Token | before |
Token ID | 878 |
Feature activation | -4.73e+00 |
Pos logit contributions | |
suddenly | +0.162 |
opened | +0.158 |
pretended | +0.154 |
Suddenly | +0.152 |
two | +0.151 |
Neg logit contributions | |
unatt | -0.217 |
forward | -0.214 |
beh | -0.212 |
ash | -0.203 |
baby | -0.202 |
Token | . |
Token ID | 13 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
suddenly | +0.100 |
opened | +0.094 |
a | +0.092 |
two | +0.088 |
pretended | +0.083 |
Neg logit contributions | |
beh | -0.213 |
rav | -0.197 |
ash | -0.197 |
unatt | -0.196 |
losers | -0.192 |
Token |
|
Token ID | 198 |
Feature activation | +7.38e-01 |
Pos logit contributions | |
beh | +0.066 |
forward | +0.065 |
unatt | +0.065 |
ash | +0.064 |
rav | +0.063 |
Neg logit contributions | |
suddenly | -0.046 |
opened | -0.044 |
pretended | -0.044 |
two | -0.041 |
Suddenly | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | +7.58e-01 |
Pos logit contributions | |
baby | +0.028 |
forward | +0.027 |
ash | +0.027 |
rav | +0.027 |
beh | +0.027 |
Neg logit contributions | |
pretended | -0.016 |
suddenly | -0.016 |
opened | -0.015 |
two | -0.013 |
Suddenly | -0.012 |
Token | " |
Token ID | 1 |
Feature activation | +8.14e+00 |
Pos logit contributions | |
beh | +0.028 |
unatt | +0.027 |
ash | +0.026 |
forward | +0.026 |
rav | +0.025 |
Neg logit contributions | |
suddenly | -0.020 |
opened | -0.019 |
two | -0.018 |
pretended | -0.018 |
Suddenly | -0.017 |
Token | Un |
Token ID | 3118 |
Feature activation | +1.67e+01 |
Pos logit contributions | |
beh | +0.265 |
unatt | +0.253 |
rav | +0.245 |
forward | +0.244 |
ash | +0.242 |
Neg logit contributions | |
suddenly | -0.228 |
opened | -0.221 |
pretended | -0.217 |
Suddenly | -0.212 |
two | -0.207 |
Token | ic |
Token ID | 291 |
Feature activation | +4.47e+00 |
Pos logit contributions | |
beh | +0.434 |
unatt | +0.417 |
ash | +0.405 |
forward | +0.402 |
rav | +0.384 |
Neg logit contributions | |
suddenly | -0.597 |
opened | -0.574 |
pretended | -0.565 |
two | -0.555 |
Suddenly | -0.543 |
Token | orn |
Token ID | 1211 |
Feature activation | -3.81e+00 |
Pos logit contributions | |
beh | +0.165 |
unatt | +0.159 |
ash | +0.158 |
forward | +0.158 |
rav | +0.155 |
Neg logit contributions | |
suddenly | -0.108 |
opened | -0.101 |
pretended | -0.100 |
Suddenly | -0.097 |
two | -0.094 |
Token | is |
Token ID | 318 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
suddenly | +0.090 |
opened | +0.085 |
pretended | +0.084 |
two | +0.080 |
Suddenly | +0.080 |
Neg logit contributions | |
beh | -0.141 |
unatt | -0.137 |
forward | -0.135 |
ash | -0.134 |
rav | -0.132 |
Token | a |
Token ID | 257 |
Feature activation | -7.15e+00 |
Pos logit contributions | |
beh | +0.071 |
unatt | +0.068 |
ash | +0.067 |
forward | +0.066 |
rav | +0.066 |
Neg logit contributions | |
suddenly | -0.048 |
opened | -0.046 |
pretended | -0.044 |
two | -0.043 |
Suddenly | -0.042 |
Token | rare |
Token ID | 4071 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
suddenly | +0.184 |
opened | +0.175 |
pretended | +0.171 |
two | +0.166 |
Suddenly | +0.161 |
Neg logit contributions | |
beh | -0.257 |
unatt | -0.248 |
ash | -0.244 |
forward | -0.242 |
rav | -0.237 |
Token | books |
Token ID | 3835 |
Feature activation | +2.93e+00 |
Pos logit contributions | |
beh | +0.058 |
ash | +0.056 |
forward | +0.055 |
unatt | +0.054 |
rav | +0.054 |
Neg logit contributions | |
suddenly | -0.049 |
opened | -0.048 |
pretended | -0.048 |
two | -0.047 |
a | -0.046 |
Token | first |
Token ID | 717 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
beh | +0.142 |
rav | +0.127 |
ocol | +0.125 |
chie | +0.123 |
mir | +0.123 |
Neg logit contributions | |
suddenly | -0.061 |
a | -0.058 |
opened | -0.057 |
two | -0.051 |
was | -0.049 |
Token | ?" |
Token ID | 1701 |
Feature activation | -6.09e-01 |
Pos logit contributions | |
beh | +0.016 |
rav | +0.014 |
chie | +0.014 |
ocol | +0.014 |
aring | +0.014 |
Neg logit contributions | |
suddenly | -0.007 |
a | -0.007 |
opened | -0.006 |
two | -0.006 |
pretended | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +6.56e-01 |
Pos logit contributions | |
suddenly | +0.011 |
opened | +0.011 |
pretended | +0.010 |
two | +0.010 |
Suddenly | +0.010 |
Neg logit contributions | |
beh | -0.026 |
unatt | -0.025 |
ash | -0.025 |
forward | -0.025 |
rav | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +6.54e-01 |
Pos logit contributions | |
beh | +0.028 |
unatt | +0.027 |
forward | +0.027 |
rav | +0.027 |
baby | +0.027 |
Neg logit contributions | |
suddenly | -0.012 |
pretended | -0.011 |
opened | -0.011 |
two | -0.010 |
Suddenly | -0.009 |
Token | Un |
Token ID | 3118 |
Feature activation | +1.66e+01 |
Pos logit contributions | |
beh | +0.024 |
ash | +0.024 |
unatt | +0.023 |
forward | +0.023 |
ocol | +0.022 |
Neg logit contributions | |
suddenly | -0.017 |
opened | -0.016 |
two | -0.016 |
pretended | -0.015 |
Suddenly | -0.015 |
Token | cle |
Token ID | 2375 |
Feature activation | +6.65e-01 |
Pos logit contributions | |
beh | +0.503 |
unatt | +0.481 |
ash | +0.478 |
losers | +0.464 |
forward | +0.459 |
Neg logit contributions | |
suddenly | -0.539 |
opened | -0.500 |
pretended | -0.488 |
two | -0.487 |
Suddenly | -0.472 |
Token | Tom |
Token ID | 4186 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
baby | +0.022 |
rav | +0.022 |
beh | +0.022 |
unatt | +0.021 |
aring | +0.021 |
Neg logit contributions | |
suddenly | -0.016 |
pretended | -0.015 |
Suddenly | -0.014 |
opened | -0.014 |
two | -0.013 |
Token | laughed |
Token ID | 13818 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
suddenly | +0.007 |
Suddenly | +0.006 |
pretended | +0.006 |
two | +0.006 |
opened | +0.005 |
Neg logit contributions | |
beh | -0.013 |
unatt | -0.013 |
forward | -0.013 |
aring | -0.013 |
ash | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
suddenly | +0.034 |
a | +0.031 |
two | +0.030 |
opened | +0.029 |
pretended | +0.028 |
Neg logit contributions | |
beh | -0.114 |
unatt | -0.106 |
ash | -0.106 |
rav | -0.105 |
mir | -0.105 |
Token | said |
Token ID | 531 |
Feature activation | -3.96e+00 |
Pos logit contributions | |
beh | +0.072 |
unatt | +0.069 |
losers | +0.068 |
forward | +0.068 |
rav | +0.068 |
Neg logit contributions | |
suddenly | -0.024 |
opened | -0.020 |
Suddenly | -0.020 |
two | -0.019 |
pretended | -0.019 |
Token | sign |
Token ID | 1051 |
Feature activation | -9.31e-01 |
Pos logit contributions | |
suddenly | +0.174 |
opened | +0.165 |
pretended | +0.164 |
Suddenly | +0.154 |
two | +0.154 |
Neg logit contributions | |
beh | -0.264 |
unatt | -0.263 |
forward | -0.258 |
ash | -0.253 |
baby | -0.253 |
Token | that |
Token ID | 326 |
Feature activation | -5.19e+00 |
Pos logit contributions | |
suddenly | +0.021 |
pretended | +0.020 |
opened | +0.019 |
Suddenly | +0.018 |
two | +0.018 |
Neg logit contributions | |
beh | -0.035 |
unatt | -0.035 |
forward | -0.035 |
ash | -0.034 |
aring | -0.033 |
Token | says |
Token ID | 1139 |
Feature activation | -4.47e+00 |
Pos logit contributions | |
suddenly | +0.122 |
opened | +0.113 |
pretended | +0.112 |
two | +0.108 |
Suddenly | +0.106 |
Neg logit contributions | |
beh | -0.194 |
unatt | -0.188 |
forward | -0.186 |
ash | -0.185 |
rav | -0.182 |
Token | : |
Token ID | 25 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
suddenly | +0.109 |
opened | +0.107 |
two | +0.102 |
a | +0.101 |
pretended | +0.095 |
Neg logit contributions | |
beh | -0.184 |
unatt | -0.174 |
ash | -0.170 |
forward | -0.162 |
off | -0.160 |
Token | " |
Token ID | 366 |
Feature activation | +9.10e+00 |
Pos logit contributions | |
beh | +0.059 |
unatt | +0.058 |
ash | +0.056 |
agging | +0.053 |
losers | +0.053 |
Neg logit contributions | |
suddenly | -0.035 |
two | -0.035 |
a | -0.034 |
opened | -0.033 |
Suddenly | -0.032 |
Token | Un |
Token ID | 3118 |
Feature activation | +1.66e+01 |
Pos logit contributions | |
beh | +0.276 |
unatt | +0.275 |
forward | +0.264 |
ash | +0.263 |
aring | +0.255 |
Neg logit contributions | |
suddenly | -0.282 |
opened | -0.270 |
two | -0.263 |
pretended | -0.261 |
Suddenly | -0.252 |
Token | ite |
Token ID | 578 |
Feature activation | -4.82e-01 |
Pos logit contributions | |
beh | +0.485 |
unatt | +0.485 |
ash | +0.467 |
forward | +0.464 |
losers | +0.459 |
Neg logit contributions | |
suddenly | -0.530 |
opened | -0.503 |
two | -0.496 |
pretended | -0.493 |
Suddenly | -0.471 |
Token | the |
Token ID | 262 |
Feature activation | +7.42e-01 |
Pos logit contributions | |
opened | +0.009 |
a | +0.009 |
suddenly | +0.009 |
two | +0.009 |
was | +0.009 |
Neg logit contributions | |
beh | -0.021 |
unatt | -0.021 |
Acc | -0.021 |
agging | -0.020 |
ksh | -0.019 |
Token | hipp |
Token ID | 18568 |
Feature activation | +4.53e+00 |
Pos logit contributions | |
beh | +0.024 |
unatt | +0.022 |
ash | +0.022 |
forward | +0.022 |
rav | +0.021 |
Neg logit contributions | |
suddenly | -0.023 |
two | -0.022 |
opened | -0.022 |
pretended | -0.021 |
a | -0.021 |
Token | os |
Token ID | 418 |
Feature activation | +4.32e+00 |
Pos logit contributions | |
unatt | +0.154 |
beh | +0.144 |
forward | +0.142 |
future | +0.139 |
mir | +0.138 |
Neg logit contributions | |
two | -0.125 |
opened | -0.122 |
suddenly | -0.121 |
a | -0.117 |
was | -0.115 |
Token | ". |
Token ID | 1911 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
unatt | +0.166 |
beh | +0.164 |
Acc | +0.162 |
rav | +0.153 |
mir | +0.151 |
Neg logit contributions | |
two | -0.113 |
a | -0.109 |
opened | -0.108 |
suddenly | -0.102 |
spread | -0.100 |
Token | . |
Token ID | 13 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
beh | +0.131 |
ocol | +0.126 |
mir | +0.124 |
rav | +0.124 |
chie | +0.123 |
Neg logit contributions | |
a | -0.052 |
suddenly | -0.052 |
two | -0.046 |
was | -0.045 |
opened | -0.044 |
Token | They |
Token ID | 1119 |
Feature activation | -8.83e-01 |
Pos logit contributions | |
beh | +0.079 |
unatt | +0.077 |
ash | +0.073 |
forward | +0.072 |
rav | +0.071 |
Neg logit contributions | |
Suddenly | -0.036 |
suddenly | -0.036 |
opened | -0.035 |
two | -0.034 |
a | -0.033 |
Token | said |
Token ID | 531 |
Feature activation | -3.99e+00 |
Pos logit contributions | |
suddenly | +0.016 |
Suddenly | +0.015 |
opened | +0.014 |
two | +0.014 |
pretended | +0.014 |
Neg logit contributions | |
beh | -0.036 |
unatt | -0.035 |
forward | -0.035 |
losers | -0.034 |
ash | -0.034 |
Token | , |
Token ID | 11 |
Feature activation | -5.38e-01 |
Pos logit contributions | |
a | +0.032 |
suddenly | +0.031 |
two | +0.027 |
opened | +0.019 |
was | +0.014 |
Neg logit contributions | |
beh | -0.238 |
ash | -0.236 |
unatt | -0.235 |
opl | -0.229 |
ksh | -0.226 |
Token | " |
Token ID | 366 |
Feature activation | +9.16e+00 |
Pos logit contributions | |
suddenly | +0.007 |
a | +0.007 |
Suddenly | +0.007 |
opened | +0.007 |
two | +0.006 |
Neg logit contributions | |
beh | -0.031 |
ash | -0.029 |
unatt | -0.028 |
off | -0.028 |
chie | -0.027 |
Token | Un |
Token ID | 3118 |
Feature activation | +1.66e+01 |
Pos logit contributions | |
unatt | +0.291 |
beh | +0.285 |
ash | +0.279 |
forward | +0.278 |
aring | +0.265 |
Neg logit contributions | |
suddenly | -0.291 |
two | -0.269 |
pretended | -0.266 |
opened | -0.261 |
Suddenly | -0.257 |
Token | cle |
Token ID | 2375 |
Feature activation | +5.77e-01 |
Pos logit contributions | |
unatt | +0.469 |
beh | +0.463 |
ash | +0.447 |
losers | +0.444 |
forward | +0.441 |
Neg logit contributions | |
suddenly | -0.550 |
two | -0.514 |
opened | -0.511 |
pretended | -0.508 |
Suddenly | -0.489 |
Token | Tom |
Token ID | 4186 |
Feature activation | -3.76e-01 |
Pos logit contributions | |
beh | +0.026 |
unatt | +0.026 |
ash | +0.025 |
agging | +0.025 |
forward | +0.025 |
Neg logit contributions | |
suddenly | -0.010 |
two | -0.010 |
opened | -0.010 |
a | -0.009 |
pretended | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | -5.82e-01 |
Pos logit contributions | |
suddenly | +0.005 |
opened | +0.005 |
pretended | +0.005 |
two | +0.005 |
a | +0.004 |
Neg logit contributions | |
beh | -0.019 |
unatt | -0.018 |
ash | -0.018 |
forward | -0.018 |
rav | -0.017 |
Token | you |
Token ID | 345 |
Feature activation | +9.07e+00 |
Pos logit contributions | |
a | +0.015 |
two | +0.014 |
was | +0.012 |
opened | +0.012 |
there | +0.012 |
Neg logit contributions | |
beh | -0.026 |
agging | -0.025 |
umpy | -0.025 |
ivery | -0.024 |
ags | -0.024 |
Token | are |
Token ID | 389 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
beh | +0.265 |
losers | +0.244 |
ash | +0.242 |
aring | +0.242 |
rav | +0.238 |
Neg logit contributions | |
suddenly | -0.283 |
Suddenly | -0.267 |
opened | -0.257 |
pretended | -0.255 |
two | -0.242 |
Token | not |
Token ID | 407 |
Feature activation | +9.83e-01 |
Pos logit contributions | |
rav | +0.226 |
beh | +0.226 |
unatt | +0.222 |
forward | +0.216 |
ash | +0.216 |
Neg logit contributions | |
suddenly | -0.099 |
a | -0.098 |
two | -0.096 |
opened | -0.094 |
pretended | -0.085 |
Token | nice |
Token ID | 3621 |
Feature activation | +8.04e+00 |
Pos logit contributions | |
rav | +0.040 |
opl | +0.040 |
unatt | +0.039 |
forward | +0.039 |
agging | +0.039 |
Neg logit contributions | |
a | -0.027 |
two | -0.024 |
opened | -0.022 |
used | -0.019 |
put | -0.019 |
Token | ." |
Token ID | 526 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
beh | +0.369 |
unatt | +0.358 |
agging | +0.356 |
chie | +0.353 |
forward | +0.346 |
Neg logit contributions | |
a | -0.156 |
two | -0.153 |
was | -0.143 |
suddenly | -0.138 |
opened | -0.131 |
Token |
|
Token ID | 198 |
Feature activation | +6.71e-01 |
Pos logit contributions | |
beh | +0.097 |
unatt | +0.092 |
ash | +0.091 |
forward | +0.089 |
rav | +0.089 |
Neg logit contributions | |
suddenly | -0.042 |
opened | -0.039 |
two | -0.037 |
Suddenly | -0.037 |
pretended | -0.036 |
Token |
|
Token ID | 198 |
Feature activation | +6.87e-01 |
Pos logit contributions | |
beh | +0.029 |
unatt | +0.028 |
forward | +0.028 |
ash | +0.028 |
baby | +0.028 |
Neg logit contributions | |
suddenly | -0.012 |
pretended | -0.011 |
opened | -0.011 |
two | -0.010 |
Suddenly | -0.009 |
Token | Un |
Token ID | 3118 |
Feature activation | +1.66e+01 |
Pos logit contributions | |
beh | +0.027 |
unatt | +0.026 |
ash | +0.026 |
forward | +0.025 |
rav | +0.024 |
Neg logit contributions | |
suddenly | -0.017 |
two | -0.016 |
opened | -0.016 |
Suddenly | -0.015 |
a | -0.015 |
Token | cle |
Token ID | 2375 |
Feature activation | +5.45e-01 |
Pos logit contributions | |
beh | +0.518 |
unatt | +0.490 |
ash | +0.478 |
losers | +0.477 |
forward | +0.470 |
Neg logit contributions | |
suddenly | -0.530 |
opened | -0.494 |
two | -0.484 |
pretended | -0.469 |
Suddenly | -0.465 |
Token | Tom |
Token ID | 4186 |
Feature activation | -4.10e-01 |
Pos logit contributions | |
beh | +0.019 |
baby | +0.019 |
unatt | +0.019 |
forward | +0.019 |
ash | +0.019 |
Neg logit contributions | |
suddenly | -0.013 |
pretended | -0.012 |
opened | -0.011 |
Suddenly | -0.011 |
two | -0.011 |
Token | laughed |
Token ID | 13818 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
suddenly | +0.007 |
pretended | +0.007 |
opened | +0.006 |
Suddenly | +0.006 |
two | +0.006 |
Neg logit contributions | |
beh | -0.017 |
unatt | -0.017 |
forward | -0.017 |
ash | -0.016 |
baby | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
a | +0.045 |
two | +0.042 |
suddenly | +0.042 |
opened | +0.035 |
Suddenly | +0.034 |
Neg logit contributions | |
beh | -0.124 |
rav | -0.117 |
chie | -0.117 |
mir | -0.117 |
ocol | -0.116 |
Token | He |
Token ID | 679 |
Feature activation | -2.53e+00 |
Pos logit contributions | |
beh | +0.083 |
unatt | +0.079 |
ash | +0.076 |
rav | +0.075 |
forward | +0.074 |
Neg logit contributions | |
suddenly | -0.036 |
Suddenly | -0.036 |
opened | -0.035 |
two | -0.034 |
a | -0.033 |
Token | little |
Token ID | 1310 |
Feature activation | -4.78e+00 |
Pos logit contributions | |
baby | +0.026 |
ambulance | +0.026 |
unatt | +0.025 |
forward | +0.025 |
emergency | +0.024 |
Neg logit contributions | |
suddenly | -0.015 |
pretended | -0.012 |
opened | -0.012 |
clicked | -0.012 |
Suddenly | -0.012 |
Token | boy |
Token ID | 2933 |
Feature activation | -4.94e+00 |
Pos logit contributions | |
suddenly | +0.147 |
clicked | +0.142 |
opened | +0.138 |
inserted | +0.134 |
Suddenly | +0.134 |
Neg logit contributions | |
baby | -0.128 |
ambulance | -0.127 |
unatt | -0.124 |
forward | -0.124 |
future | -0.120 |
Token | said |
Token ID | 531 |
Feature activation | -4.21e+00 |
Pos logit contributions | |
suddenly | +0.086 |
pretended | +0.075 |
opened | +0.075 |
two | +0.075 |
Suddenly | +0.072 |
Neg logit contributions | |
beh | -0.208 |
unatt | -0.200 |
ash | -0.199 |
forward | -0.199 |
aring | -0.194 |
Token | , |
Token ID | 11 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
suddenly | +0.068 |
a | +0.062 |
two | +0.062 |
opened | +0.061 |
Suddenly | +0.054 |
Neg logit contributions | |
beh | -0.210 |
ash | -0.205 |
unatt | -0.203 |
opl | -0.200 |
ksh | -0.194 |
Token | " |
Token ID | 366 |
Feature activation | +8.98e+00 |
Pos logit contributions | |
suddenly | +0.026 |
two | +0.024 |
Suddenly | +0.024 |
opened | +0.024 |
a | +0.024 |
Neg logit contributions | |
beh | -0.053 |
ash | -0.050 |
unatt | -0.050 |
opl | -0.047 |
ocol | -0.047 |
Token | Un |
Token ID | 3118 |
Feature activation | +1.65e+01 |
Pos logit contributions | |
beh | +0.284 |
unatt | +0.280 |
ash | +0.268 |
forward | +0.266 |
aring | +0.261 |
Neg logit contributions | |
suddenly | -0.273 |
opened | -0.256 |
pretended | -0.253 |
two | -0.251 |
Suddenly | -0.246 |
Token | cle |
Token ID | 2375 |
Feature activation | +4.51e-01 |
Pos logit contributions | |
unatt | +0.466 |
ash | +0.461 |
beh | +0.456 |
losers | +0.442 |
forward | +0.432 |
Neg logit contributions | |
suddenly | -0.546 |
opened | -0.510 |
two | -0.509 |
pretended | -0.498 |
Suddenly | -0.487 |
Token | , |
Token ID | 11 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
beh | +0.019 |
unatt | +0.019 |
ash | +0.018 |
forward | +0.018 |
rav | +0.017 |
Neg logit contributions | |
suddenly | -0.009 |
opened | -0.008 |
pretended | -0.008 |
two | -0.008 |
Suddenly | -0.008 |
Token | are |
Token ID | 389 |
Feature activation | +2.72e+00 |
Pos logit contributions | |
two | +0.027 |
a | +0.027 |
suddenly | +0.026 |
opened | +0.026 |
pretended | +0.023 |
Neg logit contributions | |
beh | -0.050 |
ash | -0.048 |
agging | -0.048 |
ags | -0.047 |
opl | -0.047 |
Token | you |
Token ID | 345 |
Feature activation | +8.98e+00 |
Pos logit contributions | |
unatt | +0.069 |
beh | +0.067 |
ash | +0.065 |
agging | +0.063 |
opl | +0.063 |
Neg logit contributions | |
two | -0.099 |
suddenly | -0.099 |
a | -0.099 |
opened | -0.097 |
pretended | -0.093 |
Token | stealing |
Token ID | 16568 |
Feature activation | +5.30e-01 |
Pos logit contributions | |
beh | +0.363 |
unatt | +0.341 |
opl | +0.337 |
ash | +0.335 |
rav | +0.330 |
Neg logit contributions | |
two | -0.221 |
suddenly | -0.217 |
a | -0.212 |
opened | -0.205 |
pretended | -0.193 |
Token | came |
Token ID | 1625 |
Feature activation | +2.77e+00 |
Pos logit contributions | |
baby | +0.029 |
beh | +0.028 |
rav | +0.028 |
forward | +0.028 |
unatt | +0.028 |
Neg logit contributions | |
suddenly | -0.018 |
Suddenly | -0.016 |
pretended | -0.016 |
opened | -0.016 |
two | -0.016 |
Token | to |
Token ID | 284 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
beh | +0.079 |
forward | +0.076 |
unatt | +0.075 |
ash | +0.074 |
baby | +0.071 |
Neg logit contributions | |
suddenly | -0.092 |
opened | -0.087 |
pretended | -0.087 |
Suddenly | -0.083 |
two | -0.082 |
Token | visit |
Token ID | 3187 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
beh | +0.087 |
unatt | +0.084 |
ash | +0.083 |
forward | +0.083 |
rav | +0.081 |
Neg logit contributions | |
suddenly | -0.058 |
opened | -0.055 |
pretended | -0.054 |
two | -0.052 |
Suddenly | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
beh | +0.061 |
rav | +0.059 |
aring | +0.058 |
chie | +0.057 |
losers | +0.057 |
Neg logit contributions | |
a | -0.028 |
suddenly | -0.026 |
opened | -0.025 |
two | -0.023 |
was | -0.022 |
Token | " |
Token ID | 366 |
Feature activation | +8.87e+00 |
Pos logit contributions | |
beh | +0.054 |
unatt | +0.053 |
forward | +0.051 |
rav | +0.050 |
ash | +0.050 |
Neg logit contributions | |
opened | -0.021 |
Suddenly | -0.020 |
a | -0.019 |
suddenly | -0.019 |
Under | -0.018 |
Token | Un |
Token ID | 3118 |
Feature activation | +1.65e+01 |
Pos logit contributions | |
beh | +0.285 |
unatt | +0.275 |
ash | +0.270 |
forward | +0.268 |
rav | +0.261 |
Neg logit contributions | |
suddenly | -0.265 |
opened | -0.252 |
pretended | -0.249 |
two | -0.243 |
Suddenly | -0.237 |
Token | cle |
Token ID | 2375 |
Feature activation | +4.35e-01 |
Pos logit contributions | |
beh | +0.452 |
unatt | +0.451 |
ash | +0.448 |
losers | +0.434 |
forward | +0.428 |
Neg logit contributions | |
suddenly | -0.554 |
opened | -0.522 |
two | -0.516 |
pretended | -0.514 |
Suddenly | -0.494 |
Token | , |
Token ID | 11 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
beh | +0.019 |
unatt | +0.018 |
forward | +0.018 |
ash | +0.018 |
rav | +0.018 |
Neg logit contributions | |
suddenly | -0.008 |
opened | -0.007 |
pretended | -0.007 |
two | -0.007 |
Suddenly | -0.007 |
Token | I |
Token ID | 314 |
Feature activation | +4.84e+00 |
Pos logit contributions | |
suddenly | +0.032 |
opened | +0.031 |
two | +0.031 |
a | +0.031 |
pretended | +0.030 |
Neg logit contributions | |
beh | -0.044 |
ash | -0.042 |
unatt | -0.042 |
agging | -0.041 |
opl | -0.040 |
Token | want |
Token ID | 765 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
beh | +0.139 |
unatt | +0.132 |
forward | +0.132 |
ash | +0.129 |
rav | +0.127 |
Neg logit contributions | |
suddenly | -0.158 |
opened | -0.149 |
pretended | -0.146 |
two | -0.146 |
Suddenly | -0.145 |
Token | to |
Token ID | 284 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
beh | +0.109 |
forward | +0.106 |
agging | +0.104 |
opl | +0.103 |
umpy | +0.103 |
Neg logit contributions | |
a | -0.081 |
two | -0.075 |
suddenly | -0.069 |
opened | -0.069 |
pretended | -0.060 |
Token | my |
Token ID | 616 |
Feature activation | +4.22e+00 |
Pos logit contributions | |
beh | +0.202 |
unatt | +0.197 |
ash | +0.191 |
forward | +0.188 |
rav | +0.188 |
Neg logit contributions | |
suddenly | -0.191 |
opened | -0.182 |
pretended | -0.181 |
two | -0.178 |
a | -0.177 |
Token | dog |
Token ID | 3290 |
Feature activation | +9.52e-01 |
Pos logit contributions | |
beh | +0.141 |
forward | +0.141 |
unatt | +0.139 |
ash | +0.136 |
baby | +0.136 |
Neg logit contributions | |
suddenly | -0.113 |
opened | -0.110 |
pretended | -0.103 |
Suddenly | -0.101 |
clicked | -0.098 |
Token | ?" |
Token ID | 1701 |
Feature activation | -6.43e-01 |
Pos logit contributions | |
beh | +0.040 |
rav | +0.038 |
ash | +0.037 |
ocol | +0.037 |
unatt | +0.036 |
Neg logit contributions | |
suddenly | -0.022 |
a | -0.020 |
pretended | -0.020 |
opened | -0.020 |
two | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | +6.06e-01 |
Pos logit contributions | |
suddenly | +0.012 |
opened | +0.012 |
pretended | +0.012 |
two | +0.011 |
Suddenly | +0.010 |
Neg logit contributions | |
beh | -0.026 |
unatt | -0.026 |
forward | -0.026 |
ash | -0.025 |
rav | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +6.37e-01 |
Pos logit contributions | |
beh | +0.026 |
unatt | +0.025 |
forward | +0.025 |
ash | +0.025 |
baby | +0.025 |
Neg logit contributions | |
suddenly | -0.011 |
pretended | -0.010 |
opened | -0.010 |
two | -0.009 |
Suddenly | -0.009 |
Token | Un |
Token ID | 3118 |
Feature activation | +1.65e+01 |
Pos logit contributions | |
beh | +0.027 |
unatt | +0.026 |
ash | +0.026 |
forward | +0.025 |
rav | +0.024 |
Neg logit contributions | |
suddenly | -0.015 |
opened | -0.013 |
two | -0.013 |
pretended | -0.012 |
Suddenly | -0.012 |
Token | cle |
Token ID | 2375 |
Feature activation | +4.74e-01 |
Pos logit contributions | |
beh | +0.484 |
ash | +0.462 |
unatt | +0.460 |
losers | +0.444 |
forward | +0.444 |
Neg logit contributions | |
suddenly | -0.550 |
opened | -0.510 |
pretended | -0.499 |
two | -0.494 |
Suddenly | -0.481 |
Token | Ben |
Token ID | 3932 |
Feature activation | -2.70e+00 |
Pos logit contributions | |
baby | +0.015 |
rav | +0.014 |
aring | +0.014 |
ash | +0.014 |
beh | +0.014 |
Neg logit contributions | |
suddenly | -0.012 |
pretended | -0.011 |
Suddenly | -0.011 |
opened | -0.011 |
blasted | -0.010 |
Token | looked |
Token ID | 3114 |
Feature activation | -4.00e+00 |
Pos logit contributions | |
suddenly | +0.048 |
opened | +0.043 |
pretended | +0.043 |
two | +0.041 |
Suddenly | +0.040 |
Neg logit contributions | |
beh | -0.114 |
unatt | -0.112 |
forward | -0.110 |
ash | -0.109 |
rav | -0.108 |
Token | at |
Token ID | 379 |
Feature activation | +3.19e+00 |
Pos logit contributions | |
suddenly | +0.124 |
opened | +0.122 |
two | +0.118 |
pretended | +0.118 |
a | +0.116 |
Neg logit contributions | |
beh | -0.132 |
rav | -0.122 |
unatt | -0.121 |
ash | -0.116 |
aring | -0.114 |
Token | Lily |
Token ID | 20037 |
Feature activation | -7.84e-01 |
Pos logit contributions | |
forward | +0.105 |
unatt | +0.104 |
baby | +0.104 |
overs | +0.101 |
beh | +0.099 |
Neg logit contributions | |
suddenly | -0.082 |
opened | -0.080 |
pretended | -0.079 |
Suddenly | -0.073 |
clicked | -0.073 |
Token | down |
Token ID | 866 |
Feature activation | +3.07e+00 |
Pos logit contributions | |
beh | +0.033 |
unatt | +0.032 |
forward | +0.031 |
ash | +0.031 |
rav | +0.030 |
Neg logit contributions | |
suddenly | -0.034 |
opened | -0.033 |
pretended | -0.032 |
two | -0.031 |
Suddenly | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
beh | +0.134 |
ash | +0.128 |
unatt | +0.126 |
rav | +0.125 |
aring | +0.124 |
Neg logit contributions | |
suddenly | -0.065 |
opened | -0.061 |
two | -0.057 |
a | -0.056 |
pretended | -0.054 |
Token | |
Token ID | 220 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
beh | +0.051 |
unatt | +0.050 |
forward | +0.049 |
ash | +0.049 |
rav | +0.048 |
Neg logit contributions | |
suddenly | -0.030 |
opened | -0.029 |
pretended | -0.028 |
two | -0.027 |
Suddenly | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
beh | +0.089 |
forward | +0.087 |
ash | +0.087 |
rav | +0.085 |
unatt | +0.085 |
Neg logit contributions | |
suddenly | -0.039 |
pretended | -0.038 |
opened | -0.036 |
two | -0.032 |
Suddenly | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
beh | +0.013 |
forward | +0.013 |
unatt | +0.013 |
ash | +0.013 |
rav | +0.013 |
Neg logit contributions | |
suddenly | -0.005 |
pretended | -0.005 |
opened | -0.005 |
two | -0.004 |
Suddenly | -0.004 |
Token | Un |
Token ID | 3118 |
Feature activation | +1.65e+01 |
Pos logit contributions | |
unatt | +0.013 |
beh | +0.013 |
ash | +0.013 |
losers | +0.012 |
overs | +0.012 |
Neg logit contributions | |
suddenly | -0.007 |
two | -0.006 |
opened | -0.006 |
Suddenly | -0.006 |
pretended | -0.005 |
Token | cle |
Token ID | 2375 |
Feature activation | +5.80e-01 |
Pos logit contributions | |
beh | +0.455 |
unatt | +0.448 |
ash | +0.428 |
losers | +0.423 |
forward | +0.419 |
Neg logit contributions | |
suddenly | -0.569 |
opened | -0.535 |
pretended | -0.522 |
two | -0.520 |
Suddenly | -0.506 |
Token | looked |
Token ID | 3114 |
Feature activation | -3.94e+00 |
Pos logit contributions | |
baby | +0.021 |
rav | +0.021 |
beh | +0.021 |
forward | +0.021 |
ash | +0.020 |
Neg logit contributions | |
suddenly | -0.012 |
pretended | -0.011 |
Under | -0.011 |
opened | -0.011 |
Suddenly | -0.011 |
Token | at |
Token ID | 379 |
Feature activation | +3.17e+00 |
Pos logit contributions | |
suddenly | +0.120 |
opened | +0.116 |
pretended | +0.112 |
two | +0.110 |
Suddenly | +0.108 |
Neg logit contributions | |
beh | -0.128 |
unatt | -0.120 |
ash | -0.117 |
rav | -0.117 |
aring | -0.114 |
Token | her |
Token ID | 607 |
Feature activation | +3.39e+00 |
Pos logit contributions | |
forward | +0.110 |
beh | +0.110 |
unatt | +0.108 |
ash | +0.107 |
baby | +0.106 |
Neg logit contributions | |
suddenly | -0.080 |
opened | -0.077 |
pretended | -0.075 |
Suddenly | -0.071 |
two | -0.070 |
Token | and |
Token ID | 290 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
unatt | +0.130 |
forward | +0.129 |
beh | +0.128 |
ash | +0.123 |
aring | +0.122 |
Neg logit contributions | |
suddenly | -0.071 |
pretended | -0.069 |
opened | -0.068 |
Suddenly | -0.063 |
two | -0.063 |
Token | the |
Token ID | 262 |
Feature activation | +5.99e-01 |
Pos logit contributions | |
beh | +0.103 |
unatt | +0.100 |
forward | +0.098 |
ash | +0.098 |
rav | +0.096 |
Neg logit contributions | |
suddenly | -0.048 |
opened | -0.045 |
pretended | -0.044 |
two | -0.042 |
Suddenly | -0.041 |
Token | backyard |
Token ID | 24296 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
beh | +0.026 |
aring | +0.024 |
ash | +0.024 |
rav | +0.024 |
forward | +0.023 |
Neg logit contributions | |
suddenly | -0.014 |
two | -0.013 |
a | -0.013 |
opened | -0.013 |
pretended | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
suddenly | +0.061 |
opened | +0.058 |
two | +0.056 |
pretended | +0.055 |
a | +0.054 |
Neg logit contributions | |
beh | -0.098 |
unatt | -0.092 |
rav | -0.091 |
ash | -0.090 |
losers | -0.090 |
Token |
|
Token ID | 198 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
beh | +0.055 |
unatt | +0.054 |
ash | +0.052 |
forward | +0.052 |
rav | +0.051 |
Neg logit contributions | |
suddenly | -0.032 |
opened | -0.030 |
pretended | -0.030 |
two | -0.029 |
Suddenly | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | +4.31e-01 |
Pos logit contributions | |
forward | +0.016 |
beh | +0.016 |
ash | +0.016 |
rav | +0.015 |
aring | +0.015 |
Neg logit contributions | |
pretended | -0.009 |
suddenly | -0.008 |
opened | -0.008 |
two | -0.007 |
clicked | -0.007 |
Token | Un |
Token ID | 3118 |
Feature activation | +1.65e+01 |
Pos logit contributions | |
beh | +0.019 |
unatt | +0.019 |
ash | +0.018 |
forward | +0.018 |
baby | +0.018 |
Neg logit contributions | |
suddenly | -0.008 |
opened | -0.007 |
two | -0.007 |
pretended | -0.007 |
Suddenly | -0.006 |
Token | cle |
Token ID | 2375 |
Feature activation | +3.89e-01 |
Pos logit contributions | |
beh | +0.473 |
unatt | +0.462 |
ash | +0.456 |
forward | +0.446 |
losers | +0.444 |
Neg logit contributions | |
suddenly | -0.562 |
opened | -0.516 |
two | -0.505 |
pretended | -0.499 |
Suddenly | -0.482 |
Token | sprayed |
Token ID | 35729 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
baby | +0.014 |
rav | +0.013 |
beh | +0.013 |
ash | +0.013 |
forward | +0.013 |
Neg logit contributions | |
suddenly | -0.008 |
pretended | -0.008 |
Suddenly | -0.008 |
opened | -0.007 |
Under | -0.007 |
Token | them |
Token ID | 606 |
Feature activation | +6.73e+00 |
Pos logit contributions | |
suddenly | +0.037 |
opened | +0.034 |
two | +0.033 |
pretended | +0.033 |
a | +0.032 |
Neg logit contributions | |
beh | -0.065 |
unatt | -0.063 |
aring | -0.061 |
rav | -0.061 |
ash | -0.061 |
Token | with |
Token ID | 351 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
beh | +0.284 |
unatt | +0.273 |
ash | +0.269 |
rav | +0.268 |
aring | +0.264 |
Neg logit contributions | |
suddenly | -0.140 |
opened | -0.132 |
two | -0.126 |
pretended | -0.125 |
a | -0.120 |
Token | the |
Token ID | 262 |
Feature activation | +6.91e-01 |
Pos logit contributions | |
suddenly | +0.042 |
opened | +0.040 |
pretended | +0.038 |
two | +0.037 |
Suddenly | +0.036 |
Neg logit contributions | |
beh | -0.079 |
unatt | -0.077 |
ash | -0.076 |
forward | -0.075 |
rav | -0.074 |
Token | the |
Token ID | 262 |
Feature activation | +8.62e-01 |
Pos logit contributions | |
suddenly | +0.081 |
opened | +0.077 |
pretended | +0.075 |
Suddenly | +0.070 |
two | +0.070 |
Neg logit contributions | |
beh | -0.158 |
unatt | -0.155 |
forward | -0.151 |
ash | -0.150 |
baby | -0.147 |
Token | coat |
Token ID | 13209 |
Feature activation | +5.38e-01 |
Pos logit contributions | |
beh | +0.037 |
rav | +0.034 |
aring | +0.033 |
ash | +0.033 |
unatt | +0.032 |
Neg logit contributions | |
two | -0.020 |
suddenly | -0.020 |
opened | -0.019 |
a | -0.019 |
pretended | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
beh | +0.022 |
rav | +0.021 |
ash | +0.020 |
mir | +0.020 |
ocol | +0.020 |
Neg logit contributions | |
suddenly | -0.013 |
opened | -0.012 |
two | -0.011 |
a | -0.011 |
pretended | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +3.45e-01 |
Pos logit contributions | |
beh | +0.060 |
unatt | +0.059 |
forward | +0.058 |
ash | +0.058 |
rav | +0.056 |
Neg logit contributions | |
suddenly | -0.027 |
opened | -0.025 |
pretended | -0.025 |
two | -0.024 |
Suddenly | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | +3.47e-01 |
Pos logit contributions | |
beh | +0.014 |
forward | +0.014 |
ash | +0.013 |
baby | +0.013 |
unatt | +0.013 |
Neg logit contributions | |
pretended | -0.007 |
suddenly | -0.007 |
opened | -0.006 |
two | -0.006 |
clicked | -0.005 |
Token | Un |
Token ID | 3118 |
Feature activation | +1.64e+01 |
Pos logit contributions | |
unatt | +0.014 |
beh | +0.014 |
ash | +0.014 |
overs | +0.013 |
rav | +0.013 |
Neg logit contributions | |
suddenly | -0.008 |
two | -0.007 |
opened | -0.007 |
Suddenly | -0.007 |
Suddenly | -0.007 |
Token | cle |
Token ID | 2375 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
beh | +0.469 |
unatt | +0.464 |
ash | +0.447 |
losers | +0.444 |
forward | +0.434 |
Neg logit contributions | |
suddenly | -0.552 |
opened | -0.513 |
two | -0.499 |
pretended | -0.495 |
Suddenly | -0.483 |
Token | Mike |
Token ID | 4995 |
Feature activation | +8.92e-01 |
Pos logit contributions | |
baby | +0.014 |
beh | +0.014 |
forward | +0.014 |
rav | +0.014 |
ash | +0.014 |
Neg logit contributions | |
suddenly | -0.008 |
pretended | -0.007 |
opened | -0.007 |
Suddenly | -0.007 |
Under | -0.007 |
Token | said |
Token ID | 531 |
Feature activation | -4.00e+00 |
Pos logit contributions | |
beh | +0.041 |
unatt | +0.040 |
forward | +0.040 |
ash | +0.040 |
rav | +0.039 |
Neg logit contributions | |
suddenly | -0.013 |
opened | -0.012 |
pretended | -0.012 |
two | -0.011 |
Suddenly | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
suddenly | +0.066 |
a | +0.059 |
two | +0.053 |
opened | +0.051 |
Suddenly | +0.047 |
Neg logit contributions | |
beh | -0.214 |
ash | -0.211 |
opl | -0.211 |
ags | -0.201 |
ksh | -0.199 |
Token | himself |
Token ID | 2241 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
beh | +0.092 |
unatt | +0.089 |
ash | +0.087 |
forward | +0.086 |
rav | +0.085 |
Neg logit contributions | |
suddenly | -0.063 |
opened | -0.059 |
two | -0.057 |
pretended | -0.057 |
Suddenly | -0.055 |
Token | it |
Token ID | 340 |
Feature activation | +3.85e+00 |
Pos logit contributions | |
beh | +0.145 |
unatt | +0.139 |
ash | +0.135 |
aring | +0.134 |
rav | +0.133 |
Neg logit contributions | |
suddenly | -0.095 |
two | -0.088 |
opened | -0.086 |
a | -0.086 |
pretended | -0.084 |
Token | back |
Token ID | 736 |
Feature activation | +1.07e+01 |
Pos logit contributions | |
ocol | +0.128 |
umm | +0.124 |
beh | +0.122 |
mir | +0.114 |
impact | +0.112 |
Neg logit contributions | |
a | -0.149 |
two | -0.132 |
opened | -0.119 |
suddenly | -0.117 |
was | -0.113 |
Token | !" |
Token ID | 2474 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
beh | +0.378 |
unatt | +0.354 |
rav | +0.351 |
forward | +0.346 |
ash | +0.345 |
Neg logit contributions | |
suddenly | -0.300 |
opened | -0.288 |
a | -0.286 |
two | -0.285 |
pretended | -0.273 |
Token |
|
Token ID | 198 |
Feature activation | +5.24e-01 |
Pos logit contributions | |
pretended | +0.008 |
suddenly | +0.008 |
opened | +0.008 |
two | +0.007 |
clicked | +0.006 |
Neg logit contributions | |
unatt | -0.015 |
beh | -0.015 |
forward | -0.015 |
baby | -0.014 |
aring | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +5.58e-01 |
Pos logit contributions | |
baby | +0.018 |
aring | +0.017 |
beh | +0.017 |
forward | +0.017 |
future | +0.017 |
Neg logit contributions | |
pretended | -0.014 |
opened | -0.013 |
suddenly | -0.012 |
spread | -0.011 |
blasted | -0.011 |
Token | Un |
Token ID | 3118 |
Feature activation | +1.64e+01 |
Pos logit contributions | |
beh | +0.018 |
unatt | +0.017 |
aring | +0.016 |
forward | +0.016 |
baby | +0.016 |
Neg logit contributions | |
pretended | -0.016 |
opened | -0.016 |
suddenly | -0.016 |
two | -0.014 |
Suddenly | -0.014 |
Token | cle |
Token ID | 2375 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
beh | +0.451 |
ash | +0.435 |
unatt | +0.433 |
forward | +0.425 |
losers | +0.410 |
Neg logit contributions | |
suddenly | -0.565 |
opened | -0.531 |
pretended | -0.516 |
two | -0.515 |
Suddenly | -0.499 |
Token | Jack |
Token ID | 3619 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
baby | +0.012 |
rav | +0.012 |
aring | +0.012 |
unatt | +0.012 |
beh | +0.012 |
Neg logit contributions | |
suddenly | -0.009 |
pretended | -0.009 |
Suddenly | -0.008 |
Under | -0.008 |
opened | -0.008 |
Token | laughed |
Token ID | 13818 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
suddenly | +0.042 |
Suddenly | +0.039 |
pretended | +0.038 |
two | +0.037 |
Under | +0.037 |
Neg logit contributions | |
forward | -0.113 |
unatt | -0.112 |
beh | -0.112 |
baby | -0.111 |
ash | -0.110 |
Token | and |
Token ID | 290 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
suddenly | +0.040 |
a | +0.039 |
two | +0.037 |
opened | +0.033 |
pretended | +0.032 |
Neg logit contributions | |
beh | -0.137 |
rav | -0.129 |
mir | -0.128 |
ocol | -0.127 |
ash | -0.127 |
Token | threw |
Token ID | 9617 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
beh | +0.056 |
unatt | +0.055 |
forward | +0.053 |
losers | +0.053 |
ash | +0.053 |
Neg logit contributions | |
suddenly | -0.018 |
opened | -0.016 |
pretended | -0.016 |
Suddenly | -0.015 |
two | -0.015 |
Token | him |
Token ID | 683 |
Feature activation | +4.85e+00 |
Pos logit contributions | |
suddenly | +0.057 |
opened | +0.054 |
pretended | +0.052 |
two | +0.051 |
Suddenly | +0.049 |
Neg logit contributions | |
beh | -0.097 |
unatt | -0.095 |
ash | -0.093 |
forward | -0.092 |
rav | -0.091 |
Token | . |
Token ID | 13 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
beh | +0.153 |
forward | +0.153 |
unatt | +0.149 |
aring | +0.147 |
ash | +0.142 |
Neg logit contributions | |
suddenly | -0.137 |
pretended | -0.132 |
opened | -0.131 |
two | -0.125 |
Suddenly | -0.122 |
Token |
|
Token ID | 198 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
beh | +0.038 |
forward | +0.037 |
unatt | +0.037 |
ash | +0.036 |
aring | +0.036 |
Neg logit contributions | |
suddenly | -0.022 |
pretended | -0.021 |
opened | -0.020 |
two | -0.019 |
clicked | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
forward | +0.005 |
off | +0.005 |
baby | +0.005 |
aring | +0.005 |
rav | +0.005 |
Neg logit contributions | |
pretended | -0.005 |
suddenly | -0.004 |
opened | -0.004 |
two | -0.004 |
clicked | -0.004 |
Token | " |
Token ID | 1 |
Feature activation | +7.31e+00 |
Pos logit contributions | |
beh | +0.007 |
unatt | +0.007 |
forward | +0.007 |
ash | +0.007 |
rav | +0.007 |
Neg logit contributions | |
suddenly | -0.003 |
opened | -0.003 |
pretended | -0.003 |
two | -0.003 |
Suddenly | -0.003 |
Token | Un |
Token ID | 3118 |
Feature activation | +1.64e+01 |
Pos logit contributions | |
beh | +0.204 |
unatt | +0.197 |
forward | +0.191 |
ash | +0.190 |
rav | +0.187 |
Neg logit contributions | |
suddenly | -0.245 |
opened | -0.237 |
pretended | -0.233 |
two | -0.227 |
Suddenly | -0.223 |
Token | cle |
Token ID | 2375 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
beh | +0.458 |
unatt | +0.451 |
ash | +0.449 |
losers | +0.439 |
forward | +0.419 |
Neg logit contributions | |
suddenly | -0.551 |
opened | -0.521 |
two | -0.508 |
pretended | -0.507 |
Suddenly | -0.493 |
Token | , |
Token ID | 11 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
beh | +0.011 |
unatt | +0.011 |
rav | +0.011 |
forward | +0.011 |
ash | +0.011 |
Neg logit contributions | |
suddenly | -0.006 |
opened | -0.005 |
pretended | -0.005 |
Suddenly | -0.005 |
two | -0.005 |
Token | what |
Token ID | 644 |
Feature activation | -3.42e+00 |
Pos logit contributions | |
suddenly | +0.037 |
opened | +0.035 |
pretended | +0.034 |
two | +0.034 |
a | +0.032 |
Neg logit contributions | |
beh | -0.050 |
unatt | -0.048 |
ash | -0.047 |
forward | -0.047 |
agging | -0.046 |
Token | are |
Token ID | 389 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
suddenly | +0.097 |
opened | +0.094 |
two | +0.091 |
pretended | +0.091 |
a | +0.087 |
Neg logit contributions | |
beh | -0.113 |
unatt | -0.112 |
ash | -0.107 |
forward | -0.107 |
rav | -0.103 |
Token | you |
Token ID | 345 |
Feature activation | +8.80e+00 |
Pos logit contributions | |
unatt | +0.066 |
ash | +0.064 |
opl | +0.063 |
impact | +0.062 |
beh | +0.061 |
Neg logit contributions | |
a | -0.095 |
two | -0.094 |
suddenly | -0.093 |
opened | -0.092 |
pretended | -0.083 |
Token | touched |
Token ID | 12615 |
Feature activation | -3.11e+00 |
Pos logit contributions | |
suddenly | +0.051 |
opened | +0.047 |
pretended | +0.045 |
two | +0.043 |
Suddenly | +0.042 |
Neg logit contributions | |
beh | -0.132 |
unatt | -0.129 |
ash | -0.127 |
forward | -0.127 |
rav | -0.124 |
Token | them |
Token ID | 606 |
Feature activation | +6.54e+00 |
Pos logit contributions | |
a | +0.053 |
suddenly | +0.050 |
two | +0.048 |
was | +0.039 |
opened | +0.038 |
Neg logit contributions | |
beh | -0.153 |
unatt | -0.146 |
rav | -0.144 |
aring | -0.144 |
vern | -0.143 |
Token | and |
Token ID | 290 |
Feature activation | +9.79e-01 |
Pos logit contributions | |
beh | +0.277 |
rav | +0.263 |
ash | +0.257 |
unatt | +0.256 |
mir | +0.253 |
Neg logit contributions | |
suddenly | -0.149 |
opened | -0.137 |
two | -0.131 |
a | -0.129 |
pretended | -0.128 |
Token | said |
Token ID | 531 |
Feature activation | -4.26e+00 |
Pos logit contributions | |
beh | +0.039 |
unatt | +0.038 |
ash | +0.037 |
forward | +0.037 |
rav | +0.036 |
Neg logit contributions | |
suddenly | -0.022 |
opened | -0.020 |
pretended | -0.020 |
two | -0.019 |
Suddenly | -0.018 |
Token | " |
Token ID | 366 |
Feature activation | +8.92e+00 |
Pos logit contributions | |
a | +0.079 |
suddenly | +0.077 |
two | +0.068 |
Suddenly | +0.065 |
opened | +0.062 |
Neg logit contributions | |
beh | -0.215 |
opl | -0.208 |
ash | -0.208 |
unatt | -0.205 |
ags | -0.194 |
Token | Un |
Token ID | 3118 |
Feature activation | +1.63e+01 |
Pos logit contributions | |
unatt | +0.266 |
beh | +0.263 |
forward | +0.255 |
overs | +0.254 |
mir | +0.254 |
Neg logit contributions | |
suddenly | -0.292 |
pretended | -0.261 |
two | -0.258 |
opened | -0.257 |
Suddenly | -0.251 |
Token | free |
Token ID | 5787 |
Feature activation | +3.61e+00 |
Pos logit contributions | |
unatt | +0.459 |
beh | +0.450 |
ash | +0.444 |
forward | +0.439 |
agging | +0.428 |
Neg logit contributions | |
suddenly | -0.540 |
pretended | -0.503 |
two | -0.500 |
opened | -0.497 |
Suddenly | -0.474 |
Token | ze |
Token ID | 2736 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
beh | +0.116 |
unatt | +0.111 |
forward | +0.108 |
ash | +0.105 |
rav | +0.099 |
Neg logit contributions | |
suddenly | -0.113 |
two | -0.106 |
a | -0.105 |
pretended | -0.103 |
opened | -0.101 |
Token | !" |
Token ID | 2474 |
Feature activation | -7.45e-01 |
Pos logit contributions | |
beh | +0.063 |
unatt | +0.060 |
ocol | +0.059 |
mir | +0.058 |
opl | +0.057 |
Neg logit contributions | |
suddenly | -0.032 |
a | -0.029 |
two | -0.028 |
was | -0.027 |
Suddenly | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | +8.78e-02 |
Pos logit contributions | |
suddenly | +0.017 |
opened | +0.016 |
pretended | +0.016 |
two | +0.015 |
Suddenly | +0.014 |
Neg logit contributions | |
beh | -0.028 |
unatt | -0.027 |
ash | -0.027 |
forward | -0.027 |
aring | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | +9.79e-02 |
Pos logit contributions | |
ash | +0.003 |
aring | +0.003 |
baby | +0.003 |
off | +0.003 |
beh | +0.003 |
Neg logit contributions | |
pretended | -0.002 |
opened | -0.002 |
suddenly | -0.002 |
two | -0.002 |
spread | -0.002 |
Token | than |
Token ID | 621 |
Feature activation | +3.45e+00 |
Pos logit contributions | |
beh | +0.051 |
unatt | +0.049 |
ash | +0.048 |
forward | +0.048 |
rav | +0.047 |
Neg logit contributions | |
suddenly | -0.051 |
opened | -0.048 |
pretended | -0.048 |
two | -0.046 |
Suddenly | -0.045 |
Token | Uncle |
Token ID | 23169 |
Feature activation | +3.01e+00 |
Pos logit contributions | |
beh | +0.141 |
unatt | +0.134 |
ash | +0.133 |
aring | +0.132 |
rav | +0.130 |
Neg logit contributions | |
suddenly | -0.077 |
opened | -0.072 |
two | -0.069 |
pretended | -0.069 |
a | -0.067 |
Token | ! |
Token ID | 0 |
Feature activation | +8.00e-01 |
Pos logit contributions | |
beh | +0.122 |
unatt | +0.119 |
ash | +0.115 |
forward | +0.114 |
losers | +0.112 |
Neg logit contributions | |
suddenly | -0.068 |
opened | -0.065 |
pretended | -0.062 |
two | -0.061 |
a | -0.059 |
Token |
|
Token ID | 198 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
beh | +0.035 |
unatt | +0.035 |
ash | +0.033 |
agging | +0.033 |
rav | +0.032 |
Neg logit contributions | |
suddenly | -0.016 |
Suddenly | -0.015 |
two | -0.015 |
opened | -0.014 |
a | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | +4.04e-01 |
Pos logit contributions | |
beh | +0.018 |
unatt | +0.018 |
forward | +0.017 |
ash | +0.017 |
rav | +0.017 |
Neg logit contributions | |
suddenly | -0.006 |
opened | -0.006 |
pretended | -0.006 |
two | -0.005 |
Suddenly | -0.005 |
Token | Un |
Token ID | 3118 |
Feature activation | +1.63e+01 |
Pos logit contributions | |
unatt | +0.016 |
beh | +0.016 |
ash | +0.016 |
ags | +0.015 |
ocol | +0.015 |
Neg logit contributions | |
suddenly | -0.010 |
two | -0.009 |
Suddenly | -0.009 |
As | -0.009 |
Suddenly | -0.009 |
Token | cle |
Token ID | 2375 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
beh | +0.470 |
unatt | +0.463 |
ash | +0.445 |
losers | +0.443 |
forward | +0.432 |
Neg logit contributions | |
suddenly | -0.550 |
opened | -0.515 |
two | -0.506 |
pretended | -0.494 |
Suddenly | -0.487 |
Token | was |
Token ID | 373 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
beh | +0.013 |
rav | +0.012 |
baby | +0.012 |
forward | +0.012 |
ash | +0.012 |
Neg logit contributions | |
suddenly | -0.006 |
pretended | -0.006 |
opened | -0.006 |
Suddenly | -0.006 |
two | -0.005 |
Token | so |
Token ID | 523 |
Feature activation | +8.84e-01 |
Pos logit contributions | |
suddenly | +0.040 |
opened | +0.038 |
pretended | +0.037 |
two | +0.036 |
Suddenly | +0.035 |
Neg logit contributions | |
beh | -0.053 |
unatt | -0.052 |
forward | -0.051 |
ash | -0.051 |
rav | -0.049 |
Token | happy |
Token ID | 3772 |
Feature activation | +4.91e+00 |
Pos logit contributions | |
beh | +0.026 |
unatt | +0.026 |
ash | +0.025 |
forward | +0.024 |
rav | +0.024 |
Neg logit contributions | |
suddenly | -0.029 |
opened | -0.028 |
two | -0.027 |
pretended | -0.027 |
Suddenly | -0.026 |
Token | he |
Token ID | 339 |
Feature activation | -6.19e-01 |
Pos logit contributions | |
beh | +0.185 |
unatt | +0.179 |
forward | +0.176 |
ash | +0.175 |
rav | +0.172 |
Neg logit contributions | |
suddenly | -0.117 |
opened | -0.111 |
pretended | -0.108 |
two | -0.104 |
Suddenly | -0.102 |
Token | shrink |
Token ID | 22085 |
Feature activation | -4.91e+00 |
Pos logit contributions | |
two | +0.031 |
a | +0.031 |
opened | +0.031 |
suddenly | +0.030 |
pretended | +0.028 |
Neg logit contributions | |
beh | -0.045 |
agging | -0.041 |
ash | -0.041 |
unatt | -0.040 |
ags | -0.039 |
Token | me |
Token ID | 502 |
Feature activation | +8.92e+00 |
Pos logit contributions | |
suddenly | +0.109 |
two | +0.106 |
a | +0.105 |
opened | +0.101 |
pretended | +0.096 |
Neg logit contributions | |
beh | -0.204 |
unatt | -0.193 |
rav | -0.189 |
forward | -0.180 |
aring | -0.179 |
Token | !" |
Token ID | 2474 |
Feature activation | -7.74e-01 |
Pos logit contributions | |
beh | +0.335 |
rav | +0.324 |
unatt | +0.317 |
mir | +0.310 |
losers | +0.310 |
Neg logit contributions | |
suddenly | -0.218 |
opened | -0.213 |
two | -0.210 |
a | -0.209 |
pretended | -0.196 |
Token |
|
Token ID | 198 |
Feature activation | -4.30e-02 |
Pos logit contributions | |
suddenly | +0.018 |
opened | +0.017 |
pretended | +0.016 |
two | +0.016 |
Suddenly | +0.015 |
Neg logit contributions | |
beh | -0.030 |
unatt | -0.029 |
forward | -0.029 |
ash | -0.029 |
rav | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | -6.17e-02 |
Pos logit contributions | |
pretended | +0.001 |
suddenly | +0.001 |
opened | +0.001 |
two | +0.001 |
clicked | +0.001 |
Neg logit contributions | |
baby | -0.002 |
beh | -0.002 |
rav | -0.002 |
forward | -0.002 |
unatt | -0.002 |
Token | Un |
Token ID | 3118 |
Feature activation | +1.63e+01 |
Pos logit contributions | |
suddenly | +0.002 |
opened | +0.001 |
two | +0.001 |
pretended | +0.001 |
Suddenly | +0.001 |
Neg logit contributions | |
beh | -0.002 |
unatt | -0.002 |
ash | -0.002 |
forward | -0.002 |
rav | -0.002 |
Token | cle |
Token ID | 2375 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
beh | +0.482 |
unatt | +0.468 |
ash | +0.462 |
losers | +0.447 |
forward | +0.441 |
Neg logit contributions | |
suddenly | -0.529 |
opened | -0.496 |
pretended | -0.484 |
two | -0.483 |
Suddenly | -0.466 |
Token | Tom |
Token ID | 4186 |
Feature activation | -9.94e-01 |
Pos logit contributions | |
baby | +0.008 |
rav | +0.008 |
beh | +0.008 |
aring | +0.008 |
ash | +0.008 |
Neg logit contributions | |
suddenly | -0.005 |
pretended | -0.005 |
Suddenly | -0.005 |
opened | -0.004 |
Under | -0.004 |
Token | smiled |
Token ID | 13541 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
suddenly | +0.017 |
Suddenly | +0.015 |
two | +0.014 |
pretended | +0.013 |
Under | +0.013 |
Neg logit contributions | |
beh | -0.038 |
aring | -0.038 |
unatt | -0.038 |
ash | -0.037 |
forward | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | +8.26e-01 |
Pos logit contributions | |
suddenly | +0.021 |
opened | +0.020 |
pretended | +0.019 |
two | +0.019 |
Suddenly | +0.018 |
Neg logit contributions | |
beh | -0.053 |
unatt | -0.051 |
ash | -0.050 |
rav | -0.049 |
losers | -0.049 |
Token | put |
Token ID | 1234 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
beh | +0.036 |
losers | +0.034 |
forward | +0.034 |
rav | +0.034 |
unatt | +0.034 |
Neg logit contributions | |
suddenly | -0.013 |
Suddenly | -0.011 |
pretended | -0.010 |
opened | -0.010 |
two | -0.010 |
Token | welcome |
Token ID | 7062 |
Feature activation | +5.39e+00 |
Pos logit contributions | |
unatt | +0.022 |
ash | +0.022 |
rav | +0.022 |
overs | +0.022 |
opl | +0.021 |
Neg logit contributions | |
a | -0.022 |
two | -0.022 |
opened | -0.020 |
suddenly | -0.020 |
pretended | -0.020 |
Token | here |
Token ID | 994 |
Feature activation | -7.36e-01 |
Pos logit contributions | |
beh | +0.228 |
ash | +0.220 |
opl | +0.219 |
rav | +0.214 |
mir | +0.213 |
Neg logit contributions | |
a | -0.128 |
two | -0.124 |
suddenly | -0.120 |
there | -0.103 |
opened | -0.098 |
Token | ." |
Token ID | 526 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
a | +0.015 |
suddenly | +0.014 |
two | +0.013 |
was | +0.012 |
opened | +0.011 |
Neg logit contributions | |
beh | -0.036 |
ocol | -0.034 |
mir | -0.033 |
ash | -0.033 |
opl | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
beh | +0.078 |
unatt | +0.075 |
ash | +0.074 |
forward | +0.073 |
rav | +0.073 |
Neg logit contributions | |
suddenly | -0.022 |
opened | -0.020 |
two | -0.018 |
Suddenly | -0.018 |
pretended | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
unatt | +0.010 |
beh | +0.010 |
baby | +0.010 |
forward | +0.010 |
ash | +0.010 |
Neg logit contributions | |
pretended | -0.005 |
suddenly | -0.005 |
opened | -0.005 |
two | -0.004 |
Suddenly | -0.004 |
Token | Un |
Token ID | 3118 |
Feature activation | +1.63e+01 |
Pos logit contributions | |
beh | +0.010 |
ash | +0.010 |
forward | +0.010 |
unatt | +0.010 |
rav | +0.010 |
Neg logit contributions | |
suddenly | -0.007 |
opened | -0.006 |
two | -0.006 |
Suddenly | -0.006 |
pretended | -0.006 |
Token | cle |
Token ID | 2375 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
beh | +0.472 |
unatt | +0.447 |
ash | +0.446 |
losers | +0.434 |
forward | +0.432 |
Neg logit contributions | |
suddenly | -0.545 |
opened | -0.513 |
two | -0.501 |
pretended | -0.494 |
Suddenly | -0.481 |
Token | Jack |
Token ID | 3619 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
baby | +0.010 |
beh | +0.010 |
unatt | +0.009 |
rav | +0.009 |
forward | +0.009 |
Neg logit contributions | |
suddenly | -0.006 |
pretended | -0.006 |
Suddenly | -0.006 |
opened | -0.006 |
Under | -0.006 |
Token | gr |
Token ID | 1036 |
Feature activation | +1.11e+01 |
Pos logit contributions | |
suddenly | +0.049 |
Suddenly | +0.045 |
pretended | +0.044 |
two | +0.042 |
opened | +0.042 |
Neg logit contributions | |
beh | -0.103 |
unatt | -0.102 |
baby | -0.100 |
aring | -0.100 |
forward | -0.099 |
Token | umbled |
Token ID | 11137 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
unatt | +0.273 |
forward | +0.263 |
beh | +0.262 |
rav | +0.250 |
mir | +0.246 |
Neg logit contributions | |
suddenly | -0.397 |
opened | -0.394 |
two | -0.382 |
pretended | -0.377 |
Suddenly | -0.374 |
Token | and |
Token ID | 290 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
suddenly | +0.003 |
opened | +0.003 |
two | +0.003 |
a | +0.003 |
pretended | +0.003 |
Neg logit contributions | |
beh | -0.006 |
ash | -0.006 |
rav | -0.006 |
unatt | -0.006 |
mir | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
suddenly | +0.049 |
opened | +0.046 |
two | +0.044 |
pretended | +0.043 |
a | +0.042 |
Neg logit contributions | |
beh | -0.102 |
rav | -0.098 |
unatt | -0.097 |
ash | -0.096 |
baby | -0.094 |
Token | Tim |
Token ID | 5045 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
ash | +0.047 |
forward | +0.046 |
beh | +0.046 |
unatt | +0.045 |
overs | +0.045 |
Neg logit contributions | |
suddenly | -0.034 |
opened | -0.034 |
pretended | -0.033 |
two | -0.030 |
clicked | -0.028 |
Token | said |
Token ID | 531 |
Feature activation | -4.39e+00 |
Pos logit contributions | |
ash | +0.036 |
beh | +0.035 |
baby | +0.034 |
unatt | +0.034 |
off | +0.033 |
Neg logit contributions | |
suddenly | -0.024 |
two | -0.022 |
Suddenly | -0.021 |
opened | -0.021 |
a | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
suddenly | +0.055 |
a | +0.048 |
two | +0.046 |
opened | +0.044 |
Suddenly | +0.036 |
Neg logit contributions | |
beh | -0.243 |
ash | -0.235 |
unatt | -0.233 |
opl | -0.231 |
ksh | -0.226 |
Token | " |
Token ID | 366 |
Feature activation | +8.85e+00 |
Pos logit contributions | |
suddenly | +0.023 |
a | +0.021 |
Suddenly | +0.021 |
two | +0.021 |
opened | +0.021 |
Neg logit contributions | |
beh | -0.051 |
unatt | -0.048 |
ash | -0.047 |
opl | -0.046 |
off | -0.045 |
Token | Un |
Token ID | 3118 |
Feature activation | +1.63e+01 |
Pos logit contributions | |
beh | +0.318 |
unatt | +0.309 |
forward | +0.301 |
ash | +0.301 |
rav | +0.294 |
Neg logit contributions | |
suddenly | -0.230 |
opened | -0.218 |
pretended | -0.214 |
two | -0.208 |
Suddenly | -0.203 |
Token | cle |
Token ID | 2375 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
unatt | +0.506 |
beh | +0.506 |
ash | +0.498 |
forward | +0.486 |
losers | +0.481 |
Neg logit contributions | |
suddenly | -0.492 |
two | -0.454 |
opened | -0.451 |
pretended | -0.446 |
Suddenly | -0.434 |
Token | Jack |
Token ID | 3619 |
Feature activation | -2.96e+00 |
Pos logit contributions | |
beh | +0.007 |
unatt | +0.007 |
ash | +0.007 |
forward | +0.007 |
rav | +0.006 |
Neg logit contributions | |
suddenly | -0.003 |
opened | -0.003 |
pretended | -0.002 |
two | -0.002 |
Suddenly | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
suddenly | +0.042 |
opened | +0.040 |
two | +0.037 |
pretended | +0.037 |
a | +0.034 |
Neg logit contributions | |
beh | -0.143 |
unatt | -0.140 |
forward | -0.136 |
ash | -0.134 |
rav | -0.132 |
Token | let |
Token ID | 1309 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
a | +0.031 |
two | +0.031 |
suddenly | +0.030 |
opened | +0.030 |
pretended | +0.028 |
Neg logit contributions | |
beh | -0.048 |
agging | -0.046 |
ash | -0.045 |
unatt | -0.044 |
ags | -0.044 |
Token | 's |
Token ID | 338 |
Feature activation | +4.81e+00 |
Pos logit contributions | |
two | +0.082 |
a | +0.079 |
suddenly | +0.075 |
opened | +0.071 |
was | +0.071 |
Neg logit contributions | |
beh | -0.099 |
unatt | -0.098 |
agging | -0.097 |
rav | -0.096 |
aring | -0.095 |
Token | 's |
Token ID | 338 |
Feature activation | +4.98e+00 |
Pos logit contributions | |
beh | +0.081 |
unatt | +0.079 |
forward | +0.077 |
ash | +0.077 |
rav | +0.075 |
Neg logit contributions | |
suddenly | -0.057 |
opened | -0.054 |
pretended | -0.053 |
two | -0.052 |
Suddenly | -0.050 |
Token | silly |
Token ID | 14397 |
Feature activation | +3.25e+00 |
Pos logit contributions | |
beh | +0.192 |
forward | +0.187 |
unatt | +0.184 |
ash | +0.181 |
losers | +0.178 |
Neg logit contributions | |
suddenly | -0.112 |
opened | -0.107 |
pretended | -0.105 |
Suddenly | -0.096 |
two | -0.091 |
Token | !" |
Token ID | 2474 |
Feature activation | -8.20e-01 |
Pos logit contributions | |
unatt | +0.141 |
beh | +0.138 |
forward | +0.130 |
ash | +0.128 |
ocol | +0.128 |
Neg logit contributions | |
suddenly | -0.064 |
two | -0.063 |
a | -0.061 |
pretended | -0.059 |
opened | -0.059 |
Token |
|
Token ID | 198 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
suddenly | +0.014 |
opened | +0.013 |
pretended | +0.013 |
two | +0.012 |
Suddenly | +0.011 |
Neg logit contributions | |
beh | -0.035 |
forward | -0.035 |
unatt | -0.035 |
ash | -0.034 |
rav | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | +9.46e-02 |
Pos logit contributions | |
beh | +0.005 |
unatt | +0.005 |
forward | +0.005 |
ash | +0.005 |
rav | +0.005 |
Neg logit contributions | |
suddenly | -0.002 |
opened | -0.001 |
pretended | -0.001 |
two | -0.001 |
Suddenly | -0.001 |
Token | Un |
Token ID | 3118 |
Feature activation | +1.62e+01 |
Pos logit contributions | |
beh | +0.004 |
unatt | +0.004 |
ocol | +0.003 |
ash | +0.003 |
losers | +0.003 |
Neg logit contributions | |
suddenly | -0.002 |
opened | -0.002 |
two | -0.002 |
Suddenly | -0.002 |
Suddenly | -0.002 |
Token | cle |
Token ID | 2375 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
beh | +0.473 |
unatt | +0.462 |
losers | +0.448 |
ash | +0.445 |
mir | +0.434 |
Neg logit contributions | |
suddenly | -0.539 |
opened | -0.505 |
pretended | -0.492 |
two | -0.492 |
Suddenly | -0.476 |
Token | smiled |
Token ID | 13541 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
beh | +0.008 |
baby | +0.008 |
rav | +0.008 |
forward | +0.008 |
ash | +0.008 |
Neg logit contributions | |
suddenly | -0.005 |
pretended | -0.004 |
opened | -0.004 |
Suddenly | -0.004 |
two | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | +8.06e-01 |
Pos logit contributions | |
suddenly | +0.031 |
opened | +0.029 |
pretended | +0.028 |
two | +0.028 |
Suddenly | +0.027 |
Neg logit contributions | |
beh | -0.052 |
unatt | -0.050 |
ash | -0.049 |
forward | -0.049 |
rav | -0.048 |
Token | said |
Token ID | 531 |
Feature activation | -4.50e+00 |
Pos logit contributions | |
beh | +0.037 |
unatt | +0.035 |
forward | +0.035 |
ash | +0.035 |
rav | +0.035 |
Neg logit contributions | |
suddenly | -0.012 |
opened | -0.010 |
pretended | -0.010 |
two | -0.010 |
Suddenly | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
suddenly | +0.047 |
a | +0.040 |
opened | +0.038 |
two | +0.035 |
Suddenly | +0.033 |
Neg logit contributions | |
beh | -0.249 |
opl | -0.246 |
unatt | -0.243 |
ash | -0.243 |
ags | -0.233 |
Token | ement |
Token ID | 972 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
suddenly | +0.158 |
opened | +0.153 |
two | +0.146 |
pretended | +0.145 |
a | +0.142 |
Neg logit contributions | |
beh | -0.111 |
unatt | -0.103 |
ash | -0.099 |
rav | -0.099 |
forward | -0.098 |
Token | . |
Token ID | 13 |
Feature activation | +8.25e-01 |
Pos logit contributions | |
beh | +0.076 |
unatt | +0.074 |
ash | +0.072 |
forward | +0.071 |
rav | +0.071 |
Neg logit contributions | |
suddenly | -0.048 |
opened | -0.045 |
pretended | -0.043 |
two | -0.043 |
a | -0.041 |
Token | |
Token ID | 220 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
beh | +0.032 |
unatt | +0.031 |
forward | +0.030 |
ash | +0.030 |
rav | +0.030 |
Neg logit contributions | |
suddenly | -0.019 |
opened | -0.018 |
pretended | -0.017 |
two | -0.017 |
Suddenly | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
beh | +0.066 |
ash | +0.065 |
aring | +0.065 |
forward | +0.065 |
rav | +0.063 |
Neg logit contributions | |
pretended | -0.037 |
suddenly | -0.037 |
opened | -0.035 |
two | -0.030 |
spread | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | +3.51e-03 |
Pos logit contributions | |
aring | +0.001 |
forward | +0.001 |
off | +0.001 |
rav | +0.001 |
beh | +0.001 |
Neg logit contributions | |
pretended | -0.001 |
suddenly | -0.000 |
opened | -0.000 |
two | -0.000 |
clicked | -0.000 |
Token | Un |
Token ID | 3118 |
Feature activation | +1.62e+01 |
Pos logit contributions | |
beh | +0.000 |
unatt | +0.000 |
ash | +0.000 |
forward | +0.000 |
rav | +0.000 |
Neg logit contributions | |
suddenly | -0.000 |
opened | -0.000 |
pretended | -0.000 |
two | -0.000 |
Suddenly | -0.000 |
Token | able |
Token ID | 540 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
beh | +0.457 |
unatt | +0.442 |
ash | +0.436 |
forward | +0.426 |
losers | +0.419 |
Neg logit contributions | |
suddenly | -0.548 |
opened | -0.516 |
two | -0.501 |
pretended | -0.492 |
Suddenly | -0.480 |
Token | to |
Token ID | 284 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
beh | +0.063 |
aring | +0.062 |
baby | +0.062 |
forward | +0.061 |
rav | +0.060 |
Neg logit contributions | |
pretended | -0.048 |
suddenly | -0.048 |
Suddenly | -0.045 |
opened | -0.045 |
clicked | -0.042 |
Token | believe |
Token ID | 1975 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
beh | +0.076 |
forward | +0.074 |
unatt | +0.073 |
ash | +0.071 |
losers | +0.071 |
Neg logit contributions | |
suddenly | -0.050 |
pretended | -0.049 |
opened | -0.048 |
Suddenly | -0.044 |
clicked | -0.043 |
Token | his |
Token ID | 465 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
suddenly | +0.032 |
opened | +0.029 |
pretended | +0.027 |
two | +0.026 |
a | +0.026 |
Neg logit contributions | |
beh | -0.067 |
unatt | -0.065 |
ash | -0.065 |
rav | -0.063 |
forward | -0.063 |
Token | luck |
Token ID | 8458 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
beh | +0.083 |
unatt | +0.081 |
ash | +0.077 |
forward | +0.075 |
overs | +0.074 |
Neg logit contributions | |
two | -0.050 |
suddenly | -0.049 |
opened | -0.049 |
a | -0.045 |
pretended | -0.045 |
Token | day |
Token ID | 1110 |
Feature activation | -6.13e+00 |
Pos logit contributions | |
beh | +0.158 |
unatt | +0.153 |
ash | +0.149 |
forward | +0.148 |
rav | +0.145 |
Neg logit contributions | |
suddenly | -0.121 |
opened | -0.117 |
pretended | -0.113 |
two | -0.112 |
Suddenly | -0.107 |
Token | , |
Token ID | 11 |
Feature activation | -4.72e+00 |
Pos logit contributions | |
suddenly | +0.123 |
opened | +0.115 |
pretended | +0.112 |
two | +0.107 |
Suddenly | +0.104 |
Neg logit contributions | |
beh | -0.252 |
unatt | -0.246 |
forward | -0.241 |
ash | -0.241 |
rav | -0.236 |
Token | a |
Token ID | 257 |
Feature activation | -1.06e+01 |
Pos logit contributions | |
suddenly | +0.097 |
opened | +0.096 |
pretended | +0.092 |
Suddenly | +0.085 |
clicked | +0.085 |
Neg logit contributions | |
beh | -0.177 |
forward | -0.176 |
unatt | -0.176 |
overs | -0.169 |
ash | -0.168 |
Token | magical |
Token ID | 10883 |
Feature activation | -3.51e+00 |
Pos logit contributions | |
suddenly | +0.171 |
two | +0.156 |
opened | +0.155 |
a | +0.145 |
pretended | +0.141 |
Neg logit contributions | |
beh | -0.523 |
ash | -0.492 |
unatt | -0.483 |
opl | -0.472 |
aring | -0.471 |
Token | fire |
Token ID | 2046 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
suddenly | +0.091 |
opened | +0.087 |
two | +0.082 |
pretended | +0.082 |
a | +0.079 |
Neg logit contributions | |
beh | -0.132 |
unatt | -0.123 |
ash | -0.122 |
forward | -0.120 |
rav | -0.119 |
Token | fly |
Token ID | 12254 |
Feature activation | -1.42e+01 |
Pos logit contributions | |
suddenly | +0.025 |
opened | +0.023 |
pretended | +0.023 |
two | +0.022 |
Suddenly | +0.021 |
Neg logit contributions | |
beh | -0.038 |
unatt | -0.037 |
ash | -0.036 |
forward | -0.036 |
rav | -0.035 |
Token | flew |
Token ID | 13112 |
Feature activation | -3.68e+00 |
Pos logit contributions | |
suddenly | +0.271 |
opened | +0.253 |
pretended | +0.244 |
two | +0.230 |
Suddenly | +0.219 |
Neg logit contributions | |
beh | -0.613 |
unatt | -0.596 |
ash | -0.586 |
forward | -0.585 |
rav | -0.577 |
Token | by |
Token ID | 416 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
suddenly | +0.090 |
opened | +0.086 |
a | +0.084 |
two | +0.082 |
pretended | +0.077 |
Neg logit contributions | |
beh | -0.143 |
rav | -0.136 |
unatt | -0.135 |
ash | -0.133 |
aring | -0.133 |
Token | and |
Token ID | 290 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
suddenly | +0.004 |
opened | +0.004 |
pretended | +0.004 |
two | +0.004 |
Suddenly | +0.004 |
Neg logit contributions | |
beh | -0.009 |
unatt | -0.009 |
forward | -0.009 |
ash | -0.009 |
rav | -0.009 |
Token | gave |
Token ID | 2921 |
Feature activation | -9.11e-02 |
Pos logit contributions | |
beh | +0.007 |
unatt | +0.007 |
forward | +0.007 |
ash | +0.007 |
losers | +0.007 |
Neg logit contributions | |
suddenly | -0.003 |
opened | -0.002 |
pretended | -0.002 |
two | -0.002 |
Suddenly | -0.002 |
Token | him |
Token ID | 683 |
Feature activation | +4.27e+00 |
Pos logit contributions | |
suddenly | +0.002 |
two | +0.002 |
opened | +0.002 |
a | +0.002 |
pretended | +0.002 |
Neg logit contributions | |
beh | -0.003 |
unatt | -0.003 |
aring | -0.003 |
ash | -0.003 |
rav | -0.003 |
Token | it |
Token ID | 340 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
beh | +0.160 |
unatt | +0.154 |
forward | +0.149 |
rav | +0.142 |
aring | +0.140 |
Neg logit contributions | |
suddenly | -0.260 |
pretended | -0.250 |
opened | -0.247 |
Suddenly | -0.244 |
two | -0.241 |
Token | . |
Token ID | 13 |
Feature activation | -9.65e-01 |
Pos logit contributions | |
beh | +0.006 |
unatt | +0.005 |
ash | +0.005 |
rav | +0.005 |
baby | +0.005 |
Neg logit contributions | |
suddenly | -0.003 |
opened | -0.003 |
pretended | -0.003 |
two | -0.003 |
a | -0.003 |
Token | But |
Token ID | 887 |
Feature activation | -6.29e+00 |
Pos logit contributions | |
suddenly | +0.022 |
opened | +0.021 |
Suddenly | +0.020 |
pretended | +0.020 |
two | +0.020 |
Neg logit contributions | |
beh | -0.038 |
unatt | -0.037 |
ash | -0.036 |
forward | -0.036 |
rav | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
suddenly | +0.124 |
opened | +0.119 |
pretended | +0.116 |
two | +0.110 |
Suddenly | +0.108 |
Neg logit contributions | |
beh | -0.255 |
unatt | -0.251 |
forward | -0.246 |
ash | -0.244 |
rav | -0.239 |
Token | fire |
Token ID | 2046 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
suddenly | +0.034 |
opened | +0.032 |
two | +0.032 |
pretended | +0.031 |
a | +0.030 |
Neg logit contributions | |
beh | -0.044 |
unatt | -0.042 |
ash | -0.041 |
forward | -0.041 |
rav | -0.040 |
Token | fly |
Token ID | 12254 |
Feature activation | -1.37e+01 |
Pos logit contributions | |
suddenly | +0.009 |
opened | +0.008 |
pretended | +0.008 |
two | +0.008 |
was | +0.007 |
Neg logit contributions | |
beh | -0.018 |
unatt | -0.018 |
forward | -0.017 |
agging | -0.017 |
rav | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
suddenly | +0.262 |
opened | +0.245 |
pretended | +0.231 |
two | +0.216 |
was | +0.201 |
Neg logit contributions | |
beh | -0.603 |
unatt | -0.594 |
forward | -0.580 |
rav | -0.578 |
ash | -0.577 |
Token | too |
Token ID | 1165 |
Feature activation | +7.99e+00 |
Pos logit contributions | |
suddenly | +0.076 |
opened | +0.073 |
two | +0.070 |
a | +0.070 |
pretended | +0.068 |
Neg logit contributions | |
beh | -0.099 |
rav | -0.097 |
unatt | -0.094 |
ash | -0.093 |
baby | -0.091 |
Token | quick |
Token ID | 2068 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
beh | +0.325 |
unatt | +0.312 |
rav | +0.310 |
ash | +0.306 |
forward | +0.296 |
Neg logit contributions | |
suddenly | -0.187 |
opened | -0.181 |
two | -0.177 |
a | -0.173 |
pretended | -0.158 |
Token | and |
Token ID | 290 |
Feature activation | +4.78e-01 |
Pos logit contributions | |
suddenly | +0.004 |
a | +0.004 |
opened | +0.004 |
two | +0.004 |
was | +0.004 |
Neg logit contributions | |
beh | -0.012 |
unatt | -0.011 |
ash | -0.011 |
rav | -0.011 |
ags | -0.011 |
Token | flew |
Token ID | 13112 |
Feature activation | -2.96e+00 |
Pos logit contributions | |
beh | +0.020 |
ash | +0.020 |
unatt | +0.020 |
agging | +0.019 |
baby | +0.019 |
Neg logit contributions | |
suddenly | -0.010 |
opened | -0.010 |
pretended | -0.009 |
two | -0.009 |
a | -0.009 |
Token | on |
Token ID | 319 |
Feature activation | -4.42e+00 |
Pos logit contributions | |
baby | +0.012 |
forward | +0.012 |
ash | +0.012 |
unatt | +0.012 |
future | +0.012 |
Neg logit contributions | |
suddenly | -0.010 |
pretended | -0.010 |
opened | -0.009 |
two | -0.009 |
clicked | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | -1.12e+01 |
Pos logit contributions | |
suddenly | +0.087 |
opened | +0.081 |
two | +0.077 |
pretended | +0.077 |
a | +0.073 |
Neg logit contributions | |
beh | -0.190 |
unatt | -0.183 |
ash | -0.180 |
forward | -0.179 |
rav | -0.178 |
Token | sunny |
Token ID | 27737 |
Feature activation | +3.17e+00 |
Pos logit contributions | |
suddenly | +0.247 |
opened | +0.236 |
a | +0.221 |
pretended | +0.220 |
two | +0.218 |
Neg logit contributions | |
beh | -0.490 |
ash | -0.434 |
rav | -0.430 |
aring | -0.423 |
unatt | -0.421 |
Token | day |
Token ID | 1110 |
Feature activation | -4.32e+00 |
Pos logit contributions | |
unatt | +0.082 |
ash | +0.079 |
opl | +0.078 |
overs | +0.076 |
future | +0.076 |
Neg logit contributions | |
suddenly | -0.103 |
pretended | -0.102 |
opened | -0.099 |
clicked | -0.096 |
two | -0.094 |
Token | . |
Token ID | 13 |
Feature activation | -4.35e-01 |
Pos logit contributions | |
suddenly | +0.098 |
opened | +0.092 |
pretended | +0.088 |
two | +0.087 |
Suddenly | +0.083 |
Neg logit contributions | |
beh | -0.174 |
unatt | -0.167 |
ash | -0.163 |
forward | -0.161 |
rav | -0.161 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | -1.33e+01 |
Pos logit contributions | |
suddenly | +0.010 |
opened | +0.009 |
pretended | +0.009 |
two | +0.009 |
Suddenly | +0.008 |
Neg logit contributions | |
beh | -0.016 |
unatt | -0.016 |
forward | -0.016 |
ash | -0.016 |
rav | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
pretended | +0.251 |
opened | +0.241 |
suddenly | +0.235 |
Suddenly | +0.218 |
clicked | +0.211 |
Neg logit contributions | |
forward | -0.514 |
beh | -0.487 |
unatt | -0.487 |
rav | -0.477 |
baby | -0.477 |
Token | he |
Token ID | 339 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
pretended | +0.057 |
opened | +0.056 |
suddenly | +0.052 |
Suddenly | +0.051 |
clicked | +0.049 |
Neg logit contributions | |
forward | -0.110 |
unatt | -0.107 |
beh | -0.107 |
overs | -0.103 |
hurricane | -0.101 |
Token | stumbled |
Token ID | 24241 |
Feature activation | -3.91e+00 |
Pos logit contributions | |
Under | +0.038 |
Suddenly | +0.038 |
pretended | +0.037 |
two | +0.034 |
suddenly | +0.034 |
Neg logit contributions | |
forward | -0.097 |
unatt | -0.095 |
beh | -0.090 |
mir | -0.090 |
ash | -0.090 |
Token | across |
Token ID | 1973 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
pretended | +0.098 |
suddenly | +0.092 |
opened | +0.090 |
Suddenly | +0.086 |
two | +0.081 |
Neg logit contributions | |
forward | -0.140 |
unatt | -0.131 |
ash | -0.129 |
agging | -0.125 |
beh | -0.125 |
Token | something |
Token ID | 1223 |
Feature activation | -4.59e+00 |
Pos logit contributions | |
pretended | +0.018 |
suddenly | +0.016 |
Suddenly | +0.015 |
opened | +0.014 |
clicked | +0.013 |
Neg logit contributions | |
unatt | -0.056 |
beh | -0.055 |
permission | -0.055 |
forward | -0.055 |
future | -0.055 |
Token | a |
Token ID | 257 |
Feature activation | -1.18e+01 |
Pos logit contributions | |
suddenly | +0.054 |
opened | +0.050 |
pretended | +0.048 |
two | +0.046 |
Suddenly | +0.044 |
Neg logit contributions | |
beh | -0.146 |
unatt | -0.142 |
forward | -0.140 |
ash | -0.140 |
rav | -0.137 |
Token | toy |
Token ID | 13373 |
Feature activation | -4.70e+00 |
Pos logit contributions | |
suddenly | +0.299 |
opened | +0.284 |
pretended | +0.277 |
two | +0.271 |
Suddenly | +0.262 |
Neg logit contributions | |
beh | -0.432 |
unatt | -0.416 |
ash | -0.408 |
forward | -0.405 |
rav | -0.398 |
Token | sword |
Token ID | 8429 |
Feature activation | -8.25e+00 |
Pos logit contributions | |
suddenly | +0.118 |
opened | +0.111 |
pretended | +0.109 |
two | +0.106 |
Suddenly | +0.102 |
Neg logit contributions | |
beh | -0.174 |
unatt | -0.166 |
ash | -0.164 |
forward | -0.163 |
rav | -0.161 |
Token | . |
Token ID | 13 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
suddenly | +0.202 |
opened | +0.191 |
pretended | +0.186 |
two | +0.182 |
Suddenly | +0.179 |
Neg logit contributions | |
beh | -0.300 |
unatt | -0.294 |
forward | -0.286 |
ash | -0.285 |
rav | -0.278 |
Token | He |
Token ID | 679 |
Feature activation | -4.59e+00 |
Pos logit contributions | |
suddenly | +0.000 |
opened | +0.000 |
pretended | +0.000 |
two | +0.000 |
Suddenly | +0.000 |
Neg logit contributions | |
beh | -0.000 |
unatt | -0.000 |
forward | -0.000 |
ash | -0.000 |
rav | -0.000 |
Token | pretended |
Token ID | 36766 |
Feature activation | -1.33e+01 |
Pos logit contributions | |
suddenly | +0.095 |
Suddenly | +0.086 |
two | +0.085 |
source | +0.079 |
As | +0.076 |
Neg logit contributions | |
losers | -0.159 |
beh | -0.152 |
forward | -0.151 |
Go | -0.148 |
mir | -0.148 |
Token | it |
Token ID | 340 |
Feature activation | +2.81e+00 |
Pos logit contributions | |
suddenly | +0.343 |
opened | +0.327 |
pretended | +0.315 |
two | +0.308 |
a | +0.299 |
Neg logit contributions | |
beh | -0.487 |
unatt | -0.466 |
ash | -0.462 |
forward | -0.457 |
rav | -0.448 |
Token | was |
Token ID | 373 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
unatt | +0.105 |
beh | +0.104 |
baby | +0.104 |
forward | +0.103 |
ash | +0.100 |
Neg logit contributions | |
Suddenly | -0.043 |
sudden | -0.040 |
Suddenly | -0.039 |
suddenly | -0.038 |
two | -0.038 |
Token | a |
Token ID | 257 |
Feature activation | -8.79e+00 |
Pos logit contributions | |
suddenly | +0.027 |
opened | +0.024 |
pretended | +0.023 |
Suddenly | +0.022 |
two | +0.020 |
Neg logit contributions | |
beh | -0.117 |
forward | -0.117 |
unatt | -0.116 |
losers | -0.113 |
ash | -0.113 |
Token | real |
Token ID | 1103 |
Feature activation | +3.42e+00 |
Pos logit contributions | |
suddenly | +0.206 |
opened | +0.196 |
pretended | +0.190 |
two | +0.184 |
Suddenly | +0.181 |
Neg logit contributions | |
beh | -0.329 |
unatt | -0.323 |
forward | -0.317 |
ash | -0.314 |
rav | -0.309 |
Token | weapon |
Token ID | 4282 |
Feature activation | -4.41e+00 |
Pos logit contributions | |
forward | +0.119 |
unatt | +0.118 |
losers | +0.116 |
baby | +0.115 |
mir | +0.115 |
Neg logit contributions | |
suddenly | -0.081 |
opened | -0.079 |
Suddenly | -0.074 |
pretended | -0.073 |
two | -0.071 |
Token | and |
Token ID | 290 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
suddenly | +0.077 |
opened | +0.067 |
two | +0.066 |
Suddenly | +0.064 |
pretended | +0.064 |
Neg logit contributions | |
beh | -0.158 |
ash | -0.154 |
unatt | -0.154 |
forward | -0.153 |
aring | -0.150 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
beh | +0.019 |
unatt | +0.019 |
forward | +0.018 |
ash | +0.018 |
rav | +0.018 |
Neg logit contributions | |
suddenly | -0.009 |
opened | -0.009 |
pretended | -0.008 |
two | -0.008 |
Suddenly | -0.008 |
Token | followed |
Token ID | 3940 |
Feature activation | -3.82e+00 |
Pos logit contributions | |
suddenly | +0.037 |
opened | +0.032 |
Suddenly | +0.032 |
two | +0.031 |
pretended | +0.031 |
Neg logit contributions | |
beh | -0.062 |
unatt | -0.060 |
forward | -0.060 |
ash | -0.060 |
rav | -0.058 |
Token | the |
Token ID | 262 |
Feature activation | -6.91e-02 |
Pos logit contributions | |
suddenly | +0.086 |
a | +0.084 |
two | +0.079 |
opened | +0.075 |
pretended | +0.073 |
Neg logit contributions | |
beh | -0.163 |
unatt | -0.160 |
aring | -0.155 |
opl | -0.154 |
rav | -0.152 |
Token | fire |
Token ID | 2046 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
suddenly | +0.002 |
two | +0.001 |
opened | +0.001 |
pretended | +0.001 |
a | +0.001 |
Neg logit contributions | |
beh | -0.003 |
unatt | -0.003 |
ash | -0.003 |
forward | -0.003 |
overs | -0.003 |
Token | fly |
Token ID | 12254 |
Feature activation | -1.32e+01 |
Pos logit contributions | |
beh | +0.012 |
unatt | +0.012 |
ash | +0.011 |
forward | +0.011 |
rav | +0.011 |
Neg logit contributions | |
suddenly | -0.009 |
opened | -0.009 |
pretended | -0.009 |
two | -0.008 |
Suddenly | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
suddenly | +0.361 |
opened | +0.348 |
pretended | +0.339 |
two | +0.336 |
a | +0.330 |
Neg logit contributions | |
beh | -0.471 |
unatt | -0.452 |
rav | -0.445 |
ash | -0.442 |
baby | -0.438 |
Token | They |
Token ID | 1119 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
forward | +0.050 |
beh | +0.050 |
ash | +0.049 |
unatt | +0.049 |
overs | +0.048 |
Neg logit contributions | |
suddenly | -0.044 |
opened | -0.042 |
pretended | -0.042 |
two | -0.039 |
clicked | -0.037 |
Token | felt |
Token ID | 2936 |
Feature activation | -4.64e+00 |
Pos logit contributions | |
Suddenly | +0.023 |
suddenly | +0.022 |
two | +0.018 |
Suddenly | +0.018 |
Under | +0.018 |
Neg logit contributions | |
unatt | -0.050 |
forward | -0.048 |
losers | -0.048 |
beh | -0.048 |
mir | -0.048 |
Token | happy |
Token ID | 3772 |
Feature activation | +4.16e+00 |
Pos logit contributions | |
suddenly | +0.125 |
opened | +0.119 |
pretended | +0.117 |
two | +0.113 |
Suddenly | +0.111 |
Neg logit contributions | |
beh | -0.160 |
unatt | -0.156 |
forward | -0.153 |
ash | -0.151 |
rav | -0.148 |
Token | . |
Token ID | 13 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
beh | +0.168 |
unatt | +0.161 |
ash | +0.160 |
forward | +0.159 |
rav | +0.156 |
Neg logit contributions | |
suddenly | -0.092 |
opened | -0.086 |
pretended | -0.084 |
two | -0.082 |
Suddenly | -0.078 |
Token | be |
Token ID | 307 |
Feature activation | +5.09e+00 |
Pos logit contributions | |
beh | +0.016 |
ash | +0.016 |
unatt | +0.016 |
rav | +0.015 |
forward | +0.015 |
Neg logit contributions | |
suddenly | -0.011 |
opened | -0.011 |
two | -0.010 |
pretended | -0.010 |
a | -0.010 |
Token | safe |
Token ID | 3338 |
Feature activation | +6.41e+00 |
Pos logit contributions | |
beh | +0.154 |
rav | +0.149 |
ash | +0.149 |
unatt | +0.145 |
aring | +0.144 |
Neg logit contributions | |
suddenly | -0.166 |
opened | -0.157 |
two | -0.157 |
pretended | -0.157 |
a | -0.156 |
Token | . |
Token ID | 13 |
Feature activation | +7.48e-01 |
Pos logit contributions | |
beh | +0.279 |
ash | +0.268 |
rav | +0.264 |
opl | +0.260 |
ocol | +0.258 |
Neg logit contributions | |
suddenly | -0.139 |
a | -0.124 |
two | -0.122 |
opened | -0.116 |
pretended | -0.114 |
Token | The |
Token ID | 383 |
Feature activation | -3.01e+00 |
Pos logit contributions | |
beh | +0.030 |
unatt | +0.029 |
ash | +0.028 |
forward | +0.028 |
rav | +0.028 |
Neg logit contributions | |
suddenly | -0.016 |
opened | -0.015 |
pretended | -0.015 |
two | -0.014 |
Suddenly | -0.013 |
Token | fire |
Token ID | 2046 |
Feature activation | +8.63e-02 |
Pos logit contributions | |
suddenly | +0.082 |
opened | +0.078 |
pretended | +0.077 |
two | +0.074 |
Suddenly | +0.073 |
Neg logit contributions | |
beh | -0.102 |
unatt | -0.099 |
forward | -0.097 |
ash | -0.097 |
rav | -0.095 |
Token | fly |
Token ID | 12254 |
Feature activation | -1.31e+01 |
Pos logit contributions | |
beh | +0.004 |
unatt | +0.004 |
forward | +0.004 |
ash | +0.004 |
rav | +0.004 |
Neg logit contributions | |
suddenly | -0.001 |
opened | -0.001 |
pretended | -0.001 |
two | -0.001 |
Suddenly | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
suddenly | +0.192 |
opened | +0.176 |
pretended | +0.168 |
two | +0.159 |
Suddenly | +0.150 |
Neg logit contributions | |
beh | -0.616 |
unatt | -0.602 |
ash | -0.591 |
forward | -0.590 |
rav | -0.581 |
Token | happy |
Token ID | 3772 |
Feature activation | +3.91e+00 |
Pos logit contributions | |
suddenly | +0.058 |
opened | +0.056 |
pretended | +0.053 |
two | +0.053 |
a | +0.052 |
Neg logit contributions | |
beh | -0.069 |
unatt | -0.066 |
rav | -0.066 |
ash | -0.066 |
forward | -0.063 |
Token | and |
Token ID | 290 |
Feature activation | +6.70e-01 |
Pos logit contributions | |
beh | +0.157 |
unatt | +0.151 |
ash | +0.150 |
forward | +0.147 |
rav | +0.146 |
Neg logit contributions | |
suddenly | -0.090 |
opened | -0.083 |
pretended | -0.081 |
two | -0.079 |
a | -0.077 |
Token | thanked |
Token ID | 26280 |
Feature activation | -4.60e+00 |
Pos logit contributions | |
beh | +0.026 |
ash | +0.025 |
unatt | +0.025 |
rav | +0.024 |
agging | +0.024 |
Neg logit contributions | |
suddenly | -0.016 |
opened | -0.016 |
pretended | -0.015 |
two | -0.015 |
a | -0.014 |
Token | Tim |
Token ID | 5045 |
Feature activation | +7.62e-01 |
Pos logit contributions | |
suddenly | +0.086 |
a | +0.080 |
two | +0.079 |
opened | +0.074 |
Suddenly | +0.068 |
Neg logit contributions | |
beh | -0.215 |
unatt | -0.206 |
aring | -0.203 |
ash | -0.202 |
rav | -0.196 |
Token | Mia |
Token ID | 32189 |
Feature activation | -7.39e+00 |
Pos logit contributions | |
suddenly | +0.017 |
Suddenly | +0.015 |
two | +0.015 |
pretended | +0.013 |
a | +0.013 |
Neg logit contributions | |
beh | -0.035 |
off | -0.033 |
ash | -0.032 |
forward | -0.032 |
unatt | -0.032 |
Token | was |
Token ID | 373 |
Feature activation | -4.94e+00 |
Pos logit contributions | |
suddenly | +0.144 |
opened | +0.134 |
pretended | +0.129 |
two | +0.126 |
Suddenly | +0.122 |
Neg logit contributions | |
beh | -0.304 |
unatt | -0.296 |
ash | -0.293 |
forward | -0.292 |
rav | -0.286 |
Token | so |
Token ID | 523 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
suddenly | +0.144 |
opened | +0.140 |
pretended | +0.139 |
Suddenly | +0.131 |
two | +0.129 |
Neg logit contributions | |
beh | -0.152 |
unatt | -0.150 |
forward | -0.149 |
ash | -0.142 |
losers | -0.141 |
Token | excited |
Token ID | 6568 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
suddenly | +0.070 |
opened | +0.067 |
two | +0.064 |
pretended | +0.064 |
Suddenly | +0.063 |
Neg logit contributions | |
beh | -0.073 |
unatt | -0.071 |
ash | -0.070 |
rav | -0.068 |
baby | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | -7.20e-02 |
Pos logit contributions | |
suddenly | +0.049 |
opened | +0.046 |
pretended | +0.044 |
two | +0.043 |
Suddenly | +0.042 |
Neg logit contributions | |
beh | -0.082 |
unatt | -0.080 |
ash | -0.078 |
forward | -0.078 |
rav | -0.077 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | -1.31e+01 |
Pos logit contributions | |
suddenly | +0.002 |
opened | +0.002 |
pretended | +0.002 |
two | +0.002 |
Suddenly | +0.002 |
Neg logit contributions | |
beh | -0.002 |
unatt | -0.002 |
forward | -0.002 |
ash | -0.002 |
rav | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
pretended | +0.271 |
opened | +0.252 |
suddenly | +0.248 |
Suddenly | +0.246 |
Under | +0.237 |
Neg logit contributions | |
forward | -0.475 |
overs | -0.444 |
beh | -0.440 |
unatt | -0.440 |
ash | -0.439 |
Token | she |
Token ID | 673 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
pretended | +0.065 |
Suddenly | +0.063 |
opened | +0.062 |
suddenly | +0.061 |
clicked | +0.058 |
Neg logit contributions | |
forward | -0.089 |
beh | -0.084 |
overs | -0.083 |
unatt | -0.083 |
ash | -0.080 |
Token | heard |
Token ID | 2982 |
Feature activation | -7.34e+00 |
Pos logit contributions | |
Suddenly | +0.024 |
Under | +0.023 |
pretended | +0.022 |
two | +0.022 |
suddenly | +0.021 |
Neg logit contributions | |
unatt | -0.062 |
forward | -0.061 |
beh | -0.059 |
ash | -0.059 |
baby | -0.059 |
Token | something |
Token ID | 1223 |
Feature activation | -4.47e+00 |
Pos logit contributions | |
pretended | +0.081 |
suddenly | +0.076 |
opened | +0.067 |
Suddenly | +0.059 |
clicked | +0.051 |
Neg logit contributions | |
beh | -0.343 |
forward | -0.340 |
unatt | -0.337 |
baby | -0.335 |
ash | -0.334 |
Token | that |
Token ID | 326 |
Feature activation | -6.49e+00 |
Pos logit contributions | |
suddenly | +0.117 |
opened | +0.111 |
pretended | +0.110 |
two | +0.106 |
Suddenly | +0.104 |
Neg logit contributions | |
beh | -0.156 |
unatt | -0.152 |
forward | -0.149 |
ash | -0.148 |
rav | -0.144 |
Token | in |
Token ID | 287 |
Feature activation | -3.01e+00 |
Pos logit contributions | |
suddenly | +0.081 |
opened | +0.076 |
pretended | +0.074 |
two | +0.071 |
Suddenly | +0.069 |
Neg logit contributions | |
beh | -0.165 |
unatt | -0.161 |
forward | -0.158 |
ash | -0.157 |
rav | -0.154 |
Token | awe |
Token ID | 25030 |
Feature activation | -6.67e+00 |
Pos logit contributions | |
suddenly | +0.069 |
opened | +0.065 |
pretended | +0.063 |
two | +0.062 |
Suddenly | +0.059 |
Neg logit contributions | |
beh | -0.119 |
unatt | -0.114 |
ash | -0.113 |
forward | -0.111 |
rav | -0.111 |
Token | as |
Token ID | 355 |
Feature activation | -6.39e+00 |
Pos logit contributions | |
suddenly | +0.149 |
opened | +0.141 |
two | +0.136 |
a | +0.136 |
pretended | +0.131 |
Neg logit contributions | |
beh | -0.275 |
unatt | -0.257 |
rav | -0.256 |
ash | -0.253 |
baby | -0.247 |
Token | the |
Token ID | 262 |
Feature activation | -4.53e-01 |
Pos logit contributions | |
suddenly | +0.124 |
pretended | +0.123 |
opened | +0.120 |
Suddenly | +0.112 |
two | +0.105 |
Neg logit contributions | |
forward | -0.252 |
beh | -0.246 |
unatt | -0.243 |
losers | -0.235 |
overs | -0.234 |
Token | dragon |
Token ID | 10441 |
Feature activation | -3.63e+00 |
Pos logit contributions | |
suddenly | +0.014 |
pretended | +0.013 |
opened | +0.013 |
Suddenly | +0.013 |
clicked | +0.012 |
Neg logit contributions | |
forward | -0.013 |
beh | -0.013 |
unatt | -0.013 |
losers | -0.013 |
ash | -0.012 |
Token | fly |
Token ID | 12254 |
Feature activation | -1.31e+01 |
Pos logit contributions | |
suddenly | +0.089 |
opened | +0.084 |
Suddenly | +0.084 |
two | +0.083 |
pretended | +0.082 |
Neg logit contributions | |
beh | -0.126 |
forward | -0.120 |
unatt | -0.120 |
ash | -0.119 |
aring | -0.118 |
Token | zoom |
Token ID | 19792 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
suddenly | +0.264 |
opened | +0.248 |
pretended | +0.240 |
two | +0.231 |
Suddenly | +0.221 |
Neg logit contributions | |
beh | -0.543 |
unatt | -0.530 |
ash | -0.519 |
forward | -0.516 |
rav | -0.509 |
Token | ed |
Token ID | 276 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
suddenly | +0.037 |
opened | +0.035 |
pretended | +0.034 |
two | +0.033 |
Suddenly | +0.033 |
Neg logit contributions | |
beh | -0.051 |
forward | -0.049 |
unatt | -0.049 |
ash | -0.049 |
rav | -0.048 |
Token | around |
Token ID | 1088 |
Feature activation | -6.84e-01 |
Pos logit contributions | |
suddenly | +0.066 |
opened | +0.063 |
pretended | +0.061 |
two | +0.060 |
a | +0.059 |
Neg logit contributions | |
beh | -0.099 |
unatt | -0.095 |
ash | -0.092 |
rav | -0.091 |
aring | -0.090 |
Token | the |
Token ID | 262 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
suddenly | +0.013 |
opened | +0.011 |
a | +0.011 |
pretended | +0.011 |
two | +0.011 |
Neg logit contributions | |
beh | -0.032 |
unatt | -0.030 |
ash | -0.030 |
rav | -0.030 |
aring | -0.030 |
Token | room |
Token ID | 2119 |
Feature activation | -3.79e+00 |
Pos logit contributions | |
suddenly | +0.001 |
opened | +0.001 |
pretended | +0.001 |
two | +0.001 |
Suddenly | +0.001 |
Neg logit contributions | |
beh | -0.001 |
unatt | -0.001 |
forward | -0.001 |
ash | -0.001 |
rav | -0.001 |
Token | a |
Token ID | 257 |
Feature activation | -1.17e+01 |
Pos logit contributions | |
suddenly | +0.144 |
opened | +0.134 |
two | +0.130 |
a | +0.126 |
pretended | +0.125 |
Neg logit contributions | |
beh | -0.339 |
unatt | -0.329 |
ash | -0.327 |
aring | -0.323 |
rav | -0.323 |
Token | blue |
Token ID | 4171 |
Feature activation | -4.45e+00 |
Pos logit contributions | |
suddenly | +0.318 |
opened | +0.304 |
pretended | +0.297 |
two | +0.289 |
Suddenly | +0.282 |
Neg logit contributions | |
beh | -0.398 |
unatt | -0.386 |
forward | -0.376 |
ash | -0.376 |
rav | -0.367 |
Token | car |
Token ID | 1097 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
suddenly | +0.116 |
opened | +0.112 |
pretended | +0.109 |
two | +0.105 |
Suddenly | +0.104 |
Neg logit contributions | |
beh | -0.152 |
unatt | -0.150 |
forward | -0.147 |
ash | -0.147 |
rav | -0.143 |
Token | . |
Token ID | 13 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
suddenly | +0.046 |
opened | +0.044 |
pretended | +0.043 |
two | +0.042 |
Suddenly | +0.041 |
Neg logit contributions | |
beh | -0.068 |
unatt | -0.067 |
forward | -0.065 |
ash | -0.065 |
aring | -0.063 |
Token | They |
Token ID | 1119 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
suddenly | +0.004 |
opened | +0.004 |
pretended | +0.004 |
two | +0.004 |
Suddenly | +0.003 |
Neg logit contributions | |
beh | -0.006 |
unatt | -0.006 |
forward | -0.006 |
ash | -0.005 |
rav | -0.005 |
Token | pretended |
Token ID | 36766 |
Feature activation | -1.31e+01 |
Pos logit contributions | |
suddenly | +0.050 |
Suddenly | +0.047 |
two | +0.044 |
opened | +0.041 |
Suddenly | +0.040 |
Neg logit contributions | |
forward | -0.097 |
beh | -0.094 |
losers | -0.094 |
unatt | -0.092 |
ash | -0.090 |
Token | they |
Token ID | 484 |
Feature activation | +5.01e-01 |
Pos logit contributions | |
suddenly | +0.385 |
opened | +0.369 |
pretended | +0.360 |
two | +0.352 |
Suddenly | +0.342 |
Neg logit contributions | |
beh | -0.426 |
unatt | -0.411 |
ash | -0.402 |
forward | -0.399 |
rav | -0.389 |
Token | were |
Token ID | 547 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
beh | +0.018 |
unatt | +0.018 |
forward | +0.018 |
rav | +0.017 |
baby | +0.017 |
Neg logit contributions | |
suddenly | -0.011 |
Suddenly | -0.011 |
two | -0.010 |
opened | -0.009 |
Under | -0.009 |
Token | racing |
Token ID | 11717 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
suddenly | +0.043 |
opened | +0.041 |
pretended | +0.040 |
two | +0.039 |
Suddenly | +0.039 |
Neg logit contributions | |
beh | -0.049 |
unatt | -0.048 |
forward | -0.047 |
ash | -0.046 |
rav | -0.045 |
Token | each |
Token ID | 1123 |
Feature activation | +6.53e+00 |
Pos logit contributions | |
beh | +0.048 |
unatt | +0.046 |
ash | +0.046 |
forward | +0.045 |
rav | +0.044 |
Neg logit contributions | |
suddenly | -0.027 |
opened | -0.025 |
pretended | -0.024 |
two | -0.024 |
a | -0.023 |
Token | other |
Token ID | 584 |
Feature activation | +3.76e+00 |
Pos logit contributions | |
beh | +0.224 |
forward | +0.223 |
unatt | +0.220 |
ash | +0.218 |
losers | +0.211 |
Neg logit contributions | |
suddenly | -0.167 |
opened | -0.162 |
pretended | -0.156 |
Suddenly | -0.150 |
two | -0.148 |
Token | Mom |
Token ID | 11254 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
suddenly | +0.171 |
opened | +0.163 |
pretended | +0.160 |
two | +0.152 |
Suddenly | +0.150 |
Neg logit contributions | |
beh | -0.310 |
unatt | -0.303 |
forward | -0.297 |
ash | -0.294 |
rav | -0.289 |
Token | my |
Token ID | 1820 |
Feature activation | -5.13e+00 |
Pos logit contributions | |
beh | +0.016 |
unatt | +0.016 |
mir | +0.015 |
rav | +0.015 |
losers | +0.015 |
Neg logit contributions | |
suddenly | -0.008 |
opened | -0.007 |
pretended | -0.006 |
two | -0.006 |
was | -0.006 |
Token | was |
Token ID | 373 |
Feature activation | -3.63e+00 |
Pos logit contributions | |
suddenly | +0.076 |
opened | +0.069 |
pretended | +0.062 |
two | +0.056 |
was | +0.056 |
Neg logit contributions | |
beh | -0.251 |
unatt | -0.250 |
rav | -0.242 |
ash | -0.242 |
forward | -0.240 |
Token | worried |
Token ID | 7960 |
Feature activation | -8.95e-01 |
Pos logit contributions | |
suddenly | +0.102 |
opened | +0.098 |
two | +0.097 |
a | +0.095 |
pretended | +0.091 |
Neg logit contributions | |
beh | -0.129 |
rav | -0.126 |
ash | -0.124 |
unatt | -0.121 |
aring | -0.119 |
Token | . |
Token ID | 13 |
Feature activation | +3.57e-01 |
Pos logit contributions | |
suddenly | +0.016 |
a | +0.016 |
two | +0.014 |
was | +0.014 |
opened | +0.013 |
Neg logit contributions | |
beh | -0.043 |
rav | -0.041 |
unatt | -0.041 |
mir | -0.040 |
losers | -0.040 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | -1.31e+01 |
Pos logit contributions | |
beh | +0.015 |
unatt | +0.014 |
ash | +0.014 |
forward | +0.014 |
rav | +0.013 |
Neg logit contributions | |
suddenly | -0.008 |
opened | -0.007 |
Suddenly | -0.007 |
two | -0.007 |
pretended | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
suddenly | +0.228 |
pretended | +0.222 |
opened | +0.219 |
Suddenly | +0.203 |
two | +0.196 |
Neg logit contributions | |
beh | -0.528 |
forward | -0.528 |
unatt | -0.526 |
rav | -0.511 |
baby | -0.510 |
Token | Baby |
Token ID | 14801 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
pretended | +0.052 |
opened | +0.051 |
Suddenly | +0.050 |
suddenly | +0.050 |
clicked | +0.045 |
Neg logit contributions | |
forward | -0.090 |
unatt | -0.089 |
beh | -0.089 |
baby | -0.085 |
losers | -0.085 |
Token | said |
Token ID | 531 |
Feature activation | -5.52e+00 |
Pos logit contributions | |
beh | +0.043 |
unatt | +0.042 |
baby | +0.042 |
forward | +0.042 |
ash | +0.042 |
Neg logit contributions | |
suddenly | -0.018 |
opened | -0.017 |
pretended | -0.017 |
Suddenly | -0.016 |
two | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | -2.36e+00 |
Pos logit contributions | |
suddenly | +0.105 |
a | +0.096 |
opened | +0.092 |
two | +0.084 |
Suddenly | +0.081 |
Neg logit contributions | |
beh | -0.271 |
ash | -0.267 |
opl | -0.265 |
unatt | -0.254 |
off | -0.253 |
Token | " |
Token ID | 366 |
Feature activation | +7.72e+00 |
Pos logit contributions | |
suddenly | +0.053 |
a | +0.049 |
opened | +0.049 |
Suddenly | +0.049 |
two | +0.047 |
Neg logit contributions | |
beh | -0.110 |
unatt | -0.102 |
ash | -0.102 |
opl | -0.101 |
off | -0.098 |
Token | toy |
Token ID | 13373 |
Feature activation | -4.57e+00 |
Pos logit contributions | |
suddenly | +0.019 |
opened | +0.018 |
pretended | +0.018 |
two | +0.017 |
Suddenly | +0.017 |
Neg logit contributions | |
beh | -0.024 |
unatt | -0.023 |
forward | -0.023 |
ash | -0.022 |
rav | -0.022 |
Token | medical |
Token ID | 3315 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
suddenly | +0.114 |
opened | +0.106 |
pretended | +0.104 |
two | +0.103 |
a | +0.097 |
Neg logit contributions | |
beh | -0.173 |
unatt | -0.164 |
ash | -0.162 |
forward | -0.161 |
rav | -0.159 |
Token | kit |
Token ID | 6220 |
Feature activation | -6.33e+00 |
Pos logit contributions | |
beh | +0.073 |
unatt | +0.069 |
rav | +0.068 |
aring | +0.067 |
ash | +0.067 |
Neg logit contributions | |
suddenly | -0.054 |
opened | -0.051 |
two | -0.051 |
pretended | -0.050 |
a | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
suddenly | +0.143 |
opened | +0.132 |
two | +0.128 |
pretended | +0.127 |
a | +0.125 |
Neg logit contributions | |
beh | -0.257 |
unatt | -0.244 |
rav | -0.241 |
ash | -0.240 |
mir | -0.238 |
Token | She |
Token ID | 1375 |
Feature activation | -2.91e+00 |
Pos logit contributions | |
beh | +0.009 |
unatt | +0.009 |
forward | +0.009 |
ash | +0.009 |
rav | +0.008 |
Neg logit contributions | |
suddenly | -0.006 |
opened | -0.005 |
pretended | -0.005 |
two | -0.005 |
Suddenly | -0.005 |
Token | pretended |
Token ID | 36766 |
Feature activation | -1.31e+01 |
Pos logit contributions | |
suddenly | +0.048 |
Suddenly | +0.042 |
two | +0.042 |
pretended | +0.040 |
opened | +0.039 |
Neg logit contributions | |
beh | -0.117 |
losers | -0.114 |
forward | -0.114 |
ash | -0.111 |
unatt | -0.111 |
Token | to |
Token ID | 284 |
Feature activation | +9.43e-01 |
Pos logit contributions | |
suddenly | +0.302 |
opened | +0.284 |
pretended | +0.269 |
a | +0.265 |
two | +0.263 |
Neg logit contributions | |
beh | -0.530 |
ash | -0.506 |
unatt | -0.504 |
forward | -0.494 |
rav | -0.490 |
Token | help |
Token ID | 1037 |
Feature activation | -3.67e-01 |
Pos logit contributions | |
beh | +0.036 |
unatt | +0.036 |
forward | +0.035 |
ash | +0.035 |
rav | +0.034 |
Neg logit contributions | |
suddenly | -0.022 |
opened | -0.021 |
pretended | -0.020 |
two | -0.020 |
a | -0.019 |
Token | her |
Token ID | 607 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
suddenly | +0.008 |
a | +0.008 |
opened | +0.008 |
two | +0.007 |
pretended | +0.007 |
Neg logit contributions | |
beh | -0.015 |
unatt | -0.015 |
rav | -0.015 |
agging | -0.014 |
ash | -0.014 |
Token | stuffed |
Token ID | 22259 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
beh | +0.084 |
unatt | +0.082 |
forward | +0.080 |
ash | +0.079 |
baby | +0.078 |
Neg logit contributions | |
suddenly | -0.061 |
opened | -0.058 |
pretended | -0.056 |
Suddenly | -0.055 |
two | -0.054 |
Token | animals |
Token ID | 4695 |
Feature activation | -9.44e-01 |
Pos logit contributions | |
beh | +0.046 |
unatt | +0.044 |
ash | +0.043 |
forward | +0.043 |
baby | +0.042 |
Neg logit contributions | |
suddenly | -0.037 |
opened | -0.035 |
pretended | -0.034 |
Suddenly | -0.033 |
two | -0.032 |
Token | walking |
Token ID | 6155 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
suddenly | +0.103 |
two | +0.097 |
opened | +0.097 |
a | +0.093 |
pretended | +0.093 |
Neg logit contributions | |
beh | -0.137 |
unatt | -0.131 |
ash | -0.131 |
rav | -0.131 |
aring | -0.128 |
Token | in |
Token ID | 287 |
Feature activation | -4.67e+00 |
Pos logit contributions | |
suddenly | +0.032 |
opened | +0.031 |
pretended | +0.030 |
two | +0.029 |
Suddenly | +0.028 |
Neg logit contributions | |
beh | -0.047 |
forward | -0.046 |
unatt | -0.046 |
ash | -0.045 |
rav | -0.044 |
Token | the |
Token ID | 262 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
pretended | +0.101 |
suddenly | +0.093 |
opened | +0.091 |
Suddenly | +0.086 |
clicked | +0.085 |
Neg logit contributions | |
forward | -0.176 |
unatt | -0.164 |
future | -0.161 |
overs | -0.161 |
ash | -0.158 |
Token | park |
Token ID | 3952 |
Feature activation | -3.15e+00 |
Pos logit contributions | |
pretended | +0.050 |
suddenly | +0.049 |
opened | +0.048 |
clicked | +0.044 |
Suddenly | +0.043 |
Neg logit contributions | |
unatt | -0.055 |
forward | -0.055 |
beh | -0.054 |
future | -0.052 |
ash | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
suddenly | +0.066 |
opened | +0.062 |
pretended | +0.060 |
two | +0.058 |
Suddenly | +0.056 |
Neg logit contributions | |
beh | -0.127 |
unatt | -0.124 |
forward | -0.122 |
ash | -0.121 |
rav | -0.119 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | -1.31e+01 |
Pos logit contributions | |
forward | +0.006 |
beh | +0.006 |
ash | +0.006 |
unatt | +0.006 |
overs | +0.006 |
Neg logit contributions | |
suddenly | -0.003 |
pretended | -0.003 |
opened | -0.003 |
two | -0.003 |
clicked | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
pretended | +0.281 |
opened | +0.256 |
pressed | +0.239 |
clicked | +0.239 |
suddenly | +0.236 |
Neg logit contributions | |
forward | -0.466 |
unatt | -0.422 |
rav | -0.419 |
overs | -0.419 |
beh | -0.418 |
Token | Kim |
Token ID | 6502 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
pretended | +0.060 |
opened | +0.059 |
clicked | +0.053 |
Suddenly | +0.053 |
inserted | +0.053 |
Neg logit contributions | |
forward | -0.097 |
unatt | -0.093 |
beh | -0.090 |
overs | -0.089 |
hurricane | -0.087 |
Token | saw |
Token ID | 2497 |
Feature activation | -7.11e+00 |
Pos logit contributions | |
two | +0.025 |
Suddenly | +0.025 |
pretended | +0.025 |
Under | +0.025 |
suddenly | +0.025 |
Neg logit contributions | |
unatt | -0.062 |
baby | -0.060 |
forward | -0.059 |
ash | -0.059 |
beh | -0.058 |
Token | something |
Token ID | 1223 |
Feature activation | -4.52e+00 |
Pos logit contributions | |
pretended | +0.117 |
opened | +0.104 |
Under | +0.093 |
suddenly | +0.091 |
Suddenly | +0.086 |
Neg logit contributions | |
unatt | -0.292 |
ambulance | -0.288 |
beh | -0.283 |
permission | -0.280 |
baby | -0.280 |
Token | that |
Token ID | 326 |
Feature activation | -6.49e+00 |
Pos logit contributions | |
suddenly | +0.111 |
pretended | +0.107 |
opened | +0.107 |
two | +0.101 |
Suddenly | +0.101 |
Neg logit contributions | |
beh | -0.161 |
unatt | -0.158 |
forward | -0.156 |
ash | -0.154 |
baby | -0.149 |
Token | playing |
Token ID | 2712 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
beh | +0.009 |
forward | +0.009 |
unatt | +0.009 |
ash | +0.009 |
baby | +0.009 |
Neg logit contributions | |
suddenly | -0.003 |
opened | -0.003 |
pretended | -0.003 |
two | -0.003 |
Suddenly | -0.003 |
Token | with |
Token ID | 351 |
Feature activation | -5.73e+00 |
Pos logit contributions | |
suddenly | +0.039 |
a | +0.037 |
pretended | +0.035 |
two | +0.034 |
opened | +0.033 |
Neg logit contributions | |
beh | -0.105 |
aring | -0.099 |
rav | -0.098 |
unatt | -0.098 |
mir | -0.096 |
Token | her |
Token ID | 607 |
Feature activation | +4.38e-01 |
Pos logit contributions | |
suddenly | +0.151 |
opened | +0.146 |
pretended | +0.141 |
Suddenly | +0.136 |
two | +0.135 |
Neg logit contributions | |
beh | -0.197 |
unatt | -0.193 |
ash | -0.188 |
forward | -0.187 |
losers | -0.182 |
Token | toys |
Token ID | 14958 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
beh | +0.016 |
unatt | +0.015 |
forward | +0.015 |
ash | +0.015 |
rav | +0.015 |
Neg logit contributions | |
suddenly | -0.011 |
opened | -0.011 |
pretended | -0.010 |
two | -0.010 |
Suddenly | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
suddenly | +0.036 |
opened | +0.033 |
pretended | +0.032 |
two | +0.032 |
a | +0.031 |
Neg logit contributions | |
beh | -0.077 |
rav | -0.073 |
ash | -0.071 |
forward | -0.070 |
aring | -0.070 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | -1.31e+01 |
Pos logit contributions | |
suddenly | +0.003 |
opened | +0.003 |
pretended | +0.003 |
two | +0.003 |
Suddenly | +0.003 |
Neg logit contributions | |
beh | -0.005 |
unatt | -0.005 |
forward | -0.005 |
ash | -0.005 |
rav | -0.005 |
Token | she |
Token ID | 673 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
pretended | +0.253 |
suddenly | +0.246 |
opened | +0.242 |
Suddenly | +0.226 |
clicked | +0.214 |
Neg logit contributions | |
forward | -0.494 |
beh | -0.484 |
unatt | -0.480 |
rav | -0.472 |
baby | -0.471 |
Token | saw |
Token ID | 2497 |
Feature activation | -7.22e+00 |
Pos logit contributions | |
Suddenly | +0.030 |
Under | +0.029 |
As | +0.028 |
two | +0.028 |
suddenly | +0.027 |
Neg logit contributions | |
unatt | -0.073 |
forward | -0.072 |
baby | -0.072 |
ash | -0.070 |
losers | -0.070 |
Token | something |
Token ID | 1223 |
Feature activation | -4.53e+00 |
Pos logit contributions | |
pretended | +0.129 |
opened | +0.120 |
blasted | +0.109 |
Under | +0.109 |
suddenly | +0.105 |
Neg logit contributions | |
ambulance | -0.274 |
unatt | -0.274 |
permission | -0.270 |
baby | -0.268 |
beh | -0.264 |
Token | brown |
Token ID | 7586 |
Feature activation | -3.48e-01 |
Pos logit contributions | |
suddenly | +0.109 |
opened | +0.104 |
pretended | +0.102 |
two | +0.099 |
Suddenly | +0.097 |
Neg logit contributions | |
beh | -0.167 |
unatt | -0.163 |
forward | -0.160 |
ash | -0.159 |
rav | -0.155 |
Token | pe |
Token ID | 613 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
suddenly | +0.008 |
pretended | +0.008 |
opened | +0.008 |
Suddenly | +0.007 |
two | +0.007 |
Neg logit contributions | |
unatt | -0.013 |
beh | -0.012 |
forward | -0.012 |
ash | -0.012 |
baby | -0.012 |
Token | home |
Token ID | 1363 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
suddenly | +0.174 |
opened | +0.168 |
pretended | +0.164 |
two | +0.163 |
Suddenly | +0.161 |
Neg logit contributions | |
beh | -0.090 |
unatt | -0.083 |
ash | -0.080 |
forward | -0.076 |
rav | -0.076 |
Token | Lucky |
Token ID | 27963 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
forward | +0.071 |
unatt | +0.070 |
aring | +0.068 |
beh | +0.068 |
overs | +0.067 |
Neg logit contributions | |
pretended | -0.042 |
suddenly | -0.039 |
two | -0.037 |
Suddenly | -0.036 |
Under | -0.036 |
Token | saw |
Token ID | 2497 |
Feature activation | -6.23e+00 |
Pos logit contributions | |
suddenly | +0.022 |
opened | +0.020 |
pretended | +0.020 |
two | +0.019 |
Suddenly | +0.019 |
Neg logit contributions | |
beh | -0.047 |
unatt | -0.047 |
forward | -0.045 |
ash | -0.045 |
rav | -0.044 |
Token | a |
Token ID | 257 |
Feature activation | -8.11e+00 |
Pos logit contributions | |
pretended | +0.165 |
opened | +0.160 |
suddenly | +0.154 |
clicked | +0.147 |
blasted | +0.144 |
Neg logit contributions | |
baby | -0.199 |
unatt | -0.195 |
ambulance | -0.193 |
beh | -0.190 |
forward | -0.189 |
Token | dragon |
Token ID | 10441 |
Feature activation | -3.60e+00 |
Pos logit contributions | |
suddenly | +0.171 |
opened | +0.160 |
two | +0.152 |
pretended | +0.148 |
a | +0.144 |
Neg logit contributions | |
beh | -0.356 |
ash | -0.335 |
unatt | -0.326 |
forward | -0.323 |
off | -0.317 |
Token | fly |
Token ID | 12254 |
Feature activation | -1.31e+01 |
Pos logit contributions | |
suddenly | +0.093 |
pretended | +0.091 |
Suddenly | +0.090 |
opened | +0.089 |
two | +0.088 |
Neg logit contributions | |
unatt | -0.118 |
beh | -0.117 |
forward | -0.113 |
aring | -0.111 |
baby | -0.110 |
Token | . |
Token ID | 13 |
Feature activation | +4.85e-01 |
Pos logit contributions | |
suddenly | +0.320 |
opened | +0.304 |
pretended | +0.294 |
two | +0.287 |
Suddenly | +0.278 |
Neg logit contributions | |
beh | -0.487 |
unatt | -0.471 |
ash | -0.463 |
forward | -0.460 |
rav | -0.454 |
Token | The |
Token ID | 383 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
beh | +0.019 |
unatt | +0.018 |
ash | +0.018 |
forward | +0.017 |
rav | +0.017 |
Neg logit contributions | |
suddenly | -0.012 |
opened | -0.011 |
Suddenly | -0.011 |
pretended | -0.011 |
two | -0.011 |
Token | dragon |
Token ID | 10441 |
Feature activation | -3.62e+00 |
Pos logit contributions | |
suddenly | +0.125 |
Suddenly | +0.122 |
pretended | +0.122 |
opened | +0.118 |
clicked | +0.118 |
Neg logit contributions | |
losers | -0.071 |
unatt | -0.070 |
future | -0.068 |
forward | -0.065 |
ambulance | -0.063 |
Token | fly |
Token ID | 12254 |
Feature activation | -1.31e+01 |
Pos logit contributions | |
Suddenly | +0.102 |
two | +0.099 |
suddenly | +0.097 |
Under | +0.096 |
pretended | +0.092 |
Neg logit contributions | |
beh | -0.102 |
aring | -0.101 |
baby | -0.098 |
unatt | -0.093 |
ash | -0.090 |
Token | told |
Token ID | 1297 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
suddenly | +0.179 |
Suddenly | +0.172 |
pretended | +0.164 |
two | +0.160 |
opened | +0.150 |
Neg logit contributions | |
beh | -0.556 |
unatt | -0.535 |
losers | -0.535 |
aring | -0.532 |
mir | -0.527 |
Token | a |
Token ID | 257 |
Feature activation | -1.18e+01 |
Pos logit contributions | |
opened | +0.072 |
suddenly | +0.071 |
two | +0.067 |
pretended | +0.066 |
a | +0.065 |
Neg logit contributions | |
beh | -0.128 |
ash | -0.127 |
unatt | -0.127 |
rav | -0.124 |
forward | -0.123 |
Token | big |
Token ID | 1263 |
Feature activation | -5.37e+00 |
Pos logit contributions | |
suddenly | +0.299 |
opened | +0.284 |
pretended | +0.277 |
two | +0.271 |
Suddenly | +0.262 |
Neg logit contributions | |
beh | -0.432 |
unatt | -0.416 |
ash | -0.408 |
forward | -0.405 |
rav | -0.398 |
Token | rocket |
Token ID | 10701 |
Feature activation | -6.99e+00 |
Pos logit contributions | |
suddenly | +0.130 |
opened | +0.127 |
two | +0.123 |
a | +0.120 |
pretended | +0.119 |
Neg logit contributions | |
beh | -0.222 |
ash | -0.200 |
rav | -0.199 |
unatt | -0.196 |
aring | -0.194 |
Token | that |
Token ID | 326 |
Feature activation | -7.77e+00 |
Pos logit contributions | |
suddenly | +0.170 |
opened | +0.161 |
pretended | +0.157 |
two | +0.151 |
Suddenly | +0.145 |
Neg logit contributions | |
beh | -0.266 |
unatt | -0.253 |
ash | -0.251 |
forward | -0.249 |
rav | -0.247 |
Token | they |
Token ID | 484 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
suddenly | +0.125 |
opened | +0.115 |
pretended | +0.111 |
two | +0.105 |
Suddenly | +0.101 |
Neg logit contributions | |
beh | -0.352 |
unatt | -0.344 |
forward | -0.338 |
ash | -0.337 |
rav | -0.332 |
Token | pretended |
Token ID | 36766 |
Feature activation | -1.30e+01 |
Pos logit contributions | |
beh | +0.005 |
unatt | +0.005 |
forward | +0.005 |
ash | +0.005 |
rav | +0.005 |
Neg logit contributions | |
suddenly | -0.002 |
opened | -0.002 |
pretended | -0.002 |
two | -0.002 |
Suddenly | -0.002 |
Token | to |
Token ID | 284 |
Feature activation | +8.89e-01 |
Pos logit contributions | |
suddenly | +0.370 |
opened | +0.353 |
pretended | +0.339 |
a | +0.334 |
two | +0.331 |
Neg logit contributions | |
beh | -0.459 |
unatt | -0.437 |
ash | -0.432 |
forward | -0.424 |
rav | -0.414 |
Token | fly |
Token ID | 6129 |
Feature activation | -9.72e-01 |
Pos logit contributions | |
beh | +0.030 |
unatt | +0.030 |
forward | +0.029 |
losers | +0.029 |
ash | +0.028 |
Neg logit contributions | |
suddenly | -0.022 |
opened | -0.021 |
pretended | -0.021 |
Suddenly | -0.020 |
two | -0.020 |
Token | to |
Token ID | 284 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
suddenly | +0.023 |
opened | +0.021 |
pretended | +0.021 |
a | +0.021 |
two | +0.020 |
Neg logit contributions | |
beh | -0.039 |
unatt | -0.037 |
ash | -0.036 |
aring | -0.036 |
rav | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
beh | +0.059 |
unatt | +0.057 |
ash | +0.056 |
forward | +0.055 |
aring | +0.055 |
Neg logit contributions | |
suddenly | -0.020 |
opened | -0.019 |
pretended | -0.018 |
two | -0.018 |
a | -0.017 |
Token | moon |
Token ID | 8824 |
Feature activation | -7.11e+00 |
Pos logit contributions | |
suddenly | +0.005 |
opened | +0.005 |
pretended | +0.005 |
two | +0.005 |
Suddenly | +0.005 |
Neg logit contributions | |
beh | -0.009 |
unatt | -0.008 |
ash | -0.008 |
forward | -0.008 |
rav | -0.008 |
Token | fun |
Token ID | 1257 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
beh | +0.080 |
rav | +0.078 |
ash | +0.077 |
forward | +0.075 |
unatt | +0.075 |
Neg logit contributions | |
suddenly | -0.064 |
opened | -0.061 |
two | -0.059 |
a | -0.058 |
pretended | -0.057 |
Token | ! |
Token ID | 0 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
beh | +0.091 |
unatt | +0.085 |
ash | +0.084 |
forward | +0.084 |
aring | +0.082 |
Neg logit contributions | |
suddenly | -0.057 |
pretended | -0.051 |
two | -0.050 |
opened | -0.050 |
a | -0.050 |
Token | |
Token ID | 220 |
Feature activation | -9.89e-02 |
Pos logit contributions | |
suddenly | +0.030 |
opened | +0.028 |
pretended | +0.028 |
two | +0.027 |
Suddenly | +0.027 |
Neg logit contributions | |
beh | -0.049 |
unatt | -0.048 |
ash | -0.047 |
forward | -0.047 |
rav | -0.046 |
Token |
|
Token ID | 198 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
suddenly | +0.001 |
opened | +0.001 |
pretended | +0.001 |
two | +0.001 |
Suddenly | +0.001 |
Neg logit contributions | |
beh | -0.005 |
unatt | -0.005 |
forward | -0.004 |
ash | -0.004 |
rav | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
suddenly | +0.029 |
pretended | +0.029 |
opened | +0.029 |
two | +0.025 |
clicked | +0.024 |
Neg logit contributions | |
beh | -0.053 |
ash | -0.052 |
forward | -0.051 |
unatt | -0.051 |
baby | -0.051 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | -1.30e+01 |
Pos logit contributions | |
suddenly | +0.026 |
opened | +0.025 |
pretended | +0.024 |
two | +0.023 |
Suddenly | +0.023 |
Neg logit contributions | |
beh | -0.047 |
unatt | -0.046 |
ash | -0.045 |
forward | -0.045 |
rav | -0.044 |
Token | , |
Token ID | 11 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
pretended | +0.240 |
suddenly | +0.235 |
opened | +0.229 |
Suddenly | +0.214 |
clicked | +0.200 |
Neg logit contributions | |
forward | -0.509 |
beh | -0.495 |
unatt | -0.491 |
rav | -0.485 |
baby | -0.479 |
Token | L |
Token ID | 406 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
pretended | +0.053 |
opened | +0.052 |
suddenly | +0.051 |
Suddenly | +0.051 |
clicked | +0.047 |
Neg logit contributions | |
forward | -0.098 |
beh | -0.095 |
unatt | -0.094 |
overs | -0.091 |
losers | -0.091 |
Token | ila |
Token ID | 10102 |
Feature activation | -3.09e+00 |
Pos logit contributions | |
pretended | +0.050 |
suddenly | +0.050 |
opened | +0.047 |
Suddenly | +0.047 |
two | +0.046 |
Neg logit contributions | |
ash | -0.039 |
beh | -0.037 |
rav | -0.036 |
baby | -0.036 |
aring | -0.036 |
Token | heard |
Token ID | 2982 |
Feature activation | -7.17e+00 |
Pos logit contributions | |
suddenly | +0.047 |
pretended | +0.044 |
opened | +0.043 |
Suddenly | +0.042 |
two | +0.042 |
Neg logit contributions | |
unatt | -0.132 |
beh | -0.132 |
forward | -0.128 |
ash | -0.127 |
aring | -0.126 |
Token | a |
Token ID | 257 |
Feature activation | -8.62e+00 |
Pos logit contributions | |
suddenly | +0.083 |
pretended | +0.083 |
opened | +0.076 |
Suddenly | +0.064 |
two | +0.058 |
Neg logit contributions | |
beh | -0.336 |
forward | -0.329 |
unatt | -0.329 |
baby | -0.326 |
ash | -0.325 |
Token | on |
Token ID | 319 |
Feature activation | -4.42e+00 |
Pos logit contributions | |
aring | +0.005 |
baby | +0.005 |
beh | +0.005 |
unatt | +0.005 |
losers | +0.005 |
Neg logit contributions | |
suddenly | -0.002 |
two | -0.002 |
pretended | -0.002 |
Suddenly | -0.002 |
opened | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | -1.12e+01 |
Pos logit contributions | |
suddenly | +0.087 |
opened | +0.081 |
two | +0.077 |
pretended | +0.077 |
a | +0.073 |
Neg logit contributions | |
beh | -0.190 |
unatt | -0.183 |
ash | -0.180 |
forward | -0.179 |
rav | -0.178 |
Token | blanket |
Token ID | 18447 |
Feature activation | -3.09e+00 |
Pos logit contributions | |
suddenly | +0.247 |
opened | +0.236 |
a | +0.221 |
pretended | +0.220 |
two | +0.218 |
Neg logit contributions | |
beh | -0.490 |
ash | -0.434 |
rav | -0.430 |
aring | -0.423 |
unatt | -0.421 |
Token | . |
Token ID | 13 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
suddenly | +0.070 |
opened | +0.066 |
pretended | +0.064 |
two | +0.062 |
Suddenly | +0.060 |
Neg logit contributions | |
beh | -0.120 |
unatt | -0.116 |
ash | -0.114 |
forward | -0.114 |
rav | -0.112 |
Token | She |
Token ID | 1375 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
suddenly | +0.003 |
opened | +0.003 |
pretended | +0.003 |
two | +0.003 |
Suddenly | +0.003 |
Neg logit contributions | |
beh | -0.005 |
unatt | -0.005 |
forward | -0.005 |
ash | -0.005 |
rav | -0.005 |
Token | pretended |
Token ID | 36766 |
Feature activation | -1.30e+01 |
Pos logit contributions | |
suddenly | +0.060 |
Suddenly | +0.054 |
two | +0.052 |
pretended | +0.049 |
opened | +0.048 |
Neg logit contributions | |
beh | -0.123 |
ash | -0.121 |
losers | -0.121 |
mir | -0.120 |
forward | -0.120 |
Token | they |
Token ID | 484 |
Feature activation | +6.61e-01 |
Pos logit contributions | |
suddenly | +0.355 |
opened | +0.340 |
pretended | +0.325 |
two | +0.321 |
a | +0.317 |
Neg logit contributions | |
beh | -0.463 |
unatt | -0.442 |
ash | -0.438 |
forward | -0.432 |
rav | -0.422 |
Token | were |
Token ID | 547 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
beh | +0.026 |
unatt | +0.025 |
forward | +0.025 |
ash | +0.024 |
rav | +0.024 |
Neg logit contributions | |
suddenly | -0.013 |
Suddenly | -0.013 |
opened | -0.012 |
two | -0.012 |
pretended | -0.012 |
Token | having |
Token ID | 1719 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
suddenly | +0.035 |
opened | +0.033 |
pretended | +0.033 |
two | +0.032 |
Suddenly | +0.031 |
Neg logit contributions | |
beh | -0.048 |
unatt | -0.046 |
ash | -0.045 |
forward | -0.045 |
rav | -0.045 |
Token | a |
Token ID | 257 |
Feature activation | -8.69e+00 |
Pos logit contributions | |
suddenly | +0.024 |
opened | +0.022 |
pretended | +0.022 |
two | +0.021 |
a | +0.020 |
Neg logit contributions | |
beh | -0.074 |
unatt | -0.072 |
ash | -0.071 |
forward | -0.070 |
aring | -0.070 |
Token | tea |
Token ID | 8887 |
Feature activation | -4.42e-01 |
Pos logit contributions | |
suddenly | +0.273 |
opened | +0.262 |
pretended | +0.256 |
two | +0.251 |
Suddenly | +0.247 |
Neg logit contributions | |
beh | -0.256 |
unatt | -0.250 |
forward | -0.245 |
ash | -0.242 |
rav | -0.237 |
Token | had |
Token ID | 550 |
Feature activation | -6.28e+00 |
Pos logit contributions | |
suddenly | +0.073 |
opened | +0.070 |
pretended | +0.068 |
two | +0.064 |
Suddenly | +0.061 |
Neg logit contributions | |
beh | -0.161 |
unatt | -0.156 |
ash | -0.154 |
forward | -0.153 |
rav | -0.151 |
Token | so |
Token ID | 523 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
suddenly | +0.097 |
two | +0.094 |
opened | +0.093 |
a | +0.093 |
pretended | +0.088 |
Neg logit contributions | |
beh | -0.301 |
unatt | -0.293 |
ash | -0.291 |
rav | -0.289 |
aring | -0.289 |
Token | much |
Token ID | 881 |
Feature activation | +4.96e+00 |
Pos logit contributions | |
suddenly | +0.052 |
opened | +0.051 |
pretended | +0.050 |
two | +0.049 |
Suddenly | +0.048 |
Neg logit contributions | |
beh | -0.042 |
unatt | -0.041 |
ash | -0.040 |
forward | -0.039 |
rav | -0.038 |
Token | fun |
Token ID | 1257 |
Feature activation | +5.07e+00 |
Pos logit contributions | |
beh | +0.142 |
unatt | +0.137 |
ash | +0.133 |
forward | +0.133 |
rav | +0.129 |
Neg logit contributions | |
suddenly | -0.165 |
opened | -0.158 |
pretended | -0.155 |
two | -0.152 |
Suddenly | -0.149 |
Token | ! |
Token ID | 0 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
beh | +0.202 |
unatt | +0.196 |
ash | +0.192 |
rav | +0.191 |
forward | +0.189 |
Neg logit contributions | |
suddenly | -0.119 |
pretended | -0.105 |
opened | -0.105 |
two | -0.104 |
a | -0.101 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | -1.30e+01 |
Pos logit contributions | |
suddenly | +0.010 |
opened | +0.010 |
pretended | +0.009 |
two | +0.009 |
Suddenly | +0.009 |
Neg logit contributions | |
beh | -0.021 |
unatt | -0.021 |
ash | -0.020 |
forward | -0.020 |
rav | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
suddenly | +0.233 |
opened | +0.227 |
pretended | +0.226 |
Suddenly | +0.208 |
two | +0.196 |
Neg logit contributions | |
beh | -0.523 |
forward | -0.523 |
unatt | -0.516 |
rav | -0.507 |
ash | -0.504 |
Token | they |
Token ID | 484 |
Feature activation | +5.03e-01 |
Pos logit contributions | |
opened | +0.052 |
suddenly | +0.051 |
pretended | +0.051 |
Suddenly | +0.049 |
clicked | +0.045 |
Neg logit contributions | |
forward | -0.085 |
beh | -0.085 |
unatt | -0.083 |
losers | -0.081 |
overs | -0.081 |
Token | heard |
Token ID | 2982 |
Feature activation | -7.28e+00 |
Pos logit contributions | |
unatt | +0.022 |
forward | +0.022 |
beh | +0.021 |
ash | +0.021 |
baby | +0.021 |
Neg logit contributions | |
Suddenly | -0.007 |
suddenly | -0.007 |
pretended | -0.006 |
opened | -0.006 |
two | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | -8.82e+00 |
Pos logit contributions | |
suddenly | +0.073 |
pretended | +0.066 |
opened | +0.065 |
Suddenly | +0.053 |
two | +0.050 |
Neg logit contributions | |
beh | -0.360 |
forward | -0.351 |
unatt | -0.351 |
ash | -0.349 |
baby | -0.347 |
Token | loud |
Token ID | 7812 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
suddenly | +0.196 |
opened | +0.190 |
two | +0.176 |
a | +0.174 |
was | +0.161 |
Neg logit contributions | |
beh | -0.381 |
ash | -0.368 |
unatt | -0.357 |
overs | -0.349 |
ags | -0.344 |
Token | outside |
Token ID | 2354 |
Feature activation | -4.45e+00 |
Pos logit contributions | |
suddenly | +0.014 |
pretended | +0.013 |
opened | +0.012 |
Suddenly | +0.011 |
clicked | +0.010 |
Neg logit contributions | |
forward | -0.043 |
beh | -0.043 |
unatt | -0.043 |
overs | -0.042 |
ash | -0.042 |
Token | with |
Token ID | 351 |
Feature activation | -5.38e+00 |
Pos logit contributions | |
suddenly | +0.096 |
opened | +0.090 |
pretended | +0.086 |
two | +0.085 |
Suddenly | +0.081 |
Neg logit contributions | |
beh | -0.183 |
unatt | -0.175 |
ash | -0.172 |
forward | -0.171 |
rav | -0.170 |
Token | his |
Token ID | 465 |
Feature activation | -8.16e-01 |
Pos logit contributions | |
suddenly | +0.121 |
opened | +0.114 |
pretended | +0.111 |
two | +0.109 |
Suddenly | +0.103 |
Neg logit contributions | |
beh | -0.211 |
unatt | -0.205 |
ash | -0.201 |
forward | -0.200 |
rav | -0.198 |
Token | friends |
Token ID | 2460 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
suddenly | +0.021 |
opened | +0.020 |
pretended | +0.020 |
two | +0.019 |
Suddenly | +0.019 |
Neg logit contributions | |
beh | -0.028 |
unatt | -0.028 |
forward | -0.027 |
ash | -0.027 |
rav | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
suddenly | +0.034 |
opened | +0.032 |
pretended | +0.031 |
two | +0.030 |
a | +0.029 |
Neg logit contributions | |
beh | -0.055 |
unatt | -0.052 |
ash | -0.051 |
rav | -0.051 |
forward | -0.051 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | -1.30e+01 |
Pos logit contributions | |
beh | +0.001 |
unatt | +0.001 |
forward | +0.001 |
ash | +0.001 |
rav | +0.001 |
Neg logit contributions | |
suddenly | -0.000 |
opened | -0.000 |
pretended | -0.000 |
two | -0.000 |
Suddenly | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
pretended | +0.247 |
opened | +0.244 |
suddenly | +0.241 |
Suddenly | +0.223 |
clicked | +0.217 |
Neg logit contributions | |
forward | -0.492 |
beh | -0.475 |
unatt | -0.471 |
rav | -0.463 |
overs | -0.458 |
Token | another |
Token ID | 1194 |
Feature activation | -2.71e+00 |
Pos logit contributions | |
opened | +0.060 |
pretended | +0.057 |
Suddenly | +0.055 |
clicked | +0.054 |
inserted | +0.054 |
Neg logit contributions | |
forward | -0.097 |
beh | -0.094 |
unatt | -0.094 |
hurricane | -0.091 |
overs | -0.090 |
Token | boy |
Token ID | 2933 |
Feature activation | -6.04e+00 |
Pos logit contributions | |
suddenly | +0.054 |
opened | +0.052 |
pretended | +0.051 |
Suddenly | +0.048 |
two | +0.046 |
Neg logit contributions | |
beh | -0.106 |
forward | -0.105 |
unatt | -0.103 |
ash | -0.103 |
baby | -0.102 |
Token | , |
Token ID | 11 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
two | +0.099 |
suddenly | +0.097 |
Suddenly | +0.092 |
pretended | +0.091 |
opened | +0.091 |
Neg logit contributions | |
beh | -0.242 |
unatt | -0.240 |
forward | -0.238 |
ash | -0.233 |
baby | -0.231 |
Token | who |
Token ID | 508 |
Feature activation | -5.17e+00 |
Pos logit contributions | |
suddenly | +0.045 |
pretended | +0.045 |
opened | +0.044 |
Suddenly | +0.042 |
two | +0.041 |
Neg logit contributions | |
forward | -0.086 |
beh | -0.085 |
unatt | -0.084 |
baby | -0.080 |
rav | -0.079 |
Token | on |
Token ID | 319 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
suddenly | +0.018 |
opened | +0.017 |
pretended | +0.017 |
two | +0.016 |
Suddenly | +0.016 |
Neg logit contributions | |
beh | -0.038 |
unatt | -0.037 |
ash | -0.036 |
forward | -0.036 |
rav | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
suddenly | +0.038 |
opened | +0.034 |
two | +0.033 |
a | +0.031 |
pretended | +0.031 |
Neg logit contributions | |
beh | -0.108 |
unatt | -0.103 |
ash | -0.102 |
rav | -0.102 |
aring | -0.101 |
Token | jungle |
Token ID | 20712 |
Feature activation | -5.29e+00 |
Pos logit contributions | |
suddenly | +0.041 |
two | +0.039 |
opened | +0.038 |
a | +0.037 |
pretended | +0.036 |
Neg logit contributions | |
beh | -0.077 |
rav | -0.071 |
aring | -0.071 |
ash | -0.071 |
unatt | -0.070 |
Token | gym |
Token ID | 11550 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
suddenly | +0.124 |
opened | +0.118 |
two | +0.112 |
pretended | +0.111 |
a | +0.109 |
Neg logit contributions | |
beh | -0.209 |
rav | -0.199 |
unatt | -0.197 |
ash | -0.196 |
forward | -0.194 |
Token | . |
Token ID | 13 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
suddenly | +0.047 |
opened | +0.044 |
pretended | +0.042 |
two | +0.041 |
Suddenly | +0.040 |
Neg logit contributions | |
beh | -0.080 |
unatt | -0.076 |
ash | -0.075 |
rav | -0.075 |
forward | -0.074 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | -1.29e+01 |
Pos logit contributions | |
beh | +0.004 |
unatt | +0.004 |
ash | +0.004 |
forward | +0.004 |
rav | +0.004 |
Neg logit contributions | |
suddenly | -0.002 |
opened | -0.002 |
pretended | -0.002 |
two | -0.002 |
Suddenly | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -2.63e+00 |
Pos logit contributions | |
suddenly | +0.230 |
pretended | +0.225 |
opened | +0.221 |
Suddenly | +0.202 |
two | +0.192 |
Neg logit contributions | |
beh | -0.517 |
forward | -0.516 |
unatt | -0.512 |
rav | -0.499 |
ash | -0.498 |
Token | he |
Token ID | 339 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
opened | +0.053 |
pretended | +0.052 |
suddenly | +0.050 |
Suddenly | +0.049 |
clicked | +0.047 |
Neg logit contributions | |
forward | -0.095 |
unatt | -0.095 |
beh | -0.094 |
overs | -0.091 |
losers | -0.090 |
Token | heard |
Token ID | 2982 |
Feature activation | -7.27e+00 |
Pos logit contributions | |
suddenly | +0.030 |
pretended | +0.029 |
Suddenly | +0.029 |
opened | +0.029 |
two | +0.028 |
Neg logit contributions | |
unatt | -0.087 |
forward | -0.086 |
beh | -0.084 |
ash | -0.084 |
mir | -0.083 |
Token | a |
Token ID | 257 |
Feature activation | -8.89e+00 |
Pos logit contributions | |
suddenly | +0.073 |
pretended | +0.073 |
opened | +0.065 |
Suddenly | +0.052 |
clicked | +0.048 |
Neg logit contributions | |
beh | -0.346 |
unatt | -0.343 |
baby | -0.340 |
forward | -0.339 |
ash | -0.334 |
Token | very |
Token ID | 845 |
Feature activation | +6.46e+00 |
Pos logit contributions | |
suddenly | +0.200 |
opened | +0.196 |
two | +0.180 |
a | +0.177 |
Suddenly | +0.168 |
Neg logit contributions | |
beh | -0.373 |
ash | -0.364 |
unatt | -0.352 |
overs | -0.349 |
ags | -0.341 |
Token | You |
Token ID | 921 |
Feature activation | +3.41e+00 |
Pos logit contributions | |
beh | +0.027 |
unatt | +0.025 |
forward | +0.025 |
ash | +0.024 |
rav | +0.024 |
Neg logit contributions | |
suddenly | -0.056 |
opened | -0.055 |
pretended | -0.054 |
two | -0.052 |
Suddenly | -0.051 |
Token | are |
Token ID | 389 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
losers | +0.076 |
beh | +0.073 |
unatt | +0.071 |
rav | +0.071 |
forward | +0.070 |
Neg logit contributions | |
suddenly | -0.129 |
Suddenly | -0.124 |
Suddenly | -0.112 |
pretended | -0.111 |
opened | -0.110 |
Token | the |
Token ID | 262 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
beh | +0.105 |
unatt | +0.103 |
forward | +0.102 |
losers | +0.099 |
ash | +0.098 |
Neg logit contributions | |
suddenly | -0.058 |
opened | -0.054 |
pretended | -0.053 |
Suddenly | -0.051 |
two | -0.049 |
Token | best |
Token ID | 1266 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
beh | +0.020 |
ash | +0.019 |
unatt | +0.019 |
aring | +0.019 |
rav | +0.018 |
Neg logit contributions | |
two | -0.016 |
suddenly | -0.016 |
a | -0.014 |
opened | -0.014 |
pretended | -0.014 |
Token | co |
Token ID | 763 |
Feature activation | +6.53e+00 |
Pos logit contributions | |
beh | +0.090 |
unatt | +0.086 |
ash | +0.084 |
forward | +0.081 |
rav | +0.080 |
Neg logit contributions | |
suddenly | -0.073 |
two | -0.069 |
opened | -0.069 |
pretended | -0.067 |
a | -0.065 |
Token | - |
Token ID | 12 |
Feature activation | +1.38e+01 |
Pos logit contributions | |
beh | +0.194 |
unatt | +0.192 |
mir | +0.189 |
future | +0.189 |
ash | +0.187 |
Neg logit contributions | |
suddenly | -0.200 |
two | -0.185 |
opened | -0.185 |
a | -0.181 |
pretended | -0.181 |
Token | p |
Token ID | 79 |
Feature activation | -3.80e-01 |
Pos logit contributions | |
unatt | +0.309 |
beh | +0.299 |
ash | +0.294 |
mir | +0.283 |
agging | +0.282 |
Neg logit contributions | |
suddenly | -0.526 |
opened | -0.513 |
two | -0.504 |
a | -0.501 |
pretended | -0.494 |
Token | ilot |
Token ID | 23439 |
Feature activation | -7.86e+00 |
Pos logit contributions | |
suddenly | +0.009 |
two | +0.009 |
opened | +0.009 |
a | +0.009 |
pretended | +0.008 |
Neg logit contributions | |
beh | -0.015 |
unatt | -0.014 |
ash | -0.013 |
forward | -0.013 |
future | -0.013 |
Token | ever |
Token ID | 1683 |
Feature activation | +5.30e+00 |
Pos logit contributions | |
suddenly | +0.194 |
a | +0.194 |
two | +0.193 |
was | +0.176 |
pretended | +0.176 |
Neg logit contributions | |
beh | -0.322 |
ocol | -0.304 |
ash | -0.301 |
unatt | -0.285 |
mir | -0.284 |
Token | ." |
Token ID | 526 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
beh | +0.175 |
ash | +0.164 |
unatt | +0.164 |
forward | +0.160 |
rav | +0.159 |
Neg logit contributions | |
suddenly | -0.156 |
opened | -0.151 |
pretended | -0.148 |
two | -0.145 |
a | -0.141 |
Token | They |
Token ID | 1119 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
forward | +0.058 |
ash | +0.056 |
unatt | +0.056 |
beh | +0.055 |
overs | +0.055 |
Neg logit contributions | |
suddenly | -0.038 |
opened | -0.035 |
pretended | -0.035 |
clicked | -0.030 |
two | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | +4.92e-01 |
Pos logit contributions | |
suddenly | +0.011 |
opened | +0.010 |
two | +0.009 |
pretended | +0.009 |
a | +0.009 |
Neg logit contributions | |
beh | -0.026 |
ash | -0.025 |
unatt | -0.025 |
aring | -0.024 |
agging | -0.024 |
Token | butterfly |
Token ID | 35113 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
beh | +0.017 |
unatt | +0.017 |
ash | +0.016 |
forward | +0.016 |
rav | +0.016 |
Neg logit contributions | |
suddenly | -0.013 |
opened | -0.013 |
two | -0.012 |
pretended | -0.012 |
a | -0.012 |
Token | could |
Token ID | 714 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
suddenly | +0.023 |
opened | +0.021 |
pretended | +0.018 |
was | +0.015 |
two | +0.014 |
Neg logit contributions | |
beh | -0.117 |
ash | -0.114 |
unatt | -0.113 |
rav | -0.113 |
forward | -0.112 |
Token | enjoy |
Token ID | 2883 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
suddenly | +0.026 |
opened | +0.025 |
pretended | +0.023 |
two | +0.023 |
a | +0.022 |
Neg logit contributions | |
beh | -0.044 |
ash | -0.043 |
unatt | -0.042 |
forward | -0.041 |
baby | -0.041 |
Token | it |
Token ID | 340 |
Feature activation | +4.06e+00 |
Pos logit contributions | |
beh | +0.072 |
ash | +0.070 |
aring | +0.069 |
rav | +0.068 |
baby | +0.068 |
Neg logit contributions | |
suddenly | -0.025 |
a | -0.024 |
two | -0.021 |
opened | -0.019 |
was | -0.017 |
Token | too |
Token ID | 1165 |
Feature activation | +9.23e+00 |
Pos logit contributions | |
beh | +0.181 |
rav | +0.177 |
mir | +0.174 |
baby | +0.172 |
ocol | +0.172 |
Neg logit contributions | |
suddenly | -0.083 |
a | -0.080 |
two | -0.075 |
opened | -0.074 |
pretended | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
beh | +0.383 |
unatt | +0.364 |
rav | +0.363 |
ash | +0.362 |
losers | +0.357 |
Neg logit contributions | |
suddenly | -0.211 |
opened | -0.197 |
two | -0.191 |
a | -0.189 |
Suddenly | -0.174 |
Token | She |
Token ID | 1375 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
beh | +0.050 |
unatt | +0.049 |
forward | +0.048 |
ash | +0.048 |
rav | +0.047 |
Neg logit contributions | |
suddenly | -0.036 |
opened | -0.035 |
pretended | -0.034 |
two | -0.033 |
Suddenly | -0.031 |
Token | watched |
Token ID | 7342 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
suddenly | +0.030 |
two | +0.029 |
Suddenly | +0.029 |
pretended | +0.027 |
opened | +0.027 |
Neg logit contributions | |
beh | -0.062 |
unatt | -0.061 |
forward | -0.060 |
losers | -0.059 |
mir | -0.058 |
Token | as |
Token ID | 355 |
Feature activation | -5.18e+00 |
Pos logit contributions | |
suddenly | +0.044 |
opened | +0.040 |
pretended | +0.038 |
two | +0.038 |
Suddenly | +0.036 |
Neg logit contributions | |
beh | -0.107 |
unatt | -0.103 |
ash | -0.102 |
rav | -0.100 |
forward | -0.099 |
Token | the |
Token ID | 262 |
Feature activation | +5.84e-01 |
Pos logit contributions | |
suddenly | +0.105 |
pretended | +0.104 |
opened | +0.103 |
Suddenly | +0.096 |
two | +0.090 |
Neg logit contributions | |
forward | -0.201 |
beh | -0.199 |
unatt | -0.196 |
rav | -0.191 |
overs | -0.189 |
Token | , |
Token ID | 11 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
suddenly | +0.050 |
a | +0.046 |
two | +0.044 |
opened | +0.040 |
Suddenly | +0.036 |
Neg logit contributions | |
beh | -0.222 |
ash | -0.216 |
unatt | -0.216 |
opl | -0.216 |
ksh | -0.208 |
Token | " |
Token ID | 366 |
Feature activation | +8.96e+00 |
Pos logit contributions | |
suddenly | +0.024 |
Suddenly | +0.023 |
opened | +0.022 |
two | +0.022 |
a | +0.022 |
Neg logit contributions | |
beh | -0.058 |
unatt | -0.054 |
ash | -0.054 |
opl | -0.051 |
off | -0.051 |
Token | You |
Token ID | 1639 |
Feature activation | +3.91e+00 |
Pos logit contributions | |
beh | +0.282 |
unatt | +0.278 |
forward | +0.267 |
ash | +0.266 |
aring | +0.259 |
Neg logit contributions | |
suddenly | -0.272 |
opened | -0.255 |
pretended | -0.254 |
two | -0.250 |
Suddenly | -0.245 |
Token | are |
Token ID | 389 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
beh | +0.110 |
losers | +0.108 |
forward | +0.105 |
unatt | +0.101 |
ash | +0.099 |
Neg logit contributions | |
suddenly | -0.124 |
Suddenly | -0.116 |
opened | -0.111 |
pretended | -0.110 |
clicked | -0.104 |
Token | getting |
Token ID | 1972 |
Feature activation | +6.88e+00 |
Pos logit contributions | |
beh | +0.112 |
unatt | +0.109 |
ash | +0.108 |
rav | +0.108 |
aring | +0.106 |
Neg logit contributions | |
suddenly | -0.063 |
opened | -0.062 |
two | -0.061 |
a | -0.060 |
pretended | -0.059 |
Token | too |
Token ID | 1165 |
Feature activation | +9.28e+00 |
Pos logit contributions | |
beh | +0.267 |
rav | +0.262 |
unatt | +0.260 |
ash | +0.259 |
opl | +0.259 |
Neg logit contributions | |
suddenly | -0.160 |
opened | -0.152 |
two | -0.150 |
a | -0.148 |
pretended | -0.142 |
Token | crazy |
Token ID | 7165 |
Feature activation | +3.37e+00 |
Pos logit contributions | |
beh | +0.337 |
unatt | +0.328 |
ash | +0.320 |
rav | +0.316 |
forward | +0.314 |
Neg logit contributions | |
suddenly | -0.241 |
opened | -0.233 |
two | -0.226 |
pretended | -0.219 |
a | -0.214 |
Token | , |
Token ID | 11 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
beh | +0.134 |
unatt | +0.129 |
rav | +0.126 |
ash | +0.124 |
mir | +0.124 |
Neg logit contributions | |
suddenly | -0.077 |
two | -0.071 |
opened | -0.070 |
pretended | -0.070 |
a | -0.069 |
Token | I |
Token ID | 314 |
Feature activation | +5.09e+00 |
Pos logit contributions | |
suddenly | +0.023 |
two | +0.022 |
opened | +0.021 |
a | +0.021 |
pretended | +0.021 |
Neg logit contributions | |
beh | -0.050 |
unatt | -0.047 |
ash | -0.046 |
opl | -0.045 |
rav | -0.045 |
Token | don |
Token ID | 836 |
Feature activation | +4.63e+00 |
Pos logit contributions | |
beh | +0.165 |
unatt | +0.160 |
ash | +0.156 |
forward | +0.156 |
rav | +0.152 |
Neg logit contributions | |
suddenly | -0.148 |
opened | -0.141 |
pretended | -0.138 |
two | -0.135 |
Suddenly | -0.132 |
Token | 't |
Token ID | 470 |
Feature activation | +7.53e-03 |
Pos logit contributions | |
unatt | +0.224 |
beh | +0.217 |
ash | +0.215 |
mir | +0.215 |
rav | +0.212 |
Neg logit contributions | |
suddenly | -0.075 |
two | -0.072 |
a | -0.066 |
opened | -0.062 |
was | -0.055 |
Token | said |
Token ID | 531 |
Feature activation | -4.78e+00 |
Pos logit contributions | |
suddenly | +0.040 |
two | +0.039 |
pretended | +0.036 |
Suddenly | +0.036 |
opened | +0.036 |
Neg logit contributions | |
beh | -0.096 |
losers | -0.091 |
mir | -0.090 |
rav | -0.089 |
ash | -0.089 |
Token | to |
Token ID | 284 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
suddenly | +0.066 |
a | +0.059 |
two | +0.056 |
opened | +0.055 |
Suddenly | +0.045 |
Neg logit contributions | |
beh | -0.256 |
ash | -0.256 |
unatt | -0.249 |
opl | -0.248 |
ksh | -0.239 |
Token | her |
Token ID | 607 |
Feature activation | +2.87e+00 |
Pos logit contributions | |
beh | +0.062 |
unatt | +0.060 |
forward | +0.059 |
ash | +0.059 |
baby | +0.058 |
Neg logit contributions | |
suddenly | -0.050 |
opened | -0.048 |
pretended | -0.047 |
two | -0.044 |
Suddenly | -0.044 |
Token | dad |
Token ID | 9955 |
Feature activation | +3.00e+00 |
Pos logit contributions | |
forward | +0.079 |
ambulance | +0.078 |
unatt | +0.077 |
baby | +0.076 |
beh | +0.075 |
Neg logit contributions | |
suddenly | -0.087 |
Suddenly | -0.083 |
pretended | -0.082 |
opened | -0.081 |
clicked | -0.078 |
Token | , |
Token ID | 11 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
beh | +0.148 |
unatt | +0.144 |
ash | +0.140 |
losers | +0.138 |
forward | +0.137 |
Neg logit contributions | |
suddenly | -0.041 |
opened | -0.037 |
two | -0.035 |
pretended | -0.034 |
a | -0.032 |
Token | " |
Token ID | 366 |
Feature activation | +8.69e+00 |
Pos logit contributions | |
suddenly | +0.032 |
opened | +0.029 |
a | +0.029 |
two | +0.029 |
Suddenly | +0.029 |
Neg logit contributions | |
beh | -0.058 |
ash | -0.055 |
unatt | -0.054 |
off | -0.052 |
rav | -0.051 |
Token | Look |
Token ID | 8567 |
Feature activation | -4.52e+00 |
Pos logit contributions | |
beh | +0.297 |
unatt | +0.289 |
ash | +0.279 |
forward | +0.278 |
aring | +0.271 |
Neg logit contributions | |
suddenly | -0.247 |
opened | -0.230 |
pretended | -0.227 |
two | -0.224 |
Suddenly | -0.219 |
Token | , |
Token ID | 11 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
suddenly | +0.088 |
opened | +0.081 |
pretended | +0.078 |
two | +0.077 |
a | +0.073 |
Neg logit contributions | |
beh | -0.195 |
unatt | -0.187 |
ash | -0.183 |
forward | -0.180 |
rav | -0.179 |
Token | monkeys |
Token ID | 26697 |
Feature activation | -6.73e-01 |
Pos logit contributions | |
suddenly | +0.030 |
opened | +0.028 |
two | +0.028 |
pretended | +0.027 |
a | +0.026 |
Neg logit contributions | |
beh | -0.044 |
unatt | -0.041 |
ash | -0.041 |
forward | -0.040 |
agging | -0.039 |
Token | ! |
Token ID | 0 |
Feature activation | +8.33e-01 |
Pos logit contributions | |
suddenly | +0.008 |
opened | +0.008 |
two | +0.007 |
a | +0.007 |
pretended | +0.007 |
Neg logit contributions | |
beh | -0.034 |
unatt | -0.033 |
ash | -0.032 |
forward | -0.032 |
ksh | -0.032 |
Token | Can |
Token ID | 1680 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
beh | +0.027 |
unatt | +0.026 |
ash | +0.025 |
forward | +0.025 |
agging | +0.024 |
Neg logit contributions | |
suddenly | -0.025 |
opened | -0.024 |
two | -0.023 |
pretended | -0.023 |
Suddenly | -0.023 |
Token | I |
Token ID | 314 |
Feature activation | +5.06e+00 |
Pos logit contributions | |
beh | +0.085 |
ash | +0.083 |
forward | +0.078 |
opl | +0.078 |
agging | +0.077 |
Neg logit contributions | |
suddenly | -0.088 |
opened | -0.083 |
two | -0.080 |
a | -0.079 |
pretended | -0.077 |
Token | have |
Token ID | 423 |
Feature activation | -5.01e-03 |
Pos logit contributions | |
beh | +0.218 |
unatt | +0.209 |
ash | +0.200 |
rav | +0.199 |
opl | +0.199 |
Neg logit contributions | |
opened | -0.114 |
suddenly | -0.112 |
pretended | -0.103 |
a | -0.100 |
two | -0.100 |
Token | some |
Token ID | 617 |
Feature activation | -5.52e-01 |
Pos logit contributions | |
suddenly | +0.000 |
opened | +0.000 |
pretended | +0.000 |
two | +0.000 |
Suddenly | +0.000 |
Neg logit contributions | |
beh | -0.000 |
unatt | -0.000 |
forward | -0.000 |
ash | -0.000 |
rav | -0.000 |
Token | water |
Token ID | 1660 |
Feature activation | -5.97e-01 |
Pos logit contributions | |
suddenly | +0.013 |
opened | +0.012 |
two | +0.012 |
pretended | +0.011 |
a | +0.011 |
Neg logit contributions | |
beh | -0.022 |
unatt | -0.021 |
forward | -0.021 |
rav | -0.021 |
ash | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | -5.06e-01 |
Pos logit contributions | |
suddenly | +0.012 |
a | +0.012 |
opened | +0.011 |
two | +0.011 |
pretended | +0.010 |
Neg logit contributions | |
beh | -0.028 |
rav | -0.025 |
opl | -0.025 |
aring | -0.025 |
agging | -0.024 |
Token | please |
Token ID | 3387 |
Feature activation | +9.28e+00 |
Pos logit contributions | |
suddenly | +0.012 |
opened | +0.011 |
two | +0.011 |
pretended | +0.011 |
a | +0.011 |
Neg logit contributions | |
beh | -0.021 |
unatt | -0.020 |
ash | -0.019 |
forward | -0.019 |
rav | -0.019 |
Token | ?" |
Token ID | 1701 |
Feature activation | -6.09e-01 |
Pos logit contributions | |
beh | +0.293 |
unatt | +0.286 |
forward | +0.279 |
ash | +0.279 |
rav | +0.272 |
Neg logit contributions | |
suddenly | -0.268 |
opened | -0.255 |
pretended | -0.253 |
two | -0.244 |
Suddenly | -0.244 |
Token | the |
Token ID | 262 |
Feature activation | +9.98e-01 |
Pos logit contributions | |
inserted | +0.017 |
pretended | +0.017 |
opened | +0.017 |
suddenly | +0.017 |
clicked | +0.016 |
Neg logit contributions | |
Go | -0.016 |
baby | -0.015 |
losers | -0.015 |
ambulance | -0.015 |
future | -0.015 |
Token | yellow |
Token ID | 7872 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
unatt | +0.031 |
beh | +0.030 |
forward | +0.030 |
baby | +0.030 |
ash | +0.029 |
Neg logit contributions | |
suddenly | -0.028 |
pretended | -0.027 |
opened | -0.026 |
Suddenly | -0.025 |
clicked | -0.024 |
Token | flower |
Token ID | 15061 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
suddenly | +0.013 |
opened | +0.012 |
pretended | +0.012 |
two | +0.011 |
Suddenly | +0.011 |
Neg logit contributions | |
beh | -0.018 |
unatt | -0.017 |
ash | -0.017 |
forward | -0.017 |
rav | -0.016 |
Token | asks |
Token ID | 7893 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
suddenly | +0.044 |
opened | +0.042 |
pretended | +0.040 |
two | +0.039 |
Suddenly | +0.037 |
Neg logit contributions | |
beh | -0.076 |
unatt | -0.073 |
ash | -0.072 |
forward | -0.071 |
rav | -0.070 |
Token | swing |
Token ID | 9628 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
beh | +0.095 |
unatt | +0.092 |
forward | +0.090 |
ash | +0.090 |
rav | +0.088 |
Neg logit contributions | |
suddenly | -0.068 |
opened | -0.065 |
pretended | -0.064 |
two | -0.062 |
Suddenly | -0.060 |
Token | , |
Token ID | 11 |
Feature activation | -6.05e-01 |
Pos logit contributions | |
a | +0.001 |
suddenly | +0.001 |
two | +0.000 |
pretended | +0.000 |
opened | +0.000 |
Neg logit contributions | |
beh | -0.001 |
mir | -0.001 |
ife | -0.001 |
aring | -0.001 |
ash | -0.001 |
Token | but |
Token ID | 475 |
Feature activation | -3.65e+00 |
Pos logit contributions | |
a | +0.014 |
suddenly | +0.013 |
two | +0.013 |
opened | +0.013 |
pretended | +0.013 |
Neg logit contributions | |
beh | -0.027 |
ash | -0.026 |
opl | -0.025 |
agging | -0.024 |
ocol | -0.024 |
Token | not |
Token ID | 407 |
Feature activation | +9.57e-01 |
Pos logit contributions | |
suddenly | +0.106 |
two | +0.102 |
opened | +0.100 |
a | +0.098 |
pretended | +0.094 |
Neg logit contributions | |
beh | -0.129 |
ash | -0.124 |
unatt | -0.120 |
opl | -0.120 |
agging | -0.116 |
Token | too |
Token ID | 1165 |
Feature activation | +9.14e+00 |
Pos logit contributions | |
beh | +0.035 |
ash | +0.034 |
forward | +0.034 |
overs | +0.034 |
unatt | +0.034 |
Neg logit contributions | |
suddenly | -0.024 |
opened | -0.023 |
two | -0.023 |
a | -0.023 |
pretended | -0.022 |
Token | high |
Token ID | 1029 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
beh | +0.383 |
ksh | +0.379 |
ash | +0.379 |
rav | +0.374 |
unatt | +0.371 |
Neg logit contributions | |
a | -0.207 |
two | -0.204 |
opened | -0.190 |
suddenly | -0.183 |
little | -0.167 |
Token | . |
Token ID | 13 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
beh | +0.089 |
ash | +0.083 |
rav | +0.082 |
unatt | +0.082 |
mir | +0.082 |
Neg logit contributions | |
suddenly | -0.060 |
two | -0.057 |
a | -0.057 |
pretended | -0.054 |
opened | -0.053 |
Token | We |
Token ID | 775 |
Feature activation | +4.94e+00 |
Pos logit contributions | |
beh | +0.048 |
unatt | +0.047 |
ash | +0.044 |
forward | +0.043 |
agging | +0.043 |
Neg logit contributions | |
suddenly | -0.056 |
two | -0.054 |
opened | -0.054 |
Suddenly | -0.053 |
pretended | -0.053 |
Token | will |
Token ID | 481 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
unatt | +0.179 |
beh | +0.179 |
ash | +0.174 |
overs | +0.171 |
baby | +0.169 |
Neg logit contributions | |
opened | -0.127 |
suddenly | -0.126 |
pretended | -0.123 |
two | -0.121 |
a | -0.116 |
Token | be |
Token ID | 307 |
Feature activation | +6.18e+00 |
Pos logit contributions | |
beh | +0.097 |
unatt | +0.094 |
ash | +0.092 |
forward | +0.092 |
rav | +0.090 |
Neg logit contributions | |
suddenly | -0.063 |
opened | -0.060 |
pretended | -0.058 |
two | -0.057 |
Suddenly | -0.055 |
Token | safe |
Token ID | 3338 |
Feature activation | +7.62e+00 |
Pos logit contributions | |
rav | +0.205 |
ash | +0.197 |
umm | +0.195 |
baby | +0.195 |
beh | +0.191 |
Neg logit contributions | |
two | -0.195 |
a | -0.195 |
suddenly | -0.187 |
opened | -0.181 |
pretended | -0.176 |
Token | pictures |
Token ID | 5986 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
suddenly | +0.046 |
pretended | +0.041 |
Suddenly | +0.041 |
opened | +0.041 |
two | +0.039 |
Neg logit contributions | |
beh | -0.038 |
unatt | -0.038 |
baby | -0.037 |
forward | -0.037 |
ash | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | +8.38e-01 |
Pos logit contributions | |
suddenly | +0.060 |
opened | +0.057 |
pretended | +0.056 |
two | +0.054 |
Suddenly | +0.053 |
Neg logit contributions | |
beh | -0.086 |
unatt | -0.083 |
ash | -0.081 |
forward | -0.081 |
rav | -0.079 |
Token | words |
Token ID | 2456 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
beh | +0.022 |
losers | +0.021 |
mir | +0.021 |
baby | +0.021 |
rav | +0.021 |
Neg logit contributions | |
suddenly | -0.026 |
Suddenly | -0.024 |
opened | -0.024 |
pretended | -0.023 |
two | -0.022 |
Token | that |
Token ID | 326 |
Feature activation | -5.65e+00 |
Pos logit contributions | |
suddenly | +0.033 |
opened | +0.031 |
pretended | +0.030 |
two | +0.030 |
a | +0.028 |
Neg logit contributions | |
beh | -0.059 |
unatt | -0.058 |
ash | -0.056 |
rav | -0.055 |
forward | -0.054 |
Token | their |
Token ID | 511 |
Feature activation | +4.32e+00 |
Pos logit contributions | |
suddenly | +0.104 |
opened | +0.100 |
pretended | +0.095 |
two | +0.092 |
a | +0.085 |
Neg logit contributions | |
beh | -0.248 |
unatt | -0.243 |
ash | -0.239 |
forward | -0.237 |
aring | -0.233 |
Token | mom |
Token ID | 1995 |
Feature activation | +4.12e+00 |
Pos logit contributions | |
beh | +0.154 |
unatt | +0.145 |
ash | +0.144 |
forward | +0.141 |
aring | +0.138 |
Neg logit contributions | |
suddenly | -0.116 |
opened | -0.116 |
pretended | -0.114 |
two | -0.112 |
a | -0.108 |
Token | read |
Token ID | 1100 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
beh | +0.177 |
unatt | +0.170 |
ash | +0.170 |
baby | +0.169 |
aring | +0.167 |
Neg logit contributions | |
suddenly | -0.068 |
Suddenly | -0.061 |
two | -0.057 |
opened | -0.054 |
pretended | -0.054 |
Token | to |
Token ID | 284 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
suddenly | +0.055 |
a | +0.054 |
opened | +0.053 |
two | +0.051 |
pretended | +0.050 |
Neg logit contributions | |
beh | -0.114 |
unatt | -0.107 |
aring | -0.104 |
rav | -0.104 |
ash | -0.104 |
Token | them |
Token ID | 606 |
Feature activation | +6.52e+00 |
Pos logit contributions | |
beh | +0.083 |
unatt | +0.080 |
ash | +0.079 |
rav | +0.078 |
aring | +0.078 |
Neg logit contributions | |
suddenly | -0.053 |
opened | -0.051 |
a | -0.050 |
two | -0.050 |
pretended | -0.049 |
Token | every |
Token ID | 790 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
beh | +0.269 |
unatt | +0.259 |
rav | +0.253 |
ash | +0.250 |
mir | +0.250 |
Neg logit contributions | |
suddenly | -0.143 |
opened | -0.135 |
a | -0.133 |
two | -0.132 |
pretended | -0.131 |
Token | night |
Token ID | 1755 |
Feature activation | -3.11e+00 |
Pos logit contributions | |
suddenly | +0.073 |
opened | +0.071 |
pretended | +0.068 |
two | +0.068 |
a | +0.066 |
Neg logit contributions | |
beh | -0.086 |
unatt | -0.081 |
rav | -0.079 |
ash | -0.079 |
aring | -0.078 |
Token | a |
Token ID | 257 |
Feature activation | -7.53e+00 |
Pos logit contributions | |
unatt | +0.170 |
beh | +0.170 |
mir | +0.163 |
forward | +0.161 |
losers | +0.159 |
Neg logit contributions | |
suddenly | -0.096 |
pretended | -0.092 |
opened | -0.088 |
Suddenly | -0.085 |
two | -0.079 |
Token | shelter |
Token ID | 11772 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
suddenly | +0.208 |
pretended | +0.194 |
Suddenly | +0.191 |
opened | +0.190 |
two | +0.190 |
Neg logit contributions | |
mir | -0.228 |
unatt | -0.228 |
forward | -0.223 |
hurricane | -0.217 |
rav | -0.217 |
Token | to |
Token ID | 284 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
beh | +0.038 |
unatt | +0.038 |
forward | +0.037 |
ash | +0.036 |
aring | +0.035 |
Neg logit contributions | |
suddenly | -0.030 |
opened | -0.029 |
pretended | -0.028 |
two | -0.028 |
Suddenly | -0.027 |
Token | hide |
Token ID | 7808 |
Feature activation | +9.58e-01 |
Pos logit contributions | |
beh | +0.074 |
unatt | +0.073 |
forward | +0.070 |
ash | +0.069 |
rav | +0.067 |
Neg logit contributions | |
suddenly | -0.070 |
opened | -0.066 |
pretended | -0.066 |
two | -0.064 |
Suddenly | -0.064 |
Token | from |
Token ID | 422 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
beh | +0.041 |
unatt | +0.040 |
ash | +0.039 |
forward | +0.039 |
rav | +0.039 |
Neg logit contributions | |
suddenly | -0.018 |
opened | -0.017 |
pretended | -0.017 |
two | -0.016 |
Suddenly | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | +7.24e-01 |
Pos logit contributions | |
pretended | +0.034 |
suddenly | +0.034 |
opened | +0.034 |
clicked | +0.032 |
Suddenly | +0.032 |
Neg logit contributions | |
unatt | -0.050 |
future | -0.045 |
forward | -0.045 |
mir | -0.044 |
beh | -0.043 |
Token | rain |
Token ID | 6290 |
Feature activation | +5.77e+00 |
Pos logit contributions | |
beh | +0.029 |
ash | +0.027 |
rav | +0.027 |
aring | +0.026 |
forward | +0.026 |
Neg logit contributions | |
suddenly | -0.019 |
opened | -0.018 |
two | -0.018 |
a | -0.017 |
pretended | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
beh | +0.209 |
unatt | +0.201 |
forward | +0.198 |
ash | +0.198 |
rav | +0.194 |
Neg logit contributions | |
suddenly | -0.147 |
opened | -0.140 |
pretended | -0.137 |
two | -0.132 |
Suddenly | -0.129 |
Token | Twe |
Token ID | 24205 |
Feature activation | +7.93e+00 |
Pos logit contributions | |
beh | +0.048 |
unatt | +0.046 |
forward | +0.046 |
aring | +0.045 |
baby | +0.044 |
Neg logit contributions | |
suddenly | -0.033 |
pretended | -0.032 |
opened | -0.030 |
two | -0.029 |
pressed | -0.027 |
Token | ety |
Token ID | 2963 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
beh | +0.181 |
unatt | +0.161 |
ash | +0.160 |
aring | +0.153 |
baby | +0.149 |
Neg logit contributions | |
suddenly | -0.324 |
opened | -0.315 |
pretended | -0.308 |
two | -0.305 |
Suddenly | -0.298 |
Token | saw |
Token ID | 2497 |
Feature activation | -5.61e+00 |
Pos logit contributions | |
Two | +0.023 |
Melvin | +0.022 |
Under | +0.022 |
disc | +0.022 |
growth | +0.022 |
Neg logit contributions | |
unatt | -0.043 |
beh | -0.042 |
mir | -0.042 |
aring | -0.041 |
off | -0.040 |
Token | ball |
Token ID | 2613 |
Feature activation | +2.70e-01 |
Pos logit contributions | |
forward | +0.144 |
unatt | +0.143 |
ash | +0.140 |
losers | +0.140 |
beh | +0.140 |
Neg logit contributions | |
suddenly | -0.178 |
opened | -0.174 |
pretended | -0.168 |
Suddenly | -0.161 |
clicked | -0.158 |
Token | all |
Token ID | 477 |
Feature activation | +4.11e+00 |
Pos logit contributions | |
beh | +0.010 |
unatt | +0.010 |
ash | +0.009 |
rav | +0.009 |
forward | +0.009 |
Neg logit contributions | |
suddenly | -0.007 |
opened | -0.006 |
pretended | -0.006 |
two | -0.006 |
Suddenly | -0.006 |
Token | day |
Token ID | 1110 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
forward | +0.105 |
beh | +0.105 |
ash | +0.103 |
unatt | +0.103 |
baby | +0.096 |
Neg logit contributions | |
suddenly | -0.145 |
opened | -0.139 |
pretended | -0.136 |
two | -0.132 |
Suddenly | -0.132 |
Token | , |
Token ID | 11 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
suddenly | +0.054 |
opened | +0.051 |
pretended | +0.050 |
two | +0.048 |
Suddenly | +0.046 |
Neg logit contributions | |
beh | -0.099 |
unatt | -0.096 |
forward | -0.094 |
ash | -0.094 |
rav | -0.092 |
Token | and |
Token ID | 290 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
suddenly | +0.029 |
opened | +0.027 |
pretended | +0.026 |
two | +0.026 |
Suddenly | +0.025 |
Neg logit contributions | |
beh | -0.048 |
unatt | -0.047 |
forward | -0.046 |
ash | -0.046 |
rav | -0.045 |
Token | they |
Token ID | 484 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
beh | +0.037 |
forward | +0.037 |
unatt | +0.037 |
losers | +0.036 |
rav | +0.036 |
Neg logit contributions | |
opened | -0.025 |
suddenly | -0.025 |
two | -0.023 |
pretended | -0.023 |
Suddenly | -0.023 |
Token | became |
Token ID | 2627 |
Feature activation | -7.89e-01 |
Pos logit contributions | |
beh | +0.050 |
unatt | +0.049 |
forward | +0.048 |
ash | +0.047 |
rav | +0.047 |
Neg logit contributions | |
suddenly | -0.028 |
opened | -0.026 |
Suddenly | -0.025 |
two | -0.025 |
pretended | -0.025 |
Token | best |
Token ID | 1266 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
suddenly | +0.023 |
opened | +0.022 |
pretended | +0.022 |
two | +0.022 |
a | +0.021 |
Neg logit contributions | |
beh | -0.025 |
unatt | -0.025 |
ash | -0.024 |
forward | -0.024 |
rav | -0.024 |
Token | friends |
Token ID | 2460 |
Feature activation | +8.64e-01 |
Pos logit contributions | |
unatt | +0.102 |
beh | +0.099 |
ash | +0.097 |
ife | +0.094 |
ksh | +0.093 |
Neg logit contributions | |
two | -0.049 |
suddenly | -0.048 |
opened | -0.044 |
pretended | -0.044 |
a | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
beh | +0.039 |
ash | +0.037 |
ocol | +0.037 |
unatt | +0.037 |
mir | +0.037 |
Neg logit contributions | |
suddenly | -0.016 |
a | -0.016 |
two | -0.015 |
pretended | -0.014 |
opened | -0.014 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
beh | +0.045 |
forward | +0.043 |
ash | +0.043 |
unatt | +0.043 |
rav | +0.042 |
Neg logit contributions | |
suddenly | -0.025 |
opened | -0.024 |
pretended | -0.024 |
two | -0.021 |
Suddenly | -0.020 |
Token | bird |
Token ID | 6512 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
agging | +0.032 |
opl | +0.032 |
ash | +0.032 |
unatt | +0.031 |
baby | +0.030 |
Neg logit contributions | |
pretended | -0.031 |
suddenly | -0.030 |
opened | -0.029 |
Suddenly | -0.027 |
led | -0.027 |
Token | in |
Token ID | 287 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
pretended | +0.039 |
suddenly | +0.039 |
opened | +0.038 |
two | +0.037 |
Suddenly | +0.036 |
Neg logit contributions | |
beh | -0.048 |
unatt | -0.046 |
aring | -0.045 |
ash | -0.044 |
forward | -0.044 |
Token | a |
Token ID | 257 |
Feature activation | -7.69e+00 |
Pos logit contributions | |
pretended | +0.037 |
suddenly | +0.032 |
Suddenly | +0.032 |
opened | +0.030 |
clicked | +0.030 |
Neg logit contributions | |
forward | -0.062 |
public | -0.061 |
unatt | -0.061 |
overs | -0.061 |
future | -0.060 |
Token | tree |
Token ID | 5509 |
Feature activation | -5.62e-01 |
Pos logit contributions | |
suddenly | +0.184 |
opened | +0.175 |
two | +0.168 |
a | +0.162 |
Suddenly | +0.156 |
Neg logit contributions | |
beh | -0.329 |
rav | -0.301 |
ash | -0.298 |
aring | -0.290 |
unatt | -0.290 |
Token | . |
Token ID | 13 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
suddenly | +0.015 |
pretended | +0.015 |
opened | +0.015 |
two | +0.014 |
Suddenly | +0.014 |
Neg logit contributions | |
beh | -0.019 |
unatt | -0.018 |
forward | -0.018 |
ash | -0.018 |
aring | -0.017 |
Token | Beth |
Token ID | 19223 |
Feature activation | +3.55e-01 |
Pos logit contributions | |
beh | +0.032 |
forward | +0.031 |
baby | +0.031 |
unatt | +0.030 |
ash | +0.030 |
Neg logit contributions | |
pretended | -0.029 |
suddenly | -0.029 |
opened | -0.027 |
two | -0.026 |
imagined | -0.024 |
Token | couldn |
Token ID | 3521 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
beh | +0.014 |
aring | +0.013 |
baby | +0.013 |
unatt | +0.013 |
off | +0.013 |
Neg logit contributions | |
two | -0.005 |
pretended | -0.005 |
suddenly | -0.005 |
a | -0.005 |
Under | -0.005 |
Token | 't |
Token ID | 470 |
Feature activation | -4.46e-01 |
Pos logit contributions | |
suddenly | +0.034 |
opened | +0.031 |
two | +0.030 |
a | +0.030 |
Suddenly | +0.028 |
Neg logit contributions | |
beh | -0.075 |
unatt | -0.075 |
ash | -0.072 |
rav | -0.070 |
forward | -0.070 |
Token | help |
Token ID | 1037 |
Feature activation | +8.47e-01 |
Pos logit contributions | |
suddenly | +0.015 |
opened | +0.014 |
pretended | +0.014 |
two | +0.014 |
Suddenly | +0.014 |
Neg logit contributions | |
beh | -0.012 |
unatt | -0.012 |
forward | -0.012 |
ash | -0.011 |
aring | -0.011 |
Token | but |
Token ID | 475 |
Feature activation | -4.05e+00 |
Pos logit contributions | |
beh | +0.039 |
unatt | +0.038 |
ash | +0.037 |
rav | +0.037 |
forward | +0.037 |
Neg logit contributions | |
suddenly | -0.014 |
opened | -0.013 |
pretended | -0.012 |
two | -0.012 |
a | -0.011 |
Token | be |
Token ID | 307 |
Feature activation | +5.83e+00 |
Pos logit contributions | |
suddenly | +0.096 |
opened | +0.091 |
pretended | +0.089 |
two | +0.086 |
Suddenly | +0.084 |
Neg logit contributions | |
beh | -0.154 |
unatt | -0.150 |
ash | -0.147 |
forward | -0.146 |
rav | -0.143 |
Token | sad |
Token ID | 6507 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
beh | +0.251 |
ash | +0.247 |
unatt | +0.247 |
rav | +0.237 |
forward | +0.234 |
Neg logit contributions | |
suddenly | -0.229 |
opened | -0.227 |
two | -0.217 |
pretended | -0.214 |
a | -0.209 |
Token | and |
Token ID | 290 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
beh | +0.089 |
unatt | +0.088 |
forward | +0.086 |
ash | +0.085 |
aring | +0.084 |
Neg logit contributions | |
suddenly | -0.079 |
opened | -0.075 |
pretended | -0.074 |
Suddenly | -0.071 |
two | -0.070 |
Token | angry |
Token ID | 7954 |
Feature activation | +6.18e+00 |
Pos logit contributions | |
beh | +0.044 |
unatt | +0.042 |
forward | +0.042 |
losers | +0.039 |
baby | +0.039 |
Neg logit contributions | |
suddenly | -0.058 |
opened | -0.055 |
pretended | -0.054 |
Suddenly | -0.054 |
two | -0.054 |
Token | . |
Token ID | 13 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
beh | +0.229 |
unatt | +0.223 |
ash | +0.218 |
forward | +0.215 |
rav | +0.213 |
Neg logit contributions | |
suddenly | -0.154 |
opened | -0.145 |
pretended | -0.142 |
two | -0.141 |
Suddenly | -0.135 |
Token | They |
Token ID | 1119 |
Feature activation | -7.81e-01 |
Pos logit contributions | |
beh | +0.051 |
forward | +0.050 |
baby | +0.049 |
overs | +0.049 |
ash | +0.048 |
Neg logit contributions | |
suddenly | -0.059 |
opened | -0.056 |
pretended | -0.055 |
two | -0.052 |
clicked | -0.050 |
Token | cried |
Token ID | 16896 |
Feature activation | -9.17e-01 |
Pos logit contributions | |
suddenly | +0.018 |
Suddenly | +0.017 |
two | +0.016 |
Melvin | +0.014 |
Far | +0.014 |
Neg logit contributions | |
losers | -0.026 |
beh | -0.025 |
unatt | -0.024 |
forward | -0.024 |
dishonest | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
a | +0.013 |
two | +0.013 |
suddenly | +0.012 |
opened | +0.011 |
pretended | +0.011 |
Neg logit contributions | |
beh | -0.046 |
unatt | -0.044 |
opl | -0.043 |
ash | -0.043 |
rav | -0.043 |
Token | shouted |
Token ID | 17293 |
Feature activation | -3.17e+00 |
Pos logit contributions | |
beh | +0.067 |
unatt | +0.064 |
losers | +0.063 |
forward | +0.063 |
baby | +0.060 |
Neg logit contributions | |
suddenly | -0.035 |
Suddenly | -0.032 |
two | -0.030 |
opened | -0.030 |
pretended | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
a | +0.070 |
two | +0.058 |
suddenly | +0.052 |
Suddenly | +0.050 |
was | +0.049 |
Neg logit contributions | |
beh | -0.162 |
umm | -0.150 |
opl | -0.149 |
mir | -0.148 |
chie | -0.148 |
Token |
|
Token ID | 198 |
Feature activation | +7.02e-01 |
Pos logit contributions | |
beh | +0.058 |
unatt | +0.057 |
forward | +0.056 |
ash | +0.055 |
rav | +0.054 |
Neg logit contributions | |
suddenly | -0.052 |
opened | -0.050 |
pretended | -0.049 |
two | -0.047 |
Suddenly | -0.045 |
Token |
|
Token ID | 198 |
Feature activation | +6.79e-01 |
Pos logit contributions | |
forward | +0.025 |
beh | +0.025 |
aring | +0.025 |
baby | +0.025 |
rav | +0.025 |
Neg logit contributions | |
pretended | -0.016 |
suddenly | -0.015 |
opened | -0.015 |
clicked | -0.012 |
two | -0.012 |
Token | things |
Token ID | 1243 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
beh | +0.056 |
rav | +0.054 |
aring | +0.054 |
opl | +0.053 |
baby | +0.052 |
Neg logit contributions | |
suddenly | -0.029 |
two | -0.028 |
a | -0.027 |
opened | -0.027 |
pretended | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
beh | +0.010 |
rav | +0.010 |
ash | +0.010 |
ksh | +0.009 |
baby | +0.009 |
Neg logit contributions | |
suddenly | -0.004 |
opened | -0.004 |
a | -0.004 |
was | -0.003 |
two | -0.003 |
Token | She |
Token ID | 1375 |
Feature activation | -3.19e+00 |
Pos logit contributions | |
suddenly | +0.002 |
opened | +0.002 |
Suddenly | +0.002 |
two | +0.002 |
pretended | +0.002 |
Neg logit contributions | |
beh | -0.005 |
unatt | -0.005 |
ash | -0.005 |
forward | -0.005 |
rav | -0.005 |
Token | told |
Token ID | 1297 |
Feature activation | -2.86e+00 |
Pos logit contributions | |
suddenly | +0.050 |
opened | +0.046 |
pretended | +0.044 |
two | +0.042 |
Suddenly | +0.040 |
Neg logit contributions | |
beh | -0.145 |
unatt | -0.142 |
ash | -0.139 |
forward | -0.139 |
rav | -0.137 |
Token | him |
Token ID | 683 |
Feature activation | +4.03e+00 |
Pos logit contributions | |
suddenly | +0.068 |
a | +0.065 |
two | +0.065 |
opened | +0.064 |
pretended | +0.059 |
Neg logit contributions | |
beh | -0.113 |
unatt | -0.110 |
ash | -0.107 |
rav | -0.106 |
forward | -0.105 |
Token | that |
Token ID | 326 |
Feature activation | -6.44e+00 |
Pos logit contributions | |
beh | +0.147 |
unatt | +0.144 |
ash | +0.141 |
forward | +0.139 |
rav | +0.136 |
Neg logit contributions | |
suddenly | -0.102 |
opened | -0.097 |
pretended | -0.094 |
two | -0.093 |
Suddenly | -0.089 |
Token | he |
Token ID | 339 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
suddenly | +0.127 |
a | +0.115 |
two | +0.114 |
opened | +0.109 |
pretended | +0.100 |
Neg logit contributions | |
beh | -0.292 |
ash | -0.292 |
agging | -0.291 |
ags | -0.284 |
ksh | -0.283 |
Token | should |
Token ID | 815 |
Feature activation | -3.35e+00 |
Pos logit contributions | |
suddenly | +0.033 |
opened | +0.030 |
pretended | +0.029 |
two | +0.028 |
Suddenly | +0.028 |
Neg logit contributions | |
beh | -0.074 |
unatt | -0.072 |
forward | -0.071 |
ash | -0.071 |
rav | -0.069 |
Token | say |
Token ID | 910 |
Feature activation | -2.84e+00 |
Pos logit contributions | |
suddenly | +0.077 |
opened | +0.073 |
two | +0.071 |
pretended | +0.067 |
a | +0.066 |
Neg logit contributions | |
beh | -0.136 |
unatt | -0.135 |
ash | -0.134 |
forward | -0.131 |
rav | -0.131 |
Token | sorry |
Token ID | 7926 |
Feature activation | +4.48e+00 |
Pos logit contributions | |
a | +0.074 |
two | +0.069 |
suddenly | +0.069 |
opened | +0.060 |
pretended | +0.056 |
Neg logit contributions | |
beh | -0.121 |
ash | -0.115 |
ksh | -0.112 |
umm | -0.109 |
unatt | -0.108 |
Token | to |
Token ID | 284 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
beh | +0.172 |
unatt | +0.166 |
opl | +0.163 |
ocol | +0.162 |
agging | +0.159 |
Neg logit contributions | |
suddenly | -0.112 |
a | -0.107 |
pretended | -0.105 |
opened | -0.104 |
two | -0.103 |
Token | m |
Token ID | 285 |
Feature activation | +9.36e+00 |
Pos logit contributions | |
beh | +0.095 |
unatt | +0.086 |
ash | +0.085 |
aring | +0.085 |
off | +0.084 |
Neg logit contributions | |
two | -0.056 |
suddenly | -0.054 |
a | -0.053 |
pretended | -0.051 |
opened | -0.051 |
Token | ama |
Token ID | 1689 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
unatt | +0.315 |
beh | +0.313 |
forward | +0.312 |
losers | +0.301 |
dishonest | +0.274 |
Neg logit contributions | |
suddenly | -0.253 |
two | -0.241 |
opened | -0.232 |
Suddenly | -0.227 |
was | -0.219 |
Token | loved |
Token ID | 6151 |
Feature activation | +8.98e-01 |
Pos logit contributions | |
beh | +0.050 |
unatt | +0.048 |
ash | +0.047 |
rav | +0.046 |
forward | +0.046 |
Neg logit contributions | |
suddenly | -0.024 |
opened | -0.023 |
pretended | -0.023 |
two | -0.023 |
a | -0.022 |
Token | him |
Token ID | 683 |
Feature activation | +4.86e+00 |
Pos logit contributions | |
beh | +0.032 |
unatt | +0.030 |
ash | +0.030 |
aring | +0.030 |
rav | +0.029 |
Neg logit contributions | |
suddenly | -0.026 |
a | -0.026 |
two | -0.024 |
opened | -0.024 |
pretended | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
beh | +0.206 |
rav | +0.195 |
ash | +0.193 |
ocol | +0.192 |
off | +0.190 |
Neg logit contributions | |
a | -0.118 |
suddenly | -0.111 |
two | -0.107 |
opened | -0.099 |
pretended | -0.096 |
Token | They |
Token ID | 1119 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
beh | +0.056 |
unatt | +0.054 |
forward | +0.054 |
ash | +0.053 |
rav | +0.053 |
Neg logit contributions | |
suddenly | -0.024 |
opened | -0.023 |
pretended | -0.022 |
two | -0.021 |
Suddenly | -0.019 |
Token | were |
Token ID | 547 |
Feature activation | -6.87e-01 |
Pos logit contributions | |
suddenly | +0.023 |
opened | +0.022 |
pretended | +0.021 |
two | +0.020 |
Suddenly | +0.020 |
Neg logit contributions | |
beh | -0.051 |
unatt | -0.050 |
forward | -0.049 |
ash | -0.049 |
rav | -0.048 |
Token | happy |
Token ID | 3772 |
Feature activation | +4.66e+00 |
Pos logit contributions | |
suddenly | +0.020 |
opened | +0.019 |
pretended | +0.018 |
two | +0.018 |
a | +0.017 |
Neg logit contributions | |
beh | -0.023 |
unatt | -0.022 |
ash | -0.022 |
forward | -0.021 |
rav | -0.021 |
Token | ot |
Token ID | 30972 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
beh | +0.205 |
ash | +0.196 |
unatt | +0.195 |
forward | +0.188 |
opl | +0.186 |
Neg logit contributions | |
suddenly | -0.095 |
two | -0.086 |
opened | -0.086 |
pretended | -0.085 |
a | -0.085 |
Token | ters |
Token ID | 1010 |
Feature activation | -3.06e+00 |
Pos logit contributions | |
two | +0.058 |
suddenly | +0.058 |
opened | +0.054 |
pretended | +0.053 |
a | +0.052 |
Neg logit contributions | |
unatt | -0.069 |
opl | -0.065 |
beh | -0.064 |
forward | -0.064 |
mir | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
a | +0.048 |
suddenly | +0.045 |
two | +0.045 |
was | +0.042 |
opened | +0.042 |
Neg logit contributions | |
beh | -0.151 |
ash | -0.142 |
rav | -0.141 |
ocol | -0.139 |
unatt | -0.139 |
Token | look |
Token ID | 804 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
beh | +0.049 |
unatt | +0.047 |
ash | +0.046 |
forward | +0.046 |
rav | +0.045 |
Neg logit contributions | |
suddenly | -0.034 |
opened | -0.033 |
pretended | -0.032 |
two | -0.031 |
Suddenly | -0.030 |
Token | for |
Token ID | 329 |
Feature activation | +4.49e+00 |
Pos logit contributions | |
suddenly | +0.049 |
opened | +0.048 |
pretended | +0.047 |
a | +0.046 |
two | +0.046 |
Neg logit contributions | |
beh | -0.054 |
rav | -0.051 |
unatt | -0.050 |
ash | -0.049 |
opl | -0.048 |
Token | sp |
Token ID | 599 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
beh | +0.144 |
unatt | +0.141 |
ash | +0.138 |
forward | +0.137 |
baby | +0.133 |
Neg logit contributions | |
suddenly | -0.127 |
opened | -0.121 |
pretended | -0.118 |
Suddenly | -0.116 |
two | -0.113 |
Token | oons |
Token ID | 13022 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
beh | +0.007 |
unatt | +0.007 |
rav | +0.007 |
forward | +0.007 |
baby | +0.006 |
Neg logit contributions | |
suddenly | -0.006 |
opened | -0.006 |
pretended | -0.005 |
two | -0.005 |
a | -0.005 |
Token | in |
Token ID | 287 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
beh | +0.069 |
rav | +0.065 |
opl | +0.063 |
aring | +0.063 |
ife | +0.062 |
Neg logit contributions | |
a | -0.024 |
suddenly | -0.024 |
two | -0.023 |
pretended | -0.023 |
opened | -0.022 |
Token | every |
Token ID | 790 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
suddenly | +0.034 |
opened | +0.033 |
pretended | +0.033 |
Suddenly | +0.032 |
clicked | +0.029 |
Neg logit contributions | |
unatt | -0.061 |
forward | -0.060 |
beh | -0.059 |
overs | -0.058 |
baby | -0.058 |
Token | corner |
Token ID | 5228 |
Feature activation | -8.65e-01 |
Pos logit contributions | |
suddenly | +0.052 |
opened | +0.051 |
two | +0.049 |
pretended | +0.048 |
a | +0.047 |
Neg logit contributions | |
beh | -0.064 |
aring | -0.060 |
rav | -0.059 |
unatt | -0.059 |
agging | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
suddenly | +0.019 |
opened | +0.018 |
a | +0.017 |
pretended | +0.017 |
two | +0.017 |
Neg logit contributions | |
beh | -0.036 |
rav | -0.035 |
losers | -0.034 |
unatt | -0.034 |
agging | -0.034 |
Token | One |
Token ID | 1881 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
beh | +0.057 |
unatt | +0.056 |
ash | +0.055 |
forward | +0.055 |
rav | +0.054 |
Neg logit contributions | |
suddenly | -0.029 |
opened | -0.027 |
pretended | -0.026 |
two | -0.025 |
Suddenly | -0.024 |
Token | day |
Token ID | 1110 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
suddenly | +0.048 |
opened | +0.046 |
pretended | +0.045 |
two | +0.044 |
Suddenly | +0.043 |
Neg logit contributions | |
beh | -0.042 |
unatt | -0.040 |
ash | -0.039 |
forward | -0.039 |
rav | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
pretended | +0.047 |
opened | +0.047 |
inserted | +0.045 |
clicked | +0.045 |
suddenly | +0.045 |
Neg logit contributions | |
forward | -0.065 |
baby | -0.060 |
rav | -0.060 |
overs | -0.059 |
ocol | -0.058 |
Token | started |
Token ID | 2067 |
Feature activation | -5.12e-01 |
Pos logit contributions | |
beh | +0.025 |
unatt | +0.024 |
ash | +0.024 |
rav | +0.024 |
forward | +0.024 |
Neg logit contributions | |
suddenly | -0.010 |
opened | -0.010 |
pretended | -0.009 |
two | -0.009 |
Suddenly | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
suddenly | +0.015 |
opened | +0.014 |
a | +0.014 |
two | +0.013 |
pretended | +0.013 |
Neg logit contributions | |
ash | -0.018 |
rav | -0.018 |
unatt | -0.018 |
beh | -0.018 |
ife | -0.018 |
Token | run |
Token ID | 1057 |
Feature activation | +4.44e+00 |
Pos logit contributions | |
beh | +0.057 |
unatt | +0.055 |
forward | +0.053 |
ash | +0.052 |
rav | +0.051 |
Neg logit contributions | |
suddenly | -0.062 |
opened | -0.059 |
pretended | -0.058 |
two | -0.057 |
Suddenly | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | +7.09e-01 |
Pos logit contributions | |
beh | +0.126 |
unatt | +0.119 |
ash | +0.118 |
rav | +0.116 |
forward | +0.111 |
Neg logit contributions | |
suddenly | -0.152 |
opened | -0.145 |
pretended | -0.143 |
two | -0.139 |
Suddenly | -0.138 |
Token |
|
Token ID | 198 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
beh | +0.030 |
unatt | +0.029 |
ash | +0.028 |
forward | +0.028 |
rav | +0.028 |
Neg logit contributions | |
suddenly | -0.014 |
opened | -0.013 |
pretended | -0.013 |
two | -0.013 |
Suddenly | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
pretended | +0.007 |
opened | +0.006 |
suddenly | +0.006 |
two | +0.005 |
clicked | +0.005 |
Neg logit contributions | |
forward | -0.008 |
aring | -0.008 |
off | -0.008 |
ash | -0.008 |
baby | -0.008 |
Token | But |
Token ID | 1537 |
Feature activation | -5.01e+00 |
Pos logit contributions | |
suddenly | +0.005 |
opened | +0.004 |
two | +0.004 |
pretended | +0.004 |
Suddenly | +0.004 |
Neg logit contributions | |
beh | -0.012 |
unatt | -0.011 |
ash | -0.011 |
forward | -0.011 |
rav | -0.011 |
Token | it |
Token ID | 340 |
Feature activation | +3.36e+00 |
Pos logit contributions | |
pretended | +0.074 |
opened | +0.072 |
suddenly | +0.064 |
clicked | +0.063 |
Under | +0.061 |
Neg logit contributions | |
forward | -0.218 |
unatt | -0.215 |
beh | -0.204 |
losers | -0.196 |
overs | -0.195 |
Token | was |
Token ID | 373 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
beh | +0.149 |
unatt | +0.147 |
forward | +0.146 |
ash | +0.142 |
losers | +0.142 |
Neg logit contributions | |
Suddenly | -0.039 |
pretended | -0.038 |
suddenly | -0.038 |
two | -0.037 |
opened | -0.037 |
Token | too |
Token ID | 1165 |
Feature activation | +8.64e+00 |
Pos logit contributions | |
suddenly | +0.066 |
opened | +0.063 |
two | +0.061 |
a | +0.061 |
pretended | +0.060 |
Neg logit contributions | |
beh | -0.062 |
rav | -0.059 |
ash | -0.059 |
unatt | -0.059 |
aring | -0.057 |
Token | late |
Token ID | 2739 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
beh | +0.245 |
unatt | +0.235 |
ash | +0.229 |
forward | +0.226 |
rav | +0.224 |
Neg logit contributions | |
suddenly | -0.291 |
opened | -0.281 |
pretended | -0.273 |
two | -0.271 |
Suddenly | -0.263 |
Token | mom |
Token ID | 1995 |
Feature activation | +3.81e+00 |
Pos logit contributions | |
beh | +0.079 |
ash | +0.075 |
ife | +0.071 |
aring | +0.071 |
rav | +0.070 |
Neg logit contributions | |
suddenly | -0.044 |
two | -0.042 |
opened | -0.041 |
a | -0.041 |
pretended | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | +6.06e-01 |
Pos logit contributions | |
beh | +0.152 |
rav | +0.148 |
unatt | +0.144 |
losers | +0.144 |
ash | +0.142 |
Neg logit contributions | |
suddenly | -0.087 |
opened | -0.083 |
pretended | -0.079 |
two | -0.078 |
a | -0.075 |
Token | Then |
Token ID | 3244 |
Feature activation | -5.91e+00 |
Pos logit contributions | |
beh | +0.025 |
unatt | +0.025 |
ash | +0.024 |
forward | +0.024 |
rav | +0.023 |
Neg logit contributions | |
suddenly | -0.012 |
opened | -0.012 |
pretended | -0.011 |
two | -0.011 |
Suddenly | -0.011 |
Token | he |
Token ID | 339 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
opened | +0.119 |
pretended | +0.118 |
clicked | +0.111 |
suddenly | +0.109 |
pressed | +0.104 |
Neg logit contributions | |
forward | -0.231 |
unatt | -0.210 |
beh | -0.206 |
baby | -0.205 |
overs | -0.205 |
Token | saw |
Token ID | 2497 |
Feature activation | -6.00e+00 |
Pos logit contributions | |
suddenly | +0.018 |
Suddenly | +0.017 |
opened | +0.017 |
pretended | +0.016 |
two | +0.016 |
Neg logit contributions | |
forward | -0.045 |
unatt | -0.045 |
beh | -0.044 |
mir | -0.044 |
ash | -0.043 |
Token | a |
Token ID | 257 |
Feature activation | -7.94e+00 |
Pos logit contributions | |
pretended | +0.121 |
opened | +0.120 |
suddenly | +0.114 |
clicked | +0.107 |
Suddenly | +0.105 |
Neg logit contributions | |
unatt | -0.233 |
forward | -0.226 |
permission | -0.216 |
overs | -0.216 |
beh | -0.214 |
Token | huge |
Token ID | 3236 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
suddenly | +0.176 |
opened | +0.165 |
pretended | +0.157 |
two | +0.157 |
Suddenly | +0.148 |
Neg logit contributions | |
beh | -0.325 |
unatt | -0.309 |
ash | -0.307 |
forward | -0.301 |
rav | -0.295 |
Token | neck |
Token ID | 7393 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
suddenly | +0.053 |
opened | +0.050 |
two | +0.048 |
pretended | +0.047 |
a | +0.047 |
Neg logit contributions | |
beh | -0.099 |
unatt | -0.090 |
ash | -0.090 |
rav | -0.088 |
forward | -0.087 |
Token | - |
Token ID | 12 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
suddenly | +0.069 |
pretended | +0.069 |
opened | +0.066 |
Suddenly | +0.064 |
two | +0.062 |
Neg logit contributions | |
unatt | -0.097 |
forward | -0.097 |
beh | -0.096 |
aring | -0.093 |
ash | -0.093 |
Token | high |
Token ID | 8929 |
Feature activation | -3.23e+00 |
Pos logit contributions | |
beh | +0.351 |
off | +0.316 |
forward | +0.309 |
ash | +0.305 |
aring | +0.297 |
Neg logit contributions | |
suddenly | -0.497 |
pretended | -0.483 |
opened | -0.468 |
led | -0.448 |
two | -0.439 |
Token | patch |
Token ID | 8529 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
suddenly | +0.080 |
pretended | +0.078 |
opened | +0.076 |
Suddenly | +0.073 |
two | +0.072 |
Neg logit contributions | |
unatt | -0.111 |
beh | -0.107 |
forward | -0.107 |
ash | -0.105 |
mir | -0.105 |
Token | She |
Token ID | 1375 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
beh | +0.036 |
unatt | +0.036 |
ash | +0.034 |
forward | +0.034 |
rav | +0.033 |
Neg logit contributions | |
suddenly | -0.030 |
opened | -0.029 |
pretended | -0.028 |
two | -0.028 |
Suddenly | -0.028 |
Token | gave |
Token ID | 2921 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
suddenly | +0.042 |
Suddenly | +0.037 |
opened | +0.037 |
pretended | +0.037 |
two | +0.036 |
Neg logit contributions | |
beh | -0.081 |
unatt | -0.077 |
forward | -0.076 |
ash | -0.076 |
rav | -0.076 |
Token | it |
Token ID | 340 |
Feature activation | +3.60e+00 |
Pos logit contributions | |
beh | +0.014 |
unatt | +0.014 |
ash | +0.013 |
forward | +0.013 |
aring | +0.013 |
Neg logit contributions | |
suddenly | -0.012 |
opened | -0.011 |
two | -0.011 |
pretended | -0.011 |
a | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
beh | +0.148 |
opl | +0.144 |
unatt | +0.143 |
agging | +0.142 |
ash | +0.140 |
Neg logit contributions | |
a | -0.089 |
two | -0.081 |
opened | -0.079 |
suddenly | -0.073 |
pretended | -0.070 |
Token | me |
Token ID | 502 |
Feature activation | +8.57e+00 |
Pos logit contributions | |
beh | +0.076 |
aring | +0.069 |
ash | +0.068 |
rav | +0.066 |
unatt | +0.066 |
Neg logit contributions | |
a | -0.069 |
two | -0.065 |
suddenly | -0.065 |
opened | -0.062 |
pretended | -0.058 |
Token | when |
Token ID | 618 |
Feature activation | -9.91e+00 |
Pos logit contributions | |
beh | +0.371 |
unatt | +0.356 |
rav | +0.347 |
ash | +0.344 |
agging | +0.340 |
Neg logit contributions | |
suddenly | -0.171 |
a | -0.163 |
opened | -0.162 |
two | -0.161 |
pretended | -0.152 |
Token | I |
Token ID | 314 |
Feature activation | +4.49e+00 |
Pos logit contributions | |
suddenly | +0.259 |
two | +0.239 |
opened | +0.235 |
a | +0.235 |
pretended | +0.212 |
Neg logit contributions | |
beh | -0.391 |
ash | -0.377 |
unatt | -0.373 |
aring | -0.366 |
agging | -0.360 |
Token | married |
Token ID | 6405 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
beh | +0.183 |
unatt | +0.178 |
ash | +0.173 |
forward | +0.170 |
rav | +0.169 |
Neg logit contributions | |
suddenly | -0.084 |
Suddenly | -0.079 |
two | -0.074 |
pretended | -0.072 |
opened | -0.069 |
Token | your |
Token ID | 534 |
Feature activation | +5.65e+00 |
Pos logit contributions | |
beh | +0.080 |
aring | +0.072 |
agging | +0.072 |
ash | +0.071 |
rav | +0.070 |
Neg logit contributions | |
suddenly | -0.042 |
a | -0.041 |
opened | -0.040 |
two | -0.040 |
pretended | -0.039 |
Token | dad |
Token ID | 9955 |
Feature activation | +2.95e+00 |
Pos logit contributions | |
beh | +0.215 |
ash | +0.204 |
unatt | +0.195 |
forward | +0.195 |
aring | +0.192 |
Neg logit contributions | |
suddenly | -0.144 |
two | -0.142 |
opened | -0.141 |
pretended | -0.139 |
a | -0.136 |
Token | . |
Token ID | 13 |
Feature activation | +9.97e-01 |
Pos logit contributions | |
beh | +0.116 |
unatt | +0.112 |
forward | +0.109 |
ash | +0.108 |
rav | +0.108 |
Neg logit contributions | |
opened | -0.068 |
suddenly | -0.067 |
pretended | -0.066 |
two | -0.064 |
a | -0.062 |
Token | park |
Token ID | 3952 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
beh | +0.003 |
unatt | +0.003 |
ash | +0.002 |
forward | +0.002 |
rav | +0.002 |
Neg logit contributions | |
suddenly | -0.002 |
opened | -0.002 |
two | -0.002 |
pretended | -0.002 |
Suddenly | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
pretended | +0.041 |
suddenly | +0.039 |
opened | +0.037 |
two | +0.036 |
clicked | +0.036 |
Neg logit contributions | |
forward | -0.072 |
unatt | -0.071 |
rav | -0.069 |
aring | -0.068 |
overs | -0.068 |
Token | the |
Token ID | 262 |
Feature activation | +6.23e-02 |
Pos logit contributions | |
pretended | +0.043 |
opened | +0.040 |
inserted | +0.040 |
dial | +0.039 |
imagined | +0.039 |
Neg logit contributions | |
forward | -0.054 |
unatt | -0.054 |
hurricane | -0.050 |
losers | -0.049 |
overs | -0.049 |
Token | mouse |
Token ID | 10211 |
Feature activation | -5.55e+00 |
Pos logit contributions | |
unatt | +0.002 |
baby | +0.002 |
law | +0.002 |
losers | +0.002 |
ambulance | +0.002 |
Neg logit contributions | |
suddenly | -0.002 |
opened | -0.002 |
pretended | -0.002 |
clicked | -0.002 |
inserted | -0.002 |
Token | saw |
Token ID | 2497 |
Feature activation | -6.20e+00 |
Pos logit contributions | |
suddenly | +0.087 |
two | +0.083 |
Suddenly | +0.078 |
opened | +0.075 |
Under | +0.073 |
Neg logit contributions | |
unatt | -0.227 |
forward | -0.210 |
mir | -0.210 |
beh | -0.210 |
off | -0.209 |
Token | a |
Token ID | 257 |
Feature activation | -8.06e+00 |
Pos logit contributions | |
pretended | +0.079 |
suddenly | +0.075 |
opened | +0.071 |
Suddenly | +0.059 |
clicked | +0.058 |
Neg logit contributions | |
unatt | -0.287 |
beh | -0.276 |
losers | -0.273 |
forward | -0.272 |
ambulance | -0.271 |
Token | big |
Token ID | 1263 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
suddenly | +0.171 |
opened | +0.161 |
pretended | +0.155 |
two | +0.152 |
Suddenly | +0.145 |
Neg logit contributions | |
beh | -0.328 |
unatt | -0.317 |
ash | -0.312 |
forward | -0.310 |
rav | -0.304 |
Token | ball |
Token ID | 2613 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
opened | +0.036 |
suddenly | +0.036 |
two | +0.035 |
a | +0.034 |
pretended | +0.034 |
Neg logit contributions | |
beh | -0.066 |
ash | -0.059 |
unatt | -0.058 |
rav | -0.057 |
forward | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | +5.18e-01 |
Pos logit contributions | |
suddenly | +0.003 |
opened | +0.003 |
two | +0.003 |
pretended | +0.003 |
a | +0.002 |
Neg logit contributions | |
beh | -0.005 |
rav | -0.005 |
unatt | -0.005 |
ash | -0.004 |
forward | -0.004 |
Token | The |
Token ID | 383 |
Feature activation | -3.35e+00 |
Pos logit contributions | |
beh | +0.021 |
unatt | +0.020 |
ash | +0.020 |
forward | +0.019 |
rav | +0.019 |
Neg logit contributions | |
suddenly | -0.012 |
opened | -0.011 |
pretended | -0.011 |
two | -0.010 |
Suddenly | -0.010 |
Token | mouse |
Token ID | 10211 |
Feature activation | -5.63e+00 |
Pos logit contributions | |
suddenly | +0.104 |
opened | +0.100 |
pretended | +0.099 |
Suddenly | +0.094 |
clicked | +0.089 |
Neg logit contributions | |
unatt | -0.097 |
losers | -0.093 |
forward | -0.093 |
baby | -0.092 |
beh | -0.091 |
Token | a |
Token ID | 257 |
Feature activation | -8.08e+00 |
Pos logit contributions | |
clicked | +0.156 |
inserted | +0.149 |
blasted | +0.147 |
pretended | +0.142 |
suddenly | +0.140 |
Neg logit contributions | |
unatt | -0.226 |
losers | -0.222 |
baby | -0.216 |
future | -0.214 |
des | -0.210 |
Token | time |
Token ID | 640 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
suddenly | +0.219 |
opened | +0.206 |
pretended | +0.201 |
clicked | +0.195 |
two | +0.192 |
Neg logit contributions | |
forward | -0.262 |
ash | -0.261 |
unatt | -0.261 |
agging | -0.255 |
beh | -0.253 |
Token | , |
Token ID | 11 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
pretended | +0.047 |
suddenly | +0.046 |
opened | +0.043 |
clicked | +0.041 |
dial | +0.040 |
Neg logit contributions | |
rav | -0.073 |
forward | -0.072 |
unatt | -0.071 |
aring | -0.070 |
baby | -0.069 |
Token | there |
Token ID | 612 |
Feature activation | -7.84e+00 |
Pos logit contributions | |
growth | +0.042 |
dial | +0.040 |
pretended | +0.039 |
trek | +0.038 |
blasted | +0.037 |
Neg logit contributions | |
forward | -0.041 |
back | -0.034 |
unhappy | -0.033 |
baby | -0.033 |
unatt | -0.032 |
Token | was |
Token ID | 373 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
pretended | +0.168 |
dial | +0.163 |
inserted | +0.161 |
blasted | +0.145 |
pressed | +0.144 |
Neg logit contributions | |
baby | -0.271 |
rav | -0.254 |
unatt | -0.247 |
future | -0.246 |
ash | -0.238 |
Token | a |
Token ID | 257 |
Feature activation | -8.07e+00 |
Pos logit contributions | |
dial | +0.034 |
clicked | +0.034 |
rang | +0.033 |
suddenly | +0.031 |
opened | +0.031 |
Neg logit contributions | |
forward | -0.057 |
ags | -0.057 |
overs | -0.056 |
baby | -0.055 |
future | -0.055 |
Token | boy |
Token ID | 2933 |
Feature activation | -5.13e+00 |
Pos logit contributions | |
suddenly | +0.116 |
opened | +0.106 |
pretended | +0.101 |
two | +0.096 |
Suddenly | +0.091 |
Neg logit contributions | |
beh | -0.381 |
unatt | -0.372 |
ash | -0.365 |
forward | -0.365 |
rav | -0.359 |
Token | named |
Token ID | 3706 |
Feature activation | -4.37e+00 |
Pos logit contributions | |
suddenly | +0.115 |
opened | +0.109 |
pretended | +0.105 |
two | +0.104 |
Suddenly | +0.102 |
Neg logit contributions | |
beh | -0.194 |
unatt | -0.190 |
baby | -0.188 |
ash | -0.184 |
aring | -0.184 |
Token | Tim |
Token ID | 5045 |
Feature activation | +5.67e-01 |
Pos logit contributions | |
suddenly | +0.134 |
pretended | +0.120 |
opened | +0.120 |
clicked | +0.113 |
two | +0.113 |
Neg logit contributions | |
ash | -0.131 |
baby | -0.130 |
beh | -0.128 |
unatt | -0.128 |
losers | -0.127 |
Token | my |
Token ID | 1820 |
Feature activation | -2.70e+00 |
Pos logit contributions | |
baby | +0.015 |
unatt | +0.014 |
ash | +0.014 |
beh | +0.014 |
off | +0.014 |
Neg logit contributions | |
suddenly | -0.020 |
Suddenly | -0.018 |
opened | -0.018 |
clicked | -0.018 |
pretended | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
suddenly | +0.086 |
opened | +0.081 |
pretended | +0.078 |
Suddenly | +0.077 |
clicked | +0.077 |
Neg logit contributions | |
baby | -0.077 |
beh | -0.077 |
unatt | -0.076 |
forward | -0.072 |
aring | -0.072 |
Token | with |
Token ID | 351 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
suddenly | +0.042 |
opened | +0.039 |
pretended | +0.038 |
two | +0.037 |
Suddenly | +0.036 |
Neg logit contributions | |
beh | -0.079 |
unatt | -0.076 |
ash | -0.075 |
forward | -0.074 |
rav | -0.074 |
Token | her |
Token ID | 607 |
Feature activation | +2.77e+00 |
Pos logit contributions | |
suddenly | +0.084 |
pretended | +0.081 |
opened | +0.081 |
Suddenly | +0.078 |
two | +0.075 |
Neg logit contributions | |
beh | -0.084 |
forward | -0.082 |
unatt | -0.082 |
ash | -0.079 |
baby | -0.076 |
Token | mom |
Token ID | 1995 |
Feature activation | +3.75e+00 |
Pos logit contributions | |
beh | +0.094 |
unatt | +0.093 |
forward | +0.091 |
ash | +0.090 |
rav | +0.088 |
Neg logit contributions | |
suddenly | -0.074 |
opened | -0.071 |
pretended | -0.070 |
Suddenly | -0.067 |
two | -0.066 |
Token | and |
Token ID | 290 |
Feature activation | +7.54e-01 |
Pos logit contributions | |
beh | +0.146 |
unatt | +0.142 |
ash | +0.138 |
forward | +0.138 |
rav | +0.137 |
Neg logit contributions | |
suddenly | -0.086 |
opened | -0.081 |
pretended | -0.079 |
two | -0.077 |
Suddenly | -0.074 |
Token | brought |
Token ID | 3181 |
Feature activation | +8.53e-02 |
Pos logit contributions | |
unatt | +0.021 |
beh | +0.021 |
forward | +0.021 |
rav | +0.020 |
baby | +0.019 |
Neg logit contributions | |
suddenly | -0.023 |
Suddenly | -0.022 |
opened | -0.022 |
pretended | -0.021 |
two | -0.021 |
Token | a |
Token ID | 257 |
Feature activation | -8.25e+00 |
Pos logit contributions | |
unatt | +0.004 |
beh | +0.004 |
aring | +0.004 |
rav | +0.004 |
losers | +0.004 |
Neg logit contributions | |
two | -0.002 |
suddenly | -0.002 |
a | -0.002 |
opened | -0.002 |
pretended | -0.001 |
Token | sp |
Token ID | 599 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
suddenly | +0.209 |
opened | +0.200 |
pretended | +0.194 |
two | +0.190 |
Suddenly | +0.183 |
Neg logit contributions | |
beh | -0.301 |
unatt | -0.290 |
ash | -0.283 |
forward | -0.282 |
rav | -0.276 |
Token | ade |
Token ID | 671 |
Feature activation | -6.57e-01 |
Pos logit contributions | |
opened | +0.015 |
suddenly | +0.015 |
two | +0.015 |
pretended | +0.014 |
a | +0.014 |
Neg logit contributions | |
beh | -0.014 |
unatt | -0.014 |
rav | -0.013 |
forward | -0.013 |
baby | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
a | +0.013 |
suddenly | +0.013 |
two | +0.013 |
opened | +0.013 |
pretended | +0.012 |
Neg logit contributions | |
rav | -0.028 |
beh | -0.028 |
losers | -0.026 |
baby | -0.026 |
apologies | -0.025 |
Token | As |
Token ID | 1081 |
Feature activation | -6.81e+00 |
Pos logit contributions | |
forward | +0.038 |
overs | +0.037 |
ash | +0.037 |
beh | +0.037 |
rav | +0.036 |
Neg logit contributions | |
suddenly | -0.030 |
pretended | -0.029 |
opened | -0.028 |
two | -0.027 |
clicked | -0.025 |
Token | they |
Token ID | 484 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
pretended | +0.169 |
opened | +0.168 |
suddenly | +0.161 |
Suddenly | +0.159 |
two | +0.155 |
Neg logit contributions | |
forward | -0.228 |
beh | -0.220 |
unatt | -0.218 |
rav | -0.210 |
overs | -0.207 |