 |
Feature activation | -8.23e-02 |
Pos logit contributions | |
End | +0.122 |
problem | +0.120 |
sky | +0.110 |
gloomy | +0.100 |
happiest | +0.098 |
Neg logit contributions | |
ck | -0.171 |
snatch | -0.170 |
zip | -0.168 |
zzy | -0.168 |
Flo | -0.167 |
Token | . |
Token ID | 13 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
End | +0.119 |
problem | +0.119 |
sky | +0.115 |
best | +0.110 |
gloomy | +0.106 |
Neg logit contributions | |
ck | -0.150 |
snatch | -0.149 |
zip | -0.146 |
Flo | -0.145 |
zzy | -0.139 |
Token |
|
Token ID | 198 |
Feature activation | +6.38e-02 |
Pos logit contributions | |
sky | +0.046 |
End | +0.045 |
day | +0.037 |
problem | +0.037 |
gloomy | +0.035 |
Neg logit contributions | |
ck | -0.079 |
snatch | -0.073 |
zip | -0.071 |
steer | -0.067 |
ge | -0.067 |
Token |
|
Token ID | 198 |
Feature activation | +6.28e-02 |
Pos logit contributions | |
snatch | +0.149 |
ck | +0.138 |
unw | +0.137 |
zip | +0.135 |
ried | +0.135 |
Neg logit contributions | |
End | -0.075 |
sky | -0.061 |
day | -0.055 |
end | -0.055 |
problem | -0.050 |
Token | The |
Token ID | 464 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
snatch | +0.106 |
ck | +0.102 |
unw | +0.101 |
rily | +0.099 |
onna | +0.098 |
Neg logit contributions | |
End | -0.110 |
end | -0.090 |
sky | -0.090 |
day | -0.086 |
problem | -0.083 |
Token | End |
Token ID | 5268 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
snatch | +0.022 |
zip | +0.022 |
ck | +0.022 |
Flo | +0.022 |
unw | +0.021 |
Neg logit contributions | |
End | -0.014 |
problem | -0.014 |
sky | -0.013 |
happiest | -0.012 |
gloomy | -0.012 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
End | +0.460 |
problem | +0.444 |
sky | +0.420 |
end | +0.378 |
best | +0.373 |
Neg logit contributions | |
snatch | -0.826 |
zip | -0.824 |
ck | -0.820 |
Flo | -0.814 |
unw | -0.783 |
Token | George |
Token ID | 20191 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
End | +0.047 |
sky | +0.046 |
problem | +0.043 |
end | +0.039 |
best | +0.037 |
Neg logit contributions | |
snatch | -0.086 |
ck | -0.086 |
zip | -0.083 |
Flo | -0.082 |
unw | -0.080 |
Token | and |
Token ID | 290 |
Feature activation | +8.17e-02 |
Pos logit contributions | |
snatch | +0.346 |
zip | +0.344 |
Flo | +0.343 |
ck | +0.338 |
zzy | +0.323 |
Neg logit contributions | |
End | -0.211 |
problem | -0.201 |
sky | -0.199 |
end | -0.182 |
best | -0.173 |
Token | Tim |
Token ID | 5045 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
snatch | +0.111 |
zip | +0.111 |
ck | +0.110 |
Flo | +0.110 |
zzy | +0.104 |
Neg logit contributions | |
End | -0.157 |
problem | -0.153 |
sky | -0.147 |
end | -0.139 |
happiest | -0.138 |
Token | my |
Token ID | 1820 |
Feature activation | +9.15e-02 |
Pos logit contributions | |
snatch | +0.271 |
zip | +0.270 |
ck | +0.268 |
Flo | +0.267 |
zzy | +0.255 |
Neg logit contributions | |
End | -0.299 |
problem | -0.290 |
sky | -0.279 |
end | -0.262 |
happiest | -0.260 |
Token | . |
Token ID | 13 |
Feature activation | -4.74e-02 |
Pos logit contributions | |
snatch | +0.118 |
zip | +0.118 |
ck | +0.117 |
Flo | +0.116 |
zzy | +0.112 |
Neg logit contributions | |
End | -0.063 |
problem | -0.059 |
sky | -0.057 |
end | -0.052 |
best | -0.051 |
Token | |
Token ID | 220 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
End | +0.063 |
sky | +0.062 |
best | +0.053 |
problem | +0.052 |
happiest | +0.051 |
Neg logit contributions | |
ck | -0.100 |
snatch | -0.094 |
zzy | -0.092 |
zip | -0.091 |
unw | -0.088 |
Token |
|
Token ID | 198 |
Feature activation | +5.41e-02 |
Pos logit contributions | |
End | +0.159 |
sky | +0.158 |
best | +0.144 |
problem | +0.139 |
happiest | +0.131 |
Neg logit contributions | |
snatch | -0.272 |
ck | -0.270 |
unw | -0.249 |
zzy | -0.248 |
zip | -0.248 |
Token |
|
Token ID | 198 |
Feature activation | +5.38e-02 |
Pos logit contributions | |
snatch | +0.119 |
ck | +0.116 |
unw | +0.111 |
ried | +0.109 |
zip | +0.108 |
Neg logit contributions | |
End | -0.066 |
sky | -0.057 |
end | -0.049 |
day | -0.048 |
best | -0.047 |
Token | The |
Token ID | 464 |
Feature activation | +6.03e-03 |
Pos logit contributions | |
snatch | +0.092 |
ck | +0.090 |
unw | +0.088 |
zip | +0.085 |
uzz | +0.085 |
Neg logit contributions | |
End | -0.092 |
sky | -0.077 |
end | -0.076 |
best | -0.070 |
day | -0.069 |
Token | End |
Token ID | 5268 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
snatch | +0.010 |
Flo | +0.010 |
ck | +0.010 |
zip | +0.009 |
zzy | +0.009 |
Neg logit contributions | |
problem | -0.009 |
End | -0.009 |
sky | -0.009 |
gloomy | -0.008 |
happiest | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
End | +0.538 |
problem | +0.513 |
sky | +0.490 |
happiest | +0.460 |
end | +0.449 |
Neg logit contributions | |
zip | -0.745 |
Flo | -0.743 |
snatch | -0.741 |
ck | -0.740 |
zzy | -0.727 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
End | +0.057 |
sky | +0.054 |
best | +0.044 |
end | +0.043 |
day | +0.043 |
Neg logit contributions | |
ck | -0.111 |
snatch | -0.109 |
zip | -0.103 |
unw | -0.102 |
zzy | -0.101 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
End | +0.054 |
problem | +0.052 |
sky | +0.051 |
end | +0.045 |
best | +0.045 |
Neg logit contributions | |
snatch | -0.087 |
ck | -0.087 |
zip | -0.086 |
Flo | -0.085 |
zzy | -0.082 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
ck | +0.112 |
zip | +0.110 |
Flo | +0.108 |
uzz | +0.108 |
zzy | +0.107 |
Neg logit contributions | |
End | -0.215 |
problem | -0.204 |
sky | -0.204 |
happiest | -0.198 |
end | -0.189 |
Token | a |
Token ID | 257 |
Feature activation | -2.87e-04 |
Pos logit contributions | |
snatch | +0.196 |
zip | +0.195 |
ck | +0.194 |
Flo | +0.193 |
zzy | +0.186 |
Neg logit contributions | |
End | -0.143 |
problem | -0.136 |
sky | -0.132 |
end | -0.122 |
gloomy | -0.120 |
Token | night |
Token ID | 1755 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
End | +0.119 |
problem | +0.101 |
happiest | +0.089 |
sky | +0.086 |
best | +0.083 |
Neg logit contributions | |
snatch | -0.239 |
zip | -0.236 |
unw | -0.235 |
Flo | -0.232 |
ck | -0.231 |
Token | . |
Token ID | 13 |
Feature activation | -4.09e-02 |
Pos logit contributions | |
End | +0.430 |
problem | +0.419 |
happiest | +0.358 |
best | +0.349 |
end | +0.343 |
Neg logit contributions | |
unw | -0.487 |
onna | -0.460 |
ck | -0.457 |
Flo | -0.455 |
steer | -0.452 |
Token |
|
Token ID | 198 |
Feature activation | +5.81e-02 |
Pos logit contributions | |
End | +0.061 |
sky | +0.058 |
problem | +0.055 |
end | +0.051 |
day | +0.050 |
Neg logit contributions | |
ck | -0.077 |
snatch | -0.076 |
zip | -0.074 |
Flo | -0.072 |
unw | -0.071 |
Token |
|
Token ID | 198 |
Feature activation | +5.62e-02 |
Pos logit contributions | |
snatch | +0.120 |
zip | +0.113 |
ck | +0.112 |
unw | +0.112 |
Flo | +0.108 |
Neg logit contributions | |
End | -0.084 |
sky | -0.072 |
end | -0.067 |
problem | -0.066 |
day | -0.066 |
Token | The |
Token ID | 464 |
Feature activation | +5.79e-03 |
Pos logit contributions | |
snatch | +0.093 |
ck | +0.088 |
unw | +0.088 |
zip | +0.088 |
rily | +0.085 |
Neg logit contributions | |
End | -0.103 |
end | -0.086 |
sky | -0.085 |
day | -0.081 |
problem | -0.081 |
Token | End |
Token ID | 5268 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
snatch | +0.013 |
ck | +0.013 |
Flo | +0.013 |
zip | +0.012 |
unw | +0.012 |
Neg logit contributions | |
End | -0.005 |
problem | -0.005 |
sky | -0.005 |
happiest | -0.004 |
best | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -4.16e-02 |
Pos logit contributions | |
End | +0.496 |
problem | +0.473 |
sky | +0.448 |
happiest | +0.410 |
end | +0.410 |
Neg logit contributions | |
zip | -0.788 |
snatch | -0.786 |
ck | -0.781 |
Flo | -0.780 |
zzy | -0.758 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
End | +0.056 |
sky | +0.053 |
end | +0.045 |
day | +0.045 |
problem | +0.045 |
Neg logit contributions | |
ck | -0.089 |
snatch | -0.088 |
zip | -0.083 |
unw | -0.080 |
Flo | -0.080 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
End | +0.049 |
sky | +0.046 |
problem | +0.046 |
end | +0.040 |
best | +0.040 |
Neg logit contributions | |
snatch | -0.081 |
ck | -0.081 |
zip | -0.079 |
Flo | -0.079 |
unw | -0.076 |
Token | there |
Token ID | 612 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
uzz | +0.098 |
zip | +0.098 |
ck | +0.097 |
Flo | +0.096 |
zzy | +0.095 |
Neg logit contributions | |
End | -0.217 |
problem | -0.206 |
sky | -0.206 |
happiest | -0.203 |
end | -0.192 |
Token | was |
Token ID | 373 |
Feature activation | -5.79e-02 |
Pos logit contributions | |
snatch | +0.222 |
zip | +0.222 |
Flo | +0.221 |
ck | +0.218 |
unw | +0.214 |
Neg logit contributions | |
End | -0.127 |
problem | -0.124 |
sky | -0.116 |
gloomy | -0.103 |
happiest | -0.103 |
Token | again |
Token ID | 757 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
snatch | +0.113 |
zip | +0.113 |
ck | +0.112 |
Flo | +0.111 |
zzy | +0.106 |
Neg logit contributions | |
End | -0.115 |
problem | -0.111 |
sky | -0.107 |
end | -0.099 |
happiest | -0.099 |
Token | . |
Token ID | 13 |
Feature activation | -4.43e-02 |
Pos logit contributions | |
snatch | +0.211 |
ck | +0.210 |
zip | +0.210 |
Flo | +0.208 |
zzy | +0.203 |
Neg logit contributions | |
End | -0.136 |
problem | -0.130 |
sky | -0.124 |
end | -0.111 |
happiest | -0.111 |
Token |
|
Token ID | 198 |
Feature activation | +6.02e-02 |
Pos logit contributions | |
problem | +0.077 |
End | +0.077 |
happiest | +0.068 |
sky | +0.067 |
end | +0.067 |
Neg logit contributions | |
zip | -0.068 |
Flo | -0.066 |
snatch | -0.065 |
unw | -0.064 |
zzy | -0.063 |
Token |
|
Token ID | 198 |
Feature activation | +5.90e-02 |
Pos logit contributions | |
snatch | +0.107 |
ck | +0.104 |
zip | +0.104 |
Flo | +0.103 |
unw | +0.098 |
Neg logit contributions | |
End | -0.094 |
sky | -0.085 |
problem | -0.085 |
end | -0.082 |
day | -0.076 |
Token | The |
Token ID | 464 |
Feature activation | +9.81e-03 |
Pos logit contributions | |
snatch | +0.081 |
ck | +0.080 |
zip | +0.079 |
Flo | +0.078 |
unw | +0.074 |
Neg logit contributions | |
End | -0.114 |
problem | -0.108 |
sky | -0.106 |
end | -0.102 |
happiest | -0.098 |
Token | End |
Token ID | 5268 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
snatch | +0.018 |
zip | +0.016 |
ck | +0.016 |
steer | +0.016 |
persuade | +0.016 |
Neg logit contributions | |
problem | -0.011 |
sizes | -0.011 |
sky | -0.011 |
moons | -0.011 |
jets | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -4.54e-02 |
Pos logit contributions | |
End | +0.547 |
problem | +0.515 |
sky | +0.492 |
happiest | +0.480 |
end | +0.449 |
Neg logit contributions | |
zip | -0.735 |
Flo | -0.729 |
zzy | -0.728 |
ck | -0.728 |
snatch | -0.725 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.25e-02 |
Pos logit contributions | |
End | +0.065 |
sky | +0.063 |
end | +0.056 |
problem | +0.056 |
day | +0.053 |
Neg logit contributions | |
ck | -0.093 |
snatch | -0.090 |
zip | -0.087 |
Flo | -0.084 |
steer | -0.084 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
sky | +0.061 |
End | +0.060 |
problem | +0.055 |
end | +0.052 |
day | +0.051 |
Neg logit contributions | |
ck | -0.083 |
snatch | -0.082 |
Flo | -0.078 |
zip | -0.077 |
steer | -0.074 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
zip | +0.108 |
ck | +0.107 |
uzz | +0.105 |
Flo | +0.103 |
zzy | +0.102 |
Neg logit contributions | |
End | -0.220 |
sky | -0.207 |
problem | -0.205 |
happiest | -0.205 |
gloomy | -0.191 |
Token | a |
Token ID | 257 |
Feature activation | +3.03e-03 |
Pos logit contributions | |
zip | +0.189 |
snatch | +0.189 |
ck | +0.187 |
Flo | +0.187 |
zzy | +0.180 |
Neg logit contributions | |
End | -0.153 |
problem | -0.148 |
sky | -0.141 |
end | -0.130 |
happiest | -0.130 |
Token | . |
Token ID | 13 |
Feature activation | -4.98e-02 |
Pos logit contributions | |
ck | +0.223 |
Flo | +0.223 |
snatch | +0.222 |
zip | +0.220 |
unw | +0.215 |
Neg logit contributions | |
End | -0.191 |
problem | -0.181 |
sky | -0.170 |
gloomy | -0.161 |
end | -0.157 |
Token | |
Token ID | 220 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
End | +0.067 |
sky | +0.065 |
problem | +0.058 |
best | +0.056 |
end | +0.055 |
Neg logit contributions | |
ck | -0.101 |
snatch | -0.099 |
zip | -0.097 |
zzy | -0.093 |
unw | -0.092 |
Token |
|
Token ID | 198 |
Feature activation | +5.60e-02 |
Pos logit contributions | |
End | +0.169 |
sky | +0.165 |
best | +0.152 |
problem | +0.147 |
happiest | +0.139 |
Neg logit contributions | |
snatch | -0.272 |
ck | -0.265 |
zip | -0.247 |
unw | -0.246 |
steer | -0.243 |
Token |
|
Token ID | 198 |
Feature activation | +5.47e-02 |
Pos logit contributions | |
snatch | +0.106 |
ck | +0.102 |
zip | +0.100 |
unw | +0.097 |
zzy | +0.095 |
Neg logit contributions | |
End | -0.084 |
sky | -0.076 |
end | -0.069 |
problem | -0.068 |
best | -0.068 |
Token | The |
Token ID | 464 |
Feature activation | +7.95e-03 |
Pos logit contributions | |
snatch | +0.081 |
zip | +0.078 |
ck | +0.078 |
unw | +0.074 |
uzz | +0.074 |
Neg logit contributions | |
End | -0.103 |
sky | -0.093 |
problem | -0.092 |
end | -0.090 |
happiest | -0.087 |
Token | End |
Token ID | 5268 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
snatch | +0.016 |
Flo | +0.016 |
ck | +0.016 |
zip | +0.016 |
unw | +0.015 |
Neg logit contributions | |
problem | -0.008 |
sky | -0.008 |
End | -0.008 |
starts | -0.007 |
happens | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -5.02e-02 |
Pos logit contributions | |
End | +0.537 |
problem | +0.507 |
sky | +0.482 |
happiest | +0.466 |
end | +0.442 |
Neg logit contributions | |
zip | -0.743 |
ck | -0.741 |
Flo | -0.741 |
snatch | -0.735 |
zzy | -0.734 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.35e-02 |
Pos logit contributions | |
End | +0.064 |
sky | +0.063 |
end | +0.054 |
best | +0.052 |
day | +0.051 |
Neg logit contributions | |
snatch | -0.109 |
ck | -0.109 |
zip | -0.104 |
unw | -0.099 |
zzy | -0.099 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
End | +0.060 |
problem | +0.058 |
sky | +0.056 |
end | +0.051 |
best | +0.050 |
Neg logit contributions | |
snatch | -0.082 |
ck | -0.082 |
zip | -0.082 |
Flo | -0.081 |
zzy | -0.078 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
ck | +0.103 |
uzz | +0.102 |
zip | +0.101 |
Flo | +0.100 |
zzy | +0.099 |
Neg logit contributions | |
End | -0.218 |
sky | -0.206 |
problem | -0.204 |
happiest | -0.201 |
sounds | -0.191 |
Token | a |
Token ID | 257 |
Feature activation | -9.68e-05 |
Pos logit contributions | |
snatch | +0.187 |
zip | +0.187 |
ck | +0.185 |
Flo | +0.185 |
zzy | +0.178 |
Neg logit contributions | |
End | -0.147 |
problem | -0.142 |
sky | -0.135 |
end | -0.125 |
happiest | -0.124 |
Token | to |
Token ID | 284 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
End | +0.050 |
problem | +0.046 |
sky | +0.045 |
happiest | +0.042 |
gloomy | +0.041 |
Neg logit contributions | |
snatch | -0.063 |
ck | -0.063 |
zip | -0.063 |
Flo | -0.063 |
unw | -0.061 |
Token | play |
Token ID | 711 |
Feature activation | +6.00e-03 |
Pos logit contributions | |
rily | +0.256 |
zzy | +0.251 |
xc | +0.249 |
zip | +0.245 |
uzz | +0.242 |
Neg logit contributions | |
end | -0.229 |
problem | -0.207 |
day | -0.198 |
gloomy | -0.195 |
End | -0.192 |
Token | ." |
Token ID | 526 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
zip | +0.009 |
snatch | +0.009 |
Flo | +0.009 |
ck | +0.009 |
unw | +0.008 |
Neg logit contributions | |
End | -0.011 |
problem | -0.010 |
sky | -0.010 |
happiest | -0.009 |
gloomy | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +7.78e-02 |
Pos logit contributions | |
sky | +0.014 |
End | +0.014 |
problem | +0.012 |
gloomy | +0.011 |
end | +0.011 |
Neg logit contributions | |
ck | -0.028 |
snatch | -0.027 |
zip | -0.026 |
Flo | -0.025 |
ried | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +7.61e-02 |
Pos logit contributions | |
snatch | +0.157 |
zip | +0.151 |
ck | +0.148 |
Flo | +0.146 |
zzy | +0.145 |
Neg logit contributions | |
End | -0.107 |
sky | -0.097 |
problem | -0.093 |
end | -0.092 |
gloomy | -0.085 |
Token | L |
Token ID | 43 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
snatch | +0.107 |
ck | +0.099 |
zip | +0.098 |
unw | +0.094 |
uzz | +0.092 |
Neg logit contributions | |
End | -0.158 |
end | -0.142 |
sky | -0.140 |
problem | -0.137 |
gloomy | -0.131 |
Token | ily |
Token ID | 813 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
zip | +0.306 |
snatch | +0.306 |
Flo | +0.299 |
ck | +0.296 |
unw | +0.285 |
Neg logit contributions | |
End | -0.308 |
problem | -0.293 |
sky | -0.279 |
end | -0.274 |
best | -0.260 |
Token | nodded |
Token ID | 14464 |
Feature activation | +8.27e-02 |
Pos logit contributions | |
snatch | +0.250 |
zip | +0.249 |
Flo | +0.249 |
ck | +0.246 |
uzz | +0.240 |
Neg logit contributions | |
End | -0.140 |
problem | -0.129 |
sky | -0.129 |
end | -0.118 |
best | -0.112 |
Token | and |
Token ID | 290 |
Feature activation | +7.32e-02 |
Pos logit contributions | |
zip | +0.162 |
snatch | +0.161 |
Flo | +0.159 |
ck | +0.157 |
zzy | +0.154 |
Neg logit contributions | |
End | -0.114 |
problem | -0.106 |
sky | -0.105 |
end | -0.101 |
gloomy | -0.095 |
Token | said |
Token ID | 531 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
snatch | +0.130 |
zip | +0.130 |
Flo | +0.128 |
ck | +0.126 |
zzy | +0.123 |
Neg logit contributions | |
End | -0.103 |
problem | -0.099 |
sky | -0.098 |
best | -0.095 |
happiest | -0.093 |
Token | , |
Token ID | 11 |
Feature activation | -9.27e-02 |
Pos logit contributions | |
End | +0.035 |
sky | +0.033 |
problem | +0.032 |
happiest | +0.030 |
gloomy | +0.028 |
Neg logit contributions | |
Flo | -0.065 |
snatch | -0.063 |
ck | -0.063 |
zip | -0.062 |
uzz | -0.061 |
Token | it |
Token ID | 340 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
zzy | +0.133 |
zip | +0.130 |
Flo | +0.130 |
snatch | +0.130 |
ck | +0.125 |
Neg logit contributions | |
problem | -0.145 |
End | -0.143 |
end | -0.141 |
sky | -0.140 |
best | -0.137 |
Token | with |
Token ID | 351 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
End | +0.174 |
problem | +0.171 |
sky | +0.161 |
end | +0.160 |
best | +0.159 |
Neg logit contributions | |
snatch | -0.194 |
zip | -0.194 |
Flo | -0.189 |
ck | -0.189 |
zzy | -0.188 |
Token | his |
Token ID | 465 |
Feature activation | +4.06e-02 |
Pos logit contributions | |
zip | +0.331 |
snatch | +0.330 |
ck | +0.328 |
Flo | +0.327 |
zzy | +0.312 |
Neg logit contributions | |
End | -0.323 |
problem | -0.313 |
sky | -0.301 |
happiest | -0.278 |
end | -0.278 |
Token | friends |
Token ID | 2460 |
Feature activation | +2.31e-04 |
Pos logit contributions | |
ck | +0.075 |
Flo | +0.075 |
snatch | +0.075 |
zip | +0.074 |
unw | +0.071 |
Neg logit contributions | |
End | -0.056 |
sky | -0.050 |
problem | -0.050 |
happiest | -0.047 |
end | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
snatch | +0.000 |
ck | +0.000 |
zip | +0.000 |
Flo | +0.000 |
unw | +0.000 |
Neg logit contributions | |
End | -0.000 |
problem | -0.000 |
sky | -0.000 |
happiest | -0.000 |
gloomy | -0.000 |
Token | Tim |
Token ID | 5045 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
sky | +0.037 |
End | +0.037 |
problem | +0.034 |
day | +0.032 |
best | +0.032 |
Neg logit contributions | |
ck | -0.056 |
snatch | -0.055 |
zzy | -0.053 |
zip | -0.052 |
Flo | -0.051 |
Token | wanted |
Token ID | 2227 |
Feature activation | -5.64e-02 |
Pos logit contributions | |
ck | +0.352 |
zip | +0.347 |
Flo | +0.343 |
snatch | +0.338 |
unw | +0.331 |
Neg logit contributions | |
End | -0.231 |
problem | -0.222 |
sky | -0.211 |
gloomy | -0.191 |
happiest | -0.185 |
Token | to |
Token ID | 284 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
problem | +0.084 |
End | +0.084 |
sky | +0.082 |
end | +0.076 |
best | +0.076 |
Neg logit contributions | |
zip | -0.100 |
Flo | -0.100 |
ck | -0.100 |
snatch | -0.100 |
zzy | -0.097 |
Token | test |
Token ID | 1332 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
uzz | +0.287 |
rily | +0.286 |
Flo | +0.286 |
igi | +0.286 |
ried | +0.281 |
Neg logit contributions | |
end | -0.216 |
sky | -0.189 |
problem | -0.185 |
End | -0.182 |
day | -0.182 |
Token | his |
Token ID | 465 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
zip | +0.032 |
snatch | +0.032 |
Flo | +0.032 |
ck | +0.031 |
zzy | +0.031 |
Neg logit contributions | |
End | -0.034 |
problem | -0.034 |
sky | -0.032 |
end | -0.030 |
best | -0.030 |
Token | story |
Token ID | 1621 |
Feature activation | -4.12e-02 |
Pos logit contributions | |
snatch | +0.075 |
ck | +0.074 |
zip | +0.072 |
Flo | +0.071 |
unw | +0.071 |
Neg logit contributions | |
End | -0.055 |
sky | -0.048 |
problem | -0.046 |
happiest | -0.044 |
end | -0.043 |
Token | kins |
Token ID | 5331 |
Feature activation | -6.41e-02 |
Pos logit contributions | |
snatch | +0.165 |
zip | +0.164 |
ck | +0.163 |
Flo | +0.162 |
zzy | +0.155 |
Neg logit contributions | |
End | -0.130 |
problem | -0.125 |
sky | -0.119 |
best | -0.112 |
end | -0.111 |
Token | . |
Token ID | 13 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
End | +0.087 |
problem | +0.083 |
sky | +0.079 |
end | +0.072 |
happiest | +0.072 |
Neg logit contributions | |
zip | -0.122 |
snatch | -0.121 |
ck | -0.121 |
Flo | -0.120 |
zzy | -0.117 |
Token | |
Token ID | 220 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
sky | +0.050 |
End | +0.048 |
problem | +0.045 |
day | +0.042 |
gloomy | +0.042 |
Neg logit contributions | |
ck | -0.070 |
snatch | -0.066 |
zzy | -0.063 |
zip | -0.062 |
steer | -0.061 |
Token |
|
Token ID | 198 |
Feature activation | +6.91e-02 |
Pos logit contributions | |
End | +0.149 |
sky | +0.143 |
best | +0.133 |
problem | +0.133 |
end | +0.122 |
Neg logit contributions | |
snatch | -0.256 |
ck | -0.247 |
zip | -0.233 |
unw | -0.232 |
steer | -0.229 |
Token |
|
Token ID | 198 |
Feature activation | +6.69e-02 |
Pos logit contributions | |
snatch | +0.172 |
ck | +0.171 |
steer | +0.152 |
ried | +0.151 |
rily | +0.150 |
Neg logit contributions | |
sky | -0.063 |
End | -0.055 |
day | -0.052 |
s | -0.049 |
best | -0.047 |
Token | L |
Token ID | 43 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
ck | +0.104 |
snatch | +0.104 |
zzy | +0.095 |
onna | +0.095 |
uzz | +0.093 |
Neg logit contributions | |
End | -0.115 |
sky | -0.104 |
end | -0.104 |
problem | -0.101 |
day | -0.100 |
Token | ily |
Token ID | 813 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
snatch | +0.329 |
Flo | +0.322 |
zip | +0.321 |
steer | +0.312 |
ck | +0.312 |
Neg logit contributions | |
End | -0.286 |
problem | -0.264 |
end | -0.250 |
sky | -0.242 |
best | -0.233 |
Token | 's |
Token ID | 338 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
zip | +0.232 |
snatch | +0.231 |
Flo | +0.227 |
ck | +0.226 |
zzy | +0.216 |
Neg logit contributions | |
End | -0.164 |
problem | -0.158 |
sky | -0.151 |
end | -0.141 |
best | -0.138 |
Token | mom |
Token ID | 1995 |
Feature activation | -5.17e-03 |
Pos logit contributions | |
End | +0.217 |
problem | +0.209 |
sky | +0.202 |
happiest | +0.186 |
gloomy | +0.185 |
Neg logit contributions | |
snatch | -0.225 |
zip | -0.222 |
ck | -0.222 |
Flo | -0.220 |
unw | -0.214 |
Token | said |
Token ID | 531 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
End | +0.007 |
problem | +0.007 |
sky | +0.006 |
end | +0.006 |
best | +0.006 |
Neg logit contributions | |
Flo | -0.010 |
zip | -0.010 |
snatch | -0.010 |
ck | -0.010 |
zzy | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -8.76e-02 |
Pos logit contributions | |
End | +0.039 |
sky | +0.037 |
problem | +0.037 |
happiest | +0.033 |
end | +0.032 |
Neg logit contributions | |
Flo | -0.058 |
snatch | -0.057 |
ck | -0.056 |
unw | -0.056 |
zzy | -0.054 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
ck | +0.105 |
snatch | +0.103 |
zip | +0.101 |
Flo | +0.099 |
unw | +0.099 |
Neg logit contributions | |
End | -0.088 |
problem | -0.081 |
sky | -0.079 |
end | -0.070 |
happiest | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
End | +0.210 |
problem | +0.193 |
sky | +0.184 |
happiest | +0.177 |
gloomy | +0.167 |
Neg logit contributions | |
snatch | -0.234 |
ck | -0.233 |
Flo | -0.228 |
zip | -0.227 |
unw | -0.227 |
Token | She |
Token ID | 1375 |
Feature activation | +5.30e-02 |
Pos logit contributions | |
End | +0.043 |
sky | +0.041 |
problem | +0.040 |
end | +0.036 |
best | +0.036 |
Neg logit contributions | |
ck | -0.061 |
snatch | -0.061 |
zip | -0.059 |
Flo | -0.058 |
zzy | -0.057 |
Token | wrote |
Token ID | 2630 |
Feature activation | +6.86e-02 |
Pos logit contributions | |
snatch | +0.108 |
zip | +0.108 |
Flo | +0.107 |
izer | +0.105 |
uzz | +0.103 |
Neg logit contributions | |
End | -0.062 |
problem | -0.061 |
sky | -0.059 |
end | -0.054 |
best | -0.054 |
Token | her |
Token ID | 607 |
Feature activation | -3.03e-02 |
Pos logit contributions | |
zip | +0.112 |
snatch | +0.112 |
ck | +0.111 |
Flo | +0.110 |
zzy | +0.106 |
Neg logit contributions | |
End | -0.112 |
problem | -0.109 |
sky | -0.105 |
happiest | -0.097 |
end | -0.097 |
Token | name |
Token ID | 1438 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
End | +0.060 |
problem | +0.057 |
sky | +0.055 |
happiest | +0.053 |
gloomy | +0.052 |
Neg logit contributions | |
zip | -0.039 |
snatch | -0.039 |
ck | -0.039 |
Flo | -0.037 |
unw | -0.036 |
Token | on |
Token ID | 319 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
zip | +0.342 |
snatch | +0.339 |
Flo | +0.339 |
ck | +0.337 |
unw | +0.326 |
Neg logit contributions | |
End | -0.326 |
problem | -0.316 |
sky | -0.291 |
happiest | -0.279 |
gloomy | -0.275 |
Token | it |
Token ID | 340 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
snatch | +0.225 |
zip | +0.224 |
ck | +0.224 |
Flo | +0.220 |
zzy | +0.215 |
Neg logit contributions | |
End | -0.184 |
problem | -0.180 |
sky | -0.177 |
end | -0.163 |
best | -0.160 |
Token | . |
Token ID | 13 |
Feature activation | -3.37e-02 |
Pos logit contributions | |
End | +0.149 |
problem | +0.143 |
sky | +0.136 |
end | +0.124 |
happiest | +0.123 |
Neg logit contributions | |
snatch | -0.225 |
zip | -0.225 |
ck | -0.224 |
Flo | -0.223 |
zzy | -0.215 |
Token | She |
Token ID | 1375 |
Feature activation | +4.99e-02 |
Pos logit contributions | |
sky | +0.039 |
End | +0.038 |
problem | +0.035 |
best | +0.032 |
gloomy | +0.032 |
Neg logit contributions | |
ck | -0.075 |
snatch | -0.073 |
zip | -0.069 |
Flo | -0.068 |
zzy | -0.068 |
Token | asked |
Token ID | 1965 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
snatch | +0.095 |
Flo | +0.094 |
zip | +0.093 |
ck | +0.090 |
izer | +0.088 |
Neg logit contributions | |
End | -0.067 |
problem | -0.065 |
sky | -0.064 |
best | -0.059 |
end | -0.058 |
Token | 's |
Token ID | 338 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
ck | +0.053 |
Flo | +0.053 |
snatch | +0.052 |
zip | +0.052 |
zzy | +0.051 |
Neg logit contributions | |
End | -0.042 |
problem | -0.037 |
sky | -0.036 |
happiest | -0.034 |
end | -0.033 |
Token | favorite |
Token ID | 4004 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
End | +0.194 |
problem | +0.181 |
sky | +0.172 |
end | +0.162 |
happiest | +0.161 |
Neg logit contributions | |
snatch | -0.223 |
zip | -0.221 |
Flo | -0.220 |
ck | -0.220 |
unw | -0.213 |
Token | spot |
Token ID | 4136 |
Feature activation | +5.83e-02 |
Pos logit contributions | |
snatch | +0.068 |
Flo | +0.067 |
zip | +0.065 |
ck | +0.064 |
unw | +0.062 |
Neg logit contributions | |
End | -0.055 |
problem | -0.048 |
happiest | -0.045 |
sky | -0.044 |
best | -0.041 |
Token | was |
Token ID | 373 |
Feature activation | -5.54e-02 |
Pos logit contributions | |
snatch | +0.106 |
ck | +0.106 |
zip | +0.105 |
zzy | +0.105 |
unw | +0.103 |
Neg logit contributions | |
End | -0.080 |
problem | -0.072 |
sky | -0.067 |
happiest | -0.067 |
goals | -0.062 |
Token | taken |
Token ID | 2077 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
End | +0.093 |
problem | +0.086 |
sky | +0.082 |
end | +0.078 |
happiest | +0.077 |
Neg logit contributions | |
snatch | -0.088 |
ck | -0.088 |
Flo | -0.088 |
zip | -0.087 |
unw | -0.086 |
Token | by |
Token ID | 416 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
snatch | +0.189 |
zip | +0.189 |
ck | +0.188 |
Flo | +0.187 |
zzy | +0.179 |
Neg logit contributions | |
End | -0.183 |
problem | -0.177 |
sky | -0.170 |
end | -0.159 |
happiest | -0.158 |
Token | a |
Token ID | 257 |
Feature activation | +1.33e-03 |
Pos logit contributions | |
zip | +0.356 |
snatch | +0.356 |
ck | +0.352 |
Flo | +0.350 |
zzy | +0.340 |
Neg logit contributions | |
End | -0.279 |
problem | -0.270 |
sky | -0.259 |
end | -0.240 |
best | -0.237 |
Token | big |
Token ID | 1263 |
Feature activation | +3.13e-04 |
Pos logit contributions | |
snatch | +0.003 |
Flo | +0.002 |
unw | +0.002 |
zip | +0.002 |
onna | +0.002 |
Neg logit contributions | |
End | -0.002 |
sky | -0.001 |
happiest | -0.001 |
problem | -0.001 |
best | -0.001 |
Token | dog |
Token ID | 3290 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
snatch | +0.001 |
unw | +0.001 |
Flo | +0.001 |
ck | +0.001 |
pat | +0.001 |
Neg logit contributions | |
End | -0.000 |
happiest | -0.000 |
best | -0.000 |
sky | -0.000 |
problem | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
zip | +0.023 |
snatch | +0.023 |
ck | +0.023 |
Flo | +0.023 |
zzy | +0.022 |
Neg logit contributions | |
problem | -0.013 |
End | -0.013 |
sky | -0.012 |
end | -0.011 |
gloomy | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +7.67e-02 |
Pos logit contributions | |
sky | +0.040 |
End | +0.038 |
problem | +0.035 |
gloomy | +0.033 |
day | +0.032 |
Neg logit contributions | |
ck | -0.065 |
snatch | -0.062 |
zip | -0.059 |
zzy | -0.059 |
Flo | -0.058 |
Token | The |
Token ID | 383 |
Feature activation | +4.07e-02 |
Pos logit contributions | |
sky | +0.028 |
End | +0.027 |
problem | +0.025 |
gloomy | +0.023 |
end | +0.022 |
Neg logit contributions | |
ck | -0.050 |
snatch | -0.048 |
zip | -0.047 |
zzy | -0.047 |
Flo | -0.046 |
Token | chicken |
Token ID | 9015 |
Feature activation | +4.07e-02 |
Pos logit contributions | |
Flo | +0.051 |
ck | +0.051 |
snatch | +0.047 |
izer | +0.045 |
zip | +0.044 |
Neg logit contributions | |
End | -0.080 |
problem | -0.077 |
happiest | -0.073 |
sky | -0.071 |
goals | -0.070 |
Token | was |
Token ID | 373 |
Feature activation | -5.30e-02 |
Pos logit contributions | |
ck | +0.081 |
Flo | +0.081 |
snatch | +0.080 |
zip | +0.079 |
unw | +0.077 |
Neg logit contributions | |
End | -0.053 |
problem | -0.049 |
sky | -0.046 |
happiest | -0.045 |
end | -0.043 |
Token | very |
Token ID | 845 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
problem | +0.034 |
End | +0.032 |
sky | +0.032 |
gloomy | +0.030 |
best | +0.030 |
Neg logit contributions | |
zip | -0.142 |
zzy | -0.138 |
snatch | -0.138 |
Flo | -0.136 |
ck | -0.136 |
Token | pretty |
Token ID | 2495 |
Feature activation | -4.22e-03 |
Pos logit contributions | |
problem | +0.144 |
gloomy | +0.135 |
End | +0.126 |
best | +0.123 |
sky | +0.120 |
Neg logit contributions | |
zip | -0.408 |
snatch | -0.405 |
zzy | -0.395 |
Flo | -0.387 |
Wh | -0.382 |
Token | and |
Token ID | 290 |
Feature activation | +8.53e-02 |
Pos logit contributions | |
End | +0.009 |
problem | +0.008 |
happiest | +0.008 |
end | +0.007 |
ending | +0.007 |
Neg logit contributions | |
unw | -0.005 |
Flo | -0.005 |
ck | -0.005 |
snatch | -0.005 |
zip | -0.005 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
Flo | +0.162 |
ck | +0.161 |
zip | +0.159 |
snatch | +0.159 |
unw | +0.156 |
Neg logit contributions | |
End | -0.119 |
problem | -0.115 |
sky | -0.105 |
end | -0.101 |
happiest | -0.095 |
Token | wanted |
Token ID | 2227 |
Feature activation | -5.57e-02 |
Pos logit contributions | |
ck | +0.202 |
unw | +0.201 |
zip | +0.196 |
Flo | +0.196 |
snatch | +0.192 |
Neg logit contributions | |
End | -0.190 |
problem | -0.176 |
sky | -0.165 |
happiest | -0.158 |
end | -0.158 |
Token | to |
Token ID | 284 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
End | +0.091 |
problem | +0.089 |
sky | +0.086 |
end | +0.080 |
best | +0.079 |
Neg logit contributions | |
zip | -0.092 |
snatch | -0.091 |
ck | -0.090 |
Flo | -0.090 |
zzy | -0.087 |
Token | play |
Token ID | 711 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
xc | +0.250 |
ge | +0.247 |
rily | +0.246 |
igi | +0.244 |
ried | +0.242 |
Neg logit contributions | |
end | -0.237 |
sky | -0.223 |
sun | -0.206 |
problem | -0.206 |
day | -0.205 |
Token | with |
Token ID | 351 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
snatch | +0.022 |
ried | +0.021 |
zip | +0.021 |
ge | +0.021 |
igi | +0.021 |
Neg logit contributions | |
end | -0.021 |
End | -0.021 |
sky | -0.020 |
problem | -0.020 |
gloomy | -0.020 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
End | +0.163 |
problem | +0.157 |
sky | +0.151 |
end | +0.138 |
gloomy | +0.136 |
Neg logit contributions | |
snatch | -0.238 |
zip | -0.237 |
Flo | -0.236 |
ck | -0.236 |
zzy | -0.226 |
Token | something |
Token ID | 1223 |
Feature activation | -6.76e-02 |
Pos logit contributions | |
problem | +0.022 |
End | +0.022 |
happiest | +0.019 |
sky | +0.019 |
ending | +0.019 |
Neg logit contributions | |
ck | -0.014 |
Flo | -0.014 |
zip | -0.014 |
unw | -0.013 |
zzy | -0.013 |
Token | in |
Token ID | 287 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
End | +0.083 |
problem | +0.071 |
sky | +0.069 |
happiest | +0.064 |
end | +0.062 |
Neg logit contributions | |
snatch | -0.141 |
ck | -0.141 |
zip | -0.138 |
Flo | -0.137 |
unw | -0.137 |
Token | the |
Token ID | 262 |
Feature activation | -9.48e-02 |
Pos logit contributions | |
Flo | +0.384 |
zip | +0.382 |
ck | +0.379 |
snatch | +0.374 |
zzy | +0.369 |
Neg logit contributions | |
End | -0.238 |
sky | -0.231 |
problem | -0.223 |
end | -0.213 |
gloomy | -0.206 |
Token | corner |
Token ID | 5228 |
Feature activation | -9.52e-02 |
Pos logit contributions | |
End | +0.124 |
problem | +0.121 |
ending | +0.096 |
goals | +0.096 |
happens | +0.094 |
Neg logit contributions | |
Flo | -0.182 |
zip | -0.178 |
snatch | -0.176 |
unw | -0.174 |
zzy | -0.171 |
Token | of |
Token ID | 286 |
Feature activation | +9.62e-02 |
Pos logit contributions | |
End | +0.165 |
problem | +0.160 |
happiest | +0.153 |
best | +0.144 |
sky | +0.144 |
Neg logit contributions | |
ck | -0.142 |
zip | -0.141 |
snatch | -0.139 |
unw | -0.134 |
zzy | -0.133 |
Token | her |
Token ID | 607 |
Feature activation | -2.33e-02 |
Pos logit contributions | |
Flo | +0.214 |
zip | +0.213 |
ck | +0.211 |
zzy | +0.209 |
snatch | +0.209 |
Neg logit contributions | |
sky | -0.109 |
End | -0.103 |
end | -0.100 |
gloomy | -0.090 |
night | -0.088 |
Token | garden |
Token ID | 11376 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
End | +0.034 |
problem | +0.028 |
happiest | +0.027 |
end | +0.025 |
ending | +0.025 |
Neg logit contributions | |
unw | -0.043 |
zip | -0.041 |
Flo | -0.041 |
zzy | -0.040 |
ighed | -0.039 |
Token | that |
Token ID | 326 |
Feature activation | -8.96e-02 |
Pos logit contributions | |
End | +0.226 |
problem | +0.213 |
happiest | +0.200 |
best | +0.184 |
goals | +0.178 |
Neg logit contributions | |
unw | -0.198 |
snatch | -0.194 |
zip | -0.185 |
ck | -0.184 |
Flo | -0.180 |
Token | she |
Token ID | 673 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
End | +0.126 |
problem | +0.122 |
sky | +0.105 |
goals | +0.105 |
happiest | +0.097 |
Neg logit contributions | |
zip | -0.162 |
snatch | -0.159 |
ck | -0.159 |
unw | -0.156 |
zzy | -0.156 |
Token | hadn |
Token ID | 8020 |
Feature activation | -8.63e-03 |
Pos logit contributions | |
End | +0.149 |
problem | +0.142 |
sky | +0.135 |
end | +0.122 |
happiest | +0.121 |
Neg logit contributions | |
snatch | -0.255 |
zip | -0.255 |
ck | -0.253 |
Flo | -0.252 |
zzy | -0.244 |
Token | was |
Token ID | 373 |
Feature activation | -5.23e-02 |
Pos logit contributions | |
Flo | +0.237 |
zip | +0.237 |
ck | +0.233 |
snatch | +0.231 |
zzy | +0.227 |
Neg logit contributions | |
End | -0.152 |
problem | -0.151 |
sky | -0.135 |
happiest | -0.128 |
gloomy | -0.125 |
Token | a |
Token ID | 257 |
Feature activation | +6.49e-03 |
Pos logit contributions | |
End | +0.068 |
sky | +0.064 |
problem | +0.063 |
best | +0.058 |
gloomy | +0.058 |
Neg logit contributions | |
snatch | -0.106 |
zip | -0.104 |
ck | -0.103 |
Flo | -0.101 |
zzy | -0.098 |
Token | little |
Token ID | 1310 |
Feature activation | +8.98e-02 |
Pos logit contributions | |
snatch | +0.013 |
zip | +0.013 |
ck | +0.013 |
Flo | +0.013 |
zzy | +0.012 |
Neg logit contributions | |
End | -0.008 |
problem | -0.008 |
sky | -0.007 |
end | -0.007 |
happiest | -0.007 |
Token | bear |
Token ID | 6842 |
Feature activation | +8.49e-02 |
Pos logit contributions | |
Flo | +0.172 |
zip | +0.167 |
snatch | +0.167 |
ck | +0.166 |
unw | +0.160 |
Neg logit contributions | |
End | -0.123 |
problem | -0.112 |
sky | -0.104 |
happiest | -0.099 |
end | -0.097 |
Token | named |
Token ID | 3706 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
snatch | +0.127 |
zip | +0.125 |
Flo | +0.119 |
ck | +0.118 |
zzy | +0.115 |
Neg logit contributions | |
End | -0.150 |
problem | -0.145 |
sky | -0.144 |
end | -0.140 |
day | -0.136 |
Token | Teddy |
Token ID | 29345 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
snatch | +0.165 |
zzy | +0.160 |
ck | +0.159 |
har | +0.155 |
zip | +0.155 |
Neg logit contributions | |
sky | -0.183 |
End | -0.181 |
problem | -0.173 |
end | -0.171 |
best | -0.169 |
Token | . |
Token ID | 13 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
ck | +0.271 |
Flo | +0.266 |
zip | +0.265 |
snatch | +0.261 |
unw | +0.257 |
Neg logit contributions | |
End | -0.202 |
problem | -0.198 |
sky | -0.184 |
happiest | -0.171 |
gloomy | -0.167 |
Token | Teddy |
Token ID | 29345 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
End | +0.035 |
sky | +0.034 |
problem | +0.034 |
gloomy | +0.031 |
end | +0.031 |
Neg logit contributions | |
ck | -0.031 |
snatch | -0.031 |
zip | -0.029 |
zzy | -0.029 |
Flo | -0.029 |
Token | liked |
Token ID | 8288 |
Feature activation | -3.88e-02 |
Pos logit contributions | |
Flo | +0.295 |
ck | +0.292 |
zip | +0.290 |
unw | +0.281 |
zzy | +0.281 |
Neg logit contributions | |
problem | -0.175 |
End | -0.174 |
sky | -0.152 |
happiest | -0.142 |
gloomy | -0.136 |
Token | to |
Token ID | 284 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
End | +0.067 |
problem | +0.064 |
sky | +0.062 |
end | +0.058 |
happiest | +0.058 |
Neg logit contributions | |
snatch | -0.061 |
zip | -0.060 |
ck | -0.060 |
Flo | -0.060 |
zzy | -0.057 |
Token | hug |
Token ID | 16225 |
Feature activation | +8.82e-02 |
Pos logit contributions | |
Flo | +0.255 |
zip | +0.251 |
uzz | +0.251 |
zzy | +0.249 |
ck | +0.248 |
Neg logit contributions | |
end | -0.232 |
End | -0.228 |
sky | -0.228 |
problem | -0.226 |
day | -0.219 |
Token | never |
Token ID | 1239 |
Feature activation | +5.49e-03 |
Pos logit contributions | |
snatch | +0.149 |
ck | +0.148 |
Flo | +0.145 |
zip | +0.145 |
unw | +0.141 |
Neg logit contributions | |
End | -0.110 |
problem | -0.105 |
sky | -0.104 |
end | -0.094 |
happiest | -0.092 |
Token | wanted |
Token ID | 2227 |
Feature activation | -6.22e-02 |
Pos logit contributions | |
ck | +0.009 |
snatch | +0.008 |
unw | +0.008 |
zip | +0.008 |
steer | +0.008 |
Neg logit contributions | |
End | -0.010 |
sky | -0.009 |
problem | -0.009 |
best | -0.008 |
happiest | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
End | +0.104 |
problem | +0.101 |
sky | +0.097 |
end | +0.091 |
happiest | +0.090 |
Neg logit contributions | |
snatch | -0.099 |
zip | -0.099 |
ck | -0.098 |
Flo | -0.097 |
zzy | -0.093 |
Token | smile |
Token ID | 8212 |
Feature activation | +4.98e-02 |
Pos logit contributions | |
Flo | +0.245 |
zzy | +0.233 |
zip | +0.232 |
snatch | +0.228 |
ck | +0.228 |
Neg logit contributions | |
problem | -0.235 |
End | -0.234 |
sky | -0.225 |
end | -0.224 |
gloomy | -0.208 |
Token | . |
Token ID | 13 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
zip | +0.087 |
ck | +0.086 |
Flo | +0.086 |
snatch | +0.085 |
zzy | +0.082 |
Neg logit contributions | |
End | -0.078 |
problem | -0.071 |
sky | -0.069 |
happiest | -0.063 |
end | -0.063 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
End | +0.048 |
problem | +0.046 |
sky | +0.045 |
gloomy | +0.041 |
best | +0.041 |
Neg logit contributions | |
ck | -0.057 |
zip | -0.055 |
snatch | -0.055 |
zzy | -0.053 |
steer | -0.052 |
Token | tried |
Token ID | 3088 |
Feature activation | -5.71e-03 |
Pos logit contributions | |
snatch | +0.209 |
zip | +0.209 |
ck | +0.208 |
Flo | +0.206 |
unw | +0.200 |
Neg logit contributions | |
End | -0.154 |
problem | -0.147 |
sky | -0.141 |
end | -0.129 |
gloomy | -0.129 |
Token | to |
Token ID | 284 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
problem | +0.009 |
End | +0.008 |
sky | +0.008 |
best | +0.008 |
end | +0.008 |
Neg logit contributions | |
Flo | -0.010 |
zip | -0.010 |
ck | -0.010 |
snatch | -0.010 |
zzy | -0.010 |
Token | make |
Token ID | 787 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
igi | +0.292 |
Flo | +0.289 |
rily | +0.277 |
uzz | +0.275 |
zzy | +0.271 |
Neg logit contributions | |
end | -0.238 |
End | -0.194 |
problem | -0.191 |
sky | -0.182 |
best | -0.175 |
Token | Mitt |
Token ID | 16627 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
zip | +0.040 |
snatch | +0.040 |
Flo | +0.039 |
ck | +0.039 |
zzy | +0.038 |
Neg logit contributions | |
problem | -0.046 |
End | -0.046 |
sky | -0.044 |
best | -0.042 |
gloomy | -0.041 |
Token | ens |
Token ID | 641 |
Feature activation | +6.70e-02 |
Pos logit contributions | |
snatch | +0.178 |
zip | +0.177 |
ck | +0.176 |
Flo | +0.171 |
unw | +0.166 |
Neg logit contributions | |
End | -0.198 |
problem | -0.195 |
sky | -0.187 |
end | -0.176 |
best | -0.176 |
Token | back |
Token ID | 736 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
End | +0.014 |
problem | +0.013 |
sky | +0.013 |
best | +0.012 |
end | +0.012 |
Neg logit contributions | |
Flo | -0.020 |
snatch | -0.019 |
zip | -0.019 |
ck | -0.019 |
zzy | -0.019 |
Token | at |
Token ID | 379 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
Flo | +0.059 |
snatch | +0.059 |
ck | +0.058 |
zip | +0.057 |
zzy | +0.055 |
Neg logit contributions | |
problem | -0.038 |
End | -0.037 |
sky | -0.036 |
best | -0.036 |
end | -0.034 |
Token | Jack |
Token ID | 3619 |
Feature activation | +8.49e-02 |
Pos logit contributions | |
zip | +0.178 |
snatch | +0.177 |
ck | +0.175 |
unw | +0.171 |
Flo | +0.168 |
Neg logit contributions | |
End | -0.171 |
problem | -0.165 |
sky | -0.156 |
happiest | -0.150 |
end | -0.145 |
Token | . |
Token ID | 13 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
zip | +0.173 |
snatch | +0.173 |
ck | +0.172 |
Flo | +0.171 |
zzy | +0.166 |
Neg logit contributions | |
End | -0.104 |
problem | -0.100 |
sky | -0.094 |
end | -0.086 |
happiest | -0.085 |
Token | So |
Token ID | 1406 |
Feature activation | +9.61e-02 |
Pos logit contributions | |
End | +0.055 |
problem | +0.053 |
sky | +0.049 |
happiest | +0.048 |
end | +0.048 |
Neg logit contributions | |
zip | -0.048 |
snatch | -0.047 |
unw | -0.045 |
Flo | -0.044 |
zzy | -0.043 |
Token | Jill |
Token ID | 22772 |
Feature activation | +7.24e-02 |
Pos logit contributions | |
snatch | +0.182 |
zip | +0.180 |
ck | +0.178 |
unw | +0.177 |
zzy | +0.171 |
Neg logit contributions | |
End | -0.126 |
problem | -0.121 |
sky | -0.119 |
happiest | -0.112 |
gloomy | -0.106 |
Token | sn |
Token ID | 3013 |
Feature activation | +4.85e-02 |
Pos logit contributions | |
Flo | +0.128 |
snatch | +0.128 |
zip | +0.128 |
ck | +0.123 |
uzz | +0.119 |
Neg logit contributions | |
problem | -0.107 |
End | -0.105 |
sky | -0.100 |
end | -0.094 |
best | -0.093 |
Token | uck |
Token ID | 1347 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
snatch | +0.084 |
Flo | +0.082 |
ck | +0.082 |
unw | +0.080 |
zip | +0.077 |
Neg logit contributions | |
End | -0.081 |
problem | -0.072 |
sky | -0.070 |
end | -0.067 |
day | -0.065 |
Token | into |
Token ID | 656 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
Flo | +0.359 |
snatch | +0.352 |
Twe | +0.348 |
ck | +0.342 |
zip | +0.336 |
Neg logit contributions | |
End | -0.264 |
sky | -0.260 |
problem | -0.258 |
best | -0.245 |
days | -0.243 |
Token | Jack |
Token ID | 3619 |
Feature activation | +8.34e-02 |
Pos logit contributions | |
Flo | +0.325 |
igi | +0.296 |
ck | +0.296 |
zip | +0.292 |
zzy | +0.292 |
Neg logit contributions | |
End | -0.153 |
end | -0.143 |
sky | -0.140 |
problem | -0.134 |
gloomy | -0.129 |
Token | 's |
Token ID | 338 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
snatch | +0.193 |
zip | +0.193 |
ck | +0.192 |
Flo | +0.191 |
zzy | +0.186 |
Neg logit contributions | |
End | -0.079 |
problem | -0.075 |
sky | -0.070 |
end | -0.061 |
happiest | -0.060 |
Token | big |
Token ID | 1263 |
Feature activation | -8.63e-03 |
Pos logit contributions | |
End | +0.015 |
problem | +0.013 |
sky | +0.013 |
happiest | +0.013 |
best | +0.012 |
Neg logit contributions | |
snatch | -0.020 |
Flo | -0.019 |
zip | -0.019 |
ck | -0.019 |
unw | -0.019 |
Token | shadow |
Token ID | 9082 |
Feature activation | -4.27e-02 |
Pos logit contributions | |
End | +0.013 |
problem | +0.011 |
happiest | +0.011 |
sky | +0.011 |
best | +0.011 |
Neg logit contributions | |
snatch | -0.016 |
ck | -0.015 |
Flo | -0.015 |
zip | -0.015 |
unw | -0.015 |
Token | behind |
Token ID | 2157 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
End | +0.066 |
problem | +0.063 |
sky | +0.060 |
happiest | +0.057 |
best | +0.056 |
Neg logit contributions | |
snatch | -0.074 |
ck | -0.074 |
zip | -0.073 |
Flo | -0.073 |
unw | -0.070 |
Token | a |
Token ID | 257 |
Feature activation | -9.99e-03 |
Pos logit contributions | |
zip | +0.311 |
snatch | +0.306 |
ck | +0.305 |
Flo | +0.305 |
zzy | +0.299 |
Neg logit contributions | |
End | -0.228 |
sky | -0.223 |
end | -0.210 |
problem | -0.202 |
gloomy | -0.189 |
Token | bush |
Token ID | 24484 |
Feature activation | -7.25e-02 |
Pos logit contributions | |
End | +0.019 |
happiest | +0.017 |
problem | +0.016 |
ending | +0.016 |
best | +0.016 |
Neg logit contributions | |
Flo | -0.015 |
snatch | -0.014 |
unw | -0.014 |
zip | -0.014 |
steer | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -3.59e-02 |
Pos logit contributions | |
End | +0.096 |
problem | +0.091 |
happiest | +0.083 |
sky | +0.082 |
best | +0.080 |
Neg logit contributions | |
snatch | -0.141 |
ck | -0.139 |
Flo | -0.138 |
zip | -0.138 |
unw | -0.136 |
Token | It |
Token ID | 632 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
sky | +0.036 |
End | +0.031 |
problem | +0.027 |
gloomy | +0.027 |
sun | +0.024 |
Neg logit contributions | |
ck | -0.092 |
snatch | -0.086 |
zip | -0.084 |
ge | -0.084 |
steer | -0.082 |
Token | was |
Token ID | 373 |
Feature activation | -6.63e-02 |
Pos logit contributions | |
snatch | +0.036 |
zip | +0.036 |
ck | +0.036 |
Flo | +0.036 |
zzy | +0.035 |
Neg logit contributions | |
End | -0.018 |
problem | -0.017 |
sky | -0.016 |
end | -0.014 |
happiest | -0.014 |
Token | a |
Token ID | 257 |
Feature activation | -8.98e-03 |
Pos logit contributions | |
End | +0.088 |
problem | +0.085 |
sky | +0.083 |
gloomy | +0.077 |
best | +0.076 |
Neg logit contributions | |
zip | -0.133 |
snatch | -0.131 |
Flo | -0.129 |
ck | -0.128 |
zzy | -0.125 |
Token | cat |
Token ID | 3797 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
End | +0.013 |
problem | +0.012 |
sky | +0.011 |
happiest | +0.011 |
best | +0.011 |
Neg logit contributions | |
snatch | -0.016 |
Flo | -0.016 |
zip | -0.016 |
ck | -0.016 |
unw | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | -8.42e-02 |
Pos logit contributions | |
zip | +0.047 |
snatch | +0.047 |
ck | +0.047 |
Flo | +0.046 |
zzy | +0.045 |
Neg logit contributions | |
End | -0.029 |
problem | -0.028 |
sky | -0.027 |
end | -0.025 |
gloomy | -0.024 |
Token | y |
Token ID | 88 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
snatch | +0.240 |
zip | +0.239 |
ck | +0.236 |
Flo | +0.235 |
zzy | +0.227 |
Neg logit contributions | |
End | -0.181 |
problem | -0.176 |
sky | -0.168 |
end | -0.155 |
best | -0.153 |
Token | went |
Token ID | 1816 |
Feature activation | +5.46e-04 |
Pos logit contributions | |
unw | +0.043 |
ck | +0.043 |
Flo | +0.042 |
snatch | +0.042 |
zip | +0.042 |
Neg logit contributions | |
End | -0.044 |
problem | -0.041 |
gloomy | -0.038 |
happiest | -0.037 |
sky | -0.037 |
Token | back |
Token ID | 736 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
ck | +0.001 |
snatch | +0.001 |
zip | +0.001 |
Flo | +0.001 |
zzy | +0.001 |
Neg logit contributions | |
End | -0.001 |
problem | -0.001 |
sky | -0.001 |
end | -0.001 |
best | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
snatch | +0.060 |
ck | +0.059 |
Flo | +0.059 |
zip | +0.057 |
zzy | +0.055 |
Neg logit contributions | |
problem | -0.044 |
sky | -0.043 |
End | -0.043 |
best | -0.039 |
day | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
zip | +0.225 |
snatch | +0.225 |
ck | +0.223 |
Flo | +0.222 |
zzy | +0.213 |
Neg logit contributions | |
End | -0.249 |
problem | -0.242 |
sky | -0.233 |
end | -0.219 |
happiest | -0.216 |
Token | hive |
Token ID | 35881 |
Feature activation | -9.83e-02 |
Pos logit contributions | |
End | +0.139 |
problem | +0.125 |
sky | +0.116 |
happiest | +0.111 |
gloomy | +0.107 |
Neg logit contributions | |
Flo | -0.196 |
snatch | -0.190 |
zip | -0.185 |
unw | -0.184 |
ck | -0.183 |
Token | , |
Token ID | 11 |
Feature activation | -9.10e-02 |
Pos logit contributions | |
End | +0.151 |
problem | +0.145 |
sky | +0.137 |
happiest | +0.128 |
end | +0.128 |
Neg logit contributions | |
snatch | -0.169 |
zip | -0.169 |
ck | -0.168 |
Flo | -0.167 |
unw | -0.162 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
sky | +0.111 |
End | +0.107 |
problem | +0.101 |
gloomy | +0.100 |
happiest | +0.097 |
Neg logit contributions | |
zip | -0.190 |
zzy | -0.189 |
snatch | -0.186 |
ck | -0.186 |
Flo | -0.183 |
Token | and |
Token ID | 290 |
Feature activation | +7.40e-02 |
Pos logit contributions | |
End | +0.277 |
problem | +0.259 |
happiest | +0.240 |
sky | +0.234 |
end | +0.231 |
Neg logit contributions | |
Flo | -0.185 |
snatch | -0.185 |
ck | -0.185 |
zip | -0.180 |
unw | -0.177 |
Token | ready |
Token ID | 3492 |
Feature activation | -7.86e-02 |
Pos logit contributions | |
zip | +0.123 |
snatch | +0.119 |
ck | +0.117 |
unw | +0.117 |
steer | +0.113 |
Neg logit contributions | |
End | -0.124 |
problem | -0.121 |
sky | -0.108 |
end | -0.105 |
happiest | -0.095 |
Token | to |
Token ID | 284 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
End | +0.130 |
problem | +0.124 |
sky | +0.119 |
happiest | +0.111 |
end | +0.110 |
Neg logit contributions | |
snatch | -0.128 |
zip | -0.127 |
ck | -0.127 |
Flo | -0.126 |
unw | -0.122 |
Token | friend |
Token ID | 1545 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
problem | +0.079 |
End | +0.077 |
sky | +0.073 |
best | +0.071 |
end | +0.069 |
Neg logit contributions | |
ck | -0.126 |
zip | -0.126 |
snatch | -0.125 |
Flo | -0.124 |
unw | -0.119 |
Token | . |
Token ID | 13 |
Feature activation | -3.53e-02 |
Pos logit contributions | |
snatch | +0.318 |
zip | +0.318 |
ck | +0.316 |
Flo | +0.310 |
zzy | +0.302 |
Neg logit contributions | |
End | -0.149 |
problem | -0.144 |
sky | -0.135 |
best | -0.126 |
end | -0.126 |
Token | They |
Token ID | 1119 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
End | +0.050 |
sky | +0.049 |
problem | +0.046 |
best | +0.043 |
end | +0.043 |
Neg logit contributions | |
ck | -0.068 |
snatch | -0.068 |
zip | -0.065 |
zzy | -0.064 |
unw | -0.063 |
Token | knew |
Token ID | 2993 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
zip | +0.031 |
ck | +0.031 |
snatch | +0.031 |
Flo | +0.031 |
zzy | +0.030 |
Neg logit contributions | |
End | -0.021 |
problem | -0.021 |
sky | -0.019 |
gloomy | -0.018 |
happiest | -0.017 |
Token | they |
Token ID | 484 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
End | +0.166 |
problem | +0.161 |
sky | +0.148 |
best | +0.136 |
end | +0.133 |
Neg logit contributions | |
zip | -0.384 |
snatch | -0.382 |
Flo | -0.378 |
ck | -0.378 |
zzy | -0.370 |
Token | could |
Token ID | 714 |
Feature activation | -9.42e-02 |
Pos logit contributions | |
End | +0.185 |
problem | +0.178 |
sky | +0.170 |
gloomy | +0.144 |
goals | +0.137 |
Neg logit contributions | |
zip | -0.267 |
zzy | -0.264 |
ck | -0.263 |
snatch | -0.255 |
unw | -0.252 |
Token | always |
Token ID | 1464 |
Feature activation | +4.14e-02 |
Pos logit contributions | |
end | +0.127 |
End | +0.118 |
best | +0.117 |
problem | +0.110 |
day | +0.108 |
Neg logit contributions | |
rily | -0.183 |
zip | -0.182 |
snatch | -0.181 |
ck | -0.179 |
zzy | -0.178 |
Token | count |
Token ID | 954 |
Feature activation | +5.29e-02 |
Pos logit contributions | |
Flo | +0.071 |
zip | +0.070 |
snatch | +0.069 |
uzz | +0.069 |
zzy | +0.069 |
Neg logit contributions | |
End | -0.062 |
end | -0.061 |
problem | -0.061 |
best | -0.059 |
happiest | -0.057 |
Token | on |
Token ID | 319 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
snatch | +0.091 |
zip | +0.090 |
Flo | +0.090 |
ck | +0.090 |
zzy | +0.086 |
Neg logit contributions | |
End | -0.082 |
problem | -0.080 |
sky | -0.077 |
end | -0.071 |
best | -0.071 |
Token | each |
Token ID | 1123 |
Feature activation | -4.64e-02 |
Pos logit contributions | |
zip | +0.212 |
snatch | +0.212 |
ck | +0.211 |
Flo | +0.210 |
unw | +0.201 |
Neg logit contributions | |
End | -0.196 |
problem | -0.188 |
sky | -0.181 |
happiest | -0.168 |
end | -0.167 |
Token | other |
Token ID | 584 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
End | +0.043 |
problem | +0.039 |
sky | +0.036 |
happiest | +0.034 |
gloomy | +0.032 |
Neg logit contributions | |
snatch | -0.108 |
ck | -0.107 |
zip | -0.106 |
Flo | -0.105 |
steer | -0.102 |
Token |
|
Token ID | 198 |
Feature activation | +7.49e-02 |
Pos logit contributions | |
snatch | +0.179 |
ck | +0.165 |
har | +0.162 |
ried | +0.162 |
zip | +0.161 |
Neg logit contributions | |
End | -0.080 |
sky | -0.076 |
end | -0.074 |
problem | -0.073 |
day | -0.072 |
Token | His |
Token ID | 6653 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
snatch | +0.093 |
ck | +0.090 |
zip | +0.086 |
Flo | +0.085 |
unw | +0.080 |
Neg logit contributions | |
End | -0.157 |
problem | -0.150 |
sky | -0.149 |
end | -0.146 |
gloomy | -0.140 |
Token | mom |
Token ID | 1995 |
Feature activation | -4.45e-03 |
Pos logit contributions | |
snatch | +0.260 |
zip | +0.259 |
Flo | +0.258 |
ck | +0.258 |
zzy | +0.248 |
Neg logit contributions | |
End | -0.215 |
problem | -0.210 |
sky | -0.199 |
end | -0.187 |
best | -0.185 |
Token | smiled |
Token ID | 13541 |
Feature activation | -4.09e-02 |
Pos logit contributions | |
End | +0.006 |
problem | +0.006 |
sky | +0.006 |
end | +0.005 |
best | +0.005 |
Neg logit contributions | |
snatch | -0.009 |
Flo | -0.009 |
zip | -0.008 |
ck | -0.008 |
zzy | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | +7.85e-02 |
Pos logit contributions | |
End | +0.085 |
problem | +0.083 |
happiest | +0.077 |
sky | +0.075 |
tallest | +0.073 |
Neg logit contributions | |
unw | -0.043 |
zip | -0.042 |
ck | -0.041 |
zzy | -0.041 |
Flo | -0.040 |
Token | said |
Token ID | 531 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
snatch | +0.137 |
zip | +0.136 |
Flo | +0.135 |
ck | +0.135 |
zzy | +0.129 |
Neg logit contributions | |
End | -0.120 |
problem | -0.116 |
sky | -0.112 |
end | -0.104 |
best | -0.103 |
Token | , |
Token ID | 11 |
Feature activation | -8.35e-02 |
Pos logit contributions | |
End | +0.036 |
sky | +0.032 |
happiest | +0.031 |
problem | +0.031 |
end | +0.028 |
Neg logit contributions | |
Flo | -0.055 |
onna | -0.053 |
uzz | -0.053 |
snatch | -0.052 |
unw | -0.052 |
Token | " |
Token ID | 366 |
Feature activation | -9.97e-02 |
Pos logit contributions | |
End | +0.095 |
problem | +0.094 |
sky | +0.082 |
happiest | +0.079 |
end | +0.078 |
Neg logit contributions | |
Flo | -0.176 |
snatch | -0.174 |
zip | -0.174 |
ck | -0.170 |
unw | -0.169 |
Token | That |
Token ID | 2504 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
End | +0.201 |
problem | +0.200 |
end | +0.178 |
sky | +0.176 |
gloomy | +0.175 |
Neg logit contributions | |
Flo | -0.134 |
zip | -0.132 |
snatch | -0.122 |
zzy | -0.121 |
Wh | -0.120 |
Token | is |
Token ID | 318 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
End | +0.037 |
lessness | +0.027 |
gloomy | +0.025 |
happiest | +0.025 |
days | +0.024 |
Neg logit contributions | |
zip | -0.027 |
steer | -0.026 |
shave | -0.026 |
unw | -0.026 |
ask | -0.025 |
Token | a |
Token ID | 257 |
Feature activation | -3.93e-03 |
Pos logit contributions | |
End | +0.193 |
problem | +0.181 |
sky | +0.178 |
happiest | +0.157 |
gloomy | +0.153 |
Neg logit contributions | |
zip | -0.232 |
Flo | -0.230 |
unw | -0.227 |
snatch | -0.220 |
ck | -0.218 |
Token | scary |
Token ID | 14343 |
Feature activation | -7.23e-02 |
Pos logit contributions | |
snatch | +0.137 |
ck | +0.134 |
steer | +0.131 |
zip | +0.129 |
Flo | +0.128 |
Neg logit contributions | |
End | -0.130 |
problem | -0.113 |
sky | -0.111 |
end | -0.105 |
best | -0.104 |
Token | . |
Token ID | 13 |
Feature activation | -4.62e-02 |
Pos logit contributions | |
End | +0.124 |
problem | +0.110 |
happiest | +0.110 |
sky | +0.108 |
end | +0.102 |
Neg logit contributions | |
ck | -0.116 |
snatch | -0.115 |
Flo | -0.112 |
zip | -0.112 |
unw | -0.110 |
Token | I |
Token ID | 314 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
End | +0.056 |
sky | +0.054 |
problem | +0.050 |
happiest | +0.046 |
gloomy | +0.046 |
Neg logit contributions | |
ck | -0.097 |
snatch | -0.095 |
zip | -0.095 |
zzy | -0.093 |
unw | -0.092 |
Token | like |
Token ID | 588 |
Feature activation | +5.08e-02 |
Pos logit contributions | |
problem | +0.020 |
end | +0.018 |
End | +0.017 |
sky | +0.016 |
day | +0.016 |
Neg logit contributions | |
rily | -0.032 |
zip | -0.032 |
Flo | -0.031 |
izer | -0.031 |
snatch | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
snatch | +0.081 |
ck | +0.080 |
zip | +0.080 |
Flo | +0.078 |
unw | +0.076 |
Neg logit contributions | |
End | -0.087 |
problem | -0.080 |
sky | -0.076 |
happiest | -0.074 |
end | -0.073 |
Token | flowers |
Token ID | 12734 |
Feature activation | -7.41e-02 |
Pos logit contributions | |
End | +0.136 |
problem | +0.118 |
happiest | +0.104 |
sky | +0.104 |
starts | +0.103 |
Neg logit contributions | |
snatch | -0.200 |
Flo | -0.192 |
zip | -0.189 |
steer | -0.184 |
unw | -0.184 |
Token | better |
Token ID | 1365 |
Feature activation | -5.13e-02 |
Pos logit contributions | |
End | +0.120 |
problem | +0.107 |
sky | +0.101 |
happiest | +0.098 |
end | +0.094 |
Neg logit contributions | |
ck | -0.128 |
Flo | -0.125 |
snatch | -0.125 |
zip | -0.122 |
unw | -0.121 |
Token | . |
Token ID | 13 |
Feature activation | -4.59e-02 |
Pos logit contributions | |
End | +0.089 |
problem | +0.080 |
sky | +0.080 |
happiest | +0.078 |
end | +0.074 |
Neg logit contributions | |
ck | -0.083 |
snatch | -0.079 |
Flo | -0.079 |
zip | -0.079 |
steer | -0.076 |
Token | They |
Token ID | 1119 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
End | +0.056 |
sky | +0.055 |
problem | +0.048 |
happiest | +0.046 |
end | +0.046 |
Neg logit contributions | |
ck | -0.097 |
snatch | -0.094 |
zip | -0.094 |
zzy | -0.092 |
ge | -0.091 |
Token | are |
Token ID | 389 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
snatch | +0.027 |
ck | +0.027 |
zip | +0.027 |
Flo | +0.027 |
unw | +0.026 |
Neg logit contributions | |
End | -0.017 |
problem | -0.015 |
sky | -0.015 |
gloomy | -0.013 |
happiest | -0.013 |
Token | soft |
Token ID | 2705 |
Feature activation | +2.28e-03 |
Pos logit contributions | |
End | +0.218 |
problem | +0.204 |
sky | +0.194 |
end | +0.180 |
happiest | +0.175 |
Neg logit contributions | |
ck | -0.319 |
snatch | -0.316 |
Flo | -0.313 |
zip | -0.312 |
unw | -0.302 |
Token | the |
Token ID | 262 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
End | +0.094 |
sky | +0.091 |
problem | +0.091 |
best | +0.085 |
day | +0.081 |
Neg logit contributions | |
zip | -0.204 |
snatch | -0.204 |
ck | -0.203 |
Flo | -0.200 |
zzy | -0.193 |
Token | game |
Token ID | 983 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
End | +0.124 |
problem | +0.123 |
sky | +0.115 |
gloomy | +0.112 |
happens | +0.108 |
Neg logit contributions | |
snatch | -0.196 |
zip | -0.186 |
unw | -0.180 |
steer | -0.177 |
grab | -0.175 |
Token | came |
Token ID | 1625 |
Feature activation | +4.98e-02 |
Pos logit contributions | |
End | +0.156 |
problem | +0.142 |
sky | +0.130 |
gloomy | +0.128 |
happiest | +0.128 |
Neg logit contributions | |
snatch | -0.235 |
zip | -0.230 |
ck | -0.226 |
unw | -0.226 |
Flo | -0.225 |
Token | to |
Token ID | 284 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
snatch | +0.084 |
zip | +0.084 |
ck | +0.083 |
Flo | +0.083 |
zzy | +0.079 |
Neg logit contributions | |
End | -0.079 |
problem | -0.076 |
sky | -0.073 |
end | -0.068 |
happiest | -0.068 |
Token | an |
Token ID | 281 |
Feature activation | +5.32e-02 |
Pos logit contributions | |
zip | +0.292 |
Flo | +0.282 |
ck | +0.279 |
snatch | +0.274 |
zzy | +0.274 |
Neg logit contributions | |
End | -0.180 |
end | -0.179 |
sky | -0.174 |
problem | -0.172 |
gloomy | -0.159 |
Token | end |
Token ID | 886 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
snatch | +0.110 |
ck | +0.109 |
unw | +0.103 |
steer | +0.100 |
grab | +0.099 |
Neg logit contributions | |
sky | -0.055 |
problem | -0.055 |
End | -0.054 |
best | -0.048 |
happens | -0.047 |
Token | and |
Token ID | 290 |
Feature activation | +7.14e-02 |
Pos logit contributions | |
End | +0.425 |
problem | +0.398 |
sky | +0.376 |
happiest | +0.353 |
end | +0.346 |
Neg logit contributions | |
zip | -0.611 |
snatch | -0.608 |
ck | -0.607 |
Flo | -0.603 |
zzy | -0.590 |
Token | they |
Token ID | 484 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
snatch | +0.130 |
zip | +0.129 |
ck | +0.128 |
Flo | +0.128 |
zzy | +0.123 |
Neg logit contributions | |
End | -0.104 |
problem | -0.100 |
sky | -0.096 |
end | -0.088 |
happiest | -0.088 |
Token | both |
Token ID | 1111 |
Feature activation | -5.21e-02 |
Pos logit contributions | |
End | +0.188 |
problem | +0.181 |
sky | +0.170 |
end | +0.154 |
happiest | +0.153 |
Neg logit contributions | |
zip | -0.292 |
snatch | -0.292 |
ck | -0.291 |
Flo | -0.288 |
zzy | -0.281 |
Token | collapsed |
Token ID | 14707 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
End | +0.079 |
problem | +0.077 |
sky | +0.070 |
end | +0.065 |
best | +0.064 |
Neg logit contributions | |
ck | -0.092 |
snatch | -0.091 |
zip | -0.091 |
zzy | -0.090 |
Flo | -0.089 |
Token | in |
Token ID | 287 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
End | +0.024 |
problem | +0.023 |
sky | +0.022 |
end | +0.021 |
happiest | +0.021 |
Neg logit contributions | |
snatch | -0.025 |
zip | -0.024 |
ck | -0.024 |
Flo | -0.024 |
unw | -0.023 |
Token | club |
Token ID | 3430 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
snatch | +0.058 |
unw | +0.057 |
zip | +0.057 |
ck | +0.056 |
Flo | +0.056 |
Neg logit contributions | |
End | -0.041 |
problem | -0.037 |
sky | -0.036 |
gloomy | -0.033 |
end | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | +7.59e-02 |
Pos logit contributions | |
End | +0.044 |
problem | +0.043 |
sky | +0.042 |
end | +0.038 |
best | +0.038 |
Neg logit contributions | |
snatch | -0.056 |
ck | -0.056 |
zip | -0.056 |
Flo | -0.055 |
zzy | -0.053 |
Token | part |
Token ID | 636 |
Feature activation | -2.42e-02 |
Pos logit contributions | |
Flo | +0.136 |
snatch | +0.135 |
zip | +0.134 |
ck | +0.132 |
zzy | +0.129 |
Neg logit contributions | |
End | -0.107 |
problem | -0.107 |
sky | -0.104 |
gloomy | -0.097 |
end | -0.096 |
Token | ied |
Token ID | 798 |
Feature activation | +9.50e-02 |
Pos logit contributions | |
End | +0.036 |
problem | +0.035 |
sky | +0.033 |
end | +0.031 |
happiest | +0.031 |
Neg logit contributions | |
zip | -0.043 |
snatch | -0.043 |
ck | -0.042 |
Flo | -0.042 |
zzy | -0.041 |
Token | all |
Token ID | 477 |
Feature activation | -7.61e-02 |
Pos logit contributions | |
zip | +0.169 |
snatch | +0.168 |
ck | +0.167 |
Flo | +0.167 |
zzy | +0.164 |
Neg logit contributions | |
End | -0.139 |
problem | -0.134 |
sky | -0.132 |
end | -0.124 |
gloomy | -0.122 |
Token | night |
Token ID | 1755 |
Feature activation | -2.88e-01 |
Pos logit contributions | |
End | +0.117 |
problem | +0.113 |
sky | +0.105 |
happiest | +0.099 |
end | +0.098 |
Neg logit contributions | |
zip | -0.133 |
ck | -0.130 |
snatch | -0.130 |
Flo | -0.128 |
unw | -0.124 |
Token | long |
Token ID | 890 |
Feature activation | -6.66e-02 |
Pos logit contributions | |
End | +0.465 |
problem | +0.419 |
happiest | +0.396 |
end | +0.383 |
sky | +0.367 |
Neg logit contributions | |
unw | -0.507 |
Flo | -0.486 |
zip | -0.481 |
ck | -0.480 |
onna | -0.480 |
Token | . |
Token ID | 13 |
Feature activation | -3.67e-02 |
Pos logit contributions | |
End | +0.094 |
problem | +0.090 |
happiest | +0.083 |
sky | +0.082 |
end | +0.078 |
Neg logit contributions | |
snatch | -0.122 |
ck | -0.121 |
Flo | -0.121 |
zip | -0.121 |
unw | -0.120 |
Token | But |
Token ID | 887 |
Feature activation | +6.27e-02 |
Pos logit contributions | |
End | +0.060 |
problem | +0.057 |
sky | +0.057 |
end | +0.051 |
gloomy | +0.051 |
Neg logit contributions | |
ck | -0.063 |
snatch | -0.062 |
zip | -0.061 |
Flo | -0.061 |
zzy | -0.059 |
Token | the |
Token ID | 262 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
ck | +0.106 |
snatch | +0.105 |
zip | +0.104 |
Flo | +0.103 |
unw | +0.100 |
Neg logit contributions | |
problem | -0.091 |
End | -0.091 |
sky | -0.087 |
happiest | -0.080 |
gloomy | -0.078 |
Token | next |
Token ID | 1306 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
End | +0.117 |
problem | +0.114 |
starts | +0.104 |
sizes | +0.103 |
sky | +0.101 |
Neg logit contributions | |
snatch | -0.197 |
unw | -0.187 |
Wh | -0.180 |
zip | -0.178 |
ck | -0.177 |
Token | tasty |
Token ID | 25103 |
Feature activation | +1.03e-03 |
Pos logit contributions | |
snatch | +0.178 |
zip | +0.177 |
ck | +0.177 |
Flo | +0.176 |
unw | +0.171 |
Neg logit contributions | |
End | -0.131 |
problem | -0.123 |
sky | -0.117 |
happiest | -0.109 |
end | -0.108 |
Token | foods |
Token ID | 9013 |
Feature activation | -7.27e-03 |
Pos logit contributions | |
Flo | +0.002 |
ck | +0.002 |
unw | +0.002 |
snatch | +0.002 |
zip | +0.002 |
Neg logit contributions | |
End | -0.002 |
problem | -0.001 |
happiest | -0.001 |
sky | -0.001 |
end | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -2.76e-02 |
Pos logit contributions | |
End | +0.012 |
problem | +0.011 |
sky | +0.010 |
happiest | +0.010 |
gloomy | +0.010 |
Neg logit contributions | |
ck | -0.012 |
Flo | -0.012 |
unw | -0.012 |
zip | -0.012 |
zzy | -0.012 |
Token | But |
Token ID | 887 |
Feature activation | +6.82e-02 |
Pos logit contributions | |
sky | +0.034 |
End | +0.031 |
problem | +0.027 |
day | +0.027 |
end | +0.027 |
Neg logit contributions | |
ck | -0.063 |
snatch | -0.061 |
zip | -0.059 |
zzy | -0.058 |
har | -0.056 |
Token | one |
Token ID | 530 |
Feature activation | -8.64e-02 |
Pos logit contributions | |
snatch | +0.143 |
ck | +0.142 |
Flo | +0.142 |
zip | +0.141 |
unw | +0.136 |
Neg logit contributions | |
End | -0.079 |
problem | -0.076 |
sky | -0.072 |
happiest | -0.065 |
best | -0.063 |
Token | day |
Token ID | 1110 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
End | +0.040 |
problem | +0.035 |
sky | +0.030 |
happiest | +0.020 |
best | +0.018 |
Neg logit contributions | |
ck | -0.240 |
snatch | -0.239 |
Flo | -0.238 |
zip | -0.236 |
zzy | -0.233 |
Token | the |
Token ID | 262 |
Feature activation | -9.94e-02 |
Pos logit contributions | |
End | +0.376 |
happiest | +0.339 |
problem | +0.331 |
sky | +0.329 |
tallest | +0.296 |
Neg logit contributions | |
Flo | -0.453 |
snatch | -0.444 |
ck | -0.444 |
unw | -0.435 |
zip | -0.433 |
Token | sky |
Token ID | 6766 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
End | +0.161 |
problem | +0.155 |
sky | +0.147 |
happiest | +0.138 |
end | +0.137 |
Neg logit contributions | |
snatch | -0.162 |
Flo | -0.160 |
zip | -0.159 |
ck | -0.159 |
unw | -0.151 |
Token | was |
Token ID | 373 |
Feature activation | -5.99e-02 |
Pos logit contributions | |
End | +0.256 |
problem | +0.215 |
happiest | +0.209 |
best | +0.202 |
goals | +0.195 |
Neg logit contributions | |
ck | -0.361 |
snatch | -0.354 |
zip | -0.340 |
steer | -0.338 |
unw | -0.334 |
Token | dark |
Token ID | 3223 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
End | +0.064 |
problem | +0.057 |
sky | +0.052 |
happiest | +0.049 |
end | +0.048 |
Neg logit contributions | |
snatch | -0.129 |
ck | -0.128 |
zip | -0.127 |
Flo | -0.126 |
unw | -0.126 |
Token | and |
Token ID | 290 |
Feature activation | +7.82e-02 |
Pos logit contributions | |
End | +0.231 |
problem | +0.208 |
happiest | +0.200 |
best | +0.186 |
goals | +0.184 |
Neg logit contributions | |
unw | -0.145 |
zip | -0.140 |
ck | -0.139 |
Flo | -0.138 |
snatch | -0.136 |
Token | lonely |
Token ID | 21757 |
Feature activation | -9.66e-02 |
Pos logit contributions | |
End | +0.059 |
problem | +0.056 |
sky | +0.053 |
end | +0.046 |
happiest | +0.045 |
Neg logit contributions | |
snatch | -0.135 |
zip | -0.134 |
ck | -0.134 |
Flo | -0.133 |
zzy | -0.129 |
Token | . |
Token ID | 13 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
End | +0.184 |
problem | +0.157 |
sky | +0.154 |
happiest | +0.148 |
starts | +0.147 |
Neg logit contributions | |
unw | -0.125 |
zip | -0.122 |
snatch | -0.116 |
zzy | -0.115 |
Flo | -0.113 |
Token |
|
Token ID | 198 |
Feature activation | +7.89e-02 |
Pos logit contributions | |
End | +0.039 |
problem | +0.037 |
sky | +0.036 |
end | +0.034 |
happiest | +0.033 |
Neg logit contributions | |
snatch | -0.036 |
zip | -0.036 |
ck | -0.036 |
Flo | -0.036 |
zzy | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | +7.65e-02 |
Pos logit contributions | |
snatch | +0.175 |
ck | +0.164 |
zip | +0.162 |
unw | +0.159 |
Flo | +0.159 |
Neg logit contributions | |
End | -0.098 |
sky | -0.090 |
problem | -0.086 |
day | -0.084 |
end | -0.083 |
Token | One |
Token ID | 3198 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
snatch | +0.113 |
zip | +0.111 |
ck | +0.110 |
Flo | +0.108 |
unw | +0.105 |
Neg logit contributions | |
End | -0.143 |
problem | -0.135 |
sky | -0.132 |
end | -0.127 |
gloomy | -0.123 |
Token | day |
Token ID | 1110 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
zzy | +0.069 |
ck | +0.069 |
Flo | +0.068 |
zip | +0.068 |
snatch | +0.067 |
Neg logit contributions | |
End | -0.019 |
problem | -0.017 |
sky | -0.015 |
happiest | -0.014 |
goals | -0.013 |
Token | she |
Token ID | 673 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
End | +0.298 |
problem | +0.273 |
happiest | +0.258 |
sky | +0.254 |
goals | +0.234 |
Neg logit contributions | |
Flo | -0.510 |
zip | -0.503 |
unw | -0.497 |
snatch | -0.493 |
ck | -0.489 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
End | +0.168 |
problem | +0.163 |
sky | +0.155 |
end | +0.142 |
gloomy | +0.140 |
Neg logit contributions | |
snatch | -0.246 |
zip | -0.246 |
Flo | -0.244 |
ck | -0.244 |
zzy | -0.234 |
Token | something |
Token ID | 1223 |
Feature activation | -6.80e-02 |
Pos logit contributions | |
End | +0.018 |
problem | +0.017 |
sky | +0.016 |
happiest | +0.015 |
end | +0.015 |
Neg logit contributions | |
ck | -0.015 |
snatch | -0.015 |
zip | -0.015 |
Flo | -0.015 |
unw | -0.015 |
Token | special |
Token ID | 2041 |
Feature activation | -7.17e-02 |
Pos logit contributions | |
End | +0.085 |
problem | +0.074 |
sky | +0.073 |
happiest | +0.066 |
end | +0.065 |
Neg logit contributions | |
snatch | -0.139 |
ck | -0.138 |
zip | -0.135 |
Flo | -0.135 |
unw | -0.135 |
Token | . |
Token ID | 13 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
End | +0.125 |
problem | +0.115 |
happiest | +0.110 |
sky | +0.109 |
gloomy | +0.103 |
Neg logit contributions | |
ck | -0.107 |
snatch | -0.106 |
unw | -0.106 |
zip | -0.105 |
zzy | -0.103 |
Token | sometimes |
Token ID | 3360 |
Feature activation | -9.27e-02 |
Pos logit contributions | |
ck | +0.295 |
snatch | +0.286 |
Flo | +0.284 |
zip | +0.280 |
uzz | +0.277 |
Neg logit contributions | |
problem | -0.204 |
End | -0.204 |
sky | -0.204 |
happiest | -0.178 |
end | -0.168 |
Token | they |
Token ID | 484 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
End | +0.152 |
sky | +0.137 |
problem | +0.124 |
end | +0.117 |
happiest | +0.116 |
Neg logit contributions | |
ck | -0.150 |
steer | -0.146 |
snatch | -0.146 |
zip | -0.142 |
Flo | -0.140 |
Token | fight |
Token ID | 1907 |
Feature activation | +3.91e-02 |
Pos logit contributions | |
End | +0.192 |
problem | +0.186 |
sky | +0.178 |
end | +0.166 |
happiest | +0.164 |
Neg logit contributions | |
zip | -0.216 |
snatch | -0.216 |
ck | -0.214 |
Flo | -0.214 |
zzy | -0.205 |
Token | . |
Token ID | 13 |
Feature activation | -2.55e-02 |
Pos logit contributions | |
zip | +0.062 |
ck | +0.061 |
snatch | +0.060 |
Flo | +0.057 |
steer | +0.056 |
Neg logit contributions | |
End | -0.068 |
problem | -0.063 |
sky | -0.062 |
happiest | -0.060 |
gloomy | -0.056 |
Token | One |
Token ID | 1881 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
End | +0.037 |
problem | +0.035 |
sky | +0.035 |
end | +0.032 |
gloomy | +0.031 |
Neg logit contributions | |
ck | -0.047 |
snatch | -0.047 |
zip | -0.047 |
Flo | -0.046 |
zzy | -0.045 |
Token | day |
Token ID | 1110 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
zzy | +0.045 |
snatch | +0.045 |
ck | +0.045 |
steer | +0.045 |
zip | +0.045 |
Neg logit contributions | |
End | -0.016 |
problem | -0.014 |
sky | -0.013 |
happiest | -0.013 |
moons | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | -8.57e-02 |
Pos logit contributions | |
End | +0.289 |
problem | +0.267 |
sky | +0.267 |
happiest | +0.260 |
tallest | +0.228 |
Neg logit contributions | |
Flo | -0.508 |
zip | -0.503 |
unw | -0.498 |
snatch | -0.498 |
ck | -0.497 |
Token | they |
Token ID | 484 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
End | +0.098 |
problem | +0.094 |
sky | +0.089 |
end | +0.080 |
happiest | +0.079 |
Neg logit contributions | |
snatch | -0.183 |
zip | -0.182 |
ck | -0.181 |
Flo | -0.181 |
zzy | -0.175 |
Token | find |
Token ID | 1064 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
End | +0.200 |
problem | +0.195 |
sky | +0.188 |
end | +0.180 |
happiest | +0.174 |
Neg logit contributions | |
Flo | -0.214 |
zip | -0.210 |
ck | -0.210 |
snatch | -0.209 |
zzy | -0.201 |
Token | a |
Token ID | 257 |
Feature activation | +1.81e-03 |
Pos logit contributions | |
ck | +0.040 |
zip | +0.040 |
snatch | +0.040 |
Flo | +0.040 |
zzy | +0.039 |
Neg logit contributions | |
End | -0.039 |
problem | -0.037 |
sky | -0.036 |
happiest | -0.033 |
end | -0.033 |
Token | wallet |
Token ID | 13008 |
Feature activation | +3.03e-02 |
Pos logit contributions | |
snatch | +0.004 |
zip | +0.004 |
Flo | +0.004 |
ck | +0.004 |
unw | +0.004 |
Neg logit contributions | |
End | -0.002 |
sky | -0.002 |
problem | -0.002 |
happiest | -0.002 |
end | -0.002 |