Token |
|
Token ID | 198 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
ey | +0.041 |
ad | +0.037 |
idges | +0.034 |
odic | +0.034 |
ement | +0.032 |
Neg logit contributions | |
gy | -0.046 |
bank | -0.038 |
neither | -0.037 |
station | -0.035 |
drown | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | +5.00e-01 |
Pos logit contributions | |
ey | +0.014 |
yogurt | +0.012 |
honey | +0.012 |
idges | +0.011 |
ad | +0.011 |
Neg logit contributions | |
gy | -0.024 |
bank | -0.019 |
lves | -0.017 |
zag | -0.017 |
station | -0.016 |
Token | " |
Token ID | 1 |
Feature activation | -7.62e-01 |
Pos logit contributions | |
ey | +0.030 |
yogurt | +0.024 |
ad | +0.024 |
idges | +0.024 |
honey | +0.024 |
Neg logit contributions | |
gy | -0.053 |
bank | -0.044 |
lves | -0.039 |
zag | -0.038 |
drown | -0.037 |
Token | Sorry |
Token ID | 14385 |
Feature activation | +2.57e+00 |
Pos logit contributions | |
gy | +0.081 |
bank | +0.068 |
lves | +0.061 |
zag | +0.060 |
station | +0.058 |
Neg logit contributions | |
ey | -0.045 |
honey | -0.035 |
yogurt | -0.035 |
idges | -0.035 |
ad | -0.035 |
Token | , |
Token ID | 11 |
Feature activation | +7.05e-01 |
Pos logit contributions | |
ey | +0.238 |
idges | +0.193 |
ad | +0.190 |
ement | +0.186 |
yogurt | +0.185 |
Neg logit contributions | |
gy | -0.198 |
bank | -0.146 |
neither | -0.119 |
station | -0.118 |
zag | -0.116 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
ey | +0.046 |
idges | +0.038 |
ad | +0.037 |
yogurt | +0.037 |
honey | +0.036 |
Neg logit contributions | |
gy | -0.070 |
bank | -0.058 |
lves | -0.051 |
zag | -0.050 |
station | -0.049 |
Token | , |
Token ID | 11 |
Feature activation | +5.83e-01 |
Pos logit contributions | |
ey | +0.751 |
honey | +0.664 |
yogurt | +0.651 |
ad | +0.631 |
idges | +0.625 |
Neg logit contributions | |
gy | -1.023 |
bank | -0.847 |
lves | -0.768 |
zag | -0.719 |
drown | -0.689 |
Token | we |
Token ID | 356 |
Feature activation | -4.02e+00 |
Pos logit contributions | |
ey | +0.029 |
ad | +0.024 |
idges | +0.023 |
ement | +0.022 |
ango | +0.021 |
Neg logit contributions | |
gy | -0.064 |
bank | -0.055 |
drown | -0.049 |
neither | -0.048 |
station | -0.048 |
Token | can |
Token ID | 460 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
gy | +0.372 |
bank | +0.300 |
lves | +0.268 |
zag | +0.263 |
station | +0.247 |
Neg logit contributions | |
ey | -0.287 |
yogurt | -0.251 |
honey | -0.249 |
ad | -0.243 |
idges | -0.242 |
Token | 't |
Token ID | 470 |
Feature activation | -4.49e+00 |
Pos logit contributions | |
ey | +0.267 |
yogurt | +0.234 |
honey | +0.232 |
ad | +0.230 |
idges | +0.228 |
Neg logit contributions | |
gy | -0.301 |
bank | -0.241 |
lves | -0.219 |
zag | -0.204 |
station | -0.195 |
Token | . |
Token ID | 13 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
gy | +0.418 |
bank | +0.338 |
lves | +0.302 |
zag | +0.292 |
station | +0.278 |
Neg logit contributions | |
ey | -0.319 |
yogurt | -0.275 |
honey | -0.273 |
ad | -0.269 |
idges | -0.267 |
Token | , |
Token ID | 11 |
Feature activation | +5.27e-01 |
Pos logit contributions | |
ey | +0.026 |
idges | +0.022 |
ad | +0.021 |
yogurt | +0.020 |
honey | +0.020 |
Neg logit contributions | |
gy | -0.041 |
bank | -0.033 |
zag | -0.029 |
station | -0.029 |
lves | -0.028 |
Token | " |
Token ID | 366 |
Feature activation | +2.52e+00 |
Pos logit contributions | |
ey | +0.038 |
idges | +0.032 |
ad | +0.032 |
yogurt | +0.031 |
honey | +0.030 |
Neg logit contributions | |
gy | -0.049 |
bank | -0.040 |
lves | -0.034 |
zag | -0.034 |
station | -0.033 |
Token | What |
Token ID | 2061 |
Feature activation | +7.18e+00 |
Pos logit contributions | |
ey | +0.148 |
yogurt | +0.118 |
honey | +0.118 |
idges | +0.116 |
ad | +0.115 |
Neg logit contributions | |
gy | -0.268 |
bank | -0.224 |
lves | -0.201 |
zag | -0.197 |
station | -0.190 |
Token | 's |
Token ID | 338 |
Feature activation | -5.92e+00 |
Pos logit contributions | |
ey | +0.603 |
ad | +0.505 |
idges | +0.504 |
honey | +0.491 |
yogurt | +0.488 |
Neg logit contributions | |
gy | -0.583 |
bank | -0.458 |
zag | -0.384 |
lves | -0.372 |
neither | -0.371 |
Token | wrong |
Token ID | 2642 |
Feature activation | -4.10e-01 |
Pos logit contributions | |
gy | +0.538 |
bank | +0.434 |
lves | +0.376 |
zag | +0.367 |
station | +0.353 |
Neg logit contributions | |
ey | -0.447 |
idges | -0.370 |
yogurt | -0.370 |
ad | -0.369 |
honey | -0.366 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
gy | +0.041 |
bank | +0.033 |
zag | +0.029 |
lves | +0.029 |
station | +0.028 |
Neg logit contributions | |
ey | -0.028 |
idges | -0.022 |
yogurt | -0.022 |
ad | -0.022 |
honey | -0.022 |
Token | ?" |
Token ID | 1701 |
Feature activation | +9.16e-01 |
Pos logit contributions | |
ey | +0.699 |
yogurt | +0.565 |
ad | +0.557 |
honey | +0.557 |
idges | +0.557 |
Neg logit contributions | |
gy | -1.122 |
bank | -0.926 |
lves | -0.816 |
zag | -0.812 |
station | -0.778 |
Token | Ana |
Token ID | 17639 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ey | +0.061 |
idges | +0.050 |
yogurt | +0.049 |
honey | +0.049 |
ad | +0.048 |
Neg logit contributions | |
gy | -0.091 |
bank | -0.074 |
lves | -0.065 |
zag | -0.064 |
station | -0.062 |
Token | replied |
Token ID | 8712 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
ey | +0.081 |
yogurt | +0.069 |
idges | +0.068 |
honey | +0.068 |
ad | +0.066 |
Neg logit contributions | |
gy | -0.096 |
bank | -0.078 |
lves | -0.066 |
zag | -0.065 |
station | -0.063 |
Token | , |
Token ID | 11 |
Feature activation | +4.72e-01 |
Pos logit contributions | |
ey | +0.131 |
idges | +0.108 |
yogurt | +0.106 |
ad | +0.106 |
honey | +0.103 |
Neg logit contributions | |
gy | -0.114 |
bank | -0.084 |
zag | -0.070 |
neither | -0.069 |
lves | -0.066 |
Token | " |
Token ID | 366 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
ey | +0.032 |
idges | +0.025 |
yogurt | +0.025 |
ad | +0.025 |
honey | +0.025 |
Neg logit contributions | |
gy | -0.048 |
bank | -0.039 |
zag | -0.034 |
lves | -0.033 |
station | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
gy | +0.073 |
bank | +0.059 |
lves | +0.054 |
drown | +0.051 |
zag | +0.050 |
Neg logit contributions | |
ey | -0.051 |
yogurt | -0.048 |
ad | -0.046 |
honey | -0.045 |
sweet | -0.043 |
Token | said |
Token ID | 531 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
gy | +0.013 |
bank | +0.011 |
lves | +0.009 |
zag | +0.009 |
station | +0.009 |
Neg logit contributions | |
ey | -0.011 |
idges | -0.009 |
yogurt | -0.009 |
ad | -0.009 |
honey | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +8.29e-01 |
Pos logit contributions | |
ey | +0.015 |
idges | +0.012 |
yogurt | +0.012 |
ad | +0.012 |
honey | +0.012 |
Neg logit contributions | |
gy | -0.015 |
bank | -0.012 |
zag | -0.010 |
lves | -0.010 |
neither | -0.010 |
Token | " |
Token ID | 366 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
ey | +0.053 |
idges | +0.043 |
ad | +0.042 |
yogurt | +0.042 |
honey | +0.040 |
Neg logit contributions | |
gy | -0.085 |
bank | -0.070 |
zag | -0.061 |
lves | -0.060 |
station | -0.060 |
Token | No |
Token ID | 2949 |
Feature activation | +4.42e-02 |
Pos logit contributions | |
ey | +0.158 |
honey | +0.129 |
yogurt | +0.125 |
idges | +0.124 |
ad | +0.117 |
Neg logit contributions | |
gy | -0.256 |
bank | -0.212 |
lves | -0.192 |
zag | -0.187 |
station | -0.180 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
ey | +0.004 |
yogurt | +0.003 |
idges | +0.003 |
ad | +0.003 |
honey | +0.003 |
Neg logit contributions | |
gy | -0.004 |
bank | -0.003 |
zag | -0.003 |
lves | -0.003 |
drown | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | +5.83e-01 |
Pos logit contributions | |
ey | +0.742 |
honey | +0.655 |
yogurt | +0.647 |
ad | +0.629 |
idges | +0.625 |
Neg logit contributions | |
gy | -1.019 |
bank | -0.851 |
lves | -0.762 |
zag | -0.711 |
drown | -0.692 |
Token | you |
Token ID | 345 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
ey | +0.040 |
ad | +0.032 |
idges | +0.032 |
yogurt | +0.030 |
ement | +0.030 |
Neg logit contributions | |
gy | -0.056 |
bank | -0.046 |
lves | -0.040 |
neither | -0.039 |
drown | -0.039 |
Token | can |
Token ID | 460 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
ey | +0.161 |
honey | +0.161 |
sweet | +0.157 |
yogurt | +0.153 |
idges | +0.146 |
Neg logit contributions | |
gy | -0.280 |
bank | -0.228 |
lves | -0.210 |
zag | -0.196 |
station | -0.190 |
Token | 't |
Token ID | 470 |
Feature activation | -4.49e+00 |
Pos logit contributions | |
ey | +0.270 |
yogurt | +0.242 |
ad | +0.242 |
honey | +0.237 |
idges | +0.234 |
Neg logit contributions | |
gy | -0.293 |
bank | -0.232 |
lves | -0.213 |
zag | -0.192 |
station | -0.188 |
Token | catch |
Token ID | 4929 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
gy | +0.419 |
bank | +0.339 |
lves | +0.325 |
zag | +0.297 |
station | +0.293 |
Neg logit contributions | |
ey | -0.279 |
ad | -0.266 |
yogurt | -0.264 |
honey | -0.260 |
idges | -0.247 |
Token | ing |
Token ID | 278 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
gy | +0.041 |
bank | +0.034 |
lves | +0.030 |
zag | +0.030 |
station | +0.028 |
Neg logit contributions | |
ey | -0.025 |
yogurt | -0.020 |
idges | -0.019 |
honey | -0.019 |
ad | -0.019 |
Token | ? |
Token ID | 30 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
ey | +0.162 |
yogurt | +0.139 |
honey | +0.138 |
ad | +0.137 |
idges | +0.136 |
Neg logit contributions | |
gy | -0.178 |
bank | -0.141 |
lves | -0.123 |
zag | -0.120 |
drown | -0.113 |
Token | Are |
Token ID | 4231 |
Feature activation | +4.05e+00 |
Pos logit contributions | |
ey | +0.166 |
yogurt | +0.137 |
honey | +0.137 |
idges | +0.135 |
ad | +0.134 |
Neg logit contributions | |
gy | -0.236 |
bank | -0.193 |
lves | -0.169 |
zag | -0.166 |
station | -0.160 |
Token | you |
Token ID | 345 |
Feature activation | +3.07e+00 |
Pos logit contributions | |
ey | +0.248 |
honey | +0.191 |
idges | +0.187 |
ad | +0.187 |
yogurt | +0.187 |
Neg logit contributions | |
gy | -0.422 |
bank | -0.348 |
lves | -0.311 |
zag | -0.309 |
neither | -0.305 |
Token | tired |
Token ID | 10032 |
Feature activation | -3.33e+00 |
Pos logit contributions | |
ey | +0.231 |
yogurt | +0.197 |
honey | +0.196 |
ad | +0.194 |
idges | +0.193 |
Neg logit contributions | |
gy | -0.274 |
bank | -0.221 |
lves | -0.192 |
zag | -0.188 |
station | -0.179 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
gy | +0.333 |
bank | +0.275 |
zag | +0.242 |
drown | +0.233 |
lves | +0.233 |
Neg logit contributions | |
ey | -0.219 |
idges | -0.176 |
yogurt | -0.172 |
honey | -0.172 |
ad | -0.170 |
Token | ?" |
Token ID | 1701 |
Feature activation | +9.16e-01 |
Pos logit contributions | |
ey | +0.702 |
yogurt | +0.571 |
honey | +0.564 |
idges | +0.561 |
ad | +0.560 |
Neg logit contributions | |
gy | -1.115 |
bank | -0.921 |
lves | -0.813 |
zag | -0.806 |
station | -0.770 |
Token | she |
Token ID | 673 |
Feature activation | -9.72e-02 |
Pos logit contributions | |
ey | +0.066 |
idges | +0.055 |
honey | +0.053 |
yogurt | +0.052 |
ad | +0.051 |
Neg logit contributions | |
gy | -0.085 |
bank | -0.070 |
lves | -0.059 |
neither | -0.057 |
zag | -0.057 |
Token | said |
Token ID | 531 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
gy | +0.006 |
bank | +0.005 |
argue | +0.004 |
station | +0.004 |
zag | +0.004 |
Neg logit contributions | |
ey | -0.009 |
idges | -0.008 |
mmmm | -0.008 |
yogurt | -0.008 |
ad | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | +4.72e-01 |
Pos logit contributions | |
gy | +0.010 |
bank | +0.008 |
zag | +0.006 |
neither | +0.006 |
station | +0.006 |
Neg logit contributions | |
ey | -0.011 |
idges | -0.010 |
yogurt | -0.009 |
honey | -0.009 |
ad | -0.009 |
Token | smiling |
Token ID | 16755 |
Feature activation | +7.64e+00 |
Pos logit contributions | |
ey | +0.035 |
idges | +0.029 |
yogurt | +0.028 |
ad | +0.028 |
honey | +0.027 |
Neg logit contributions | |
gy | -0.044 |
bank | -0.035 |
zag | -0.029 |
station | -0.029 |
neither | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
ey | +0.030 |
idges | +0.028 |
ad | +0.026 |
honey | +0.025 |
heart | +0.024 |
Neg logit contributions | |
gy | -0.034 |
bank | -0.027 |
drown | -0.024 |
neither | -0.022 |
lves | -0.022 |
Token | jar |
Token ID | 17379 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
ey | +0.017 |
idges | +0.014 |
yogurt | +0.014 |
ad | +0.014 |
honey | +0.014 |
Neg logit contributions | |
gy | -0.019 |
bank | -0.016 |
lves | -0.013 |
zag | -0.013 |
station | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
ey | +0.032 |
yogurt | +0.026 |
honey | +0.026 |
idges | +0.026 |
ad | +0.026 |
Neg logit contributions | |
gy | -0.043 |
bank | -0.035 |
lves | -0.031 |
zag | -0.030 |
station | -0.029 |
Token | have |
Token ID | 423 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
ey | +0.013 |
idges | +0.011 |
ad | +0.011 |
yogurt | +0.011 |
honey | +0.010 |
Neg logit contributions | |
gy | -0.013 |
bank | -0.010 |
lves | -0.008 |
zag | -0.008 |
station | -0.008 |
Token | some |
Token ID | 617 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
gy | +0.249 |
bank | +0.215 |
drowned | +0.168 |
station | +0.165 |
neither | +0.165 |
Neg logit contributions | |
ey | -0.242 |
idges | -0.195 |
ad | -0.187 |
ango | -0.184 |
ement | -0.182 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
gy | +0.247 |
bank | +0.235 |
station | +0.181 |
drown | +0.176 |
fight | +0.175 |
Neg logit contributions | |
ey | -0.214 |
ango | -0.180 |
idges | -0.175 |
ad | -0.174 |
Jamie | -0.167 |
Token | ," |
Token ID | 553 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
ey | +0.697 |
honey | +0.647 |
yogurt | +0.642 |
sweet | +0.592 |
ad | +0.592 |
Neg logit contributions | |
gy | -1.038 |
bank | -0.855 |
lves | -0.796 |
zag | -0.736 |
drown | -0.731 |
Token | he |
Token ID | 339 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
gy | +0.018 |
bank | +0.015 |
lves | +0.013 |
zag | +0.013 |
neither | +0.012 |
Neg logit contributions | |
ey | -0.013 |
idges | -0.010 |
honey | -0.010 |
yogurt | -0.010 |
ad | -0.010 |
Token | said |
Token ID | 531 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
ey | +0.081 |
idges | +0.067 |
yogurt | +0.066 |
ad | +0.066 |
honey | +0.065 |
Neg logit contributions | |
gy | -0.094 |
bank | -0.077 |
zag | -0.064 |
lves | -0.064 |
station | -0.062 |
Token | to |
Token ID | 284 |
Feature activation | +7.18e-01 |
Pos logit contributions | |
gy | +0.012 |
bank | +0.010 |
lves | +0.008 |
zag | +0.008 |
drown | +0.008 |
Neg logit contributions | |
ey | -0.009 |
yogurt | -0.008 |
honey | -0.008 |
idges | -0.008 |
ad | -0.008 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
ey | +0.046 |
idges | +0.036 |
yogurt | +0.035 |
honey | +0.035 |
layers | +0.033 |
Neg logit contributions | |
gy | -0.075 |
bank | -0.061 |
zag | -0.053 |
drown | -0.052 |
lves | -0.052 |
Token | " |
Token ID | 366 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
gy | +0.107 |
bank | +0.088 |
lves | +0.078 |
zag | +0.076 |
station | +0.073 |
Neg logit contributions | |
ey | -0.073 |
yogurt | -0.059 |
idges | -0.058 |
honey | -0.058 |
ad | -0.058 |
Token | Don |
Token ID | 3987 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
ey | +0.134 |
honey | +0.107 |
idges | +0.105 |
yogurt | +0.105 |
ad | +0.101 |
Neg logit contributions | |
gy | -0.238 |
bank | -0.199 |
lves | -0.177 |
zag | -0.174 |
station | -0.168 |
Token | 't |
Token ID | 470 |
Feature activation | -4.13e+00 |
Pos logit contributions | |
ey | +0.113 |
yogurt | +0.098 |
honey | +0.097 |
idges | +0.096 |
ad | +0.095 |
Neg logit contributions | |
gy | -0.096 |
bank | -0.073 |
zag | -0.061 |
lves | -0.061 |
station | -0.057 |
Token | worry |
Token ID | 5490 |
Feature activation | -4.17e-01 |
Pos logit contributions | |
gy | +0.379 |
bank | +0.310 |
lves | +0.274 |
zag | +0.268 |
station | +0.255 |
Neg logit contributions | |
ey | -0.299 |
ad | -0.258 |
honey | -0.256 |
yogurt | -0.255 |
idges | -0.247 |
Token | , |
Token ID | 11 |
Feature activation | +7.05e-01 |
Pos logit contributions | |
gy | +0.034 |
bank | +0.026 |
zag | +0.021 |
neither | +0.021 |
lves | +0.021 |
Neg logit contributions | |
ey | -0.037 |
ad | -0.030 |
idges | -0.030 |
yogurt | -0.029 |
layers | -0.028 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
ey | +0.054 |
ad | +0.044 |
idges | +0.043 |
ement | +0.042 |
yogurt | +0.042 |
Neg logit contributions | |
gy | -0.063 |
bank | -0.050 |
lves | -0.043 |
neither | -0.043 |
station | -0.043 |
Token | ! |
Token ID | 0 |
Feature activation | +2.54e+00 |
Pos logit contributions | |
ey | +0.802 |
yogurt | +0.674 |
honey | +0.667 |
idges | +0.663 |
ad | +0.662 |
Neg logit contributions | |
gy | -1.015 |
bank | -0.821 |
lves | -0.715 |
zag | -0.704 |
station | -0.670 |
Token | I |
Token ID | 314 |
Feature activation | +3.68e+00 |
Pos logit contributions | |
ey | +0.154 |
yogurt | +0.123 |
idges | +0.121 |
ad | +0.121 |
honey | +0.121 |
Neg logit contributions | |
gy | -0.267 |
bank | -0.220 |
lves | -0.196 |
zag | -0.193 |
drown | -0.187 |
Token | 'm |
Token ID | 1101 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
ey | +0.262 |
ad | +0.237 |
honey | +0.237 |
yogurt | +0.231 |
sweet | +0.229 |
Neg logit contributions | |
gy | -0.309 |
bank | -0.250 |
lves | -0.232 |
zag | -0.215 |
station | -0.204 |
Token | going |
Token ID | 1016 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
gy | +0.268 |
bank | +0.213 |
lves | +0.186 |
drown | +0.180 |
zag | +0.178 |
Neg logit contributions | |
ey | -0.238 |
ad | -0.196 |
idges | -0.195 |
yogurt | -0.195 |
honey | -0.187 |
Token | to |
Token ID | 284 |
Feature activation | +6.96e-01 |
Pos logit contributions | |
ey | +0.197 |
yogurt | +0.178 |
honey | +0.177 |
ad | +0.170 |
idges | +0.168 |
Neg logit contributions | |
gy | -0.289 |
bank | -0.236 |
lves | -0.219 |
zag | -0.205 |
drown | -0.199 |
Token |
|
Token ID | 198 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
ey | +0.275 |
honey | +0.242 |
yogurt | +0.237 |
ad | +0.236 |
sweet | +0.228 |
Neg logit contributions | |
gy | -0.385 |
bank | -0.326 |
lves | -0.293 |
zag | -0.278 |
station | -0.269 |
Token |
|
Token ID | 198 |
Feature activation | +5.00e-01 |
Pos logit contributions | |
ey | +0.014 |
yogurt | +0.012 |
honey | +0.012 |
idges | +0.011 |
ad | +0.011 |
Neg logit contributions | |
gy | -0.024 |
bank | -0.019 |
lves | -0.017 |
zag | -0.017 |
station | -0.016 |
Token | " |
Token ID | 1 |
Feature activation | -7.62e-01 |
Pos logit contributions | |
ey | +0.030 |
yogurt | +0.024 |
ad | +0.024 |
idges | +0.024 |
honey | +0.024 |
Neg logit contributions | |
gy | -0.053 |
bank | -0.044 |
lves | -0.039 |
zag | -0.038 |
drown | -0.037 |
Token | Sure |
Token ID | 19457 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
gy | +0.081 |
bank | +0.068 |
lves | +0.061 |
zag | +0.060 |
station | +0.058 |
Neg logit contributions | |
ey | -0.045 |
honey | -0.035 |
yogurt | -0.035 |
idges | -0.035 |
ad | -0.035 |
Token | , |
Token ID | 11 |
Feature activation | +7.05e-01 |
Pos logit contributions | |
ey | +0.155 |
yogurt | +0.130 |
idges | +0.128 |
honey | +0.128 |
ad | +0.127 |
Neg logit contributions | |
gy | -0.181 |
bank | -0.145 |
lves | -0.125 |
zag | -0.123 |
station | -0.117 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
ey | +0.052 |
idges | +0.041 |
ad | +0.041 |
yogurt | +0.040 |
ement | +0.040 |
Neg logit contributions | |
gy | -0.064 |
bank | -0.052 |
lves | -0.046 |
station | -0.045 |
zag | -0.043 |
Token | , |
Token ID | 11 |
Feature activation | +5.83e-01 |
Pos logit contributions | |
ey | +0.835 |
yogurt | +0.697 |
idges | +0.693 |
ad | +0.684 |
honey | +0.684 |
Neg logit contributions | |
gy | -1.000 |
bank | -0.799 |
zag | -0.687 |
lves | -0.682 |
station | -0.650 |
Token | but |
Token ID | 475 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
ey | +0.040 |
ad | +0.032 |
ement | +0.031 |
idges | +0.031 |
odic | +0.029 |
Neg logit contributions | |
gy | -0.055 |
bank | -0.045 |
station | -0.040 |
lves | -0.039 |
neither | -0.039 |
Token | be |
Token ID | 307 |
Feature activation | +3.85e-01 |
Pos logit contributions | |
gy | +0.101 |
bank | +0.081 |
lves | +0.069 |
zag | +0.068 |
station | +0.065 |
Neg logit contributions | |
ey | -0.087 |
idges | -0.072 |
ad | -0.072 |
yogurt | -0.072 |
honey | -0.070 |
Token | careful |
Token ID | 8161 |
Feature activation | +7.79e-01 |
Pos logit contributions | |
ey | +0.031 |
ad | +0.025 |
idges | +0.025 |
yogurt | +0.025 |
ango | +0.023 |
Neg logit contributions | |
gy | -0.034 |
bank | -0.027 |
lves | -0.023 |
zag | -0.022 |
neither | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
ey | +0.060 |
yogurt | +0.050 |
ad | +0.049 |
idges | +0.049 |
honey | +0.048 |
Neg logit contributions | |
gy | -0.071 |
bank | -0.056 |
zag | -0.049 |
lves | -0.048 |
station | -0.046 |
Token | be |
Token ID | 307 |
Feature activation | +6.92e-01 |
Pos logit contributions | |
ey | +0.032 |
idges | +0.027 |
ad | +0.027 |
yogurt | +0.027 |
honey | +0.026 |
Neg logit contributions | |
gy | -0.042 |
bank | -0.034 |
lves | -0.030 |
zag | -0.029 |
station | -0.028 |
Token | eh |
Token ID | 17231 |
Feature activation | -3.91e+00 |
Pos logit contributions | |
ey | +0.051 |
ad | +0.045 |
yogurt | +0.045 |
honey | +0.044 |
idges | +0.043 |
Neg logit contributions | |
gy | -0.060 |
bank | -0.049 |
lves | -0.043 |
drown | -0.041 |
zag | -0.041 |
Token | ive |
Token ID | 425 |
Feature activation | -5.00e-01 |
Pos logit contributions | |
gy | +0.321 |
bank | +0.251 |
lves | +0.219 |
zag | +0.207 |
station | +0.200 |
Neg logit contributions | |
ey | -0.317 |
yogurt | -0.279 |
ad | -0.276 |
honey | -0.274 |
idges | -0.268 |
Token | full |
Token ID | 1336 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
gy | +0.045 |
bank | +0.036 |
lves | +0.032 |
zag | +0.031 |
drown | +0.030 |
Neg logit contributions | |
ey | -0.037 |
yogurt | -0.031 |
honey | -0.031 |
ad | -0.031 |
idges | -0.031 |
Token | of |
Token ID | 286 |
Feature activation | +3.15e+00 |
Pos logit contributions | |
ey | +0.155 |
idges | +0.121 |
ango | +0.113 |
ad | +0.111 |
subtract | +0.111 |
Neg logit contributions | |
gy | -0.192 |
bank | -0.169 |
station | -0.137 |
zag | -0.135 |
neither | -0.134 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
ey | +0.226 |
idges | +0.180 |
ement | +0.176 |
ango | +0.174 |
ad | +0.173 |
Neg logit contributions | |
gy | -0.276 |
bank | -0.247 |
lves | -0.198 |
drown | -0.197 |
station | -0.186 |
Token | ! |
Token ID | 0 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
ey | +0.808 |
yogurt | +0.667 |
idges | +0.661 |
ad | +0.660 |
honey | +0.652 |
Neg logit contributions | |
gy | -1.011 |
bank | -0.819 |
zag | -0.704 |
lves | -0.702 |
station | -0.667 |
Token | He |
Token ID | 679 |
Feature activation | +3.01e+00 |
Pos logit contributions | |
ey | +0.166 |
idges | +0.136 |
yogurt | +0.135 |
honey | +0.134 |
ad | +0.133 |
Neg logit contributions | |
gy | -0.244 |
bank | -0.200 |
lves | -0.176 |
zag | -0.173 |
station | -0.166 |
Token | was |
Token ID | 373 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
ey | +0.248 |
idges | +0.211 |
yogurt | +0.207 |
ad | +0.205 |
honey | +0.199 |
Neg logit contributions | |
gy | -0.252 |
bank | -0.203 |
zag | -0.166 |
lves | -0.165 |
station | -0.153 |
Token | so |
Token ID | 523 |
Feature activation | +7.38e-01 |
Pos logit contributions | |
ey | +0.108 |
idges | +0.090 |
yogurt | +0.087 |
ad | +0.086 |
honey | +0.083 |
Neg logit contributions | |
gy | -0.115 |
bank | -0.091 |
lves | -0.076 |
zag | -0.076 |
drown | -0.073 |
Token | happy |
Token ID | 3772 |
Feature activation | +9.74e-01 |
Pos logit contributions | |
ey | +0.066 |
idges | +0.056 |
ad | +0.055 |
yogurt | +0.053 |
ango | +0.052 |
Neg logit contributions | |
gy | -0.056 |
bank | -0.043 |
lves | -0.036 |
drown | -0.036 |
zag | -0.035 |
Token | € |
Token ID | 26391 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
gy | +0.497 |
bank | +0.400 |
lves | +0.359 |
zag | +0.341 |
station | +0.328 |
Neg logit contributions | |
ey | -0.416 |
ad | -0.374 |
yogurt | -0.347 |
idges | -0.345 |
honey | -0.338 |
Token | ™ |
Token ID | 8151 |
Feature activation | -4.15e+00 |
Pos logit contributions | |
gy | +0.122 |
bank | +0.103 |
lves | +0.092 |
station | +0.086 |
zag | +0.084 |
Neg logit contributions | |
ey | -0.087 |
ad | -0.074 |
yogurt | -0.074 |
honey | -0.070 |
idges | -0.070 |
Token | m |
Token ID | 76 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
gy | +0.320 |
bank | +0.248 |
lves | +0.210 |
zag | +0.197 |
station | +0.193 |
Neg logit contributions | |
ey | -0.355 |
yogurt | -0.314 |
ad | -0.309 |
idges | -0.307 |
honey | -0.304 |
Token | sorry |
Token ID | 7926 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
ey | +0.161 |
yogurt | +0.135 |
honey | +0.134 |
idges | +0.133 |
ad | +0.130 |
Neg logit contributions | |
gy | -0.184 |
bank | -0.144 |
zag | -0.126 |
lves | -0.123 |
drown | -0.119 |
Token | , |
Token ID | 11 |
Feature activation | +6.34e-01 |
Pos logit contributions | |
ey | +0.130 |
idges | +0.108 |
yogurt | +0.107 |
ad | +0.107 |
honey | +0.105 |
Neg logit contributions | |
gy | -0.146 |
bank | -0.117 |
zag | -0.100 |
lves | -0.098 |
station | -0.094 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
ey | +0.045 |
idges | +0.037 |
ad | +0.036 |
yogurt | +0.036 |
honey | +0.035 |
Neg logit contributions | |
gy | -0.059 |
bank | -0.049 |
lves | -0.042 |
zag | -0.041 |
station | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
ey | +0.758 |
honey | +0.655 |
yogurt | +0.653 |
ad | +0.635 |
idges | +0.629 |
Neg logit contributions | |
gy | -1.015 |
bank | -0.832 |
lves | -0.751 |
zag | -0.707 |
drown | -0.679 |
Token | I |
Token ID | 314 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
ey | +0.024 |
idges | +0.019 |
honey | +0.019 |
yogurt | +0.019 |
ad | +0.019 |
Neg logit contributions | |
gy | -0.037 |
bank | -0.030 |
lves | -0.027 |
zag | -0.026 |
station | -0.026 |
Token | â |
Token ID | 22940 |
Feature activation | -5.69e+00 |
Pos logit contributions | |
ey | +0.271 |
yogurt | +0.240 |
honey | +0.237 |
ad | +0.237 |
sweet | +0.229 |
Neg logit contributions | |
gy | -0.315 |
bank | -0.249 |
lves | -0.227 |
zag | -0.215 |
station | -0.202 |
Token | € |
Token ID | 26391 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
gy | +0.513 |
bank | +0.413 |
lves | +0.359 |
zag | +0.352 |
drown | +0.335 |
Neg logit contributions | |
ey | -0.427 |
yogurt | -0.362 |
honey | -0.359 |
ad | -0.356 |
idges | -0.355 |
Token | ™ |
Token ID | 8151 |
Feature activation | -4.55e+00 |
Pos logit contributions | |
gy | +0.139 |
bank | +0.118 |
lves | +0.104 |
station | +0.094 |
zag | +0.093 |
Neg logit contributions | |
ey | -0.118 |
yogurt | -0.107 |
ad | -0.103 |
goodness | -0.102 |
idges | -0.100 |
Token | . |
Token ID | 13 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
ey | +0.395 |
idges | +0.328 |
yogurt | +0.321 |
ad | +0.318 |
honey | +0.316 |
Neg logit contributions | |
gy | -0.455 |
bank | -0.365 |
zag | -0.310 |
lves | -0.303 |
station | -0.295 |
Token | He |
Token ID | 679 |
Feature activation | +3.05e+00 |
Pos logit contributions | |
ey | +0.027 |
idges | +0.026 |
ad | +0.024 |
illy | +0.022 |
inges | +0.022 |
Neg logit contributions | |
gy | -0.030 |
bank | -0.024 |
drown | -0.021 |
neither | -0.019 |
zag | -0.019 |
Token | does |
Token ID | 857 |
Feature activation | +4.59e+00 |
Pos logit contributions | |
ey | +0.216 |
idges | +0.190 |
ad | +0.188 |
importantly | +0.178 |
mmmm | +0.176 |
Neg logit contributions | |
gy | -0.267 |
bank | -0.232 |
lves | -0.185 |
station | -0.181 |
drown | -0.180 |
Token | not |
Token ID | 407 |
Feature activation | -2.97e+00 |
Pos logit contributions | |
ey | +0.435 |
idges | +0.367 |
ad | +0.367 |
yogurt | +0.365 |
honey | +0.361 |
Neg logit contributions | |
gy | -0.325 |
bank | -0.251 |
zag | -0.202 |
lves | -0.197 |
drown | -0.196 |
Token | like |
Token ID | 588 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
gy | +0.252 |
bank | +0.201 |
lves | +0.170 |
zag | +0.167 |
drown | +0.162 |
Neg logit contributions | |
ey | -0.239 |
idges | -0.202 |
yogurt | -0.201 |
honey | -0.199 |
ad | -0.198 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
ey | +0.124 |
ement | +0.101 |
idges | +0.100 |
ad | +0.099 |
odic | +0.095 |
Neg logit contributions | |
gy | -0.150 |
bank | -0.128 |
drown | -0.108 |
neither | -0.106 |
lves | -0.105 |
Token | . |
Token ID | 13 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
ey | +0.782 |
honey | +0.701 |
yogurt | +0.697 |
ad | +0.670 |
idges | +0.662 |
Neg logit contributions | |
gy | -0.972 |
bank | -0.784 |
lves | -0.707 |
zag | -0.673 |
drown | -0.641 |
Token | But |
Token ID | 887 |
Feature activation | -2.98e+00 |
Pos logit contributions | |
ey | +0.031 |
idges | +0.029 |
ad | +0.028 |
illy | +0.025 |
ement | +0.024 |
Neg logit contributions | |
gy | -0.032 |
bank | -0.025 |
drown | -0.022 |
neither | -0.022 |
house | -0.020 |
Token | he |
Token ID | 339 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
gy | +0.263 |
bank | +0.212 |
neither | +0.196 |
zag | +0.183 |
drown | +0.180 |
Neg logit contributions | |
ey | -0.229 |
idges | -0.189 |
ad | -0.184 |
layers | -0.170 |
ement | -0.168 |
Token | listens |
Token ID | 35019 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
ey | +0.078 |
idges | +0.066 |
ad | +0.065 |
yogurt | +0.062 |
honey | +0.061 |
Neg logit contributions | |
gy | -0.097 |
bank | -0.082 |
lves | -0.068 |
zag | -0.067 |
drown | -0.067 |
Token | to |
Token ID | 284 |
Feature activation | +8.10e-01 |
Pos logit contributions | |
ey | +0.271 |
idges | +0.225 |
ad | +0.223 |
yogurt | +0.210 |
honey | +0.206 |
Neg logit contributions | |
gy | -0.184 |
bank | -0.143 |
zag | -0.121 |
neither | -0.107 |
station | -0.106 |
Token | with |
Token ID | 351 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
ey | +0.683 |
yogurt | +0.637 |
honey | +0.635 |
sweet | +0.589 |
ad | +0.576 |
Neg logit contributions | |
gy | -1.046 |
bank | -0.862 |
lves | -0.806 |
zag | -0.741 |
drown | -0.735 |
Token | me |
Token ID | 502 |
Feature activation | +6.08e+00 |
Pos logit contributions | |
ey | +0.006 |
ement | +0.004 |
yogurt | +0.004 |
idges | +0.004 |
ad | +0.004 |
Neg logit contributions | |
gy | -0.013 |
bank | -0.011 |
lves | -0.009 |
drowned | -0.009 |
neither | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
ey | +0.452 |
yogurt | +0.378 |
honey | +0.375 |
idges | +0.373 |
ad | +0.373 |
Neg logit contributions | |
gy | -0.556 |
bank | -0.447 |
lves | -0.388 |
zag | -0.384 |
station | -0.363 |
Token | I |
Token ID | 314 |
Feature activation | +3.69e+00 |
Pos logit contributions | |
ey | +0.040 |
ad | +0.034 |
idges | +0.033 |
yogurt | +0.032 |
ement | +0.032 |
Neg logit contributions | |
gy | -0.046 |
bank | -0.037 |
station | -0.033 |
lves | -0.032 |
drown | -0.031 |
Token | love |
Token ID | 1842 |
Feature activation | +4.54e+00 |
Pos logit contributions | |
ey | +0.235 |
yogurt | +0.222 |
sweet | +0.214 |
honey | +0.212 |
ad | +0.207 |
Neg logit contributions | |
gy | -0.334 |
bank | -0.271 |
lves | -0.256 |
zag | -0.243 |
station | -0.234 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
ey | +0.356 |
ement | +0.301 |
ad | +0.285 |
idges | +0.275 |
Jamie | +0.267 |
Neg logit contributions | |
gy | -0.384 |
bank | -0.316 |
lves | -0.258 |
neither | -0.250 |
house | -0.245 |
Token | too |
Token ID | 1165 |
Feature activation | -8.64e-01 |
Pos logit contributions | |
ey | +0.715 |
honey | +0.644 |
yogurt | +0.640 |
ad | +0.602 |
idges | +0.601 |
Neg logit contributions | |
gy | -1.036 |
bank | -0.855 |
lves | -0.773 |
zag | -0.733 |
drown | -0.714 |
Token | !" |
Token ID | 2474 |
Feature activation | +6.48e-01 |
Pos logit contributions | |
gy | +0.079 |
bank | +0.064 |
lves | +0.055 |
zag | +0.055 |
drown | +0.052 |
Neg logit contributions | |
ey | -0.064 |
yogurt | -0.053 |
ad | -0.053 |
idges | -0.052 |
honey | -0.052 |
Token | But |
Token ID | 887 |
Feature activation | -2.98e+00 |
Pos logit contributions | |
ey | +0.042 |
yogurt | +0.036 |
honey | +0.036 |
ad | +0.035 |
idges | +0.034 |
Neg logit contributions | |
gy | -0.063 |
bank | -0.051 |
lves | -0.046 |
zag | -0.046 |
drown | -0.044 |
Token | the |
Token ID | 262 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
gy | +0.318 |
bank | +0.276 |
zag | +0.255 |
lves | +0.252 |
drown | +0.241 |
Neg logit contributions | |
ey | -0.140 |
honey | -0.128 |
yogurt | -0.125 |
sweet | -0.123 |
ad | -0.122 |
Token | bird |
Token ID | 6512 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
ey | +0.024 |
idges | +0.021 |
ad | +0.020 |
yogurt | +0.020 |
honey | +0.019 |
Neg logit contributions | |
gy | -0.018 |
bank | -0.014 |
zag | -0.011 |
lves | -0.011 |
station | -0.010 |
Token | a |
Token ID | 257 |
Feature activation | +4.42e-01 |
Pos logit contributions | |
gy | +0.221 |
bank | +0.173 |
lves | +0.154 |
drown | +0.145 |
zag | +0.144 |
Neg logit contributions | |
ey | -0.191 |
idges | -0.163 |
ad | -0.162 |
yogurt | -0.157 |
honey | -0.153 |
Token | little |
Token ID | 1310 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
ey | +0.029 |
yogurt | +0.025 |
honey | +0.025 |
sweet | +0.024 |
ad | +0.023 |
Neg logit contributions | |
gy | -0.044 |
bank | -0.036 |
lves | -0.032 |
zag | -0.031 |
drown | -0.029 |
Token | bit |
Token ID | 1643 |
Feature activation | -5.96e+00 |
Pos logit contributions | |
ey | +0.108 |
honey | +0.105 |
sweet | +0.103 |
yogurt | +0.102 |
mirror | +0.093 |
Neg logit contributions | |
gy | -0.108 |
bank | -0.082 |
zag | -0.075 |
lves | -0.074 |
drown | -0.063 |
Token | sick |
Token ID | 6639 |
Feature activation | -7.33e+00 |
Pos logit contributions | |
gy | +0.477 |
bank | +0.380 |
lves | +0.324 |
zag | +0.304 |
drown | +0.299 |
Neg logit contributions | |
ey | -0.504 |
ad | -0.428 |
idges | -0.425 |
yogurt | -0.420 |
honey | -0.409 |
Token | , |
Token ID | 11 |
Feature activation | +5.40e-01 |
Pos logit contributions | |
gy | +0.609 |
bank | +0.487 |
lves | +0.393 |
zag | +0.393 |
station | +0.386 |
Neg logit contributions | |
ey | -0.607 |
idges | -0.511 |
ad | -0.500 |
yogurt | -0.499 |
honey | -0.485 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
ey | +0.048 |
ad | +0.043 |
idges | +0.040 |
ango | +0.039 |
odic | +0.038 |
Neg logit contributions | |
gy | -0.037 |
bank | -0.029 |
neither | -0.028 |
house | -0.025 |
station | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
ey | +0.725 |
honey | +0.665 |
yogurt | +0.643 |
ad | +0.616 |
sweet | +0.612 |
Neg logit contributions | |
gy | -1.008 |
bank | -0.834 |
lves | -0.765 |
zag | -0.714 |
drown | -0.691 |
Token | But |
Token ID | 887 |
Feature activation | -2.98e+00 |
Pos logit contributions | |
ey | +0.028 |
ad | +0.024 |
idges | +0.023 |
yogurt | +0.022 |
ement | +0.021 |
Neg logit contributions | |
gy | -0.037 |
bank | -0.029 |
lves | -0.026 |
station | -0.025 |
house | -0.025 |
Token | don |
Token ID | 836 |
Feature activation | -2.53e+00 |
Pos logit contributions | |
gy | +0.258 |
bank | +0.200 |
neither | +0.182 |
station | +0.168 |
lves | +0.167 |
Neg logit contributions | |
ey | -0.235 |
ad | -0.199 |
idges | -0.191 |
ango | -0.184 |
ement | -0.184 |
Token | 't |
Token ID | 470 |
Feature activation | -4.40e+00 |
Pos logit contributions | |
gy | +0.216 |
bank | +0.172 |
lves | +0.148 |
zag | +0.141 |
station | +0.135 |
Neg logit contributions | |
ey | -0.200 |
idges | -0.174 |
honey | -0.174 |
yogurt | -0.174 |
ad | -0.171 |
Token | worry |
Token ID | 5490 |
Feature activation | -6.30e-01 |
Pos logit contributions | |
gy | +0.365 |
bank | +0.288 |
lves | +0.244 |
zag | +0.238 |
drown | +0.229 |
Neg logit contributions | |
ey | -0.364 |
yogurt | -0.310 |
idges | -0.309 |
ad | -0.308 |
honey | -0.306 |
Token | jar |
Token ID | 17379 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
ey | +0.737 |
yogurt | +0.634 |
honey | +0.631 |
ad | +0.608 |
idges | +0.590 |
Neg logit contributions | |
gy | -1.032 |
bank | -0.833 |
lves | -0.768 |
zag | -0.722 |
drown | -0.710 |
Token | . |
Token ID | 13 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
ey | +0.013 |
yogurt | +0.011 |
honey | +0.011 |
ad | +0.011 |
idges | +0.011 |
Neg logit contributions | |
gy | -0.018 |
bank | -0.014 |
lves | -0.013 |
zag | -0.012 |
drown | -0.012 |
Token | The |
Token ID | 383 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
ey | +0.024 |
idges | +0.020 |
ad | +0.019 |
yogurt | +0.019 |
honey | +0.019 |
Neg logit contributions | |
gy | -0.036 |
bank | -0.030 |
lves | -0.026 |
zag | -0.025 |
station | -0.025 |
Token | bear |
Token ID | 6842 |
Feature activation | -9.05e-01 |
Pos logit contributions | |
gy | +0.202 |
bank | +0.164 |
lves | +0.132 |
station | +0.128 |
zag | +0.127 |
Neg logit contributions | |
ey | -0.217 |
idges | -0.187 |
ad | -0.183 |
yogurt | -0.180 |
honey | -0.172 |
Token | loved |
Token ID | 6151 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
gy | +0.086 |
bank | +0.069 |
lves | +0.065 |
drown | +0.062 |
zag | +0.061 |
Neg logit contributions | |
ey | -0.057 |
yogurt | -0.050 |
honey | -0.049 |
ad | -0.049 |
sweet | -0.048 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
ey | +0.207 |
ement | +0.177 |
idges | +0.168 |
ad | +0.163 |
odic | +0.157 |
Neg logit contributions | |
gy | -0.227 |
bank | -0.191 |
lves | -0.157 |
fight | -0.150 |
neither | -0.149 |
Token | , |
Token ID | 11 |
Feature activation | +5.52e-01 |
Pos logit contributions | |
ey | +0.826 |
idges | +0.680 |
ad | +0.666 |
yogurt | +0.665 |
honey | +0.651 |
Neg logit contributions | |
gy | -1.001 |
bank | -0.806 |
zag | -0.686 |
lves | -0.680 |
station | -0.652 |
Token | so |
Token ID | 523 |
Feature activation | +8.41e-01 |
Pos logit contributions | |
ey | +0.046 |
idges | +0.039 |
ad | +0.038 |
ement | +0.037 |
yogurt | +0.036 |
Neg logit contributions | |
gy | -0.044 |
bank | -0.037 |
lves | -0.029 |
neither | -0.029 |
station | -0.029 |
Token | he |
Token ID | 339 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ey | +0.047 |
yogurt | +0.038 |
honey | +0.037 |
ad | +0.036 |
idges | +0.036 |
Neg logit contributions | |
gy | -0.092 |
bank | -0.078 |
lves | -0.070 |
zag | -0.069 |
station | -0.066 |
Token | decided |
Token ID | 3066 |
Feature activation | -9.75e-01 |
Pos logit contributions | |
ey | +0.082 |
honey | +0.077 |
yogurt | +0.075 |
ad | +0.073 |
sweet | +0.072 |
Neg logit contributions | |
gy | -0.091 |
bank | -0.069 |
lves | -0.063 |
zag | -0.061 |
drown | -0.059 |
Token | to |
Token ID | 284 |
Feature activation | +8.02e-01 |
Pos logit contributions | |
gy | +0.084 |
bank | +0.067 |
lves | +0.057 |
zag | +0.056 |
drown | +0.053 |
Neg logit contributions | |
ey | -0.079 |
yogurt | -0.066 |
idges | -0.066 |
honey | -0.066 |
ad | -0.066 |
Token | It |
Token ID | 1026 |
Feature activation | +4.42e+00 |
Pos logit contributions | |
gy | +0.123 |
bank | +0.105 |
lves | +0.093 |
zag | +0.092 |
station | +0.089 |
Neg logit contributions | |
ey | -0.054 |
yogurt | -0.043 |
idges | -0.041 |
ad | -0.041 |
honey | -0.040 |
Token | 's |
Token ID | 338 |
Feature activation | -6.18e+00 |
Pos logit contributions | |
ey | +0.341 |
honey | +0.332 |
sweet | +0.314 |
yogurt | +0.305 |
idges | +0.303 |
Neg logit contributions | |
gy | -0.325 |
bank | -0.284 |
lves | -0.243 |
drown | -0.219 |
zag | -0.216 |
Token | an |
Token ID | 281 |
Feature activation | -3.01e+00 |
Pos logit contributions | |
gy | +0.603 |
bank | +0.507 |
lves | +0.448 |
zag | +0.438 |
drown | +0.414 |
Neg logit contributions | |
ey | -0.408 |
honey | -0.353 |
yogurt | -0.346 |
ad | -0.334 |
idges | -0.332 |
Token | earthquake |
Token ID | 16295 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
gy | +0.288 |
bank | +0.233 |
lves | +0.209 |
zag | +0.208 |
drown | +0.199 |
Neg logit contributions | |
ey | -0.207 |
honey | -0.178 |
yogurt | -0.176 |
ad | -0.176 |
idges | -0.168 |
Token | , |
Token ID | 11 |
Feature activation | +5.37e-01 |
Pos logit contributions | |
gy | +0.118 |
bank | +0.098 |
lves | +0.087 |
zag | +0.084 |
drown | +0.081 |
Neg logit contributions | |
ey | -0.083 |
yogurt | -0.069 |
honey | -0.069 |
ad | -0.068 |
idges | -0.068 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
ey | +0.038 |
ad | +0.032 |
idges | +0.030 |
odic | +0.030 |
subtract | +0.030 |
Neg logit contributions | |
gy | -0.047 |
bank | -0.038 |
neither | -0.036 |
station | -0.034 |
drown | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
ey | +0.681 |
honey | +0.615 |
yogurt | +0.584 |
ad | +0.569 |
sweet | +0.558 |
Neg logit contributions | |
gy | -1.039 |
bank | -0.877 |
lves | -0.817 |
zag | -0.740 |
drown | -0.719 |
Token | It |
Token ID | 632 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
ey | +0.025 |
yogurt | +0.021 |
idges | +0.021 |
honey | +0.020 |
ad | +0.020 |
Neg logit contributions | |
gy | -0.040 |
bank | -0.032 |
lves | -0.028 |
zag | -0.028 |
drown | -0.027 |
Token | 's |
Token ID | 338 |
Feature activation | -6.15e+00 |
Pos logit contributions | |
ey | +0.091 |
honey | +0.084 |
idges | +0.078 |
sweet | +0.077 |
ad | +0.075 |
Neg logit contributions | |
gy | -0.128 |
bank | -0.112 |
lves | -0.095 |
zag | -0.093 |
suits | -0.089 |
Token | when |
Token ID | 618 |
Feature activation | -6.71e-01 |
Pos logit contributions | |
gy | +0.572 |
bank | +0.487 |
lves | +0.420 |
zag | +0.411 |
station | +0.382 |
Neg logit contributions | |
ey | -0.417 |
honey | -0.384 |
sweet | -0.357 |
yogurt | -0.355 |
idges | -0.351 |
Token | the |
Token ID | 262 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
gy | +0.067 |
bank | +0.055 |
lves | +0.049 |
zag | +0.048 |
station | +0.045 |
Neg logit contributions | |
ey | -0.044 |
yogurt | -0.037 |
honey | -0.037 |
idges | -0.036 |
ad | -0.035 |
Token | and |
Token ID | 290 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
gy | +0.288 |
bank | +0.244 |
lves | +0.230 |
drown | +0.227 |
neither | +0.220 |
Neg logit contributions | |
ey | -0.165 |
ad | -0.156 |
honey | -0.154 |
sweet | -0.152 |
yogurt | -0.145 |
Token | Red |
Token ID | 2297 |
Feature activation | -2.41e+00 |
Pos logit contributions | |
gy | +0.013 |
bank | +0.010 |
zag | +0.010 |
drown | +0.009 |
lves | +0.009 |
Neg logit contributions | |
ey | -0.010 |
honey | -0.009 |
sweet | -0.009 |
yogurt | -0.009 |
ad | -0.009 |
Token | went |
Token ID | 1816 |
Feature activation | -8.23e-01 |
Pos logit contributions | |
gy | +0.197 |
lves | +0.163 |
bank | +0.159 |
zag | +0.158 |
drown | +0.157 |
Neg logit contributions | |
ad | -0.138 |
sweet | -0.132 |
yogurt | -0.125 |
honey | -0.124 |
ey | -0.116 |
Token | on |
Token ID | 319 |
Feature activation | -8.21e-01 |
Pos logit contributions | |
gy | +0.077 |
lves | +0.063 |
bank | +0.062 |
neither | +0.058 |
drown | +0.055 |
Neg logit contributions | |
ey | -0.048 |
yogurt | -0.048 |
honey | -0.046 |
sweet | -0.043 |
pasta | -0.043 |
Token | a |
Token ID | 257 |
Feature activation | +4.60e-01 |
Pos logit contributions | |
gy | +0.090 |
bank | +0.075 |
lves | +0.071 |
zag | +0.069 |
drown | +0.066 |
Neg logit contributions | |
ey | -0.041 |
yogurt | -0.037 |
honey | -0.035 |
sweet | -0.034 |
ad | -0.034 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
sweet | +0.026 |
yogurt | +0.025 |
honey | +0.024 |
pasta | +0.023 |
delicious | +0.023 |
Neg logit contributions | |
gy | -0.047 |
neither | -0.035 |
bank | -0.034 |
lves | -0.034 |
ung | -0.032 |
Token | moon |
Token ID | 22977 |
Feature activation | -5.67e+00 |
Pos logit contributions | |
honey | +0.628 |
yogurt | +0.621 |
ad | +0.604 |
ey | +0.596 |
pasta | +0.593 |
Neg logit contributions | |
gy | -0.932 |
lves | -0.797 |
bank | -0.730 |
drown | -0.715 |
argue | -0.686 |
Token | . |
Token ID | 13 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
gy | +0.603 |
lves | +0.509 |
bank | +0.500 |
drown | +0.489 |
argue | +0.448 |
Neg logit contributions | |
ad | -0.239 |
sweet | -0.225 |
honey | -0.225 |
ey | -0.206 |
yogurt | -0.204 |
Token | They |
Token ID | 1119 |
Feature activation | -3.59e+00 |
Pos logit contributions | |
ey | +0.025 |
honey | +0.022 |
sweet | +0.022 |
yogurt | +0.022 |
cereal | +0.022 |
Neg logit contributions | |
gy | -0.035 |
bank | -0.027 |
zag | -0.026 |
lves | -0.026 |
drown | -0.025 |
Token | sw |
Token ID | 1509 |
Feature activation | -5.35e+00 |
Pos logit contributions | |
gy | +0.334 |
bank | +0.254 |
zag | +0.253 |
drown | +0.240 |
lves | +0.238 |
Neg logit contributions | |
yogurt | -0.215 |
honey | -0.214 |
ey | -0.214 |
sweet | -0.204 |
ad | -0.204 |
Token | am |
Token ID | 321 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
gy | +0.439 |
bank | +0.369 |
lves | +0.324 |
drown | +0.316 |
station | +0.312 |
Neg logit contributions | |
ey | -0.402 |
idges | -0.351 |
ad | -0.351 |
honey | -0.345 |
yogurt | -0.336 |
Token | he |
Token ID | 339 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
gy | +0.013 |
bank | +0.011 |
lves | +0.009 |
neither | +0.009 |
zag | +0.009 |
Neg logit contributions | |
ey | -0.012 |
idges | -0.010 |
ad | -0.010 |
ement | -0.010 |
yogurt | -0.010 |
Token | picked |
Token ID | 6497 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
ey | +0.084 |
idges | +0.070 |
yogurt | +0.070 |
ad | +0.069 |
honey | +0.068 |
Neg logit contributions | |
gy | -0.093 |
bank | -0.075 |
lves | -0.064 |
zag | -0.062 |
station | -0.058 |
Token | some |
Token ID | 617 |
Feature activation | -3.18e+00 |
Pos logit contributions | |
ey | +0.172 |
importantly | +0.143 |
idges | +0.142 |
ad | +0.134 |
subtract | +0.134 |
Neg logit contributions | |
gy | -0.141 |
bank | -0.114 |
lves | -0.088 |
neither | -0.084 |
zag | -0.080 |
Token | of |
Token ID | 286 |
Feature activation | +3.05e+00 |
Pos logit contributions | |
bank | +0.237 |
gy | +0.233 |
away | +0.190 |
bit | +0.189 |
after | +0.180 |
Neg logit contributions | |
ey | -0.226 |
Adam | -0.188 |
Jake | -0.187 |
David | -0.185 |
ad | -0.184 |
Token | the |
Token ID | 262 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
ey | +0.249 |
idges | +0.203 |
ement | +0.199 |
subtract | +0.194 |
ention | +0.194 |
Neg logit contributions | |
gy | -0.251 |
bank | -0.204 |
neither | -0.168 |
drown | -0.164 |
lves | -0.160 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
ey | +0.010 |
ad | +0.008 |
idges | +0.008 |
Adam | +0.007 |
subtract | +0.007 |
Neg logit contributions | |
gy | -0.026 |
bank | -0.023 |
lves | -0.020 |
zag | -0.020 |
house | -0.019 |
Token | to |
Token ID | 284 |
Feature activation | +7.98e-01 |
Pos logit contributions | |
ey | +0.836 |
idges | +0.696 |
yogurt | +0.693 |
ad | +0.691 |
honey | +0.683 |
Neg logit contributions | |
gy | -0.980 |
bank | -0.790 |
zag | -0.670 |
lves | -0.669 |
station | -0.632 |
Token | eat |
Token ID | 4483 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
ey | +0.076 |
idges | +0.066 |
ement | +0.066 |
ango | +0.064 |
importantly | +0.063 |
Neg logit contributions | |
gy | -0.055 |
bank | -0.045 |
drown | -0.032 |
house | -0.031 |
fight | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | +3.86e-01 |
Pos logit contributions | |
ey | +0.180 |
idges | +0.147 |
ad | +0.141 |
importantly | +0.140 |
subtract | +0.140 |
Neg logit contributions | |
gy | -0.137 |
bank | -0.121 |
neither | -0.084 |
station | -0.081 |
house | -0.080 |
Token | He |
Token ID | 679 |
Feature activation | +3.07e+00 |
Pos logit contributions | |
ey | +0.027 |
idges | +0.022 |
ad | +0.022 |
honey | +0.021 |
yogurt | +0.021 |
Neg logit contributions | |
gy | -0.038 |
bank | -0.032 |
lves | -0.027 |
zag | -0.026 |
station | -0.026 |
Token | had |
Token ID | 550 |
Feature activation | -7.26e-01 |
Pos logit contributions | |
ey | +0.256 |
idges | +0.216 |
yogurt | +0.213 |
ad | +0.212 |
honey | +0.209 |
Neg logit contributions | |
gy | -0.255 |
bank | -0.204 |
lves | -0.169 |
zag | -0.165 |
station | -0.156 |
Token | woods |
Token ID | 16479 |
Feature activation | -6.21e-01 |
Pos logit contributions | |
ey | +0.016 |
idges | +0.015 |
ad | +0.014 |
yogurt | +0.014 |
honey | +0.014 |
Neg logit contributions | |
gy | -0.018 |
bank | -0.015 |
lves | -0.012 |
zag | -0.012 |
station | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
gy | +0.062 |
bank | +0.050 |
lves | +0.046 |
zag | +0.044 |
drown | +0.042 |
Neg logit contributions | |
ey | -0.040 |
yogurt | -0.033 |
ad | -0.033 |
honey | -0.033 |
idges | -0.032 |
Token | It |
Token ID | 632 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
ey | +0.027 |
idges | +0.025 |
ad | +0.023 |
honey | +0.022 |
heart | +0.022 |
Neg logit contributions | |
gy | -0.033 |
bank | -0.026 |
drown | -0.022 |
station | -0.021 |
lves | -0.021 |
Token | smelled |
Token ID | 37609 |
Feature activation | +3.80e+00 |
Pos logit contributions | |
ey | +0.116 |
idges | +0.100 |
yogurt | +0.098 |
honey | +0.096 |
ad | +0.095 |
Neg logit contributions | |
gy | -0.116 |
bank | -0.090 |
zag | -0.076 |
lves | -0.075 |
station | -0.070 |
Token | the |
Token ID | 262 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
ey | +0.334 |
idges | +0.298 |
subtract | +0.289 |
ango | +0.286 |
ement | +0.276 |
Neg logit contributions | |
gy | -0.252 |
bank | -0.228 |
drown | -0.205 |
neither | -0.186 |
drowned | -0.178 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
ey | +0.014 |
idges | +0.013 |
ad | +0.013 |
mmmm | +0.012 |
Jamie | +0.012 |
Neg logit contributions | |
gy | -0.020 |
bank | -0.019 |
drown | -0.015 |
lves | -0.015 |
station | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +3.80e-01 |
Pos logit contributions | |
ey | +0.815 |
idges | +0.677 |
yogurt | +0.673 |
ad | +0.667 |
honey | +0.660 |
Neg logit contributions | |
gy | -1.001 |
bank | -0.811 |
zag | -0.695 |
lves | -0.682 |
station | -0.658 |
Token | It |
Token ID | 632 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
ey | +0.029 |
idges | +0.027 |
ad | +0.025 |
honey | +0.023 |
heart | +0.023 |
Neg logit contributions | |
gy | -0.032 |
bank | -0.025 |
drown | -0.022 |
neither | -0.021 |
station | -0.020 |
Token | saw |
Token ID | 2497 |
Feature activation | -8.52e-01 |
Pos logit contributions | |
ey | +0.121 |
idges | +0.106 |
yogurt | +0.101 |
ummy | +0.100 |
ad | +0.100 |
Neg logit contributions | |
gy | -0.116 |
bank | -0.090 |
zag | -0.074 |
lves | -0.074 |
sank | -0.071 |
Token | the |
Token ID | 262 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
gy | +0.064 |
bank | +0.053 |
lves | +0.046 |
neither | +0.044 |
drown | +0.044 |
Neg logit contributions | |
ey | -0.070 |
idges | -0.060 |
subtract | -0.057 |
mmmm | -0.057 |
ad | -0.057 |
Token | broken |
Token ID | 5445 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
ey | +0.018 |
idges | +0.017 |
ad | +0.016 |
mmmm | +0.015 |
ement | +0.015 |
Neg logit contributions | |
gy | -0.020 |
bank | -0.017 |
lves | -0.014 |
drown | -0.014 |
station | -0.014 |
Token | are |
Token ID | 389 |
Feature activation | +4.65e-01 |
Pos logit contributions | |
gy | +0.138 |
bank | +0.112 |
lves | +0.105 |
zag | +0.100 |
drown | +0.096 |
Neg logit contributions | |
ey | -0.070 |
honey | -0.059 |
yogurt | -0.058 |
ad | -0.058 |
idges | -0.057 |
Token | no |
Token ID | 645 |
Feature activation | -4.47e+00 |
Pos logit contributions | |
ey | +0.031 |
honey | +0.027 |
yogurt | +0.026 |
ad | +0.025 |
idges | +0.025 |
Neg logit contributions | |
gy | -0.045 |
bank | -0.037 |
lves | -0.033 |
zag | -0.032 |
drown | -0.031 |
Token | more |
Token ID | 517 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
gy | +0.465 |
bank | +0.377 |
lves | +0.344 |
zag | +0.333 |
drown | +0.317 |
Neg logit contributions | |
ey | -0.273 |
honey | -0.235 |
yogurt | -0.232 |
ad | -0.224 |
idges | -0.222 |
Token | seats |
Token ID | 8632 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
ey | +0.128 |
yogurt | +0.106 |
honey | +0.106 |
idges | +0.104 |
ad | +0.103 |
Neg logit contributions | |
gy | -0.199 |
bank | -0.163 |
lves | -0.146 |
zag | -0.142 |
drown | -0.136 |
Token | , |
Token ID | 11 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
gy | +0.115 |
bank | +0.093 |
lves | +0.084 |
zag | +0.080 |
drown | +0.077 |
Neg logit contributions | |
ey | -0.086 |
yogurt | -0.075 |
honey | -0.074 |
ad | -0.073 |
idges | -0.072 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
ey | +0.039 |
ad | +0.033 |
idges | +0.032 |
yogurt | +0.032 |
honey | +0.032 |
Neg logit contributions | |
gy | -0.048 |
bank | -0.039 |
lves | -0.034 |
zag | -0.033 |
station | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | +3.80e-01 |
Pos logit contributions | |
ey | +0.622 |
honey | +0.599 |
yogurt | +0.572 |
sweet | +0.539 |
ad | +0.536 |
Neg logit contributions | |
gy | -1.047 |
bank | -0.866 |
lves | -0.842 |
zag | -0.751 |
drown | -0.738 |
Token | The |
Token ID | 383 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
ey | +0.027 |
yogurt | +0.022 |
honey | +0.022 |
idges | +0.022 |
ad | +0.022 |
Neg logit contributions | |
gy | -0.036 |
bank | -0.029 |
lves | -0.025 |
zag | -0.025 |
station | -0.024 |
Token | theater |
Token ID | 13766 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
gy | +0.221 |
bank | +0.188 |
station | +0.160 |
zag | +0.160 |
lves | +0.155 |
Neg logit contributions | |
ey | -0.182 |
idges | -0.152 |
ad | -0.150 |
yogurt | -0.148 |
honey | -0.144 |
Token | is |
Token ID | 318 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
ey | +0.070 |
yogurt | +0.058 |
honey | +0.058 |
idges | +0.056 |
ad | +0.055 |
Neg logit contributions | |
gy | -0.128 |
bank | -0.105 |
lves | -0.097 |
zag | -0.092 |
drown | -0.090 |
Token | full |
Token ID | 1336 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
ey | +0.137 |
honey | +0.119 |
yogurt | +0.118 |
ad | +0.113 |
idges | +0.113 |
Neg logit contributions | |
gy | -0.184 |
bank | -0.149 |
lves | -0.133 |
zag | -0.129 |
drown | -0.122 |
Token | for |
Token ID | 329 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
gy | +0.061 |
bank | +0.049 |
lves | +0.043 |
zag | +0.042 |
drown | +0.040 |
Neg logit contributions | |
ey | -0.051 |
yogurt | -0.044 |
honey | -0.043 |
ad | -0.042 |
idges | -0.042 |
Token | dinner |
Token ID | 8073 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
gy | +0.128 |
bank | +0.109 |
fight | +0.089 |
argue | +0.089 |
drown | +0.087 |
Neg logit contributions | |
ey | -0.128 |
idges | -0.097 |
ement | -0.096 |
elling | -0.091 |
ad | -0.089 |
Token | . |
Token ID | 13 |
Feature activation | +3.61e-01 |
Pos logit contributions | |
ey | +0.170 |
yogurt | +0.144 |
honey | +0.143 |
ad | +0.140 |
idges | +0.140 |
Neg logit contributions | |
gy | -0.237 |
bank | -0.193 |
lves | -0.171 |
zag | -0.167 |
drown | -0.160 |
Token | She |
Token ID | 1375 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
ey | +0.032 |
ad | +0.029 |
idges | +0.029 |
odic | +0.027 |
honey | +0.027 |
Neg logit contributions | |
gy | -0.026 |
bank | -0.020 |
lves | -0.016 |
house | -0.016 |
zag | -0.015 |
Token | liked |
Token ID | 8288 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
ey | +0.144 |
yogurt | +0.126 |
honey | +0.125 |
ad | +0.122 |
idges | +0.120 |
Neg logit contributions | |
gy | -0.173 |
bank | -0.138 |
lves | -0.122 |
zag | -0.119 |
drown | -0.114 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
ey | +0.146 |
ement | +0.130 |
ad | +0.125 |
idges | +0.120 |
mmmm | +0.116 |
Neg logit contributions | |
gy | -0.079 |
bank | -0.059 |
neither | -0.045 |
fight | -0.044 |
argue | -0.042 |
Token | so |
Token ID | 523 |
Feature activation | +8.19e-01 |
Pos logit contributions | |
ey | +0.856 |
idges | +0.700 |
ad | +0.697 |
yogurt | +0.695 |
honey | +0.670 |
Neg logit contributions | |
gy | -0.972 |
bank | -0.772 |
zag | -0.666 |
lves | -0.650 |
station | -0.625 |
Token | much |
Token ID | 881 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
ey | +0.074 |
idges | +0.060 |
ad | +0.060 |
yogurt | +0.058 |
honey | +0.058 |
Neg logit contributions | |
gy | -0.061 |
bank | -0.046 |
lves | -0.039 |
zag | -0.039 |
neither | -0.038 |
Token | that |
Token ID | 326 |
Feature activation | -5.79e-01 |
Pos logit contributions | |
gy | +0.096 |
bank | +0.077 |
zag | +0.066 |
neither | +0.063 |
they | +0.062 |
Neg logit contributions | |
ey | -0.162 |
idges | -0.134 |
ad | -0.132 |
ement | -0.128 |
Jamie | -0.127 |
Token | she |
Token ID | 673 |
Feature activation | -7.37e-02 |
Pos logit contributions | |
gy | +0.054 |
bank | +0.044 |
neither | +0.038 |
zag | +0.037 |
lves | +0.037 |
Neg logit contributions | |
ey | -0.042 |
idges | -0.034 |
ad | -0.034 |
ement | -0.032 |
yogurt | -0.031 |
Token | did |
Token ID | 750 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
gy | +0.006 |
bank | +0.005 |
lves | +0.004 |
zag | +0.004 |
drown | +0.004 |
Neg logit contributions | |
ey | -0.006 |
honey | -0.005 |
yogurt | -0.005 |
ad | -0.005 |
idges | -0.005 |
Token | hive |
Token ID | 35881 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
ey | +0.015 |
honey | +0.014 |
yogurt | +0.013 |
sweet | +0.013 |
ad | +0.012 |
Neg logit contributions | |
gy | -0.018 |
bank | -0.014 |
lves | -0.013 |
zag | -0.013 |
drown | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
gy | +0.217 |
bank | +0.180 |
lves | +0.168 |
zag | +0.155 |
drown | +0.154 |
Neg logit contributions | |
ey | -0.135 |
yogurt | -0.116 |
honey | -0.115 |
ad | -0.111 |
sweet | -0.109 |
Token | They |
Token ID | 1119 |
Feature activation | -3.62e+00 |
Pos logit contributions | |
ey | +0.030 |
idges | +0.026 |
ad | +0.025 |
honey | +0.025 |
yogurt | +0.025 |
Neg logit contributions | |
gy | -0.029 |
bank | -0.023 |
lves | -0.019 |
zag | -0.019 |
station | -0.019 |
Token | made |
Token ID | 925 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
gy | +0.349 |
bank | +0.282 |
lves | +0.259 |
zag | +0.255 |
drown | +0.244 |
Neg logit contributions | |
ey | -0.228 |
honey | -0.206 |
yogurt | -0.205 |
ad | -0.199 |
sweet | -0.198 |
Token | sweet |
Token ID | 6029 |
Feature activation | +9.49e+00 |
Pos logit contributions | |
gy | +0.119 |
bank | +0.100 |
lves | +0.087 |
zag | +0.084 |
drown | +0.083 |
Neg logit contributions | |
ey | -0.077 |
ad | -0.060 |
idges | -0.060 |
yogurt | -0.059 |
ement | -0.055 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
ey | +0.701 |
ad | +0.578 |
idges | +0.563 |
subtract | +0.538 |
yogurt | +0.534 |
Neg logit contributions | |
gy | -0.846 |
bank | -0.706 |
lves | -0.571 |
station | -0.552 |
zag | -0.551 |
Token | . |
Token ID | 13 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
honey | +0.666 |
ey | +0.653 |
sweet | +0.617 |
yogurt | +0.584 |
ad | +0.547 |
Neg logit contributions | |
gy | -1.029 |
bank | -0.857 |
lves | -0.842 |
drowned | -0.740 |
zag | -0.739 |
Token | They |
Token ID | 1119 |
Feature activation | -3.60e+00 |
Pos logit contributions | |
ey | +0.030 |
idges | +0.025 |
ad | +0.025 |
honey | +0.025 |
yogurt | +0.024 |
Neg logit contributions | |
gy | -0.032 |
bank | -0.025 |
lves | -0.021 |
zag | -0.021 |
station | -0.021 |
Token | worked |
Token ID | 3111 |
Feature activation | -7.05e-01 |
Pos logit contributions | |
gy | +0.333 |
bank | +0.266 |
lves | +0.244 |
zag | +0.239 |
drown | +0.226 |
Neg logit contributions | |
ey | -0.247 |
honey | -0.223 |
yogurt | -0.222 |
ad | -0.213 |
sweet | -0.210 |
Token | hard |
Token ID | 1327 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
gy | +0.065 |
bank | +0.053 |
lves | +0.048 |
drown | +0.045 |
zag | +0.045 |
Neg logit contributions | |
ey | -0.049 |
honey | -0.043 |
yogurt | -0.043 |
sweet | -0.042 |
idges | -0.042 |
Token | every |
Token ID | 790 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
ey | +0.089 |
honey | +0.082 |
ad | +0.079 |
yogurt | +0.079 |
sweet | +0.077 |
Neg logit contributions | |
gy | -0.164 |
bank | -0.135 |
lves | -0.130 |
drown | -0.121 |
station | -0.118 |
Token | |
Token ID | 220 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
ey | +0.135 |
honey | +0.112 |
yogurt | +0.112 |
ad | +0.110 |
idges | +0.106 |
Neg logit contributions | |
gy | -0.222 |
bank | -0.184 |
lves | -0.164 |
zag | -0.163 |
drown | -0.156 |
Token |
|
Token ID | 198 |
Feature activation | -5.44e-02 |
Pos logit contributions | |
gy | +0.126 |
bank | +0.101 |
zag | +0.090 |
lves | +0.089 |
drown | +0.085 |
Neg logit contributions | |
ey | -0.089 |
yogurt | -0.073 |
honey | -0.072 |
idges | -0.071 |
ad | -0.071 |
Token |
|
Token ID | 198 |
Feature activation | -5.55e-02 |
Pos logit contributions | |
gy | +0.005 |
bank | +0.004 |
zag | +0.004 |
lves | +0.004 |
drown | +0.003 |
Neg logit contributions | |
ey | -0.004 |
yogurt | -0.003 |
idges | -0.003 |
honey | -0.003 |
ad | -0.003 |
Token | T |
Token ID | 51 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
gy | +0.006 |
bank | +0.005 |
lves | +0.004 |
zag | +0.004 |
drown | +0.004 |
Neg logit contributions | |
ey | -0.004 |
honey | -0.003 |
yogurt | -0.003 |
ad | -0.003 |
idges | -0.003 |
Token | ig |
Token ID | 328 |
Feature activation | -3.99e+00 |
Pos logit contributions | |
gy | +0.252 |
bank | +0.209 |
lves | +0.188 |
zag | +0.186 |
station | +0.178 |
Neg logit contributions | |
ey | -0.132 |
yogurt | -0.106 |
honey | -0.104 |
idges | -0.103 |
ad | -0.101 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
gy | +0.298 |
bank | +0.239 |
lves | +0.206 |
drown | +0.190 |
station | +0.186 |
Neg logit contributions | |
ey | -0.340 |
ad | -0.302 |
honey | -0.298 |
yogurt | -0.297 |
idges | -0.295 |
Token | was |
Token ID | 373 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
gy | +1.512 |
bank | +1.223 |
lves | +1.095 |
zag | +1.068 |
drown | +1.036 |
Neg logit contributions | |
ey | -1.092 |
yogurt | -0.928 |
ad | -0.926 |
honey | -0.923 |
idges | -0.902 |
Token | so |
Token ID | 523 |
Feature activation | +5.34e-01 |
Pos logit contributions | |
ey | +0.084 |
idges | +0.070 |
yogurt | +0.069 |
honey | +0.067 |
ad | +0.066 |
Neg logit contributions | |
gy | -0.099 |
bank | -0.079 |
zag | -0.068 |
lves | -0.068 |
drown | -0.065 |
Token | happy |
Token ID | 3772 |
Feature activation | +7.94e-01 |
Pos logit contributions | |
ey | +0.046 |
idges | +0.039 |
yogurt | +0.038 |
honey | +0.038 |
ad | +0.038 |
Neg logit contributions | |
gy | -0.041 |
bank | -0.032 |
zag | -0.028 |
drown | -0.028 |
neither | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
ey | +0.069 |
idges | +0.057 |
yogurt | +0.055 |
ad | +0.054 |
layers | +0.053 |
Neg logit contributions | |
gy | -0.065 |
bank | -0.050 |
zag | -0.042 |
neither | -0.040 |
lves | -0.040 |
Token | the |
Token ID | 262 |
Feature activation | -5.80e-02 |
Pos logit contributions | |
gy | +0.040 |
bank | +0.032 |
lves | +0.028 |
zag | +0.027 |
neither | +0.026 |
Neg logit contributions | |
ey | -0.032 |
idges | -0.026 |
yogurt | -0.026 |
ad | -0.026 |
honey | -0.025 |
Token | day |
Token ID | 1110 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
gy | +0.473 |
bank | +0.415 |
lves | +0.369 |
drown | +0.348 |
argue | +0.343 |
Neg logit contributions | |
ey | -0.373 |
sweet | -0.372 |
honey | -0.365 |
yogurt | -0.354 |
ad | -0.340 |
Token | , |
Token ID | 11 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
gy | +0.242 |
bank | +0.188 |
zag | +0.159 |
lves | +0.157 |
neither | +0.149 |
Neg logit contributions | |
ey | -0.244 |
idges | -0.205 |
ad | -0.204 |
honey | -0.202 |
yogurt | -0.202 |
Token | a |
Token ID | 257 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
ey | +0.017 |
idges | +0.014 |
ad | +0.013 |
yogurt | +0.013 |
honey | +0.013 |
Neg logit contributions | |
gy | -0.024 |
bank | -0.019 |
lves | -0.017 |
zag | -0.017 |
neither | -0.016 |
Token | little |
Token ID | 1310 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
ey | +0.009 |
honey | +0.008 |
sweet | +0.008 |
yogurt | +0.008 |
idges | +0.007 |
Neg logit contributions | |
gy | -0.018 |
bank | -0.015 |
lves | -0.014 |
zag | -0.014 |
drown | -0.013 |
Token | pig |
Token ID | 12967 |
Feature activation | -5.40e+00 |
Pos logit contributions | |
honey | +0.058 |
sweet | +0.056 |
ey | +0.056 |
yogurt | +0.050 |
ummy | +0.045 |
Neg logit contributions | |
gy | -0.105 |
bank | -0.089 |
zag | -0.083 |
drown | -0.080 |
drowned | -0.080 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
gy | +0.420 |
bank | +0.411 |
lves | +0.394 |
drown | +0.392 |
argue | +0.370 |
Neg logit contributions | |
ad | -0.302 |
ey | -0.292 |
yogurt | -0.251 |
sweet | -0.239 |
honey | -0.233 |
Token | named |
Token ID | 3706 |
Feature activation | -6.48e+00 |
Pos logit contributions | |
gy | +1.458 |
drown | +1.246 |
lves | +1.236 |
zag | +1.226 |
argue | +1.190 |
Neg logit contributions | |
ad | -0.749 |
honey | -0.687 |
sweet | -0.686 |
ey | -0.659 |
yogurt | -0.658 |
Token | Pe |
Token ID | 2631 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
gy | +0.628 |
bank | +0.515 |
lves | +0.453 |
zag | +0.446 |
drown | +0.425 |
Neg logit contributions | |
ey | -0.446 |
yogurt | -0.373 |
honey | -0.370 |
ad | -0.365 |
idges | -0.364 |
Token | ppa |
Token ID | 44989 |
Feature activation | -3.78e+00 |
Pos logit contributions | |
gy | +0.240 |
bank | +0.200 |
drown | +0.172 |
zag | +0.167 |
lves | +0.167 |
Neg logit contributions | |
ey | -0.189 |
idges | -0.171 |
honey | -0.170 |
ad | -0.168 |
yogurt | -0.164 |
Token | went |
Token ID | 1816 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
gy | +0.336 |
bank | +0.296 |
lves | +0.275 |
drown | +0.267 |
zag | +0.262 |
Neg logit contributions | |
ey | -0.221 |
ad | -0.206 |
yogurt | -0.199 |
honey | -0.195 |
sweet | -0.183 |
Token | for |
Token ID | 329 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
gy | +0.111 |
bank | +0.092 |
lves | +0.091 |
neither | +0.084 |
drown | +0.083 |
Neg logit contributions | |
ey | -0.059 |
honey | -0.057 |
yogurt | -0.054 |
sweet | -0.050 |
ad | -0.049 |
Token | that |
Token ID | 326 |
Feature activation | -8.72e-01 |
Pos logit contributions | |
ey | +0.078 |
yogurt | +0.065 |
idges | +0.064 |
honey | +0.063 |
layers | +0.061 |
Neg logit contributions | |
gy | -0.083 |
bank | -0.065 |
lves | -0.057 |
zag | -0.054 |
station | -0.053 |
Token | day |
Token ID | 1110 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
gy | +0.071 |
bank | +0.057 |
lves | +0.050 |
drown | +0.046 |
zag | +0.045 |
Neg logit contributions | |
ey | -0.071 |
honey | -0.063 |
ad | -0.063 |
yogurt | -0.062 |
sweet | -0.062 |
Token | on |
Token ID | 319 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
gy | +0.275 |
bank | +0.242 |
lves | +0.205 |
drown | +0.196 |
station | +0.190 |
Neg logit contributions | |
ey | -0.184 |
ad | -0.171 |
idges | -0.164 |
honey | -0.161 |
importantly | -0.157 |
Token | , |
Token ID | 11 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
gy | +0.103 |
bank | +0.084 |
lves | +0.075 |
zag | +0.072 |
drown | +0.070 |
Neg logit contributions | |
ey | -0.079 |
honey | -0.067 |
yogurt | -0.066 |
ad | -0.066 |
idges | -0.065 |
Token | Mag |
Token ID | 2944 |
Feature activation | -5.23e+00 |
Pos logit contributions | |
ey | +0.014 |
ad | +0.014 |
sweet | +0.014 |
honey | +0.013 |
idges | +0.013 |
Neg logit contributions | |
gy | -0.022 |
bank | -0.019 |
drown | -0.016 |
argue | -0.016 |
zag | -0.016 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
gy | +0.035 |
bank | +0.028 |
drown | -0.039 |
argue | -0.045 |
neither | -0.052 |
Neg logit contributions | |
ey | -0.748 |
ad | -0.702 |
idges | -0.685 |
ummy | -0.661 |
ement | -0.656 |
Token | was |
Token ID | 373 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
gy | +1.515 |
bank | +1.246 |
lves | +1.174 |
drown | +1.113 |
zag | +1.101 |
Neg logit contributions | |
ey | -0.958 |
ad | -0.892 |
honey | -0.843 |
yogurt | -0.840 |
sweet | -0.815 |
Token | never |
Token ID | 1239 |
Feature activation | -4.90e+00 |
Pos logit contributions | |
ey | +0.085 |
honey | +0.076 |
ad | +0.076 |
yogurt | +0.075 |
idges | +0.074 |
Neg logit contributions | |
gy | -0.090 |
bank | -0.072 |
lves | -0.062 |
zag | -0.060 |
drown | -0.057 |
Token | fearful |
Token ID | 26844 |
Feature activation | -4.69e+00 |
Pos logit contributions | |
gy | +0.420 |
bank | +0.342 |
zag | +0.296 |
lves | +0.291 |
station | +0.275 |
Neg logit contributions | |
ey | -0.375 |
honey | -0.331 |
yogurt | -0.329 |
ad | -0.321 |
sweet | -0.314 |
Token | of |
Token ID | 286 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
gy | +0.427 |
bank | +0.344 |
lves | +0.302 |
zag | +0.295 |
drown | +0.286 |
Neg logit contributions | |
ey | -0.346 |
yogurt | -0.283 |
idges | -0.282 |
ad | -0.278 |
honey | -0.277 |
Token | boards |
Token ID | 11490 |
Feature activation | +9.65e-02 |
Pos logit contributions | |
ey | +0.194 |
idges | +0.159 |
yogurt | +0.157 |
ad | +0.155 |
honey | +0.151 |
Neg logit contributions | |
gy | -0.258 |
bank | -0.212 |
lves | -0.184 |
zag | -0.179 |
drown | -0.176 |
Token | together |
Token ID | 1978 |
Feature activation | -3.59e+00 |
Pos logit contributions | |
gy | +0.015 |
bank | +0.011 |
lves | +0.009 |
zag | +0.009 |
station | +0.008 |
Neg logit contributions | |
ey | -0.021 |
yogurt | -0.018 |
honey | -0.018 |
idges | -0.018 |
ad | -0.018 |
Token | all |
Token ID | 477 |
Feature activation | -3.96e+00 |
Pos logit contributions | |
gy | +0.352 |
bank | +0.278 |
lves | +0.274 |
drown | +0.250 |
zag | +0.243 |
Neg logit contributions | |
ey | -0.217 |
honey | -0.201 |
yogurt | -0.200 |
ad | -0.195 |
sweet | -0.188 |
Token | day |
Token ID | 1110 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
gy | +0.320 |
bank | +0.268 |
lves | +0.228 |
drown | +0.223 |
zag | +0.215 |
Neg logit contributions | |
ey | -0.308 |
ad | -0.280 |
honey | -0.274 |
yogurt | -0.268 |
sweet | -0.266 |
Token | . |
Token ID | 13 |
Feature activation | +8.16e-02 |
Pos logit contributions | |
gy | +0.314 |
lves | +0.254 |
bank | +0.251 |
drown | +0.238 |
station | +0.236 |
Neg logit contributions | |
sweet | -0.124 |
ey | -0.121 |
honey | -0.115 |
yogurt | -0.114 |
ummy | -0.109 |
Token | Bug |
Token ID | 15217 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
ey | +0.004 |
honey | +0.004 |
ad | +0.004 |
yogurt | +0.004 |
sweet | +0.004 |
Neg logit contributions | |
gy | -0.008 |
bank | -0.007 |
zag | -0.006 |
lves | -0.006 |
argue | -0.006 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
gy | +0.040 |
bank | +0.017 |
lves | -0.016 |
drown | -0.017 |
station | -0.026 |
Neg logit contributions | |
ey | -0.450 |
honey | -0.432 |
ad | -0.432 |
sweet | -0.426 |
idges | -0.416 |
Token | learned |
Token ID | 4499 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
gy | +1.530 |
bank | +1.245 |
lves | +1.188 |
drown | +1.133 |
zag | +1.120 |
Neg logit contributions | |
ey | -0.911 |
ad | -0.871 |
honey | -0.849 |
sweet | -0.829 |
yogurt | -0.825 |
Token | that |
Token ID | 326 |
Feature activation | -8.81e-01 |
Pos logit contributions | |
gy | +0.203 |
bank | +0.164 |
zag | +0.139 |
lves | +0.138 |
station | +0.129 |
Neg logit contributions | |
ey | -0.188 |
yogurt | -0.153 |
idges | -0.152 |
ad | -0.148 |
honey | -0.148 |
Token | being |
Token ID | 852 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
gy | +0.081 |
bank | +0.065 |
lves | +0.057 |
zag | +0.056 |
station | +0.053 |
Neg logit contributions | |
ey | -0.065 |
yogurt | -0.055 |
honey | -0.054 |
idges | -0.054 |
ad | -0.054 |
Token | loyal |
Token ID | 9112 |
Feature activation | -8.71e-01 |
Pos logit contributions | |
ey | +0.112 |
honey | +0.090 |
yogurt | +0.090 |
idges | +0.088 |
ad | +0.087 |
Neg logit contributions | |
gy | -0.213 |
bank | -0.180 |
lves | -0.159 |
zag | -0.158 |
station | -0.152 |
Token | is |
Token ID | 318 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
gy | +0.081 |
bank | +0.067 |
lves | +0.059 |
zag | +0.058 |
drown | +0.055 |
Neg logit contributions | |
ey | -0.061 |
honey | -0.052 |
yogurt | -0.051 |
ad | -0.050 |
idges | -0.050 |
Token | body |
Token ID | 1767 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
gy | +0.043 |
lves | +0.035 |
bank | +0.034 |
zag | +0.031 |
drown | +0.030 |
Neg logit contributions | |
ey | -0.039 |
honey | -0.035 |
sweet | -0.034 |
yogurt | -0.034 |
ad | -0.032 |
Token | and |
Token ID | 290 |
Feature activation | -4.30e-01 |
Pos logit contributions | |
gy | +0.103 |
bank | +0.086 |
lves | +0.084 |
drown | +0.079 |
argue | +0.076 |
Neg logit contributions | |
ey | -0.055 |
yogurt | -0.047 |
ad | -0.046 |
honey | -0.044 |
sweet | -0.044 |
Token | a |
Token ID | 257 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
gy | +0.037 |
bank | +0.029 |
zag | +0.025 |
lves | +0.023 |
neither | +0.023 |
Neg logit contributions | |
ey | -0.035 |
idges | -0.030 |
ement | -0.029 |
ad | -0.029 |
yogurt | -0.028 |
Token | w |
Token ID | 266 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
ey | +0.011 |
honey | +0.009 |
yogurt | +0.009 |
sweet | +0.009 |
ad | +0.008 |
Neg logit contributions | |
gy | -0.016 |
bank | -0.013 |
lves | -0.012 |
zag | -0.011 |
drown | -0.011 |
Token | ag |
Token ID | 363 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
ey | +0.095 |
yogurt | +0.079 |
honey | +0.078 |
idges | +0.075 |
ad | +0.074 |
Neg logit contributions | |
gy | -0.152 |
bank | -0.125 |
zag | -0.112 |
lves | -0.112 |
drown | -0.106 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
gy | +0.085 |
bank | +0.078 |
lves | +0.067 |
drown | +0.062 |
station | +0.061 |
Neg logit contributions | |
ey | -0.095 |
ad | -0.089 |
yogurt | -0.079 |
honey | -0.079 |
idges | -0.079 |
Token | tail |
Token ID | 7894 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
gy | +1.253 |
lves | +0.999 |
drown | +0.944 |
neither | +0.903 |
bank | +0.891 |
Neg logit contributions | |
ey | -1.180 |
honey | -1.092 |
yogurt | -1.085 |
sweet | -1.053 |
ad | -1.008 |
Token | . |
Token ID | 13 |
Feature activation | +7.72e-02 |
Pos logit contributions | |
gy | +0.281 |
bank | +0.229 |
lves | +0.213 |
zag | +0.200 |
drown | +0.198 |
Neg logit contributions | |
ey | -0.173 |
yogurt | -0.148 |
ad | -0.146 |
honey | -0.145 |
idges | -0.138 |
Token | One |
Token ID | 1881 |
Feature activation | -5.54e+00 |
Pos logit contributions | |
ey | +0.007 |
idges | +0.006 |
honey | +0.006 |
ad | +0.006 |
illy | +0.006 |
Neg logit contributions | |
gy | -0.006 |
bank | -0.005 |
station | -0.004 |
zag | -0.003 |
house | -0.003 |
Token | day |
Token ID | 1110 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
gy | +0.436 |
bank | +0.354 |
lves | +0.304 |
zag | +0.275 |
drown | +0.275 |
Neg logit contributions | |
ey | -0.469 |
yogurt | -0.421 |
honey | -0.417 |
ad | -0.408 |
sweet | -0.404 |
Token | , |
Token ID | 11 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
gy | +0.240 |
bank | +0.184 |
zag | +0.156 |
lves | +0.153 |
neither | +0.147 |
Neg logit contributions | |
ey | -0.248 |
idges | -0.210 |
ad | -0.208 |
honey | -0.207 |
yogurt | -0.205 |
Token | was |
Token ID | 373 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
gy | +0.371 |
bank | +0.257 |
zag | +0.213 |
lves | +0.209 |
station | +0.184 |
Neg logit contributions | |
ey | -0.518 |
yogurt | -0.457 |
idges | -0.443 |
honey | -0.438 |
ad | -0.438 |
Token | happy |
Token ID | 3772 |
Feature activation | +7.98e-01 |
Pos logit contributions | |
ey | +0.081 |
yogurt | +0.067 |
idges | +0.067 |
honey | +0.066 |
ad | +0.065 |
Neg logit contributions | |
gy | -0.104 |
bank | -0.084 |
lves | -0.073 |
zag | -0.072 |
station | -0.069 |
Token | to |
Token ID | 284 |
Feature activation | +4.81e-01 |
Pos logit contributions | |
ey | +0.075 |
idges | +0.062 |
yogurt | +0.061 |
ad | +0.060 |
layers | +0.059 |
Neg logit contributions | |
gy | -0.061 |
bank | -0.044 |
neither | -0.036 |
lves | -0.036 |
zag | -0.036 |
Token | see |
Token ID | 766 |
Feature activation | +7.86e-01 |
Pos logit contributions | |
ey | +0.035 |
ad | +0.032 |
honey | +0.030 |
yogurt | +0.030 |
sweet | +0.029 |
Neg logit contributions | |
gy | -0.042 |
bank | -0.035 |
lves | -0.032 |
zag | -0.029 |
station | -0.028 |
Token | Bug |
Token ID | 15217 |
Feature activation | -3.33e+00 |
Pos logit contributions | |
ey | +0.068 |
yogurt | +0.055 |
ad | +0.053 |
idges | +0.053 |
subtract | +0.052 |
Neg logit contributions | |
gy | -0.066 |
bank | -0.051 |
lves | -0.047 |
neither | -0.040 |
drown | -0.040 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
gy | +0.012 |
bank | -0.013 |
lves | -0.041 |
drown | -0.051 |
zag | -0.056 |
Neg logit contributions | |
ey | -0.472 |
honey | -0.461 |
ad | -0.456 |
sweet | -0.454 |
yogurt | -0.443 |
Token | and |
Token ID | 290 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
gy | +1.434 |
bank | +1.134 |
lves | +1.029 |
zag | +0.994 |
drown | +0.939 |
Neg logit contributions | |
ey | -1.173 |
yogurt | -1.015 |
honey | -1.013 |
ad | -0.995 |
idges | -0.978 |
Token | they |
Token ID | 484 |
Feature activation | -7.28e+00 |
Pos logit contributions | |
gy | +0.041 |
bank | +0.033 |
lves | +0.029 |
zag | +0.029 |
station | +0.028 |
Neg logit contributions | |
ey | -0.030 |
honey | -0.026 |
yogurt | -0.025 |
ad | -0.025 |
idges | -0.025 |
Token | played |
Token ID | 2826 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
gy | +0.666 |
bank | +0.559 |
zag | +0.504 |
drown | +0.496 |
lves | +0.492 |
Neg logit contributions | |
ey | -0.427 |
ad | -0.425 |
honey | -0.413 |
sweet | -0.401 |
yogurt | -0.380 |
Token | together |
Token ID | 1978 |
Feature activation | -3.59e+00 |
Pos logit contributions | |
gy | +0.015 |
bank | +0.011 |
lves | +0.009 |
zag | +0.009 |
station | +0.008 |
Neg logit contributions | |
ey | -0.021 |
yogurt | -0.018 |
honey | -0.018 |
idges | -0.018 |
ad | -0.018 |
Token | all |
Token ID | 477 |
Feature activation | -3.96e+00 |
Pos logit contributions | |
gy | +0.352 |
bank | +0.278 |
lves | +0.274 |
drown | +0.250 |
zag | +0.243 |
Neg logit contributions | |
ey | -0.217 |
honey | -0.201 |
yogurt | -0.200 |
ad | -0.195 |
sweet | -0.188 |
Token | what |
Token ID | 644 |
Feature activation | +5.14e+00 |
Pos logit contributions | |
ey | +0.053 |
yogurt | +0.040 |
ad | +0.038 |
idges | +0.037 |
ement | +0.037 |
Neg logit contributions | |
gy | -0.085 |
bank | -0.068 |
lves | -0.063 |
drown | -0.059 |
neither | -0.058 |
Token | was |
Token ID | 373 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
ey | +0.428 |
idges | +0.347 |
yogurt | +0.344 |
honey | +0.344 |
ad | +0.339 |
Neg logit contributions | |
gy | -0.443 |
bank | -0.340 |
zag | -0.291 |
lves | -0.290 |
neither | -0.283 |
Token | wrong |
Token ID | 2642 |
Feature activation | -8.79e-01 |
Pos logit contributions | |
ey | +0.089 |
yogurt | +0.076 |
honey | +0.076 |
idges | +0.075 |
ad | +0.075 |
Neg logit contributions | |
gy | -0.094 |
bank | -0.074 |
lves | -0.064 |
zag | -0.063 |
station | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | +8.53e-02 |
Pos logit contributions | |
gy | +0.087 |
bank | +0.073 |
lves | +0.067 |
zag | +0.063 |
drown | +0.061 |
Neg logit contributions | |
ey | -0.055 |
honey | -0.047 |
yogurt | -0.047 |
ad | -0.046 |
idges | -0.044 |
Token | Bug |
Token ID | 15217 |
Feature activation | -3.33e+00 |
Pos logit contributions | |
ey | +0.005 |
honey | +0.005 |
sweet | +0.005 |
idges | +0.005 |
yogurt | +0.004 |
Neg logit contributions | |
gy | -0.008 |
bank | -0.006 |
zag | -0.006 |
lves | -0.006 |
station | -0.005 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
gy | +0.025 |
bank | +0.008 |
lves | -0.018 |
drown | -0.023 |
station | -0.032 |
Neg logit contributions | |
ey | -0.449 |
ad | -0.435 |
honey | -0.431 |
sweet | -0.426 |
idges | -0.415 |
Token | told |
Token ID | 1297 |
Feature activation | -3.61e+00 |
Pos logit contributions | |
gy | +1.480 |
bank | +1.174 |
lves | +1.096 |
zag | +1.073 |
drown | +1.039 |
Neg logit contributions | |
ey | -1.052 |
ad | -0.944 |
honey | -0.941 |
yogurt | -0.931 |
idges | -0.898 |
Token | them |
Token ID | 606 |
Feature activation | -9.40e-01 |
Pos logit contributions | |
gy | +0.385 |
bank | +0.318 |
lves | +0.281 |
zag | +0.272 |
drown | +0.269 |
Neg logit contributions | |
ey | -0.225 |
yogurt | -0.180 |
idges | -0.172 |
ad | -0.166 |
honey | -0.164 |
Token | about |
Token ID | 546 |
Feature activation | +4.46e+00 |
Pos logit contributions | |
gy | +0.089 |
bank | +0.072 |
lves | +0.062 |
zag | +0.061 |
drown | +0.059 |
Neg logit contributions | |
ey | -0.067 |
yogurt | -0.055 |
ad | -0.054 |
idges | -0.054 |
honey | -0.053 |
Token | the |
Token ID | 262 |
Feature activation | -6.13e-02 |
Pos logit contributions | |
ey | +0.306 |
ad | +0.240 |
idges | +0.239 |
yogurt | +0.233 |
ement | +0.230 |
Neg logit contributions | |
gy | -0.442 |
bank | -0.364 |
lves | -0.312 |
drown | -0.305 |
neither | -0.303 |
Token | cart |
Token ID | 6383 |
Feature activation | -4.62e+00 |
Pos logit contributions | |
gy | +0.006 |
bank | +0.004 |
lves | +0.004 |
zag | +0.004 |
station | +0.004 |
Neg logit contributions | |
ey | -0.005 |
yogurt | -0.004 |
honey | -0.004 |
idges | -0.004 |
ad | -0.004 |
Token | Tim |
Token ID | 5045 |
Feature activation | -3.99e-01 |
Pos logit contributions | |
ey | +0.008 |
idges | +0.007 |
honey | +0.007 |
ad | +0.006 |
yogurt | +0.006 |
Neg logit contributions | |
gy | -0.008 |
bank | -0.006 |
zag | -0.005 |
house | -0.005 |
station | -0.005 |
Token | my |
Token ID | 1820 |
Feature activation | +2.99e+00 |
Pos logit contributions | |
gy | +0.040 |
bank | +0.031 |
lves | +0.027 |
zag | +0.027 |
station | +0.026 |
Neg logit contributions | |
ey | -0.028 |
idges | -0.024 |
honey | -0.023 |
yogurt | -0.022 |
ad | -0.022 |
Token | had |
Token ID | 550 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
ey | +0.174 |
yogurt | +0.149 |
ad | +0.145 |
sweet | +0.140 |
honey | +0.136 |
Neg logit contributions | |
gy | -0.296 |
bank | -0.258 |
lves | -0.236 |
drown | -0.226 |
zag | -0.225 |
Token | a |
Token ID | 257 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
gy | +0.104 |
bank | +0.085 |
lves | +0.075 |
zag | +0.073 |
station | +0.070 |
Neg logit contributions | |
ey | -0.070 |
idges | -0.057 |
honey | -0.056 |
yogurt | -0.055 |
ad | -0.055 |
Token | pig |
Token ID | 12967 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
ey | +0.012 |
yogurt | +0.010 |
honey | +0.010 |
sweet | +0.010 |
pasta | +0.009 |
Neg logit contributions | |
gy | -0.017 |
bank | -0.014 |
lves | -0.013 |
zag | -0.012 |
drown | -0.012 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
gy | +0.281 |
drown | +0.269 |
bank | +0.260 |
lves | +0.250 |
argue | +0.233 |
Neg logit contributions | |
ad | -0.438 |
ey | -0.427 |
sweet | -0.425 |
yogurt | -0.417 |
honey | -0.401 |
Token | bank |
Token ID | 3331 |
Feature activation | -1.24e+01 |
Pos logit contributions | |
gy | +1.038 |
lves | +0.824 |
drown | +0.751 |
bank | +0.716 |
zag | +0.705 |
Neg logit contributions | |
ey | -1.346 |
yogurt | -1.310 |
honey | -1.300 |
sweet | -1.253 |
ad | -1.221 |
Token | that |
Token ID | 326 |
Feature activation | -8.77e-01 |
Pos logit contributions | |
gy | +1.194 |
bank | +0.985 |
lves | +0.848 |
zag | +0.841 |
station | +0.806 |
Neg logit contributions | |
ey | -0.873 |
honey | -0.713 |
idges | -0.710 |
yogurt | -0.708 |
ad | -0.697 |
Token | was |
Token ID | 373 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
gy | +0.093 |
bank | +0.078 |
lves | +0.071 |
zag | +0.070 |
drown | +0.067 |
Neg logit contributions | |
ey | -0.050 |
yogurt | -0.042 |
honey | -0.041 |
ad | -0.040 |
idges | -0.039 |
Token | full |
Token ID | 1336 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
ey | +0.080 |
yogurt | +0.070 |
honey | +0.069 |
ad | +0.067 |
idges | +0.066 |
Neg logit contributions | |
gy | -0.100 |
bank | -0.081 |
lves | -0.072 |
zag | -0.071 |
drown | -0.067 |
Token | of |
Token ID | 286 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
ey | +0.122 |
idges | +0.095 |
ad | +0.090 |
Jamie | +0.088 |
yogurt | +0.086 |
Neg logit contributions | |
gy | -0.161 |
bank | -0.141 |
station | -0.120 |
zag | -0.111 |
lves | -0.110 |
Token | hurt |
Token ID | 5938 |
Feature activation | -4.94e+00 |
Pos logit contributions | |
gy | +0.036 |
bank | +0.028 |
zag | +0.027 |
station | +0.026 |
lves | +0.025 |
Neg logit contributions | |
honey | -0.031 |
sweet | -0.030 |
yogurt | -0.030 |
ey | -0.030 |
ad | -0.027 |
Token | her |
Token ID | 607 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
gy | +0.558 |
bank | +0.467 |
lves | +0.420 |
zag | +0.414 |
drown | +0.398 |
Neg logit contributions | |
ey | -0.268 |
yogurt | -0.208 |
idges | -0.204 |
honey | -0.203 |
ad | -0.198 |
Token | knee |
Token ID | 10329 |
Feature activation | -7.69e-01 |
Pos logit contributions | |
ey | +0.070 |
honey | +0.063 |
yogurt | +0.062 |
sweet | +0.061 |
ad | +0.058 |
Neg logit contributions | |
gy | -0.092 |
bank | -0.075 |
lves | -0.073 |
zag | -0.067 |
station | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | +8.08e-02 |
Pos logit contributions | |
gy | +0.081 |
bank | +0.067 |
lves | +0.066 |
station | +0.061 |
zag | +0.060 |
Neg logit contributions | |
ey | -0.035 |
sweet | -0.035 |
honey | -0.034 |
yogurt | -0.032 |
ad | -0.031 |
Token | Dog |
Token ID | 8532 |
Feature activation | -5.04e+00 |
Pos logit contributions | |
ey | +0.005 |
yogurt | +0.004 |
honey | +0.004 |
ad | +0.004 |
idges | +0.004 |
Neg logit contributions | |
gy | -0.008 |
bank | -0.007 |
lves | -0.006 |
zag | -0.006 |
drown | -0.006 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
gy | +0.012 |
bank | -0.009 |
drown | -0.057 |
lves | -0.061 |
station | -0.078 |
Neg logit contributions | |
ey | -0.696 |
ad | -0.687 |
honey | -0.677 |
sweet | -0.676 |
idges | -0.651 |
Token | helped |
Token ID | 4193 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
gy | +1.539 |
bank | +1.213 |
lves | +1.180 |
zag | +1.165 |
station | +1.093 |
Neg logit contributions | |
ey | -0.927 |
honey | -0.885 |
yogurt | -0.865 |
ad | -0.850 |
sweet | -0.847 |
Token | Kitty |
Token ID | 21168 |
Feature activation | -3.30e+00 |
Pos logit contributions | |
ey | +0.088 |
honey | +0.083 |
ad | +0.081 |
sweet | +0.075 |
idges | +0.071 |
Neg logit contributions | |
gy | -0.226 |
bank | -0.201 |
lves | -0.183 |
zag | -0.181 |
drown | -0.175 |
Token | stand |
Token ID | 1302 |
Feature activation | +3.06e+00 |
Pos logit contributions | |
gy | +0.268 |
bank | +0.212 |
lves | +0.199 |
zag | +0.183 |
station | +0.177 |
Neg logit contributions | |
ey | -0.250 |
honey | -0.231 |
yogurt | -0.226 |
ad | -0.222 |
sweet | -0.218 |
Token | up |
Token ID | 510 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
ey | +0.208 |
honey | +0.184 |
yogurt | +0.183 |
ad | +0.172 |
sweet | +0.172 |
Neg logit contributions | |
gy | -0.285 |
bank | -0.236 |
lves | -0.214 |
zag | -0.203 |
station | -0.200 |
Token | and |
Token ID | 290 |
Feature activation | -4.29e-01 |
Pos logit contributions | |
gy | +0.023 |
bank | +0.019 |
lves | +0.016 |
zag | +0.016 |
drown | +0.015 |
Neg logit contributions | |
ey | -0.017 |
yogurt | -0.014 |
honey | -0.014 |
ad | -0.014 |
idges | -0.014 |
Token | After |
Token ID | 3260 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
gy | +0.005 |
bank | +0.004 |
zag | +0.004 |
lves | +0.004 |
station | +0.004 |
Neg logit contributions | |
ey | -0.003 |
idges | -0.002 |
honey | -0.002 |
yogurt | -0.002 |
ad | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
ey | +0.117 |
yogurt | +0.096 |
honey | +0.095 |
idges | +0.094 |
ad | +0.094 |
Neg logit contributions | |
gy | -0.186 |
bank | -0.154 |
lves | -0.137 |
zag | -0.135 |
station | -0.129 |
Token | while |
Token ID | 981 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
ey | +0.012 |
honey | +0.012 |
sweet | +0.012 |
yogurt | +0.011 |
idges | +0.010 |
Neg logit contributions | |
gy | -0.014 |
bank | -0.011 |
lves | -0.009 |
zag | -0.009 |
drown | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
gy | +0.215 |
bank | +0.182 |
lves | +0.168 |
zag | +0.165 |
station | +0.156 |
Neg logit contributions | |
ey | -0.109 |
honey | -0.104 |
sweet | -0.096 |
ad | -0.095 |
yogurt | -0.094 |
Token | Bug |
Token ID | 15217 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
honey | +0.015 |
sweet | +0.014 |
ey | +0.014 |
yogurt | +0.013 |
goodness | +0.013 |
Neg logit contributions | |
gy | -0.022 |
bank | -0.018 |
zag | -0.018 |
station | -0.016 |
lves | -0.016 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
gy | +0.010 |
bank | -0.000 |
lves | -0.028 |
argue | -0.032 |
drown | -0.032 |
Neg logit contributions | |
ey | -0.447 |
ad | -0.441 |
honey | -0.435 |
sweet | -0.434 |
himself | -0.416 |
Token | was |
Token ID | 373 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
gy | +1.478 |
bank | +1.158 |
lves | +1.127 |
drown | +1.105 |
zag | +1.075 |
Neg logit contributions | |
ad | -0.929 |
ey | -0.919 |
honey | -0.854 |
yogurt | -0.844 |
sweet | -0.836 |
Token | tired |
Token ID | 10032 |
Feature activation | -3.78e+00 |
Pos logit contributions | |
ey | +0.083 |
yogurt | +0.072 |
honey | +0.072 |
ad | +0.072 |
idges | +0.070 |
Neg logit contributions | |
gy | -0.098 |
bank | -0.080 |
lves | -0.070 |
zag | -0.066 |
station | -0.064 |
Token | and |
Token ID | 290 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
gy | +0.350 |
bank | +0.283 |
lves | +0.248 |
zag | +0.243 |
drown | +0.231 |
Neg logit contributions | |
ey | -0.275 |
yogurt | -0.232 |
honey | -0.231 |
ad | -0.228 |
idges | -0.227 |
Token | wanted |
Token ID | 2227 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
gy | +0.036 |
bank | +0.028 |
lves | +0.025 |
zag | +0.025 |
station | +0.023 |
Neg logit contributions | |
ey | -0.033 |
yogurt | -0.029 |
honey | -0.029 |
ad | -0.028 |
idges | -0.028 |
Token | to |
Token ID | 284 |
Feature activation | +4.98e-01 |
Pos logit contributions | |
gy | +0.118 |
bank | +0.094 |
lves | +0.074 |
zag | +0.074 |
drown | +0.072 |
Neg logit contributions | |
ey | -0.134 |
idges | -0.111 |
ad | -0.109 |
yogurt | -0.107 |
honey | -0.106 |
Token | Rob |
Token ID | 3851 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
ey | +0.006 |
yogurt | +0.005 |
honey | +0.004 |
idges | +0.004 |
ad | +0.004 |
Neg logit contributions | |
gy | -0.009 |
bank | -0.008 |
lves | -0.007 |
zag | -0.007 |
station | -0.007 |
Token | pulled |
Token ID | 5954 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
gy | +0.013 |
bank | +0.010 |
drown | +0.010 |
lves | +0.010 |
zag | +0.009 |
Neg logit contributions | |
ey | -0.009 |
ad | -0.009 |
yogurt | -0.008 |
honey | -0.008 |
sweet | -0.008 |
Token | out |
Token ID | 503 |
Feature activation | -3.48e+00 |
Pos logit contributions | |
gy | +0.142 |
bank | +0.116 |
lves | +0.101 |
zag | +0.099 |
station | +0.095 |
Neg logit contributions | |
ey | -0.111 |
yogurt | -0.092 |
idges | -0.091 |
honey | -0.091 |
ad | -0.090 |
Token | his |
Token ID | 465 |
Feature activation | +9.02e-01 |
Pos logit contributions | |
gy | +0.333 |
bank | +0.276 |
lves | +0.232 |
zag | +0.230 |
station | +0.228 |
Neg logit contributions | |
ey | -0.243 |
idges | -0.197 |
yogurt | -0.194 |
honey | -0.192 |
ad | -0.188 |
Token | pig |
Token ID | 12967 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
ey | +0.069 |
yogurt | +0.060 |
honey | +0.059 |
ad | +0.057 |
sweet | +0.057 |
Neg logit contributions | |
gy | -0.079 |
bank | -0.061 |
lves | -0.056 |
zag | -0.055 |
drown | -0.052 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
gy | +0.272 |
lves | +0.206 |
bank | +0.192 |
drown | +0.191 |
zag | +0.169 |
Neg logit contributions | |
ey | -0.510 |
honey | -0.480 |
ad | -0.472 |
yogurt | -0.468 |
sweet | -0.465 |
Token | bank |
Token ID | 17796 |
Feature activation | -9.09e+00 |
Pos logit contributions | |
gy | +0.561 |
lves | +0.254 |
zag | +0.204 |
bank | +0.198 |
drown | +0.197 |
Neg logit contributions | |
ey | -1.933 |
yogurt | -1.831 |
honey | -1.817 |
sweet | -1.746 |
ad | -1.745 |
Token | and |
Token ID | 290 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
gy | +0.800 |
bank | +0.646 |
zag | +0.536 |
lves | +0.526 |
station | +0.513 |
Neg logit contributions | |
ey | -0.721 |
idges | -0.597 |
yogurt | -0.593 |
honey | -0.587 |
ad | -0.585 |
Token | emptied |
Token ID | 46705 |
Feature activation | +6.48e-01 |
Pos logit contributions | |
gy | +0.039 |
bank | +0.031 |
lves | +0.029 |
zag | +0.029 |
drown | +0.028 |
Neg logit contributions | |
ey | -0.027 |
yogurt | -0.025 |
honey | -0.025 |
ad | -0.024 |
sweet | -0.023 |
Token | it |
Token ID | 340 |
Feature activation | -8.80e-02 |
Pos logit contributions | |
ey | +0.041 |
yogurt | +0.034 |
honey | +0.034 |
ad | +0.033 |
idges | +0.033 |
Neg logit contributions | |
gy | -0.066 |
bank | -0.054 |
lves | -0.048 |
zag | -0.047 |
drown | -0.045 |
Token | on |
Token ID | 319 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
gy | +0.008 |
bank | +0.006 |
lves | +0.005 |
zag | +0.005 |
drown | +0.005 |
Neg logit contributions | |
ey | -0.007 |
yogurt | -0.006 |
honey | -0.006 |
ad | -0.006 |
idges | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -4.26e-01 |
Pos logit contributions | |
ey | +0.134 |
yogurt | +0.110 |
honey | +0.108 |
ad | +0.106 |
idges | +0.104 |
Neg logit contributions | |
gy | -0.214 |
bank | -0.173 |
lves | -0.159 |
zag | -0.154 |
drown | -0.150 |
Token | Ben |
Token ID | 3932 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
gy | +0.044 |
bank | +0.036 |
lves | +0.032 |
zag | +0.031 |
station | +0.031 |
Neg logit contributions | |
ey | -0.027 |
idges | -0.022 |
honey | -0.021 |
ad | -0.021 |
yogurt | -0.021 |
Token | get |
Token ID | 651 |
Feature activation | -3.51e+00 |
Pos logit contributions | |
ey | +0.149 |
yogurt | +0.128 |
honey | +0.126 |
ad | +0.126 |
idges | +0.123 |
Neg logit contributions | |
gy | -0.177 |
bank | -0.137 |
lves | -0.124 |
zag | -0.123 |
drown | -0.112 |
Token | their |
Token ID | 511 |
Feature activation | -5.52e+00 |
Pos logit contributions | |
gy | +0.352 |
zag | +0.280 |
bank | +0.267 |
lves | +0.259 |
drown | +0.232 |
Neg logit contributions | |
yogurt | -0.216 |
ey | -0.214 |
honey | -0.202 |
sweet | -0.192 |
pasta | -0.191 |
Token | pig |
Token ID | 12967 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
gy | +0.538 |
lves | +0.402 |
zag | +0.402 |
bank | +0.399 |
drown | +0.363 |
Neg logit contributions | |
ey | -0.368 |
yogurt | -0.342 |
honey | -0.315 |
sweet | -0.305 |
pasta | -0.297 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
lves | +0.313 |
gy | +0.291 |
drown | +0.283 |
nes | +0.269 |
zag | +0.266 |
Neg logit contributions | |
sweet | -0.376 |
honey | -0.370 |
himself | -0.369 |
yogurt | -0.365 |
ad | -0.355 |
Token | bank |
Token ID | 3331 |
Feature activation | -1.24e+01 |
Pos logit contributions | |
gy | +0.582 |
lves | +0.355 |
zag | +0.271 |
drown | +0.230 |
neither | +0.189 |
Neg logit contributions | |
ey | -1.848 |
yogurt | -1.827 |
honey | -1.782 |
cereal | -1.745 |
pasta | -1.712 |
Token | and |
Token ID | 290 |
Feature activation | -4.36e-01 |
Pos logit contributions | |
gy | +1.151 |
bank | +0.942 |
lves | +0.810 |
zag | +0.800 |
station | +0.766 |
Neg logit contributions | |
ey | -0.905 |
yogurt | -0.754 |
honey | -0.751 |
idges | -0.747 |
ad | -0.742 |
Token | put |
Token ID | 1234 |
Feature activation | +6.05e-01 |
Pos logit contributions | |
gy | +0.039 |
zag | +0.030 |
lves | +0.030 |
bank | +0.029 |
drown | +0.025 |
Neg logit contributions | |
ey | -0.031 |
yogurt | -0.029 |
honey | -0.029 |
ad | -0.026 |
sweet | -0.026 |
Token | it |
Token ID | 340 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
ey | +0.033 |
yogurt | +0.031 |
honey | +0.031 |
ad | +0.028 |
sweet | +0.028 |
Neg logit contributions | |
gy | -0.063 |
bank | -0.051 |
zag | -0.049 |
lves | -0.048 |
drown | -0.046 |
Token | in |
Token ID | 287 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
gy | +0.010 |
bank | +0.008 |
lves | +0.008 |
drown | +0.007 |
zag | +0.007 |
Neg logit contributions | |
ey | -0.006 |
yogurt | -0.006 |
honey | -0.006 |
sweet | -0.005 |
cereal | -0.005 |
Token | some |
Token ID | 617 |
Feature activation | -3.46e+00 |
Pos logit contributions | |
gy | +0.107 |
bank | +0.088 |
lves | +0.080 |
zag | +0.078 |
drown | +0.075 |
Neg logit contributions | |
ey | -0.064 |
yogurt | -0.054 |
honey | -0.053 |
ad | -0.052 |
idges | -0.051 |
Token | money |
Token ID | 1637 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
gy | +0.308 |
bank | +0.265 |
station | +0.223 |
zag | +0.218 |
house | +0.212 |
Neg logit contributions | |
ey | -0.241 |
idges | -0.211 |
ad | -0.207 |
mmmm | -0.202 |
ango | -0.200 |
Token | from |
Token ID | 422 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
gy | +0.280 |
bank | +0.228 |
lves | +0.210 |
zag | +0.198 |
drown | +0.196 |
Neg logit contributions | |
ey | -0.196 |
yogurt | -0.173 |
honey | -0.169 |
ad | -0.165 |
idges | -0.163 |
Token | their |
Token ID | 511 |
Feature activation | -5.52e+00 |
Pos logit contributions | |
gy | +0.143 |
bank | +0.117 |
lves | +0.105 |
zag | +0.104 |
drown | +0.099 |
Neg logit contributions | |
ey | -0.101 |
yogurt | -0.088 |
honey | -0.086 |
ad | -0.085 |
idges | -0.082 |
Token | pig |
Token ID | 12967 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
gy | +0.485 |
bank | +0.387 |
lves | +0.338 |
zag | +0.331 |
drown | +0.313 |
Neg logit contributions | |
ey | -0.427 |
yogurt | -0.365 |
honey | -0.362 |
ad | -0.355 |
idges | -0.354 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
gy | +0.126 |
drown | +0.114 |
lves | +0.112 |
bank | +0.085 |
neither | +0.070 |
Neg logit contributions | |
ey | -0.566 |
honey | -0.563 |
sweet | -0.552 |
yogurt | -0.548 |
ad | -0.544 |
Token | bank |
Token ID | 3331 |
Feature activation | -1.24e+01 |
Pos logit contributions | |
gy | +1.054 |
lves | +0.786 |
bank | +0.715 |
drown | +0.708 |
zag | +0.695 |
Neg logit contributions | |
ey | -1.400 |
yogurt | -1.318 |
honey | -1.304 |
ad | -1.234 |
cereal | -1.227 |
Token | . |
Token ID | 13 |
Feature activation | +7.82e-02 |
Pos logit contributions | |
gy | +1.124 |
bank | +0.922 |
zag | +0.776 |
lves | +0.769 |
station | +0.747 |
Neg logit contributions | |
ey | -0.938 |
yogurt | -0.777 |
idges | -0.776 |
honey | -0.775 |
ad | -0.768 |
Token | They |
Token ID | 1119 |
Feature activation | -3.91e+00 |
Pos logit contributions | |
ey | +0.005 |
yogurt | +0.004 |
honey | +0.004 |
idges | +0.004 |
ad | +0.004 |
Neg logit contributions | |
gy | -0.008 |
bank | -0.007 |
lves | -0.006 |
zag | -0.006 |
drown | -0.006 |
Token | asked |
Token ID | 1965 |
Feature activation | -9.14e-01 |
Pos logit contributions | |
gy | +0.355 |
zag | +0.283 |
bank | +0.277 |
lves | +0.271 |
drown | +0.267 |
Neg logit contributions | |
ey | -0.244 |
yogurt | -0.235 |
ad | -0.231 |
honey | -0.216 |
sweet | -0.212 |
Token | their |
Token ID | 511 |
Feature activation | -5.51e+00 |
Pos logit contributions | |
gy | +0.082 |
bank | +0.066 |
lves | +0.056 |
zag | +0.055 |
station | +0.053 |
Neg logit contributions | |
ey | -0.071 |
yogurt | -0.059 |
idges | -0.058 |
honey | -0.058 |
ad | -0.058 |
Token | One |
Token ID | 3198 |
Feature activation | -3.35e+00 |
Pos logit contributions | |
gy | +0.005 |
bank | +0.004 |
lves | +0.004 |
zag | +0.004 |
drown | +0.004 |
Neg logit contributions | |
ey | -0.004 |
yogurt | -0.004 |
ad | -0.003 |
idges | -0.003 |
honey | -0.003 |
Token | day |
Token ID | 1110 |
Feature activation | -2.91e+00 |
Pos logit contributions | |
gy | +0.288 |
bank | +0.247 |
lves | +0.212 |
drown | +0.197 |
zag | +0.192 |
Neg logit contributions | |
ey | -0.242 |
honey | -0.229 |
sweet | -0.226 |
yogurt | -0.223 |
ad | -0.214 |
Token | , |
Token ID | 11 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
gy | +0.267 |
bank | +0.217 |
lves | +0.189 |
zag | +0.185 |
station | +0.177 |
Neg logit contributions | |
ey | -0.214 |
yogurt | -0.183 |
honey | -0.182 |
idges | -0.178 |
ad | -0.178 |
Token | the |
Token ID | 262 |
Feature activation | -6.13e-02 |
Pos logit contributions | |
ey | +0.014 |
yogurt | +0.012 |
honey | +0.012 |
idges | +0.011 |
ad | +0.011 |
Neg logit contributions | |
gy | -0.026 |
bank | -0.022 |
zag | -0.019 |
lves | -0.019 |
drown | -0.019 |
Token | pig |
Token ID | 12967 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
gy | +0.005 |
bank | +0.004 |
zag | +0.004 |
lves | +0.003 |
drown | +0.003 |
Neg logit contributions | |
honey | -0.005 |
sweet | -0.005 |
ey | -0.004 |
yogurt | -0.004 |
ad | -0.004 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
gy | +0.412 |
lves | +0.380 |
drown | +0.380 |
bank | +0.377 |
argue | +0.364 |
Neg logit contributions | |
ad | -0.281 |
ey | -0.273 |
honey | -0.254 |
yogurt | -0.252 |
sweet | -0.251 |
Token | decided |
Token ID | 3066 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
gy | +1.489 |
drown | +1.272 |
zag | +1.255 |
argue | +1.237 |
lves | +1.230 |
Neg logit contributions | |
sweet | -0.684 |
honey | -0.677 |
ad | -0.672 |
yogurt | -0.617 |
me | -0.602 |
Token | to |
Token ID | 284 |
Feature activation | +4.91e-01 |
Pos logit contributions | |
gy | +0.112 |
bank | +0.090 |
lves | +0.078 |
zag | +0.076 |
station | +0.072 |
Neg logit contributions | |
ey | -0.100 |
yogurt | -0.085 |
honey | -0.084 |
idges | -0.084 |
ad | -0.084 |
Token | take |
Token ID | 1011 |
Feature activation | -4.96e-01 |
Pos logit contributions | |
ey | +0.035 |
ad | +0.033 |
honey | +0.031 |
yogurt | +0.031 |
sweet | +0.029 |
Neg logit contributions | |
gy | -0.043 |
lves | -0.034 |
bank | -0.033 |
zag | -0.030 |
neither | -0.028 |
Token | a |
Token ID | 257 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
gy | +0.054 |
bank | +0.045 |
lves | +0.041 |
zag | +0.040 |
station | +0.039 |
Neg logit contributions | |
ey | -0.028 |
honey | -0.024 |
yogurt | -0.023 |
ad | -0.023 |
idges | -0.022 |
Token | stroll |
Token ID | 41360 |
Feature activation | -9.39e-02 |
Pos logit contributions | |
ey | +0.012 |
honey | +0.011 |
yogurt | +0.011 |
sweet | +0.011 |
pasta | +0.010 |
Neg logit contributions | |
gy | -0.013 |
bank | -0.009 |
lves | -0.009 |
neither | -0.008 |
zag | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | -6.71e-02 |
Pos logit contributions | |
gy | +0.038 |
bank | +0.030 |
lves | +0.026 |
zag | +0.026 |
drown | +0.024 |
Neg logit contributions | |
ey | -0.033 |
yogurt | -0.030 |
honey | -0.029 |
ad | -0.028 |
idges | -0.028 |
Token | raced |
Token ID | 32897 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
gy | +0.005 |
bank | +0.004 |
lves | +0.003 |
zag | +0.003 |
station | +0.003 |
Neg logit contributions | |
ey | -0.006 |
idges | -0.005 |
ad | -0.005 |
yogurt | -0.005 |
honey | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | -6.70e-02 |
Pos logit contributions | |
gy | +0.200 |
bank | +0.158 |
lves | +0.137 |
zag | +0.134 |
drown | +0.127 |
Neg logit contributions | |
ey | -0.188 |
yogurt | -0.163 |
honey | -0.161 |
ad | -0.160 |
idges | -0.159 |
Token | s |
Token ID | 264 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
gy | +0.005 |
bank | +0.004 |
lves | +0.003 |
zag | +0.003 |
drown | +0.003 |
Neg logit contributions | |
ey | -0.006 |
yogurt | -0.005 |
honey | -0.005 |
ad | -0.005 |
idges | -0.005 |
Token | og |
Token ID | 519 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
gy | +0.096 |
bank | +0.079 |
zag | +0.071 |
station | +0.069 |
house | +0.065 |
Neg logit contributions | |
ey | -0.077 |
honey | -0.069 |
sweet | -0.065 |
yogurt | -0.063 |
idges | -0.061 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
gy | +0.181 |
bank | +0.138 |
lves | +0.134 |
drown | +0.121 |
neither | +0.099 |
Neg logit contributions | |
ey | -0.321 |
ad | -0.320 |
yogurt | -0.306 |
honey | -0.301 |
idges | -0.296 |
Token | race |
Token ID | 3234 |
Feature activation | -5.55e+00 |
Pos logit contributions | |
gy | +0.865 |
lves | +0.573 |
drown | +0.491 |
neither | +0.488 |
zag | +0.461 |
Neg logit contributions | |
ey | -1.550 |
yogurt | -1.486 |
honey | -1.472 |
sweet | -1.409 |
cereal | -1.400 |
Token | car |
Token ID | 1097 |
Feature activation | -6.99e+00 |
Pos logit contributions | |
gy | +0.544 |
bank | +0.439 |
lves | +0.419 |
drown | +0.387 |
zag | +0.385 |
Neg logit contributions | |
ey | -0.338 |
ad | -0.304 |
yogurt | -0.296 |
honey | -0.281 |
idges | -0.279 |
Token | around |
Token ID | 1088 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
gy | +0.708 |
bank | +0.584 |
lves | +0.583 |
drown | +0.548 |
drowned | +0.513 |
Neg logit contributions | |
ey | -0.338 |
ad | -0.332 |
yogurt | -0.308 |
honey | -0.292 |
himself | -0.292 |
Token | the |
Token ID | 262 |
Feature activation | -8.00e-02 |
Pos logit contributions | |
gy | +0.239 |
bank | +0.195 |
lves | +0.165 |
zag | +0.165 |
station | +0.161 |
Neg logit contributions | |
ey | -0.178 |
idges | -0.146 |
yogurt | -0.145 |
honey | -0.145 |
ad | -0.139 |
Token | bowl |
Token ID | 9396 |
Feature activation | +3.44e+00 |
Pos logit contributions | |
gy | +0.007 |
bank | +0.005 |
lves | +0.005 |
zag | +0.005 |
station | +0.004 |
Neg logit contributions | |
ey | -0.006 |
yogurt | -0.006 |
honey | -0.005 |
idges | -0.005 |
ad | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
ey | +0.033 |
honey | +0.031 |
yogurt | +0.030 |
sweet | +0.029 |
ad | +0.028 |
Neg logit contributions | |
gy | -0.055 |
bank | -0.045 |
lves | -0.044 |
drown | -0.040 |
zag | -0.039 |
Token | get |
Token ID | 651 |
Feature activation | -3.51e+00 |
Pos logit contributions | |
gy | +0.037 |
bank | +0.029 |
lves | +0.025 |
zag | +0.024 |
drown | +0.023 |
Neg logit contributions | |
ey | -0.035 |
yogurt | -0.030 |
idges | -0.030 |
ad | -0.030 |
honey | -0.030 |
Token | warm |
Token ID | 5814 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
gy | +0.308 |
bank | +0.248 |
lves | +0.210 |
drown | +0.208 |
neither | +0.198 |
Neg logit contributions | |
ey | -0.278 |
idges | -0.228 |
ad | -0.227 |
yogurt | -0.218 |
ango | -0.215 |
Token | ." |
Token ID | 526 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
ey | +0.257 |
idges | +0.216 |
ad | +0.215 |
yogurt | +0.207 |
subtract | +0.200 |
Neg logit contributions | |
gy | -0.225 |
bank | -0.173 |
station | -0.136 |
zag | -0.135 |
lves | -0.133 |
Token | Dog |
Token ID | 8532 |
Feature activation | -5.04e+00 |
Pos logit contributions | |
gy | +0.169 |
bank | +0.144 |
lves | +0.133 |
zag | +0.127 |
drown | +0.125 |
Neg logit contributions | |
ey | -0.126 |
honey | -0.115 |
ad | -0.112 |
sweet | -0.109 |
yogurt | -0.105 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
gy | +0.032 |
bank | -0.015 |
lves | -0.063 |
drown | -0.066 |
station | -0.085 |
Neg logit contributions | |
ey | -0.731 |
ad | -0.698 |
honey | -0.691 |
sweet | -0.679 |
idges | -0.677 |
Token | agreed |
Token ID | 4987 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
gy | +1.505 |
bank | +1.215 |
lves | +1.127 |
zag | +1.086 |
drown | +1.058 |
Neg logit contributions | |
ey | -1.042 |
honey | -0.936 |
yogurt | -0.933 |
ad | -0.915 |
sweet | -0.877 |
Token | , |
Token ID | 11 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
gy | +0.002 |
bank | +0.001 |
zag | +0.001 |
lves | +0.001 |
station | +0.001 |
Neg logit contributions | |
ey | -0.001 |
idges | -0.001 |
yogurt | -0.001 |
honey | -0.001 |
ad | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | -4.36e-01 |
Pos logit contributions | |
ey | +0.017 |
yogurt | +0.014 |
honey | +0.014 |
ad | +0.014 |
idges | +0.014 |
Neg logit contributions | |
gy | -0.023 |
bank | -0.019 |
lves | -0.017 |
zag | -0.016 |
drown | -0.016 |
Token | they |
Token ID | 484 |
Feature activation | -7.30e+00 |
Pos logit contributions | |
gy | +0.034 |
bank | +0.028 |
zag | +0.026 |
argue | +0.025 |
lves | +0.024 |
Neg logit contributions | |
honey | -0.033 |
sweet | -0.031 |
yogurt | -0.028 |
ey | -0.028 |
ad | -0.028 |
Token | both |
Token ID | 1111 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
gy | +0.643 |
bank | +0.516 |
lves | +0.466 |
zag | +0.462 |
drown | +0.431 |
Neg logit contributions | |
ey | -0.504 |
honey | -0.478 |
yogurt | -0.469 |
ad | -0.461 |
sweet | -0.449 |
Token | . |
Token ID | 13 |
Feature activation | +8.37e-02 |
Pos logit contributions | |
gy | +0.595 |
lves | +0.536 |
bank | +0.525 |
zag | +0.496 |
harbor | +0.492 |
Neg logit contributions | |
sweet | -0.142 |
honey | -0.134 |
yogurt | -0.112 |
himself | -0.103 |
him | -0.090 |
Token |
|
Token ID | 198 |
Feature activation | -4.71e-02 |
Pos logit contributions | |
ey | +0.005 |
idges | +0.005 |
ad | +0.005 |
yogurt | +0.004 |
honey | +0.004 |
Neg logit contributions | |
gy | -0.008 |
bank | -0.007 |
lves | -0.006 |
neither | -0.006 |
zag | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -5.55e-02 |
Pos logit contributions | |
gy | +0.004 |
bank | +0.003 |
zag | +0.003 |
lves | +0.003 |
drown | +0.003 |
Neg logit contributions | |
ey | -0.003 |
yogurt | -0.003 |
honey | -0.003 |
idges | -0.003 |
ad | -0.003 |
Token | " |
Token ID | 1 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
gy | +0.006 |
bank | +0.005 |
lves | +0.004 |
zag | +0.004 |
drown | +0.004 |
Neg logit contributions | |
ey | -0.004 |
yogurt | -0.003 |
honey | -0.003 |
idges | -0.003 |
ad | -0.003 |
Token | Dog |
Token ID | 32942 |
Feature activation | -8.17e+00 |
Pos logit contributions | |
gy | +0.150 |
bank | +0.125 |
lves | +0.112 |
zag | +0.112 |
station | +0.107 |
Neg logit contributions | |
ey | -0.076 |
yogurt | -0.061 |
honey | -0.060 |
ad | -0.059 |
idges | -0.059 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
gy | +0.667 |
bank | +0.567 |
lves | +0.502 |
zag | +0.470 |
drown | +0.460 |
Neg logit contributions | |
ey | -0.607 |
ad | -0.550 |
honey | -0.540 |
yogurt | -0.526 |
sweet | -0.515 |
Token | , |
Token ID | 11 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
gy | +1.527 |
bank | +1.243 |
lves | +1.139 |
zag | +1.109 |
drown | +1.057 |
Neg logit contributions | |
ey | -1.041 |
honey | -0.905 |
ad | -0.894 |
yogurt | -0.893 |
idges | -0.853 |
Token | you |
Token ID | 345 |
Feature activation | +2.54e+00 |
Pos logit contributions | |
ey | +0.016 |
honey | +0.014 |
yogurt | +0.013 |
ad | +0.013 |
idges | +0.013 |
Neg logit contributions | |
gy | -0.024 |
bank | -0.020 |
lves | -0.017 |
zag | -0.017 |
station | -0.016 |
Token | are |
Token ID | 389 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
ey | +0.153 |
honey | +0.135 |
yogurt | +0.132 |
sweet | +0.131 |
ad | +0.129 |
Neg logit contributions | |
gy | -0.258 |
bank | -0.214 |
zag | -0.193 |
lves | -0.187 |
station | -0.183 |
Token | a |
Token ID | 257 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
ey | +0.014 |
honey | +0.012 |
yogurt | +0.012 |
ad | +0.012 |
idges | +0.012 |
Neg logit contributions | |
gy | -0.019 |
bank | -0.016 |
zag | -0.013 |
lves | -0.013 |
station | -0.013 |
Token | bad |
Token ID | 2089 |
Feature activation | -3.80e+00 |
Pos logit contributions | |
sweet | +0.010 |
honey | +0.009 |
ey | +0.008 |
yogurt | +0.008 |
mirror | +0.007 |
Neg logit contributions | |
gy | -0.019 |
bank | -0.015 |
zag | -0.014 |
banks | -0.014 |
lves | -0.013 |
Token | her |
Token ID | 607 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
ey | +0.152 |
yogurt | +0.125 |
honey | +0.124 |
idges | +0.123 |
ad | +0.122 |
Neg logit contributions | |
gy | -0.225 |
bank | -0.186 |
lves | -0.164 |
zag | -0.159 |
station | -0.153 |
Token | penny |
Token ID | 26369 |
Feature activation | -2.84e+00 |
Pos logit contributions | |
ey | +0.084 |
yogurt | +0.072 |
honey | +0.072 |
idges | +0.071 |
ad | +0.071 |
Neg logit contributions | |
gy | -0.084 |
bank | -0.066 |
lves | -0.057 |
zag | -0.055 |
station | -0.052 |
Token | in |
Token ID | 287 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
gy | +0.241 |
bank | +0.191 |
lves | +0.170 |
zag | +0.163 |
drown | +0.162 |
Neg logit contributions | |
ey | -0.213 |
yogurt | -0.194 |
honey | -0.189 |
ad | -0.186 |
idges | -0.182 |
Token | the |
Token ID | 262 |
Feature activation | -7.06e-02 |
Pos logit contributions | |
ey | +0.079 |
idges | +0.066 |
honey | +0.061 |
yogurt | +0.061 |
ad | +0.060 |
Neg logit contributions | |
gy | -0.119 |
bank | -0.099 |
lves | -0.084 |
zag | -0.081 |
station | -0.080 |
Token | pig |
Token ID | 12967 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
gy | +0.006 |
bank | +0.005 |
lves | +0.004 |
zag | +0.004 |
station | +0.004 |
Neg logit contributions | |
ey | -0.005 |
idges | -0.005 |
yogurt | -0.005 |
honey | -0.004 |
ad | -0.004 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
gy | +0.092 |
bank | +0.032 |
lves | +0.029 |
drown | +0.027 |
neither | +0.002 |
Neg logit contributions | |
ey | -0.666 |
ad | -0.653 |
honey | -0.634 |
yogurt | -0.633 |
sweet | -0.620 |
Token | bank |
Token ID | 3331 |
Feature activation | -1.24e+01 |
Pos logit contributions | |
gy | +0.448 |
lves | +0.092 |
zag | +0.066 |
bank | +0.055 |
drown | +0.037 |
Neg logit contributions | |
ey | -2.067 |
yogurt | -1.997 |
honey | -1.964 |
ad | -1.896 |
sweet | -1.880 |
Token | ever |
Token ID | 1683 |
Feature activation | -7.20e-01 |
Pos logit contributions | |
gy | +1.098 |
bank | +0.914 |
lves | +0.760 |
zag | +0.757 |
station | +0.728 |
Neg logit contributions | |
ey | -0.955 |
idges | -0.791 |
honey | -0.785 |
yogurt | -0.780 |
ad | -0.774 |
Token | again |
Token ID | 757 |
Feature activation | +5.27e-01 |
Pos logit contributions | |
gy | +0.058 |
bank | +0.048 |
lves | +0.038 |
zag | +0.036 |
argue | +0.036 |
Neg logit contributions | |
ey | -0.062 |
idges | -0.053 |
ad | -0.051 |
honey | -0.050 |
yogurt | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | +9.44e-02 |
Pos logit contributions | |
ey | +0.034 |
yogurt | +0.029 |
honey | +0.029 |
ad | +0.028 |
idges | +0.028 |
Neg logit contributions | |
gy | -0.052 |
bank | -0.042 |
lves | -0.038 |
zag | -0.037 |
drown | -0.036 |
Token | She |
Token ID | 1375 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
ey | +0.006 |
honey | +0.005 |
idges | +0.005 |
yogurt | +0.005 |
ad | +0.005 |
Neg logit contributions | |
gy | -0.009 |
bank | -0.008 |
lves | -0.007 |
zag | -0.007 |
station | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
gy | +0.329 |
bank | +0.270 |
lves | +0.236 |
zag | +0.230 |
drown | +0.220 |
Neg logit contributions | |
ey | -0.262 |
yogurt | -0.224 |
honey | -0.224 |
ad | -0.221 |
idges | -0.219 |
Token | a |
Token ID | 257 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
ey | +0.014 |
yogurt | +0.012 |
honey | +0.012 |
ad | +0.012 |
idges | +0.011 |
Neg logit contributions | |
gy | -0.025 |
bank | -0.021 |
lves | -0.019 |
zag | -0.019 |
drown | -0.018 |
Token | dog |
Token ID | 3290 |
Feature activation | -5.02e+00 |
Pos logit contributions | |
ey | +0.010 |
honey | +0.009 |
yogurt | +0.009 |
sweet | +0.009 |
ad | +0.008 |
Neg logit contributions | |
gy | -0.017 |
bank | -0.014 |
lves | -0.012 |
zag | -0.012 |
drown | -0.012 |
Token | named |
Token ID | 3706 |
Feature activation | -6.49e+00 |
Pos logit contributions | |
gy | +0.332 |
bank | +0.270 |
lves | +0.223 |
zag | +0.208 |
drown | +0.197 |
Neg logit contributions | |
ey | -0.454 |
ad | -0.424 |
honey | -0.420 |
yogurt | -0.416 |
idges | -0.406 |
Token | Dog |
Token ID | 8532 |
Feature activation | -5.05e+00 |
Pos logit contributions | |
gy | +0.598 |
bank | +0.487 |
lves | +0.424 |
zag | +0.417 |
drown | +0.397 |
Neg logit contributions | |
ey | -0.470 |
yogurt | -0.396 |
honey | -0.394 |
ad | -0.389 |
idges | -0.389 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
gy | +0.205 |
bank | +0.189 |
lves | +0.147 |
drown | +0.134 |
zag | +0.119 |
Neg logit contributions | |
ad | -0.502 |
ey | -0.499 |
honey | -0.485 |
sweet | -0.472 |
idges | -0.458 |
Token | , |
Token ID | 11 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
gy | +1.544 |
lves | +1.273 |
bank | +1.265 |
zag | +1.205 |
drown | +1.171 |
Neg logit contributions | |
ey | -0.895 |
ad | -0.818 |
yogurt | -0.779 |
honey | -0.778 |
sweet | -0.731 |
Token | came |
Token ID | 1625 |
Feature activation | +5.04e-02 |
Pos logit contributions | |
ey | +0.022 |
yogurt | +0.019 |
idges | +0.019 |
ad | +0.019 |
honey | +0.019 |
Neg logit contributions | |
gy | -0.019 |
bank | -0.014 |
lves | -0.012 |
zag | -0.011 |
station | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | +4.87e-01 |
Pos logit contributions | |
ey | +0.003 |
honey | +0.003 |
ad | +0.003 |
sweet | +0.002 |
idges | +0.002 |
Neg logit contributions | |
gy | -0.005 |
bank | -0.004 |
lves | -0.004 |
drown | -0.004 |
sank | -0.004 |
Token | play |
Token ID | 711 |
Feature activation | +9.92e-01 |
Pos logit contributions | |
ey | +0.036 |
yogurt | +0.030 |
honey | +0.030 |
ad | +0.029 |
idges | +0.029 |
Neg logit contributions | |
gy | -0.045 |
bank | -0.036 |
lves | -0.032 |
zag | -0.031 |
station | -0.030 |
Token | with |
Token ID | 351 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
ey | +0.054 |
yogurt | +0.048 |
honey | +0.047 |
ad | +0.047 |
sweet | +0.046 |
Neg logit contributions | |
gy | -0.105 |
bank | -0.086 |
lves | -0.080 |
zag | -0.079 |
drown | -0.075 |
Token | met |
Token ID | 1138 |
Feature activation | -4.95e+00 |
Pos logit contributions | |
gy | +0.265 |
bank | +0.207 |
lves | +0.180 |
zag | +0.179 |
drown | +0.170 |
Neg logit contributions | |
ey | -0.269 |
yogurt | -0.236 |
honey | -0.235 |
ad | -0.232 |
idges | -0.227 |
Token | a |
Token ID | 257 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
gy | +0.434 |
bank | +0.326 |
lves | +0.279 |
zag | +0.277 |
house | +0.272 |
Neg logit contributions | |
ey | -0.424 |
idges | -0.334 |
yogurt | -0.324 |
layers | -0.314 |
ad | -0.314 |
Token | dog |
Token ID | 3290 |
Feature activation | -5.02e+00 |
Pos logit contributions | |
ey | +0.010 |
yogurt | +0.008 |
ad | +0.008 |
idges | +0.008 |
honey | +0.008 |
Neg logit contributions | |
gy | -0.017 |
bank | -0.014 |
lves | -0.012 |
zag | -0.012 |
station | -0.012 |
Token | named |
Token ID | 3706 |
Feature activation | -6.49e+00 |
Pos logit contributions | |
gy | +0.399 |
bank | +0.321 |
lves | +0.283 |
zag | +0.272 |
drown | +0.266 |
Neg logit contributions | |
ey | -0.379 |
honey | -0.356 |
ad | -0.355 |
yogurt | -0.353 |
idges | -0.334 |
Token | Dog |
Token ID | 8532 |
Feature activation | -5.04e+00 |
Pos logit contributions | |
gy | +0.581 |
bank | +0.457 |
lves | +0.401 |
zag | +0.387 |
station | +0.371 |
Neg logit contributions | |
ey | -0.505 |
yogurt | -0.420 |
idges | -0.418 |
ad | -0.412 |
honey | -0.411 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
gy | +0.338 |
bank | +0.304 |
drown | +0.264 |
lves | +0.261 |
zag | +0.248 |
Neg logit contributions | |
ad | -0.368 |
ey | -0.367 |
honey | -0.358 |
sweet | -0.348 |
himself | -0.333 |
Token | . |
Token ID | 13 |
Feature activation | +8.26e-02 |
Pos logit contributions | |
gy | +1.490 |
bank | +1.190 |
lves | +1.122 |
zag | +1.089 |
drown | +1.062 |
Neg logit contributions | |
ey | -1.053 |
honey | -0.921 |
yogurt | -0.920 |
ad | -0.904 |
idges | -0.856 |
Token | Dog |
Token ID | 8532 |
Feature activation | -5.05e+00 |
Pos logit contributions | |
ey | +0.007 |
honey | +0.006 |
ad | +0.006 |
yogurt | +0.006 |
idges | +0.006 |
Neg logit contributions | |
gy | -0.006 |
bank | -0.005 |
lves | -0.004 |
zag | -0.004 |
drown | -0.004 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
gy | +0.025 |
bank | -0.024 |
drown | -0.075 |
lves | -0.077 |
zag | -0.089 |
Neg logit contributions | |
ey | -0.738 |
ad | -0.705 |
honey | -0.697 |
sweet | -0.689 |
idges | -0.688 |
Token | was |
Token ID | 373 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
gy | +1.528 |
bank | +1.241 |
lves | +1.147 |
zag | +1.117 |
drown | +1.111 |
Neg logit contributions | |
ey | -0.999 |
ad | -0.879 |
honey | -0.870 |
yogurt | -0.868 |
idges | -0.837 |
Token | very |
Token ID | 845 |
Feature activation | -8.44e-01 |
Pos logit contributions | |
ey | +0.068 |
honey | +0.057 |
yogurt | +0.057 |
ad | +0.056 |
idges | +0.055 |
Neg logit contributions | |
gy | -0.110 |
bank | -0.093 |
lves | -0.082 |
zag | -0.080 |
drown | -0.077 |
Token | She |
Token ID | 1375 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
ey | +0.013 |
idges | +0.012 |
ad | +0.012 |
honey | +0.011 |
ement | +0.011 |
Neg logit contributions | |
gy | -0.012 |
bank | -0.009 |
neither | -0.008 |
zag | -0.008 |
lves | -0.007 |
Token | says |
Token ID | 1139 |
Feature activation | +8.19e-01 |
Pos logit contributions | |
ey | +0.126 |
idges | +0.107 |
yogurt | +0.106 |
ad | +0.106 |
honey | +0.105 |
Neg logit contributions | |
gy | -0.155 |
bank | -0.129 |
lves | -0.109 |
zag | -0.107 |
station | -0.105 |
Token | to |
Token ID | 284 |
Feature activation | +5.78e-01 |
Pos logit contributions | |
ey | +0.076 |
idges | +0.064 |
ad | +0.063 |
layers | +0.063 |
ango | +0.061 |
Neg logit contributions | |
gy | -0.052 |
bank | -0.042 |
neither | -0.038 |
zag | -0.034 |
argue | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
ey | +0.046 |
importantly | +0.041 |
idges | +0.040 |
layers | +0.040 |
ement | +0.039 |
Neg logit contributions | |
gy | -0.045 |
bank | -0.035 |
neither | -0.030 |
fight | -0.029 |
drown | -0.029 |
Token | ice |
Token ID | 4771 |
Feature activation | +4.70e+00 |
Pos logit contributions | |
ey | +0.002 |
idges | +0.002 |
ad | +0.001 |
honey | +0.001 |
ango | +0.001 |
Neg logit contributions | |
gy | -0.002 |
bank | -0.001 |
station | -0.001 |
zag | -0.001 |
drown | -0.001 |
Token | cream |
Token ID | 8566 |
Feature activation | +7.49e+00 |
Pos logit contributions | |
ey | +0.195 |
idges | +0.137 |
honey | +0.135 |
ad | +0.133 |
yogurt | +0.133 |
Neg logit contributions | |
gy | -0.583 |
bank | -0.504 |
zag | -0.453 |
lves | -0.451 |
station | -0.438 |
Token | man |
Token ID | 582 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
ey | +0.605 |
idges | +0.518 |
honey | +0.497 |
ad | +0.492 |
yogurt | +0.488 |
Neg logit contributions | |
gy | -0.630 |
bank | -0.527 |
zag | -0.429 |
station | -0.427 |
lves | -0.416 |
Token | , |
Token ID | 11 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
gy | +0.142 |
bank | +0.113 |
zag | +0.097 |
lves | +0.097 |
station | +0.093 |
Neg logit contributions | |
ey | -0.120 |
idges | -0.099 |
yogurt | -0.099 |
honey | -0.098 |
ad | -0.098 |
Token | " |
Token ID | 366 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
ey | +0.024 |
idges | +0.020 |
ad | +0.020 |
yogurt | +0.019 |
honey | +0.018 |
Neg logit contributions | |
gy | -0.033 |
bank | -0.027 |
lves | -0.023 |
station | -0.023 |
zag | -0.023 |
Token | Can |
Token ID | 6090 |
Feature activation | +5.87e+00 |
Pos logit contributions | |
ey | +0.134 |
honey | +0.110 |
idges | +0.108 |
yogurt | +0.104 |
ad | +0.100 |
Neg logit contributions | |
gy | -0.204 |
bank | -0.176 |
lves | -0.155 |
zag | -0.147 |
station | -0.145 |
Token | I |
Token ID | 314 |
Feature activation | +3.51e+00 |
Pos logit contributions | |
ey | +0.338 |
yogurt | +0.263 |
honey | +0.262 |
idges | +0.261 |
ad | +0.260 |
Neg logit contributions | |
gy | -0.633 |
bank | -0.531 |
lves | -0.471 |
zag | -0.468 |
drown | -0.450 |
Token | little |
Token ID | 1310 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
gy | +0.243 |
bank | +0.203 |
lves | +0.170 |
station | +0.169 |
zag | +0.165 |
Neg logit contributions | |
ey | -0.199 |
idges | -0.168 |
ad | -0.165 |
yogurt | -0.158 |
honey | -0.155 |
Token | girl |
Token ID | 2576 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
ey | +0.081 |
honey | +0.071 |
yogurt | +0.071 |
ad | +0.068 |
sweet | +0.066 |
Neg logit contributions | |
gy | -0.104 |
bank | -0.085 |
lves | -0.076 |
zag | -0.075 |
drown | -0.072 |
Token | said |
Token ID | 531 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
gy | +0.104 |
bank | +0.086 |
lves | +0.075 |
zag | +0.074 |
station | +0.071 |
Neg logit contributions | |
ey | -0.071 |
yogurt | -0.058 |
honey | -0.058 |
idges | -0.058 |
ad | -0.057 |
Token | , |
Token ID | 11 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
gy | +0.025 |
bank | +0.020 |
zag | +0.016 |
neither | +0.016 |
lves | +0.016 |
Neg logit contributions | |
ey | -0.027 |
ad | -0.022 |
idges | -0.022 |
yogurt | -0.022 |
honey | -0.021 |
Token | " |
Token ID | 366 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
ey | +0.020 |
yogurt | +0.016 |
idges | +0.016 |
ad | +0.016 |
honey | +0.015 |
Neg logit contributions | |
gy | -0.033 |
bank | -0.027 |
zag | -0.024 |
lves | -0.024 |
station | -0.023 |
Token | Ok |
Token ID | 18690 |
Feature activation | +5.26e+00 |
Pos logit contributions | |
ey | +0.143 |
honey | +0.118 |
yogurt | +0.114 |
idges | +0.113 |
ad | +0.108 |
Neg logit contributions | |
gy | -0.191 |
bank | -0.158 |
lves | -0.143 |
zag | -0.138 |
station | -0.133 |
Token | ! |
Token ID | 0 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
ey | +0.430 |
yogurt | +0.352 |
idges | +0.350 |
ad | +0.348 |
honey | +0.340 |
Neg logit contributions | |
gy | -0.461 |
bank | -0.358 |
zag | -0.306 |
lves | -0.301 |
station | -0.295 |
Token | I |
Token ID | 314 |
Feature activation | +3.45e+00 |
Pos logit contributions | |
ey | +0.150 |
yogurt | +0.124 |
honey | +0.122 |
idges | +0.121 |
ad | +0.121 |
Neg logit contributions | |
gy | -0.221 |
bank | -0.179 |
lves | -0.158 |
zag | -0.157 |
station | -0.152 |
Token | 'll |
Token ID | 1183 |
Feature activation | -8.53e-01 |
Pos logit contributions | |
ey | +0.238 |
yogurt | +0.216 |
honey | +0.216 |
sweet | +0.215 |
ad | +0.215 |
Neg logit contributions | |
gy | -0.299 |
bank | -0.236 |
lves | -0.222 |
zag | -0.204 |
jet | -0.193 |
Token | put |
Token ID | 1234 |
Feature activation | +6.56e-01 |
Pos logit contributions | |
gy | +0.073 |
bank | +0.059 |
lves | +0.050 |
zag | +0.049 |
drown | +0.047 |
Neg logit contributions | |
ey | -0.068 |
yogurt | -0.058 |
honey | -0.057 |
idges | -0.057 |
ad | -0.056 |
Token | the |
Token ID | 262 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
ey | +0.048 |
ad | +0.039 |
idges | +0.039 |
yogurt | +0.038 |
honey | +0.038 |
Neg logit contributions | |
gy | -0.061 |
bank | -0.050 |
lves | -0.043 |
zag | -0.041 |
station | -0.040 |
Token | smiled |
Token ID | 13541 |
Feature activation | +6.06e+00 |
Pos logit contributions | |
gy | +0.373 |
bank | +0.308 |
zag | +0.264 |
lves | +0.262 |
station | +0.252 |
Neg logit contributions | |
ey | -0.255 |
idges | -0.202 |
yogurt | -0.194 |
ad | -0.193 |
honey | -0.191 |
Token | and |
Token ID | 290 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
ey | +0.555 |
idges | +0.498 |
yogurt | +0.468 |
ad | +0.466 |
mmmm | +0.452 |
Neg logit contributions | |
gy | -0.467 |
bank | -0.330 |
neither | -0.271 |
zag | -0.262 |
lves | -0.261 |
Token | said |
Token ID | 531 |
Feature activation | -3.11e-01 |
Pos logit contributions | |
gy | +0.032 |
bank | +0.026 |
lves | +0.023 |
zag | +0.022 |
station | +0.021 |
Neg logit contributions | |
ey | -0.027 |
idges | -0.023 |
ad | -0.022 |
yogurt | -0.022 |
honey | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
gy | +0.021 |
bank | +0.015 |
neither | +0.013 |
zag | +0.012 |
argue | +0.011 |
Neg logit contributions | |
ey | -0.032 |
idges | -0.026 |
ad | -0.026 |
ango | -0.026 |
layers | -0.025 |
Token | " |
Token ID | 366 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
ey | +0.021 |
ad | +0.017 |
idges | +0.017 |
yogurt | +0.017 |
honey | +0.016 |
Neg logit contributions | |
gy | -0.030 |
bank | -0.024 |
zag | -0.021 |
lves | -0.021 |
station | -0.020 |
Token | Would |
Token ID | 17353 |
Feature activation | +5.35e+00 |
Pos logit contributions | |
ey | +0.153 |
honey | +0.127 |
idges | +0.124 |
yogurt | +0.118 |
ad | +0.115 |
Neg logit contributions | |
gy | -0.180 |
bank | -0.146 |
lves | -0.134 |
zag | -0.125 |
neither | -0.121 |
Token | you |
Token ID | 345 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
ey | +0.422 |
layers | +0.303 |
idges | +0.289 |
amaz | +0.287 |
honey | +0.286 |
Neg logit contributions | |
gy | -0.452 |
bank | -0.379 |
lves | -0.344 |
neither | -0.333 |
zag | -0.315 |
Token | like |
Token ID | 588 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
ey | +0.249 |
idges | +0.209 |
ement | +0.205 |
ango | +0.196 |
ad | +0.195 |
Neg logit contributions | |
gy | -0.179 |
bank | -0.141 |
argue | -0.112 |
neither | -0.106 |
fight | -0.103 |
Token | to |
Token ID | 284 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
ey | +0.162 |
ement | +0.145 |
ango | +0.133 |
idges | +0.132 |
Jamie | +0.131 |
Neg logit contributions | |
gy | -0.067 |
bank | -0.056 |
station | -0.038 |
neither | -0.037 |
fight | -0.035 |
Token | share |
Token ID | 2648 |
Feature activation | +3.56e+00 |
Pos logit contributions | |
ey | +0.053 |
ement | +0.046 |
idges | +0.045 |
importantly | +0.044 |
ango | +0.043 |
Neg logit contributions | |
gy | -0.041 |
bank | -0.033 |
argue | -0.025 |
fight | -0.025 |
drown | -0.024 |
Token | some |
Token ID | 617 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
ey | +0.281 |
idges | +0.220 |
ad | +0.219 |
importantly | +0.218 |
ement | +0.216 |
Neg logit contributions | |
gy | -0.302 |
bank | -0.251 |
lves | -0.202 |
station | -0.200 |
house | -0.196 |
Token | that |
Token ID | 326 |
Feature activation | -7.64e-01 |
Pos logit contributions | |
gy | +0.023 |
bank | +0.019 |
lves | +0.016 |
zag | +0.016 |
neither | +0.015 |
Neg logit contributions | |
ey | -0.018 |
yogurt | -0.014 |
ad | -0.014 |
idges | -0.014 |
honey | -0.014 |
Token | her |
Token ID | 607 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
gy | +0.075 |
bank | +0.062 |
lves | +0.054 |
zag | +0.054 |
drown | +0.051 |
Neg logit contributions | |
ey | -0.051 |
yogurt | -0.043 |
honey | -0.042 |
idges | -0.042 |
ad | -0.042 |
Token | mom |
Token ID | 1995 |
Feature activation | -5.94e-01 |
Pos logit contributions | |
ey | +0.072 |
yogurt | +0.063 |
sweet | +0.062 |
honey | +0.060 |
idges | +0.055 |
Neg logit contributions | |
gy | -0.109 |
bank | -0.089 |
lves | -0.084 |
zag | -0.077 |
drown | -0.074 |
Token | always |
Token ID | 1464 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
gy | +0.055 |
bank | +0.046 |
lves | +0.040 |
zag | +0.039 |
drown | +0.039 |
Neg logit contributions | |
ey | -0.041 |
yogurt | -0.034 |
ad | -0.034 |
honey | -0.034 |
idges | -0.034 |
Token | knew |
Token ID | 2993 |
Feature activation | -8.67e-01 |
Pos logit contributions | |
gy | +0.028 |
bank | +0.022 |
lves | +0.019 |
zag | +0.019 |
station | +0.018 |
Neg logit contributions | |
ey | -0.027 |
yogurt | -0.023 |
idges | -0.023 |
honey | -0.023 |
ad | -0.022 |
Token | what |
Token ID | 644 |
Feature activation | +5.25e+00 |
Pos logit contributions | |
gy | +0.082 |
bank | +0.066 |
lves | +0.059 |
zag | +0.059 |
neither | +0.059 |
Neg logit contributions | |
ey | -0.060 |
ad | -0.047 |
yogurt | -0.045 |
idges | -0.044 |
layers | -0.044 |
Token | was |
Token ID | 373 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
ey | +0.416 |
ad | +0.343 |
idges | +0.341 |
yogurt | +0.339 |
honey | +0.337 |
Neg logit contributions | |
gy | -0.455 |
bank | -0.363 |
zag | -0.309 |
lves | -0.309 |
station | -0.297 |
Token | best |
Token ID | 1266 |
Feature activation | -9.22e-01 |
Pos logit contributions | |
ey | +0.084 |
honey | +0.073 |
yogurt | +0.072 |
idges | +0.070 |
ad | +0.070 |
Neg logit contributions | |
gy | -0.116 |
bank | -0.095 |
lves | -0.083 |
zag | -0.082 |
drown | -0.078 |
Token | for |
Token ID | 329 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
gy | +0.078 |
bank | +0.060 |
zag | +0.053 |
lves | +0.050 |
station | +0.049 |
Neg logit contributions | |
ey | -0.076 |
yogurt | -0.064 |
ad | -0.063 |
idges | -0.063 |
honey | -0.061 |
Token | her |
Token ID | 607 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
gy | +0.179 |
bank | +0.145 |
lves | +0.130 |
zag | +0.128 |
station | +0.126 |
Neg logit contributions | |
ey | -0.107 |
ad | -0.081 |
yogurt | -0.081 |
idges | -0.078 |
honey | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
ey | +0.074 |
honey | +0.065 |
yogurt | +0.065 |
sweet | +0.063 |
ad | +0.061 |
Neg logit contributions | |
gy | -0.110 |
bank | -0.090 |
lves | -0.083 |
zag | -0.078 |
drown | -0.075 |
Token | nap |
Token ID | 25422 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
ey | +0.013 |
honey | +0.011 |
yogurt | +0.011 |
ad | +0.011 |
idges | +0.011 |
Neg logit contributions | |
gy | -0.015 |
bank | -0.012 |
lves | -0.010 |
zag | -0.010 |
station | -0.009 |
Token | now |
Token ID | 783 |
Feature activation | +2.54e+00 |
Pos logit contributions | |
ey | +0.085 |
yogurt | +0.072 |
honey | +0.072 |
ad | +0.071 |
idges | +0.071 |
Neg logit contributions | |
gy | -0.114 |
bank | -0.093 |
lves | -0.083 |
zag | -0.080 |
drown | -0.077 |
Token | ." |
Token ID | 526 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
ey | +0.169 |
honey | +0.147 |
yogurt | +0.144 |
ad | +0.142 |
idges | +0.140 |
Neg logit contributions | |
gy | -0.242 |
bank | -0.197 |
lves | -0.178 |
zag | -0.170 |
drown | -0.163 |
Token |
|
Token ID | 198 |
Feature activation | -4.26e-02 |
Pos logit contributions | |
gy | +0.180 |
bank | +0.146 |
zag | +0.120 |
station | +0.118 |
lves | +0.114 |
Neg logit contributions | |
ey | -0.150 |
idges | -0.129 |
honey | -0.129 |
yogurt | -0.122 |
ad | -0.119 |
Token |
|
Token ID | 198 |
Feature activation | -4.56e-02 |
Pos logit contributions | |
gy | +0.004 |
bank | +0.003 |
zag | +0.003 |
lves | +0.002 |
drown | +0.002 |
Neg logit contributions | |
ey | -0.004 |
honey | -0.003 |
yogurt | -0.003 |
idges | -0.003 |
pasta | -0.003 |
Token | John |
Token ID | 7554 |
Feature activation | +5.76e+00 |
Pos logit contributions | |
gy | +0.005 |
bank | +0.004 |
zag | +0.003 |
lves | +0.003 |
station | +0.003 |
Neg logit contributions | |
ey | -0.003 |
honey | -0.003 |
yogurt | -0.002 |
idges | -0.002 |
ad | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | -4.22e-01 |
Pos logit contributions | |
ey | +0.438 |
yogurt | +0.372 |
honey | +0.369 |
ad | +0.366 |
idges | +0.366 |
Neg logit contributions | |
gy | -0.511 |
bank | -0.410 |
lves | -0.356 |
zag | -0.349 |
drown | -0.331 |
Token | the |
Token ID | 262 |
Feature activation | -5.80e-02 |
Pos logit contributions | |
gy | +0.050 |
bank | +0.042 |
lves | +0.037 |
zag | +0.037 |
station | +0.036 |
Neg logit contributions | |
ey | -0.021 |
idges | -0.016 |
yogurt | -0.016 |
honey | -0.016 |
ad | -0.015 |
Token | little |
Token ID | 1310 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
gy | +0.005 |
bank | +0.004 |
lves | +0.003 |
zag | +0.003 |
jet | +0.003 |
Neg logit contributions | |
ey | -0.005 |
idges | -0.004 |
yogurt | -0.004 |
ad | -0.004 |
honey | -0.004 |
Token | squirrel |
Token ID | 33039 |
Feature activation | -5.18e-02 |
Pos logit contributions | |
ey | +0.091 |
idges | +0.078 |
yogurt | +0.077 |
honey | +0.076 |
ad | +0.075 |
Neg logit contributions | |
gy | -0.086 |
bank | -0.068 |
lves | -0.057 |
zag | -0.054 |
station | -0.053 |
Token | stayed |
Token ID | 9658 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
gy | +0.004 |
bank | +0.003 |
lves | +0.003 |
drown | +0.003 |
zag | +0.003 |
Neg logit contributions | |
ad | -0.003 |
honey | -0.003 |
ey | -0.003 |
yogurt | -0.003 |
sweet | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +4.81e-01 |
Pos logit contributions | |
ey | +0.282 |
yogurt | +0.240 |
honey | +0.239 |
idges | +0.235 |
ad | +0.235 |
Neg logit contributions | |
gy | -0.346 |
bank | -0.280 |
lves | -0.244 |
zag | -0.239 |
station | -0.227 |
Token | it |
Token ID | 340 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
ey | +0.036 |
idges | +0.029 |
ad | +0.029 |
ement | +0.027 |
yogurt | +0.026 |
Neg logit contributions | |
gy | -0.043 |
bank | -0.036 |
lves | -0.030 |
neither | -0.030 |
drown | -0.029 |
Token | 's |
Token ID | 338 |
Feature activation | -6.21e+00 |
Pos logit contributions | |
ey | +0.012 |
yogurt | +0.010 |
honey | +0.010 |
ad | +0.010 |
idges | +0.010 |
Neg logit contributions | |
gy | -0.012 |
bank | -0.010 |
lves | -0.008 |
zag | -0.008 |
drown | -0.008 |
Token | so |
Token ID | 523 |
Feature activation | +7.72e-01 |
Pos logit contributions | |
gy | +0.547 |
bank | +0.421 |
lves | +0.350 |
station | +0.346 |
zag | +0.337 |
Neg logit contributions | |
ey | -0.503 |
idges | -0.434 |
yogurt | -0.411 |
ement | -0.391 |
ad | -0.391 |
Token | big |
Token ID | 1263 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
ey | +0.047 |
yogurt | +0.037 |
idges | +0.037 |
honey | +0.036 |
ad | +0.036 |
Neg logit contributions | |
gy | -0.082 |
bank | -0.068 |
lves | -0.061 |
zag | -0.060 |
drown | -0.057 |
Token | and |
Token ID | 290 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
ey | +0.093 |
yogurt | +0.077 |
idges | +0.077 |
ad | +0.075 |
honey | +0.075 |
Neg logit contributions | |
gy | -0.119 |
bank | -0.096 |
lves | -0.082 |
zag | -0.082 |
station | -0.079 |
Token | strong |
Token ID | 1913 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
gy | +0.018 |
bank | +0.015 |
lves | +0.013 |
zag | +0.012 |
station | +0.012 |
Neg logit contributions | |
ey | -0.013 |
idges | -0.011 |
ad | -0.011 |
yogurt | -0.010 |
honey | -0.010 |
Token | !" |
Token ID | 2474 |
Feature activation | +6.00e-01 |
Pos logit contributions | |
gy | +0.024 |
bank | +0.019 |
zag | +0.016 |
lves | +0.016 |
station | +0.016 |
Neg logit contributions | |
ey | -0.019 |
yogurt | -0.016 |
idges | -0.016 |
ad | -0.016 |
honey | -0.015 |
Token | Ben |
Token ID | 3932 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
ey | +0.036 |
yogurt | +0.029 |
idges | +0.029 |
honey | +0.029 |
ad | +0.028 |
Neg logit contributions | |
gy | -0.063 |
bank | -0.053 |
lves | -0.047 |
zag | -0.046 |
drown | -0.044 |
Token | said |
Token ID | 531 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
ey | +0.136 |
yogurt | +0.115 |
ad | +0.114 |
honey | +0.114 |
idges | +0.108 |
Neg logit contributions | |
gy | -0.227 |
bank | -0.187 |
lves | -0.171 |
zag | -0.165 |
drown | -0.162 |
Token | , |
Token ID | 11 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
gy | +0.011 |
bank | +0.009 |
lves | +0.008 |
zag | +0.008 |
station | +0.008 |
Neg logit contributions | |
ey | -0.007 |
yogurt | -0.006 |
honey | -0.006 |
ad | -0.006 |
sweet | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
ey | +0.589 |
idges | +0.538 |
mmmm | +0.506 |
yogurt | +0.505 |
subtract | +0.496 |
Neg logit contributions | |
gy | -0.405 |
bank | -0.272 |
neither | -0.220 |
zag | -0.215 |
station | -0.214 |
Token | hugged |
Token ID | 41130 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
gy | +0.034 |
bank | +0.027 |
lves | +0.024 |
station | +0.023 |
neither | +0.023 |
Neg logit contributions | |
ey | -0.037 |
idges | -0.030 |
ad | -0.030 |
yogurt | -0.029 |
ummy | -0.029 |
Token | them |
Token ID | 606 |
Feature activation | -9.45e-01 |
Pos logit contributions | |
ey | +0.194 |
importantly | +0.166 |
odic | +0.163 |
idges | +0.162 |
mmmm | +0.158 |
Neg logit contributions | |
gy | -0.207 |
bank | -0.154 |
neither | -0.141 |
argue | -0.128 |
lves | -0.126 |
Token | . |
Token ID | 13 |
Feature activation | +8.37e-02 |
Pos logit contributions | |
gy | +0.089 |
bank | +0.073 |
lves | +0.064 |
zag | +0.063 |
drown | +0.060 |
Neg logit contributions | |
ey | -0.067 |
yogurt | -0.056 |
honey | -0.056 |
ad | -0.055 |
idges | -0.055 |
Token | " |
Token ID | 366 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
ey | +0.007 |
ad | +0.006 |
idges | +0.006 |
odic | +0.006 |
yogurt | +0.006 |
Neg logit contributions | |
gy | -0.006 |
bank | -0.004 |
neither | -0.004 |
station | -0.004 |
zag | -0.004 |
Token | We |
Token ID | 1135 |
Feature activation | -9.06e-01 |
Pos logit contributions | |
ey | +0.120 |
yogurt | +0.096 |
honey | +0.096 |
idges | +0.095 |
ad | +0.094 |
Neg logit contributions | |
gy | -0.202 |
bank | -0.170 |
lves | -0.154 |
zag | -0.149 |
station | -0.144 |
Token | can |
Token ID | 460 |
Feature activation | +3.25e+00 |
Pos logit contributions | |
gy | +0.093 |
bank | +0.075 |
zag | +0.070 |
lves | +0.069 |
drown | +0.065 |
Neg logit contributions | |
ey | -0.053 |
ad | -0.047 |
yogurt | -0.047 |
honey | -0.046 |
idges | -0.044 |
Token | learn |
Token ID | 2193 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
ey | +0.261 |
yogurt | +0.221 |
honey | +0.220 |
idges | +0.217 |
ad | +0.214 |
Neg logit contributions | |
gy | -0.279 |
bank | -0.223 |
zag | -0.188 |
lves | -0.187 |
station | -0.178 |
Token | to |
Token ID | 284 |
Feature activation | +4.89e-01 |
Pos logit contributions | |
gy | +0.144 |
bank | +0.122 |
zag | +0.104 |
station | +0.102 |
lves | +0.102 |
Neg logit contributions | |
ey | -0.114 |
ad | -0.093 |
yogurt | -0.090 |
idges | -0.090 |
ango | -0.084 |
Token | be |
Token ID | 307 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
ey | +0.037 |
yogurt | +0.032 |
honey | +0.031 |
idges | +0.031 |
ad | +0.031 |
Neg logit contributions | |
gy | -0.044 |
bank | -0.035 |
lves | -0.030 |
zag | -0.030 |
station | -0.028 |
Token | modest |
Token ID | 12949 |
Feature activation | +7.11e-02 |
Pos logit contributions | |
ey | +0.008 |
honey | +0.007 |
yogurt | +0.006 |
sweet | +0.006 |
ad | +0.006 |
Neg logit contributions | |
gy | -0.015 |
bank | -0.012 |
lves | -0.011 |
zag | -0.011 |
drown | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
gy | +0.077 |
bank | +0.064 |
lves | +0.056 |
zag | +0.055 |
station | +0.053 |
Neg logit contributions | |
ey | -0.050 |
yogurt | -0.041 |
idges | -0.041 |
honey | -0.040 |
ad | -0.040 |
Token | ground |
Token ID | 2323 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
ey | +0.016 |
idges | +0.013 |
yogurt | +0.013 |
ad | +0.013 |
honey | +0.013 |
Neg logit contributions | |
gy | -0.021 |
bank | -0.017 |
lves | -0.015 |
zag | -0.014 |
station | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
ey | +0.079 |
yogurt | +0.067 |
honey | +0.066 |
ad | +0.065 |
idges | +0.065 |
Neg logit contributions | |
gy | -0.106 |
bank | -0.087 |
lves | -0.076 |
zag | -0.075 |
drown | -0.071 |
Token | He |
Token ID | 679 |
Feature activation | +2.97e+00 |
Pos logit contributions | |
ey | +0.022 |
idges | +0.018 |
honey | +0.018 |
ad | +0.018 |
yogurt | +0.017 |
Neg logit contributions | |
gy | -0.029 |
bank | -0.024 |
lves | -0.021 |
station | -0.020 |
zag | -0.020 |
Token | saw |
Token ID | 2497 |
Feature activation | -9.81e-01 |
Pos logit contributions | |
ey | +0.252 |
idges | +0.220 |
honey | +0.217 |
yogurt | +0.216 |
ad | +0.214 |
Neg logit contributions | |
gy | -0.232 |
bank | -0.188 |
lves | -0.152 |
zag | -0.149 |
station | -0.143 |
Token | the |
Token ID | 262 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
gy | +0.090 |
bank | +0.076 |
lves | +0.064 |
station | +0.062 |
neither | +0.061 |
Neg logit contributions | |
ey | -0.072 |
idges | -0.059 |
ad | -0.056 |
yogurt | -0.054 |
ement | -0.054 |
Token | scissors |
Token ID | 41786 |
Feature activation | +3.33e+00 |
Pos logit contributions | |
ey | +0.008 |
idges | +0.007 |
ad | +0.007 |
yogurt | +0.006 |
honey | +0.006 |
Neg logit contributions | |
gy | -0.010 |
bank | -0.009 |
lves | -0.007 |
zag | -0.007 |
station | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
ey | +0.228 |
yogurt | +0.195 |
ad | +0.193 |
honey | +0.192 |
idges | +0.187 |
Neg logit contributions | |
gy | -0.315 |
bank | -0.255 |
lves | -0.235 |
zag | -0.219 |
drown | -0.213 |
Token | the |
Token ID | 262 |
Feature activation | +7.13e-02 |
Pos logit contributions | |
gy | +0.026 |
bank | +0.022 |
lves | +0.018 |
station | +0.017 |
neither | +0.017 |
Neg logit contributions | |
ey | -0.022 |
idges | -0.019 |
ement | -0.018 |
ad | -0.018 |
ummy | -0.018 |
Token | hat |
Token ID | 6877 |
Feature activation | +3.22e-01 |
Pos logit contributions | |
ey | +0.005 |
idges | +0.004 |
ad | +0.004 |
yogurt | +0.004 |
ement | +0.004 |
Neg logit contributions | |
gy | -0.007 |
bank | -0.006 |
lves | -0.005 |
station | -0.005 |
zag | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
ey | +0.022 |
yogurt | +0.019 |
honey | +0.019 |
ad | +0.019 |
idges | +0.018 |
Neg logit contributions | |
gy | -0.030 |
bank | -0.025 |
lves | -0.022 |
zag | -0.021 |
drown | -0.020 |
Token | can |
Token ID | 460 |
Feature activation | +3.48e+00 |
Pos logit contributions | |
ey | +0.218 |
ad | +0.189 |
yogurt | +0.189 |
honey | +0.187 |
idges | +0.185 |
Neg logit contributions | |
gy | -0.234 |
bank | -0.188 |
zag | -0.166 |
lves | -0.162 |
station | -0.151 |
Token | make |
Token ID | 787 |
Feature activation | +9.73e-01 |
Pos logit contributions | |
ey | +0.245 |
ad | +0.232 |
idges | +0.214 |
yogurt | +0.212 |
sweet | +0.205 |
Neg logit contributions | |
gy | -0.300 |
bank | -0.247 |
lves | -0.233 |
zag | -0.197 |
drown | -0.196 |
Token | another |
Token ID | 1194 |
Feature activation | -2.38e+00 |
Pos logit contributions | |
ey | +0.061 |
yogurt | +0.049 |
honey | +0.049 |
ad | +0.048 |
idges | +0.048 |
Neg logit contributions | |
gy | -0.101 |
bank | -0.083 |
lves | -0.074 |
zag | -0.073 |
station | -0.070 |
Token | car |
Token ID | 1097 |
Feature activation | -6.75e+00 |
Pos logit contributions | |
gy | +0.223 |
lves | +0.173 |
bank | +0.167 |
zag | +0.149 |
nes | +0.148 |
Neg logit contributions | |
ey | -0.161 |
yogurt | -0.149 |
honey | -0.149 |
sweet | -0.149 |
pasta | -0.138 |
Token | with |
Token ID | 351 |
Feature activation | +7.78e-02 |
Pos logit contributions | |
gy | +0.657 |
bank | +0.543 |
lves | +0.520 |
drown | +0.460 |
zag | +0.459 |
Neg logit contributions | |
ey | -0.389 |
ad | -0.362 |
honey | -0.358 |
yogurt | -0.355 |
sweet | -0.346 |
Token | another |
Token ID | 1194 |
Feature activation | -2.38e+00 |
Pos logit contributions | |
ey | +0.005 |
yogurt | +0.004 |
honey | +0.004 |
idges | +0.004 |
ad | +0.004 |
Neg logit contributions | |
gy | -0.008 |
bank | -0.006 |
lves | -0.006 |
zag | -0.006 |
station | -0.005 |
Token | box |
Token ID | 3091 |
Feature activation | -3.53e-01 |
Pos logit contributions | |
gy | +0.205 |
lves | +0.153 |
bank | +0.148 |
zag | +0.132 |
drown | +0.129 |
Neg logit contributions | |
ey | -0.171 |
honey | -0.161 |
yogurt | -0.159 |
sweet | -0.157 |
surprises | -0.148 |
Token | . |
Token ID | 13 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
gy | +0.035 |
bank | +0.029 |
lves | +0.027 |
zag | +0.025 |
drown | +0.025 |
Neg logit contributions | |
ey | -0.021 |
honey | -0.019 |
yogurt | -0.019 |
ad | -0.018 |
sweet | -0.018 |
Token | And |
Token ID | 843 |
Feature activation | -4.05e-01 |
Pos logit contributions | |
ey | +0.024 |
honey | +0.020 |
idges | +0.020 |
yogurt | +0.020 |
ad | +0.019 |
Neg logit contributions | |
gy | -0.032 |
bank | -0.025 |
lves | -0.022 |
station | -0.022 |
drown | -0.022 |
Token | you |
Token ID | 345 |
Feature activation | +2.80e+00 |
Pos logit contributions | |
gy | +0.040 |
bank | +0.033 |
lves | +0.029 |
zag | +0.029 |
drown | +0.027 |
Neg logit contributions | |
ey | -0.027 |
yogurt | -0.022 |
honey | -0.022 |
idges | -0.022 |
ad | -0.022 |
Token | can |
Token ID | 460 |
Feature activation | +3.52e+00 |
Pos logit contributions | |
ey | +0.161 |
ad | +0.149 |
honey | +0.144 |
yogurt | +0.142 |
sweet | +0.141 |
Neg logit contributions | |
gy | -0.266 |
bank | -0.216 |
lves | -0.202 |
zag | -0.201 |
ene | -0.187 |
Token | is |
Token ID | 318 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
ey | +0.122 |
idges | +0.107 |
yogurt | +0.104 |
ad | +0.102 |
ummy | +0.100 |
Neg logit contributions | |
gy | -0.157 |
bank | -0.133 |
lves | -0.115 |
zag | -0.110 |
station | -0.108 |
Token | happy |
Token ID | 3772 |
Feature activation | +8.89e-01 |
Pos logit contributions | |
ey | +0.141 |
idges | +0.119 |
yogurt | +0.116 |
ad | +0.115 |
honey | +0.113 |
Neg logit contributions | |
gy | -0.152 |
bank | -0.117 |
lves | -0.103 |
zag | -0.102 |
drown | -0.098 |
Token | . |
Token ID | 13 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
ey | +0.091 |
idges | +0.079 |
mmmm | +0.076 |
layers | +0.074 |
ango | +0.074 |
Neg logit contributions | |
gy | -0.052 |
bank | -0.034 |
neither | -0.033 |
they | -0.029 |
drown | -0.028 |
Token | She |
Token ID | 1375 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
ey | +0.014 |
idges | +0.013 |
yogurt | +0.012 |
pasta | +0.012 |
layers | +0.012 |
Neg logit contributions | |
gy | -0.016 |
bank | -0.011 |
neither | -0.010 |
zag | -0.010 |
lves | -0.010 |
Token | thinks |
Token ID | 6834 |
Feature activation | +4.81e-01 |
Pos logit contributions | |
ey | +0.122 |
idges | +0.105 |
yogurt | +0.103 |
ad | +0.101 |
ummy | +0.098 |
Neg logit contributions | |
gy | -0.161 |
bank | -0.137 |
lves | -0.119 |
zag | -0.114 |
station | -0.112 |
Token | the |
Token ID | 262 |
Feature activation | +3.70e-02 |
Pos logit contributions | |
ey | +0.038 |
idges | +0.031 |
layers | +0.030 |
yogurt | +0.030 |
ad | +0.030 |
Neg logit contributions | |
gy | -0.038 |
bank | -0.030 |
neither | -0.028 |
drown | -0.027 |
zag | -0.027 |
Token | lizard |
Token ID | 42406 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
ey | +0.003 |
idges | +0.003 |
subtract | +0.003 |
ad | +0.003 |
mmmm | +0.002 |
Neg logit contributions | |
gy | -0.003 |
bank | -0.002 |
zag | -0.002 |
lves | -0.002 |
drown | -0.002 |
Token | is |
Token ID | 318 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
ey | +0.002 |
yogurt | +0.002 |
honey | +0.002 |
idges | +0.002 |
ad | +0.002 |
Neg logit contributions | |
gy | -0.003 |
bank | -0.003 |
lves | -0.002 |
zag | -0.002 |
station | -0.002 |
Token | fine |
Token ID | 3734 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
ey | +0.140 |
idges | +0.114 |
ement | +0.111 |
ad | +0.111 |
ummy | +0.110 |
Neg logit contributions | |
gy | -0.156 |
bank | -0.117 |
lves | -0.109 |
neither | -0.102 |
drown | -0.100 |
Token | . |
Token ID | 13 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
ey | +0.139 |
idges | +0.117 |
ad | +0.113 |
yogurt | +0.111 |
honey | +0.107 |
Neg logit contributions | |
gy | -0.131 |
bank | -0.102 |
zag | -0.085 |
neither | -0.083 |
station | -0.081 |
Token | She |
Token ID | 1375 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
ey | +0.015 |
idges | +0.014 |
yogurt | +0.013 |
pasta | +0.013 |
ad | +0.013 |
Neg logit contributions | |
gy | -0.017 |
bank | -0.012 |
neither | -0.011 |
lves | -0.011 |
zag | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
gy | +0.034 |
bank | +0.027 |
lves | +0.024 |
zag | +0.023 |
drown | +0.022 |
Neg logit contributions | |
ey | -0.029 |
yogurt | -0.025 |
honey | -0.025 |
ad | -0.025 |
idges | -0.024 |
Token | always |
Token ID | 1464 |
Feature activation | -3.96e-01 |
Pos logit contributions | |
ey | +0.077 |
honey | +0.071 |
yogurt | +0.070 |
sweet | +0.069 |
ad | +0.067 |
Neg logit contributions | |
gy | -0.104 |
bank | -0.085 |
lves | -0.076 |
zag | -0.074 |
drown | -0.069 |
Token | playing |
Token ID | 2712 |
Feature activation | +3.02e+00 |
Pos logit contributions | |
gy | +0.034 |
bank | +0.027 |
lves | +0.023 |
zag | +0.023 |
drown | +0.022 |
Neg logit contributions | |
ey | -0.032 |
yogurt | -0.027 |
idges | -0.026 |
ad | -0.026 |
honey | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
ey | +0.239 |
yogurt | +0.199 |
idges | +0.199 |
ad | +0.197 |
honey | +0.197 |
Neg logit contributions | |
gy | -0.264 |
bank | -0.214 |
lves | -0.180 |
zag | -0.177 |
station | -0.172 |
Token | One |
Token ID | 1881 |
Feature activation | -5.48e+00 |
Pos logit contributions | |
ey | +0.012 |
idges | +0.011 |
honey | +0.010 |
yogurt | +0.010 |
ad | +0.010 |
Neg logit contributions | |
gy | -0.011 |
bank | -0.008 |
station | -0.007 |
zag | -0.007 |
lves | -0.007 |
Token | day |
Token ID | 1110 |
Feature activation | -2.86e+00 |
Pos logit contributions | |
gy | +0.435 |
bank | +0.348 |
lves | +0.301 |
zag | +0.280 |
drown | +0.274 |
Neg logit contributions | |
ey | -0.460 |
yogurt | -0.421 |
honey | -0.413 |
ad | -0.399 |
sweet | -0.397 |
Token | he |
Token ID | 339 |
Feature activation | +8.52e-01 |
Pos logit contributions | |
gy | +0.233 |
bank | +0.181 |
lves | +0.149 |
zag | +0.148 |
neither | +0.142 |
Neg logit contributions | |
ey | -0.247 |
idges | -0.209 |
ad | -0.206 |
honey | -0.203 |
yogurt | -0.201 |
Token | was |
Token ID | 373 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
ey | +0.055 |
yogurt | +0.050 |
ad | +0.048 |
honey | +0.046 |
pasta | +0.045 |
Neg logit contributions | |
gy | -0.079 |
bank | -0.064 |
lves | -0.061 |
zag | -0.060 |
drown | -0.055 |
Token | playing |
Token ID | 2712 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
ey | +0.058 |
yogurt | +0.054 |
honey | +0.053 |
sweet | +0.051 |
ad | +0.048 |
Neg logit contributions | |
gy | -0.121 |
bank | -0.101 |
lves | -0.095 |
zag | -0.090 |
mer | -0.085 |
Token | outside |
Token ID | 2354 |
Feature activation | +2.52e+00 |
Pos logit contributions | |
ey | +0.241 |
idges | +0.200 |
yogurt | +0.199 |
ad | +0.199 |
honey | +0.197 |
Neg logit contributions | |
gy | -0.264 |
bank | -0.213 |
lves | -0.179 |
zag | -0.178 |
station | -0.171 |
Token | when |
Token ID | 618 |
Feature activation | -9.13e-01 |
Pos logit contributions | |
ey | +0.206 |
yogurt | +0.175 |
honey | +0.174 |
idges | +0.174 |
ad | +0.172 |
Neg logit contributions | |
gy | -0.213 |
bank | -0.168 |
lves | -0.142 |
zag | -0.140 |
station | -0.134 |
Token | kind |
Token ID | 1611 |
Feature activation | +4.31e+00 |
Pos logit contributions | |
ey | +0.021 |
yogurt | +0.017 |
idges | +0.017 |
ad | +0.016 |
honey | +0.016 |
Neg logit contributions | |
gy | -0.033 |
bank | -0.027 |
lves | -0.024 |
zag | -0.023 |
station | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
ey | +0.303 |
yogurt | +0.254 |
honey | +0.253 |
ad | +0.250 |
idges | +0.249 |
Neg logit contributions | |
gy | -0.406 |
bank | -0.331 |
lves | -0.292 |
zag | -0.286 |
drown | -0.273 |
Token | helpful |
Token ID | 7613 |
Feature activation | +4.27e+00 |
Pos logit contributions | |
gy | +0.025 |
bank | +0.021 |
lves | +0.019 |
zag | +0.019 |
drown | +0.018 |
Neg logit contributions | |
ey | -0.015 |
yogurt | -0.012 |
honey | -0.012 |
idges | -0.011 |
ad | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
ey | +0.315 |
yogurt | +0.266 |
honey | +0.264 |
ad | +0.261 |
idges | +0.261 |
Neg logit contributions | |
gy | -0.391 |
bank | -0.316 |
lves | -0.276 |
zag | -0.271 |
drown | -0.257 |
Token | not |
Token ID | 407 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
gy | +0.024 |
bank | +0.020 |
lves | +0.018 |
zag | +0.018 |
station | +0.017 |
Neg logit contributions | |
ey | -0.014 |
honey | -0.012 |
yogurt | -0.012 |
ad | -0.011 |
idges | -0.011 |
Token | gr |
Token ID | 1036 |
Feature activation | -3.91e+00 |
Pos logit contributions | |
gy | +0.251 |
bank | +0.195 |
lves | +0.172 |
zag | +0.169 |
station | +0.158 |
Neg logit contributions | |
ey | -0.248 |
yogurt | -0.220 |
honey | -0.218 |
ad | -0.213 |
idges | -0.209 |
Token | umpy |
Token ID | 32152 |
Feature activation | -6.06e+00 |
Pos logit contributions | |
gy | +0.354 |
bank | +0.303 |
station | +0.255 |
lves | +0.250 |
zag | +0.249 |
Neg logit contributions | |
ey | -0.267 |
ad | -0.242 |
idges | -0.235 |
ummy | -0.215 |
ement | -0.211 |
Token | . |
Token ID | 13 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
gy | +0.509 |
bank | +0.399 |
lves | +0.356 |
zag | +0.343 |
drown | +0.319 |
Neg logit contributions | |
ey | -0.468 |
yogurt | -0.424 |
honey | -0.421 |
ad | -0.411 |
idges | -0.406 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
ey | +0.023 |
yogurt | +0.019 |
ad | +0.019 |
honey | +0.019 |
idges | +0.019 |
Neg logit contributions | |
gy | -0.023 |
bank | -0.019 |
zag | -0.016 |
lves | -0.016 |
drown | -0.015 |
Token | S |
Token ID | 50 |
Feature activation | -3.42e+00 |
Pos logit contributions | |
ey | +0.101 |
idges | +0.089 |
honey | +0.089 |
layers | +0.084 |
subtract | +0.082 |
Neg logit contributions | |
gy | -0.099 |
bank | -0.075 |
zag | -0.066 |
station | -0.062 |
drown | -0.059 |
Token | ara |
Token ID | 3301 |
Feature activation | +9.48e-01 |
Pos logit contributions | |
gy | +0.281 |
bank | +0.227 |
station | +0.190 |
zag | +0.189 |
lves | +0.188 |
Neg logit contributions | |
ey | -0.272 |
honey | -0.231 |
idges | -0.227 |
layers | -0.222 |
importantly | -0.222 |
Token | and |
Token ID | 290 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
ey | +0.052 |
idges | +0.044 |
yogurt | +0.044 |
ad | +0.044 |
honey | +0.043 |
Neg logit contributions | |
gy | -0.060 |
bank | -0.048 |
lves | -0.041 |
zag | -0.040 |
station | -0.039 |
Token | a |
Token ID | 257 |
Feature activation | +3.40e-01 |
Pos logit contributions | |
gy | +0.023 |
bank | +0.019 |
lves | +0.016 |
zag | +0.015 |
drown | +0.015 |
Neg logit contributions | |
ey | -0.020 |
yogurt | -0.017 |
ad | -0.017 |
idges | -0.017 |
honey | -0.016 |
Token | bit |
Token ID | 1643 |
Feature activation | -6.07e+00 |
Pos logit contributions | |
ey | +0.025 |
honey | +0.021 |
yogurt | +0.021 |
ad | +0.020 |
sweet | +0.020 |
Neg logit contributions | |
gy | -0.031 |
bank | -0.025 |
lves | -0.022 |
zag | -0.021 |
drown | -0.020 |
Token | scary |
Token ID | 14343 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
gy | +0.518 |
bank | +0.415 |
lves | +0.357 |
zag | +0.347 |
drown | +0.332 |
Neg logit contributions | |
ey | -0.482 |
yogurt | -0.411 |
idges | -0.406 |
ad | -0.405 |
honey | -0.404 |
Token | . |
Token ID | 13 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
gy | +0.200 |
bank | +0.162 |
lves | +0.141 |
zag | +0.139 |
station | +0.132 |
Neg logit contributions | |
ey | -0.160 |
yogurt | -0.135 |
honey | -0.133 |
idges | -0.133 |
ad | -0.132 |
Token | But |
Token ID | 887 |
Feature activation | -3.10e+00 |
Pos logit contributions | |
ey | +0.018 |
yogurt | +0.015 |
honey | +0.015 |
idges | +0.014 |
ad | +0.014 |
Neg logit contributions | |
gy | -0.026 |
bank | -0.021 |
lves | -0.019 |
zag | -0.018 |
station | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
gy | +0.326 |
bank | +0.273 |
lves | +0.247 |
zag | +0.242 |
drown | +0.231 |
Neg logit contributions | |
ey | -0.179 |
honey | -0.154 |
ad | -0.146 |
yogurt | -0.145 |
sweet | -0.142 |
Token | kitten |
Token ID | 42143 |
Feature activation | -6.92e+00 |
Pos logit contributions | |
ey | +0.009 |
honey | +0.008 |
yogurt | +0.008 |
sweet | +0.008 |
ad | +0.008 |
Neg logit contributions | |
gy | -0.010 |
bank | -0.008 |
lves | -0.007 |
zag | -0.007 |
drown | -0.006 |
Token | was |
Token ID | 373 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
gy | +0.625 |
bank | +0.496 |
lves | +0.455 |
zag | +0.435 |
drown | +0.427 |
Neg logit contributions | |
ey | -0.476 |
honey | -0.425 |
ad | -0.419 |
yogurt | -0.412 |
sweet | -0.402 |
Token | brave |
Token ID | 14802 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ey | +0.095 |
honey | +0.082 |
yogurt | +0.081 |
ad | +0.081 |
idges | +0.079 |
Neg logit contributions | |
gy | -0.117 |
bank | -0.094 |
lves | -0.082 |
zag | -0.080 |
station | -0.076 |
Token | , |
Token ID | 11 |
Feature activation | +4.41e-01 |
Pos logit contributions | |
ey | +0.088 |
yogurt | +0.075 |
idges | +0.074 |
ad | +0.073 |
honey | +0.072 |
Neg logit contributions | |
gy | -0.091 |
bank | -0.072 |
lves | -0.061 |
zag | -0.059 |
station | -0.058 |
Token | One |
Token ID | 1881 |
Feature activation | -5.50e+00 |
Pos logit contributions | |
ey | +0.010 |
idges | +0.009 |
honey | +0.009 |
ad | +0.009 |
yogurt | +0.008 |
Neg logit contributions | |
gy | -0.011 |
bank | -0.009 |
station | -0.007 |
zag | -0.007 |
lves | -0.007 |
Token | day |
Token ID | 1110 |
Feature activation | -2.88e+00 |
Pos logit contributions | |
gy | +0.450 |
bank | +0.375 |
lves | +0.332 |
drown | +0.306 |
zag | +0.292 |
Neg logit contributions | |
ey | -0.434 |
yogurt | -0.403 |
honey | -0.384 |
sweet | -0.384 |
ad | -0.381 |
Token | , |
Token ID | 11 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
gy | +0.261 |
bank | +0.212 |
lves | +0.185 |
zag | +0.180 |
station | +0.172 |
Neg logit contributions | |
ey | -0.213 |
yogurt | -0.182 |
honey | -0.180 |
ad | -0.177 |
idges | -0.177 |
Token | Tim |
Token ID | 5045 |
Feature activation | -3.74e-01 |
Pos logit contributions | |
ey | +0.020 |
idges | +0.017 |
ad | +0.017 |
honey | +0.017 |
yogurt | +0.016 |
Neg logit contributions | |
gy | -0.027 |
bank | -0.022 |
lves | -0.019 |
zag | -0.019 |
station | -0.018 |
Token | my |
Token ID | 1820 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
gy | +0.042 |
bank | +0.036 |
lves | +0.033 |
zag | +0.032 |
drown | +0.032 |
Neg logit contributions | |
ey | -0.019 |
yogurt | -0.015 |
ad | -0.015 |
honey | -0.014 |
idges | -0.014 |
Token | 's |
Token ID | 338 |
Feature activation | -6.41e+00 |
Pos logit contributions | |
ey | +0.187 |
yogurt | +0.183 |
pasta | +0.176 |
sweet | +0.166 |
honey | +0.164 |
Neg logit contributions | |
gy | -0.258 |
bank | -0.217 |
lves | -0.212 |
zag | -0.206 |
drown | -0.200 |
Token | little |
Token ID | 1310 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
gy | +0.575 |
bank | +0.467 |
lves | +0.426 |
drown | +0.393 |
zag | +0.388 |
Neg logit contributions | |
ey | -0.464 |
yogurt | -0.415 |
honey | -0.396 |
sweet | -0.395 |
ad | -0.383 |
Token | sister |
Token ID | 6621 |
Feature activation | -4.04e+00 |
Pos logit contributions | |
ey | +0.083 |
yogurt | +0.077 |
honey | +0.075 |
sweet | +0.073 |
eyes | +0.071 |
Neg logit contributions | |
gy | -0.089 |
bank | -0.071 |
lves | -0.071 |
zag | -0.069 |
drown | -0.062 |
Token | , |
Token ID | 11 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
gy | +0.366 |
bank | +0.311 |
lves | +0.281 |
zag | +0.281 |
drown | +0.265 |
Neg logit contributions | |
ey | -0.261 |
yogurt | -0.249 |
honey | -0.240 |
ad | -0.236 |
sweet | -0.233 |
Token | Sally |
Token ID | 25737 |
Feature activation | -9.37e-01 |
Pos logit contributions | |
ey | +0.020 |
yogurt | +0.016 |
idges | +0.016 |
honey | +0.016 |
ad | +0.016 |
Neg logit contributions | |
gy | -0.028 |
bank | -0.023 |
lves | -0.020 |
zag | -0.020 |
station | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
gy | +0.089 |
bank | +0.076 |
lves | +0.068 |
zag | +0.068 |
drown | +0.065 |
Neg logit contributions | |
ey | -0.058 |
honey | -0.051 |
yogurt | -0.051 |
ad | -0.050 |
idges | -0.048 |
Token | play |
Token ID | 711 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
gy | +0.595 |
bank | +0.483 |
lves | +0.440 |
zag | +0.428 |
drown | +0.412 |
Neg logit contributions | |
ey | -0.425 |
honey | -0.360 |
yogurt | -0.358 |
ad | -0.342 |
idges | -0.339 |
Token | room |
Token ID | 3823 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
ey | +0.057 |
yogurt | +0.052 |
ad | +0.051 |
honey | +0.051 |
idges | +0.048 |
Neg logit contributions | |
gy | -0.101 |
bank | -0.084 |
lves | -0.079 |
drown | -0.075 |
zag | -0.073 |
Token | . |
Token ID | 13 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
ey | +0.142 |
honey | +0.132 |
yogurt | +0.130 |
ad | +0.124 |
idges | +0.123 |
Neg logit contributions | |
gy | -0.252 |
bank | -0.203 |
lves | -0.190 |
drown | -0.184 |
zag | -0.180 |
Token | So |
Token ID | 1406 |
Feature activation | +5.08e-01 |
Pos logit contributions | |
ey | +0.008 |
yogurt | +0.006 |
honey | +0.006 |
idges | +0.006 |
ad | +0.006 |
Neg logit contributions | |
gy | -0.012 |
bank | -0.010 |
lves | -0.009 |
zag | -0.009 |
drown | -0.009 |
Token | one |
Token ID | 530 |
Feature activation | -9.62e+00 |
Pos logit contributions | |
ey | +0.031 |
honey | +0.028 |
yogurt | +0.028 |
sweet | +0.027 |
ad | +0.026 |
Neg logit contributions | |
gy | -0.051 |
bank | -0.043 |
lves | -0.040 |
drown | -0.038 |
zag | -0.037 |
Token | day |
Token ID | 1110 |
Feature activation | -2.88e+00 |
Pos logit contributions | |
gy | +0.558 |
bank | +0.374 |
zag | +0.309 |
station | +0.289 |
lves | +0.286 |
Neg logit contributions | |
ey | -1.024 |
idges | -0.911 |
ad | -0.887 |
yogurt | -0.861 |
honey | -0.855 |
Token | , |
Token ID | 11 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
gy | +0.291 |
bank | +0.247 |
lves | +0.220 |
zag | +0.218 |
drown | +0.209 |
Neg logit contributions | |
ey | -0.171 |
yogurt | -0.159 |
honey | -0.156 |
sweet | -0.147 |
ad | -0.144 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
ey | +0.018 |
yogurt | +0.016 |
honey | +0.016 |
sweet | +0.015 |
idges | +0.015 |
Neg logit contributions | |
gy | -0.029 |
bank | -0.024 |
lves | -0.021 |
zag | -0.021 |
drown | -0.020 |
Token | asked |
Token ID | 1965 |
Feature activation | -8.79e-01 |
Pos logit contributions | |
gy | +0.166 |
bank | +0.135 |
lves | +0.124 |
drown | +0.122 |
zag | +0.121 |
Neg logit contributions | |
ey | -0.126 |
yogurt | -0.110 |
honey | -0.107 |
ad | -0.106 |
sweet | -0.102 |
Token | her |
Token ID | 607 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
gy | +0.094 |
bank | +0.077 |
lves | +0.069 |
station | +0.068 |
zag | +0.067 |
Neg logit contributions | |
ey | -0.054 |
yogurt | -0.041 |
idges | -0.041 |
ad | -0.040 |
honey | -0.039 |
Token | grand |
Token ID | 4490 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
ey | +0.075 |
honey | +0.065 |
yogurt | +0.064 |
idges | +0.062 |
ad | +0.062 |
Neg logit contributions | |
gy | -0.099 |
bank | -0.081 |
lves | -0.071 |
zag | -0.070 |
drown | -0.067 |
Token | fastest |
Token ID | 14162 |
Feature activation | -3.42e+00 |
Pos logit contributions | |
gy | +0.255 |
bank | +0.200 |
zag | +0.187 |
lves | +0.184 |
drown | +0.174 |
Neg logit contributions | |
ey | -0.159 |
honey | -0.144 |
yogurt | -0.136 |
sweet | -0.133 |
ad | -0.131 |
Token | pig |
Token ID | 12967 |
Feature activation | -5.15e+00 |
Pos logit contributions | |
gy | +0.311 |
bank | +0.244 |
lves | +0.222 |
zag | +0.221 |
argue | +0.216 |
Neg logit contributions | |
ey | -0.226 |
honey | -0.211 |
ad | -0.205 |
yogurt | -0.204 |
idges | -0.199 |
Token | , |
Token ID | 11 |
Feature activation | +4.94e-01 |
Pos logit contributions | |
gy | +0.473 |
bank | +0.382 |
lves | +0.365 |
drown | +0.345 |
argue | +0.327 |
Neg logit contributions | |
ey | -0.314 |
ad | -0.296 |
honey | -0.271 |
yogurt | -0.270 |
idges | -0.268 |
Token | the |
Token ID | 262 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
ey | +0.032 |
honey | +0.028 |
yogurt | +0.027 |
ad | +0.027 |
idges | +0.026 |
Neg logit contributions | |
gy | -0.049 |
bank | -0.039 |
zag | -0.035 |
lves | -0.035 |
drown | -0.033 |
Token | smallest |
Token ID | 18197 |
Feature activation | -3.23e-01 |
Pos logit contributions | |
ey | +0.013 |
honey | +0.011 |
yogurt | +0.011 |
ad | +0.010 |
sweet | +0.010 |
Neg logit contributions | |
gy | -0.017 |
bank | -0.014 |
lves | -0.012 |
zag | -0.012 |
drown | -0.012 |
Token | one |
Token ID | 530 |
Feature activation | -9.41e+00 |
Pos logit contributions | |
gy | +0.030 |
bank | +0.024 |
lves | +0.022 |
zag | +0.021 |
drown | +0.020 |
Neg logit contributions | |
ey | -0.023 |
honey | -0.019 |
yogurt | -0.019 |
ad | -0.019 |
idges | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | +5.02e-01 |
Pos logit contributions | |
gy | +0.844 |
bank | +0.682 |
zag | +0.592 |
lves | +0.585 |
station | +0.559 |
Neg logit contributions | |
ey | -0.720 |
yogurt | -0.594 |
idges | -0.590 |
honey | -0.588 |
ad | -0.577 |
Token | won |
Token ID | 1839 |
Feature activation | -3.11e+00 |
Pos logit contributions | |
ey | +0.038 |
yogurt | +0.032 |
honey | +0.031 |
idges | +0.031 |
ad | +0.031 |
Neg logit contributions | |
gy | -0.045 |
bank | -0.036 |
lves | -0.032 |
zag | -0.031 |
station | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
gy | +0.325 |
bank | +0.273 |
lves | +0.245 |
zag | +0.238 |
station | +0.230 |
Neg logit contributions | |
ey | -0.180 |
ad | -0.150 |
honey | -0.148 |
yogurt | -0.148 |
idges | -0.147 |
Token | race |
Token ID | 3234 |
Feature activation | -5.28e+00 |
Pos logit contributions | |
ey | +0.016 |
honey | +0.015 |
yogurt | +0.014 |
sweet | +0.014 |
ad | +0.014 |
Neg logit contributions | |
gy | -0.017 |
bank | -0.012 |
lves | -0.011 |
zag | -0.011 |
drown | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
gy | +0.543 |
bank | +0.439 |
lves | +0.418 |
drown | +0.389 |
zag | +0.388 |
Neg logit contributions | |
ey | -0.291 |
ad | -0.268 |
yogurt | -0.252 |
idges | -0.244 |
honey | -0.242 |
Token | stay |
Token ID | 2652 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
gy | +0.022 |
bank | +0.018 |
lves | +0.015 |
zag | +0.015 |
drown | +0.014 |
Neg logit contributions | |
ey | -0.020 |
idges | -0.017 |
ad | -0.016 |
yogurt | -0.016 |
honey | -0.016 |
Token | with |
Token ID | 351 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
gy | +0.010 |
bank | +0.008 |
zag | +0.007 |
lves | +0.007 |
drown | +0.007 |
Neg logit contributions | |
ey | -0.007 |
sweet | -0.006 |
honey | -0.006 |
yogurt | -0.006 |
ad | -0.006 |
Token | me |
Token ID | 502 |
Feature activation | +5.99e+00 |
Pos logit contributions | |
ey | +0.002 |
ad | +0.001 |
yogurt | +0.001 |
idges | +0.001 |
ement | +0.001 |
Neg logit contributions | |
gy | -0.003 |
bank | -0.003 |
lves | -0.002 |
neither | -0.002 |
station | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
ey | +0.329 |
honey | +0.329 |
sweet | +0.309 |
yogurt | +0.309 |
idges | +0.287 |
Neg logit contributions | |
gy | -0.591 |
bank | -0.492 |
lves | -0.464 |
drown | -0.430 |
zag | -0.422 |
Token | The |
Token ID | 383 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
ey | +0.018 |
idges | +0.015 |
yogurt | +0.015 |
honey | +0.015 |
ad | +0.015 |
Neg logit contributions | |
gy | -0.028 |
bank | -0.022 |
lves | -0.020 |
zag | -0.019 |
station | -0.019 |
Token | bank |
Token ID | 3331 |
Feature activation | -1.23e+01 |
Pos logit contributions | |
gy | +0.214 |
bank | +0.173 |
lves | +0.144 |
zag | +0.141 |
station | +0.138 |
Neg logit contributions | |
ey | -0.213 |
idges | -0.182 |
ad | -0.179 |
yogurt | -0.179 |
honey | -0.178 |
Token | is |
Token ID | 318 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
gy | +1.231 |
bank | +1.024 |
lves | +0.896 |
zag | +0.890 |
station | +0.855 |
Neg logit contributions | |
ey | -0.790 |
yogurt | -0.642 |
honey | -0.636 |
idges | -0.633 |
ad | -0.630 |
Token | not |
Token ID | 407 |
Feature activation | -3.06e+00 |
Pos logit contributions | |
ey | +0.142 |
yogurt | +0.120 |
honey | +0.119 |
idges | +0.118 |
ad | +0.118 |
Neg logit contributions | |
gy | -0.164 |
bank | -0.132 |
lves | -0.114 |
zag | -0.112 |
station | -0.106 |
Token | a |
Token ID | 257 |
Feature activation | +3.66e-01 |
Pos logit contributions | |
gy | +0.315 |
bank | +0.261 |
lves | +0.233 |
zag | +0.232 |
drown | +0.221 |
Neg logit contributions | |
ey | -0.180 |
honey | -0.161 |
yogurt | -0.157 |
sweet | -0.151 |
ad | -0.147 |
Token | safe |
Token ID | 3338 |
Feature activation | -7.70e-01 |
Pos logit contributions | |
ey | +0.025 |
honey | +0.022 |
yogurt | +0.022 |
sweet | +0.022 |
mirror | +0.020 |
Neg logit contributions | |
gy | -0.035 |
bank | -0.027 |
zag | -0.025 |
lves | -0.024 |
drown | -0.023 |
Token | place |
Token ID | 1295 |
Feature activation | -2.98e+00 |
Pos logit contributions | |
gy | +0.065 |
lves | +0.053 |
bank | +0.050 |
zag | +0.048 |
drown | +0.047 |
Neg logit contributions | |
ey | -0.052 |
honey | -0.049 |
yogurt | -0.048 |
sweet | -0.047 |
ad | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | -4.12e-02 |
Pos logit contributions | |
ey | +0.210 |
idges | +0.175 |
mmmm | +0.169 |
ement | +0.169 |
odic | +0.162 |
Neg logit contributions | |
gy | -0.219 |
bank | -0.187 |
lves | -0.161 |
station | -0.154 |
jet | -0.151 |
Token | toes |
Token ID | 23932 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
gy | +0.004 |
bank | +0.003 |
station | +0.002 |
lves | +0.002 |
zag | +0.002 |
Neg logit contributions | |
ey | -0.003 |
idges | -0.003 |
ad | -0.003 |
honey | -0.003 |
yogurt | -0.003 |
Token | were |
Token ID | 547 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
ey | +0.155 |
honey | +0.130 |
yogurt | +0.129 |
idges | +0.129 |
ad | +0.128 |
Neg logit contributions | |
gy | -0.188 |
bank | -0.151 |
zag | -0.130 |
lves | -0.129 |
station | -0.123 |
Token | pink |
Token ID | 11398 |
Feature activation | +2.42e+00 |
Pos logit contributions | |
gy | +0.003 |
bank | +0.002 |
zag | +0.002 |
station | +0.002 |
neither | +0.002 |
Neg logit contributions | |
ey | -0.003 |
idges | -0.002 |
mmmm | -0.002 |
ad | -0.002 |
ement | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | -4.04e-01 |
Pos logit contributions | |
ey | +0.202 |
idges | +0.178 |
ad | +0.171 |
honey | +0.162 |
mmmm | +0.161 |
Neg logit contributions | |
gy | -0.202 |
bank | -0.161 |
zag | -0.131 |
station | -0.124 |
house | -0.119 |
Token | one |
Token ID | 530 |
Feature activation | -9.64e+00 |
Pos logit contributions | |
gy | +0.038 |
bank | +0.030 |
zag | +0.026 |
neither | +0.026 |
lves | +0.025 |
Neg logit contributions | |
ey | -0.030 |
idges | -0.024 |
ad | -0.023 |
ement | -0.023 |
mmmm | -0.021 |
Token | of |
Token ID | 286 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
gy | +0.800 |
bank | +0.644 |
zag | +0.558 |
station | +0.525 |
house | +0.519 |
Neg logit contributions | |
ey | -0.777 |
idges | -0.665 |
ad | -0.637 |
mmmm | -0.605 |
ango | -0.604 |
Token | them |
Token ID | 606 |
Feature activation | -9.08e-01 |
Pos logit contributions | |
ey | +0.224 |
idges | +0.182 |
yogurt | +0.179 |
honey | +0.172 |
ement | +0.167 |
Neg logit contributions | |
gy | -0.228 |
bank | -0.187 |
lves | -0.160 |
neither | -0.153 |
station | -0.151 |
Token | was |
Token ID | 373 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
gy | +0.075 |
bank | +0.060 |
zag | +0.049 |
lves | +0.049 |
station | +0.046 |
Neg logit contributions | |
ey | -0.076 |
idges | -0.064 |
honey | -0.064 |
yogurt | -0.063 |
ad | -0.063 |
Token | black |
Token ID | 2042 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
ey | +0.088 |
idges | +0.072 |
ad | +0.069 |
yogurt | +0.068 |
honey | +0.066 |
Neg logit contributions | |
gy | -0.105 |
bank | -0.083 |
zag | -0.072 |
lves | -0.069 |
station | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
ey | +0.025 |
idges | +0.022 |
ad | +0.020 |
mmmm | +0.020 |
Jamie | +0.020 |
Neg logit contributions | |
gy | -0.023 |
bank | -0.018 |
zag | -0.014 |
station | -0.014 |
house | -0.013 |
Token | so |
Token ID | 523 |
Feature activation | +5.39e-01 |
Pos logit contributions | |
ey | +0.127 |
yogurt | +0.104 |
ad | +0.102 |
idges | +0.100 |
ango | +0.098 |
Neg logit contributions | |
gy | -0.144 |
bank | -0.120 |
zag | -0.103 |
lves | -0.098 |
station | -0.097 |
Token | you |
Token ID | 345 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
ey | +0.039 |
yogurt | +0.033 |
honey | +0.033 |
idges | +0.033 |
ad | +0.033 |
Neg logit contributions | |
gy | -0.050 |
bank | -0.040 |
lves | -0.035 |
zag | -0.034 |
station | -0.033 |
Token | can |
Token ID | 460 |
Feature activation | +3.27e+00 |
Pos logit contributions | |
ey | +0.171 |
ad | +0.149 |
yogurt | +0.146 |
honey | +0.146 |
idges | +0.144 |
Neg logit contributions | |
gy | -0.240 |
bank | -0.193 |
lves | -0.177 |
zag | -0.165 |
station | -0.160 |
Token | manage |
Token ID | 6687 |
Feature activation | +9.69e-02 |
Pos logit contributions | |
ey | +0.243 |
ad | +0.214 |
yogurt | +0.210 |
honey | +0.207 |
idges | +0.207 |
Neg logit contributions | |
gy | -0.287 |
bank | -0.227 |
lves | -0.205 |
zag | -0.190 |
station | -0.187 |
Token | it |
Token ID | 340 |
Feature activation | -8.68e-02 |
Pos logit contributions | |
ey | +0.007 |
yogurt | +0.006 |
ad | +0.006 |
idges | +0.006 |
honey | +0.006 |
Neg logit contributions | |
gy | -0.009 |
bank | -0.007 |
zag | -0.006 |
lves | -0.006 |
station | -0.006 |
Token | one |
Token ID | 530 |
Feature activation | -9.64e+00 |
Pos logit contributions | |
gy | +0.008 |
bank | +0.006 |
lves | +0.005 |
zag | +0.005 |
station | +0.005 |
Neg logit contributions | |
ey | -0.006 |
yogurt | -0.005 |
honey | -0.005 |
idges | -0.005 |
ad | -0.005 |
Token | day |
Token ID | 1110 |
Feature activation | -2.90e+00 |
Pos logit contributions | |
gy | +0.806 |
bank | +0.657 |
zag | +0.566 |
lves | +0.542 |
station | +0.530 |
Neg logit contributions | |
ey | -0.777 |
idges | -0.643 |
ad | -0.640 |
yogurt | -0.632 |
honey | -0.625 |
Token | ." |
Token ID | 526 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
gy | +0.260 |
bank | +0.211 |
lves | +0.180 |
zag | +0.179 |
station | +0.171 |
Neg logit contributions | |
ey | -0.219 |
yogurt | -0.182 |
honey | -0.181 |
idges | -0.180 |
ad | -0.180 |
Token |
|
Token ID | 198 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
gy | +0.162 |
bank | +0.129 |
zag | +0.110 |
station | +0.105 |
neither | +0.102 |
Neg logit contributions | |
ey | -0.159 |
idges | -0.136 |
honey | -0.134 |
yogurt | -0.132 |
ad | -0.126 |
Token |
|
Token ID | 198 |
Feature activation | -3.03e-02 |
Pos logit contributions | |
gy | +0.003 |
bank | +0.002 |
zag | +0.002 |
station | +0.002 |
lves | +0.002 |
Neg logit contributions | |
ey | -0.003 |
honey | -0.003 |
yogurt | -0.003 |
idges | -0.003 |
umm | -0.003 |
Token | Lisa |
Token ID | 44203 |
Feature activation | +2.59e+00 |
Pos logit contributions | |
gy | +0.003 |
bank | +0.002 |
zag | +0.002 |
station | +0.002 |
house | +0.002 |
Neg logit contributions | |
ey | -0.002 |
honey | -0.002 |
yogurt | -0.002 |
ummy | -0.002 |
idges | -0.002 |
Token | there |
Token ID | 612 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
ey | +0.025 |
ad | +0.022 |
idges | +0.022 |
yogurt | +0.021 |
honey | +0.021 |
Neg logit contributions | |
gy | -0.022 |
bank | -0.017 |
lves | -0.014 |
zag | -0.014 |
station | -0.013 |
Token | were |
Token ID | 547 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
gy | +0.127 |
bank | +0.102 |
zag | +0.084 |
station | +0.083 |
lves | +0.082 |
Neg logit contributions | |
ey | -0.124 |
idges | -0.104 |
honey | -0.101 |
yogurt | -0.101 |
ad | -0.099 |
Token | two |
Token ID | 734 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
gy | +0.001 |
bank | +0.001 |
lves | +0.001 |
station | +0.001 |
zag | +0.001 |
Neg logit contributions | |
ey | -0.001 |
idges | -0.001 |
ad | -0.001 |
ement | -0.001 |
yogurt | -0.001 |
Token | stickers |
Token ID | 28568 |
Feature activation | +4.15e+00 |
Pos logit contributions | |
gy | +0.027 |
bank | +0.022 |
lves | +0.020 |
station | +0.019 |
zag | +0.019 |
Neg logit contributions | |
ey | -0.017 |
ad | -0.014 |
yogurt | -0.013 |
idges | -0.013 |
honey | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
ey | +0.269 |
yogurt | +0.233 |
honey | +0.231 |
ad | +0.230 |
idges | +0.222 |
Neg logit contributions | |
gy | -0.400 |
bank | -0.331 |
lves | -0.305 |
drown | -0.289 |
zag | -0.278 |
Token | one |
Token ID | 530 |
Feature activation | -9.61e+00 |
Pos logit contributions | |
ey | +0.020 |
idges | +0.017 |
ad | +0.016 |
yogurt | +0.015 |
honey | +0.015 |
Neg logit contributions | |
gy | -0.028 |
bank | -0.023 |
lves | -0.020 |
zag | -0.019 |
station | -0.019 |
Token | for |
Token ID | 329 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
gy | +0.827 |
bank | +0.674 |
zag | +0.574 |
station | +0.561 |
lves | +0.548 |
Neg logit contributions | |
ey | -0.746 |
idges | -0.638 |
ad | -0.616 |
yogurt | -0.603 |
honey | -0.587 |
Token | Ben |
Token ID | 3932 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
gy | +0.170 |
bank | +0.140 |
lves | +0.122 |
zag | +0.118 |
station | +0.117 |
Neg logit contributions | |
ey | -0.126 |
yogurt | -0.101 |
idges | -0.100 |
ad | -0.099 |
honey | -0.099 |
Token | and |
Token ID | 290 |
Feature activation | -3.84e-01 |
Pos logit contributions | |
ey | +0.117 |
honey | +0.108 |
yogurt | +0.106 |
sweet | +0.100 |
ad | +0.099 |
Neg logit contributions | |
gy | -0.198 |
bank | -0.167 |
lves | -0.155 |
zag | -0.147 |
drown | -0.146 |
Token | one |
Token ID | 530 |
Feature activation | -9.61e+00 |
Pos logit contributions | |
gy | +0.039 |
bank | +0.032 |
lves | +0.029 |
zag | +0.028 |
station | +0.027 |
Neg logit contributions | |
ey | -0.024 |
yogurt | -0.020 |
idges | -0.020 |
honey | -0.020 |
ad | -0.019 |
Token | for |
Token ID | 329 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
gy | +0.816 |
bank | +0.670 |
zag | +0.563 |
station | +0.546 |
lves | +0.530 |
Neg logit contributions | |
ey | -0.765 |
idges | -0.652 |
ad | -0.627 |
yogurt | -0.601 |
honey | -0.588 |