Token | when |
Token ID | 618 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
OM | +0.076 |
happened | +0.070 |
emon | +0.063 |
ained | +0.062 |
OO | +0.061 |
Neg logit contributions | |
grate | -0.104 |
rily | -0.086 |
active | -0.085 |
positive | -0.082 |
tightly | -0.081 |
Token | she |
Token ID | 673 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
happened | +0.051 |
end | +0.046 |
OM | +0.044 |
ABC | +0.041 |
been | +0.040 |
Neg logit contributions | |
grate | -0.060 |
rily | -0.053 |
eyed | -0.052 |
away | -0.052 |
Follow | -0.047 |
Token | got |
Token ID | 1392 |
Feature activation | +2.35e-01 |
Pos logit contributions | |
grate | +0.053 |
rily | +0.047 |
away | +0.046 |
eyed | +0.045 |
freely | +0.043 |
Neg logit contributions | |
happened | -0.028 |
OM | -0.026 |
end | -0.026 |
ABC | -0.024 |
consequences | -0.021 |
Token | to |
Token ID | 284 |
Feature activation | -7.29e-02 |
Pos logit contributions | |
happened | +0.291 |
end | +0.256 |
done | +0.238 |
OM | +0.231 |
been | +0.230 |
Neg logit contributions | |
grate | -0.444 |
rily | -0.400 |
away | -0.398 |
oc | -0.393 |
eyed | -0.380 |
Token | the |
Token ID | 262 |
Feature activation | +9.95e-02 |
Pos logit contributions | |
grate | +0.133 |
rily | +0.116 |
away | +0.114 |
eyed | +0.111 |
freely | +0.107 |
Neg logit contributions | |
happened | -0.094 |
OM | -0.089 |
end | -0.087 |
ABC | -0.081 |
consequences | -0.074 |
Token | end |
Token ID | 886 |
Feature activation | +3.24e-01 |
Pos logit contributions | |
happened | +0.107 |
OM | +0.100 |
end | +0.098 |
ABC | +0.090 |
consequences | +0.080 |
Neg logit contributions | |
grate | -0.201 |
rily | -0.178 |
away | -0.175 |
eyed | -0.172 |
freely | -0.166 |
Token | of |
Token ID | 286 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
happened | +0.475 |
end | +0.429 |
OM | +0.415 |
ABC | +0.396 |
time | +0.377 |
Neg logit contributions | |
grate | -0.554 |
rily | -0.490 |
away | -0.470 |
eyed | -0.467 |
freely | -0.434 |
Token | the |
Token ID | 262 |
Feature activation | +9.58e-02 |
Pos logit contributions | |
grate | +0.278 |
rily | +0.241 |
away | +0.237 |
eyed | +0.230 |
freely | +0.225 |
Neg logit contributions | |
happened | -0.192 |
OM | -0.185 |
end | -0.175 |
ABC | -0.167 |
OO | -0.152 |
Token | shortcut |
Token ID | 29401 |
Feature activation | +2.77e-04 |
Pos logit contributions | |
happened | +0.097 |
end | +0.096 |
OM | +0.091 |
ABC | +0.083 |
time | +0.075 |
Neg logit contributions | |
grate | -0.201 |
rily | -0.183 |
eyed | -0.174 |
away | -0.174 |
freely | -0.166 |
Token | , |
Token ID | 11 |
Feature activation | +3.52e-02 |
Pos logit contributions | |
happened | +0.000 |
OM | +0.000 |
end | +0.000 |
ABC | +0.000 |
been | +0.000 |
Neg logit contributions | |
grate | -0.001 |
rily | -0.000 |
away | -0.000 |
eyed | -0.000 |
active | -0.000 |
Token | she |
Token ID | 673 |
Feature activation | -4.18e-02 |
Pos logit contributions | |
happened | +0.042 |
OM | +0.038 |
end | +0.037 |
ABC | +0.034 |
been | +0.033 |
Neg logit contributions | |
grate | -0.068 |
eyed | -0.060 |
rily | -0.060 |
away | -0.059 |
Follow | -0.055 |
Token | a |
Token ID | 257 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
happened | +0.180 |
end | +0.153 |
ABC | +0.148 |
been | +0.145 |
OM | +0.144 |
Neg logit contributions | |
grate | -0.271 |
rily | -0.251 |
away | -0.251 |
Follow | -0.232 |
eyed | -0.231 |
Token | big |
Token ID | 1263 |
Feature activation | -9.36e-02 |
Pos logit contributions | |
happened | +0.185 |
OM | +0.176 |
end | +0.168 |
ABC | +0.158 |
consequences | +0.143 |
Neg logit contributions | |
grate | -0.297 |
rily | -0.259 |
away | -0.257 |
eyed | -0.250 |
freely | -0.242 |
Token | surprise |
Token ID | 5975 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
grate | +0.180 |
rily | +0.161 |
eyed | +0.156 |
away | +0.155 |
freely | +0.147 |
Neg logit contributions | |
happened | -0.114 |
end | -0.107 |
OM | -0.102 |
ABC | -0.094 |
time | -0.088 |
Token | at |
Token ID | 379 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
grate | +0.050 |
rily | +0.045 |
away | +0.044 |
eyed | +0.044 |
active | +0.042 |
Neg logit contributions | |
happened | -0.031 |
end | -0.028 |
OM | -0.026 |
been | -0.023 |
ABC | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +9.42e-02 |
Pos logit contributions | |
grate | +0.373 |
rily | +0.326 |
away | +0.321 |
eyed | +0.313 |
freely | +0.301 |
Neg logit contributions | |
happened | -0.258 |
OM | -0.244 |
end | -0.236 |
ABC | -0.222 |
consequences | -0.202 |
Token | end |
Token ID | 886 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
happened | +0.107 |
OM | +0.098 |
ABC | +0.086 |
ained | +0.084 |
OO | +0.081 |
Neg logit contributions | |
grate | -0.181 |
away | -0.161 |
rily | -0.158 |
eyed | -0.151 |
greet | -0.150 |
Token | of |
Token ID | 286 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
happened | +0.424 |
end | +0.370 |
OM | +0.367 |
ABC | +0.343 |
is | +0.340 |
Neg logit contributions | |
grate | -0.585 |
rily | -0.512 |
away | -0.509 |
eyed | -0.498 |
active | -0.471 |
Token | the |
Token ID | 262 |
Feature activation | +9.16e-02 |
Pos logit contributions | |
grate | +0.312 |
away | +0.280 |
rily | +0.278 |
eyed | +0.277 |
lived | +0.270 |
Neg logit contributions | |
happened | -0.172 |
end | -0.163 |
OM | -0.148 |
been | -0.141 |
time | -0.137 |
Token | saf |
Token ID | 1932 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
happened | +0.091 |
OM | +0.084 |
end | +0.083 |
ABC | +0.075 |
consequences | +0.066 |
Neg logit contributions | |
grate | -0.195 |
rily | -0.175 |
away | -0.171 |
eyed | -0.169 |
freely | -0.163 |
Token | ari |
Token ID | 2743 |
Feature activation | +7.19e-03 |
Pos logit contributions | |
happened | +0.141 |
OM | +0.130 |
end | +0.129 |
ABC | +0.120 |
consequences | +0.116 |
Neg logit contributions | |
grate | -0.173 |
away | -0.147 |
rily | -0.146 |
eyed | -0.142 |
freely | -0.136 |
Token | . |
Token ID | 13 |
Feature activation | +2.57e-02 |
Pos logit contributions | |
happened | +0.008 |
OM | +0.007 |
end | +0.007 |
ABC | +0.007 |
been | +0.006 |
Neg logit contributions | |
grate | -0.014 |
rily | -0.013 |
away | -0.013 |
eyed | -0.012 |
freely | -0.012 |
Token | won |
Token ID | 1839 |
Feature activation | +6.24e-02 |
Pos logit contributions | |
grate | +0.067 |
rily | +0.063 |
eyed | +0.060 |
away | +0.058 |
lived | +0.055 |
Neg logit contributions | |
end | -0.039 |
happened | -0.038 |
ABC | -0.035 |
OM | -0.035 |
problem | -0.031 |
Token | a |
Token ID | 257 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
OM | +0.083 |
happened | +0.083 |
ABC | +0.077 |
end | +0.076 |
dentist | +0.072 |
Neg logit contributions | |
grate | -0.103 |
eyed | -0.088 |
away | -0.088 |
rily | -0.086 |
freely | -0.086 |
Token | prize |
Token ID | 11596 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
happened | +0.186 |
end | +0.176 |
OM | +0.174 |
ABC | +0.159 |
time | +0.146 |
Neg logit contributions | |
grate | -0.274 |
rily | -0.245 |
eyed | -0.235 |
away | -0.232 |
freely | -0.218 |
Token | at |
Token ID | 379 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
happened | +0.126 |
end | +0.117 |
OM | +0.114 |
ABC | +0.102 |
been | +0.099 |
Neg logit contributions | |
grate | -0.190 |
rily | -0.168 |
eyed | -0.162 |
away | -0.162 |
active | -0.154 |
Token | the |
Token ID | 262 |
Feature activation | +9.36e-02 |
Pos logit contributions | |
grate | +0.355 |
away | +0.309 |
rily | +0.307 |
eyed | +0.295 |
freely | +0.294 |
Neg logit contributions | |
happened | -0.268 |
OM | -0.255 |
end | -0.233 |
ABC | -0.233 |
OO | -0.214 |
Token | end |
Token ID | 886 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
happened | +0.108 |
OM | +0.093 |
ABC | +0.079 |
ained | +0.078 |
OO | +0.077 |
Neg logit contributions | |
grate | -0.181 |
away | -0.164 |
rily | -0.157 |
freely | -0.154 |
positive | -0.151 |
Token | . |
Token ID | 13 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
happened | +0.439 |
OM | +0.401 |
end | +0.400 |
ABC | +0.376 |
is | +0.359 |
Neg logit contributions | |
grate | -0.556 |
rily | -0.479 |
away | -0.470 |
eyed | -0.460 |
lived | -0.433 |
Token |
|
Token ID | 198 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
happened | +0.017 |
OM | +0.016 |
end | +0.016 |
ABC | +0.015 |
been | +0.014 |
Neg logit contributions | |
grate | -0.027 |
away | -0.025 |
rily | -0.025 |
eyed | -0.023 |
active | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
happened | +0.016 |
OM | +0.015 |
end | +0.014 |
ABC | +0.014 |
consequences | +0.013 |
Neg logit contributions | |
grate | -0.026 |
eyed | -0.022 |
rily | -0.022 |
away | -0.022 |
freely | -0.021 |
Token | After |
Token ID | 3260 |
Feature activation | -4.94e-05 |
Pos logit contributions | |
happened | +0.018 |
OM | +0.018 |
end | +0.017 |
ABC | +0.015 |
been | +0.015 |
Neg logit contributions | |
grate | -0.022 |
rily | -0.020 |
away | -0.018 |
eyed | -0.017 |
freely | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | +8.92e-02 |
Pos logit contributions | |
grate | +0.000 |
rily | +0.000 |
away | +0.000 |
eyed | +0.000 |
freely | +0.000 |
Neg logit contributions | |
happened | -0.000 |
OM | -0.000 |
end | -0.000 |
ABC | -0.000 |
consequences | -0.000 |
Token | until |
Token ID | 1566 |
Feature activation | +8.88e-03 |
Pos logit contributions | |
grate | +0.033 |
rily | +0.029 |
away | +0.027 |
eyed | +0.027 |
oc | +0.026 |
Neg logit contributions | |
happened | -0.020 |
end | -0.020 |
OM | -0.019 |
ABC | -0.017 |
been | -0.017 |
Token | he |
Token ID | 339 |
Feature activation | -6.66e-02 |
Pos logit contributions | |
happened | +0.012 |
OM | +0.011 |
end | +0.011 |
ABC | +0.010 |
been | +0.009 |
Neg logit contributions | |
grate | -0.016 |
rily | -0.014 |
away | -0.014 |
eyed | -0.014 |
active | -0.013 |
Token | got |
Token ID | 1392 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
grate | +0.144 |
rily | +0.129 |
away | +0.127 |
eyed | +0.125 |
freely | +0.121 |
Neg logit contributions | |
happened | -0.062 |
OM | -0.058 |
end | -0.055 |
ABC | -0.051 |
consequences | -0.044 |
Token | to |
Token ID | 284 |
Feature activation | -8.17e-02 |
Pos logit contributions | |
happened | +0.288 |
end | +0.255 |
OM | +0.248 |
done | +0.236 |
been | +0.235 |
Neg logit contributions | |
grate | -0.417 |
rily | -0.374 |
away | -0.373 |
oc | -0.361 |
eyed | -0.353 |
Token | the |
Token ID | 262 |
Feature activation | +9.33e-02 |
Pos logit contributions | |
grate | +0.146 |
rily | +0.127 |
away | +0.125 |
eyed | +0.123 |
freely | +0.117 |
Neg logit contributions | |
happened | -0.108 |
OM | -0.102 |
end | -0.101 |
ABC | -0.094 |
been | -0.086 |
Token | end |
Token ID | 886 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
happened | +0.104 |
OM | +0.095 |
end | +0.087 |
ABC | +0.084 |
OO | +0.076 |
Neg logit contributions | |
grate | -0.187 |
rily | -0.164 |
away | -0.163 |
eyed | -0.158 |
freely | -0.154 |
Token | . |
Token ID | 13 |
Feature activation | +2.57e-02 |
Pos logit contributions | |
happened | +0.418 |
end | +0.357 |
OM | +0.350 |
time | +0.344 |
is | +0.337 |
Neg logit contributions | |
grate | -0.581 |
rily | -0.513 |
away | -0.494 |
eyed | -0.490 |
greet | -0.468 |
Token |
|
Token ID | 198 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
happened | +0.034 |
OM | +0.033 |
end | +0.031 |
ABC | +0.030 |
been | +0.028 |
Neg logit contributions | |
grate | -0.044 |
away | -0.040 |
rily | -0.039 |
eyed | -0.037 |
active | -0.036 |
Token |
|
Token ID | 198 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
happened | +0.027 |
OM | +0.026 |
end | +0.025 |
ABC | +0.023 |
consequences | +0.021 |
Neg logit contributions | |
grate | -0.046 |
rily | -0.040 |
away | -0.039 |
eyed | -0.039 |
freely | -0.038 |
Token | The |
Token ID | 464 |
Feature activation | -4.70e-02 |
Pos logit contributions | |
happened | +0.030 |
OM | +0.030 |
end | +0.030 |
ABC | +0.026 |
been | +0.026 |
Neg logit contributions | |
grate | -0.037 |
rily | -0.034 |
away | -0.032 |
freely | -0.029 |
eyed | -0.029 |
Token | second |
Token ID | 1218 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
grate | +0.096 |
rily | +0.083 |
away | +0.083 |
eyed | +0.081 |
freely | +0.079 |
Neg logit contributions | |
happened | -0.050 |
OM | -0.047 |
end | -0.041 |
ABC | -0.040 |
OO | -0.037 |
Token | can |
Token ID | 460 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
happened | +0.014 |
OM | +0.013 |
end | +0.012 |
ABC | +0.011 |
consequences | +0.010 |
Neg logit contributions | |
grate | -0.031 |
rily | -0.028 |
away | -0.027 |
eyed | -0.027 |
freely | -0.026 |
Token | make |
Token ID | 787 |
Feature activation | -4.70e-02 |
Pos logit contributions | |
end | +0.052 |
happened | +0.050 |
OM | +0.046 |
ABC | +0.044 |
been | +0.043 |
Neg logit contributions | |
grate | -0.072 |
rily | -0.068 |
away | -0.066 |
eyed | -0.063 |
mer | -0.061 |
Token | it |
Token ID | 340 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
grate | +0.087 |
rily | +0.076 |
away | +0.075 |
eyed | +0.073 |
freely | +0.070 |
Neg logit contributions | |
happened | -0.059 |
OM | -0.056 |
end | -0.054 |
ABC | -0.050 |
consequences | -0.045 |
Token | to |
Token ID | 284 |
Feature activation | -8.71e-02 |
Pos logit contributions | |
happened | +0.163 |
end | +0.154 |
OM | +0.149 |
ABC | +0.139 |
done | +0.131 |
Neg logit contributions | |
grate | -0.220 |
rily | -0.194 |
away | -0.189 |
eyed | -0.188 |
Follow | -0.177 |
Token | the |
Token ID | 262 |
Feature activation | +9.36e-02 |
Pos logit contributions | |
grate | +0.155 |
rily | +0.144 |
eyed | +0.135 |
Follow | +0.134 |
away | +0.133 |
Neg logit contributions | |
end | -0.116 |
happened | -0.113 |
OM | -0.102 |
ABC | -0.096 |
been | -0.095 |
Token | end |
Token ID | 886 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
happened | +0.110 |
OM | +0.103 |
end | +0.094 |
ABC | +0.091 |
OO | +0.082 |
Neg logit contributions | |
grate | -0.180 |
rily | -0.157 |
away | -0.156 |
eyed | -0.151 |
freely | -0.147 |
Token | . |
Token ID | 13 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
happened | +0.389 |
end | +0.343 |
OM | +0.335 |
is | +0.322 |
ABC | +0.309 |
Neg logit contributions | |
grate | -0.618 |
rily | -0.547 |
eyed | -0.518 |
away | -0.517 |
greet | -0.501 |
Token | Can |
Token ID | 1680 |
Feature activation | -9.53e-02 |
Pos logit contributions | |
happened | +0.023 |
OM | +0.023 |
end | +0.022 |
ABC | +0.021 |
is | +0.020 |
Neg logit contributions | |
grate | -0.042 |
away | -0.042 |
rily | -0.040 |
eyed | -0.037 |
ized | -0.036 |
Token | we |
Token ID | 356 |
Feature activation | -7.46e-02 |
Pos logit contributions | |
grate | +0.192 |
rily | +0.190 |
away | +0.180 |
eyed | +0.176 |
mer | +0.169 |
Neg logit contributions | |
end | -0.107 |
happened | -0.098 |
OM | -0.096 |
is | -0.093 |
ABC | -0.085 |
Token | rest |
Token ID | 1334 |
Feature activation | +9.05e-02 |
Pos logit contributions | |
rily | +0.177 |
mer | +0.151 |
oc | +0.145 |
away | +0.145 |
oing | +0.144 |
Neg logit contributions | |
end | -0.097 |
time | -0.063 |
was | -0.063 |
is | -0.062 |
OM | -0.062 |
Token | a |
Token ID | 257 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
happened | +0.101 |
end | +0.096 |
OM | +0.088 |
done | +0.081 |
time | +0.080 |
Neg logit contributions | |
grate | -0.183 |
rily | -0.163 |
eyed | -0.158 |
away | -0.154 |
lived | -0.151 |
Token | a |
Token ID | 257 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
happened | +0.068 |
end | +0.064 |
OM | +0.060 |
ABC | +0.058 |
consequences | +0.050 |
Neg logit contributions | |
grate | -0.146 |
rily | -0.133 |
away | -0.132 |
eyed | -0.129 |
freely | -0.125 |
Token | large |
Token ID | 1588 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
happened | +0.185 |
OM | +0.176 |
end | +0.168 |
ABC | +0.158 |
consequences | +0.143 |
Neg logit contributions | |
grate | -0.297 |
rily | -0.259 |
away | -0.257 |
eyed | -0.250 |
freely | -0.242 |
Token | door |
Token ID | 3420 |
Feature activation | +6.97e-02 |
Pos logit contributions | |
grate | +0.208 |
rily | +0.187 |
eyed | +0.182 |
away | +0.179 |
freely | +0.171 |
Neg logit contributions | |
happened | -0.128 |
end | -0.121 |
OM | -0.115 |
ABC | -0.107 |
time | -0.099 |
Token | at |
Token ID | 379 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
happened | +0.081 |
end | +0.071 |
OM | +0.066 |
been | +0.062 |
is | +0.061 |
Neg logit contributions | |
grate | -0.142 |
rily | -0.128 |
away | -0.123 |
active | -0.122 |
eyed | -0.120 |
Token | the |
Token ID | 262 |
Feature activation | +9.14e-02 |
Pos logit contributions | |
grate | +0.340 |
away | +0.298 |
rily | +0.296 |
eyed | +0.283 |
freely | +0.278 |
Neg logit contributions | |
happened | -0.292 |
OM | -0.278 |
end | -0.256 |
ABC | -0.254 |
OO | -0.241 |
Token | end |
Token ID | 886 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
happened | +0.111 |
OM | +0.098 |
ABC | +0.087 |
ained | +0.087 |
OO | +0.085 |
Neg logit contributions | |
grate | -0.168 |
away | -0.153 |
rily | -0.147 |
eyed | -0.144 |
greet | -0.142 |
Token | of |
Token ID | 286 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
happened | +0.436 |
end | +0.388 |
OM | +0.387 |
ABC | +0.359 |
time | +0.347 |
Neg logit contributions | |
grate | -0.570 |
rily | -0.492 |
away | -0.484 |
eyed | -0.474 |
active | -0.451 |
Token | the |
Token ID | 262 |
Feature activation | +8.85e-02 |
Pos logit contributions | |
grate | +0.307 |
away | +0.271 |
rily | +0.271 |
eyed | +0.270 |
lived | +0.258 |
Neg logit contributions | |
happened | -0.182 |
end | -0.171 |
OM | -0.165 |
ABC | -0.148 |
been | -0.147 |
Token | porch |
Token ID | 33179 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
happened | +0.090 |
end | +0.084 |
OM | +0.083 |
ABC | +0.075 |
consequences | +0.066 |
Neg logit contributions | |
grate | -0.187 |
rily | -0.168 |
away | -0.163 |
eyed | -0.161 |
freely | -0.155 |
Token | . |
Token ID | 13 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
happened | +0.133 |
OM | +0.125 |
end | +0.122 |
ABC | +0.111 |
been | +0.102 |
Neg logit contributions | |
grate | -0.227 |
rily | -0.202 |
away | -0.201 |
eyed | -0.194 |
freely | -0.187 |
Token | She |
Token ID | 1375 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
happened | +0.030 |
OM | +0.029 |
end | +0.027 |
ABC | +0.026 |
been | +0.025 |
Neg logit contributions | |
grate | -0.042 |
away | -0.039 |
rily | -0.037 |
eyed | -0.036 |
positive | -0.034 |
Token | sky |
Token ID | 6766 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
happened | +0.187 |
end | +0.168 |
been | +0.153 |
best | +0.150 |
time | +0.149 |
Neg logit contributions | |
grate | -0.304 |
away | -0.284 |
rily | -0.271 |
oc | -0.263 |
eyed | -0.262 |
Token | was |
Token ID | 373 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
happened | +0.175 |
end | +0.155 |
OM | +0.155 |
ABC | +0.136 |
been | +0.134 |
Neg logit contributions | |
grate | -0.270 |
rily | -0.237 |
away | -0.234 |
eyed | -0.228 |
active | -0.221 |
Token | never |
Token ID | 1239 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
happened | +0.301 |
OM | +0.289 |
end | +0.277 |
ABC | +0.266 |
consequences | +0.242 |
Neg logit contributions | |
grate | -0.409 |
rily | -0.355 |
away | -0.347 |
eyed | -0.341 |
freely | -0.329 |
Token | going |
Token ID | 1016 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
OM | +0.033 |
happened | +0.030 |
ABC | +0.030 |
omm | +0.029 |
end | +0.029 |
Neg logit contributions | |
grate | -0.036 |
away | -0.032 |
rily | -0.030 |
freely | -0.030 |
eyed | -0.029 |
Token | to |
Token ID | 284 |
Feature activation | -8.61e-02 |
Pos logit contributions | |
grate | +0.050 |
rily | +0.043 |
away | +0.041 |
eyed | +0.041 |
oc | +0.039 |
Neg logit contributions | |
happened | -0.042 |
OM | -0.039 |
end | -0.039 |
ABC | -0.036 |
been | -0.034 |
Token | end |
Token ID | 886 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
grate | +0.148 |
rily | +0.128 |
away | +0.125 |
eyed | +0.122 |
freely | +0.117 |
Neg logit contributions | |
happened | -0.120 |
OM | -0.114 |
end | -0.111 |
ABC | -0.104 |
consequences | -0.096 |
Token | . |
Token ID | 13 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
happened | +0.367 |
end | +0.322 |
OM | +0.316 |
ABC | +0.293 |
time | +0.275 |
Neg logit contributions | |
grate | -0.644 |
rily | -0.569 |
eyed | -0.546 |
away | -0.542 |
Follow | -0.515 |
Token |
|
Token ID | 198 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
happened | +0.021 |
OM | +0.020 |
end | +0.019 |
ABC | +0.018 |
been | +0.017 |
Neg logit contributions | |
grate | -0.027 |
away | -0.025 |
rily | -0.024 |
eyed | -0.023 |
active | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
happened | +0.017 |
OM | +0.016 |
end | +0.015 |
ABC | +0.014 |
consequences | +0.013 |
Neg logit contributions | |
grate | -0.030 |
rily | -0.026 |
away | -0.026 |
eyed | -0.025 |
freely | -0.024 |
Token | The |
Token ID | 464 |
Feature activation | -4.64e-02 |
Pos logit contributions | |
happened | +0.020 |
OM | +0.020 |
end | +0.020 |
ABC | +0.018 |
been | +0.017 |
Neg logit contributions | |
grate | -0.026 |
rily | -0.023 |
away | -0.022 |
eyed | -0.020 |
freely | -0.020 |
Token | girl |
Token ID | 2576 |
Feature activation | +3.43e-02 |
Pos logit contributions | |
grate | +0.093 |
rily | +0.081 |
away | +0.080 |
eyed | +0.079 |
freely | +0.076 |
Neg logit contributions | |
happened | -0.051 |
OM | -0.048 |
end | -0.044 |
ABC | -0.042 |
OO | -0.038 |
Token | time |
Token ID | 640 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
end | +0.008 |
happened | +0.008 |
OM | +0.007 |
bottom | +0.006 |
time | +0.006 |
Neg logit contributions | |
grate | -0.011 |
rily | -0.010 |
Follow | -0.010 |
active | -0.010 |
away | -0.009 |
Token | for |
Token ID | 329 |
Feature activation | -6.00e-02 |
Pos logit contributions | |
happened | +0.226 |
OM | +0.212 |
end | +0.206 |
ABC | +0.192 |
consequences | +0.175 |
Neg logit contributions | |
grate | -0.339 |
rily | -0.297 |
away | -0.292 |
eyed | -0.286 |
freely | -0.274 |
Token | the |
Token ID | 262 |
Feature activation | +9.00e-02 |
Pos logit contributions | |
grate | +0.122 |
rily | +0.109 |
away | +0.107 |
eyed | +0.105 |
freely | +0.101 |
Neg logit contributions | |
happened | -0.065 |
OM | -0.059 |
end | -0.059 |
ABC | -0.053 |
been | -0.049 |
Token | party |
Token ID | 2151 |
Feature activation | +8.12e-02 |
Pos logit contributions | |
happened | +0.104 |
OM | +0.098 |
end | +0.093 |
ABC | +0.088 |
consequences | +0.078 |
Neg logit contributions | |
grate | -0.174 |
rily | -0.152 |
away | -0.151 |
eyed | -0.147 |
freely | -0.142 |
Token | to |
Token ID | 284 |
Feature activation | -8.49e-02 |
Pos logit contributions | |
happened | +0.099 |
OM | +0.094 |
end | +0.091 |
ABC | +0.085 |
consequences | +0.077 |
Neg logit contributions | |
grate | -0.152 |
rily | -0.133 |
away | -0.131 |
eyed | -0.128 |
freely | -0.124 |
Token | end |
Token ID | 886 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
grate | +0.139 |
rily | +0.120 |
away | +0.117 |
eyed | +0.115 |
freely | +0.109 |
Neg logit contributions | |
happened | -0.125 |
end | -0.119 |
OM | -0.118 |
ABC | -0.109 |
been | -0.102 |
Token | . |
Token ID | 13 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
happened | +0.398 |
OM | +0.363 |
end | +0.360 |
ABC | +0.336 |
time | +0.303 |
Neg logit contributions | |
grate | -0.596 |
rily | -0.523 |
away | -0.515 |
eyed | -0.505 |
freely | -0.483 |
Token | Ell |
Token ID | 7122 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
happened | +0.029 |
OM | +0.028 |
end | +0.026 |
ABC | +0.025 |
been | +0.023 |
Neg logit contributions | |
grate | -0.044 |
away | -0.040 |
rily | -0.039 |
eyed | -0.038 |
freely | -0.036 |
Token | a |
Token ID | 64 |
Feature activation | +6.30e-02 |
Pos logit contributions | |
happened | +0.141 |
OM | +0.135 |
end | +0.124 |
ABC | +0.115 |
consequences | +0.109 |
Neg logit contributions | |
grate | -0.270 |
rily | -0.241 |
eyed | -0.237 |
away | -0.236 |
freely | -0.220 |
Token | was |
Token ID | 373 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
happened | +0.056 |
end | +0.050 |
OM | +0.049 |
ABC | +0.044 |
been | +0.039 |
Neg logit contributions | |
grate | -0.139 |
rily | -0.127 |
away | -0.125 |
eyed | -0.122 |
freely | -0.118 |
Token | sad |
Token ID | 6507 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
happened | +0.279 |
OM | +0.271 |
end | +0.251 |
ABC | +0.240 |
consequences | +0.220 |
Neg logit contributions | |
grate | -0.412 |
rily | -0.354 |
eyed | -0.346 |
away | -0.343 |
active | -0.331 |
Token | 't |
Token ID | 470 |
Feature activation | +5.86e-03 |
Pos logit contributions | |
happened | +0.237 |
OM | +0.224 |
end | +0.215 |
ABC | +0.203 |
consequences | +0.198 |
Neg logit contributions | |
grate | -0.302 |
rily | -0.258 |
eyed | -0.255 |
away | -0.252 |
freely | -0.245 |
Token | want |
Token ID | 765 |
Feature activation | -6.42e-02 |
Pos logit contributions | |
end | +0.009 |
happened | +0.008 |
time | +0.007 |
been | +0.007 |
OM | +0.007 |
Neg logit contributions | |
grate | -0.010 |
rily | -0.009 |
away | -0.008 |
Follow | -0.008 |
eyed | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | +8.93e-02 |
Pos logit contributions | |
grate | +0.110 |
rily | +0.096 |
away | +0.093 |
eyed | +0.092 |
freely | +0.086 |
Neg logit contributions | |
happened | -0.092 |
end | -0.085 |
OM | -0.084 |
ABC | -0.079 |
been | -0.074 |
Token | day |
Token ID | 1110 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
happened | +0.098 |
OM | +0.097 |
end | +0.084 |
ABC | +0.082 |
OO | +0.081 |
Neg logit contributions | |
grate | -0.169 |
away | -0.147 |
rily | -0.146 |
freely | -0.141 |
active | -0.140 |
Token | to |
Token ID | 284 |
Feature activation | -8.35e-02 |
Pos logit contributions | |
happened | +0.212 |
OM | +0.195 |
end | +0.194 |
ABC | +0.179 |
been | +0.168 |
Neg logit contributions | |
grate | -0.281 |
rily | -0.245 |
away | -0.238 |
eyed | -0.234 |
freely | -0.223 |
Token | end |
Token ID | 886 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
grate | +0.160 |
rily | +0.148 |
eyed | +0.138 |
away | +0.138 |
Follow | +0.135 |
Neg logit contributions | |
end | -0.104 |
happened | -0.094 |
OM | -0.092 |
ABC | -0.083 |
time | -0.081 |
Token | . |
Token ID | 13 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
happened | +0.338 |
end | +0.280 |
OM | +0.268 |
ABC | +0.254 |
time | +0.253 |
Neg logit contributions | |
grate | -0.684 |
rily | -0.607 |
eyed | -0.587 |
away | -0.581 |
Follow | -0.565 |
Token | But |
Token ID | 887 |
Feature activation | -9.47e-02 |
Pos logit contributions | |
happened | +0.028 |
OM | +0.027 |
end | +0.025 |
ABC | +0.024 |
been | +0.023 |
Neg logit contributions | |
grate | -0.045 |
away | -0.041 |
rily | -0.040 |
eyed | -0.038 |
active | -0.037 |
Token | soon |
Token ID | 2582 |
Feature activation | +4.39e-02 |
Pos logit contributions | |
grate | +0.182 |
eyed | +0.178 |
rily | +0.178 |
away | +0.173 |
lived | +0.162 |
Neg logit contributions | |
end | -0.107 |
happened | -0.104 |
something | -0.091 |
was | -0.090 |
time | -0.088 |
Token | the |
Token ID | 262 |
Feature activation | +8.51e-02 |
Pos logit contributions | |
OM | +0.056 |
happened | +0.055 |
end | +0.052 |
ABC | +0.051 |
OO | +0.046 |
Neg logit contributions | |
grate | -0.079 |
away | -0.067 |
rily | -0.067 |
freely | -0.063 |
eyed | -0.063 |
Token | sun |
Token ID | 4252 |
Feature activation | +5.56e-02 |
Pos logit contributions | |
happened | +0.095 |
OM | +0.091 |
end | +0.081 |
ABC | +0.079 |
OO | +0.076 |
Neg logit contributions | |
grate | -0.166 |
rily | -0.143 |
away | -0.142 |
eyed | -0.138 |
active | -0.136 |
Token | had |
Token ID | 550 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
grate | +0.053 |
rily | +0.048 |
away | +0.044 |
eyed | +0.043 |
active | +0.042 |
Neg logit contributions | |
happened | -0.032 |
OM | -0.031 |
end | -0.030 |
ABC | -0.025 |
been | -0.025 |
Token | a |
Token ID | 257 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
OM | +0.147 |
happened | +0.137 |
end | +0.135 |
ABC | +0.129 |
omm | +0.116 |
Neg logit contributions | |
grate | -0.309 |
away | -0.269 |
rily | -0.267 |
eyed | -0.265 |
active | -0.258 |
Token | star |
Token ID | 3491 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
happened | +0.194 |
OM | +0.181 |
end | +0.178 |
ABC | +0.165 |
consequences | +0.149 |
Neg logit contributions | |
grate | -0.291 |
rily | -0.256 |
away | -0.250 |
eyed | -0.247 |
freely | -0.235 |
Token | on |
Token ID | 319 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
happened | +0.031 |
end | +0.028 |
OM | +0.027 |
ABC | +0.025 |
been | +0.024 |
Neg logit contributions | |
grate | -0.042 |
rily | -0.037 |
away | -0.036 |
eyed | -0.036 |
freely | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | +9.17e-02 |
Pos logit contributions | |
grate | +0.242 |
rily | +0.209 |
eyed | +0.207 |
away | +0.206 |
active | +0.197 |
Neg logit contributions | |
happened | -0.163 |
end | -0.155 |
OM | -0.151 |
ABC | -0.140 |
time | -0.130 |
Token | end |
Token ID | 886 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
happened | +0.103 |
OM | +0.092 |
ABC | +0.077 |
end | +0.077 |
OO | +0.075 |
Neg logit contributions | |
grate | -0.187 |
away | -0.166 |
rily | -0.166 |
eyed | -0.159 |
freely | -0.154 |
Token | . |
Token ID | 13 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
happened | +0.468 |
end | +0.429 |
OM | +0.415 |
ABC | +0.394 |
time | +0.384 |
Neg logit contributions | |
grate | -0.530 |
rily | -0.452 |
away | -0.439 |
eyed | -0.437 |
active | -0.414 |
Token | " |
Token ID | 366 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
happened | +0.024 |
OM | +0.023 |
end | +0.022 |
ABC | +0.021 |
been | +0.020 |
Neg logit contributions | |
grate | -0.033 |
away | -0.030 |
rily | -0.029 |
eyed | -0.028 |
positive | -0.027 |
Token | Maybe |
Token ID | 13300 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
grate | +0.024 |
rily | +0.021 |
away | +0.020 |
eyed | +0.020 |
freely | +0.019 |
Neg logit contributions | |
happened | -0.018 |
OM | -0.018 |
end | -0.017 |
ABC | -0.016 |
been | -0.015 |
Token | it |
Token ID | 340 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
grate | +0.025 |
rily | +0.022 |
eyed | +0.022 |
away | +0.022 |
freely | +0.021 |
Neg logit contributions | |
happened | -0.018 |
end | -0.017 |
OM | -0.016 |
ABC | -0.015 |
been | -0.014 |
Token | can |
Token ID | 460 |
Feature activation | +3.83e-02 |
Pos logit contributions | |
happened | +0.105 |
OM | +0.093 |
end | +0.091 |
ABC | +0.081 |
been | +0.070 |
Neg logit contributions | |
grate | -0.272 |
rily | -0.245 |
away | -0.242 |
eyed | -0.237 |
freely | -0.229 |
Token | |
Token ID | 220 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
end | +0.018 |
happened | +0.018 |
time | +0.016 |
best | +0.015 |
problem | +0.015 |
Neg logit contributions | |
grate | -0.033 |
rily | -0.032 |
ishly | -0.029 |
oc | -0.029 |
greet | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | +3.06e-02 |
Pos logit contributions | |
happened | +0.188 |
OM | +0.182 |
end | +0.177 |
ABC | +0.166 |
consequences | +0.157 |
Neg logit contributions | |
grate | -0.239 |
rily | -0.208 |
eyed | -0.199 |
away | -0.192 |
freely | -0.190 |
Token |
|
Token ID | 198 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
happened | +0.035 |
OM | +0.033 |
end | +0.032 |
ABC | +0.029 |
been | +0.027 |
Neg logit contributions | |
grate | -0.060 |
rily | -0.053 |
away | -0.052 |
eyed | -0.050 |
freely | -0.049 |
Token | In |
Token ID | 818 |
Feature activation | -8.55e-02 |
Pos logit contributions | |
happened | +0.035 |
OM | +0.035 |
end | +0.035 |
ABC | +0.031 |
been | +0.031 |
Neg logit contributions | |
grate | -0.047 |
rily | -0.043 |
away | -0.040 |
eyed | -0.037 |
freely | -0.037 |
Token | the |
Token ID | 262 |
Feature activation | +9.33e-02 |
Pos logit contributions | |
grate | +0.178 |
rily | +0.159 |
away | +0.157 |
eyed | +0.156 |
freely | +0.147 |
Neg logit contributions | |
happened | -0.090 |
end | -0.087 |
OM | -0.080 |
ABC | -0.072 |
time | -0.071 |
Token | end |
Token ID | 886 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
happened | +0.036 |
end | +0.034 |
OM | +0.031 |
ABC | +0.021 |
time | +0.013 |
Neg logit contributions | |
grate | -0.253 |
rily | -0.234 |
away | -0.229 |
eyed | -0.227 |
freely | -0.219 |
Token | , |
Token ID | 11 |
Feature activation | +2.94e-02 |
Pos logit contributions | |
happened | +0.278 |
end | +0.242 |
OM | +0.216 |
time | +0.209 |
ABC | +0.204 |
Neg logit contributions | |
grate | -0.714 |
rily | -0.684 |
away | -0.659 |
eyed | -0.646 |
lived | -0.611 |
Token | they |
Token ID | 484 |
Feature activation | -5.54e-02 |
Pos logit contributions | |
happened | +0.035 |
end | +0.032 |
OM | +0.031 |
something | +0.030 |
time | +0.029 |
Neg logit contributions | |
grate | -0.056 |
eyed | -0.051 |
rily | -0.050 |
away | -0.050 |
lived | -0.046 |
Token | both |
Token ID | 1111 |
Feature activation | -3.87e-02 |
Pos logit contributions | |
grate | +0.111 |
rily | +0.105 |
away | +0.101 |
eyed | +0.097 |
lived | +0.096 |
Neg logit contributions | |
end | -0.058 |
happened | -0.058 |
OM | -0.050 |
been | -0.046 |
done | -0.045 |
Token | learned |
Token ID | 4499 |
Feature activation | +5.66e-02 |
Pos logit contributions | |
grate | +0.079 |
rily | +0.073 |
away | +0.071 |
eyed | +0.067 |
freely | +0.066 |
Neg logit contributions | |
happened | -0.042 |
end | -0.038 |
OM | -0.035 |
been | -0.032 |
ABC | -0.031 |
Token | an |
Token ID | 281 |
Feature activation | +6.72e-02 |
Pos logit contributions | |
happened | +0.065 |
OM | +0.063 |
end | +0.059 |
ABC | +0.056 |
OO | +0.050 |
Neg logit contributions | |
grate | -0.110 |
rily | -0.096 |
away | -0.096 |
eyed | -0.093 |
freely | -0.090 |
Token | who |
Token ID | 508 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
grate | +0.089 |
rily | +0.078 |
eyed | +0.076 |
away | +0.076 |
freely | +0.072 |
Neg logit contributions | |
happened | -0.064 |
end | -0.059 |
OM | -0.056 |
ABC | -0.054 |
been | -0.051 |
Token | could |
Token ID | 714 |
Feature activation | +5.32e-02 |
Pos logit contributions | |
grate | +0.246 |
rily | +0.223 |
away | +0.216 |
eyed | +0.212 |
freely | +0.203 |
Neg logit contributions | |
happened | -0.107 |
end | -0.090 |
OM | -0.089 |
ABC | -0.080 |
been | -0.075 |
Token | get |
Token ID | 651 |
Feature activation | +8.89e-02 |
Pos logit contributions | |
end | +0.072 |
happened | +0.068 |
OM | +0.062 |
been | +0.060 |
ABC | +0.058 |
Neg logit contributions | |
grate | -0.096 |
rily | -0.093 |
away | -0.085 |
oc | -0.079 |
eyed | -0.079 |
Token | to |
Token ID | 284 |
Feature activation | -8.27e-02 |
Pos logit contributions | |
happened | +0.101 |
end | +0.085 |
been | +0.080 |
done | +0.077 |
OM | +0.076 |
Neg logit contributions | |
grate | -0.181 |
rily | -0.165 |
oc | -0.157 |
away | -0.157 |
eyed | -0.156 |
Token | the |
Token ID | 262 |
Feature activation | +9.59e-02 |
Pos logit contributions | |
grate | +0.152 |
rily | +0.134 |
away | +0.129 |
eyed | +0.128 |
freely | +0.121 |
Neg logit contributions | |
happened | -0.106 |
end | -0.099 |
OM | -0.099 |
ABC | -0.091 |
been | -0.084 |
Token | end |
Token ID | 886 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
happened | +0.112 |
OM | +0.103 |
end | +0.094 |
ABC | +0.092 |
OO | +0.084 |
Neg logit contributions | |
grate | -0.187 |
away | -0.162 |
rily | -0.162 |
eyed | -0.156 |
freely | -0.154 |
Token | of |
Token ID | 286 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
happened | +0.379 |
end | +0.333 |
OM | +0.325 |
is | +0.310 |
time | +0.309 |
Neg logit contributions | |
grate | -0.613 |
rily | -0.539 |
away | -0.516 |
eyed | -0.515 |
slur | -0.493 |
Token | the |
Token ID | 262 |
Feature activation | +9.31e-02 |
Pos logit contributions | |
grate | +0.311 |
rily | +0.278 |
eyed | +0.274 |
away | +0.272 |
lived | +0.260 |
Neg logit contributions | |
happened | -0.173 |
end | -0.166 |
OM | -0.157 |
ABC | -0.142 |
time | -0.141 |
Token | wire |
Token ID | 6503 |
Feature activation | +6.56e-02 |
Pos logit contributions | |
end | +0.107 |
happened | +0.104 |
OM | +0.098 |
ABC | +0.092 |
time | +0.087 |
Neg logit contributions | |
grate | -0.187 |
rily | -0.171 |
eyed | -0.161 |
away | -0.159 |
freely | -0.150 |
Token | first |
Token ID | 717 |
Feature activation | +9.51e-02 |
Pos logit contributions | |
happened | +0.072 |
end | +0.068 |
OM | +0.063 |
ABC | +0.058 |
is | +0.056 |
Neg logit contributions | |
grate | -0.131 |
rily | -0.116 |
eyed | -0.115 |
away | -0.114 |
Follow | -0.107 |
Token | . |
Token ID | 13 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
happened | +0.086 |
OM | +0.080 |
end | +0.075 |
ABC | +0.070 |
consequences | +0.060 |
Neg logit contributions | |
grate | -0.208 |
rily | -0.186 |
away | -0.185 |
eyed | -0.180 |
freely | -0.175 |
Token | 't |
Token ID | 470 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
happened | +0.211 |
OM | +0.207 |
end | +0.196 |
ABC | +0.191 |
consequences | +0.177 |
Neg logit contributions | |
grate | -0.278 |
rily | -0.234 |
away | -0.234 |
eyed | -0.230 |
freely | -0.226 |
Token | rescue |
Token ID | 9992 |
Feature activation | -5.82e-02 |
Pos logit contributions | |
end | +0.037 |
happened | +0.035 |
OM | +0.032 |
been | +0.031 |
time | +0.030 |
Neg logit contributions | |
grate | -0.041 |
rily | -0.038 |
away | -0.034 |
eyed | -0.033 |
Follow | -0.032 |
Token | him |
Token ID | 683 |
Feature activation | +5.07e-02 |
Pos logit contributions | |
grate | +0.120 |
eyed | +0.108 |
rily | +0.107 |
Follow | +0.101 |
away | +0.100 |
Neg logit contributions | |
happened | -0.063 |
end | -0.060 |
time | -0.054 |
something | -0.050 |
ABC | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
end | +0.051 |
happened | +0.048 |
OM | +0.042 |
time | +0.040 |
ABC | +0.039 |
Neg logit contributions | |
grate | -0.103 |
rily | -0.095 |
eyed | -0.092 |
away | -0.091 |
Follow | -0.087 |
Token | The |
Token ID | 383 |
Feature activation | -6.89e-02 |
Pos logit contributions | |
happened | +0.039 |
OM | +0.037 |
end | +0.036 |
ABC | +0.034 |
been | +0.033 |
Neg logit contributions | |
grate | -0.047 |
away | -0.044 |
rily | -0.043 |
eyed | -0.040 |
active | -0.038 |
Token | end |
Token ID | 886 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
grate | +0.127 |
away | +0.108 |
rily | +0.107 |
eyed | +0.103 |
active | +0.102 |
Neg logit contributions | |
happened | -0.087 |
OM | -0.083 |
ABC | -0.072 |
end | -0.071 |
ained | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
happened | +0.372 |
OM | +0.347 |
end | +0.338 |
ABC | +0.312 |
consequences | +0.282 |
Neg logit contributions | |
grate | -0.618 |
rily | -0.547 |
away | -0.535 |
eyed | -0.524 |
freely | -0.504 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
happened | +0.023 |
OM | +0.022 |
end | +0.021 |
ABC | +0.019 |
been | +0.018 |
Neg logit contributions | |
grate | -0.046 |
rily | -0.042 |
away | -0.042 |
eyed | -0.040 |
active | -0.038 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
OM | +0.180 |
end | +0.171 |
happened | +0.169 |
consequences | +0.168 |
salon | +0.166 |
Neg logit contributions | |
grate | -0.133 |
freely | -0.114 |
eyed | -0.109 |
oc | -0.108 |
away | -0.104 |
Token | upon |
Token ID | 2402 |
Feature activation | +7.35e-02 |
Pos logit contributions | |
OM | +0.050 |
happened | +0.044 |
ABC | +0.044 |
consequences | +0.043 |
emon | +0.042 |
Neg logit contributions | |
grate | -0.051 |
rily | -0.046 |
away | -0.045 |
freely | -0.044 |
closely | -0.043 |
Token | a |
Token ID | 257 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
omm | +0.111 |
OM | +0.108 |
problem | +0.104 |
anger | +0.101 |
fault | +0.100 |
Neg logit contributions | |
grate | -0.106 |
active | -0.093 |
close | -0.092 |
freely | -0.090 |
eyed | -0.089 |
Token | have |
Token ID | 423 |
Feature activation | +8.29e-03 |
Pos logit contributions | |
grate | +0.147 |
rily | +0.133 |
away | +0.126 |
eyed | +0.124 |
freely | +0.117 |
Neg logit contributions | |
happened | -0.109 |
end | -0.107 |
OM | -0.103 |
ABC | -0.096 |
been | -0.092 |
Token | a |
Token ID | 257 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
happened | +0.007 |
OM | +0.006 |
end | +0.006 |
ABC | +0.006 |
consequences | +0.005 |
Neg logit contributions | |
grate | -0.019 |
rily | -0.017 |
away | -0.017 |
eyed | -0.016 |
freely | -0.016 |
Token | surprise |
Token ID | 5975 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
happened | +0.161 |
end | +0.157 |
OM | +0.144 |
time | +0.137 |
ABC | +0.137 |
Neg logit contributions | |
grate | -0.285 |
rily | -0.266 |
eyed | -0.254 |
away | -0.242 |
freely | -0.232 |
Token | at |
Token ID | 379 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
grate | +0.059 |
rily | +0.054 |
eyed | +0.053 |
away | +0.052 |
Follow | +0.052 |
Neg logit contributions | |
happened | -0.033 |
end | -0.031 |
time | -0.026 |
been | -0.026 |
done | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | +8.88e-02 |
Pos logit contributions | |
grate | +0.365 |
rily | +0.319 |
away | +0.314 |
eyed | +0.305 |
freely | +0.296 |
Neg logit contributions | |
happened | -0.257 |
OM | -0.246 |
end | -0.233 |
ABC | -0.222 |
consequences | -0.203 |
Token | end |
Token ID | 886 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
happened | +0.099 |
OM | +0.091 |
ABC | +0.079 |
end | +0.077 |
ained | +0.075 |
Neg logit contributions | |
grate | -0.175 |
away | -0.154 |
rily | -0.151 |
eyed | -0.145 |
freely | -0.144 |
Token | of |
Token ID | 286 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
happened | +0.418 |
end | +0.354 |
OM | +0.347 |
is | +0.342 |
time | +0.342 |
Neg logit contributions | |
grate | -0.586 |
rily | -0.500 |
away | -0.493 |
eyed | -0.488 |
greet | -0.467 |
Token | the |
Token ID | 262 |
Feature activation | +8.81e-02 |
Pos logit contributions | |
grate | +0.319 |
rily | +0.284 |
away | +0.283 |
eyed | +0.282 |
lived | +0.276 |
Neg logit contributions | |
happened | -0.173 |
end | -0.167 |
OM | -0.149 |
been | -0.144 |
time | -0.141 |
Token | class |
Token ID | 1398 |
Feature activation | +9.01e-02 |
Pos logit contributions | |
happened | +0.078 |
end | +0.075 |
OM | +0.071 |
ABC | +0.064 |
time | +0.057 |
Neg logit contributions | |
grate | -0.199 |
rily | -0.181 |
eyed | -0.174 |
away | -0.173 |
freely | -0.166 |
Token | . |
Token ID | 13 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
happened | +0.105 |
end | +0.095 |
OM | +0.094 |
ABC | +0.086 |
been | +0.081 |
Neg logit contributions | |
grate | -0.177 |
rily | -0.158 |
away | -0.154 |
eyed | -0.151 |
freely | -0.144 |
Token | Tim |
Token ID | 5045 |
Feature activation | +4.64e-03 |
Pos logit contributions | |
happened | +0.028 |
OM | +0.026 |
ABC | +0.025 |
omm | +0.024 |
end | +0.024 |
Neg logit contributions | |
grate | -0.047 |
away | -0.046 |
rily | -0.042 |
eyed | -0.042 |
active | -0.041 |
Token | felt |
Token ID | 2936 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
grate | +0.059 |
rily | +0.049 |
away | +0.049 |
eyed | +0.048 |
freely | +0.047 |
Neg logit contributions | |
happened | -0.050 |
OM | -0.049 |
ABC | -0.045 |
end | -0.045 |
fault | -0.043 |
Token | very |
Token ID | 845 |
Feature activation | +4.74e-02 |
Pos logit contributions | |
happened | +0.199 |
end | +0.184 |
OM | +0.184 |
ABC | +0.169 |
consequences | +0.157 |
Neg logit contributions | |
grate | -0.281 |
rily | -0.246 |
away | -0.243 |
eyed | -0.237 |
freely | -0.226 |
Token | special |
Token ID | 2041 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
happened | +0.058 |
end | +0.054 |
done | +0.050 |
time | +0.047 |
best | +0.047 |
Neg logit contributions | |
grate | -0.087 |
rily | -0.080 |
away | -0.080 |
oc | -0.079 |
Follow | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
grate | +0.251 |
rily | +0.226 |
eyed | +0.223 |
away | +0.221 |
Follow | +0.211 |
Neg logit contributions | |
happened | -0.150 |
end | -0.138 |
OM | -0.122 |
ABC | -0.119 |
been | -0.117 |
Token | The |
Token ID | 383 |
Feature activation | -6.93e-02 |
Pos logit contributions | |
happened | +0.028 |
OM | +0.027 |
end | +0.026 |
ABC | +0.025 |
omm | +0.024 |
Neg logit contributions | |
grate | -0.037 |
away | -0.034 |
rily | -0.033 |
eyed | -0.032 |
lived | -0.030 |
Token | end |
Token ID | 886 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
grate | +0.129 |
rily | +0.111 |
away | +0.110 |
eyed | +0.106 |
freely | +0.104 |
Neg logit contributions | |
happened | -0.087 |
OM | -0.083 |
end | -0.074 |
ABC | -0.072 |
OO | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
happened | +0.356 |
OM | +0.322 |
end | +0.322 |
ABC | +0.290 |
time | +0.261 |
Neg logit contributions | |
grate | -0.636 |
rily | -0.575 |
away | -0.554 |
eyed | -0.546 |
freely | -0.521 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
happened | +0.024 |
OM | +0.023 |
end | +0.022 |
ABC | +0.020 |
been | +0.019 |
Neg logit contributions | |
grate | -0.041 |
rily | -0.037 |
away | -0.037 |
eyed | -0.035 |
freely | -0.034 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.06e-02 |
Pos logit contributions | |
OM | +0.182 |
happened | +0.171 |
end | +0.170 |
consequences | +0.169 |
fault | +0.168 |
Neg logit contributions | |
grate | -0.126 |
freely | -0.110 |
eyed | -0.107 |
away | -0.100 |
oc | -0.100 |
Token | upon |
Token ID | 2402 |
Feature activation | +7.04e-02 |
Pos logit contributions | |
OM | +0.045 |
happened | +0.040 |
ABC | +0.039 |
consequences | +0.039 |
emon | +0.037 |
Neg logit contributions | |
grate | -0.048 |
rily | -0.043 |
away | -0.043 |
freely | -0.042 |
eyed | -0.040 |
Token | a |
Token ID | 257 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
omm | +0.108 |
OM | +0.106 |
problem | +0.100 |
anger | +0.099 |
fault | +0.098 |
Neg logit contributions | |
grate | -0.101 |
active | -0.090 |
close | -0.089 |
freely | -0.085 |
eyed | -0.085 |
Token | what |
Token ID | 644 |
Feature activation | +5.35e-02 |
Pos logit contributions | |
OM | +0.026 |
happened | +0.025 |
end | +0.024 |
ABC | +0.022 |
OO | +0.019 |
Neg logit contributions | |
grate | -0.079 |
rily | -0.070 |
away | -0.068 |
eyed | -0.067 |
freely | -0.066 |
Token | they |
Token ID | 484 |
Feature activation | -4.15e-02 |
Pos logit contributions | |
happened | +0.054 |
OM | +0.054 |
end | +0.050 |
ABC | +0.046 |
OO | +0.041 |
Neg logit contributions | |
grate | -0.109 |
rily | -0.096 |
away | -0.095 |
eyed | -0.093 |
freely | -0.090 |
Token | did |
Token ID | 750 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
grate | +0.089 |
rily | +0.084 |
away | +0.080 |
eyed | +0.077 |
active | +0.075 |
Neg logit contributions | |
happened | -0.044 |
end | -0.040 |
OM | -0.033 |
done | -0.032 |
been | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
happened | +0.157 |
end | +0.145 |
OM | +0.144 |
ABC | +0.130 |
been | +0.120 |
Neg logit contributions | |
grate | -0.269 |
rily | -0.238 |
away | -0.233 |
eyed | -0.227 |
freely | -0.218 |
Token | The |
Token ID | 383 |
Feature activation | -7.54e-02 |
Pos logit contributions | |
happened | +0.031 |
OM | +0.030 |
end | +0.030 |
ABC | +0.029 |
is | +0.027 |
Neg logit contributions | |
grate | -0.033 |
away | -0.031 |
rily | -0.030 |
lived | -0.027 |
eyed | -0.027 |
Token | end |
Token ID | 886 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
grate | +0.142 |
away | +0.122 |
rily | +0.119 |
eyed | +0.116 |
freely | +0.115 |
Neg logit contributions | |
happened | -0.089 |
OM | -0.086 |
ABC | -0.072 |
ained | -0.070 |
OO | -0.069 |
Token | . |
Token ID | 13 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
happened | +0.383 |
OM | +0.366 |
end | +0.350 |
ABC | +0.330 |
consequences | +0.298 |
Neg logit contributions | |
grate | -0.600 |
rily | -0.520 |
away | -0.516 |
eyed | -0.506 |
freely | -0.487 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
happened | +0.021 |
OM | +0.020 |
end | +0.019 |
ABC | +0.018 |
been | +0.016 |
Neg logit contributions | |
grate | -0.035 |
rily | -0.031 |
away | -0.031 |
eyed | -0.030 |
freely | -0.029 |
Token | Once |
Token ID | 7454 |
Feature activation | +2.91e-02 |
Pos logit contributions | |
OM | +0.186 |
fault | +0.176 |
salon | +0.175 |
consequences | +0.172 |
OO | +0.172 |
Neg logit contributions | |
grate | -0.116 |
freely | -0.101 |
oc | -0.098 |
eyed | -0.098 |
away | -0.091 |
Token | upon |
Token ID | 2402 |
Feature activation | +6.92e-02 |
Pos logit contributions | |
OM | +0.045 |
consequences | +0.040 |
ABC | +0.039 |
happened | +0.039 |
emon | +0.037 |
Neg logit contributions | |
grate | -0.042 |
away | -0.038 |
rily | -0.038 |
freely | -0.037 |
closely | -0.036 |
Token | a |
Token ID | 257 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
omm | +0.106 |
OM | +0.103 |
problem | +0.097 |
fault | +0.095 |
anger | +0.095 |
Neg logit contributions | |
grate | -0.099 |
active | -0.087 |
close | -0.087 |
away | -0.084 |
freely | -0.084 |
Token | think |
Token ID | 892 |
Feature activation | +3.22e-02 |
Pos logit contributions | |
grate | +0.018 |
rily | +0.017 |
away | +0.016 |
eyed | +0.016 |
freely | +0.015 |
Neg logit contributions | |
happened | -0.009 |
end | -0.009 |
OM | -0.008 |
ABC | -0.007 |
been | -0.007 |
Token | it |
Token ID | 340 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
happened | +0.041 |
OM | +0.039 |
end | +0.038 |
ABC | +0.035 |
consequences | +0.032 |
Neg logit contributions | |
grate | -0.059 |
rily | -0.051 |
away | -0.050 |
eyed | -0.049 |
freely | -0.047 |
Token | 's |
Token ID | 338 |
Feature activation | +5.06e-02 |
Pos logit contributions | |
OM | +0.121 |
happened | +0.105 |
end | +0.104 |
omm | +0.100 |
ABC | +0.098 |
Neg logit contributions | |
grate | -0.257 |
rily | -0.227 |
freely | -0.223 |
away | -0.222 |
eyed | -0.215 |
Token | time |
Token ID | 640 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
happened | +0.056 |
OM | +0.055 |
end | +0.050 |
ABC | +0.048 |
OO | +0.043 |
Neg logit contributions | |
grate | -0.100 |
rily | -0.088 |
away | -0.085 |
eyed | -0.084 |
freely | -0.082 |
Token | to |
Token ID | 284 |
Feature activation | -8.89e-02 |
Pos logit contributions | |
happened | +0.238 |
OM | +0.222 |
end | +0.219 |
ABC | +0.205 |
been | +0.191 |
Neg logit contributions | |
grate | -0.285 |
rily | -0.247 |
away | -0.242 |
eyed | -0.235 |
freely | -0.225 |
Token | end |
Token ID | 886 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
grate | +0.159 |
rily | +0.146 |
away | +0.139 |
eyed | +0.139 |
mer | +0.133 |
Neg logit contributions | |
end | -0.120 |
happened | -0.110 |
OM | -0.108 |
time | -0.099 |
ABC | -0.096 |
Token | now |
Token ID | 783 |
Feature activation | +5.68e-02 |
Pos logit contributions | |
happened | +0.341 |
end | +0.299 |
OM | +0.291 |
ABC | +0.275 |
time | +0.253 |
Neg logit contributions | |
grate | -0.653 |
rily | -0.583 |
away | -0.570 |
eyed | -0.560 |
freely | -0.538 |
Token | Jack |
Token ID | 3619 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
happened | +0.057 |
end | +0.052 |
OM | +0.045 |
is | +0.044 |
time | +0.042 |
Neg logit contributions | |
grate | -0.122 |
rily | -0.112 |
eyed | -0.109 |
away | -0.108 |
lived | -0.104 |
Token | - |
Token ID | 532 |
Feature activation | +2.63e-02 |
Pos logit contributions | |
grate | +0.065 |
rily | +0.058 |
away | +0.057 |
eyed | +0.056 |
freely | +0.054 |
Neg logit contributions | |
happened | -0.038 |
OM | -0.036 |
end | -0.035 |
ABC | -0.032 |
consequences | -0.029 |
Token | it |
Token ID | 340 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
happened | +0.035 |
end | +0.032 |
been | +0.029 |
was | +0.029 |
something | +0.029 |
Neg logit contributions | |
grate | -0.047 |
rily | -0.046 |
away | -0.045 |
eyed | -0.044 |
mer | -0.042 |
Token | was |
Token ID | 373 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
happened | +0.090 |
end | +0.075 |
OM | +0.074 |
ABC | +0.063 |
been | +0.055 |
Neg logit contributions | |
grate | -0.300 |
rily | -0.273 |
away | -0.271 |
eyed | -0.267 |
freely | -0.256 |
Token | nearly |
Token ID | 3016 |
Feature activation | +5.34e-02 |
Pos logit contributions | |
OM | +0.029 |
happened | +0.029 |
end | +0.028 |
ABC | +0.027 |
omm | +0.025 |
Neg logit contributions | |
grate | -0.032 |
away | -0.027 |
rily | -0.027 |
eyed | -0.027 |
freely | -0.026 |
Token | as |
Token ID | 355 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
happened | +0.079 |
OM | +0.074 |
end | +0.073 |
ABC | +0.068 |
been | +0.064 |
Neg logit contributions | |
grate | -0.088 |
rily | -0.076 |
away | -0.074 |
eyed | -0.072 |
freely | -0.068 |
Token | nice |
Token ID | 3621 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
happened | +0.139 |
end | +0.121 |
OM | +0.116 |
time | +0.113 |
been | +0.109 |
Neg logit contributions | |
grate | -0.190 |
rily | -0.171 |
away | -0.162 |
eyed | -0.162 |
oc | -0.155 |
Token | . |
Token ID | 13 |
Feature activation | +2.84e-02 |
Pos logit contributions | |
grate | +0.195 |
eyed | +0.175 |
rily | +0.171 |
away | +0.168 |
oc | +0.161 |
Neg logit contributions | |
happened | -0.131 |
end | -0.113 |
OM | -0.108 |
time | -0.104 |
ABC | -0.100 |
Token | The |
Token ID | 383 |
Feature activation | -7.44e-02 |
Pos logit contributions | |
happened | +0.037 |
OM | +0.036 |
end | +0.034 |
ABC | +0.033 |
been | +0.031 |
Neg logit contributions | |
grate | -0.049 |
away | -0.046 |
rily | -0.044 |
eyed | -0.041 |
active | -0.040 |
Token | end |
Token ID | 886 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
grate | +0.141 |
rily | +0.122 |
away | +0.121 |
eyed | +0.118 |
freely | +0.115 |
Neg logit contributions | |
happened | -0.090 |
OM | -0.085 |
end | -0.078 |
ABC | -0.075 |
OO | -0.069 |
Token | . |
Token ID | 13 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
happened | +0.366 |
OM | +0.339 |
end | +0.332 |
ABC | +0.304 |
consequences | +0.274 |
Neg logit contributions | |
grate | -0.623 |
rily | -0.554 |
away | -0.540 |
eyed | -0.530 |
freely | -0.508 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
happened | +0.021 |
OM | +0.021 |
end | +0.020 |
ABC | +0.018 |
been | +0.017 |
Neg logit contributions | |
grate | -0.045 |
rily | -0.041 |
away | -0.041 |
eyed | -0.038 |
active | -0.038 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
OM | +0.158 |
end | +0.151 |
happened | +0.148 |
consequences | +0.148 |
fault | +0.146 |
Neg logit contributions | |
grate | -0.143 |
freely | -0.125 |
eyed | -0.120 |
oc | -0.118 |
away | -0.116 |
Token | upon |
Token ID | 2402 |
Feature activation | +7.30e-02 |
Pos logit contributions | |
OM | +0.048 |
happened | +0.043 |
ABC | +0.042 |
consequences | +0.042 |
emon | +0.040 |
Neg logit contributions | |
grate | -0.053 |
rily | -0.047 |
away | -0.047 |
freely | -0.046 |
othe | -0.044 |
Token | a |
Token ID | 257 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
OM | +0.108 |
omm | +0.108 |
problem | +0.102 |
fault | +0.100 |
anger | +0.099 |
Neg logit contributions | |
grate | -0.107 |
active | -0.094 |
close | -0.093 |
freely | -0.090 |
away | -0.089 |
Token | ! |
Token ID | 0 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
happened | +0.000 |
OM | +0.000 |
end | +0.000 |
ABC | +0.000 |
omm | +0.000 |
Neg logit contributions | |
grate | -0.000 |
rily | -0.000 |
away | -0.000 |
eyed | -0.000 |
freely | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
happened | +0.042 |
OM | +0.040 |
end | +0.040 |
ABC | +0.037 |
been | +0.035 |
Neg logit contributions | |
grate | -0.052 |
rily | -0.046 |
away | -0.046 |
eyed | -0.043 |
freely | -0.041 |
Token |
|
Token ID | 198 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
happened | +0.026 |
OM | +0.024 |
end | +0.023 |
ABC | +0.022 |
consequences | +0.020 |
Neg logit contributions | |
grate | -0.044 |
rily | -0.038 |
eyed | -0.038 |
away | -0.038 |
freely | -0.036 |
Token | At |
Token ID | 2953 |
Feature activation | -5.66e-02 |
Pos logit contributions | |
OM | +0.031 |
end | +0.031 |
happened | +0.031 |
ABC | +0.027 |
been | +0.027 |
Neg logit contributions | |
grate | -0.041 |
rily | -0.037 |
away | -0.035 |
eyed | -0.032 |
freely | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | +8.85e-02 |
Pos logit contributions | |
grate | +0.098 |
away | +0.084 |
rily | +0.083 |
eyed | +0.080 |
freely | +0.079 |
Neg logit contributions | |
happened | -0.078 |
OM | -0.074 |
ABC | -0.069 |
end | -0.069 |
consequences | -0.063 |
Token | end |
Token ID | 886 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
happened | +0.067 |
OM | +0.055 |
ABC | +0.045 |
OO | +0.042 |
ained | +0.038 |
Neg logit contributions | |
grate | -0.207 |
away | -0.188 |
freely | -0.179 |
greet | -0.178 |
positive | -0.178 |
Token | of |
Token ID | 286 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
happened | +0.440 |
end | +0.413 |
OM | +0.392 |
ABC | +0.374 |
time | +0.369 |
Neg logit contributions | |
grate | -0.543 |
rily | -0.507 |
away | -0.476 |
eyed | -0.468 |
freely | -0.442 |
Token | the |
Token ID | 262 |
Feature activation | +8.98e-02 |
Pos logit contributions | |
grate | +0.283 |
rily | +0.247 |
away | +0.242 |
eyed | +0.237 |
freely | +0.227 |
Neg logit contributions | |
happened | -0.204 |
OM | -0.193 |
end | -0.188 |
ABC | -0.176 |
consequences | -0.160 |
Token | movie |
Token ID | 3807 |
Feature activation | +8.05e-02 |
Pos logit contributions | |
happened | +0.064 |
end | +0.061 |
OM | +0.057 |
ABC | +0.050 |
time | +0.043 |
Neg logit contributions | |
grate | -0.218 |
rily | -0.201 |
eyed | -0.192 |
away | -0.192 |
freely | -0.184 |
Token | , |
Token ID | 11 |
Feature activation | +2.91e-02 |
Pos logit contributions | |
happened | +0.087 |
OM | +0.081 |
end | +0.079 |
ABC | +0.072 |
been | +0.064 |
Neg logit contributions | |
grate | -0.164 |
rily | -0.145 |
away | -0.143 |
eyed | -0.140 |
freely | -0.135 |
Token | Jenny |
Token ID | 28437 |
Feature activation | +5.43e-02 |
Pos logit contributions | |
happened | +0.031 |
end | +0.028 |
OM | +0.027 |
ABC | +0.026 |
time | +0.025 |
Neg logit contributions | |
grate | -0.058 |
eyed | -0.054 |
away | -0.053 |
rily | -0.052 |
lived | -0.050 |
Token | |
Token ID | 220 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
happened | +0.049 |
OM | +0.047 |
end | +0.046 |
ABC | +0.043 |
been | +0.040 |
Neg logit contributions | |
grate | -0.060 |
away | -0.053 |
rily | -0.053 |
eyed | -0.050 |
freely | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
happened | +0.193 |
OM | +0.186 |
end | +0.183 |
ABC | +0.168 |
consequences | +0.164 |
Neg logit contributions | |
grate | -0.217 |
rily | -0.186 |
eyed | -0.183 |
freely | -0.175 |
away | -0.173 |
Token |
|
Token ID | 198 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
happened | +0.029 |
OM | +0.027 |
end | +0.026 |
ABC | +0.024 |
consequences | +0.022 |
Neg logit contributions | |
grate | -0.045 |
rily | -0.039 |
away | -0.039 |
eyed | -0.038 |
freely | -0.037 |
Token | In |
Token ID | 818 |
Feature activation | -9.34e-02 |
Pos logit contributions | |
OM | +0.032 |
happened | +0.032 |
end | +0.032 |
ABC | +0.028 |
been | +0.028 |
Neg logit contributions | |
grate | -0.041 |
rily | -0.036 |
away | -0.035 |
eyed | -0.032 |
freely | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | +8.77e-02 |
Pos logit contributions | |
grate | +0.184 |
rily | +0.163 |
away | +0.159 |
eyed | +0.158 |
freely | +0.149 |
Neg logit contributions | |
happened | -0.109 |
end | -0.105 |
OM | -0.099 |
ABC | -0.090 |
time | -0.086 |
Token | end |
Token ID | 886 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
happened | +0.027 |
OM | +0.019 |
ABC | +0.009 |
end | +0.008 |
OO | +0.001 |
Neg logit contributions | |
grate | -0.247 |
away | -0.224 |
rily | -0.222 |
eyed | -0.219 |
freely | -0.216 |
Token | , |
Token ID | 11 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
happened | +0.269 |
end | +0.240 |
OM | +0.211 |
ABC | +0.205 |
time | +0.201 |
Neg logit contributions | |
grate | -0.714 |
rily | -0.679 |
away | -0.656 |
eyed | -0.647 |
lived | -0.614 |
Token | the |
Token ID | 262 |
Feature activation | +8.71e-02 |
Pos logit contributions | |
happened | +0.033 |
end | +0.031 |
OM | +0.029 |
ABC | +0.027 |
time | +0.027 |
Neg logit contributions | |
grate | -0.056 |
eyed | -0.051 |
rily | -0.050 |
away | -0.050 |
lived | -0.046 |
Token | robot |
Token ID | 9379 |
Feature activation | -3.62e-03 |
Pos logit contributions | |
happened | +0.101 |
OM | +0.096 |
end | +0.089 |
ABC | +0.085 |
OO | +0.077 |
Neg logit contributions | |
grate | -0.170 |
rily | -0.147 |
away | -0.146 |
eyed | -0.142 |
freely | -0.138 |
Token | was |
Token ID | 373 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
grate | +0.008 |
rily | +0.008 |
away | +0.008 |
eyed | +0.007 |
freely | +0.007 |
Neg logit contributions | |
happened | -0.003 |
end | -0.003 |
OM | -0.002 |
ABC | -0.002 |
been | -0.002 |
Token | so |
Token ID | 523 |
Feature activation | +6.14e-02 |
Pos logit contributions | |
happened | +0.250 |
end | +0.240 |
OM | +0.218 |
been | +0.214 |
done | +0.213 |
Neg logit contributions | |
grate | -0.433 |
rily | -0.396 |
away | -0.395 |
eyed | -0.366 |
Follow | -0.364 |
Token | fast |
Token ID | 3049 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
happened | +0.027 |
OM | +0.025 |
end | +0.024 |
ABC | +0.022 |
been | +0.019 |
Neg logit contributions | |
grate | -0.051 |
rily | -0.045 |
away | -0.045 |
eyed | -0.043 |
freely | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
grate | +0.259 |
rily | +0.229 |
eyed | +0.226 |
away | +0.225 |
freely | +0.213 |
Neg logit contributions | |
happened | -0.144 |
end | -0.131 |
OM | -0.125 |
ABC | -0.121 |
consequences | -0.113 |
Token | It |
Token ID | 632 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
OM | +0.026 |
ABC | +0.023 |
happened | +0.023 |
is | +0.022 |
omm | +0.022 |
Neg logit contributions | |
away | -0.039 |
grate | -0.039 |
rily | -0.037 |
ized | -0.034 |
eyed | -0.034 |
Token | 's |
Token ID | 338 |
Feature activation | +4.43e-02 |
Pos logit contributions | |
grate | +0.106 |
away | +0.100 |
rily | +0.100 |
eyed | +0.095 |
freely | +0.092 |
Neg logit contributions | |
happened | -0.037 |
end | -0.029 |
is | -0.029 |
OM | -0.029 |
ABC | -0.027 |
Token | really |
Token ID | 1107 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
happened | +0.044 |
OM | +0.044 |
end | +0.038 |
ABC | +0.037 |
CA | +0.032 |
Neg logit contributions | |
grate | -0.093 |
rily | -0.080 |
away | -0.078 |
active | -0.078 |
eyed | -0.078 |
Token | active |
Token ID | 4075 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
grate | +0.056 |
rily | +0.052 |
away | +0.049 |
eyed | +0.048 |
freely | +0.046 |
Neg logit contributions | |
happened | -0.038 |
OM | -0.033 |
end | -0.033 |
ABC | -0.031 |
been | -0.029 |
Token | !" |
Token ID | 2474 |
Feature activation | +8.72e-02 |
Pos logit contributions | |
grate | +0.533 |
rily | +0.464 |
away | +0.456 |
eyed | +0.446 |
freely | +0.428 |
Neg logit contributions | |
happened | -0.384 |
OM | -0.363 |
end | -0.353 |
ABC | -0.331 |
consequences | -0.301 |
Token |
|
Token ID | 198 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
happened | +0.123 |
OM | +0.121 |
end | +0.118 |
ABC | +0.109 |
consequences | +0.107 |
Neg logit contributions | |
grate | -0.145 |
eyed | -0.121 |
rily | -0.121 |
away | -0.117 |
freely | -0.112 |
Token |
|
Token ID | 198 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
happened | +0.017 |
OM | +0.016 |
end | +0.016 |
consequences | +0.015 |
ABC | +0.015 |
Neg logit contributions | |
grate | -0.018 |
eyed | -0.016 |
away | -0.015 |
rily | -0.015 |
greet | -0.014 |
Token | Mary |
Token ID | 24119 |
Feature activation | +4.80e-02 |
Pos logit contributions | |
happened | +0.017 |
OM | +0.017 |
end | +0.016 |
ABC | +0.015 |
fault | +0.015 |
Neg logit contributions | |
grate | -0.018 |
eyed | -0.015 |
away | -0.015 |
rily | -0.015 |
freely | -0.014 |
Token | smiled |
Token ID | 13541 |
Feature activation | +6.36e-02 |
Pos logit contributions | |
happened | +0.052 |
OM | +0.048 |
end | +0.047 |
ABC | +0.043 |
been | +0.039 |
Neg logit contributions | |
grate | -0.096 |
rily | -0.087 |
away | -0.086 |
eyed | -0.083 |
freely | -0.081 |
Token | . |
Token ID | 13 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
grate | +0.287 |
rily | +0.258 |
eyed | +0.251 |
away | +0.249 |
active | +0.239 |
Neg logit contributions | |
happened | -0.147 |
end | -0.132 |
OM | -0.123 |
ABC | -0.111 |
been | -0.108 |
Token | The |
Token ID | 383 |
Feature activation | -8.93e-02 |
Pos logit contributions | |
happened | +0.031 |
OM | +0.029 |
end | +0.028 |
ABC | +0.027 |
been | +0.026 |
Neg logit contributions | |
grate | -0.027 |
rily | -0.023 |
away | -0.023 |
eyed | -0.022 |
freely | -0.020 |
Token | squirrel |
Token ID | 33039 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
grate | +0.118 |
away | +0.097 |
active | +0.095 |
eyed | +0.091 |
freely | +0.089 |
Neg logit contributions | |
OM | -0.153 |
happened | -0.151 |
ABC | -0.139 |
abella | -0.139 |
fault | -0.137 |
Token | was |
Token ID | 373 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
happened | +0.026 |
end | +0.021 |
OM | +0.019 |
was | +0.018 |
is | +0.016 |
Neg logit contributions | |
grate | -0.083 |
rily | -0.079 |
away | -0.075 |
eyed | -0.074 |
greet | -0.071 |
Token | very |
Token ID | 845 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
happened | +0.293 |
OM | +0.281 |
end | +0.271 |
ABC | +0.256 |
consequences | +0.235 |
Neg logit contributions | |
grate | -0.366 |
rily | -0.314 |
away | -0.307 |
eyed | -0.303 |
freely | -0.290 |
Token | active |
Token ID | 4075 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
happened | +0.019 |
end | +0.017 |
done | +0.017 |
best | +0.016 |
time | +0.015 |
Neg logit contributions | |
grate | -0.032 |
rily | -0.032 |
away | -0.028 |
greet | -0.028 |
ishly | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | -6.76e-02 |
Pos logit contributions | |
grate | +0.460 |
rily | +0.384 |
away | +0.380 |
freely | +0.364 |
eyed | +0.364 |
Neg logit contributions | |
OM | -0.451 |
happened | -0.445 |
end | -0.417 |
ABC | -0.414 |
omm | -0.378 |
Token | didn |
Token ID | 1422 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
grate | +0.134 |
rily | +0.126 |
away | +0.122 |
eyed | +0.119 |
freely | +0.111 |
Neg logit contributions | |
happened | -0.079 |
end | -0.065 |
OM | -0.065 |
been | -0.060 |
done | -0.059 |
Token | 't |
Token ID | 470 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
happened | +0.237 |
OM | +0.221 |
end | +0.206 |
ABC | +0.204 |
consequences | +0.204 |
Neg logit contributions | |
grate | -0.253 |
eyed | -0.208 |
rily | -0.207 |
away | -0.201 |
freely | -0.199 |
Token | want |
Token ID | 765 |
Feature activation | -7.38e-02 |
Pos logit contributions | |
grate | +0.028 |
rily | +0.025 |
away | +0.022 |
eyed | +0.022 |
Follow | +0.021 |
Neg logit contributions | |
happened | -0.028 |
end | -0.027 |
OM | -0.026 |
ABC | -0.024 |
been | -0.024 |
Token | to |
Token ID | 284 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
grate | +0.110 |
away | +0.092 |
rily | +0.090 |
eyed | +0.087 |
freely | +0.085 |
Neg logit contributions | |
happened | -0.117 |
OM | -0.116 |
end | -0.110 |
ABC | -0.106 |
OO | -0.100 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
happened | +0.262 |
end | +0.251 |
OM | +0.240 |
ABC | +0.222 |
been | +0.219 |
Neg logit contributions | |
grate | -0.376 |
rily | -0.335 |
away | -0.330 |
eyed | -0.313 |
freely | -0.305 |
Token | to |
Token ID | 284 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
grate | +0.223 |
rily | +0.213 |
away | +0.206 |
eyed | +0.205 |
Follow | +0.198 |
Neg logit contributions | |
happened | -0.131 |
end | -0.118 |
was | -0.102 |
been | -0.101 |
best | -0.098 |
Token | see |
Token ID | 766 |
Feature activation | -6.64e-02 |
Pos logit contributions | |
grate | +0.190 |
rily | +0.179 |
away | +0.161 |
eyed | +0.155 |
lived | +0.154 |
Neg logit contributions | |
end | -0.135 |
happened | -0.132 |
OM | -0.120 |
been | -0.114 |
ABC | -0.112 |
Token | him |
Token ID | 683 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
grate | +0.128 |
eyed | +0.114 |
rily | +0.114 |
away | +0.109 |
Follow | +0.108 |
Neg logit contributions | |
happened | -0.080 |
end | -0.074 |
ABC | -0.063 |
been | -0.063 |
something | -0.062 |
Token | so |
Token ID | 523 |
Feature activation | +5.17e-02 |
Pos logit contributions | |
happened | +0.031 |
end | +0.031 |
OM | +0.028 |
ABC | +0.027 |
been | +0.025 |
Neg logit contributions | |
grate | -0.048 |
rily | -0.043 |
away | -0.043 |
eyed | -0.042 |
Follow | -0.039 |
Token | active |
Token ID | 4075 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
happened | +0.061 |
end | +0.058 |
best | +0.055 |
was | +0.052 |
been | +0.051 |
Neg logit contributions | |
grate | -0.093 |
rily | -0.091 |
oc | -0.085 |
Follow | -0.083 |
Mic | -0.082 |
Token | and |
Token ID | 290 |
Feature activation | -6.95e-02 |
Pos logit contributions | |
grate | +0.587 |
rily | +0.522 |
eyed | +0.510 |
away | +0.505 |
Follow | +0.478 |
Neg logit contributions | |
happened | -0.334 |
end | -0.309 |
OM | -0.283 |
ABC | -0.263 |
been | -0.253 |
Token | healthy |
Token ID | 5448 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
rily | +0.119 |
grate | +0.116 |
away | +0.105 |
oc | +0.101 |
mer | +0.097 |
Neg logit contributions | |
happened | -0.098 |
been | -0.090 |
done | -0.089 |
end | -0.087 |
best | -0.086 |
Token | . |
Token ID | 13 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
grate | +0.420 |
rily | +0.373 |
away | +0.363 |
eyed | +0.360 |
freely | +0.342 |
Neg logit contributions | |
happened | -0.245 |
end | -0.225 |
OM | -0.221 |
ABC | -0.204 |
consequences | -0.187 |
Token | And |
Token ID | 843 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
happened | +0.027 |
end | +0.026 |
OM | +0.026 |
ABC | +0.024 |
been | +0.023 |
Neg logit contributions | |
grate | -0.027 |
rily | -0.023 |
away | -0.023 |
eyed | -0.022 |
active | -0.021 |
Token | that |
Token ID | 326 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
grate | +0.252 |
rily | +0.242 |
eyed | +0.238 |
itching | +0.231 |
away | +0.230 |
Neg logit contributions | |
happened | -0.179 |
end | -0.172 |
was | -0.166 |
best | -0.160 |
time | -0.155 |
Token | old |
Token ID | 1468 |
Feature activation | -7.67e-02 |
Pos logit contributions | |
end | +0.009 |
happened | +0.009 |
ABC | +0.007 |
time | +0.007 |
dentist | +0.006 |
Neg logit contributions | |
grate | -0.024 |
rily | -0.021 |
eyed | -0.020 |
away | -0.020 |
Follow | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
grate | +0.136 |
away | +0.115 |
rily | +0.114 |
eyed | +0.112 |
active | +0.105 |
Neg logit contributions | |
happened | -0.103 |
OM | -0.102 |
end | -0.092 |
ABC | -0.089 |
ained | -0.083 |
Token | Jack |
Token ID | 3619 |
Feature activation | -4.61e-02 |
Pos logit contributions | |
happened | +0.040 |
OM | +0.038 |
end | +0.038 |
ABC | +0.036 |
time | +0.034 |
Neg logit contributions | |
grate | -0.024 |
rily | -0.019 |
away | -0.019 |
eyed | -0.018 |
freely | -0.016 |
Token | was |
Token ID | 373 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
grate | +0.101 |
rily | +0.092 |
away | +0.090 |
eyed | +0.086 |
freely | +0.085 |
Neg logit contributions | |
happened | -0.044 |
OM | -0.038 |
end | -0.038 |
ABC | -0.034 |
been | -0.032 |
Token | very |
Token ID | 845 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
happened | +0.155 |
end | +0.134 |
OM | +0.124 |
ABC | +0.122 |
been | +0.111 |
Neg logit contributions | |
grate | -0.499 |
rily | -0.468 |
away | -0.463 |
eyed | -0.436 |
freely | -0.432 |
Token | active |
Token ID | 4075 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
happened | +0.020 |
best | +0.019 |
end | +0.018 |
ABC | +0.017 |
done | +0.017 |
Neg logit contributions | |
rily | -0.031 |
grate | -0.030 |
away | -0.028 |
freely | -0.026 |
oc | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | -6.74e-02 |
Pos logit contributions | |
grate | +0.516 |
rily | +0.448 |
away | +0.440 |
eyed | +0.430 |
freely | +0.411 |
Neg logit contributions | |
happened | -0.401 |
OM | -0.378 |
end | -0.370 |
ABC | -0.347 |
consequences | -0.318 |
Token | loved |
Token ID | 6151 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
grate | +0.146 |
rily | +0.143 |
away | +0.136 |
eyed | +0.126 |
freely | +0.126 |
Neg logit contributions | |
happened | -0.067 |
end | -0.054 |
OM | -0.053 |
been | -0.051 |
ABC | -0.051 |
Token | to |
Token ID | 284 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
grate | +0.042 |
away | +0.035 |
rily | +0.035 |
eyed | +0.034 |
active | +0.033 |
Neg logit contributions | |
happened | -0.044 |
OM | -0.043 |
end | -0.042 |
ABC | -0.039 |
OO | -0.037 |
Token | run |
Token ID | 1057 |
Feature activation | -5.31e-02 |
Pos logit contributions | |
grate | +0.155 |
rily | +0.134 |
away | +0.128 |
eyed | +0.124 |
freely | +0.118 |
Neg logit contributions | |
happened | -0.169 |
OM | -0.160 |
end | -0.160 |
ABC | -0.152 |
been | -0.140 |
Token | around |
Token ID | 1088 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
grate | +0.086 |
rily | +0.076 |
eyed | +0.070 |
away | +0.068 |
lived | +0.065 |
Neg logit contributions | |
happened | -0.079 |
end | -0.076 |
OM | -0.073 |
ABC | -0.070 |
time | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
grate | +0.041 |
away | +0.034 |
eyed | +0.034 |
rily | +0.034 |
freely | +0.032 |
Neg logit contributions | |
OM | -0.031 |
happened | -0.031 |
end | -0.028 |
ABC | -0.028 |
consequences | -0.025 |
Token | Spot |
Token ID | 15899 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
happened | +0.030 |
OM | +0.028 |
end | +0.028 |
ABC | +0.026 |
time | +0.025 |
Neg logit contributions | |
grate | -0.033 |
rily | -0.029 |
away | -0.028 |
eyed | -0.028 |
freely | -0.026 |
Token | was |
Token ID | 373 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
grate | +0.040 |
active | +0.034 |
eyed | +0.034 |
away | +0.034 |
freely | +0.033 |
Neg logit contributions | |
OM | -0.032 |
happened | -0.029 |
end | -0.028 |
ABC | -0.027 |
fault | -0.026 |
Token | always |
Token ID | 1464 |
Feature activation | -8.54e-02 |
Pos logit contributions | |
happened | +0.167 |
end | +0.145 |
OM | +0.143 |
ABC | +0.128 |
been | +0.116 |
Neg logit contributions | |
grate | -0.490 |
rily | -0.452 |
away | -0.445 |
eyed | -0.430 |
freely | -0.418 |
Token | very |
Token ID | 845 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
grate | +0.159 |
rily | +0.142 |
away | +0.138 |
eyed | +0.134 |
freely | +0.128 |
Neg logit contributions | |
happened | -0.108 |
OM | -0.099 |
end | -0.098 |
ABC | -0.091 |
been | -0.086 |
Token | active |
Token ID | 4075 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
best | +0.018 |
happened | +0.018 |
time | +0.017 |
done | +0.016 |
a | +0.016 |
Neg logit contributions | |
rily | -0.035 |
grate | -0.032 |
away | -0.029 |
ishly | -0.028 |
oc | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | -6.76e-02 |
Pos logit contributions | |
grate | +0.482 |
away | +0.403 |
rily | +0.402 |
eyed | +0.388 |
freely | +0.382 |
Neg logit contributions | |
happened | -0.430 |
OM | -0.429 |
end | -0.395 |
ABC | -0.387 |
omm | -0.353 |
Token | full |
Token ID | 1336 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
grate | +0.141 |
rily | +0.136 |
away | +0.129 |
eyed | +0.123 |
freely | +0.118 |
Neg logit contributions | |
happened | -0.070 |
end | -0.061 |
OM | -0.059 |
been | -0.056 |
ABC | -0.055 |
Token | of |
Token ID | 286 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
grate | +0.211 |
rily | +0.174 |
away | +0.171 |
eyed | +0.165 |
freely | +0.157 |
Neg logit contributions | |
happened | -0.241 |
OM | -0.234 |
end | -0.225 |
ABC | -0.216 |
OO | -0.201 |
Token | energy |
Token ID | 2568 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
grate | +0.320 |
rily | +0.278 |
away | +0.274 |
eyed | +0.268 |
freely | +0.258 |
Neg logit contributions | |
happened | -0.220 |
OM | -0.209 |
end | -0.201 |
ABC | -0.188 |
consequences | -0.171 |
Token | . |
Token ID | 13 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
happened | +0.019 |
end | +0.017 |
OM | +0.016 |
ABC | +0.015 |
been | +0.014 |
Neg logit contributions | |
grate | -0.038 |
rily | -0.034 |
away | -0.033 |
eyed | -0.033 |
freely | -0.031 |
Token | run |
Token ID | 1057 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
grate | +0.129 |
rily | +0.128 |
mer | +0.117 |
away | +0.116 |
eyed | +0.113 |
Neg logit contributions | |
end | -0.117 |
happened | -0.111 |
time | -0.101 |
ABC | -0.100 |
OM | -0.099 |
Token | and |
Token ID | 290 |
Feature activation | -6.55e-02 |
Pos logit contributions | |
grate | +0.066 |
rily | +0.056 |
away | +0.055 |
eyed | +0.053 |
freely | +0.051 |
Neg logit contributions | |
happened | -0.073 |
OM | -0.069 |
end | -0.068 |
ABC | -0.064 |
consequences | -0.060 |
Token | play |
Token ID | 711 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
grate | +0.105 |
rily | +0.096 |
away | +0.091 |
eyed | +0.086 |
freely | +0.079 |
Neg logit contributions | |
happened | -0.099 |
end | -0.096 |
OM | -0.093 |
ABC | -0.087 |
been | -0.085 |
Token | and |
Token ID | 290 |
Feature activation | -6.84e-02 |
Pos logit contributions | |
grate | +0.042 |
rily | +0.036 |
eyed | +0.035 |
away | +0.034 |
lived | +0.031 |
Neg logit contributions | |
happened | -0.042 |
end | -0.040 |
OM | -0.039 |
ABC | -0.036 |
time | -0.035 |
Token | stay |
Token ID | 2652 |
Feature activation | -2.48e-03 |
Pos logit contributions | |
grate | +0.112 |
rily | +0.103 |
away | +0.096 |
eyed | +0.093 |
lived | +0.085 |
Neg logit contributions | |
happened | -0.101 |
end | -0.098 |
OM | -0.093 |
ABC | -0.090 |
been | -0.086 |
Token | active |
Token ID | 4075 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
grate | +0.004 |
rily | +0.004 |
Follow | +0.004 |
lived | +0.004 |
othe | +0.004 |
Neg logit contributions | |
time | -0.003 |
best | -0.003 |
end | -0.003 |
a | -0.003 |
something | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
grate | +0.535 |
rily | +0.466 |
away | +0.460 |
eyed | +0.444 |
freely | +0.435 |
Neg logit contributions | |
happened | -0.382 |
OM | -0.368 |
end | -0.341 |
ABC | -0.328 |
OO | -0.299 |
Token | Pet |
Token ID | 4767 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
happened | +0.030 |
OM | +0.028 |
end | +0.028 |
ABC | +0.026 |
been | +0.024 |
Neg logit contributions | |
grate | -0.037 |
rily | -0.032 |
away | -0.032 |
eyed | -0.031 |
freely | -0.029 |
Token | ey |
Token ID | 2959 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
grate | +0.027 |
away | +0.022 |
rily | +0.022 |
eyed | +0.021 |
freely | +0.019 |
Neg logit contributions | |
happened | -0.047 |
OM | -0.047 |
end | -0.045 |
ABC | -0.044 |
fault | -0.042 |
Token | on |
Token ID | 319 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
grate | +0.225 |
rily | +0.203 |
away | +0.200 |
eyed | +0.194 |
freely | +0.189 |
Neg logit contributions | |
happened | -0.095 |
end | -0.086 |
OM | -0.082 |
ABC | -0.076 |
been | -0.069 |
Token | the |
Token ID | 262 |
Feature activation | +8.21e-02 |
Pos logit contributions | |
grate | +0.270 |
rily | +0.240 |
eyed | +0.238 |
away | +0.232 |
lived | +0.222 |
Neg logit contributions | |
happened | -0.138 |
end | -0.133 |
ABC | -0.120 |
time | -0.117 |
OM | -0.116 |
Token | 't |
Token ID | 470 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
grate | +0.133 |
rily | +0.115 |
away | +0.113 |
eyed | +0.112 |
freely | +0.107 |
Neg logit contributions | |
happened | -0.088 |
OM | -0.082 |
end | -0.079 |
ABC | -0.075 |
fault | -0.070 |
Token | be |
Token ID | 307 |
Feature activation | +9.21e-02 |
Pos logit contributions | |
grate | +0.024 |
rily | +0.024 |
away | +0.020 |
eyed | +0.020 |
oc | +0.020 |
Neg logit contributions | |
end | -0.021 |
happened | -0.019 |
OM | -0.019 |
been | -0.018 |
time | -0.018 |
Token | lazy |
Token ID | 16931 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
happened | +0.136 |
OM | +0.128 |
end | +0.126 |
ABC | +0.119 |
consequences | +0.111 |
Neg logit contributions | |
grate | -0.151 |
rily | -0.131 |
away | -0.128 |
eyed | -0.124 |
freely | -0.119 |
Token | , |
Token ID | 11 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
grate | +0.326 |
rily | +0.293 |
eyed | +0.288 |
away | +0.285 |
oc | +0.268 |
Neg logit contributions | |
happened | -0.146 |
end | -0.130 |
OM | -0.115 |
time | -0.112 |
consequences | -0.109 |
Token | being |
Token ID | 852 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
end | +0.025 |
happened | +0.021 |
wasn | +0.021 |
is | +0.020 |
something | +0.020 |
Neg logit contributions | |
eyed | -0.060 |
grate | -0.055 |
away | -0.055 |
mer | -0.052 |
oc | -0.052 |
Token | active |
Token ID | 4075 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
happened | +0.093 |
OM | +0.087 |
end | +0.086 |
ABC | +0.081 |
been | +0.077 |
Neg logit contributions | |
grate | -0.088 |
rily | -0.076 |
away | -0.073 |
eyed | -0.071 |
freely | -0.067 |
Token | and |
Token ID | 290 |
Feature activation | -6.38e-02 |
Pos logit contributions | |
grate | +0.537 |
rily | +0.469 |
away | +0.462 |
eyed | +0.451 |
freely | +0.437 |
Neg logit contributions | |
happened | -0.369 |
OM | -0.357 |
end | -0.334 |
ABC | -0.318 |
OO | -0.288 |
Token | doing |
Token ID | 1804 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
rily | +0.116 |
grate | +0.111 |
away | +0.103 |
oc | +0.096 |
mer | +0.093 |
Neg logit contributions | |
happened | -0.084 |
end | -0.082 |
done | -0.081 |
been | -0.079 |
is | -0.076 |
Token | exercise |
Token ID | 5517 |
Feature activation | -7.88e-03 |
Pos logit contributions | |
grate | +0.034 |
rily | +0.030 |
away | +0.030 |
eyed | +0.029 |
freely | +0.028 |
Neg logit contributions | |
happened | -0.019 |
OM | -0.018 |
end | -0.017 |
ABC | -0.016 |
consequences | -0.014 |
Token | makes |
Token ID | 1838 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
grate | +0.015 |
rily | +0.013 |
away | +0.012 |
eyed | +0.012 |
greet | +0.011 |
Neg logit contributions | |
happened | -0.010 |
end | -0.009 |
OM | -0.009 |
ABC | -0.008 |
time | -0.008 |
Token | us |
Token ID | 514 |
Feature activation | +5.55e-02 |
Pos logit contributions | |
grate | +0.030 |
rily | +0.025 |
away | +0.025 |
eyed | +0.024 |
freely | +0.023 |
Neg logit contributions | |
happened | -0.027 |
OM | -0.026 |
end | -0.025 |
ABC | -0.024 |
OO | -0.022 |
Token | reminder |
Token ID | 15438 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
grate | +0.200 |
away | +0.177 |
rily | +0.177 |
eyed | +0.172 |
freely | +0.169 |
Neg logit contributions | |
happened | -0.081 |
OM | -0.076 |
end | -0.069 |
ABC | -0.065 |
consequences | -0.055 |
Token | to |
Token ID | 284 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
happened | +0.048 |
end | +0.045 |
OM | +0.044 |
ABC | +0.041 |
been | +0.039 |
Neg logit contributions | |
grate | -0.054 |
rily | -0.047 |
away | -0.045 |
eyed | -0.045 |
freely | -0.043 |
Token | be |
Token ID | 307 |
Feature activation | +8.83e-02 |
Pos logit contributions | |
grate | +0.180 |
rily | +0.166 |
away | +0.161 |
eyed | +0.158 |
oc | +0.153 |
Neg logit contributions | |
end | -0.128 |
happened | -0.128 |
OM | -0.122 |
ABC | -0.113 |
been | -0.111 |
Token | wise |
Token ID | 10787 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
happened | +0.124 |
end | +0.118 |
been | +0.111 |
done | +0.107 |
OM | +0.107 |
Neg logit contributions | |
grate | -0.143 |
rily | -0.135 |
away | -0.126 |
eyed | -0.118 |
oc | -0.116 |
Token | and |
Token ID | 290 |
Feature activation | -6.83e-02 |
Pos logit contributions | |
grate | +0.408 |
rily | +0.362 |
away | +0.350 |
eyed | +0.346 |
freely | +0.325 |
Neg logit contributions | |
happened | -0.301 |
end | -0.276 |
OM | -0.266 |
ABC | -0.257 |
been | -0.242 |
Token | active |
Token ID | 4075 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
rily | +0.122 |
grate | +0.117 |
away | +0.109 |
oc | +0.105 |
mer | +0.104 |
Neg logit contributions | |
happened | -0.091 |
end | -0.083 |
been | -0.083 |
done | -0.081 |
was | -0.076 |
Token | even |
Token ID | 772 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
grate | +0.585 |
rily | +0.530 |
eyed | +0.518 |
away | +0.512 |
lived | +0.477 |
Neg logit contributions | |
happened | -0.326 |
end | -0.305 |
OM | -0.261 |
is | -0.259 |
time | -0.256 |
Token | when |
Token ID | 618 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
grate | +0.040 |
rily | +0.037 |
eyed | +0.034 |
away | +0.033 |
freely | +0.032 |
Neg logit contributions | |
happened | -0.028 |
end | -0.027 |
time | -0.023 |
something | -0.023 |
best | -0.023 |
Token | having |
Token ID | 1719 |
Feature activation | -5.00e-03 |
Pos logit contributions | |
OM | +0.018 |
happened | +0.018 |
end | +0.017 |
ABC | +0.016 |
OO | +0.015 |
Neg logit contributions | |
grate | -0.021 |
rily | -0.018 |
away | -0.018 |
freely | -0.017 |
eyed | -0.017 |
Token | fun |
Token ID | 1257 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
grate | +0.010 |
rily | +0.009 |
away | +0.009 |
eyed | +0.009 |
freely | +0.009 |
Neg logit contributions | |
happened | -0.005 |
OM | -0.005 |
end | -0.005 |
ABC | -0.004 |
OO | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +1.93e-02 |
Pos logit contributions | |
grate | +0.045 |
rily | +0.039 |
eyed | +0.039 |
away | +0.038 |
Follow | +0.036 |
Neg logit contributions | |
happened | -0.025 |
end | -0.021 |
OM | -0.019 |
time | -0.019 |
ABC | -0.019 |
Token | would |
Token ID | 561 |
Feature activation | +9.24e-02 |
Pos logit contributions | |
grate | +0.008 |
away | +0.007 |
rily | +0.007 |
eyed | +0.007 |
freely | +0.007 |
Neg logit contributions | |
happened | -0.004 |
end | -0.003 |
OM | -0.003 |
ABC | -0.003 |
was | -0.003 |
Token | remind |
Token ID | 7101 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
end | +0.121 |
time | +0.096 |
happened | +0.096 |
been | +0.096 |
is | +0.093 |
Neg logit contributions | |
rily | -0.168 |
away | -0.162 |
mer | -0.161 |
oc | -0.159 |
grate | -0.156 |
Token | her |
Token ID | 607 |
Feature activation | +5.67e-02 |
Pos logit contributions | |
grate | +0.217 |
rily | +0.185 |
away | +0.180 |
eyed | +0.175 |
freely | +0.170 |
Neg logit contributions | |
happened | -0.171 |
OM | -0.166 |
end | -0.151 |
ABC | -0.146 |
OO | -0.145 |
Token | to |
Token ID | 284 |
Feature activation | -9.97e-02 |
Pos logit contributions | |
happened | +0.081 |
end | +0.075 |
OM | +0.074 |
ABC | +0.070 |
been | +0.065 |
Neg logit contributions | |
grate | -0.096 |
rily | -0.084 |
away | -0.082 |
eyed | -0.079 |
freely | -0.077 |
Token | stay |
Token ID | 2652 |
Feature activation | -1.68e-03 |
Pos logit contributions | |
rily | +0.180 |
grate | +0.177 |
away | +0.162 |
oc | +0.160 |
mer | +0.158 |
Neg logit contributions | |
end | -0.130 |
happened | -0.119 |
OM | -0.118 |
been | -0.114 |
done | -0.110 |
Token | active |
Token ID | 4075 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
Follow | +0.003 |
grate | +0.003 |
lived | +0.003 |
oc | +0.003 |
othe | +0.002 |
Neg logit contributions | |
something | -0.002 |
a | -0.002 |
end | -0.002 |
best | -0.002 |
time | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | -6.74e-02 |
Pos logit contributions | |
grate | +0.579 |
rily | +0.522 |
eyed | +0.514 |
away | +0.508 |
lived | +0.483 |
Neg logit contributions | |
end | -0.312 |
happened | -0.306 |
OM | -0.265 |
ABC | -0.261 |
time | -0.260 |
Token | have |
Token ID | 423 |
Feature activation | -3.38e-03 |
Pos logit contributions | |
rily | +0.123 |
grate | +0.112 |
away | +0.107 |
oc | +0.106 |
mer | +0.105 |
Neg logit contributions | |
end | -0.086 |
happened | -0.085 |
done | -0.083 |
been | -0.082 |
something | -0.077 |
Token | fun |
Token ID | 1257 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
grate | +0.007 |
rily | +0.006 |
away | +0.006 |
eyed | +0.006 |
active | +0.006 |
Neg logit contributions | |
happened | -0.003 |
OM | -0.003 |
end | -0.003 |
ABC | -0.003 |
OO | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
grate | +0.046 |
rily | +0.043 |
away | +0.040 |
eyed | +0.040 |
Follow | +0.039 |
Neg logit contributions | |
happened | -0.021 |
end | -0.019 |
time | -0.017 |
been | -0.016 |
is | -0.015 |
Token | but |
Token ID | 475 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
end | +0.023 |
happened | +0.023 |
been | +0.021 |
time | +0.021 |
best | +0.020 |
Neg logit contributions | |
grate | -0.041 |
away | -0.039 |
eyed | -0.039 |
rily | -0.038 |
mer | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
OM | +0.056 |
happened | +0.053 |
ABC | +0.048 |
end | +0.047 |
consequences | +0.046 |
Neg logit contributions | |
grate | -0.063 |
away | -0.056 |
eyed | -0.055 |
rily | -0.053 |
active | -0.050 |
Token | She |
Token ID | 1375 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
happened | +0.042 |
OM | +0.041 |
end | +0.040 |
ABC | +0.039 |
consequences | +0.037 |
Neg logit contributions | |
grate | -0.029 |
rily | -0.024 |
away | -0.023 |
eyed | -0.023 |
freely | -0.022 |
Token | was |
Token ID | 373 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
grate | +0.204 |
active | +0.193 |
eyed | +0.174 |
away | +0.158 |
slur | +0.157 |
Neg logit contributions | |
fault | -0.292 |
OM | -0.286 |
reating | -0.262 |
consequences | -0.258 |
omm | -0.258 |
Token | a |
Token ID | 257 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
happened | +0.205 |
OM | +0.194 |
end | +0.179 |
ABC | +0.162 |
consequences | +0.148 |
Neg logit contributions | |
grate | -0.462 |
rily | -0.405 |
away | -0.397 |
eyed | -0.396 |
active | -0.386 |
Token | very |
Token ID | 845 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
end | +0.139 |
happened | +0.137 |
OM | +0.122 |
ABC | +0.120 |
time | +0.114 |
Neg logit contributions | |
grate | -0.257 |
rily | -0.247 |
eyed | -0.231 |
away | -0.225 |
freely | -0.217 |
Token | active |
Token ID | 4075 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
happened | +0.025 |
end | +0.024 |
time | +0.022 |
best | +0.022 |
done | +0.021 |
Neg logit contributions | |
grate | -0.039 |
rily | -0.038 |
away | -0.034 |
afloat | -0.034 |
eyed | -0.034 |
Token | little |
Token ID | 1310 |
Feature activation | -3.65e-02 |
Pos logit contributions | |
grate | +0.564 |
away | +0.494 |
rily | +0.490 |
eyed | +0.471 |
freely | +0.465 |
Neg logit contributions | |
happened | -0.348 |
OM | -0.328 |
end | -0.303 |
ABC | -0.289 |
consequences | -0.266 |
Token | girl |
Token ID | 2576 |
Feature activation | +3.02e-02 |
Pos logit contributions | |
grate | +0.083 |
rily | +0.078 |
eyed | +0.075 |
away | +0.073 |
greet | +0.069 |
Neg logit contributions | |
happened | -0.030 |
end | -0.029 |
OM | -0.024 |
ABC | -0.023 |
best | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
happened | +0.037 |
end | +0.036 |
OM | +0.032 |
ABC | +0.031 |
been | +0.030 |
Neg logit contributions | |
grate | -0.055 |
rily | -0.052 |
away | -0.048 |
eyed | -0.046 |
freely | -0.045 |
Token | She |
Token ID | 1375 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
happened | +0.041 |
OM | +0.040 |
end | +0.038 |
ABC | +0.037 |
OO | +0.036 |
Neg logit contributions | |
grate | -0.030 |
eyed | -0.024 |
rily | -0.024 |
away | -0.023 |
freely | -0.023 |
Token | liked |
Token ID | 8288 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
grate | +0.249 |
active | +0.213 |
eyed | +0.206 |
away | +0.193 |
freely | +0.191 |
Neg logit contributions | |
OM | -0.252 |
end | -0.230 |
fault | -0.227 |
happened | -0.227 |
ABC | -0.221 |
Token | to |
Token ID | 284 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
happened | +0.050 |
end | +0.044 |
OM | +0.042 |
been | +0.040 |
done | +0.039 |
Neg logit contributions | |
grate | -0.077 |
rily | -0.071 |
eyed | -0.067 |
away | -0.066 |
Follow | -0.064 |
Token | have |
Token ID | 423 |
Feature activation | -5.59e-03 |
Pos logit contributions | |
grate | +0.148 |
rily | +0.137 |
eyed | +0.124 |
away | +0.121 |
lived | +0.115 |
Neg logit contributions | |
end | -0.169 |
happened | -0.166 |
OM | -0.157 |
ABC | -0.146 |
been | -0.145 |
Token | fun |
Token ID | 1257 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
grate | +0.011 |
away | +0.010 |
rily | +0.010 |
eyed | +0.009 |
freely | +0.009 |
Neg logit contributions | |
OM | -0.006 |
happened | -0.006 |
end | -0.006 |
ABC | -0.005 |
OO | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | -6.71e-02 |
Pos logit contributions | |
grate | +0.041 |
rily | +0.037 |
eyed | +0.036 |
away | +0.034 |
Follow | +0.034 |
Neg logit contributions | |
happened | -0.028 |
end | -0.025 |
OM | -0.023 |
time | -0.023 |
ABC | -0.022 |
Token | be |
Token ID | 307 |
Feature activation | +8.67e-02 |
Pos logit contributions | |
grate | +0.114 |
rily | +0.104 |
away | +0.099 |
eyed | +0.096 |
freely | +0.091 |
Neg logit contributions | |
happened | -0.096 |
end | -0.088 |
OM | -0.087 |
ABC | -0.081 |
been | -0.080 |
Token | active |
Token ID | 4075 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
happened | +0.110 |
end | +0.101 |
OM | +0.096 |
been | +0.095 |
done | +0.092 |
Neg logit contributions | |
grate | -0.154 |
rily | -0.140 |
away | -0.134 |
eyed | -0.129 |
othe | -0.127 |
Token | . |
Token ID | 13 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
grate | +0.584 |
rily | +0.525 |
away | +0.521 |
eyed | +0.518 |
othe | +0.489 |
Neg logit contributions | |
happened | -0.308 |
end | -0.293 |
OM | -0.266 |
ABC | -0.249 |
time | -0.241 |
Token | The |
Token ID | 383 |
Feature activation | -9.11e-02 |
Pos logit contributions | |
happened | +0.043 |
OM | +0.042 |
end | +0.041 |
ABC | +0.039 |
consequences | +0.038 |
Neg logit contributions | |
grate | -0.027 |
rily | -0.019 |
eyed | -0.019 |
oc | -0.019 |
Follow | -0.019 |
Token | lazy |
Token ID | 16931 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
grate | +0.226 |
away | +0.201 |
rily | +0.199 |
eyed | +0.198 |
active | +0.195 |
Neg logit contributions | |
happened | -0.060 |
OM | -0.054 |
ABC | -0.041 |
end | -0.038 |
fault | -0.035 |
Token | human |
Token ID | 1692 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
grate | +0.254 |
rily | +0.217 |
away | +0.216 |
eyed | +0.210 |
freely | +0.202 |
Neg logit contributions | |
happened | -0.211 |
OM | -0.203 |
end | -0.193 |
ABC | -0.184 |
been | -0.169 |
Token | then |
Token ID | 788 |
Feature activation | +5.28e-02 |
Pos logit contributions | |
grate | +0.066 |
rily | +0.062 |
away | +0.061 |
eyed | +0.059 |
freely | +0.057 |
Neg logit contributions | |
happened | -0.023 |
end | -0.021 |
OM | -0.019 |
ABC | -0.017 |
been | -0.017 |
Token | being |
Token ID | 852 |
Feature activation | +5.81e-02 |
Pos logit contributions | |
grate | +0.212 |
rily | +0.189 |
eyed | +0.180 |
away | +0.180 |
freely | +0.170 |
Neg logit contributions | |
happened | -0.120 |
end | -0.112 |
OM | -0.108 |
ABC | -0.101 |
consequences | -0.092 |
Token | lazy |
Token ID | 16931 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
happened | +0.076 |
end | +0.068 |
done | +0.067 |
been | +0.067 |
best | +0.066 |
Neg logit contributions | |
grate | -0.103 |
rily | -0.096 |
away | -0.089 |
eyed | -0.086 |
othe | -0.085 |
Token | to |
Token ID | 284 |
Feature activation | -9.99e-02 |
Pos logit contributions | |
grate | +0.268 |
rily | +0.236 |
away | +0.230 |
eyed | +0.227 |
freely | +0.214 |
Neg logit contributions | |
happened | -0.191 |
OM | -0.175 |
end | -0.175 |
ABC | -0.162 |
consequences | -0.149 |
Token | being |
Token ID | 852 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
grate | +0.182 |
rily | +0.175 |
mer | +0.162 |
eyed | +0.159 |
away | +0.156 |
Neg logit contributions | |
end | -0.127 |
happened | -0.126 |
ABC | -0.111 |
OM | -0.110 |
time | -0.110 |
Token | an |
Token ID | 281 |
Feature activation | +5.61e-02 |
Pos logit contributions | |
happened | +0.077 |
done | +0.068 |
been | +0.067 |
best | +0.064 |
a | +0.064 |
Neg logit contributions | |
grate | -0.106 |
rily | -0.101 |
away | -0.090 |
eyed | -0.089 |
freely | -0.088 |
Token | active |
Token ID | 4075 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
happened | +0.090 |
OM | +0.086 |
end | +0.083 |
ABC | +0.079 |
consequences | +0.073 |
Neg logit contributions | |
grate | -0.085 |
rily | -0.072 |
away | -0.071 |
eyed | -0.068 |
freely | -0.065 |
Token | and |
Token ID | 290 |
Feature activation | -6.87e-02 |
Pos logit contributions | |
grate | +0.572 |
rily | +0.512 |
eyed | +0.491 |
away | +0.487 |
freely | +0.462 |
Neg logit contributions | |
happened | -0.344 |
end | -0.328 |
OM | -0.311 |
ABC | -0.297 |
time | -0.272 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
rily | +0.123 |
grate | +0.120 |
away | +0.109 |
oc | +0.101 |
eyed | +0.101 |
Neg logit contributions | |
happened | -0.096 |
end | -0.084 |
been | -0.084 |
done | -0.082 |
OM | -0.077 |
Token | pet |
Token ID | 4273 |
Feature activation | -8.68e-02 |
Pos logit contributions | |
grate | +0.219 |
rily | +0.198 |
eyed | +0.192 |
away | +0.188 |
Follow | +0.182 |
Neg logit contributions | |
happened | -0.132 |
end | -0.118 |
OM | -0.110 |
ABC | -0.109 |
consequences | -0.100 |
Token | . |
Token ID | 13 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
grate | +0.167 |
rily | +0.150 |
eyed | +0.145 |
away | +0.145 |
active | +0.137 |
Neg logit contributions | |
happened | -0.102 |
end | -0.093 |
OM | -0.090 |
ABC | -0.086 |
consequences | -0.077 |
Token | He |
Token ID | 679 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
happened | +0.037 |
OM | +0.035 |
end | +0.035 |
ABC | +0.033 |
consequences | +0.032 |
Neg logit contributions | |
grate | -0.020 |
rily | -0.015 |
away | -0.014 |
eyed | -0.014 |
freely | -0.013 |
Token | Jack |
Token ID | 3619 |
Feature activation | -4.20e-02 |
Pos logit contributions | |
happened | +0.035 |
end | +0.033 |
OM | +0.033 |
time | +0.030 |
ABC | +0.030 |
Neg logit contributions | |
grate | -0.035 |
rily | -0.032 |
away | -0.031 |
eyed | -0.031 |
lived | -0.027 |
Token | was |
Token ID | 373 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
grate | +0.087 |
rily | +0.077 |
away | +0.076 |
eyed | +0.075 |
freely | +0.072 |
Neg logit contributions | |
happened | -0.043 |
OM | -0.041 |
end | -0.039 |
ABC | -0.036 |
consequences | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
happened | +0.173 |
OM | +0.154 |
end | +0.151 |
ABC | +0.135 |
been | +0.115 |
Neg logit contributions | |
grate | -0.501 |
rily | -0.455 |
away | -0.450 |
eyed | -0.438 |
freely | -0.426 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
happened | +0.150 |
end | +0.147 |
ABC | +0.129 |
OM | +0.129 |
time | +0.123 |
Neg logit contributions | |
grate | -0.256 |
rily | -0.250 |
eyed | -0.228 |
away | -0.221 |
freely | -0.215 |
Token | and |
Token ID | 290 |
Feature activation | -6.34e-02 |
Pos logit contributions | |
grate | +0.209 |
away | +0.182 |
rily | +0.181 |
eyed | +0.175 |
freely | +0.171 |
Neg logit contributions | |
happened | -0.121 |
OM | -0.116 |
end | -0.108 |
ABC | -0.102 |
consequences | -0.095 |
Token | active |
Token ID | 4075 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
grate | +0.111 |
rily | +0.101 |
away | +0.097 |
eyed | +0.093 |
freely | +0.090 |
Neg logit contributions | |
happened | -0.088 |
OM | -0.080 |
end | -0.079 |
ABC | -0.074 |
been | -0.070 |
Token | child |
Token ID | 1200 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
grate | +0.566 |
away | +0.499 |
rily | +0.487 |
eyed | +0.473 |
active | +0.468 |
Neg logit contributions | |
happened | -0.342 |
OM | -0.322 |
end | -0.294 |
ABC | -0.279 |
done | -0.258 |
Token | who |
Token ID | 508 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
happened | +0.042 |
end | +0.040 |
OM | +0.036 |
ABC | +0.035 |
been | +0.032 |
Neg logit contributions | |
grate | -0.085 |
rily | -0.078 |
away | -0.075 |
eyed | -0.072 |
freely | -0.071 |
Token | loved |
Token ID | 6151 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
grate | +0.260 |
rily | +0.249 |
away | +0.232 |
eyed | +0.221 |
freely | +0.219 |
Neg logit contributions | |
happened | -0.119 |
end | -0.102 |
OM | -0.091 |
ABC | -0.091 |
been | -0.089 |
Token | to |
Token ID | 284 |
Feature activation | -9.78e-02 |
Pos logit contributions | |
grate | +0.043 |
rily | +0.038 |
away | +0.036 |
eyed | +0.036 |
freely | +0.034 |
Neg logit contributions | |
happened | -0.035 |
OM | -0.031 |
end | -0.031 |
ABC | -0.030 |
been | -0.028 |
Token | play |
Token ID | 711 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
grate | +0.157 |
rily | +0.152 |
away | +0.136 |
eyed | +0.132 |
mer | +0.131 |
Neg logit contributions | |
happened | -0.148 |
end | -0.145 |
OM | -0.137 |
ABC | -0.136 |
time | -0.126 |
Token | lazy |
Token ID | 16931 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
happened | +0.074 |
OM | +0.069 |
end | +0.068 |
ABC | +0.064 |
consequences | +0.058 |
Neg logit contributions | |
grate | -0.103 |
rily | -0.090 |
away | -0.088 |
eyed | -0.087 |
freely | -0.083 |
Token | robot |
Token ID | 9379 |
Feature activation | -8.53e-03 |
Pos logit contributions | |
grate | +0.289 |
rily | +0.257 |
away | +0.251 |
eyed | +0.248 |
freely | +0.237 |
Neg logit contributions | |
happened | -0.168 |
OM | -0.155 |
end | -0.153 |
ABC | -0.141 |
consequences | -0.127 |
Token | had |
Token ID | 550 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
grate | +0.021 |
rily | +0.020 |
away | +0.020 |
eyed | +0.019 |
freely | +0.018 |
Neg logit contributions | |
happened | -0.005 |
end | -0.004 |
OM | -0.004 |
ABC | -0.003 |
been | -0.003 |
Token | become |
Token ID | 1716 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
OM | +0.144 |
happened | +0.141 |
end | +0.137 |
ABC | +0.128 |
OO | +0.115 |
Neg logit contributions | |
grate | -0.276 |
away | -0.241 |
rily | -0.239 |
eyed | -0.235 |
freely | -0.228 |
Token | so |
Token ID | 523 |
Feature activation | +5.36e-02 |
Pos logit contributions | |
happened | +0.112 |
end | +0.099 |
OM | +0.089 |
ABC | +0.088 |
been | +0.086 |
Neg logit contributions | |
grate | -0.206 |
rily | -0.184 |
eyed | -0.181 |
away | -0.180 |
Follow | -0.175 |
Token | active |
Token ID | 4075 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
happened | +0.068 |
been | +0.056 |
best | +0.053 |
end | +0.053 |
something | +0.052 |
Neg logit contributions | |
grate | -0.104 |
rily | -0.104 |
oc | -0.096 |
slur | -0.088 |
afloat | -0.087 |
Token | . |
Token ID | 13 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
grate | +0.573 |
rily | +0.507 |
eyed | +0.500 |
away | +0.488 |
Follow | +0.472 |
Neg logit contributions | |
happened | -0.353 |
end | -0.321 |
OM | -0.287 |
time | -0.269 |
been | -0.267 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +9.61e-02 |
Pos logit contributions | |
happened | +0.030 |
OM | +0.029 |
end | +0.028 |
ABC | +0.026 |
consequences | +0.025 |
Neg logit contributions | |
grate | -0.031 |
rily | -0.026 |
away | -0.025 |
eyed | -0.025 |
freely | -0.024 |
Token | Once |
Token ID | 7454 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
OM | +0.145 |
happened | +0.135 |
end | +0.133 |
ABC | +0.132 |
fault | +0.131 |
Neg logit contributions | |
grate | -0.141 |
freely | -0.126 |
rily | -0.118 |
eyed | -0.118 |
away | -0.117 |
Token | upon |
Token ID | 2402 |
Feature activation | +6.34e-02 |
Pos logit contributions | |
OM | +0.026 |
happened | +0.024 |
ABC | +0.023 |
consequences | +0.023 |
end | +0.022 |
Neg logit contributions | |
grate | -0.035 |
rily | -0.031 |
away | -0.031 |
freely | -0.030 |
eyed | -0.030 |
Token | a |
Token ID | 257 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
omm | +0.086 |
OM | +0.086 |
fault | +0.078 |
problem | +0.077 |
anger | +0.074 |
Neg logit contributions | |
grate | -0.105 |
eyed | -0.092 |
active | -0.091 |
away | -0.090 |
freely | -0.090 |
Token | it |
Token ID | 340 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
grate | +0.262 |
rily | +0.254 |
away | +0.241 |
eyed | +0.235 |
mer | +0.230 |
Neg logit contributions | |
happened | -0.146 |
end | -0.131 |
OM | -0.113 |
been | -0.112 |
done | -0.107 |
Token | 's |
Token ID | 338 |
Feature activation | +5.22e-02 |
Pos logit contributions | |
happened | +0.095 |
end | +0.082 |
OM | +0.079 |
ABC | +0.071 |
consequences | +0.061 |
Neg logit contributions | |
grate | -0.287 |
rily | -0.262 |
away | -0.260 |
eyed | -0.254 |
freely | -0.243 |
Token | important |
Token ID | 1593 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
happened | +0.062 |
end | +0.057 |
OM | +0.057 |
ABC | +0.052 |
consequences | +0.048 |
Neg logit contributions | |
grate | -0.101 |
rily | -0.089 |
away | -0.088 |
eyed | -0.086 |
freely | -0.082 |
Token | to |
Token ID | 284 |
Feature activation | -8.95e-02 |
Pos logit contributions | |
grate | +0.133 |
away | +0.108 |
eyed | +0.106 |
rily | +0.101 |
freely | +0.101 |
Neg logit contributions | |
OM | -0.186 |
happened | -0.173 |
end | -0.170 |
ABC | -0.168 |
CA | -0.162 |
Token | stay |
Token ID | 2652 |
Feature activation | -2.77e-03 |
Pos logit contributions | |
grate | +0.172 |
rily | +0.166 |
away | +0.149 |
mer | +0.147 |
oc | +0.144 |
Neg logit contributions | |
end | -0.114 |
happened | -0.104 |
OM | -0.099 |
been | -0.094 |
time | -0.090 |
Token | positive |
Token ID | 3967 |
Feature activation | -2.73e-01 |
Pos logit contributions | |
grate | +0.004 |
rily | +0.004 |
oc | +0.004 |
slur | +0.004 |
Follow | +0.004 |
Neg logit contributions | |
something | -0.004 |
end | -0.004 |
a | -0.004 |
best | -0.004 |
time | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | -6.79e-02 |
Pos logit contributions | |
grate | +0.545 |
rily | +0.491 |
eyed | +0.467 |
away | +0.458 |
greet | +0.449 |
Neg logit contributions | |
happened | -0.307 |
end | -0.295 |
consequences | -0.257 |
OM | -0.252 |
ABC | -0.250 |
Token | find |
Token ID | 1064 |
Feature activation | +9.09e-03 |
Pos logit contributions | |
rily | +0.121 |
grate | +0.114 |
mer | +0.100 |
oc | +0.098 |
away | +0.097 |
Neg logit contributions | |
end | -0.097 |
happened | -0.092 |
done | -0.085 |
been | -0.085 |
is | -0.084 |
Token | a |
Token ID | 257 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
happened | +0.008 |
end | +0.008 |
consequences | +0.006 |
OM | +0.006 |
ABC | +0.006 |
Neg logit contributions | |
grate | -0.020 |
rily | -0.018 |
eyed | -0.018 |
away | -0.017 |
mer | -0.017 |
Token | solution |
Token ID | 4610 |
Feature activation | -4.17e-02 |
Pos logit contributions | |
happened | +0.131 |
end | +0.130 |
OM | +0.117 |
ABC | +0.109 |
time | +0.104 |
Neg logit contributions | |
grate | -0.287 |
rily | -0.266 |
eyed | -0.250 |
away | -0.247 |
Follow | -0.238 |
Token | . |
Token ID | 13 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
grate | +0.083 |
rily | +0.074 |
away | +0.070 |
eyed | +0.070 |
lived | +0.067 |
Neg logit contributions | |
happened | -0.047 |
end | -0.044 |
OM | -0.040 |
ABC | -0.038 |
been | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | -7.00e-02 |
Pos logit contributions | |
grate | +0.014 |
rily | +0.013 |
away | +0.012 |
eyed | +0.012 |
freely | +0.012 |
Neg logit contributions | |
happened | -0.009 |
end | -0.008 |
OM | -0.008 |
ABC | -0.008 |
been | -0.007 |
Token | grow |
Token ID | 1663 |
Feature activation | -3.09e-02 |
Pos logit contributions | |
rily | +0.134 |
grate | +0.130 |
away | +0.124 |
mer | +0.118 |
oc | +0.115 |
Neg logit contributions | |
happened | -0.088 |
end | -0.077 |
been | -0.076 |
was | -0.075 |
done | -0.071 |
Token | in |
Token ID | 287 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
grate | +0.056 |
rily | +0.048 |
eyed | +0.048 |
away | +0.048 |
Follow | +0.045 |
Neg logit contributions | |
happened | -0.040 |
end | -0.036 |
OM | -0.034 |
ABC | -0.033 |
consequences | -0.031 |
Token | such |
Token ID | 884 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
grate | +0.287 |
rily | +0.271 |
eyed | +0.257 |
away | +0.253 |
Follow | +0.249 |
Neg logit contributions | |
end | -0.171 |
happened | -0.159 |
time | -0.145 |
been | -0.136 |
OM | -0.136 |
Token | a |
Token ID | 257 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
grate | +0.046 |
rily | +0.041 |
away | +0.041 |
eyed | +0.040 |
freely | +0.039 |
Neg logit contributions | |
happened | -0.013 |
OM | -0.013 |
end | -0.011 |
ABC | -0.010 |
OO | -0.008 |
Token | positive |
Token ID | 3967 |
Feature activation | -2.72e-01 |
Pos logit contributions | |
end | +0.146 |
happened | +0.138 |
time | +0.123 |
OM | +0.123 |
ABC | +0.119 |
Neg logit contributions | |
grate | -0.244 |
rily | -0.235 |
eyed | -0.216 |
away | -0.208 |
Follow | -0.202 |
Token | environment |
Token ID | 2858 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
grate | +0.600 |
rily | +0.548 |
eyed | +0.518 |
away | +0.506 |
Follow | +0.499 |
Neg logit contributions | |
end | -0.255 |
happened | -0.251 |
OM | -0.216 |
time | -0.208 |
ABC | -0.208 |
Token | . |
Token ID | 13 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
grate | +0.027 |
rily | +0.025 |
away | +0.024 |
eyed | +0.023 |
Follow | +0.023 |
Neg logit contributions | |
happened | -0.013 |
end | -0.012 |
OM | -0.010 |
ABC | -0.010 |
time | -0.009 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +9.19e-02 |
Pos logit contributions | |
happened | +0.019 |
end | +0.018 |
OM | +0.018 |
ABC | +0.016 |
been | +0.015 |
Neg logit contributions | |
grate | -0.037 |
away | -0.034 |
rily | -0.034 |
eyed | -0.031 |
active | -0.031 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
OM | +0.126 |
happened | +0.118 |
fault | +0.116 |
end | +0.115 |
ABC | +0.114 |
Neg logit contributions | |
grate | -0.145 |
freely | -0.130 |
rily | -0.127 |
eyed | -0.124 |
away | -0.121 |
Token | upon |
Token ID | 2402 |
Feature activation | +5.85e-02 |
Pos logit contributions | |
OM | +0.019 |
happened | +0.017 |
ABC | +0.017 |
consequences | +0.016 |
end | +0.016 |
Neg logit contributions | |
grate | -0.026 |
rily | -0.023 |
away | -0.023 |
freely | -0.022 |
eyed | -0.022 |
Token | encouraged |
Token ID | 10085 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
grate | +0.303 |
rily | +0.276 |
away | +0.273 |
eyed | +0.257 |
freely | +0.252 |
Neg logit contributions | |
happened | -0.233 |
end | -0.209 |
OM | -0.206 |
ABC | -0.190 |
been | -0.183 |
Token | Ros |
Token ID | 10018 |
Feature activation | -6.90e-02 |
Pos logit contributions | |
grate | +0.229 |
rily | +0.215 |
eyed | +0.199 |
away | +0.194 |
active | +0.188 |
Neg logit contributions | |
happened | -0.194 |
end | -0.184 |
OM | -0.169 |
ABC | -0.166 |
time | -0.162 |
Token | y |
Token ID | 88 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
grate | +0.098 |
rily | +0.088 |
away | +0.085 |
eyed | +0.079 |
freely | +0.075 |
Neg logit contributions | |
happened | -0.114 |
OM | -0.114 |
end | -0.104 |
ABC | -0.101 |
CA | -0.100 |
Token | to |
Token ID | 284 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
happened | +0.051 |
end | +0.047 |
OM | +0.043 |
ABC | +0.043 |
been | +0.041 |
Neg logit contributions | |
grate | -0.071 |
rily | -0.066 |
away | -0.063 |
freely | -0.060 |
eyed | -0.059 |
Token | stay |
Token ID | 2652 |
Feature activation | -4.45e-03 |
Pos logit contributions | |
grate | +0.175 |
rily | +0.174 |
oc | +0.158 |
away | +0.157 |
mer | +0.155 |
Neg logit contributions | |
end | -0.141 |
OM | -0.134 |
happened | -0.132 |
been | -0.120 |
time | -0.119 |
Token | positive |
Token ID | 3967 |
Feature activation | -2.71e-01 |
Pos logit contributions | |
oc | +0.007 |
grate | +0.007 |
Follow | +0.007 |
lived | +0.007 |
rily | +0.007 |
Neg logit contributions | |
something | -0.006 |
a | -0.005 |
end | -0.005 |
best | -0.005 |
done | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | -6.69e-02 |
Pos logit contributions | |
grate | +0.546 |
rily | +0.505 |
eyed | +0.499 |
Follow | +0.482 |
greet | +0.481 |
Neg logit contributions | |
end | -0.265 |
happened | -0.252 |
consequences | -0.230 |
ABC | -0.220 |
time | -0.220 |
Token | try |
Token ID | 1949 |
Feature activation | +1.56e-03 |
Pos logit contributions | |
rily | +0.120 |
grate | +0.110 |
away | +0.109 |
mer | +0.105 |
oc | +0.104 |
Neg logit contributions | |
happened | -0.091 |
end | -0.083 |
was | -0.082 |
done | -0.080 |
been | -0.078 |
Token | her |
Token ID | 607 |
Feature activation | +5.46e-02 |
Pos logit contributions | |
happened | +0.002 |
end | +0.002 |
OM | +0.002 |
ABC | +0.002 |
fault | +0.002 |
Neg logit contributions | |
grate | -0.003 |
rily | -0.003 |
away | -0.002 |
eyed | -0.002 |
Follow | -0.002 |
Token | best |
Token ID | 1266 |
Feature activation | +9.36e-03 |
Pos logit contributions | |
happened | +0.055 |
OM | +0.053 |
end | +0.048 |
ABC | +0.043 |
CA | +0.041 |
Neg logit contributions | |
grate | -0.114 |
rily | -0.100 |
away | -0.099 |
eyed | -0.098 |
active | -0.095 |
Token | to |
Token ID | 284 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
happened | +0.012 |
end | +0.011 |
OM | +0.011 |
ABC | +0.010 |
consequences | +0.009 |
Neg logit contributions | |
grate | -0.017 |
rily | -0.015 |
away | -0.015 |
eyed | -0.015 |
freely | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
happened | +0.020 |
end | +0.020 |
OM | +0.018 |
ABC | +0.017 |
time | +0.016 |
Neg logit contributions | |
grate | -0.038 |
rily | -0.035 |
eyed | -0.033 |
away | -0.033 |
Follow | -0.031 |
Token | He |
Token ID | 679 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
happened | +0.032 |
OM | +0.031 |
end | +0.030 |
ABC | +0.029 |
been | +0.027 |
Neg logit contributions | |
grate | -0.039 |
away | -0.036 |
rily | -0.035 |
eyed | -0.033 |
active | -0.032 |
Token | tried |
Token ID | 3088 |
Feature activation | +9.49e-02 |
Pos logit contributions | |
grate | +0.357 |
rily | +0.346 |
away | +0.342 |
eyed | +0.327 |
freely | +0.314 |
Neg logit contributions | |
happened | -0.171 |
end | -0.151 |
was | -0.145 |
wasn | -0.137 |
been | -0.133 |
Token | to |
Token ID | 284 |
Feature activation | -9.27e-02 |
Pos logit contributions | |
happened | +0.112 |
end | +0.109 |
ABC | +0.102 |
fault | +0.096 |
time | +0.095 |
Neg logit contributions | |
grate | -0.176 |
rily | -0.166 |
away | -0.157 |
mer | -0.155 |
eyed | -0.155 |
Token | stay |
Token ID | 2652 |
Feature activation | -2.58e-04 |
Pos logit contributions | |
rily | +0.152 |
grate | +0.151 |
oc | +0.141 |
away | +0.139 |
Follow | +0.138 |
Neg logit contributions | |
end | -0.143 |
been | -0.111 |
time | -0.110 |
bottom | -0.110 |
happened | -0.105 |
Token | positive |
Token ID | 3967 |
Feature activation | -2.71e-01 |
Pos logit contributions | |
Follow | +0.000 |
slur | +0.000 |
oc | +0.000 |
Here | +0.000 |
amb | +0.000 |
Neg logit contributions | |
something | -0.000 |
a | -0.000 |
best | -0.000 |
time | -0.000 |
end | -0.000 |
Token | and |
Token ID | 290 |
Feature activation | -6.51e-02 |
Pos logit contributions | |
grate | +0.551 |
rily | +0.509 |
eyed | +0.507 |
greet | +0.480 |
Follow | +0.479 |
Neg logit contributions | |
end | -0.267 |
happened | -0.255 |
consequences | -0.234 |
ABC | -0.226 |
time | -0.220 |
Token | kept |
Token ID | 4030 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
rily | +0.119 |
grate | +0.106 |
amb | +0.102 |
away | +0.099 |
oing | +0.098 |
Neg logit contributions | |
end | -0.091 |
happened | -0.083 |
done | -0.080 |
been | -0.078 |
was | -0.077 |
Token | thinking |
Token ID | 3612 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
happened | +0.027 |
end | +0.025 |
time | +0.025 |
a | +0.024 |
something | +0.024 |
Neg logit contributions | |
rily | -0.051 |
grate | -0.051 |
oc | -0.049 |
mer | -0.049 |
away | -0.049 |
Token | that |
Token ID | 326 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
grate | +0.075 |
rily | +0.070 |
eyed | +0.069 |
away | +0.067 |
Follow | +0.065 |
Neg logit contributions | |
happened | -0.049 |
end | -0.047 |
ABC | -0.041 |
time | -0.040 |
OM | -0.039 |
Token | he |
Token ID | 339 |
Feature activation | -7.75e-02 |
Pos logit contributions | |
grate | +0.049 |
rily | +0.045 |
eyed | +0.045 |
away | +0.044 |
Follow | +0.040 |
Neg logit contributions | |
happened | -0.035 |
end | -0.033 |
ABC | -0.030 |
OM | -0.029 |
time | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
grate | +0.217 |
rily | +0.194 |
away | +0.188 |
eyed | +0.187 |
active | +0.178 |
Neg logit contributions | |
happened | -0.139 |
OM | -0.122 |
end | -0.122 |
ABC | -0.109 |
been | -0.105 |
Token | It |
Token ID | 632 |
Feature activation | -4.42e-02 |
Pos logit contributions | |
happened | +0.027 |
OM | +0.026 |
end | +0.025 |
ABC | +0.024 |
been | +0.023 |
Neg logit contributions | |
grate | -0.028 |
away | -0.025 |
rily | -0.024 |
eyed | -0.023 |
active | -0.022 |
Token | could |
Token ID | 714 |
Feature activation | +3.64e-02 |
Pos logit contributions | |
grate | +0.094 |
rily | +0.091 |
away | +0.088 |
eyed | +0.084 |
othe | +0.081 |
Neg logit contributions | |
happened | -0.042 |
end | -0.036 |
was | -0.033 |
wasn | -0.032 |
been | -0.032 |
Token | feel |
Token ID | 1254 |
Feature activation | +2.99e-02 |
Pos logit contributions | |
end | +0.046 |
happened | +0.040 |
OM | +0.038 |
bottom | +0.035 |
been | +0.035 |
Neg logit contributions | |
grate | -0.066 |
rily | -0.065 |
away | -0.063 |
oc | -0.061 |
mer | -0.061 |
Token | the |
Token ID | 262 |
Feature activation | +7.58e-02 |
Pos logit contributions | |
happened | +0.036 |
end | +0.032 |
time | +0.028 |
been | +0.028 |
something | +0.027 |
Neg logit contributions | |
grate | -0.058 |
rily | -0.053 |
eyed | -0.050 |
away | -0.050 |
oc | -0.049 |
Token | positive |
Token ID | 3967 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
end | +0.083 |
happened | +0.083 |
OM | +0.076 |
ABC | +0.071 |
consequences | +0.067 |
Neg logit contributions | |
grate | -0.154 |
rily | -0.139 |
away | -0.134 |
eyed | -0.133 |
freely | -0.126 |
Token | energy |
Token ID | 2568 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
grate | +0.557 |
rily | +0.493 |
away | +0.487 |
eyed | +0.477 |
freely | +0.460 |
Neg logit contributions | |
happened | -0.285 |
OM | -0.266 |
end | -0.256 |
ABC | -0.236 |
consequences | -0.209 |
Token | in |
Token ID | 287 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
happened | +0.021 |
end | +0.020 |
been | +0.017 |
time | +0.017 |
is | +0.016 |
Neg logit contributions | |
grate | -0.032 |
rily | -0.028 |
eyed | -0.027 |
Follow | -0.027 |
away | -0.026 |
Token | the |
Token ID | 262 |
Feature activation | +7.52e-02 |
Pos logit contributions | |
grate | +0.225 |
rily | +0.194 |
eyed | +0.183 |
Follow | +0.180 |
lived | +0.180 |
Neg logit contributions | |
end | -0.233 |
happened | -0.229 |
OM | -0.209 |
time | -0.208 |
been | -0.206 |
Token | air |
Token ID | 1633 |
Feature activation | +9.87e-02 |
Pos logit contributions | |
end | +0.066 |
happened | +0.057 |
OM | +0.051 |
time | +0.048 |
consequences | +0.046 |
Neg logit contributions | |
grate | -0.180 |
rily | -0.170 |
eyed | -0.156 |
away | -0.152 |
othe | -0.150 |
Token | and |
Token ID | 290 |
Feature activation | -6.84e-02 |
Pos logit contributions | |
happened | +0.099 |
end | +0.099 |
time | +0.083 |
OM | +0.082 |
done | +0.079 |
Neg logit contributions | |
grate | -0.204 |
rily | -0.184 |
Follow | -0.183 |
eyed | -0.183 |
away | -0.178 |
Token | difficulties |
Token ID | 13156 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
grate | +0.357 |
rily | +0.339 |
away | +0.310 |
eyed | +0.308 |
mer | +0.302 |
Neg logit contributions | |
happened | -0.181 |
end | -0.173 |
consequences | -0.153 |
been | -0.151 |
time | -0.146 |
Token | , |
Token ID | 11 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
grate | +0.027 |
rily | +0.024 |
eyed | +0.023 |
away | +0.023 |
active | +0.022 |
Neg logit contributions | |
happened | -0.012 |
end | -0.011 |
OM | -0.010 |
ABC | -0.009 |
is | -0.009 |
Token | courage |
Token ID | 11917 |
Feature activation | +5.51e-03 |
Pos logit contributions | |
happened | +0.024 |
end | +0.024 |
something | +0.023 |
is | +0.023 |
best | +0.022 |
Neg logit contributions | |
grate | -0.052 |
rily | -0.050 |
eyed | -0.047 |
away | -0.045 |
mer | -0.043 |
Token | and |
Token ID | 290 |
Feature activation | -6.60e-02 |
Pos logit contributions | |
happened | +0.006 |
end | +0.005 |
OM | +0.005 |
is | +0.005 |
been | +0.005 |
Neg logit contributions | |
grate | -0.011 |
rily | -0.010 |
away | -0.010 |
eyed | -0.010 |
active | -0.009 |
Token | a |
Token ID | 257 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
rily | +0.119 |
grate | +0.116 |
away | +0.100 |
eyed | +0.100 |
mer | +0.099 |
Neg logit contributions | |
end | -0.089 |
happened | -0.088 |
been | -0.080 |
consequences | -0.078 |
done | -0.076 |
Token | positive |
Token ID | 3967 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
end | +0.163 |
happened | +0.156 |
OM | +0.142 |
problem | +0.137 |
time | +0.137 |
Neg logit contributions | |
grate | -0.237 |
rily | -0.232 |
eyed | -0.207 |
away | -0.206 |
freely | -0.195 |
Token | attitude |
Token ID | 9408 |
Feature activation | +5.63e-03 |
Pos logit contributions | |
grate | +0.542 |
rily | +0.491 |
eyed | +0.460 |
away | +0.455 |
freely | +0.435 |
Neg logit contributions | |
happened | -0.315 |
end | -0.300 |
OM | -0.267 |
time | -0.254 |
consequences | -0.251 |
Token | can |
Token ID | 460 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
happened | +0.006 |
end | +0.005 |
OM | +0.005 |
ABC | +0.004 |
is | +0.004 |
Neg logit contributions | |
grate | -0.012 |
rily | -0.011 |
away | -0.011 |
eyed | -0.010 |
active | -0.010 |
Token | help |
Token ID | 1037 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
end | +0.044 |
happened | +0.039 |
been | +0.035 |
is | +0.035 |
OM | +0.034 |
Neg logit contributions | |
grate | -0.069 |
rily | -0.068 |
away | -0.061 |
active | -0.059 |
eyed | -0.059 |
Token | us |
Token ID | 514 |
Feature activation | +5.68e-02 |
Pos logit contributions | |
grate | +0.225 |
rily | +0.209 |
eyed | +0.193 |
slur | +0.183 |
away | +0.182 |
Neg logit contributions | |
end | -0.129 |
happened | -0.127 |
is | -0.110 |
time | -0.109 |
consequences | -0.106 |
Token | make |
Token ID | 787 |
Feature activation | -5.74e-02 |
Pos logit contributions | |
end | +0.071 |
happened | +0.061 |
been | +0.056 |
is | +0.056 |
time | +0.055 |
Neg logit contributions | |
grate | -0.107 |
rily | -0.106 |
away | -0.091 |
oc | -0.091 |
slur | -0.090 |
Token | . |
Token ID | 13 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
grate | +0.113 |
away | +0.097 |
freely | +0.095 |
eyed | +0.094 |
rily | +0.093 |
Neg logit contributions | |
OM | -0.097 |
ABC | -0.088 |
omm | -0.088 |
happened | -0.086 |
OO | -0.083 |
Token |
|
Token ID | 198 |
Feature activation | +7.76e-03 |
Pos logit contributions | |
happened | +0.022 |
OM | +0.021 |
end | +0.020 |
ABC | +0.019 |
been | +0.017 |
Neg logit contributions | |
grate | -0.034 |
rily | -0.030 |
away | -0.030 |
eyed | -0.029 |
freely | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | +7.52e-03 |
Pos logit contributions | |
happened | +0.011 |
OM | +0.011 |
end | +0.010 |
ABC | +0.010 |
consequences | +0.009 |
Neg logit contributions | |
grate | -0.013 |
eyed | -0.011 |
rily | -0.011 |
away | -0.011 |
freely | -0.010 |
Token | " |
Token ID | 1 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
happened | +0.010 |
OM | +0.010 |
end | +0.009 |
ABC | +0.009 |
consequences | +0.008 |
Neg logit contributions | |
grate | -0.013 |
rily | -0.011 |
away | -0.011 |
eyed | -0.011 |
freely | -0.010 |
Token | Where |
Token ID | 8496 |
Feature activation | +3.71e-02 |
Pos logit contributions | |
happened | +0.020 |
OM | +0.019 |
end | +0.018 |
ABC | +0.017 |
consequences | +0.015 |
Neg logit contributions | |
grate | -0.027 |
rily | -0.023 |
away | -0.023 |
eyed | -0.022 |
freely | -0.021 |
Token | is |
Token ID | 318 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
OM | +0.039 |
end | +0.036 |
omm | +0.036 |
End | +0.035 |
ained | +0.034 |
Neg logit contributions | |
grate | -0.067 |
Follow | -0.061 |
away | -0.056 |
active | -0.056 |
eyed | -0.056 |
Token | the |
Token ID | 262 |
Feature activation | +6.46e-02 |
Pos logit contributions | |
happened | +0.239 |
OM | +0.214 |
end | +0.214 |
ABC | +0.197 |
been | +0.184 |
Neg logit contributions | |
grate | -0.345 |
rily | -0.299 |
away | -0.293 |
eyed | -0.293 |
freely | -0.277 |
Token | cool |
Token ID | 3608 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
happened | +0.099 |
OM | +0.099 |
ABC | +0.092 |
omm | +0.089 |
ained | +0.087 |
Neg logit contributions | |
grate | -0.097 |
away | -0.086 |
rily | -0.082 |
freely | -0.076 |
positive | -0.076 |
Token | thing |
Token ID | 1517 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
grate | +0.168 |
away | +0.150 |
rily | +0.143 |
freely | +0.130 |
eyed | +0.129 |
Neg logit contributions | |
happened | -0.170 |
OM | -0.166 |
ABC | -0.160 |
omm | -0.150 |
dentist | -0.147 |
Token | ?" |
Token ID | 1701 |
Feature activation | +8.68e-02 |
Pos logit contributions | |
happened | +0.120 |
end | +0.104 |
OM | +0.104 |
ABC | +0.091 |
been | +0.082 |
Neg logit contributions | |
grate | -0.268 |
rily | -0.240 |
away | -0.235 |
eyed | -0.232 |
freely | -0.222 |
Token | Sara |
Token ID | 24799 |
Feature activation | +9.17e-02 |
Pos logit contributions | |
happened | +0.094 |
OM | +0.083 |
end | +0.082 |
ABC | +0.074 |
been | +0.067 |
Neg logit contributions | |
grate | -0.180 |
rily | -0.160 |
away | -0.158 |
eyed | -0.155 |
freely | -0.150 |
Token | her |
Token ID | 607 |
Feature activation | +5.41e-02 |
Pos logit contributions | |
grate | +0.316 |
away | +0.265 |
rily | +0.264 |
eyed | +0.255 |
freely | +0.253 |
Neg logit contributions | |
OM | -0.231 |
happened | -0.228 |
end | -0.196 |
OO | -0.194 |
ABC | -0.191 |
Token | blue |
Token ID | 4171 |
Feature activation | -8.46e-02 |
Pos logit contributions | |
happened | +0.068 |
end | +0.063 |
OM | +0.060 |
ABC | +0.060 |
time | +0.053 |
Neg logit contributions | |
grate | -0.102 |
rily | -0.090 |
away | -0.089 |
eyed | -0.088 |
freely | -0.083 |
Token | guitar |
Token ID | 10047 |
Feature activation | -5.07e-02 |
Pos logit contributions | |
grate | +0.145 |
rily | +0.125 |
away | +0.122 |
eyed | +0.118 |
freely | +0.114 |
Neg logit contributions | |
happened | -0.118 |
OM | -0.114 |
end | -0.108 |
ABC | -0.102 |
consequences | -0.095 |
Token | and |
Token ID | 290 |
Feature activation | -6.85e-02 |
Pos logit contributions | |
grate | +0.098 |
rily | +0.089 |
away | +0.087 |
eyed | +0.086 |
freely | +0.081 |
Neg logit contributions | |
happened | -0.057 |
end | -0.054 |
OM | -0.051 |
ABC | -0.048 |
time | -0.043 |
Token | she |
Token ID | 673 |
Feature activation | -6.10e-02 |
Pos logit contributions | |
grate | +0.127 |
rily | +0.110 |
away | +0.108 |
eyed | +0.106 |
freely | +0.102 |
Neg logit contributions | |
happened | -0.086 |
OM | -0.082 |
end | -0.079 |
ABC | -0.074 |
consequences | -0.068 |
Token | was |
Token ID | 373 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
grate | +0.130 |
rily | +0.110 |
eyed | +0.109 |
away | +0.108 |
freely | +0.107 |
Neg logit contributions | |
OM | -0.060 |
happened | -0.059 |
end | -0.053 |
ABC | -0.050 |
consequences | -0.049 |
Token | very |
Token ID | 845 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
happened | +0.302 |
OM | +0.291 |
end | +0.278 |
ABC | +0.266 |
consequences | +0.245 |
Neg logit contributions | |
grate | -0.345 |
rily | -0.293 |
away | -0.286 |
eyed | -0.284 |
freely | -0.271 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
happened | +0.020 |
end | +0.019 |
done | +0.018 |
best | +0.017 |
time | +0.017 |
Neg logit contributions | |
grate | -0.030 |
rily | -0.029 |
away | -0.027 |
oc | -0.026 |
eyed | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | +2.07e-02 |
Pos logit contributions | |
grate | +0.220 |
rily | +0.195 |
away | +0.192 |
eyed | +0.189 |
freely | +0.179 |
Neg logit contributions | |
happened | -0.132 |
OM | -0.120 |
end | -0.120 |
ABC | -0.109 |
been | -0.100 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +9.47e-02 |
Pos logit contributions | |
happened | +0.032 |
OM | +0.031 |
end | +0.031 |
consequences | +0.028 |
ABC | +0.027 |
Neg logit contributions | |
grate | -0.032 |
eyed | -0.027 |
rily | -0.025 |
Follow | -0.025 |
freely | -0.024 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
OM | +0.152 |
end | +0.142 |
happened | +0.140 |
ABC | +0.140 |
fault | +0.139 |
Neg logit contributions | |
grate | -0.126 |
freely | -0.113 |
eyed | -0.106 |
rily | -0.104 |
away | -0.102 |
Token |
|
Token ID | 198 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
happened | +0.019 |
OM | +0.019 |
consequences | +0.018 |
fault | +0.018 |
end | +0.017 |
Neg logit contributions | |
grate | -0.015 |
eyed | -0.013 |
away | -0.013 |
freely | -0.013 |
greet | -0.012 |
Token | One |
Token ID | 3198 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
happened | +0.019 |
OM | +0.017 |
fault | +0.017 |
ABC | +0.016 |
consequences | +0.016 |
Neg logit contributions | |
grate | -0.017 |
away | -0.015 |
eyed | -0.014 |
active | -0.013 |
rily | -0.013 |
Token | day |
Token ID | 1110 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
OM | +0.015 |
happened | +0.014 |
fault | +0.012 |
consequences | +0.012 |
ABC | +0.012 |
Neg logit contributions | |
grate | -0.032 |
away | -0.030 |
active | -0.028 |
eyed | -0.028 |
freely | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
fault | +0.251 |
OM | +0.249 |
consequences | +0.231 |
dentist | +0.227 |
ABC | +0.223 |
Neg logit contributions | |
grate | -0.169 |
away | -0.130 |
othe | -0.126 |
toward | -0.117 |
rily | -0.115 |
Token | she |
Token ID | 673 |
Feature activation | -6.55e-02 |
Pos logit contributions | |
OM | +0.030 |
fault | +0.030 |
happened | +0.029 |
consequences | +0.029 |
end | +0.028 |
Neg logit contributions | |
grate | -0.032 |
rily | -0.027 |
away | -0.026 |
tightly | -0.025 |
toward | -0.024 |
Token | was |
Token ID | 373 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
grate | +0.111 |
away | +0.097 |
eyed | +0.094 |
rily | +0.093 |
active | +0.092 |
Neg logit contributions | |
OM | -0.086 |
end | -0.081 |
happened | -0.080 |
ABC | -0.078 |
fault | -0.077 |
Token | singing |
Token ID | 13777 |
Feature activation | -8.14e-02 |
Pos logit contributions | |
OM | +0.270 |
happened | +0.269 |
end | +0.254 |
ABC | +0.233 |
consequences | +0.222 |
Neg logit contributions | |
grate | -0.367 |
eyed | -0.315 |
rily | -0.306 |
away | -0.305 |
active | -0.304 |
Token | in |
Token ID | 287 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
grate | +0.154 |
away | +0.137 |
freely | +0.129 |
rily | +0.129 |
active | +0.128 |
Neg logit contributions | |
OM | -0.102 |
happened | -0.092 |
fault | -0.091 |
omm | -0.083 |
ABC | -0.082 |
Token | her |
Token ID | 607 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
grate | +0.252 |
away | +0.218 |
rily | +0.211 |
greet | +0.204 |
freely | +0.203 |
Neg logit contributions | |
happened | -0.201 |
OM | -0.197 |
end | -0.170 |
ABC | -0.169 |
OO | -0.168 |
Token | room |
Token ID | 2119 |
Feature activation | +5.41e-02 |
Pos logit contributions | |
happened | +0.058 |
OM | +0.057 |
end | +0.051 |
ABC | +0.047 |
OO | +0.045 |
Neg logit contributions | |
grate | -0.105 |
away | -0.092 |
rily | -0.091 |
eyed | -0.088 |
active | -0.087 |
Token | . |
Token ID | 13 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
happened | +0.075 |
OM | +0.074 |
end | +0.067 |
ABC | +0.066 |
fault | +0.066 |
Neg logit contributions | |
grate | -0.089 |
away | -0.076 |
rily | -0.073 |
eyed | -0.071 |
freely | -0.070 |
Token | a |
Token ID | 257 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
OM | +0.137 |
ops | +0.125 |
fault | +0.123 |
End | +0.120 |
uch | +0.118 |
Neg logit contributions | |
grate | -0.183 |
away | -0.170 |
eyed | -0.158 |
greet | -0.156 |
closely | -0.152 |
Token | loud |
Token ID | 7812 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
happened | +0.179 |
end | +0.168 |
OM | +0.165 |
ABC | +0.156 |
CA | +0.140 |
Neg logit contributions | |
grate | -0.226 |
rily | -0.204 |
eyed | -0.193 |
away | -0.190 |
freely | -0.182 |
Token | noise |
Token ID | 7838 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
grate | +0.333 |
rily | +0.295 |
away | +0.288 |
eyed | +0.283 |
freely | +0.272 |
Neg logit contributions | |
happened | -0.198 |
OM | -0.184 |
end | -0.180 |
ABC | -0.167 |
been | -0.148 |
Token | . |
Token ID | 13 |
Feature activation | +2.42e-02 |
Pos logit contributions | |
happened | +0.020 |
OM | +0.019 |
end | +0.018 |
ABC | +0.017 |
consequences | +0.015 |
Neg logit contributions | |
grate | -0.028 |
rily | -0.025 |
away | -0.024 |
eyed | -0.024 |
freely | -0.023 |
Token | It |
Token ID | 632 |
Feature activation | -4.39e-02 |
Pos logit contributions | |
end | +0.048 |
OM | +0.047 |
happened | +0.047 |
fault | +0.045 |
consequences | +0.045 |
Neg logit contributions | |
grate | -0.025 |
rily | -0.018 |
eyed | -0.018 |
Follow | -0.018 |
freely | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
grate | +0.100 |
away | +0.085 |
rily | +0.085 |
eyed | +0.085 |
freely | +0.084 |
Neg logit contributions | |
OM | -0.034 |
fault | -0.033 |
ABC | -0.032 |
end | -0.028 |
happened | -0.027 |
Token | a |
Token ID | 257 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
OM | +0.272 |
ops | +0.253 |
ABC | +0.233 |
uch | +0.222 |
OO | +0.222 |
Neg logit contributions | |
grate | -0.384 |
active | -0.318 |
inviting | -0.317 |
freely | -0.311 |
away | -0.308 |
Token | dog |
Token ID | 3290 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
happened | +0.164 |
OM | +0.154 |
end | +0.150 |
ABC | +0.140 |
consequences | +0.127 |
Neg logit contributions | |
grate | -0.239 |
rily | -0.209 |
away | -0.205 |
eyed | -0.201 |
freely | -0.193 |
Token | barking |
Token ID | 48961 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
happened | +0.012 |
end | +0.011 |
was | +0.010 |
is | +0.010 |
OM | +0.010 |
Neg logit contributions | |
grate | -0.024 |
rily | -0.024 |
away | -0.021 |
eyed | -0.021 |
positive | -0.021 |
Token | ! |
Token ID | 0 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
grate | +0.073 |
away | +0.061 |
rily | +0.060 |
eyed | +0.058 |
freely | +0.056 |
Neg logit contributions | |
happened | -0.083 |
OM | -0.082 |
end | -0.079 |
ABC | -0.078 |
fault | -0.074 |
Token | Lily |
Token ID | 20037 |
Feature activation | +8.20e-02 |
Pos logit contributions | |
happened | +0.037 |
OM | +0.035 |
end | +0.034 |
ABC | +0.032 |
consequences | +0.029 |
Neg logit contributions | |
grate | -0.047 |
rily | -0.041 |
away | -0.040 |
eyed | -0.039 |
freely | -0.037 |
Token | mess |
Token ID | 2085 |
Feature activation | +4.61e-02 |
Pos logit contributions | |
grate | +0.125 |
rily | +0.112 |
away | +0.111 |
eyed | +0.106 |
freely | +0.101 |
Neg logit contributions | |
happened | -0.093 |
end | -0.087 |
OM | -0.083 |
ABC | -0.079 |
been | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | +1.93e-02 |
Pos logit contributions | |
OM | +0.065 |
happened | +0.062 |
end | +0.062 |
CA | +0.058 |
ABC | +0.057 |
Neg logit contributions | |
grate | -0.075 |
rily | -0.069 |
away | -0.064 |
eyed | -0.060 |
freely | -0.060 |
Token |
|
Token ID | 198 |
Feature activation | +6.91e-03 |
Pos logit contributions | |
happened | +0.027 |
OM | +0.026 |
end | +0.025 |
ABC | +0.024 |
consequences | +0.022 |
Neg logit contributions | |
grate | -0.033 |
rily | -0.028 |
away | -0.027 |
eyed | -0.027 |
freely | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | +6.77e-03 |
Pos logit contributions | |
happened | +0.010 |
OM | +0.009 |
end | +0.009 |
ABC | +0.008 |
consequences | +0.008 |
Neg logit contributions | |
grate | -0.012 |
eyed | -0.010 |
rily | -0.010 |
away | -0.010 |
freely | -0.010 |
Token | Mom |
Token ID | 29252 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
happened | +0.009 |
OM | +0.009 |
end | +0.008 |
ABC | +0.008 |
consequences | +0.007 |
Neg logit contributions | |
grate | -0.012 |
rily | -0.010 |
away | -0.010 |
eyed | -0.010 |
freely | -0.010 |
Token | is |
Token ID | 318 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
grate | +0.049 |
rily | +0.044 |
away | +0.043 |
eyed | +0.042 |
freely | +0.041 |
Neg logit contributions | |
happened | -0.023 |
end | -0.021 |
OM | -0.021 |
ABC | -0.019 |
been | -0.017 |
Token | in |
Token ID | 287 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
happened | +0.235 |
end | +0.219 |
OM | +0.215 |
ABC | +0.202 |
done | +0.191 |
Neg logit contributions | |
grate | -0.343 |
rily | -0.307 |
away | -0.302 |
eyed | -0.284 |
freely | -0.282 |
Token | the |
Token ID | 262 |
Feature activation | +6.97e-02 |
Pos logit contributions | |
grate | +0.295 |
rily | +0.263 |
away | +0.253 |
eyed | +0.252 |
freely | +0.241 |
Neg logit contributions | |
happened | -0.169 |
end | -0.161 |
OM | -0.152 |
ABC | -0.141 |
time | -0.133 |
Token | living |
Token ID | 2877 |
Feature activation | -7.29e-02 |
Pos logit contributions | |
happened | +0.057 |
OM | +0.052 |
end | +0.050 |
ABC | +0.045 |
consequences | +0.038 |
Neg logit contributions | |
grate | -0.160 |
rily | -0.143 |
away | -0.142 |
eyed | -0.139 |
freely | -0.135 |
Token | room |
Token ID | 2119 |
Feature activation | +5.24e-02 |
Pos logit contributions | |
grate | +0.166 |
rily | +0.149 |
away | +0.146 |
eyed | +0.144 |
freely | +0.140 |
Neg logit contributions | |
happened | -0.061 |
OM | -0.055 |
end | -0.054 |
ABC | -0.048 |
consequences | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
happened | +0.065 |
OM | +0.062 |
end | +0.059 |
ABC | +0.056 |
consequences | +0.051 |
Neg logit contributions | |
grate | -0.097 |
rily | -0.084 |
away | -0.084 |
eyed | -0.082 |
freely | -0.078 |
Token | with |
Token ID | 351 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
grate | +0.053 |
rily | +0.045 |
eyed | +0.044 |
Follow | +0.042 |
away | +0.042 |
Neg logit contributions | |
happened | -0.045 |
end | -0.043 |
OM | -0.041 |
ABC | -0.038 |
time | -0.037 |
Token | her |
Token ID | 607 |
Feature activation | +5.42e-02 |
Pos logit contributions | |
grate | +0.333 |
rily | +0.293 |
away | +0.288 |
eyed | +0.283 |
freely | +0.271 |
Neg logit contributions | |
happened | -0.206 |
OM | -0.191 |
end | -0.188 |
ABC | -0.173 |
consequences | -0.157 |
Token | will |
Token ID | 481 |
Feature activation | -8.43e-03 |
Pos logit contributions | |
happened | +0.070 |
end | +0.067 |
OM | +0.063 |
ABC | +0.061 |
time | +0.058 |
Neg logit contributions | |
grate | -0.099 |
rily | -0.089 |
away | -0.088 |
eyed | -0.086 |
Follow | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
grate | +0.015 |
rily | +0.014 |
away | +0.013 |
eyed | +0.013 |
freely | +0.012 |
Neg logit contributions | |
happened | -0.011 |
OM | -0.010 |
end | -0.010 |
ABC | -0.009 |
been | -0.009 |
Token | She |
Token ID | 1375 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
happened | +0.039 |
end | +0.038 |
OM | +0.037 |
ABC | +0.035 |
consequences | +0.034 |
Neg logit contributions | |
grate | -0.028 |
rily | -0.023 |
eyed | -0.022 |
away | -0.022 |
freely | -0.020 |
Token | kept |
Token ID | 4030 |
Feature activation | +2.15e-02 |
Pos logit contributions | |
grate | +0.373 |
rily | +0.325 |
away | +0.321 |
eyed | +0.320 |
freely | +0.309 |
Neg logit contributions | |
happened | -0.161 |
OM | -0.153 |
end | -0.146 |
ABC | -0.133 |
consequences | -0.121 |
Token | pushing |
Token ID | 7796 |
Feature activation | -9.84e-02 |
Pos logit contributions | |
happened | +0.021 |
end | +0.019 |
OM | +0.017 |
done | +0.017 |
been | +0.017 |
Neg logit contributions | |
grate | -0.045 |
rily | -0.043 |
away | -0.042 |
mer | -0.040 |
oc | -0.040 |
Token | the |
Token ID | 262 |
Feature activation | +6.79e-02 |
Pos logit contributions | |
grate | +0.189 |
rily | +0.169 |
eyed | +0.163 |
away | +0.159 |
Follow | +0.157 |
Neg logit contributions | |
end | -0.112 |
happened | -0.111 |
OM | -0.099 |
ABC | -0.095 |
time | -0.093 |
Token | big |
Token ID | 1263 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
happened | +0.089 |
OM | +0.086 |
omm | +0.077 |
ABC | +0.076 |
abella | +0.075 |
Neg logit contributions | |
grate | -0.112 |
active | -0.098 |
away | -0.098 |
rily | -0.098 |
freely | -0.096 |
Token | black |
Token ID | 2042 |
Feature activation | -5.82e-02 |
Pos logit contributions | |
grate | +0.231 |
away | +0.201 |
rily | +0.199 |
eyed | +0.190 |
freely | +0.188 |
Neg logit contributions | |
happened | -0.173 |
OM | -0.166 |
ABC | -0.148 |
end | -0.146 |
omm | -0.136 |
Token | ball |
Token ID | 2613 |
Feature activation | +6.82e-02 |
Pos logit contributions | |
grate | +0.069 |
away | +0.068 |
freely | +0.064 |
toward | +0.063 |
mer | +0.063 |
Neg logit contributions | |
abella | -0.094 |
happened | -0.091 |
done | -0.089 |
ABC | -0.087 |
omm | -0.087 |
Token | began |
Token ID | 2540 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
grate | +0.125 |
away | +0.120 |
rily | +0.118 |
eyed | +0.115 |
lived | +0.110 |
Neg logit contributions | |
happened | -0.083 |
end | -0.070 |
OM | -0.070 |
was | -0.064 |
been | -0.063 |
Token | looking |
Token ID | 2045 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
happened | +0.151 |
end | +0.129 |
ABC | +0.125 |
been | +0.121 |
OM | +0.120 |
Neg logit contributions | |
grate | -0.227 |
rily | -0.216 |
away | -0.212 |
eyed | -0.200 |
lived | -0.200 |
Token | around |
Token ID | 1088 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
eyed | +0.035 |
Fox | +0.035 |
Frank | +0.035 |
Amen | +0.034 |
grate | +0.034 |
Neg logit contributions | |
end | -0.020 |
best | -0.018 |
a | -0.016 |
time | -0.016 |
happened | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
grate | +0.326 |
eyed | +0.316 |
rily | +0.314 |
Follow | +0.314 |
lived | +0.307 |
Neg logit contributions | |
end | -0.117 |
happened | -0.101 |
time | -0.096 |
a | -0.084 |
was | -0.080 |
Token | She |
Token ID | 1375 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
happened | +0.024 |
OM | +0.023 |
end | +0.022 |
ABC | +0.021 |
been | +0.019 |
Neg logit contributions | |
grate | -0.031 |
away | -0.027 |
rily | -0.027 |
eyed | -0.026 |
active | -0.025 |
Token | searched |
Token ID | 16499 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
grate | +0.334 |
rily | +0.298 |
away | +0.293 |
eyed | +0.286 |
freely | +0.275 |
Neg logit contributions | |
happened | -0.190 |
OM | -0.176 |
end | -0.172 |
ABC | -0.157 |
been | -0.142 |
Token | everywhere |
Token ID | 8347 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
end | +0.033 |
happened | +0.030 |
time | +0.030 |
done | +0.026 |
best | +0.026 |
Neg logit contributions | |
grate | -0.058 |
rily | -0.054 |
eyed | -0.054 |
lived | -0.052 |
oc | -0.051 |
Token | , |
Token ID | 11 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
grate | +0.059 |
rily | +0.055 |
eyed | +0.054 |
away | +0.053 |
Follow | +0.051 |
Neg logit contributions | |
happened | -0.031 |
end | -0.030 |
time | -0.025 |
was | -0.024 |
OM | -0.023 |
Token | but |
Token ID | 475 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
happened | +0.029 |
end | +0.029 |
done | +0.027 |
OM | +0.027 |
time | +0.026 |
Neg logit contributions | |
grate | -0.035 |
rily | -0.032 |
away | -0.031 |
eyed | -0.031 |
oc | -0.029 |
Token | she |
Token ID | 673 |
Feature activation | -5.66e-02 |
Pos logit contributions | |
grate | +0.230 |
rily | +0.203 |
away | +0.198 |
eyed | +0.194 |
active | +0.184 |
Neg logit contributions | |
happened | -0.159 |
OM | -0.147 |
end | -0.147 |
ABC | -0.131 |
been | -0.122 |
Token | still |
Token ID | 991 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
grate | +0.111 |
rily | +0.093 |
away | +0.091 |
freely | +0.091 |
eyed | +0.090 |
Neg logit contributions | |
happened | -0.062 |
OM | -0.062 |
ABC | -0.060 |
end | -0.058 |
consequences | -0.054 |
Token | man |
Token ID | 582 |
Feature activation | +5.13e-03 |
Pos logit contributions | |
happened | +0.096 |
OM | +0.096 |
omm | +0.086 |
tex | +0.086 |
ained | +0.084 |
Neg logit contributions | |
grate | -0.125 |
away | -0.107 |
greet | -0.103 |
active | -0.101 |
eyed | -0.101 |
Token | wanted |
Token ID | 2227 |
Feature activation | +4.46e-02 |
Pos logit contributions | |
happened | +0.004 |
end | +0.004 |
OM | +0.004 |
ABC | +0.003 |
been | +0.003 |
Neg logit contributions | |
grate | -0.012 |
rily | -0.011 |
away | -0.011 |
eyed | -0.010 |
freely | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
happened | +0.067 |
OM | +0.064 |
end | +0.062 |
ABC | +0.059 |
consequences | +0.055 |
Neg logit contributions | |
grate | -0.072 |
rily | -0.061 |
away | -0.060 |
eyed | -0.059 |
freely | -0.056 |
Token | help |
Token ID | 1037 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
grate | +0.174 |
rily | +0.167 |
away | +0.149 |
eyed | +0.145 |
mer | +0.141 |
Neg logit contributions | |
happened | -0.139 |
end | -0.139 |
OM | -0.129 |
ABC | -0.121 |
been | -0.118 |
Token | him |
Token ID | 683 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
grate | +0.209 |
rily | +0.193 |
eyed | +0.181 |
Follow | +0.181 |
away | +0.176 |
Neg logit contributions | |
end | -0.133 |
happened | -0.124 |
ABC | -0.110 |
time | -0.109 |
OM | -0.108 |
Token | . |
Token ID | 13 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
end | +0.024 |
happened | +0.020 |
ABC | +0.019 |
time | +0.017 |
something | +0.017 |
Neg logit contributions | |
grate | -0.061 |
rily | -0.057 |
away | -0.054 |
mer | -0.053 |
Follow | -0.053 |
Token |
|
Token ID | 198 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
happened | +0.028 |
OM | +0.027 |
end | +0.026 |
ABC | +0.025 |
been | +0.022 |
Neg logit contributions | |
grate | -0.039 |
rily | -0.034 |
away | -0.034 |
eyed | -0.033 |
freely | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
happened | +0.022 |
OM | +0.021 |
end | +0.020 |
ABC | +0.019 |
consequences | +0.018 |
Neg logit contributions | |
grate | -0.027 |
away | -0.023 |
eyed | -0.023 |
rily | -0.023 |
freely | -0.022 |
Token | One |
Token ID | 3198 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
end | +0.018 |
OM | +0.017 |
happened | +0.017 |
ABC | +0.016 |
time | +0.014 |
Neg logit contributions | |
grate | -0.030 |
rily | -0.028 |
away | -0.026 |
eyed | -0.025 |
freely | -0.024 |
Token | day |
Token ID | 1110 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
happened | +0.011 |
end | +0.010 |
OM | +0.009 |
ABC | +0.007 |
time | +0.006 |
Neg logit contributions | |
grate | -0.046 |
rily | -0.043 |
away | -0.041 |
eyed | -0.041 |
freely | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | +7.34e-02 |
Pos logit contributions | |
happened | +0.163 |
OM | +0.161 |
end | +0.150 |
ABC | +0.144 |
fault | +0.134 |
Neg logit contributions | |
grate | -0.270 |
away | -0.234 |
rily | -0.232 |
eyed | -0.225 |
freely | -0.219 |
Token | , |
Token ID | 11 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
grate | +0.027 |
rily | +0.024 |
away | +0.024 |
eyed | +0.023 |
active | +0.022 |
Neg logit contributions | |
happened | -0.021 |
end | -0.020 |
OM | -0.019 |
ABC | -0.018 |
is | -0.017 |
Token | holding |
Token ID | 4769 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
end | +0.016 |
happened | +0.016 |
ABC | +0.015 |
OM | +0.014 |
is | +0.013 |
Neg logit contributions | |
grate | -0.026 |
away | -0.025 |
rily | -0.025 |
eyed | -0.024 |
lived | -0.023 |
Token | hands |
Token ID | 2832 |
Feature activation | -5.18e-02 |
Pos logit contributions | |
grate | +0.192 |
rily | +0.171 |
eyed | +0.169 |
away | +0.169 |
lived | +0.165 |
Neg logit contributions | |
happened | -0.115 |
end | -0.109 |
ABC | -0.102 |
OM | -0.096 |
time | -0.095 |
Token | , |
Token ID | 11 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
grate | +0.112 |
Follow | +0.105 |
away | +0.104 |
rily | +0.102 |
oc | +0.100 |
Neg logit contributions | |
happened | -0.040 |
end | -0.040 |
ABC | -0.033 |
time | -0.033 |
a | -0.031 |
Token | in |
Token ID | 287 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
happened | +0.016 |
end | +0.015 |
ABC | +0.015 |
OM | +0.014 |
is | +0.013 |
Neg logit contributions | |
grate | -0.026 |
away | -0.025 |
rily | -0.025 |
eyed | -0.024 |
lived | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +6.64e-02 |
Pos logit contributions | |
grate | +0.299 |
rily | +0.276 |
eyed | +0.267 |
away | +0.262 |
lived | +0.260 |
Neg logit contributions | |
end | -0.162 |
happened | -0.146 |
time | -0.134 |
ABC | -0.132 |
OM | -0.129 |
Token | store |
Token ID | 3650 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
happened | +0.079 |
OM | +0.075 |
end | +0.069 |
ABC | +0.066 |
OO | +0.061 |
Neg logit contributions | |
grate | -0.127 |
rily | -0.110 |
away | -0.110 |
eyed | -0.107 |
freely | -0.104 |
Token | . |
Token ID | 13 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
grate | +0.040 |
rily | +0.036 |
away | +0.035 |
eyed | +0.035 |
freely | +0.033 |
Neg logit contributions | |
happened | -0.022 |
end | -0.021 |
OM | -0.021 |
ABC | -0.019 |
consequences | -0.017 |
Token | And |
Token ID | 843 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
happened | +0.025 |
OM | +0.024 |
end | +0.022 |
ABC | +0.022 |
time | +0.020 |
Neg logit contributions | |
grate | -0.032 |
away | -0.029 |
rily | -0.028 |
eyed | -0.028 |
lived | -0.026 |
Token | they |
Token ID | 484 |
Feature activation | -7.86e-02 |
Pos logit contributions | |
grate | +0.262 |
away | +0.253 |
rily | +0.250 |
eyed | +0.249 |
lived | +0.236 |
Neg logit contributions | |
happened | -0.181 |
end | -0.173 |
best | -0.160 |
ABC | -0.159 |
is | -0.155 |
Token | don |
Token ID | 836 |
Feature activation | +3.47e-03 |
Pos logit contributions | |
grate | +0.139 |
rily | +0.122 |
away | +0.119 |
eyed | +0.116 |
freely | +0.111 |
Neg logit contributions | |
happened | -0.105 |
OM | -0.099 |
end | -0.097 |
ABC | -0.090 |
consequences | -0.082 |
Token | . |
Token ID | 13 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
happened | +0.043 |
end | +0.040 |
OM | +0.035 |
time | +0.032 |
is | +0.032 |
Neg logit contributions | |
grate | -0.078 |
rily | -0.070 |
away | -0.070 |
eyed | -0.070 |
active | -0.068 |
Token | To |
Token ID | 1675 |
Feature activation | -7.88e-02 |
Pos logit contributions | |
happened | +0.037 |
OM | +0.035 |
end | +0.034 |
ABC | +0.033 |
been | +0.030 |
Neg logit contributions | |
grate | -0.041 |
away | -0.036 |
rily | -0.035 |
eyed | -0.034 |
freely | -0.032 |
Token | her |
Token ID | 607 |
Feature activation | +6.55e-02 |
Pos logit contributions | |
grate | +0.146 |
rily | +0.141 |
eyed | +0.138 |
away | +0.134 |
active | +0.133 |
Neg logit contributions | |
happened | -0.095 |
end | -0.094 |
OM | -0.078 |
done | -0.074 |
time | -0.074 |
Token | surprise |
Token ID | 5975 |
Feature activation | -4.11e-02 |
Pos logit contributions | |
happened | +0.076 |
end | +0.075 |
time | +0.063 |
is | +0.057 |
problem | +0.057 |
Neg logit contributions | |
grate | -0.126 |
rily | -0.122 |
away | -0.119 |
eyed | -0.116 |
freely | -0.116 |
Token | , |
Token ID | 11 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
grate | +0.082 |
rily | +0.072 |
away | +0.071 |
eyed | +0.070 |
freely | +0.067 |
Neg logit contributions | |
happened | -0.046 |
OM | -0.043 |
end | -0.041 |
ABC | -0.039 |
consequences | -0.035 |
Token | it |
Token ID | 340 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
happened | +0.029 |
end | +0.026 |
OM | +0.024 |
something | +0.022 |
ABC | +0.022 |
Neg logit contributions | |
grate | -0.055 |
eyed | -0.049 |
away | -0.049 |
rily | -0.049 |
active | -0.046 |
Token | was |
Token ID | 373 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
happened | +0.083 |
OM | +0.070 |
end | +0.068 |
ABC | +0.057 |
been | +0.047 |
Neg logit contributions | |
grate | -0.294 |
rily | -0.267 |
away | -0.266 |
eyed | -0.262 |
freely | -0.252 |
Token | crunch |
Token ID | 23527 |
Feature activation | -8.32e-02 |
Pos logit contributions | |
OM | +0.250 |
happened | +0.242 |
ABC | +0.225 |
end | +0.219 |
OO | +0.195 |
Neg logit contributions | |
grate | -0.438 |
rily | -0.382 |
away | -0.353 |
eyed | -0.347 |
freely | -0.346 |
Token | y |
Token ID | 88 |
Feature activation | +4.67e-02 |
Pos logit contributions | |
grate | +0.145 |
rily | +0.127 |
away | +0.125 |
eyed | +0.121 |
freely | +0.116 |
Neg logit contributions | |
happened | -0.113 |
OM | -0.109 |
end | -0.104 |
ABC | -0.100 |
consequences | -0.091 |
Token | and |
Token ID | 290 |
Feature activation | -6.23e-02 |
Pos logit contributions | |
happened | +0.054 |
end | +0.050 |
OM | +0.047 |
is | +0.042 |
time | +0.042 |
Neg logit contributions | |
grate | -0.088 |
eyed | -0.079 |
away | -0.079 |
rily | -0.076 |
greet | -0.075 |
Token | tasted |
Token ID | 29187 |
Feature activation | +9.06e-02 |
Pos logit contributions | |
grate | +0.101 |
away | +0.095 |
eyed | +0.092 |
rily | +0.091 |
freely | +0.084 |
Neg logit contributions | |
happened | -0.092 |
end | -0.086 |
OM | -0.079 |
been | -0.075 |
was | -0.073 |
Token | Ginny |
Token ID | 37877 |
Feature activation | -4.95e-03 |
Pos logit contributions | |
happened | +0.033 |
OM | +0.031 |
end | +0.030 |
ABC | +0.028 |
been | +0.026 |
Neg logit contributions | |
grate | -0.040 |
rily | -0.034 |
away | -0.034 |
eyed | -0.034 |
freely | -0.032 |
Token | shouted |
Token ID | 17293 |
Feature activation | +5.09e-02 |
Pos logit contributions | |
grate | +0.009 |
away | +0.007 |
eyed | +0.007 |
rily | +0.007 |
freely | +0.007 |
Neg logit contributions | |
OM | -0.006 |
happened | -0.006 |
end | -0.006 |
ABC | -0.006 |
fault | -0.006 |
Token | out |
Token ID | 503 |
Feature activation | -4.34e-02 |
Pos logit contributions | |
happened | +0.064 |
OM | +0.061 |
end | +0.059 |
ABC | +0.055 |
consequences | +0.050 |
Neg logit contributions | |
grate | -0.095 |
rily | -0.082 |
away | -0.081 |
eyed | -0.079 |
freely | -0.076 |
Token | " |
Token ID | 366 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
grate | +0.078 |
rily | +0.070 |
away | +0.068 |
eyed | +0.067 |
freely | +0.064 |
Neg logit contributions | |
happened | -0.057 |
end | -0.051 |
OM | -0.050 |
ABC | -0.047 |
OO | -0.043 |
Token | Look |
Token ID | 8567 |
Feature activation | +9.95e-02 |
Pos logit contributions | |
grate | +0.023 |
eyed | +0.020 |
rily | +0.020 |
away | +0.020 |
freely | +0.020 |
Neg logit contributions | |
happened | -0.027 |
fault | -0.025 |
ABC | -0.025 |
OM | -0.025 |
end | -0.025 |
Token | Jack |
Token ID | 3619 |
Feature activation | -4.45e-02 |
Pos logit contributions | |
happened | +0.138 |
OM | +0.131 |
end | +0.127 |
ABC | +0.120 |
consequences | +0.111 |
Neg logit contributions | |
grate | -0.172 |
rily | -0.148 |
away | -0.146 |
eyed | -0.142 |
freely | -0.136 |
Token | !" |
Token ID | 2474 |
Feature activation | +8.79e-02 |
Pos logit contributions | |
grate | +0.065 |
rily | +0.055 |
away | +0.050 |
oc | +0.049 |
mer | +0.048 |
Neg logit contributions | |
OM | -0.069 |
happened | -0.066 |
ABC | -0.065 |
consequences | -0.062 |
end | -0.062 |
Token | She |
Token ID | 1375 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
happened | +0.136 |
OM | +0.130 |
end | +0.125 |
ABC | +0.122 |
been | +0.113 |
Neg logit contributions | |
grate | -0.133 |
away | -0.118 |
rily | -0.115 |
eyed | -0.114 |
freely | -0.107 |
Token | pointed |
Token ID | 6235 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
grate | +0.310 |
rily | +0.254 |
eyed | +0.247 |
active | +0.241 |
away | +0.239 |
Neg logit contributions | |
OM | -0.203 |
end | -0.196 |
fault | -0.183 |
happened | -0.182 |
ABC | -0.181 |
Token | to |
Token ID | 284 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
grate | +0.053 |
oc | +0.051 |
mer | +0.048 |
fire | +0.047 |
lived | +0.047 |
Neg logit contributions | |
happened | -0.028 |
something | -0.025 |
a | -0.025 |
end | -0.024 |
time | -0.022 |
Token | an |
Token ID | 281 |
Feature activation | +5.45e-02 |
Pos logit contributions | |
grate | +0.226 |
rily | +0.202 |
away | +0.199 |
eyed | +0.196 |
freely | +0.190 |
Neg logit contributions | |
happened | -0.088 |
OM | -0.081 |
end | -0.077 |
ABC | -0.069 |
consequences | -0.060 |
Token | in |
Token ID | 287 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
grate | +0.052 |
rily | +0.046 |
eyed | +0.045 |
away | +0.044 |
lived | +0.042 |
Neg logit contributions | |
happened | -0.048 |
end | -0.048 |
OM | -0.045 |
ABC | -0.042 |
time | -0.041 |
Token | the |
Token ID | 262 |
Feature activation | +7.62e-02 |
Pos logit contributions | |
grate | +0.278 |
rily | +0.247 |
away | +0.239 |
eyed | +0.239 |
lived | +0.225 |
Neg logit contributions | |
happened | -0.166 |
end | -0.161 |
OM | -0.154 |
ABC | -0.141 |
time | -0.131 |
Token | water |
Token ID | 1660 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
happened | +0.080 |
OM | +0.074 |
end | +0.073 |
ABC | +0.066 |
consequences | +0.059 |
Neg logit contributions | |
grate | -0.158 |
rily | -0.140 |
away | -0.138 |
eyed | -0.135 |
freely | -0.130 |
Token | , |
Token ID | 11 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
end | +0.099 |
happened | +0.084 |
OM | +0.074 |
time | +0.074 |
is | +0.070 |
Neg logit contributions | |
grate | -0.232 |
rily | -0.216 |
eyed | -0.214 |
away | -0.207 |
lived | -0.204 |
Token | and |
Token ID | 290 |
Feature activation | -6.60e-02 |
Pos logit contributions | |
end | +0.024 |
happened | +0.022 |
was | +0.022 |
something | +0.021 |
wasn | +0.021 |
Neg logit contributions | |
grate | -0.042 |
away | -0.042 |
rily | -0.042 |
eyed | -0.041 |
oc | -0.038 |
Token | they |
Token ID | 484 |
Feature activation | -6.81e-02 |
Pos logit contributions | |
rily | +0.123 |
grate | +0.123 |
away | +0.116 |
eyed | +0.115 |
itching | +0.111 |
Neg logit contributions | |
happened | -0.077 |
end | -0.075 |
was | -0.070 |
been | -0.065 |
something | -0.064 |
Token | became |
Token ID | 2627 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
grate | +0.140 |
rily | +0.130 |
away | +0.126 |
eyed | +0.122 |
lived | +0.117 |
Neg logit contributions | |
happened | -0.072 |
end | -0.069 |
OM | -0.060 |
done | -0.054 |
been | -0.054 |
Token | friends |
Token ID | 2460 |
Feature activation | -4.63e-02 |
Pos logit contributions | |
happened | +0.185 |
OM | +0.174 |
end | +0.167 |
ABC | +0.155 |
consequences | +0.139 |
Neg logit contributions | |
grate | -0.325 |
rily | -0.286 |
away | -0.281 |
eyed | -0.276 |
freely | -0.266 |
Token | . |
Token ID | 13 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
grate | +0.102 |
rily | +0.099 |
away | +0.096 |
eyed | +0.094 |
lived | +0.093 |
Neg logit contributions | |
end | -0.033 |
happened | -0.029 |
was | -0.026 |
time | -0.024 |
is | -0.024 |
Token | The |
Token ID | 383 |
Feature activation | -8.53e-02 |
Pos logit contributions | |
end | +0.024 |
happened | +0.023 |
OM | +0.023 |
is | +0.022 |
ABC | +0.022 |
Neg logit contributions | |
away | -0.031 |
grate | -0.030 |
rily | -0.028 |
positive | -0.026 |
active | -0.026 |
Token | pond |
Token ID | 16723 |
Feature activation | +5.01e-02 |
Pos logit contributions | |
grate | +0.122 |
rily | +0.099 |
away | +0.097 |
eyed | +0.094 |
freely | +0.093 |
Neg logit contributions | |
happened | -0.145 |
OM | -0.142 |
ABC | -0.129 |
end | -0.126 |
OO | -0.123 |
Token | And |
Token ID | 843 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
is | +0.022 |
omm | +0.022 |
OM | +0.021 |
ABC | +0.021 |
CA | +0.020 |
Neg logit contributions | |
away | -0.042 |
grate | -0.041 |
rily | -0.040 |
ized | -0.040 |
afloat | -0.038 |
Token | this |
Token ID | 428 |
Feature activation | +8.52e-02 |
Pos logit contributions | |
rily | +0.291 |
away | +0.289 |
grate | +0.279 |
eyed | +0.276 |
freely | +0.268 |
Neg logit contributions | |
is | -0.153 |
something | -0.147 |
end | -0.145 |
a | -0.140 |
best | -0.134 |
Token | is |
Token ID | 318 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
OM | +0.076 |
happened | +0.075 |
end | +0.067 |
ABC | +0.061 |
OO | +0.055 |
Neg logit contributions | |
grate | -0.184 |
rily | -0.162 |
away | -0.160 |
eyed | -0.158 |
active | -0.153 |
Token | my |
Token ID | 616 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
happened | +0.162 |
end | +0.148 |
OM | +0.124 |
ABC | +0.118 |
time | +0.116 |
Neg logit contributions | |
grate | -0.426 |
rily | -0.401 |
away | -0.397 |
eyed | -0.390 |
freely | -0.374 |
Token | dog |
Token ID | 3290 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
happened | +0.020 |
end | +0.019 |
OM | +0.018 |
ABC | +0.017 |
fault | +0.016 |
Neg logit contributions | |
grate | -0.037 |
rily | -0.034 |
away | -0.033 |
eyed | -0.032 |
freely | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
happened | +0.013 |
end | +0.012 |
is | +0.010 |
ABC | +0.010 |
been | +0.010 |
Neg logit contributions | |
grate | -0.028 |
rily | -0.026 |
eyed | -0.025 |
away | -0.025 |
freely | -0.024 |
Token | Rex |
Token ID | 17853 |
Feature activation | -2.57e-03 |
Pos logit contributions | |
happened | +0.023 |
end | +0.022 |
OM | +0.021 |
ABC | +0.019 |
been | +0.019 |
Neg logit contributions | |
grate | -0.035 |
eyed | -0.032 |
away | -0.032 |
rily | -0.032 |
freely | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
grate | +0.005 |
rily | +0.005 |
away | +0.005 |
eyed | +0.005 |
freely | +0.005 |
Neg logit contributions | |
happened | -0.003 |
end | -0.002 |
is | -0.002 |
OM | -0.002 |
ABC | -0.002 |
Token | He |
Token ID | 679 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
is | +0.025 |
CA | +0.024 |
ABC | +0.024 |
OM | +0.024 |
omm | +0.024 |
Neg logit contributions | |
away | -0.036 |
ized | -0.035 |
grate | -0.034 |
rily | -0.034 |
positive | -0.031 |
Token | is |
Token ID | 318 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
grate | +0.448 |
rily | +0.436 |
away | +0.427 |
eyed | +0.407 |
freely | +0.402 |
Neg logit contributions | |
happened | -0.119 |
is | -0.091 |
end | -0.088 |
been | -0.076 |
was | -0.071 |
Token | a |
Token ID | 257 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
happened | +0.161 |
been | +0.141 |
time | +0.138 |
a | +0.138 |
end | +0.137 |
Neg logit contributions | |
grate | -0.411 |
away | -0.405 |
rily | -0.397 |
ized | -0.380 |
eyed | -0.376 |
Token | The |
Token ID | 383 |
Feature activation | -8.04e-02 |
Pos logit contributions | |
happened | +0.028 |
OM | +0.027 |
end | +0.026 |
ABC | +0.025 |
time | +0.024 |
Neg logit contributions | |
grate | -0.028 |
away | -0.026 |
rily | -0.024 |
eyed | -0.023 |
lived | -0.023 |
Token | m |
Token ID | 285 |
Feature activation | -3.81e-03 |
Pos logit contributions | |
grate | +0.155 |
rily | +0.132 |
away | +0.131 |
eyed | +0.128 |
active | +0.125 |
Neg logit contributions | |
happened | -0.095 |
OM | -0.091 |
ABC | -0.077 |
end | -0.076 |
OO | -0.075 |
Token | oles |
Token ID | 4316 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
grate | +0.007 |
rily | +0.006 |
away | +0.006 |
eyed | +0.006 |
freely | +0.006 |
Neg logit contributions | |
happened | -0.005 |
OM | -0.005 |
end | -0.005 |
ABC | -0.004 |
consequences | -0.004 |
Token | went |
Token ID | 1816 |
Feature activation | +7.09e-02 |
Pos logit contributions | |
rily | +0.044 |
grate | +0.044 |
away | +0.043 |
lived | +0.040 |
oc | +0.040 |
Neg logit contributions | |
end | -0.022 |
happened | -0.022 |
was | -0.019 |
done | -0.018 |
is | -0.017 |
Token | back |
Token ID | 736 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
end | +0.068 |
happened | +0.064 |
a | +0.059 |
time | +0.057 |
done | +0.057 |
Neg logit contributions | |
grate | -0.136 |
rily | -0.133 |
oc | -0.129 |
lived | -0.127 |
slur | -0.120 |
Token | to |
Token ID | 284 |
Feature activation | -9.29e-02 |
Pos logit contributions | |
grate | +0.060 |
rily | +0.055 |
away | +0.049 |
eyed | +0.049 |
lived | +0.049 |
Neg logit contributions | |
happened | -0.042 |
end | -0.037 |
OM | -0.036 |
ABC | -0.035 |
been | -0.033 |
Token | their |
Token ID | 511 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
grate | +0.175 |
rily | +0.162 |
away | +0.155 |
lived | +0.153 |
eyed | +0.153 |
Neg logit contributions | |
happened | -0.114 |
end | -0.109 |
OM | -0.098 |
ABC | -0.096 |
been | -0.094 |
Token | hole |
Token ID | 7604 |
Feature activation | +2.55e-02 |
Pos logit contributions | |
grate | +0.063 |
rily | +0.057 |
eyed | +0.055 |
away | +0.054 |
freely | +0.052 |
Neg logit contributions | |
end | -0.031 |
happened | -0.029 |
OM | -0.027 |
ABC | -0.027 |
time | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
happened | +0.030 |
end | +0.027 |
OM | +0.025 |
done | +0.024 |
been | +0.023 |
Neg logit contributions | |
grate | -0.048 |
rily | -0.042 |
away | -0.042 |
eyed | -0.042 |
Follow | -0.042 |
Token | Sara |
Token ID | 24799 |
Feature activation | +9.79e-02 |
Pos logit contributions | |
happened | +0.029 |
OM | +0.028 |
end | +0.027 |
ABC | +0.026 |
time | +0.026 |
Neg logit contributions | |
grate | -0.031 |
away | -0.029 |
rily | -0.027 |
eyed | -0.026 |
lived | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | -6.45e-02 |
Pos logit contributions | |
happened | +0.122 |
OM | +0.114 |
end | +0.111 |
ABC | +0.104 |
consequences | +0.094 |
Neg logit contributions | |
grate | -0.183 |
rily | -0.160 |
away | -0.158 |
eyed | -0.154 |
freely | -0.148 |
Token | was |
Token ID | 373 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
happened | +0.036 |
OM | +0.031 |
end | +0.030 |
ABC | +0.026 |
been | +0.023 |
Neg logit contributions | |
grate | -0.102 |
rily | -0.092 |
away | -0.091 |
eyed | -0.089 |
freely | -0.086 |
Token | . |
Token ID | 13 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
happened | +0.256 |
end | +0.233 |
OM | +0.228 |
ABC | +0.207 |
been | +0.196 |
Neg logit contributions | |
grate | -0.420 |
rily | -0.372 |
away | -0.363 |
eyed | -0.358 |
freely | -0.343 |
Token | The |
Token ID | 383 |
Feature activation | -8.44e-02 |
Pos logit contributions | |
happened | +0.028 |
is | +0.027 |
was | +0.026 |
a | +0.026 |
OM | +0.025 |
Neg logit contributions | |
grate | -0.039 |
away | -0.038 |
afloat | -0.038 |
positive | -0.037 |
eyed | -0.037 |
Token | bird |
Token ID | 6512 |
Feature activation | +5.81e-03 |
Pos logit contributions | |
grate | +0.142 |
away | +0.123 |
rily | +0.119 |
eyed | +0.115 |
active | +0.112 |
Neg logit contributions | |
OM | -0.119 |
happened | -0.117 |
ained | -0.102 |
OO | -0.101 |
ABC | -0.100 |
Token | flew |
Token ID | 13112 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
happened | +0.006 |
was | +0.005 |
end | +0.005 |
wasn | +0.004 |
been | +0.004 |
Neg logit contributions | |
grate | -0.012 |
rily | -0.012 |
away | -0.011 |
eyed | -0.011 |
freely | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | -9.87e-02 |
Pos logit contributions | |
happened | +0.057 |
end | +0.056 |
OM | +0.050 |
time | +0.049 |
a | +0.049 |
Neg logit contributions | |
grate | -0.062 |
rily | -0.055 |
oc | -0.054 |
eyed | -0.052 |
lived | -0.051 |
Token | the |
Token ID | 262 |
Feature activation | +7.66e-02 |
Pos logit contributions | |
grate | +0.201 |
rily | +0.182 |
eyed | +0.180 |
away | +0.175 |
active | +0.174 |
Neg logit contributions | |
happened | -0.101 |
end | -0.099 |
OM | -0.094 |
ABC | -0.083 |
been | -0.077 |
Token | king |
Token ID | 5822 |
Feature activation | +2.84e-02 |
Pos logit contributions | |
happened | +0.101 |
OM | +0.096 |
end | +0.089 |
ABC | +0.086 |
OO | +0.080 |
Neg logit contributions | |
grate | -0.138 |
away | -0.120 |
rily | -0.119 |
eyed | -0.115 |
freely | -0.111 |
Token | and |
Token ID | 290 |
Feature activation | -6.33e-02 |
Pos logit contributions | |
happened | +0.024 |
end | +0.023 |
is | +0.020 |
was | +0.020 |
been | +0.020 |
Neg logit contributions | |
grate | -0.062 |
rily | -0.057 |
active | -0.056 |
away | -0.054 |
freely | -0.054 |
Token | said |
Token ID | 531 |
Feature activation | +5.74e-02 |
Pos logit contributions | |
grate | +0.112 |
rily | +0.103 |
away | +0.101 |
eyed | +0.100 |
freely | +0.096 |
Neg logit contributions | |
happened | -0.084 |
end | -0.074 |
OM | -0.071 |
been | -0.066 |
was | -0.065 |
Token | , |
Token ID | 11 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
happened | +0.059 |
end | +0.051 |
OM | +0.049 |
ABC | +0.046 |
consequences | +0.039 |
Neg logit contributions | |
grate | -0.121 |
rily | -0.110 |
away | -0.109 |
eyed | -0.107 |
freely | -0.104 |
Token | even |
Token ID | 772 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
happened | +0.248 |
end | +0.227 |
OM | +0.226 |
ABC | +0.206 |
been | +0.195 |
Neg logit contributions | |
grate | -0.420 |
rily | -0.374 |
away | -0.372 |
eyed | -0.354 |
freely | -0.344 |
Token | more |
Token ID | 517 |
Feature activation | +5.89e-02 |
Pos logit contributions | |
grate | +0.038 |
rily | +0.035 |
away | +0.033 |
eyed | +0.033 |
freely | +0.032 |
Neg logit contributions | |
happened | -0.019 |
end | -0.018 |
OM | -0.016 |
ABC | -0.015 |
done | -0.015 |
Token | excited |
Token ID | 6568 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
happened | +0.084 |
time | +0.077 |
best | +0.073 |
done | +0.070 |
been | +0.069 |
Neg logit contributions | |
grate | -0.103 |
rily | -0.101 |
ishly | -0.090 |
Follow | -0.089 |
away | -0.088 |
Token | now |
Token ID | 783 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
grate | +0.291 |
rily | +0.258 |
eyed | +0.249 |
away | +0.245 |
Follow | +0.237 |
Neg logit contributions | |
happened | -0.185 |
end | -0.162 |
OM | -0.157 |
ABC | -0.144 |
been | -0.141 |
Token | . |
Token ID | 13 |
Feature activation | +2.07e-02 |
Pos logit contributions | |
happened | +0.051 |
end | +0.046 |
time | +0.039 |
been | +0.037 |
was | +0.035 |
Neg logit contributions | |
grate | -0.109 |
rily | -0.103 |
eyed | -0.095 |
away | -0.095 |
Follow | -0.094 |
Token | He |
Token ID | 679 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
happened | +0.031 |
OM | +0.031 |
end | +0.031 |
ABC | +0.030 |
time | +0.029 |
Neg logit contributions | |
grate | -0.028 |
away | -0.027 |
rily | -0.025 |
eyed | -0.024 |
active | -0.023 |
Token | liked |
Token ID | 8288 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
grate | +0.387 |
rily | +0.370 |
away | +0.361 |
eyed | +0.345 |
freely | +0.335 |
Neg logit contributions | |
happened | -0.153 |
end | -0.132 |
OM | -0.122 |
was | -0.115 |
been | -0.114 |
Token | the |
Token ID | 262 |
Feature activation | +7.78e-02 |
Pos logit contributions | |
grate | +0.044 |
rily | +0.041 |
eyed | +0.039 |
away | +0.038 |
lived | +0.038 |
Neg logit contributions | |
happened | -0.022 |
ABC | -0.018 |
end | -0.017 |
time | -0.017 |
been | -0.017 |
Token | feeling |
Token ID | 4203 |
Feature activation | +4.60e-02 |
Pos logit contributions | |
happened | +0.100 |
end | +0.096 |
OM | +0.093 |
ABC | +0.087 |
consequences | +0.081 |
Neg logit contributions | |
grate | -0.143 |
rily | -0.126 |
away | -0.122 |
eyed | -0.121 |
freely | -0.114 |
Token | of |
Token ID | 286 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
happened | +0.059 |
time | +0.048 |
was | +0.047 |
been | +0.047 |
end | +0.046 |
Neg logit contributions | |
grate | -0.091 |
rily | -0.084 |
greet | -0.074 |
Amen | -0.074 |
Fox | -0.074 |
Token | the |
Token ID | 262 |
Feature activation | +7.77e-02 |
Pos logit contributions | |
grate | +0.327 |
rily | +0.304 |
away | +0.293 |
mer | +0.289 |
eyed | +0.288 |
Neg logit contributions | |
end | -0.188 |
happened | -0.186 |
time | -0.163 |
been | -0.158 |
consequences | -0.156 |
Token | fell |
Token ID | 3214 |
Feature activation | +7.35e-02 |
Pos logit contributions | |
happened | +0.099 |
OM | +0.089 |
end | +0.085 |
ABC | +0.074 |
been | +0.064 |
Neg logit contributions | |
grate | -0.283 |
rily | -0.256 |
away | -0.253 |
eyed | -0.249 |
freely | -0.240 |
Token | off |
Token ID | 572 |
Feature activation | -5.48e-02 |
Pos logit contributions | |
happened | +0.123 |
OM | +0.117 |
end | +0.114 |
ABC | +0.108 |
been | +0.102 |
Neg logit contributions | |
grate | -0.106 |
rily | -0.088 |
away | -0.087 |
eyed | -0.085 |
freely | -0.080 |
Token | and |
Token ID | 290 |
Feature activation | -6.13e-02 |
Pos logit contributions | |
grate | +0.115 |
Follow | +0.112 |
eyed | +0.110 |
oc | +0.106 |
lived | +0.105 |
Neg logit contributions | |
end | -0.047 |
happened | -0.046 |
time | -0.041 |
a | -0.040 |
was | -0.038 |
Token | broke |
Token ID | 6265 |
Feature activation | +6.00e-02 |
Pos logit contributions | |
grate | +0.124 |
rily | +0.111 |
away | +0.110 |
eyed | +0.109 |
oc | +0.101 |
Neg logit contributions | |
happened | -0.070 |
end | -0.060 |
OM | -0.060 |
been | -0.052 |
done | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
happened | +0.082 |
OM | +0.076 |
end | +0.075 |
ABC | +0.069 |
consequences | +0.065 |
Neg logit contributions | |
grate | -0.105 |
away | -0.090 |
rily | -0.089 |
eyed | -0.088 |
Follow | -0.084 |
Token | She |
Token ID | 1375 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
happened | +0.031 |
OM | +0.029 |
end | +0.026 |
ABC | +0.026 |
omm | +0.025 |
Neg logit contributions | |
grate | -0.047 |
away | -0.043 |
rily | -0.042 |
eyed | -0.041 |
active | -0.038 |
Token | was |
Token ID | 373 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
grate | +0.353 |
rily | +0.308 |
away | +0.301 |
eyed | +0.299 |
freely | +0.288 |
Neg logit contributions | |
OM | -0.157 |
happened | -0.156 |
end | -0.142 |
ABC | -0.136 |
fault | -0.122 |
Token | sad |
Token ID | 6507 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
happened | +0.268 |
OM | +0.258 |
end | +0.243 |
ABC | +0.232 |
consequences | +0.207 |
Neg logit contributions | |
grate | -0.415 |
rily | -0.361 |
away | -0.352 |
eyed | -0.351 |
freely | -0.335 |
Token | and |
Token ID | 290 |
Feature activation | -6.38e-02 |
Pos logit contributions | |
grate | +0.208 |
rily | +0.189 |
eyed | +0.184 |
Follow | +0.180 |
away | +0.180 |
Neg logit contributions | |
happened | -0.124 |
end | -0.114 |
OM | -0.105 |
fault | -0.096 |
time | -0.093 |
Token | didn |
Token ID | 1422 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
grate | +0.117 |
rily | +0.105 |
away | +0.104 |
eyed | +0.099 |
freely | +0.096 |
Neg logit contributions | |
happened | -0.085 |
end | -0.077 |
OM | -0.072 |
been | -0.067 |
done | -0.067 |
Token | 't |
Token ID | 470 |
Feature activation | -7.81e-03 |
Pos logit contributions | |
happened | +0.240 |
OM | +0.228 |
end | +0.215 |
ABC | +0.214 |
consequences | +0.210 |
Neg logit contributions | |
grate | -0.267 |
eyed | -0.227 |
rily | -0.223 |
away | -0.213 |
freely | -0.212 |
Token | . |
Token ID | 13 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
happened | +0.102 |
end | +0.095 |
OM | +0.091 |
ABC | +0.085 |
been | +0.077 |
Neg logit contributions | |
grate | -0.183 |
rily | -0.163 |
away | -0.161 |
eyed | -0.157 |
freely | -0.152 |
Token | Come |
Token ID | 7911 |
Feature activation | -7.12e-02 |
Pos logit contributions | |
happened | +0.016 |
is | +0.015 |
end | +0.015 |
OM | +0.015 |
been | +0.014 |
Neg logit contributions | |
ized | -0.032 |
rily | -0.032 |
grate | -0.031 |
away | -0.031 |
ling | -0.028 |
Token | with |
Token ID | 351 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
rily | +0.135 |
grate | +0.133 |
lived | +0.125 |
away | +0.124 |
ized | +0.123 |
Neg logit contributions | |
end | -0.077 |
OM | -0.068 |
is | -0.067 |
happened | -0.066 |
was | -0.062 |
Token | me |
Token ID | 502 |
Feature activation | +6.99e-02 |
Pos logit contributions | |
grate | +0.274 |
away | +0.231 |
rily | +0.221 |
greet | +0.214 |
eyed | +0.209 |
Neg logit contributions | |
OM | -0.269 |
happened | -0.269 |
ABC | -0.254 |
fault | -0.237 |
ained | -0.231 |
Token | ." |
Token ID | 526 |
Feature activation | +4.38e-02 |
Pos logit contributions | |
end | +0.077 |
happened | +0.075 |
OM | +0.075 |
ABC | +0.068 |
been | +0.061 |
Neg logit contributions | |
grate | -0.133 |
rily | -0.122 |
away | -0.118 |
eyed | -0.116 |
lived | -0.111 |
Token | He |
Token ID | 679 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
happened | +0.049 |
OM | +0.047 |
end | +0.045 |
ABC | +0.044 |
is | +0.040 |
Neg logit contributions | |
grate | -0.082 |
away | -0.076 |
rily | -0.075 |
active | -0.072 |
eyed | -0.072 |
Token | lifted |
Token ID | 13663 |
Feature activation | -4.07e-02 |
Pos logit contributions | |
grate | +0.304 |
rily | +0.253 |
away | +0.245 |
eyed | +0.241 |
freely | +0.229 |
Neg logit contributions | |
happened | -0.264 |
OM | -0.257 |
end | -0.249 |
ABC | -0.234 |
consequences | -0.224 |
Token | her |
Token ID | 607 |
Feature activation | +5.35e-02 |
Pos logit contributions | |
grate | +0.085 |
rily | +0.075 |
away | +0.075 |
eyed | +0.074 |
active | +0.072 |
Neg logit contributions | |
happened | -0.039 |
end | -0.037 |
OM | -0.035 |
ABC | -0.032 |
time | -0.029 |
Token | up |
Token ID | 510 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
end | +0.064 |
happened | +0.062 |
ABC | +0.060 |
OM | +0.059 |
is | +0.055 |
Neg logit contributions | |
grate | -0.094 |
away | -0.083 |
rily | -0.081 |
freely | -0.081 |
lived | -0.081 |
Token | and |
Token ID | 290 |
Feature activation | -7.17e-02 |
Pos logit contributions | |
grate | +0.078 |
Follow | +0.072 |
lived | +0.071 |
eyed | +0.071 |
rily | +0.071 |
Neg logit contributions | |
end | -0.048 |
happened | -0.045 |
time | -0.043 |
ABC | -0.043 |
OM | -0.039 |
Token | carried |
Token ID | 5281 |
Feature activation | -8.11e-02 |
Pos logit contributions | |
rily | +0.135 |
away | +0.133 |
grate | +0.132 |
eyed | +0.131 |
active | +0.125 |
Neg logit contributions | |
happened | -0.079 |
was | -0.068 |
slides | -0.061 |
is | -0.059 |
stops | -0.058 |
Token | You |
Token ID | 921 |
Feature activation | -6.54e-02 |
Pos logit contributions | |
OM | +0.022 |
ABC | +0.021 |
end | +0.021 |
is | +0.021 |
omm | +0.020 |
Neg logit contributions | |
away | -0.030 |
rily | -0.028 |
grate | -0.027 |
ized | -0.025 |
eyed | -0.025 |
Token | are |
Token ID | 389 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
grate | +0.143 |
rily | +0.138 |
away | +0.131 |
eyed | +0.127 |
active | +0.122 |
Neg logit contributions | |
happened | -0.059 |
end | -0.057 |
OM | -0.052 |
ABC | -0.047 |
is | -0.046 |
Token | great |
Token ID | 1049 |
Feature activation | -7.32e-02 |
Pos logit contributions | |
happened | +0.128 |
end | +0.127 |
best | +0.113 |
done | +0.110 |
been | +0.108 |
Neg logit contributions | |
grate | -0.254 |
away | -0.244 |
rily | -0.242 |
eyed | -0.233 |
oc | -0.224 |
Token | at |
Token ID | 379 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
grate | +0.149 |
away | +0.139 |
rily | +0.138 |
eyed | +0.137 |
lived | +0.128 |
Neg logit contributions | |
happened | -0.077 |
end | -0.065 |
ABC | -0.060 |
fault | -0.059 |
best | -0.059 |
Token | making |
Token ID | 1642 |
Feature activation | -7.58e-02 |
Pos logit contributions | |
grate | +0.361 |
rily | +0.320 |
away | +0.308 |
eyed | +0.303 |
freely | +0.282 |
Neg logit contributions | |
happened | -0.286 |
end | -0.269 |
OM | -0.266 |
ABC | -0.249 |
been | -0.232 |
Token | beautiful |
Token ID | 4950 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
grate | +0.154 |
rily | +0.138 |
away | +0.137 |
eyed | +0.137 |
lived | +0.128 |
Neg logit contributions | |
happened | -0.080 |
end | -0.074 |
OM | -0.070 |
ABC | -0.068 |
consequences | -0.063 |
Token | cob |
Token ID | 22843 |
Feature activation | +9.35e-02 |
Pos logit contributions | |
grate | +0.319 |
rily | +0.287 |
away | +0.281 |
eyed | +0.281 |
freely | +0.266 |
Neg logit contributions | |
happened | -0.132 |
end | -0.117 |
OM | -0.117 |
ABC | -0.108 |
consequences | -0.094 |
Token | we |
Token ID | 732 |
Feature activation | -5.06e-02 |
Pos logit contributions | |
happened | +0.108 |
OM | +0.100 |
end | +0.096 |
ABC | +0.093 |
consequences | +0.084 |
Neg logit contributions | |
grate | -0.182 |
rily | -0.161 |
away | -0.158 |
eyed | -0.155 |
active | -0.146 |
Token | bs |
Token ID | 1443 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
grate | +0.079 |
rily | +0.071 |
away | +0.060 |
eyed | +0.057 |
othe | +0.056 |
Neg logit contributions | |
happened | -0.087 |
end | -0.076 |
been | -0.073 |
done | -0.073 |
time | -0.072 |
Token | , |
Token ID | 11 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
grate | +0.356 |
eyed | +0.315 |
rily | +0.311 |
away | +0.308 |
lived | +0.296 |
Neg logit contributions | |
happened | -0.198 |
end | -0.194 |
is | -0.167 |
OM | -0.165 |
ABC | -0.165 |
Token | and |
Token ID | 290 |
Feature activation | -7.00e-02 |
Pos logit contributions | |
end | +0.020 |
is | +0.018 |
happened | +0.018 |
best | +0.018 |
time | +0.018 |
Neg logit contributions | |
eyed | -0.041 |
away | -0.039 |
grate | -0.039 |
rily | -0.038 |
lived | -0.036 |
Token | being |
Token ID | 852 |
Feature activation | +5.73e-02 |
Pos logit contributions | |
grate | +0.134 |
rily | +0.121 |
away | +0.118 |
eyed | +0.115 |
freely | +0.111 |
Neg logit contributions | |
happened | -0.076 |
end | -0.071 |
OM | -0.069 |
ABC | -0.063 |
been | -0.059 |
Token | nice |
Token ID | 3621 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
happened | +0.057 |
end | +0.049 |
been | +0.046 |
best | +0.046 |
done | +0.045 |
Neg logit contributions | |
grate | -0.120 |
rily | -0.113 |
away | -0.109 |
oc | -0.106 |
eyed | -0.105 |
Token | to |
Token ID | 284 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
grate | +0.168 |
away | +0.138 |
rily | +0.134 |
eyed | +0.130 |
freely | +0.126 |
Neg logit contributions | |
OM | -0.205 |
happened | -0.188 |
end | -0.178 |
ABC | -0.175 |
CA | -0.174 |
Token | her |
Token ID | 607 |
Feature activation | +5.31e-02 |
Pos logit contributions | |
grate | +0.203 |
rily | +0.187 |
eyed | +0.177 |
away | +0.177 |
lived | +0.171 |
Neg logit contributions | |
happened | -0.111 |
end | -0.109 |
OM | -0.099 |
ABC | -0.095 |
time | -0.087 |
Token | . |
Token ID | 13 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
end | +0.055 |
happened | +0.050 |
ABC | +0.046 |
OM | +0.043 |
time | +0.041 |
Neg logit contributions | |
grate | -0.109 |
rily | -0.103 |
eyed | -0.101 |
away | -0.099 |
lived | -0.096 |
Token | He |
Token ID | 679 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
happened | +0.018 |
OM | +0.017 |
end | +0.017 |
ABC | +0.016 |
been | +0.014 |
Neg logit contributions | |
grate | -0.026 |
rily | -0.023 |
away | -0.022 |
eyed | -0.021 |
freely | -0.020 |
Token | thanked |
Token ID | 26280 |
Feature activation | -9.91e-02 |
Pos logit contributions | |
grate | +0.304 |
rily | +0.247 |
away | +0.237 |
eyed | +0.236 |
freely | +0.223 |
Neg logit contributions | |
OM | -0.265 |
happened | -0.263 |
end | -0.247 |
ABC | -0.239 |
OO | -0.225 |
Token | her |
Token ID | 607 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
grate | +0.206 |
eyed | +0.189 |
rily | +0.189 |
away | +0.184 |
lived | +0.181 |
Neg logit contributions | |
end | -0.097 |
happened | -0.094 |
OM | -0.082 |
ABC | -0.080 |
time | -0.072 |
Token | for |
Token ID | 329 |
Feature activation | -6.75e-02 |
Pos logit contributions | |
happened | +0.070 |
end | +0.066 |
OM | +0.066 |
ABC | +0.062 |
consequences | +0.056 |
Neg logit contributions | |
grate | -0.092 |
rily | -0.081 |
away | -0.079 |
eyed | -0.078 |
freely | -0.074 |
Token | sharing |
Token ID | 7373 |
Feature activation | -9.34e-02 |
Pos logit contributions | |
grate | +0.133 |
rily | +0.119 |
away | +0.116 |
eyed | +0.114 |
freely | +0.110 |
Neg logit contributions | |
happened | -0.077 |
end | -0.071 |
OM | -0.071 |
ABC | -0.064 |
been | -0.059 |
Token | her |
Token ID | 607 |
Feature activation | +5.27e-02 |
Pos logit contributions | |
grate | +0.156 |
rily | +0.127 |
away | +0.127 |
eyed | +0.119 |
freely | +0.118 |
Neg logit contributions | |
OM | -0.143 |
happened | -0.136 |
OO | -0.124 |
ABC | -0.120 |
end | -0.119 |