Token | € |
Token ID | 26391 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
gy | +0.519 |
bank | +0.421 |
lves | +0.366 |
station | +0.340 |
zag | +0.338 |
Neg logit contributions | |
ey | -0.427 |
ad | -0.373 |
idges | -0.358 |
yogurt | -0.347 |
ement | -0.346 |
Token | ™ |
Token ID | 8151 |
Feature activation | -4.11e+00 |
Pos logit contributions | |
gy | +0.120 |
bank | +0.101 |
lves | +0.087 |
station | +0.082 |
argue | +0.080 |
Neg logit contributions | |
ey | -0.090 |
ad | -0.075 |
yogurt | -0.074 |
idges | -0.074 |
ement | -0.072 |
Token | m |
Token ID | 76 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
gy | +0.320 |
bank | +0.251 |
lves | +0.205 |
station | +0.191 |
drown | +0.188 |
Neg logit contributions | |
ey | -0.354 |
idges | -0.308 |
yogurt | -0.306 |
ad | -0.302 |
ement | -0.298 |
Token | sorry |
Token ID | 7926 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
ey | +0.186 |
idges | +0.156 |
yogurt | +0.152 |
honey | +0.149 |
ement | +0.147 |
Neg logit contributions | |
gy | -0.214 |
bank | -0.168 |
lves | -0.140 |
zag | -0.138 |
drown | -0.136 |
Token | , |
Token ID | 11 |
Feature activation | +5.50e-01 |
Pos logit contributions | |
ey | +0.183 |
idges | +0.155 |
yogurt | +0.148 |
ad | +0.148 |
ement | +0.145 |
Neg logit contributions | |
gy | -0.190 |
bank | -0.152 |
lves | -0.123 |
zag | -0.121 |
station | -0.119 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.13e+01 |
Pos logit contributions | |
ey | +0.040 |
idges | +0.034 |
ement | +0.032 |
ad | +0.032 |
yogurt | +0.032 |
Neg logit contributions | |
gy | -0.051 |
bank | -0.042 |
lves | -0.035 |
station | -0.034 |
drown | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +4.27e-02 |
Pos logit contributions | |
ey | +0.826 |
idges | +0.697 |
yogurt | +0.695 |
honey | +0.684 |
ad | +0.680 |
Neg logit contributions | |
gy | -1.025 |
bank | -0.838 |
lves | -0.731 |
zag | -0.667 |
station | -0.664 |
Token | I |
Token ID | 314 |
Feature activation | +3.87e+00 |
Pos logit contributions | |
ey | +0.003 |
idges | +0.002 |
ad | +0.002 |
yogurt | +0.002 |
ement | +0.002 |
Neg logit contributions | |
gy | -0.004 |
bank | -0.004 |
lves | -0.003 |
station | -0.003 |
drown | -0.003 |
Token | â |
Token ID | 22940 |
Feature activation | -5.53e+00 |
Pos logit contributions | |
ey | +0.294 |
yogurt | +0.254 |
ad | +0.252 |
idges | +0.248 |
honey | +0.247 |
Neg logit contributions | |
gy | -0.333 |
bank | -0.264 |
lves | -0.234 |
zag | -0.213 |
station | -0.210 |
Token | € |
Token ID | 26391 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
gy | +0.505 |
bank | +0.410 |
lves | +0.346 |
zag | +0.329 |
station | +0.328 |
Neg logit contributions | |
ey | -0.416 |
idges | -0.349 |
yogurt | -0.342 |
ad | -0.335 |
honey | -0.334 |
Token | ™ |
Token ID | 8151 |
Feature activation | -4.79e+00 |
Pos logit contributions | |
gy | +0.133 |
bank | +0.114 |
lves | +0.097 |
station | +0.089 |
drown | +0.086 |
Neg logit contributions | |
ey | -0.115 |
yogurt | -0.101 |
goodness | -0.101 |
idges | -0.099 |
ad | -0.098 |
Token | , |
Token ID | 11 |
Feature activation | +5.89e-01 |
Pos logit contributions | |
ey | +0.024 |
idges | +0.020 |
ement | +0.019 |
ad | +0.019 |
yogurt | +0.019 |
Neg logit contributions | |
gy | -0.036 |
bank | -0.029 |
station | -0.025 |
lves | -0.024 |
zag | -0.024 |
Token | " |
Token ID | 366 |
Feature activation | +2.74e+00 |
Pos logit contributions | |
ey | +0.044 |
idges | +0.037 |
ad | +0.036 |
ement | +0.036 |
yogurt | +0.035 |
Neg logit contributions | |
gy | -0.053 |
bank | -0.044 |
lves | -0.036 |
station | -0.035 |
drown | -0.034 |
Token | What |
Token ID | 2061 |
Feature activation | +7.28e+00 |
Pos logit contributions | |
ey | +0.167 |
idges | +0.133 |
yogurt | +0.130 |
honey | +0.126 |
ad | +0.126 |
Neg logit contributions | |
gy | -0.290 |
bank | -0.243 |
lves | -0.212 |
station | -0.203 |
zag | -0.202 |
Token | 's |
Token ID | 338 |
Feature activation | -5.60e+00 |
Pos logit contributions | |
ey | +0.630 |
idges | +0.533 |
ement | +0.521 |
ad | +0.519 |
yogurt | +0.500 |
Neg logit contributions | |
gy | -0.581 |
bank | -0.457 |
fight | -0.361 |
argue | -0.359 |
station | -0.358 |
Token | wrong |
Token ID | 2642 |
Feature activation | -2.72e-01 |
Pos logit contributions | |
gy | +0.505 |
bank | +0.409 |
lves | +0.344 |
station | +0.326 |
zag | +0.324 |
Neg logit contributions | |
ey | -0.433 |
idges | -0.363 |
ad | -0.351 |
yogurt | -0.350 |
ement | -0.346 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.12e+01 |
Pos logit contributions | |
gy | +0.027 |
bank | +0.022 |
lves | +0.018 |
station | +0.018 |
zag | +0.018 |
Neg logit contributions | |
ey | -0.019 |
idges | -0.016 |
ad | -0.015 |
yogurt | -0.015 |
ement | -0.015 |
Token | ?" |
Token ID | 1701 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
ey | +0.739 |
idges | +0.599 |
yogurt | +0.580 |
ad | +0.578 |
ement | +0.559 |
Neg logit contributions | |
gy | -1.122 |
bank | -0.926 |
lves | -0.791 |
zag | -0.768 |
station | -0.765 |
Token | Ana |
Token ID | 17639 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
ey | +0.086 |
idges | +0.072 |
yogurt | +0.069 |
ad | +0.068 |
honey | +0.067 |
Neg logit contributions | |
gy | -0.120 |
bank | -0.099 |
lves | -0.084 |
station | -0.080 |
zag | -0.080 |
Token | replied |
Token ID | 8712 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
ey | +0.115 |
idges | +0.098 |
yogurt | +0.096 |
ad | +0.093 |
honey | +0.093 |
Neg logit contributions | |
gy | -0.127 |
bank | -0.103 |
lves | -0.084 |
station | -0.081 |
argue | -0.080 |
Token | , |
Token ID | 11 |
Feature activation | +8.22e-01 |
Pos logit contributions | |
ey | +0.152 |
idges | +0.127 |
yogurt | +0.121 |
ad | +0.121 |
ango | +0.118 |
Neg logit contributions | |
gy | -0.123 |
bank | -0.091 |
neither | -0.071 |
zag | -0.069 |
station | -0.068 |
Token | " |
Token ID | 366 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
ey | +0.057 |
idges | +0.046 |
yogurt | +0.044 |
ad | +0.043 |
honey | +0.042 |
Neg logit contributions | |
gy | -0.083 |
bank | -0.067 |
lves | -0.056 |
station | -0.055 |
zag | -0.055 |
Token | be |
Token ID | 307 |
Feature activation | +9.14e-01 |
Pos logit contributions | |
ey | +0.049 |
idges | +0.042 |
ad | +0.040 |
yogurt | +0.040 |
ement | +0.039 |
Neg logit contributions | |
gy | -0.061 |
bank | -0.050 |
lves | -0.042 |
station | -0.041 |
zag | -0.040 |
Token | eh |
Token ID | 17231 |
Feature activation | -3.51e+00 |
Pos logit contributions | |
ey | +0.068 |
ad | +0.059 |
idges | +0.058 |
yogurt | +0.058 |
honey | +0.056 |
Neg logit contributions | |
gy | -0.080 |
bank | -0.066 |
lves | -0.056 |
drown | -0.053 |
argue | -0.053 |
Token | ive |
Token ID | 425 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
gy | +0.296 |
bank | +0.235 |
lves | +0.198 |
station | +0.185 |
drown | +0.183 |
Neg logit contributions | |
ey | -0.282 |
yogurt | -0.239 |
idges | -0.239 |
ad | -0.238 |
honey | -0.232 |
Token | full |
Token ID | 1336 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
gy | +0.038 |
bank | +0.030 |
lves | +0.026 |
zag | +0.024 |
station | +0.024 |
Neg logit contributions | |
ey | -0.033 |
idges | -0.028 |
yogurt | -0.028 |
ad | -0.027 |
honey | -0.027 |
Token | of |
Token ID | 286 |
Feature activation | +3.05e+00 |
Pos logit contributions | |
ey | +0.157 |
idges | +0.125 |
ement | +0.119 |
ango | +0.116 |
ad | +0.110 |
Neg logit contributions | |
gy | -0.188 |
bank | -0.166 |
station | -0.132 |
house | -0.127 |
neither | -0.127 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.11e+01 |
Pos logit contributions | |
ey | +0.225 |
ement | +0.185 |
idges | +0.182 |
ango | +0.175 |
mmmm | +0.174 |
Neg logit contributions | |
gy | -0.264 |
bank | -0.236 |
lves | -0.184 |
drown | -0.183 |
fight | -0.180 |
Token | ! |
Token ID | 0 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
ey | +0.868 |
idges | +0.723 |
yogurt | +0.702 |
ad | +0.700 |
ement | +0.682 |
Neg logit contributions | |
gy | -0.993 |
bank | -0.803 |
lves | -0.661 |
zag | -0.644 |
station | -0.636 |
Token | He |
Token ID | 679 |
Feature activation | +3.00e+00 |
Pos logit contributions | |
ey | +0.167 |
idges | +0.138 |
yogurt | +0.132 |
ad | +0.131 |
ement | +0.130 |
Neg logit contributions | |
gy | -0.230 |
bank | -0.189 |
lves | -0.160 |
station | -0.153 |
zag | -0.153 |
Token | was |
Token ID | 373 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
ey | +0.257 |
idges | +0.221 |
ement | +0.212 |
yogurt | +0.211 |
ad | +0.209 |
Neg logit contributions | |
gy | -0.246 |
bank | -0.199 |
lves | -0.155 |
zag | -0.150 |
station | -0.146 |
Token | so |
Token ID | 523 |
Feature activation | +6.51e-01 |
Pos logit contributions | |
ey | +0.105 |
idges | +0.090 |
ement | +0.085 |
yogurt | +0.084 |
ad | +0.083 |
Neg logit contributions | |
gy | -0.104 |
bank | -0.082 |
lves | -0.066 |
zag | -0.063 |
drown | -0.063 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
ey | +0.060 |
idges | +0.052 |
ad | +0.049 |
ement | +0.049 |
ango | +0.047 |
Neg logit contributions | |
gy | -0.048 |
bank | -0.037 |
drown | -0.030 |
lves | -0.030 |
neither | -0.028 |
Token | big |
Token ID | 1263 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
ey | +0.088 |
idges | +0.079 |
ad | +0.075 |
ement | +0.074 |
yogurt | +0.070 |
Neg logit contributions | |
gy | -0.111 |
bank | -0.096 |
station | -0.078 |
lves | -0.077 |
zag | -0.075 |
Token | bowl |
Token ID | 9396 |
Feature activation | +4.34e+00 |
Pos logit contributions | |
ey | +0.134 |
idges | +0.123 |
ad | +0.112 |
ango | +0.112 |
ement | +0.112 |
Neg logit contributions | |
gy | -0.144 |
bank | -0.120 |
station | -0.096 |
fight | -0.093 |
lves | -0.090 |
Token | of |
Token ID | 286 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
ey | +0.314 |
idges | +0.261 |
yogurt | +0.244 |
honey | +0.242 |
ad | +0.241 |
Neg logit contributions | |
gy | -0.414 |
bank | -0.340 |
station | -0.278 |
zag | -0.276 |
lves | -0.275 |
Token | sweet |
Token ID | 6029 |
Feature activation | +9.71e+00 |
Pos logit contributions | |
ey | +0.243 |
ango | +0.216 |
idges | +0.212 |
ement | +0.211 |
lad | +0.208 |
Neg logit contributions | |
bank | -0.281 |
gy | -0.279 |
fight | -0.204 |
station | -0.204 |
drown | -0.199 |
Token | sticky |
Token ID | 23408 |
Feature activation | +2.74e+00 |
Pos logit contributions | |
ey | +0.678 |
idges | +0.582 |
ango | +0.559 |
ad | +0.537 |
importantly | +0.534 |
Neg logit contributions | |
gy | -0.886 |
bank | -0.801 |
fight | -0.593 |
house | -0.589 |
station | -0.577 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.11e+01 |
Pos logit contributions | |
ey | +0.201 |
idges | +0.173 |
ango | +0.165 |
importantly | +0.160 |
ad | +0.159 |
Neg logit contributions | |
gy | -0.250 |
bank | -0.219 |
station | -0.161 |
fight | -0.161 |
neither | -0.161 |
Token | . |
Token ID | 13 |
Feature activation | +4.13e-01 |
Pos logit contributions | |
ey | +0.837 |
idges | +0.702 |
yogurt | +0.688 |
ad | +0.679 |
honey | +0.670 |
Neg logit contributions | |
gy | -1.007 |
bank | -0.817 |
lves | -0.689 |
zag | -0.655 |
station | -0.651 |
Token | It |
Token ID | 632 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
ey | +0.028 |
idges | +0.024 |
ad | +0.022 |
ement | +0.021 |
honey | +0.021 |
Neg logit contributions | |
gy | -0.041 |
bank | -0.034 |
lves | -0.028 |
station | -0.027 |
house | -0.027 |
Token | looked |
Token ID | 3114 |
Feature activation | +8.18e-01 |
Pos logit contributions | |
ey | +0.120 |
idges | +0.101 |
ement | +0.096 |
ad | +0.095 |
yogurt | +0.094 |
Neg logit contributions | |
gy | -0.118 |
bank | -0.094 |
lves | -0.074 |
zag | -0.072 |
sank | -0.071 |
Token | so |
Token ID | 523 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
ey | +0.066 |
ad | +0.057 |
idges | +0.056 |
ango | +0.055 |
odic | +0.052 |
Neg logit contributions | |
gy | -0.065 |
bank | -0.056 |
lves | -0.046 |
neither | -0.044 |
away | -0.042 |
Token | tasty |
Token ID | 25103 |
Feature activation | +6.57e+00 |
Pos logit contributions | |
ey | +0.043 |
ad | +0.034 |
ango | +0.034 |
idges | +0.033 |
odic | +0.030 |
Neg logit contributions | |
gy | -0.047 |
bank | -0.042 |
neither | -0.037 |
sank | -0.035 |
drown | -0.035 |
Token |
|
Token ID | 198 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
ey | +0.316 |
honey | +0.267 |
yogurt | +0.267 |
ad | +0.267 |
idges | +0.262 |
Neg logit contributions | |
gy | -0.408 |
bank | -0.345 |
lves | -0.301 |
station | -0.279 |
zag | -0.276 |
Token |
|
Token ID | 198 |
Feature activation | +5.46e-01 |
Pos logit contributions | |
ey | +0.019 |
idges | +0.015 |
yogurt | +0.015 |
ad | +0.015 |
honey | +0.015 |
Neg logit contributions | |
gy | -0.029 |
bank | -0.024 |
lves | -0.021 |
zag | -0.020 |
station | -0.020 |
Token | " |
Token ID | 1 |
Feature activation | -9.34e-01 |
Pos logit contributions | |
ey | +0.034 |
idges | +0.027 |
yogurt | +0.027 |
ad | +0.027 |
honey | +0.026 |
Neg logit contributions | |
gy | -0.057 |
bank | -0.047 |
lves | -0.041 |
zag | -0.039 |
station | -0.039 |
Token | Sure |
Token ID | 19457 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
gy | +0.098 |
bank | +0.083 |
lves | +0.072 |
station | +0.069 |
zag | +0.069 |
Neg logit contributions | |
ey | -0.057 |
idges | -0.045 |
yogurt | -0.044 |
ad | -0.043 |
honey | -0.043 |
Token | , |
Token ID | 11 |
Feature activation | +6.35e-01 |
Pos logit contributions | |
ey | +0.160 |
idges | +0.135 |
yogurt | +0.132 |
ad | +0.130 |
honey | +0.128 |
Neg logit contributions | |
gy | -0.175 |
bank | -0.141 |
lves | -0.117 |
zag | -0.111 |
station | -0.111 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.11e+01 |
Pos logit contributions | |
ey | +0.048 |
idges | +0.039 |
ement | +0.039 |
ad | +0.037 |
yogurt | +0.037 |
Neg logit contributions | |
gy | -0.057 |
bank | -0.047 |
lves | -0.039 |
station | -0.039 |
neither | -0.037 |
Token | , |
Token ID | 11 |
Feature activation | +6.10e-01 |
Pos logit contributions | |
ey | +0.887 |
idges | +0.749 |
yogurt | +0.726 |
ad | +0.717 |
ement | +0.705 |
Neg logit contributions | |
gy | -0.978 |
bank | -0.778 |
lves | -0.638 |
zag | -0.623 |
station | -0.617 |
Token | but |
Token ID | 475 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
ey | +0.044 |
ement | +0.036 |
idges | +0.034 |
ad | +0.034 |
odic | +0.032 |
Neg logit contributions | |
gy | -0.056 |
bank | -0.046 |
station | -0.040 |
lves | -0.039 |
neither | -0.039 |
Token | be |
Token ID | 307 |
Feature activation | +4.39e-01 |
Pos logit contributions | |
gy | +0.112 |
bank | +0.090 |
lves | +0.074 |
station | +0.072 |
drown | +0.071 |
Neg logit contributions | |
ey | -0.102 |
idges | -0.086 |
ad | -0.083 |
ement | -0.083 |
yogurt | -0.082 |
Token | careful |
Token ID | 8161 |
Feature activation | +8.80e-01 |
Pos logit contributions | |
ey | +0.035 |
idges | +0.029 |
ad | +0.028 |
ement | +0.028 |
yogurt | +0.028 |
Neg logit contributions | |
gy | -0.039 |
bank | -0.031 |
lves | -0.026 |
station | -0.024 |
drown | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
ey | +0.071 |
idges | +0.059 |
yogurt | +0.058 |
ad | +0.057 |
ement | +0.056 |
Neg logit contributions | |
gy | -0.078 |
bank | -0.061 |
lves | -0.050 |
zag | -0.050 |
station | -0.049 |
Token | ing |
Token ID | 278 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
gy | +0.033 |
bank | +0.028 |
lves | +0.024 |
zag | +0.023 |
station | +0.023 |
Neg logit contributions | |
ey | -0.022 |
idges | -0.017 |
yogurt | -0.017 |
ad | -0.016 |
ement | -0.016 |
Token | ? |
Token ID | 30 |
Feature activation | +2.93e+00 |
Pos logit contributions | |
ey | +0.191 |
idges | +0.163 |
yogurt | +0.160 |
ad | +0.158 |
honey | +0.157 |
Neg logit contributions | |
gy | -0.196 |
bank | -0.156 |
lves | -0.130 |
zag | -0.122 |
station | -0.121 |
Token | Are |
Token ID | 4231 |
Feature activation | +4.19e+00 |
Pos logit contributions | |
ey | +0.206 |
idges | +0.171 |
yogurt | +0.166 |
ad | +0.163 |
honey | +0.163 |
Neg logit contributions | |
gy | -0.281 |
bank | -0.231 |
lves | -0.196 |
station | -0.187 |
zag | -0.187 |
Token | you |
Token ID | 345 |
Feature activation | +3.45e+00 |
Pos logit contributions | |
ey | +0.267 |
idges | +0.206 |
ad | +0.197 |
ement | +0.196 |
honey | +0.195 |
Neg logit contributions | |
gy | -0.431 |
bank | -0.356 |
lves | -0.310 |
neither | -0.306 |
drown | -0.300 |
Token | tired |
Token ID | 10032 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
ey | +0.268 |
idges | +0.227 |
yogurt | +0.224 |
ad | +0.221 |
honey | +0.219 |
Neg logit contributions | |
gy | -0.302 |
bank | -0.244 |
lves | -0.205 |
zag | -0.194 |
station | -0.193 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.11e+01 |
Pos logit contributions | |
gy | +0.283 |
bank | +0.234 |
zag | +0.195 |
station | +0.195 |
drown | +0.194 |
Neg logit contributions | |
ey | -0.192 |
idges | -0.157 |
yogurt | -0.148 |
ement | -0.147 |
ad | -0.146 |
Token | ?" |
Token ID | 1701 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
ey | +0.734 |
idges | +0.597 |
yogurt | +0.580 |
ad | +0.573 |
honey | +0.560 |
Neg logit contributions | |
gy | -1.112 |
bank | -0.919 |
lves | -0.788 |
zag | -0.761 |
station | -0.755 |
Token | she |
Token ID | 673 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
ey | +0.088 |
idges | +0.075 |
ement | +0.069 |
honey | +0.069 |
yogurt | +0.068 |
Neg logit contributions | |
gy | -0.107 |
bank | -0.087 |
lves | -0.071 |
neither | -0.069 |
station | -0.069 |
Token | said |
Token ID | 531 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
ey | +0.031 |
idges | +0.029 |
mmmm | +0.027 |
umm | +0.027 |
ummy | +0.027 |
Neg logit contributions | |
gy | -0.020 |
bank | -0.015 |
argue | -0.011 |
station | -0.011 |
crashed | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | +9.20e-01 |
Pos logit contributions | |
ey | +0.013 |
idges | +0.011 |
yogurt | +0.010 |
mmmm | +0.010 |
ad | +0.010 |
Neg logit contributions | |
gy | -0.011 |
bank | -0.008 |
neither | -0.006 |
zag | -0.006 |
station | -0.006 |
Token | smiling |
Token ID | 16755 |
Feature activation | +8.00e+00 |
Pos logit contributions | |
ey | +0.070 |
idges | +0.059 |
yogurt | +0.055 |
ad | +0.054 |
ement | +0.054 |
Neg logit contributions | |
gy | -0.085 |
bank | -0.067 |
station | -0.056 |
argue | -0.055 |
neither | -0.054 |
Token |
|
Token ID | 198 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
ey | +0.044 |
ad | +0.038 |
idges | +0.037 |
ement | +0.036 |
odic | +0.036 |
Neg logit contributions | |
gy | -0.047 |
bank | -0.039 |
neither | -0.036 |
station | -0.035 |
drown | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | +5.46e-01 |
Pos logit contributions | |
ey | +0.019 |
idges | +0.015 |
yogurt | +0.015 |
ad | +0.015 |
honey | +0.015 |
Neg logit contributions | |
gy | -0.029 |
bank | -0.024 |
lves | -0.021 |
zag | -0.020 |
station | -0.020 |
Token | " |
Token ID | 1 |
Feature activation | -9.34e-01 |
Pos logit contributions | |
ey | +0.034 |
idges | +0.027 |
yogurt | +0.027 |
ad | +0.027 |
honey | +0.026 |
Neg logit contributions | |
gy | -0.057 |
bank | -0.047 |
lves | -0.041 |
zag | -0.039 |
station | -0.039 |
Token | Sorry |
Token ID | 14385 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
gy | +0.098 |
bank | +0.083 |
lves | +0.072 |
station | +0.069 |
zag | +0.069 |
Neg logit contributions | |
ey | -0.057 |
idges | -0.045 |
yogurt | -0.044 |
ad | -0.043 |
honey | -0.043 |
Token | , |
Token ID | 11 |
Feature activation | +6.68e-01 |
Pos logit contributions | |
ey | +0.212 |
idges | +0.175 |
ement | +0.174 |
ad | +0.168 |
odic | +0.167 |
Neg logit contributions | |
gy | -0.167 |
bank | -0.124 |
argue | -0.097 |
station | -0.096 |
neither | -0.095 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.11e+01 |
Pos logit contributions | |
ey | +0.045 |
idges | +0.037 |
ad | +0.036 |
yogurt | +0.036 |
ement | +0.035 |
Neg logit contributions | |
gy | -0.065 |
bank | -0.054 |
lves | -0.046 |
station | -0.045 |
drown | -0.044 |
Token | , |
Token ID | 11 |
Feature activation | +6.74e-01 |
Pos logit contributions | |
ey | +0.796 |
honey | +0.676 |
yogurt | +0.676 |
idges | +0.675 |
ad | +0.659 |
Neg logit contributions | |
gy | -0.999 |
bank | -0.826 |
lves | -0.725 |
zag | -0.655 |
station | -0.653 |
Token | we |
Token ID | 356 |
Feature activation | -3.93e+00 |
Pos logit contributions | |
ey | +0.035 |
ement | +0.029 |
idges | +0.028 |
ad | +0.028 |
ango | +0.025 |
Neg logit contributions | |
gy | -0.073 |
bank | -0.063 |
drown | -0.055 |
station | -0.054 |
neither | -0.054 |
Token | can |
Token ID | 460 |
Feature activation | +3.48e+00 |
Pos logit contributions | |
gy | +0.361 |
bank | +0.292 |
lves | +0.254 |
zag | +0.241 |
station | +0.236 |
Neg logit contributions | |
ey | -0.287 |
yogurt | -0.246 |
idges | -0.246 |
honey | -0.239 |
ad | -0.238 |
Token | 't |
Token ID | 470 |
Feature activation | -4.39e+00 |
Pos logit contributions | |
ey | +0.273 |
idges | +0.236 |
yogurt | +0.235 |
ad | +0.232 |
honey | +0.229 |
Neg logit contributions | |
gy | -0.294 |
bank | -0.235 |
lves | -0.208 |
station | -0.186 |
zag | -0.185 |
Token | . |
Token ID | 13 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
gy | +0.403 |
bank | +0.326 |
lves | +0.286 |
zag | +0.265 |
station | +0.264 |
Neg logit contributions | |
ey | -0.320 |
idges | -0.273 |
yogurt | -0.271 |
ad | -0.267 |
honey | -0.266 |
Token | saw |
Token ID | 2497 |
Feature activation | -4.72e-01 |
Pos logit contributions | |
ey | +0.118 |
idges | +0.102 |
yogurt | +0.101 |
ad | +0.100 |
honey | +0.099 |
Neg logit contributions | |
gy | -0.112 |
bank | -0.088 |
lves | -0.074 |
zag | -0.069 |
station | -0.067 |
Token | a |
Token ID | 257 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
gy | +0.041 |
bank | +0.034 |
lves | +0.027 |
station | +0.026 |
neither | +0.026 |
Neg logit contributions | |
ey | -0.038 |
idges | -0.032 |
ement | -0.030 |
ummy | -0.030 |
mmmm | -0.030 |
Token | big |
Token ID | 1263 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
ey | +0.074 |
idges | +0.065 |
ad | +0.062 |
ement | +0.061 |
yogurt | +0.059 |
Neg logit contributions | |
gy | -0.094 |
bank | -0.079 |
lves | -0.066 |
station | -0.065 |
zag | -0.063 |
Token | cup |
Token ID | 6508 |
Feature activation | -6.03e-01 |
Pos logit contributions | |
ey | +0.118 |
idges | +0.104 |
ad | +0.098 |
yogurt | +0.097 |
ement | +0.097 |
Neg logit contributions | |
gy | -0.126 |
bank | -0.102 |
lves | -0.082 |
station | -0.081 |
drown | -0.078 |
Token | of |
Token ID | 286 |
Feature activation | +3.22e+00 |
Pos logit contributions | |
gy | +0.058 |
bank | +0.049 |
station | +0.041 |
zag | +0.041 |
house | +0.039 |
Neg logit contributions | |
ey | -0.042 |
idges | -0.034 |
ad | -0.031 |
honey | -0.031 |
importantly | -0.031 |
Token | yogurt |
Token ID | 32132 |
Feature activation | +1.11e+01 |
Pos logit contributions | |
ey | +0.228 |
ement | +0.206 |
idges | +0.204 |
lad | +0.189 |
ango | +0.189 |
Neg logit contributions | |
gy | -0.275 |
bank | -0.259 |
station | -0.205 |
lves | -0.194 |
jet | -0.188 |
Token | . |
Token ID | 13 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
ey | +0.908 |
idges | +0.773 |
ad | +0.742 |
honey | +0.718 |
ement | +0.718 |
Neg logit contributions | |
gy | -0.934 |
bank | -0.768 |
lves | -0.599 |
zag | -0.597 |
station | -0.592 |
Token | It |
Token ID | 632 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
ey | +0.030 |
idges | +0.026 |
ad | +0.024 |
ement | +0.024 |
honey | +0.023 |
Neg logit contributions | |
gy | -0.045 |
bank | -0.038 |
lves | -0.031 |
station | -0.031 |
house | -0.030 |
Token | was |
Token ID | 373 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
ey | +0.142 |
idges | +0.124 |
ement | +0.116 |
importantly | +0.115 |
ad | +0.115 |
Neg logit contributions | |
gy | -0.126 |
bank | -0.100 |
house | -0.073 |
zag | -0.073 |
lves | -0.072 |
Token | his |
Token ID | 465 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
ey | +0.125 |
idges | +0.116 |
ement | +0.111 |
odic | +0.104 |
lad | +0.104 |
Neg logit contributions | |
gy | -0.119 |
bank | -0.103 |
neither | -0.078 |
station | -0.077 |
crashed | -0.074 |
Token | favorite |
Token ID | 4004 |
Feature activation | -4.72e+00 |
Pos logit contributions | |
ey | +0.108 |
idges | +0.100 |
ad | +0.092 |
odic | +0.091 |
ement | +0.091 |
Neg logit contributions | |
gy | -0.101 |
bank | -0.083 |
station | -0.063 |
fight | -0.061 |
zag | -0.060 |
Token | be |
Token ID | 307 |
Feature activation | +7.56e-01 |
Pos logit contributions | |
ey | +0.074 |
idges | +0.063 |
ad | +0.061 |
ement | +0.060 |
yogurt | +0.059 |
Neg logit contributions | |
gy | -0.096 |
bank | -0.080 |
lves | -0.067 |
station | -0.065 |
zag | -0.065 |
Token | eh |
Token ID | 17231 |
Feature activation | -3.72e+00 |
Pos logit contributions | |
ey | +0.062 |
idges | +0.053 |
yogurt | +0.052 |
ad | +0.051 |
honey | +0.050 |
Neg logit contributions | |
gy | -0.064 |
bank | -0.051 |
lves | -0.043 |
zag | -0.040 |
station | -0.040 |
Token | ive |
Token ID | 425 |
Feature activation | -5.50e-01 |
Pos logit contributions | |
gy | +0.311 |
bank | +0.247 |
lves | +0.206 |
station | +0.193 |
zag | +0.192 |
Neg logit contributions | |
ey | -0.304 |
idges | -0.259 |
yogurt | -0.258 |
ad | -0.254 |
honey | -0.251 |
Token | full |
Token ID | 1336 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
gy | +0.050 |
bank | +0.040 |
lves | +0.035 |
zag | +0.032 |
drown | +0.032 |
Neg logit contributions | |
ey | -0.041 |
idges | -0.034 |
yogurt | -0.034 |
ad | -0.033 |
honey | -0.033 |
Token | of |
Token ID | 286 |
Feature activation | +2.91e+00 |
Pos logit contributions | |
ey | +0.147 |
idges | +0.121 |
ement | +0.115 |
ango | +0.114 |
Jamie | +0.111 |
Neg logit contributions | |
gy | -0.162 |
bank | -0.145 |
station | -0.113 |
house | -0.111 |
neither | -0.106 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
ey | +0.203 |
ement | +0.174 |
idges | +0.166 |
mmmm | +0.163 |
ango | +0.161 |
Neg logit contributions | |
gy | -0.252 |
bank | -0.233 |
drown | -0.185 |
lves | -0.181 |
fight | -0.179 |
Token | . |
Token ID | 13 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
ey | +0.832 |
idges | +0.694 |
yogurt | +0.676 |
ad | +0.671 |
ement | +0.654 |
Neg logit contributions | |
gy | -1.004 |
bank | -0.816 |
lves | -0.683 |
zag | -0.656 |
station | -0.650 |
Token | He |
Token ID | 679 |
Feature activation | +2.90e+00 |
Pos logit contributions | |
ey | +0.009 |
idges | +0.007 |
yogurt | +0.007 |
ad | +0.007 |
ement | +0.007 |
Neg logit contributions | |
gy | -0.012 |
bank | -0.010 |
lves | -0.009 |
station | -0.008 |
zag | -0.008 |
Token | was |
Token ID | 373 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
ey | +0.249 |
idges | +0.214 |
ement | +0.206 |
yogurt | +0.205 |
ad | +0.204 |
Neg logit contributions | |
gy | -0.237 |
bank | -0.190 |
lves | -0.150 |
zag | -0.143 |
station | -0.141 |
Token | so |
Token ID | 523 |
Feature activation | +5.07e-01 |
Pos logit contributions | |
ey | +0.100 |
idges | +0.086 |
ement | +0.081 |
yogurt | +0.080 |
ad | +0.080 |
Neg logit contributions | |
gy | -0.097 |
bank | -0.076 |
lves | -0.061 |
drown | -0.058 |
zag | -0.058 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
ey | +0.047 |
idges | +0.041 |
ad | +0.039 |
ement | +0.039 |
ango | +0.037 |
Neg logit contributions | |
gy | -0.037 |
bank | -0.029 |
lves | -0.023 |
drown | -0.023 |
neither | -0.022 |
Token | 's |
Token ID | 338 |
Feature activation | -6.08e+00 |
Pos logit contributions | |
ey | +0.078 |
idges | +0.067 |
yogurt | +0.065 |
ad | +0.065 |
honey | +0.064 |
Neg logit contributions | |
gy | -0.072 |
bank | -0.057 |
lves | -0.046 |
zag | -0.044 |
station | -0.043 |
Token | go |
Token ID | 467 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
gy | +0.494 |
bank | +0.390 |
lves | +0.325 |
zag | +0.303 |
station | +0.299 |
Neg logit contributions | |
ey | -0.511 |
idges | -0.438 |
yogurt | -0.433 |
ad | -0.427 |
honey | -0.423 |
Token | have |
Token ID | 423 |
Feature activation | -3.35e+00 |
Pos logit contributions | |
ey | +0.018 |
yogurt | +0.016 |
ad | +0.015 |
idges | +0.015 |
honey | +0.015 |
Neg logit contributions | |
gy | -0.021 |
bank | -0.017 |
lves | -0.015 |
drown | -0.013 |
neither | -0.013 |
Token | some |
Token ID | 617 |
Feature activation | -3.11e+00 |
Pos logit contributions | |
gy | +0.284 |
bank | +0.239 |
station | +0.186 |
drowned | +0.186 |
lves | +0.185 |
Neg logit contributions | |
ey | -0.277 |
idges | -0.224 |
ement | -0.217 |
ad | -0.212 |
ango | -0.210 |
Token | more |
Token ID | 517 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
gy | +0.267 |
bank | +0.248 |
station | +0.200 |
fight | +0.196 |
drown | +0.187 |
Neg logit contributions | |
ey | -0.228 |
idges | -0.181 |
ad | -0.178 |
ango | -0.178 |
ement | -0.174 |
Token | yogurt |
Token ID | 32132 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
ey | +0.131 |
idges | +0.109 |
ad | +0.105 |
ement | +0.102 |
ango | +0.100 |
Neg logit contributions | |
gy | -0.183 |
bank | -0.158 |
station | -0.127 |
lves | -0.126 |
argue | -0.125 |
Token | and |
Token ID | 290 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
ey | +0.791 |
yogurt | +0.667 |
idges | +0.663 |
honey | +0.645 |
ad | +0.641 |
Neg logit contributions | |
gy | -1.020 |
bank | -0.827 |
lves | -0.715 |
zag | -0.669 |
drown | -0.667 |
Token | play |
Token ID | 711 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
gy | +0.019 |
bank | +0.016 |
station | +0.012 |
neither | +0.012 |
argue | +0.012 |
Neg logit contributions | |
ey | -0.018 |
idges | -0.015 |
ement | -0.015 |
ad | -0.014 |
ango | -0.014 |
Token | together |
Token ID | 1978 |
Feature activation | -3.46e+00 |
Pos logit contributions | |
ey | +0.113 |
idges | +0.093 |
ad | +0.090 |
ement | +0.087 |
importantly | +0.087 |
Neg logit contributions | |
gy | -0.097 |
bank | -0.081 |
lves | -0.061 |
station | -0.061 |
house | -0.060 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
gy | +0.304 |
bank | +0.246 |
lves | +0.204 |
zag | +0.196 |
station | +0.194 |
Neg logit contributions | |
ey | -0.273 |
idges | -0.228 |
yogurt | -0.223 |
ad | -0.221 |
honey | -0.217 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
ey | +0.020 |
idges | +0.016 |
yogurt | +0.016 |
ad | +0.016 |
honey | +0.016 |
Neg logit contributions | |
gy | -0.029 |
bank | -0.024 |
lves | -0.021 |
zag | -0.020 |
station | -0.020 |
Token | liked |
Token ID | 8288 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
gy | +0.101 |
bank | +0.081 |
lves | +0.067 |
zag | +0.063 |
station | +0.063 |
Neg logit contributions | |
ey | -0.099 |
idges | -0.084 |
yogurt | -0.083 |
ad | -0.082 |
honey | -0.081 |
Token | to |
Token ID | 284 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
ey | +0.200 |
ement | +0.172 |
idges | +0.169 |
mmmm | +0.164 |
ad | +0.163 |
Neg logit contributions | |
gy | -0.156 |
bank | -0.124 |
lves | -0.090 |
station | -0.090 |
fight | -0.089 |
Token | eat |
Token ID | 4483 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
ey | +0.111 |
idges | +0.096 |
yogurt | +0.093 |
ement | +0.093 |
ad | +0.091 |
Neg logit contributions | |
gy | -0.095 |
bank | -0.075 |
lves | -0.058 |
zag | -0.056 |
station | -0.056 |
Token | y |
Token ID | 331 |
Feature activation | +4.71e+00 |
Pos logit contributions | |
ey | +0.169 |
idges | +0.139 |
ement | +0.134 |
ad | +0.132 |
importantly | +0.130 |
Neg logit contributions | |
gy | -0.184 |
bank | -0.173 |
banks | -0.124 |
lves | -0.122 |
station | -0.121 |
Token | ummy |
Token ID | 13513 |
Feature activation | +9.03e+00 |
Pos logit contributions | |
ey | +0.191 |
Jamie | +0.126 |
idges | +0.120 |
lad | +0.115 |
Adam | +0.113 |
Neg logit contributions | |
gy | -0.569 |
bank | -0.497 |
drown | -0.412 |
banks | -0.410 |
neither | -0.408 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
ey | +0.641 |
idges | +0.508 |
ad | +0.497 |
ement | +0.495 |
importantly | +0.483 |
Neg logit contributions | |
gy | -0.853 |
bank | -0.755 |
fight | -0.571 |
house | -0.570 |
station | -0.566 |
Token | . |
Token ID | 13 |
Feature activation | +5.78e-01 |
Pos logit contributions | |
ey | +0.765 |
yogurt | +0.658 |
idges | +0.651 |
honey | +0.648 |
ad | +0.629 |
Neg logit contributions | |
gy | -1.019 |
bank | -0.828 |
lves | -0.732 |
drown | -0.680 |
station | -0.674 |
Token | One |
Token ID | 1881 |
Feature activation | -5.24e+00 |
Pos logit contributions | |
ey | +0.044 |
idges | +0.037 |
ad | +0.035 |
yogurt | +0.035 |
ement | +0.035 |
Neg logit contributions | |
gy | -0.053 |
bank | -0.043 |
lves | -0.036 |
station | -0.034 |
zag | -0.033 |
Token | day |
Token ID | 1110 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
gy | +0.379 |
bank | +0.288 |
lves | +0.227 |
zag | +0.212 |
station | +0.210 |
Neg logit contributions | |
ey | -0.496 |
idges | -0.433 |
yogurt | -0.423 |
ad | -0.420 |
honey | -0.414 |
Token | , |
Token ID | 11 |
Feature activation | +5.17e-01 |
Pos logit contributions | |
gy | +0.192 |
bank | +0.154 |
lves | +0.129 |
zag | +0.122 |
station | +0.121 |
Neg logit contributions | |
ey | -0.176 |
idges | -0.149 |
yogurt | -0.146 |
ad | -0.144 |
honey | -0.142 |
Token | he |
Token ID | 339 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
ey | +0.030 |
yogurt | +0.026 |
honey | +0.025 |
idges | +0.025 |
ad | +0.024 |
Neg logit contributions | |
gy | -0.053 |
bank | -0.045 |
lves | -0.039 |
zag | -0.038 |
drown | -0.037 |
Token | � |
Token ID | 241 |
Feature activation | -5.47e-01 |
Pos logit contributions | |
gy | +0.189 |
bank | +0.155 |
lves | +0.132 |
drown | +0.126 |
zag | +0.126 |
Neg logit contributions | |
ey | -0.138 |
idges | -0.115 |
ad | -0.113 |
yogurt | -0.112 |
ement | -0.109 |
Token | I |
Token ID | 40 |
Feature activation | +3.86e+00 |
Pos logit contributions | |
gy | +0.052 |
bank | +0.043 |
lves | +0.037 |
zag | +0.036 |
station | +0.035 |
Neg logit contributions | |
ey | -0.038 |
idges | -0.032 |
honey | -0.030 |
yogurt | -0.030 |
ement | -0.029 |
Token | 'm |
Token ID | 1101 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
ey | +0.313 |
ad | +0.269 |
idges | +0.267 |
yogurt | +0.267 |
honey | +0.259 |
Neg logit contributions | |
gy | -0.320 |
bank | -0.256 |
lves | -0.216 |
station | -0.196 |
zag | -0.195 |
Token | scattering |
Token ID | 45765 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
gy | +0.276 |
bank | +0.226 |
lves | +0.193 |
zag | +0.185 |
station | +0.184 |
Neg logit contributions | |
ey | -0.206 |
idges | -0.169 |
yogurt | -0.165 |
ad | -0.162 |
honey | -0.160 |
Token | the |
Token ID | 262 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
gy | +0.090 |
bank | +0.078 |
lves | +0.064 |
station | +0.060 |
neither | +0.060 |
Neg logit contributions | |
ey | -0.080 |
idges | -0.067 |
ango | -0.065 |
ement | -0.064 |
importantly | -0.063 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
ey | +0.008 |
idges | +0.007 |
ad | +0.007 |
ement | +0.007 |
mmmm | +0.007 |
Neg logit contributions | |
gy | -0.011 |
bank | -0.009 |
station | -0.007 |
lves | -0.007 |
zag | -0.007 |
Token | so |
Token ID | 523 |
Feature activation | +6.40e-01 |
Pos logit contributions | |
ey | +0.790 |
idges | +0.658 |
yogurt | +0.646 |
ad | +0.636 |
honey | +0.628 |
Neg logit contributions | |
gy | -1.023 |
bank | -0.836 |
lves | -0.712 |
zag | -0.677 |
station | -0.673 |
Token | I |
Token ID | 314 |
Feature activation | +3.65e+00 |
Pos logit contributions | |
ey | +0.049 |
idges | +0.037 |
ad | +0.037 |
ement | +0.036 |
ango | +0.035 |
Neg logit contributions | |
gy | -0.057 |
bank | -0.047 |
lves | -0.041 |
neither | -0.040 |
zag | -0.039 |
Token | can |
Token ID | 460 |
Feature activation | +3.25e+00 |
Pos logit contributions | |
ey | +0.296 |
idges | +0.252 |
yogurt | +0.251 |
ad | +0.250 |
honey | +0.243 |
Neg logit contributions | |
gy | -0.305 |
bank | -0.242 |
lves | -0.204 |
zag | -0.190 |
station | -0.189 |
Token | feed |
Token ID | 3745 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
ey | +0.280 |
idges | +0.238 |
yogurt | +0.233 |
ement | +0.229 |
honey | +0.229 |
Neg logit contributions | |
gy | -0.263 |
bank | -0.209 |
lves | -0.165 |
zag | -0.162 |
drown | -0.158 |
Token | the |
Token ID | 262 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
gy | +0.211 |
bank | +0.185 |
lves | +0.147 |
neither | +0.143 |
station | +0.141 |
Neg logit contributions | |
ey | -0.201 |
importantly | -0.164 |
idges | -0.157 |
ement | -0.157 |
ad | -0.148 |
Token | " |
Token ID | 366 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
gy | +0.122 |
bank | +0.100 |
lves | +0.086 |
zag | +0.081 |
station | +0.081 |
Neg logit contributions | |
ey | -0.090 |
idges | -0.074 |
yogurt | -0.072 |
ad | -0.071 |
ement | -0.070 |
Token | Don |
Token ID | 3987 |
Feature activation | +8.93e-01 |
Pos logit contributions | |
ey | +0.147 |
idges | +0.118 |
honey | +0.113 |
yogurt | +0.112 |
ement | +0.109 |
Neg logit contributions | |
gy | -0.252 |
bank | -0.212 |
lves | -0.184 |
station | -0.176 |
zag | -0.176 |
Token | 't |
Token ID | 470 |
Feature activation | -4.27e+00 |
Pos logit contributions | |
ey | +0.081 |
idges | +0.070 |
yogurt | +0.069 |
honey | +0.068 |
ad | +0.067 |
Neg logit contributions | |
gy | -0.068 |
bank | -0.052 |
lves | -0.042 |
zag | -0.040 |
station | -0.039 |
Token | worry |
Token ID | 5490 |
Feature activation | -6.82e-01 |
Pos logit contributions | |
gy | +0.387 |
bank | +0.317 |
lves | +0.273 |
zag | +0.259 |
station | +0.256 |
Neg logit contributions | |
ey | -0.318 |
ad | -0.270 |
idges | -0.266 |
yogurt | -0.265 |
honey | -0.262 |
Token | , |
Token ID | 11 |
Feature activation | +6.22e-01 |
Pos logit contributions | |
gy | +0.055 |
bank | +0.041 |
neither | +0.032 |
station | +0.032 |
lves | +0.032 |
Neg logit contributions | |
ey | -0.062 |
idges | -0.051 |
ad | -0.050 |
ement | -0.049 |
yogurt | -0.049 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
ey | +0.049 |
ement | +0.040 |
idges | +0.040 |
ad | +0.039 |
yogurt | +0.037 |
Neg logit contributions | |
gy | -0.055 |
bank | -0.044 |
station | -0.037 |
lves | -0.036 |
neither | -0.036 |
Token | ! |
Token ID | 0 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
ey | +0.845 |
idges | +0.711 |
yogurt | +0.695 |
ad | +0.687 |
honey | +0.674 |
Neg logit contributions | |
gy | -0.975 |
bank | -0.784 |
lves | -0.655 |
zag | -0.627 |
station | -0.623 |
Token | I |
Token ID | 314 |
Feature activation | +3.80e+00 |
Pos logit contributions | |
ey | +0.148 |
idges | +0.118 |
yogurt | +0.114 |
ad | +0.113 |
ement | +0.111 |
Neg logit contributions | |
gy | -0.249 |
bank | -0.205 |
lves | -0.178 |
station | -0.171 |
drown | -0.171 |
Token | 'm |
Token ID | 1101 |
Feature activation | -3.09e+00 |
Pos logit contributions | |
ey | +0.276 |
ad | +0.244 |
honey | +0.239 |
yogurt | +0.237 |
idges | +0.237 |
Neg logit contributions | |
gy | -0.319 |
bank | -0.260 |
lves | -0.233 |
zag | -0.209 |
station | -0.207 |
Token | going |
Token ID | 1016 |
Feature activation | +3.16e+00 |
Pos logit contributions | |
gy | +0.269 |
bank | +0.215 |
lves | +0.182 |
drown | +0.176 |
station | +0.169 |
Neg logit contributions | |
ey | -0.247 |
idges | -0.206 |
ad | -0.199 |
yogurt | -0.198 |
ement | -0.196 |
Token | to |
Token ID | 284 |
Feature activation | +8.00e-01 |
Pos logit contributions | |
ey | +0.214 |
yogurt | +0.188 |
idges | +0.185 |
honey | +0.184 |
ad | +0.181 |
Neg logit contributions | |
gy | -0.288 |
bank | -0.235 |
lves | -0.211 |
drown | -0.192 |
zag | -0.191 |
Token | and |
Token ID | 290 |
Feature activation | -7.64e-02 |
Pos logit contributions | |
gy | +0.075 |
bank | +0.060 |
lves | +0.054 |
drown | +0.052 |
station | +0.050 |
Neg logit contributions | |
ey | -0.055 |
yogurt | -0.050 |
ad | -0.048 |
idges | -0.047 |
honey | -0.046 |
Token | said |
Token ID | 531 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
gy | +0.007 |
bank | +0.005 |
lves | +0.005 |
station | +0.004 |
zag | +0.004 |
Neg logit contributions | |
ey | -0.006 |
idges | -0.005 |
ad | -0.005 |
yogurt | -0.005 |
ement | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | +9.71e-01 |
Pos logit contributions | |
ey | +0.016 |
idges | +0.013 |
yogurt | +0.012 |
ad | +0.012 |
ement | +0.012 |
Neg logit contributions | |
gy | -0.015 |
bank | -0.012 |
lves | -0.010 |
neither | -0.010 |
zag | -0.009 |
Token | " |
Token ID | 366 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
ey | +0.065 |
idges | +0.053 |
ad | +0.050 |
yogurt | +0.050 |
ement | +0.049 |
Neg logit contributions | |
gy | -0.098 |
bank | -0.081 |
station | -0.068 |
lves | -0.068 |
zag | -0.067 |
Token | No |
Token ID | 2949 |
Feature activation | -7.05e-02 |
Pos logit contributions | |
ey | +0.172 |
idges | +0.137 |
honey | +0.134 |
yogurt | +0.132 |
ement | +0.130 |
Neg logit contributions | |
gy | -0.266 |
bank | -0.221 |
lves | -0.196 |
zag | -0.185 |
station | -0.184 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
gy | +0.006 |
bank | +0.005 |
lves | +0.004 |
zag | +0.004 |
station | +0.004 |
Neg logit contributions | |
ey | -0.006 |
idges | -0.005 |
yogurt | -0.005 |
ad | -0.005 |
ement | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | +6.50e-01 |
Pos logit contributions | |
ey | +0.784 |
idges | +0.671 |
yogurt | +0.667 |
honey | +0.661 |
ad | +0.652 |
Neg logit contributions | |
gy | -0.981 |
bank | -0.815 |
lves | -0.705 |
station | -0.645 |
drown | -0.641 |
Token | you |
Token ID | 345 |
Feature activation | +2.94e+00 |
Pos logit contributions | |
ey | +0.046 |
idges | +0.037 |
ement | +0.037 |
ad | +0.036 |
yogurt | +0.034 |
Neg logit contributions | |
gy | -0.061 |
bank | -0.050 |
lves | -0.043 |
drown | -0.042 |
neither | -0.042 |
Token | can |
Token ID | 460 |
Feature activation | +3.44e+00 |
Pos logit contributions | |
ey | +0.172 |
honey | +0.164 |
yogurt | +0.158 |
idges | +0.158 |
sweet | +0.157 |
Neg logit contributions | |
gy | -0.290 |
bank | -0.237 |
lves | -0.213 |
station | -0.195 |
zag | -0.193 |
Token | 't |
Token ID | 470 |
Feature activation | -4.40e+00 |
Pos logit contributions | |
ey | +0.270 |
ad | +0.238 |
yogurt | +0.237 |
idges | +0.237 |
honey | +0.229 |
Neg logit contributions | |
gy | -0.286 |
bank | -0.228 |
lves | -0.203 |
station | -0.181 |
zag | -0.175 |
Token | catch |
Token ID | 4929 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
gy | +0.402 |
bank | +0.326 |
lves | +0.304 |
station | +0.276 |
sank | +0.269 |
Neg logit contributions | |
ey | -0.285 |
ad | -0.267 |
yogurt | -0.263 |
idges | -0.257 |
honey | -0.255 |
Token | said |
Token ID | 531 |
Feature activation | -7.12e-02 |
Pos logit contributions | |
ey | +0.341 |
umm | +0.301 |
mmmm | +0.295 |
idges | +0.294 |
goodness | +0.286 |
Neg logit contributions | |
gy | -0.239 |
bank | -0.182 |
argue | -0.144 |
station | -0.143 |
zag | -0.133 |
Token | , |
Token ID | 11 |
Feature activation | +6.96e-01 |
Pos logit contributions | |
gy | +0.005 |
bank | +0.004 |
neither | +0.003 |
zag | +0.003 |
argue | +0.003 |
Neg logit contributions | |
ey | -0.007 |
idges | -0.006 |
ango | -0.006 |
ad | -0.006 |
mmmm | -0.006 |
Token | " |
Token ID | 366 |
Feature activation | +2.42e+00 |
Pos logit contributions | |
ey | +0.058 |
idges | +0.049 |
ad | +0.046 |
ement | +0.045 |
ango | +0.044 |
Neg logit contributions | |
gy | -0.058 |
bank | -0.046 |
station | -0.038 |
neither | -0.037 |
argue | -0.037 |
Token | Yes |
Token ID | 5297 |
Feature activation | +3.82e+00 |
Pos logit contributions | |
ey | +0.195 |
idges | +0.159 |
ement | +0.155 |
honey | +0.153 |
ocol | +0.147 |
Neg logit contributions | |
gy | -0.190 |
bank | -0.161 |
lves | -0.139 |
neither | -0.136 |
argue | -0.133 |
Token | , |
Token ID | 11 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
ey | +0.368 |
idges | +0.297 |
ad | +0.290 |
yogurt | +0.285 |
ango | +0.284 |
Neg logit contributions | |
gy | -0.275 |
bank | -0.208 |
neither | -0.173 |
station | -0.162 |
lves | -0.160 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
ey | +0.052 |
ement | +0.042 |
ad | +0.041 |
odic | +0.040 |
idges | +0.039 |
Neg logit contributions | |
gy | -0.040 |
neither | -0.032 |
bank | -0.031 |
they | -0.030 |
station | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | +5.54e-01 |
Pos logit contributions | |
ey | +0.939 |
idges | +0.778 |
ement | +0.738 |
ad | +0.738 |
importantly | +0.733 |
Neg logit contributions | |
gy | -0.950 |
bank | -0.714 |
zag | -0.596 |
station | -0.586 |
lves | -0.550 |
Token | you |
Token ID | 345 |
Feature activation | +2.95e+00 |
Pos logit contributions | |
ey | +0.048 |
ement | +0.040 |
ad | +0.038 |
odic | +0.038 |
ango | +0.036 |
Neg logit contributions | |
gy | -0.040 |
bank | -0.031 |
neither | -0.030 |
they | -0.028 |
station | -0.028 |
Token | look |
Token ID | 804 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
ey | +0.226 |
idges | +0.190 |
yogurt | +0.185 |
ad | +0.183 |
honey | +0.180 |
Neg logit contributions | |
gy | -0.266 |
bank | -0.216 |
lves | -0.182 |
zag | -0.172 |
station | -0.172 |
Token | very |
Token ID | 845 |
Feature activation | -6.18e-01 |
Pos logit contributions | |
gy | +0.015 |
bank | +0.012 |
lves | +0.010 |
station | +0.009 |
away | +0.009 |
Neg logit contributions | |
ey | -0.015 |
ad | -0.013 |
idges | -0.013 |
subtract | -0.012 |
mmmm | -0.012 |
Token | nice |
Token ID | 3621 |
Feature activation | +4.83e+00 |
Pos logit contributions | |
gy | +0.050 |
bank | +0.044 |
lves | +0.038 |
drown | +0.035 |
neither | +0.035 |
Neg logit contributions | |
ey | -0.046 |
ad | -0.039 |
idges | -0.037 |
mmmm | -0.036 |
ango | -0.036 |
Token | he |
Token ID | 339 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
gy | +0.148 |
bank | +0.122 |
lves | +0.097 |
station | +0.094 |
zag | +0.092 |
Neg logit contributions | |
ey | -0.142 |
idges | -0.120 |
ad | -0.114 |
ement | -0.114 |
yogurt | -0.113 |
Token | would |
Token ID | 561 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
ey | +0.109 |
idges | +0.093 |
yogurt | +0.089 |
ement | +0.089 |
ad | +0.088 |
Neg logit contributions | |
gy | -0.094 |
bank | -0.074 |
lves | -0.059 |
zag | -0.055 |
station | -0.055 |
Token | turn |
Token ID | 1210 |
Feature activation | -3.84e-01 |
Pos logit contributions | |
ey | +0.141 |
idges | +0.121 |
yogurt | +0.118 |
ement | +0.117 |
ad | +0.115 |
Neg logit contributions | |
gy | -0.130 |
bank | -0.103 |
lves | -0.082 |
zag | -0.078 |
drown | -0.078 |
Token | into |
Token ID | 656 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
gy | +0.030 |
bank | +0.024 |
lves | +0.018 |
station | +0.017 |
zag | +0.017 |
Neg logit contributions | |
ey | -0.034 |
idges | -0.029 |
ad | -0.028 |
ement | -0.027 |
mmmm | -0.027 |
Token | delicious |
Token ID | 12625 |
Feature activation | +8.75e+00 |
Pos logit contributions | |
gy | +0.086 |
bank | +0.073 |
station | +0.055 |
lves | +0.054 |
drown | +0.053 |
Neg logit contributions | |
ey | -0.079 |
idges | -0.068 |
ement | -0.063 |
ango | -0.062 |
ad | -0.062 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
ey | +0.604 |
idges | +0.517 |
ango | +0.491 |
ad | +0.480 |
importantly | +0.463 |
Neg logit contributions | |
gy | -0.794 |
bank | -0.731 |
station | -0.545 |
fight | -0.538 |
banks | -0.531 |
Token | . |
Token ID | 13 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
ey | +0.811 |
idges | +0.682 |
yogurt | +0.672 |
ad | +0.660 |
honey | +0.657 |
Neg logit contributions | |
gy | -0.986 |
bank | -0.799 |
lves | -0.681 |
zag | -0.642 |
station | -0.637 |
Token |
|
Token ID | 198 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
ey | +0.017 |
idges | +0.014 |
yogurt | +0.013 |
ad | +0.013 |
honey | +0.013 |
Neg logit contributions | |
gy | -0.022 |
bank | -0.018 |
lves | -0.015 |
station | -0.014 |
zag | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
ey | +0.021 |
idges | +0.017 |
honey | +0.017 |
yogurt | +0.017 |
ement | +0.017 |
Neg logit contributions | |
gy | -0.018 |
bank | -0.015 |
zag | -0.012 |
lves | -0.012 |
station | -0.011 |
Token | One |
Token ID | 3198 |
Feature activation | -3.22e+00 |
Pos logit contributions | |
ey | +0.021 |
ement | +0.016 |
honey | +0.016 |
idges | +0.016 |
yogurt | +0.016 |
Neg logit contributions | |
gy | -0.023 |
bank | -0.019 |
lves | -0.016 |
zag | -0.016 |
station | -0.016 |
Token | day |
Token ID | 1110 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
gy | +0.209 |
bank | +0.152 |
lves | +0.113 |
zag | +0.111 |
station | +0.108 |
Neg logit contributions | |
ey | -0.328 |
idges | -0.289 |
ement | -0.276 |
ad | -0.275 |
yogurt | -0.272 |
Token | replied |
Token ID | 8712 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
ey | +0.260 |
idges | +0.206 |
importantly | +0.201 |
mmmm | +0.199 |
yogurt | +0.194 |
Neg logit contributions | |
gy | -0.306 |
bank | -0.245 |
station | -0.207 |
argue | -0.197 |
zag | -0.193 |
Token | , |
Token ID | 11 |
Feature activation | +4.14e-01 |
Pos logit contributions | |
ey | +0.146 |
idges | +0.122 |
ango | +0.116 |
ad | +0.116 |
yogurt | +0.114 |
Neg logit contributions | |
gy | -0.106 |
bank | -0.079 |
neither | -0.061 |
zag | -0.056 |
station | -0.056 |
Token | " |
Token ID | 366 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
ey | +0.033 |
idges | +0.027 |
ad | +0.026 |
yogurt | +0.025 |
ement | +0.025 |
Neg logit contributions | |
gy | -0.038 |
bank | -0.030 |
station | -0.025 |
neither | -0.024 |
zag | -0.024 |
Token | No |
Token ID | 2949 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
ey | +0.159 |
idges | +0.127 |
ement | +0.124 |
honey | +0.122 |
yogurt | +0.121 |
Neg logit contributions | |
gy | -0.192 |
bank | -0.162 |
lves | -0.146 |
neither | -0.134 |
station | -0.132 |
Token | , |
Token ID | 11 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
gy | +0.024 |
bank | +0.019 |
lves | +0.015 |
station | +0.015 |
zag | +0.015 |
Neg logit contributions | |
ey | -0.024 |
idges | -0.020 |
yogurt | -0.019 |
ad | -0.019 |
ement | -0.019 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
ey | +0.031 |
ad | +0.025 |
ement | +0.024 |
idges | +0.024 |
ango | +0.023 |
Neg logit contributions | |
gy | -0.036 |
bank | -0.030 |
lves | -0.025 |
neither | -0.025 |
station | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
ey | +0.820 |
idges | +0.701 |
yogurt | +0.694 |
honey | +0.688 |
ad | +0.678 |
Neg logit contributions | |
gy | -0.953 |
bank | -0.781 |
lves | -0.667 |
zag | -0.614 |
station | -0.613 |
Token | Let |
Token ID | 3914 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
ey | +0.011 |
idges | +0.009 |
yogurt | +0.008 |
ad | +0.008 |
honey | +0.008 |
Neg logit contributions | |
gy | -0.020 |
bank | -0.017 |
lves | -0.014 |
zag | -0.014 |
station | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
ey | +0.123 |
honey | +0.119 |
idges | +0.115 |
yogurt | +0.115 |
sweet | +0.115 |
Neg logit contributions | |
gy | -0.160 |
bank | -0.125 |
zag | -0.112 |
lves | -0.108 |
drown | -0.104 |
Token | flower |
Token ID | 15061 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
ey | +0.010 |
idges | +0.009 |
ad | +0.008 |
ement | +0.008 |
yogurt | +0.008 |
Neg logit contributions | |
gy | -0.013 |
bank | -0.011 |
lves | -0.009 |
station | -0.009 |
zag | -0.008 |
Token | bloom |
Token ID | 29955 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
ey | +0.125 |
honey | +0.119 |
yogurt | +0.110 |
idges | +0.108 |
sweet | +0.106 |
Neg logit contributions | |
gy | -0.171 |
bank | -0.139 |
lves | -0.130 |
drown | -0.118 |
drowned | -0.118 |
Token | sad |
Token ID | 6507 |
Feature activation | -5.56e+00 |
Pos logit contributions | |
ey | +0.124 |
idges | +0.105 |
yogurt | +0.103 |
ad | +0.102 |
honey | +0.101 |
Neg logit contributions | |
gy | -0.134 |
bank | -0.107 |
lves | -0.090 |
zag | -0.085 |
station | -0.084 |
Token | because |
Token ID | 780 |
Feature activation | -6.15e+00 |
Pos logit contributions | |
gy | +0.441 |
bank | +0.345 |
lves | +0.264 |
neither | +0.263 |
zag | +0.261 |
Neg logit contributions | |
ey | -0.492 |
idges | -0.423 |
ad | -0.408 |
yogurt | -0.401 |
ement | -0.392 |
Token | he |
Token ID | 339 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
gy | +0.589 |
bank | +0.484 |
lves | +0.407 |
station | +0.398 |
neither | +0.395 |
Neg logit contributions | |
ey | -0.429 |
idges | -0.348 |
ement | -0.330 |
ad | -0.327 |
yogurt | -0.324 |
Token | wanted |
Token ID | 2227 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
ey | +0.101 |
idges | +0.087 |
yogurt | +0.085 |
ad | +0.084 |
ement | +0.083 |
Neg logit contributions | |
gy | -0.086 |
bank | -0.067 |
lves | -0.054 |
zag | -0.050 |
station | -0.049 |
Token | the |
Token ID | 262 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
gy | +0.082 |
bank | +0.066 |
station | +0.049 |
fight | +0.048 |
drown | +0.048 |
Neg logit contributions | |
ey | -0.098 |
idges | -0.083 |
ement | -0.083 |
mmmm | -0.078 |
ad | -0.077 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
gy | +0.010 |
bank | +0.008 |
station | +0.007 |
lves | +0.007 |
drown | +0.007 |
Neg logit contributions | |
ey | -0.008 |
idges | -0.007 |
ad | -0.007 |
ement | -0.007 |
mmmm | -0.007 |
Token | so |
Token ID | 523 |
Feature activation | +6.18e-01 |
Pos logit contributions | |
ey | +0.835 |
yogurt | +0.718 |
idges | +0.710 |
honey | +0.705 |
ad | +0.694 |
Neg logit contributions | |
gy | -0.939 |
bank | -0.756 |
lves | -0.655 |
zag | -0.600 |
station | -0.599 |
Token | much |
Token ID | 881 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
ey | +0.059 |
idges | +0.051 |
yogurt | +0.051 |
ad | +0.050 |
honey | +0.050 |
Neg logit contributions | |
gy | -0.044 |
bank | -0.034 |
lves | -0.027 |
zag | -0.025 |
station | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | +4.64e-01 |
Pos logit contributions | |
gy | +0.115 |
bank | +0.100 |
neither | +0.073 |
fight | +0.069 |
they | +0.066 |
Neg logit contributions | |
ey | -0.188 |
idges | -0.161 |
ango | -0.153 |
Jamie | -0.152 |
ad | -0.148 |
Token | but |
Token ID | 475 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
ey | +0.035 |
idges | +0.029 |
yogurt | +0.028 |
ad | +0.028 |
ement | +0.028 |
Neg logit contributions | |
gy | -0.042 |
bank | -0.034 |
lves | -0.029 |
station | -0.027 |
drown | -0.027 |
Token | he |
Token ID | 339 |
Feature activation | +9.25e-01 |
Pos logit contributions | |
gy | +0.148 |
bank | +0.124 |
lves | +0.109 |
zag | +0.105 |
drown | +0.103 |
Neg logit contributions | |
ey | -0.081 |
yogurt | -0.066 |
idges | -0.064 |
honey | -0.064 |
ad | -0.063 |
Token | the |
Token ID | 262 |
Feature activation | -6.09e-02 |
Pos logit contributions | |
ey | +0.040 |
idges | +0.038 |
ad | +0.034 |
ement | +0.034 |
heart | +0.033 |
Neg logit contributions | |
gy | -0.044 |
bank | -0.035 |
drown | -0.030 |
neither | -0.028 |
lves | -0.028 |
Token | jar |
Token ID | 17379 |
Feature activation | +6.72e-01 |
Pos logit contributions | |
gy | +0.005 |
bank | +0.004 |
lves | +0.004 |
station | +0.003 |
zag | +0.003 |
Neg logit contributions | |
ey | -0.005 |
idges | -0.004 |
ad | -0.004 |
yogurt | -0.004 |
ement | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
ey | +0.049 |
idges | +0.041 |
yogurt | +0.040 |
ad | +0.040 |
honey | +0.039 |
Neg logit contributions | |
gy | -0.062 |
bank | -0.050 |
lves | -0.043 |
zag | -0.041 |
station | -0.040 |
Token | have |
Token ID | 423 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
ey | +0.015 |
idges | +0.013 |
ement | +0.013 |
ad | +0.013 |
yogurt | +0.012 |
Neg logit contributions | |
gy | -0.014 |
bank | -0.012 |
lves | -0.009 |
station | -0.009 |
drown | -0.009 |
Token | some |
Token ID | 617 |
Feature activation | -3.06e+00 |
Pos logit contributions | |
gy | +0.248 |
bank | +0.214 |
drowned | +0.164 |
station | +0.160 |
lves | +0.157 |
Neg logit contributions | |
ey | -0.260 |
idges | -0.213 |
ement | -0.208 |
ango | -0.200 |
ad | -0.200 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.08e+01 |
Pos logit contributions | |
gy | +0.245 |
bank | +0.236 |
fight | +0.177 |
station | +0.177 |
house | +0.170 |
Neg logit contributions | |
ey | -0.228 |
ango | -0.193 |
idges | -0.188 |
ad | -0.181 |
ement | -0.179 |
Token | ," |
Token ID | 553 |
Feature activation | -4.28e-01 |
Pos logit contributions | |
ey | +0.730 |
yogurt | +0.654 |
honey | +0.647 |
idges | +0.627 |
ad | +0.609 |
Neg logit contributions | |
gy | -0.994 |
bank | -0.818 |
lves | -0.737 |
drown | -0.677 |
argue | -0.662 |
Token | he |
Token ID | 339 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
gy | +0.041 |
bank | +0.034 |
lves | +0.028 |
station | +0.027 |
argue | +0.027 |
Neg logit contributions | |
ey | -0.031 |
idges | -0.025 |
yogurt | -0.024 |
ement | -0.024 |
ad | -0.024 |
Token | said |
Token ID | 531 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
ey | +0.097 |
idges | +0.082 |
yogurt | +0.078 |
ement | +0.078 |
ad | +0.077 |
Neg logit contributions | |
gy | -0.105 |
bank | -0.086 |
lves | -0.069 |
station | -0.067 |
argue | -0.067 |
Token | to |
Token ID | 284 |
Feature activation | +7.66e-01 |
Pos logit contributions | |
gy | +0.012 |
bank | +0.010 |
lves | +0.008 |
zag | +0.008 |
station | +0.008 |
Neg logit contributions | |
ey | -0.010 |
idges | -0.009 |
yogurt | -0.009 |
ad | -0.008 |
honey | -0.008 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
ey | +0.052 |
idges | +0.042 |
ement | +0.040 |
yogurt | +0.039 |
importantly | +0.038 |
Neg logit contributions | |
gy | -0.078 |
bank | -0.063 |
lves | -0.052 |
drown | -0.052 |
station | -0.052 |
Token | the |
Token ID | 262 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
gy | +0.320 |
bank | +0.265 |
lves | +0.227 |
station | +0.217 |
zag | +0.217 |
Neg logit contributions | |
ey | -0.217 |
idges | -0.177 |
yogurt | -0.173 |
ad | -0.170 |
honey | -0.166 |
Token | bees |
Token ID | 17002 |
Feature activation | +3.98e+00 |
Pos logit contributions | |
ey | +0.029 |
idges | +0.024 |
ad | +0.024 |
ement | +0.023 |
yogurt | +0.023 |
Neg logit contributions | |
gy | -0.036 |
bank | -0.031 |
lves | -0.025 |
station | -0.025 |
argue | -0.024 |
Token | started |
Token ID | 2067 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
ey | +0.302 |
idges | +0.260 |
honey | +0.252 |
ad | +0.251 |
yogurt | +0.251 |
Neg logit contributions | |
gy | -0.351 |
bank | -0.279 |
lves | -0.238 |
zag | -0.224 |
station | -0.220 |
Token | eating |
Token ID | 6600 |
Feature activation | +5.02e+00 |
Pos logit contributions | |
ey | +0.017 |
idges | +0.015 |
yogurt | +0.014 |
ad | +0.014 |
honey | +0.014 |
Neg logit contributions | |
gy | -0.019 |
bank | -0.016 |
lves | -0.013 |
zag | -0.012 |
station | -0.012 |
Token | more |
Token ID | 517 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
ey | +0.425 |
idges | +0.362 |
ad | +0.351 |
importantly | +0.345 |
ango | +0.344 |
Neg logit contributions | |
gy | -0.393 |
bank | -0.346 |
argue | -0.247 |
fight | -0.246 |
lves | -0.243 |
Token | honey |
Token ID | 12498 |
Feature activation | +1.08e+01 |
Pos logit contributions | |
ey | +0.132 |
idges | +0.109 |
ad | +0.108 |
ango | +0.107 |
ement | +0.104 |
Neg logit contributions | |
gy | -0.168 |
bank | -0.151 |
drown | -0.113 |
lves | -0.113 |
house | -0.112 |
Token | too |
Token ID | 1165 |
Feature activation | -9.55e-01 |
Pos logit contributions | |
ey | +0.785 |
idges | +0.663 |
yogurt | +0.663 |
honey | +0.655 |
ad | +0.643 |
Neg logit contributions | |
gy | -0.989 |
bank | -0.801 |
lves | -0.700 |
station | -0.649 |
drown | -0.647 |
Token | . |
Token ID | 13 |
Feature activation | +5.87e-02 |
Pos logit contributions | |
gy | +0.083 |
bank | +0.068 |
lves | +0.055 |
zag | +0.053 |
drown | +0.053 |
Neg logit contributions | |
ey | -0.077 |
idges | -0.064 |
ad | -0.062 |
yogurt | -0.061 |
ement | -0.060 |
Token | They |
Token ID | 1119 |
Feature activation | -3.86e+00 |
Pos logit contributions | |
ey | +0.004 |
idges | +0.004 |
yogurt | +0.004 |
ad | +0.004 |
honey | +0.003 |
Neg logit contributions | |
gy | -0.005 |
bank | -0.004 |
lves | -0.004 |
zag | -0.003 |
station | -0.003 |
Token | all |
Token ID | 477 |
Feature activation | -3.82e+00 |
Pos logit contributions | |
gy | +0.333 |
bank | +0.261 |
lves | +0.225 |
zag | +0.216 |
station | +0.210 |
Neg logit contributions | |
ey | -0.299 |
yogurt | -0.260 |
idges | -0.257 |
honey | -0.256 |
ad | -0.256 |
Token | became |
Token ID | 2627 |
Feature activation | -2.54e+00 |
Pos logit contributions | |
gy | +0.320 |
bank | +0.254 |
lves | +0.212 |
zag | +0.200 |
station | +0.198 |
Neg logit contributions | |
ey | -0.312 |
idges | -0.267 |
yogurt | -0.264 |
ad | -0.261 |
honey | -0.258 |
Token | that |
Token ID | 326 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
ey | +0.073 |
idges | +0.060 |
yogurt | +0.060 |
honey | +0.057 |
layers | +0.056 |
Neg logit contributions | |
gy | -0.074 |
bank | -0.058 |
lves | -0.049 |
station | -0.046 |
zag | -0.044 |
Token | day |
Token ID | 1110 |
Feature activation | -2.71e+00 |
Pos logit contributions | |
gy | +0.089 |
bank | +0.072 |
lves | +0.061 |
drown | +0.056 |
argue | +0.054 |
Neg logit contributions | |
ey | -0.089 |
ad | -0.076 |
yogurt | -0.076 |
honey | -0.076 |
idges | -0.075 |
Token | on |
Token ID | 319 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
gy | +0.252 |
bank | +0.222 |
lves | +0.182 |
drown | +0.175 |
argue | +0.175 |
Neg logit contributions | |
ey | -0.178 |
ad | -0.162 |
idges | -0.161 |
importantly | -0.160 |
ement | -0.153 |
Token | , |
Token ID | 11 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
gy | +0.121 |
bank | +0.099 |
lves | +0.085 |
drown | +0.079 |
zag | +0.079 |
Neg logit contributions | |
ey | -0.102 |
idges | -0.086 |
yogurt | -0.084 |
ad | -0.084 |
honey | -0.083 |
Token | Mag |
Token ID | 2944 |
Feature activation | -5.54e+00 |
Pos logit contributions | |
ey | +0.008 |
importantly | +0.007 |
ad | +0.007 |
idges | +0.007 |
sweet | +0.007 |
Neg logit contributions | |
gy | -0.011 |
bank | -0.010 |
argue | -0.009 |
drown | -0.008 |
fight | -0.008 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.63e+01 |
Pos logit contributions | |
gy | +0.039 |
bank | +0.028 |
argue | -0.047 |
drown | -0.052 |
fight | -0.062 |
Neg logit contributions | |
ey | -0.803 |
ad | -0.744 |
idges | -0.738 |
ement | -0.721 |
ummy | -0.721 |
Token | was |
Token ID | 373 |
Feature activation | +9.19e-01 |
Pos logit contributions | |
gy | +1.505 |
bank | +1.238 |
lves | +1.131 |
drown | +1.074 |
argue | +1.048 |
Neg logit contributions | |
ey | -1.031 |
ad | -0.934 |
idges | -0.892 |
yogurt | -0.880 |
honey | -0.865 |
Token | never |
Token ID | 1239 |
Feature activation | -5.07e+00 |
Pos logit contributions | |
ey | +0.074 |
idges | +0.065 |
ad | +0.064 |
yogurt | +0.064 |
honey | +0.064 |
Neg logit contributions | |
gy | -0.075 |
bank | -0.061 |
lves | -0.050 |
zag | -0.047 |
station | -0.047 |
Token | fearful |
Token ID | 26844 |
Feature activation | -4.78e+00 |
Pos logit contributions | |
gy | +0.429 |
bank | +0.352 |
lves | +0.289 |
zag | +0.285 |
station | +0.276 |
Neg logit contributions | |
ey | -0.396 |
yogurt | -0.341 |
honey | -0.338 |
ad | -0.334 |
idges | -0.334 |
Token | of |
Token ID | 286 |
Feature activation | +2.57e+00 |
Pos logit contributions | |
gy | +0.429 |
bank | +0.345 |
lves | +0.294 |
drown | +0.279 |
neither | +0.278 |
Neg logit contributions | |
ey | -0.364 |
idges | -0.302 |
yogurt | -0.292 |
ad | -0.288 |
honey | -0.280 |
Token | boards |
Token ID | 11490 |
Feature activation | -5.93e-02 |
Pos logit contributions | |
ey | +0.188 |
idges | +0.157 |
mmmm | +0.151 |
ement | +0.150 |
yogurt | +0.150 |
Neg logit contributions | |
gy | -0.235 |
bank | -0.193 |
lves | -0.162 |
station | -0.156 |
drown | -0.155 |
Token | board |
Token ID | 3096 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
ey | +0.001 |
idges | +0.001 |
ad | +0.001 |
honey | +0.001 |
ement | +0.001 |
Neg logit contributions | |
gy | -0.001 |
bank | -0.001 |
station | -0.000 |
lves | -0.000 |
zag | -0.000 |
Token | far |
Token ID | 1290 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
gy | +0.092 |
bank | +0.072 |
lves | +0.069 |
drown | +0.064 |
argue | +0.061 |
Neg logit contributions | |
ey | -0.080 |
yogurt | -0.074 |
ad | -0.074 |
idges | -0.070 |
honey | -0.070 |
Token | away |
Token ID | 1497 |
Feature activation | -7.01e+00 |
Pos logit contributions | |
gy | +0.202 |
bank | +0.149 |
lves | +0.130 |
drown | +0.129 |
zag | +0.121 |
Neg logit contributions | |
ey | -0.269 |
ad | -0.248 |
ement | -0.246 |
honey | -0.240 |
ummy | -0.237 |
Token | . |
Token ID | 13 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
gy | +0.664 |
bank | +0.547 |
lves | +0.488 |
drown | +0.451 |
zag | +0.450 |
Neg logit contributions | |
ey | -0.474 |
idges | -0.393 |
ad | -0.391 |
yogurt | -0.390 |
honey | -0.386 |
Token | Mag |
Token ID | 2944 |
Feature activation | -5.51e+00 |
Pos logit contributions | |
ey | +0.008 |
ad | +0.007 |
idges | +0.007 |
importantly | +0.007 |
ement | +0.007 |
Neg logit contributions | |
gy | -0.012 |
bank | -0.011 |
lves | -0.009 |
drown | -0.009 |
argue | -0.009 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.62e+01 |
Pos logit contributions | |
gy | +0.055 |
bank | +0.038 |
argue | -0.035 |
drown | -0.036 |
fight | -0.043 |
Neg logit contributions | |
ey | -0.780 |
ad | -0.722 |
idges | -0.712 |
ement | -0.702 |
ummy | -0.694 |
Token | was |
Token ID | 373 |
Feature activation | +9.44e-01 |
Pos logit contributions | |
gy | +1.522 |
bank | +1.221 |
lves | +1.111 |
drown | +1.068 |
zag | +1.046 |
Neg logit contributions | |
ey | -1.037 |
ad | -0.922 |
idges | -0.886 |
yogurt | -0.886 |
honey | -0.857 |
Token | so |
Token ID | 523 |
Feature activation | +3.81e-01 |
Pos logit contributions | |
ey | +0.062 |
ad | +0.052 |
idges | +0.051 |
yogurt | +0.051 |
honey | +0.050 |
Neg logit contributions | |
gy | -0.091 |
bank | -0.076 |
lves | -0.066 |
zag | -0.062 |
argue | -0.062 |
Token | relieved |
Token ID | 22598 |
Feature activation | +8.50e-01 |
Pos logit contributions | |
ey | +0.027 |
idges | +0.023 |
yogurt | +0.022 |
ad | +0.022 |
honey | +0.022 |
Neg logit contributions | |
gy | -0.036 |
bank | -0.030 |
lves | -0.025 |
zag | -0.024 |
station | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
ey | +0.072 |
idges | +0.060 |
yogurt | +0.058 |
ement | +0.056 |
ad | +0.056 |
Neg logit contributions | |
gy | -0.072 |
bank | -0.056 |
lves | -0.045 |
station | -0.044 |
zag | -0.044 |
Token | she |
Token ID | 673 |
Feature activation | -5.26e-01 |
Pos logit contributions | |
ey | +0.012 |
idges | +0.010 |
ad | +0.010 |
yogurt | +0.010 |
honey | +0.010 |
Neg logit contributions | |
gy | -0.020 |
bank | -0.016 |
lves | -0.014 |
zag | -0.014 |
drown | -0.013 |
Token | After |
Token ID | 3260 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
ey | +0.006 |
idges | +0.005 |
yogurt | +0.005 |
honey | +0.005 |
ement | +0.005 |
Neg logit contributions | |
gy | -0.010 |
bank | -0.008 |
lves | -0.007 |
zag | -0.007 |
station | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
ey | +0.119 |
idges | +0.098 |
yogurt | +0.093 |
ement | +0.093 |
ad | +0.092 |
Neg logit contributions | |
gy | -0.144 |
bank | -0.118 |
lves | -0.097 |
station | -0.094 |
zag | -0.093 |
Token | while |
Token ID | 981 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
ey | +0.020 |
honey | +0.018 |
yogurt | +0.018 |
sweet | +0.017 |
idges | +0.017 |
Neg logit contributions | |
gy | -0.023 |
bank | -0.018 |
lves | -0.015 |
drown | -0.014 |
zag | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
gy | +0.190 |
bank | +0.156 |
lves | +0.137 |
zag | +0.126 |
station | +0.125 |
Neg logit contributions | |
ey | -0.145 |
ad | -0.123 |
idges | -0.123 |
yogurt | -0.123 |
honey | -0.122 |
Token | Mag |
Token ID | 2944 |
Feature activation | -5.51e+00 |
Pos logit contributions | |
ey | +0.013 |
yogurt | +0.012 |
honey | +0.012 |
ad | +0.011 |
idges | +0.011 |
Neg logit contributions | |
gy | -0.024 |
bank | -0.020 |
lves | -0.018 |
zag | -0.018 |
argue | -0.018 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.62e+01 |
Pos logit contributions | |
gy | +0.349 |
bank | +0.306 |
argue | +0.232 |
lves | +0.230 |
drown | +0.228 |
Neg logit contributions | |
ey | -0.503 |
ad | -0.450 |
ement | -0.434 |
idges | -0.433 |
goodness | -0.425 |
Token | and |
Token ID | 290 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
gy | +1.439 |
bank | +1.138 |
lves | +1.071 |
drown | +1.006 |
zag | +0.985 |
Neg logit contributions | |
ey | -1.066 |
ad | -0.981 |
yogurt | -0.937 |
honey | -0.907 |
idges | -0.906 |
Token | her |
Token ID | 607 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
gy | +0.032 |
bank | +0.028 |
drown | +0.025 |
lves | +0.024 |
argue | +0.024 |
Neg logit contributions | |
ey | -0.021 |
ad | -0.019 |
honey | -0.018 |
sweet | -0.017 |
importantly | -0.017 |
Token | friend |
Token ID | 1545 |
Feature activation | -3.33e+00 |
Pos logit contributions | |
ey | +0.083 |
idges | +0.072 |
ad | +0.067 |
yogurt | +0.067 |
mmmm | +0.067 |
Neg logit contributions | |
gy | -0.087 |
bank | -0.070 |
station | -0.056 |
lves | -0.055 |
zag | -0.055 |
Token | had |
Token ID | 550 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
gy | +0.296 |
bank | +0.240 |
lves | +0.206 |
zag | +0.194 |
drown | +0.193 |
Neg logit contributions | |
ey | -0.247 |
idges | -0.211 |
ad | -0.209 |
yogurt | -0.209 |
honey | -0.205 |
Token | carried |
Token ID | 5281 |
Feature activation | -6.75e-01 |
Pos logit contributions | |
gy | +0.104 |
bank | +0.086 |
station | +0.068 |
lves | +0.068 |
drowned | +0.066 |
Neg logit contributions | |
ey | -0.098 |
idges | -0.081 |
ement | -0.078 |
ad | -0.074 |
yogurt | -0.074 |
Token | was |
Token ID | 373 |
Feature activation | +9.19e-01 |
Pos logit contributions | |
gy | +0.129 |
bank | +0.097 |
zag | +0.076 |
station | +0.076 |
lves | +0.076 |
Neg logit contributions | |
ey | -0.137 |
idges | -0.120 |
honey | -0.112 |
yogurt | -0.110 |
ad | -0.109 |
Token | a |
Token ID | 257 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
ey | +0.069 |
idges | +0.058 |
ement | +0.055 |
ad | +0.054 |
yogurt | +0.053 |
Neg logit contributions | |
gy | -0.085 |
bank | -0.068 |
lves | -0.058 |
station | -0.055 |
zag | -0.054 |
Token | bug |
Token ID | 5434 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
ey | +0.012 |
yogurt | +0.009 |
idges | +0.009 |
honey | +0.009 |
ad | +0.009 |
Neg logit contributions | |
gy | -0.019 |
bank | -0.016 |
lves | -0.014 |
zag | -0.013 |
drown | -0.013 |
Token | named |
Token ID | 3706 |
Feature activation | -6.65e+00 |
Pos logit contributions | |
gy | +0.002 |
bank | +0.002 |
lves | +0.002 |
argue | +0.002 |
drown | +0.002 |
Neg logit contributions | |
ey | -0.003 |
yogurt | -0.003 |
ad | -0.003 |
honey | -0.003 |
idges | -0.003 |
Token | Bug |
Token ID | 15217 |
Feature activation | -3.27e+00 |
Pos logit contributions | |
gy | +0.617 |
bank | +0.486 |
lves | +0.419 |
zag | +0.393 |
station | +0.384 |
Neg logit contributions | |
ey | -0.504 |
idges | -0.421 |
ement | -0.412 |
ad | -0.404 |
yogurt | -0.394 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
gy | +0.160 |
bank | +0.151 |
drown | +0.121 |
argue | +0.118 |
lves | +0.118 |
Neg logit contributions | |
ey | -0.281 |
ad | -0.276 |
sweet | -0.272 |
honey | -0.271 |
ango | -0.261 |
Token | . |
Token ID | 13 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
gy | +1.510 |
bank | +1.241 |
lves | +1.198 |
drown | +1.148 |
argue | +1.103 |
Neg logit contributions | |
ey | -0.951 |
ad | -0.850 |
yogurt | -0.828 |
honey | -0.810 |
idges | -0.761 |
Token | Bug |
Token ID | 15217 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
gy | +0.002 |
bank | +0.001 |
lves | +0.001 |
drown | +0.001 |
house | +0.001 |
Neg logit contributions | |
ey | -0.003 |
idges | -0.003 |
yogurt | -0.003 |
ement | -0.003 |
ad | -0.003 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
gy | +0.029 |
bank | +0.005 |
lves | -0.033 |
drown | -0.034 |
station | -0.036 |
Neg logit contributions | |
ey | -0.461 |
ad | -0.431 |
honey | -0.424 |
idges | -0.422 |
yogurt | -0.419 |
Token | had |
Token ID | 550 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
gy | +1.532 |
bank | +1.268 |
lves | +1.149 |
drown | +1.109 |
zag | +1.068 |
Neg logit contributions | |
ey | -1.032 |
ad | -0.883 |
yogurt | -0.868 |
idges | -0.847 |
honey | -0.842 |
Token | a |
Token ID | 257 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
gy | +0.111 |
bank | +0.089 |
lves | +0.075 |
zag | +0.072 |
drown | +0.071 |
Neg logit contributions | |
ey | -0.088 |
idges | -0.076 |
ement | -0.073 |
yogurt | -0.071 |
ad | -0.070 |
Token | wanted |
Token ID | 2227 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ey | +0.067 |
idges | +0.057 |
yogurt | +0.057 |
ad | +0.057 |
honey | +0.056 |
Neg logit contributions | |
gy | -0.070 |
bank | -0.055 |
lves | -0.046 |
zag | -0.043 |
station | -0.043 |
Token | to |
Token ID | 284 |
Feature activation | +4.60e-01 |
Pos logit contributions | |
gy | +0.076 |
bank | +0.063 |
drowned | +0.041 |
neither | +0.041 |
fight | +0.040 |
Neg logit contributions | |
ey | -0.145 |
ement | -0.121 |
idges | -0.120 |
mmmm | -0.115 |
importantly | -0.114 |
Token | give |
Token ID | 1577 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
ey | +0.040 |
idges | +0.034 |
yogurt | +0.033 |
ement | +0.033 |
honey | +0.032 |
Neg logit contributions | |
gy | -0.037 |
bank | -0.030 |
lves | -0.023 |
drown | -0.023 |
zag | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
gy | +0.153 |
bank | +0.127 |
lves | +0.110 |
zag | +0.106 |
station | +0.105 |
Neg logit contributions | |
ey | -0.097 |
idges | -0.079 |
yogurt | -0.077 |
ad | -0.076 |
honey | -0.075 |
Token | pig |
Token ID | 12967 |
Feature activation | -5.52e+00 |
Pos logit contributions | |
ey | +0.015 |
idges | +0.013 |
yogurt | +0.013 |
ad | +0.012 |
honey | +0.012 |
Neg logit contributions | |
gy | -0.011 |
bank | -0.008 |
lves | -0.007 |
zag | -0.006 |
station | -0.006 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
gy | +0.527 |
bank | +0.445 |
lves | +0.397 |
drown | +0.372 |
zag | +0.370 |
Neg logit contributions | |
ey | -0.353 |
yogurt | -0.304 |
honey | -0.297 |
ad | -0.297 |
idges | -0.290 |
Token | a |
Token ID | 257 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
gy | +1.470 |
bank | +1.152 |
lves | +1.066 |
zag | +1.009 |
drown | +0.988 |
Neg logit contributions | |
ey | -1.125 |
yogurt | -0.998 |
honey | -0.962 |
ad | -0.940 |
idges | -0.910 |
Token | gift |
Token ID | 6979 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
ey | +0.019 |
idges | +0.016 |
ad | +0.015 |
ement | +0.015 |
yogurt | +0.015 |
Neg logit contributions | |
gy | -0.025 |
bank | -0.020 |
lves | -0.017 |
zag | -0.017 |
station | -0.016 |
Token | too |
Token ID | 1165 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
ey | +0.188 |
idges | +0.158 |
yogurt | +0.154 |
ad | +0.152 |
honey | +0.150 |
Neg logit contributions | |
gy | -0.219 |
bank | -0.177 |
lves | -0.148 |
zag | -0.141 |
station | -0.141 |
Token | . |
Token ID | 13 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
gy | +0.087 |
bank | +0.070 |
lves | +0.056 |
zag | +0.055 |
station | +0.055 |
Neg logit contributions | |
ey | -0.085 |
idges | -0.069 |
ad | -0.066 |
yogurt | -0.065 |
honey | -0.064 |
Token | But |
Token ID | 887 |
Feature activation | -3.28e+00 |
Pos logit contributions | |
ey | +0.010 |
idges | +0.008 |
yogurt | +0.008 |
ad | +0.008 |
honey | +0.007 |
Neg logit contributions | |
gy | -0.014 |
bank | -0.012 |
lves | -0.010 |
zag | -0.010 |
station | -0.010 |
Token | money |
Token ID | 1637 |
Feature activation | -3.11e+00 |
Pos logit contributions | |
gy | +0.193 |
bank | +0.153 |
lves | +0.131 |
drown | +0.122 |
zag | +0.121 |
Neg logit contributions | |
ey | -0.186 |
yogurt | -0.160 |
idges | -0.158 |
honey | -0.156 |
ad | -0.154 |
Token | away |
Token ID | 1497 |
Feature activation | -7.04e+00 |
Pos logit contributions | |
gy | +0.275 |
bank | +0.221 |
lves | +0.189 |
zag | +0.177 |
drown | +0.177 |
Neg logit contributions | |
ey | -0.235 |
yogurt | -0.200 |
idges | -0.200 |
ad | -0.195 |
honey | -0.194 |
Token | in |
Token ID | 287 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
gy | +0.627 |
bank | +0.506 |
lves | +0.419 |
zag | +0.402 |
station | +0.400 |
Neg logit contributions | |
ey | -0.545 |
idges | -0.459 |
yogurt | -0.446 |
ad | -0.442 |
honey | -0.434 |
Token | her |
Token ID | 607 |
Feature activation | +9.58e-01 |
Pos logit contributions | |
ey | +0.078 |
idges | +0.063 |
yogurt | +0.062 |
ad | +0.061 |
honey | +0.060 |
Neg logit contributions | |
gy | -0.125 |
bank | -0.104 |
lves | -0.090 |
zag | -0.086 |
station | -0.085 |
Token | pig |
Token ID | 12967 |
Feature activation | -5.38e+00 |
Pos logit contributions | |
ey | +0.077 |
yogurt | +0.064 |
idges | +0.064 |
ad | +0.063 |
honey | +0.062 |
Neg logit contributions | |
gy | -0.083 |
bank | -0.066 |
lves | -0.056 |
zag | -0.052 |
station | -0.052 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
gy | +0.317 |
lves | +0.262 |
bank | +0.246 |
drown | +0.244 |
argue | +0.221 |
Neg logit contributions | |
ey | -0.438 |
honey | -0.410 |
yogurt | -0.409 |
ad | -0.406 |
sweet | -0.383 |
Token | bank |
Token ID | 17796 |
Feature activation | -9.31e+00 |
Pos logit contributions | |
gy | +0.974 |
lves | +0.659 |
bank | +0.618 |
drown | +0.580 |
neither | +0.545 |
Neg logit contributions | |
ey | -1.556 |
yogurt | -1.442 |
honey | -1.388 |
ad | -1.358 |
pasta | -1.323 |
Token | every |
Token ID | 790 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
gy | +0.834 |
bank | +0.675 |
lves | +0.564 |
zag | +0.539 |
station | +0.535 |
Neg logit contributions | |
ey | -0.711 |
idges | -0.597 |
yogurt | -0.584 |
ad | -0.576 |
honey | -0.569 |
Token | week |
Token ID | 1285 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
gy | +0.233 |
bank | +0.179 |
lves | +0.158 |
zag | +0.140 |
drown | +0.137 |
Neg logit contributions | |
ey | -0.235 |
yogurt | -0.207 |
ad | -0.202 |
honey | -0.200 |
idges | -0.199 |
Token | . |
Token ID | 13 |
Feature activation | +4.70e-02 |
Pos logit contributions | |
gy | +0.235 |
bank | +0.191 |
lves | +0.160 |
zag | +0.153 |
station | +0.152 |
Neg logit contributions | |
ey | -0.196 |
idges | -0.165 |
yogurt | -0.161 |
ad | -0.159 |
honey | -0.157 |
Token | This |
Token ID | 770 |
Feature activation | -3.98e+00 |
Pos logit contributions | |
ey | +0.004 |
idges | +0.003 |
ad | +0.003 |
yogurt | +0.003 |
ement | +0.003 |
Neg logit contributions | |
gy | -0.004 |
bank | -0.003 |
lves | -0.003 |
station | -0.003 |
zag | -0.003 |
Token | them |
Token ID | 606 |
Feature activation | -8.79e-01 |
Pos logit contributions | |
ey | +0.098 |
idges | +0.081 |
yogurt | +0.079 |
ad | +0.078 |
honey | +0.077 |
Neg logit contributions | |
gy | -0.133 |
bank | -0.110 |
lves | -0.094 |
station | -0.088 |
zag | -0.088 |
Token | safe |
Token ID | 3338 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
gy | +0.084 |
bank | +0.068 |
lves | +0.060 |
drown | +0.059 |
zag | +0.058 |
Neg logit contributions | |
ey | -0.059 |
yogurt | -0.051 |
honey | -0.050 |
idges | -0.049 |
sweet | -0.048 |
Token | in |
Token ID | 287 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
gy | +0.113 |
bank | +0.092 |
lves | +0.081 |
zag | +0.077 |
drown | +0.076 |
Neg logit contributions | |
ey | -0.085 |
yogurt | -0.071 |
idges | -0.070 |
honey | -0.070 |
ad | -0.069 |
Token | her |
Token ID | 607 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
ey | +0.082 |
idges | +0.067 |
yogurt | +0.063 |
ement | +0.062 |
ad | +0.062 |
Neg logit contributions | |
gy | -0.139 |
bank | -0.117 |
lves | -0.100 |
station | -0.097 |
drown | -0.096 |
Token | pig |
Token ID | 12967 |
Feature activation | -5.38e+00 |
Pos logit contributions | |
ey | +0.073 |
yogurt | +0.060 |
honey | +0.058 |
idges | +0.057 |
ad | +0.056 |
Neg logit contributions | |
gy | -0.095 |
bank | -0.078 |
lves | -0.067 |
zag | -0.065 |
drown | -0.064 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
gy | +0.219 |
drown | +0.208 |
lves | +0.192 |
crashed | +0.171 |
neither | +0.169 |
Neg logit contributions | |
ey | -0.495 |
ad | -0.471 |
yogurt | -0.461 |
honey | -0.459 |
sweet | -0.455 |
Token | bank |
Token ID | 3331 |
Feature activation | -1.26e+01 |
Pos logit contributions | |
gy | +0.580 |
lves | +0.365 |
drown | +0.331 |
neither | +0.289 |
zag | +0.246 |
Neg logit contributions | |
ey | -1.793 |
yogurt | -1.739 |
honey | -1.697 |
cereal | -1.664 |
pasta | -1.647 |
Token | . |
Token ID | 13 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
gy | +1.131 |
bank | +0.921 |
lves | +0.767 |
zag | +0.731 |
station | +0.728 |
Neg logit contributions | |
ey | -0.967 |
idges | -0.813 |
yogurt | -0.794 |
ad | -0.784 |
honey | -0.775 |
Token | One |
Token ID | 1881 |
Feature activation | -5.47e+00 |
Pos logit contributions | |
ey | +0.010 |
idges | +0.009 |
ad | +0.008 |
yogurt | +0.008 |
ement | +0.008 |
Neg logit contributions | |
gy | -0.008 |
bank | -0.006 |
lves | -0.005 |
station | -0.005 |
zag | -0.005 |
Token | day |
Token ID | 1110 |
Feature activation | -2.54e+00 |
Pos logit contributions | |
gy | +0.448 |
bank | +0.368 |
lves | +0.309 |
drown | +0.293 |
argue | +0.286 |
Neg logit contributions | |
ey | -0.445 |
yogurt | -0.399 |
honey | -0.383 |
ad | -0.378 |
ummy | -0.371 |
Token | , |
Token ID | 11 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
gy | +0.235 |
bank | +0.194 |
lves | +0.166 |
station | +0.157 |
zag | +0.155 |
Neg logit contributions | |
ey | -0.185 |
yogurt | -0.158 |
idges | -0.155 |
honey | -0.152 |
ad | -0.149 |
Token | named |
Token ID | 3706 |
Feature activation | -6.65e+00 |
Pos logit contributions | |
gy | +0.002 |
bank | +0.002 |
lves | +0.002 |
argue | +0.002 |
drown | +0.002 |
Neg logit contributions | |
ey | -0.003 |
yogurt | -0.003 |
ad | -0.003 |
honey | -0.003 |
idges | -0.003 |
Token | Bug |
Token ID | 15217 |
Feature activation | -3.27e+00 |
Pos logit contributions | |
gy | +0.617 |
bank | +0.486 |
lves | +0.419 |
zag | +0.393 |
station | +0.384 |
Neg logit contributions | |
ey | -0.504 |
idges | -0.421 |
ement | -0.412 |
ad | -0.404 |
yogurt | -0.394 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
gy | +0.160 |
bank | +0.151 |
drown | +0.121 |
argue | +0.118 |
lves | +0.118 |
Neg logit contributions | |
ey | -0.281 |
ad | -0.276 |
sweet | -0.272 |
honey | -0.271 |
ango | -0.261 |
Token | . |
Token ID | 13 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
gy | +1.510 |
bank | +1.241 |
lves | +1.198 |
drown | +1.148 |
argue | +1.103 |
Neg logit contributions | |
ey | -0.951 |
ad | -0.850 |
yogurt | -0.828 |
honey | -0.810 |
idges | -0.761 |
Token | Bug |
Token ID | 15217 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
gy | +0.002 |
bank | +0.001 |
lves | +0.001 |
drown | +0.001 |
house | +0.001 |
Neg logit contributions | |
ey | -0.003 |
idges | -0.003 |
yogurt | -0.003 |
ement | -0.003 |
ad | -0.003 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
gy | +0.029 |
bank | +0.005 |
lves | -0.033 |
drown | -0.034 |
station | -0.036 |
Neg logit contributions | |
ey | -0.461 |
ad | -0.431 |
honey | -0.424 |
idges | -0.422 |
yogurt | -0.419 |
Token | had |
Token ID | 550 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
gy | +1.532 |
bank | +1.268 |
lves | +1.149 |
drown | +1.109 |
zag | +1.068 |
Neg logit contributions | |
ey | -1.032 |
ad | -0.883 |
yogurt | -0.868 |
idges | -0.847 |
honey | -0.842 |
Token | a |
Token ID | 257 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
gy | +0.111 |
bank | +0.089 |
lves | +0.075 |
zag | +0.072 |
drown | +0.071 |
Neg logit contributions | |
ey | -0.088 |
idges | -0.076 |
ement | -0.073 |
yogurt | -0.071 |
ad | -0.070 |
Token | lot |
Token ID | 1256 |
Feature activation | -4.96e+00 |
Pos logit contributions | |
ey | +0.013 |
yogurt | +0.010 |
honey | +0.010 |
idges | +0.010 |
sweet | +0.010 |
Neg logit contributions | |
gy | -0.020 |
bank | -0.017 |
lves | -0.015 |
station | -0.014 |
zag | -0.014 |
Token | of |
Token ID | 286 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
gy | +0.401 |
bank | +0.333 |
zag | +0.279 |
house | +0.273 |
lves | +0.259 |
Neg logit contributions | |
ey | -0.362 |
ement | -0.321 |
ango | -0.315 |
Jamie | -0.307 |
idges | -0.307 |
Token | friends |
Token ID | 2460 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
ey | +0.206 |
ement | +0.203 |
idges | +0.194 |
Jamie | +0.186 |
mmmm | +0.184 |
Neg logit contributions | |
gy | -0.209 |
bank | -0.184 |
fight | -0.142 |
argue | -0.135 |
drown | -0.135 |
Token | body |
Token ID | 1767 |
Feature activation | -9.39e-01 |
Pos logit contributions | |
gy | +0.044 |
bank | +0.035 |
lves | +0.034 |
drown | +0.030 |
sank | +0.029 |
Neg logit contributions | |
ey | -0.041 |
honey | -0.035 |
yogurt | -0.034 |
sweet | -0.034 |
ad | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | -4.17e-01 |
Pos logit contributions | |
gy | +0.094 |
bank | +0.078 |
lves | +0.074 |
drown | +0.070 |
argue | +0.069 |
Neg logit contributions | |
ey | -0.055 |
yogurt | -0.046 |
ad | -0.045 |
idges | -0.042 |
honey | -0.042 |
Token | a |
Token ID | 257 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
gy | +0.036 |
bank | +0.028 |
zag | +0.022 |
lves | +0.021 |
neither | +0.021 |
Neg logit contributions | |
ey | -0.035 |
idges | -0.031 |
ement | -0.031 |
ummy | -0.029 |
ad | -0.029 |
Token | w |
Token ID | 266 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
ey | +0.015 |
honey | +0.013 |
yogurt | +0.013 |
sweet | +0.012 |
idges | +0.012 |
Neg logit contributions | |
gy | -0.022 |
bank | -0.018 |
lves | -0.016 |
zag | -0.015 |
drown | -0.014 |
Token | ag |
Token ID | 363 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
ey | +0.095 |
yogurt | +0.076 |
idges | +0.076 |
honey | +0.074 |
ad | +0.071 |
Neg logit contributions | |
gy | -0.155 |
bank | -0.129 |
lves | -0.113 |
zag | -0.112 |
drown | -0.108 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
gy | +0.079 |
bank | +0.073 |
lves | +0.060 |
drown | +0.056 |
station | +0.056 |
Neg logit contributions | |
ey | -0.090 |
ad | -0.083 |
idges | -0.076 |
ement | -0.075 |
yogurt | -0.074 |
Token | tail |
Token ID | 7894 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
gy | +1.249 |
lves | +0.965 |
drown | +0.914 |
bank | +0.893 |
neither | +0.869 |
Neg logit contributions | |
ey | -1.218 |
yogurt | -1.092 |
honey | -1.081 |
sweet | -1.030 |
goodness | -1.027 |
Token | . |
Token ID | 13 |
Feature activation | +1.39e-03 |
Pos logit contributions | |
gy | +0.250 |
bank | +0.203 |
lves | +0.183 |
drown | +0.170 |
argue | +0.168 |
Neg logit contributions | |
ey | -0.171 |
yogurt | -0.143 |
ad | -0.143 |
idges | -0.140 |
honey | -0.138 |
Token | One |
Token ID | 1881 |
Feature activation | -5.50e+00 |
Pos logit contributions | |
ey | +0.000 |
idges | +0.000 |
ement | +0.000 |
illy | +0.000 |
odic | +0.000 |
Neg logit contributions | |
gy | -0.000 |
bank | -0.000 |
station | -0.000 |
house | -0.000 |
zag | -0.000 |
Token | day |
Token ID | 1110 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
gy | +0.438 |
bank | +0.356 |
lves | +0.295 |
drown | +0.269 |
argue | +0.269 |
Neg logit contributions | |
ey | -0.468 |
yogurt | -0.408 |
honey | -0.398 |
ad | -0.398 |
idges | -0.394 |
Token | , |
Token ID | 11 |
Feature activation | +2.66e-01 |
Pos logit contributions | |
gy | +0.207 |
bank | +0.158 |
lves | +0.125 |
zag | +0.123 |
neither | +0.120 |
Neg logit contributions | |
ey | -0.231 |
idges | -0.198 |
ad | -0.191 |
ement | -0.189 |
yogurt | -0.187 |
Token | and |
Token ID | 290 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
ey | +0.021 |
idges | +0.018 |
yogurt | +0.017 |
ad | +0.017 |
ement | +0.016 |
Neg logit contributions | |
gy | -0.017 |
bank | -0.013 |
lves | -0.010 |
neither | -0.010 |
argue | -0.009 |
Token | said |
Token ID | 531 |
Feature activation | -2.78e-01 |
Pos logit contributions | |
gy | +0.033 |
bank | +0.027 |
lves | +0.023 |
zag | +0.022 |
station | +0.022 |
Neg logit contributions | |
ey | -0.025 |
idges | -0.021 |
yogurt | -0.021 |
ad | -0.021 |
honey | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
gy | +0.019 |
bank | +0.014 |
neither | +0.011 |
zag | +0.011 |
argue | +0.010 |
Neg logit contributions | |
ey | -0.028 |
idges | -0.023 |
ad | -0.022 |
ement | -0.022 |
ango | -0.022 |
Token | " |
Token ID | 366 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
ey | +0.016 |
idges | +0.013 |
ad | +0.013 |
ement | +0.012 |
yogurt | +0.012 |
Neg logit contributions | |
gy | -0.019 |
bank | -0.015 |
station | -0.012 |
zag | -0.012 |
lves | -0.012 |
Token | Dog |
Token ID | 32942 |
Feature activation | -8.41e+00 |
Pos logit contributions | |
ey | +0.142 |
idges | +0.115 |
honey | +0.112 |
yogurt | +0.110 |
ement | +0.109 |
Neg logit contributions | |
gy | -0.201 |
bank | -0.162 |
lves | -0.146 |
zag | -0.135 |
neither | -0.133 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
gy | +0.387 |
bank | +0.251 |
lves | +0.157 |
zag | +0.125 |
station | +0.123 |
Neg logit contributions | |
ey | -0.998 |
idges | -0.898 |
yogurt | -0.888 |
ad | -0.887 |
honey | -0.877 |
Token | , |
Token ID | 11 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
gy | +1.508 |
bank | +1.234 |
lves | +1.052 |
zag | +1.001 |
station | +0.994 |
Neg logit contributions | |
ey | -1.156 |
idges | -0.963 |
yogurt | -0.944 |
ad | -0.930 |
honey | -0.919 |
Token | stop |
Token ID | 2245 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
ey | +0.016 |
ement | +0.013 |
idges | +0.013 |
ad | +0.012 |
importantly | +0.012 |
Neg logit contributions | |
gy | -0.020 |
bank | -0.017 |
station | -0.014 |
lves | -0.014 |
drown | -0.014 |
Token | ! |
Token ID | 0 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
gy | +0.020 |
bank | +0.016 |
lves | +0.014 |
drown | +0.013 |
station | +0.013 |
Neg logit contributions | |
ey | -0.014 |
idges | -0.012 |
ement | -0.011 |
yogurt | -0.011 |
ad | -0.011 |
Token | You |
Token ID | 921 |
Feature activation | +7.36e-01 |
Pos logit contributions | |
ey | +0.135 |
idges | +0.110 |
yogurt | +0.106 |
ement | +0.104 |
honey | +0.103 |
Neg logit contributions | |
gy | -0.214 |
bank | -0.173 |
lves | -0.149 |
drown | -0.147 |
zag | -0.145 |
Token | 're |
Token ID | 821 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
ey | +0.048 |
idges | +0.042 |
honey | +0.042 |
yogurt | +0.041 |
ad | +0.041 |
Neg logit contributions | |
gy | -0.071 |
bank | -0.058 |
lves | -0.050 |
zag | -0.048 |
station | -0.047 |
Token | ly |
Token ID | 306 |
Feature activation | +3.93e+00 |
Pos logit contributions | |
gy | +0.048 |
bank | +0.041 |
lves | +0.036 |
drown | +0.036 |
argue | +0.036 |
Neg logit contributions | |
ey | -0.022 |
idges | -0.018 |
ad | -0.018 |
honey | -0.018 |
yogurt | -0.017 |
Token | said |
Token ID | 531 |
Feature activation | -4.08e-01 |
Pos logit contributions | |
ey | +0.344 |
idges | +0.297 |
yogurt | +0.292 |
ad | +0.289 |
honey | +0.286 |
Neg logit contributions | |
gy | -0.306 |
bank | -0.239 |
lves | -0.194 |
zag | -0.182 |
station | -0.181 |
Token | , |
Token ID | 11 |
Feature activation | +3.23e-01 |
Pos logit contributions | |
gy | +0.031 |
bank | +0.024 |
zag | +0.019 |
neither | +0.019 |
drown | +0.018 |
Neg logit contributions | |
ey | -0.038 |
idges | -0.032 |
ement | -0.030 |
ad | -0.030 |
ango | -0.029 |
Token | " |
Token ID | 366 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
ey | +0.021 |
idges | +0.017 |
yogurt | +0.016 |
ad | +0.016 |
ement | +0.016 |
Neg logit contributions | |
gy | -0.033 |
bank | -0.027 |
lves | -0.023 |
station | -0.023 |
zag | -0.023 |
Token | Bug |
Token ID | 25624 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
ey | +0.150 |
idges | +0.122 |
yogurt | +0.119 |
honey | +0.118 |
ement | +0.117 |
Neg logit contributions | |
gy | -0.198 |
bank | -0.161 |
lves | -0.142 |
zag | -0.132 |
station | -0.131 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
gy | +0.208 |
bank | +0.168 |
lves | +0.140 |
zag | +0.132 |
station | +0.131 |
Neg logit contributions | |
ey | -0.195 |
idges | -0.166 |
yogurt | -0.163 |
ad | -0.162 |
honey | -0.160 |
Token | , |
Token ID | 11 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
gy | +1.440 |
bank | +1.159 |
lves | +0.945 |
zag | +0.914 |
station | +0.913 |
Neg logit contributions | |
ey | -1.267 |
idges | -1.063 |
yogurt | -1.024 |
ad | -1.005 |
ement | -1.003 |
Token | today |
Token ID | 1909 |
Feature activation | -4.17e+00 |
Pos logit contributions | |
ey | +0.024 |
ement | +0.020 |
idges | +0.020 |
ad | +0.019 |
yogurt | +0.019 |
Neg logit contributions | |
gy | -0.028 |
bank | -0.023 |
lves | -0.019 |
station | -0.019 |
drown | -0.019 |
Token | was |
Token ID | 373 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
gy | +0.370 |
bank | +0.303 |
lves | +0.246 |
station | +0.244 |
drown | +0.241 |
Neg logit contributions | |
ey | -0.324 |
idges | -0.268 |
ad | -0.260 |
yogurt | -0.257 |
ement | -0.253 |
Token | so |
Token ID | 523 |
Feature activation | +3.98e-01 |
Pos logit contributions | |
ey | +0.089 |
idges | +0.074 |
ement | +0.072 |
ad | +0.071 |
yogurt | +0.069 |
Neg logit contributions | |
gy | -0.105 |
bank | -0.085 |
lves | -0.071 |
drown | -0.069 |
station | -0.068 |
Token | much |
Token ID | 881 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
ey | +0.036 |
idges | +0.032 |
ement | +0.031 |
ad | +0.030 |
ango | +0.030 |
Neg logit contributions | |
gy | -0.028 |
bank | -0.022 |
drown | -0.019 |
drowned | -0.017 |
lves | -0.017 |
Token | Tim |
Token ID | 5045 |
Feature activation | -5.14e-01 |
Pos logit contributions | |
gy | +0.001 |
bank | +0.001 |
house | +0.001 |
station | +0.001 |
zag | +0.001 |
Neg logit contributions | |
ey | -0.001 |
idges | -0.001 |
ement | -0.001 |
honey | -0.001 |
ad | -0.001 |
Token | my |
Token ID | 1820 |
Feature activation | +2.54e+00 |
Pos logit contributions | |
gy | +0.050 |
bank | +0.039 |
lves | +0.033 |
station | +0.032 |
zag | +0.032 |
Neg logit contributions | |
ey | -0.037 |
idges | -0.032 |
ement | -0.030 |
honey | -0.030 |
yogurt | -0.030 |
Token | had |
Token ID | 550 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
ey | +0.152 |
yogurt | +0.126 |
ad | +0.124 |
idges | +0.115 |
sweet | +0.114 |
Neg logit contributions | |
gy | -0.249 |
bank | -0.217 |
lves | -0.194 |
drown | -0.186 |
argue | -0.182 |
Token | a |
Token ID | 257 |
Feature activation | +3.24e-01 |
Pos logit contributions | |
gy | +0.107 |
bank | +0.088 |
lves | +0.075 |
zag | +0.071 |
station | +0.071 |
Neg logit contributions | |
ey | -0.079 |
idges | -0.066 |
ement | -0.062 |
yogurt | -0.062 |
ad | -0.062 |
Token | pig |
Token ID | 12967 |
Feature activation | -5.30e+00 |
Pos logit contributions | |
ey | +0.022 |
yogurt | +0.019 |
honey | +0.018 |
sweet | +0.018 |
pasta | +0.017 |
Neg logit contributions | |
gy | -0.031 |
bank | -0.025 |
lves | -0.022 |
drown | -0.021 |
zag | -0.021 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
gy | +0.273 |
drown | +0.255 |
bank | +0.254 |
lves | +0.235 |
argue | +0.231 |
Neg logit contributions | |
ad | -0.431 |
ey | -0.428 |
yogurt | -0.410 |
sweet | -0.408 |
ummy | -0.390 |
Token | bank |
Token ID | 3331 |
Feature activation | -1.26e+01 |
Pos logit contributions | |
gy | +1.030 |
lves | +0.788 |
drown | +0.718 |
bank | +0.713 |
neither | +0.651 |
Neg logit contributions | |
ey | -1.386 |
yogurt | -1.318 |
honey | -1.291 |
ad | -1.234 |
sweet | -1.232 |
Token | that |
Token ID | 326 |
Feature activation | -8.68e-01 |
Pos logit contributions | |
gy | +1.188 |
bank | +0.982 |
lves | +0.819 |
zag | +0.788 |
station | +0.787 |
Neg logit contributions | |
ey | -0.916 |
idges | -0.758 |
yogurt | -0.727 |
ad | -0.719 |
honey | -0.718 |
Token | was |
Token ID | 373 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
gy | +0.090 |
bank | +0.076 |
lves | +0.067 |
zag | +0.064 |
drown | +0.064 |
Neg logit contributions | |
ey | -0.052 |
yogurt | -0.042 |
idges | -0.041 |
honey | -0.041 |
ad | -0.040 |
Token | full |
Token ID | 1336 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
ey | +0.080 |
yogurt | +0.069 |
honey | +0.068 |
idges | +0.067 |
ad | +0.067 |
Neg logit contributions | |
gy | -0.096 |
bank | -0.077 |
lves | -0.067 |
zag | -0.064 |
drown | -0.063 |
Token | of |
Token ID | 286 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
ey | +0.127 |
idges | +0.101 |
ement | +0.094 |
Jamie | +0.092 |
ad | +0.092 |
Neg logit contributions | |
gy | -0.162 |
bank | -0.142 |
station | -0.119 |
house | -0.109 |
lves | -0.107 |
Token | day |
Token ID | 1110 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
gy | +0.482 |
bank | +0.425 |
lves | +0.368 |
argue | +0.354 |
drown | +0.348 |
Neg logit contributions | |
ey | -0.375 |
sweet | -0.350 |
honey | -0.349 |
yogurt | -0.344 |
ummy | -0.335 |
Token | , |
Token ID | 11 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
gy | +0.212 |
bank | +0.163 |
lves | +0.130 |
zag | +0.128 |
station | +0.124 |
Neg logit contributions | |
ey | -0.231 |
idges | -0.197 |
ad | -0.191 |
ement | -0.188 |
yogurt | -0.187 |
Token | a |
Token ID | 257 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
ey | +0.016 |
idges | +0.014 |
ad | +0.013 |
ement | +0.013 |
yogurt | +0.013 |
Neg logit contributions | |
gy | -0.021 |
bank | -0.017 |
lves | -0.014 |
station | -0.014 |
zag | -0.014 |
Token | little |
Token ID | 1310 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
ey | +0.014 |
honey | +0.012 |
yogurt | +0.012 |
idges | +0.011 |
sweet | +0.011 |
Neg logit contributions | |
gy | -0.027 |
bank | -0.023 |
lves | -0.020 |
argue | -0.019 |
zag | -0.019 |
Token | pig |
Token ID | 12967 |
Feature activation | -5.36e+00 |
Pos logit contributions | |
ey | +0.057 |
honey | +0.056 |
sweet | +0.054 |
yogurt | +0.050 |
ummy | +0.049 |
Neg logit contributions | |
gy | -0.102 |
bank | -0.087 |
argue | -0.078 |
zag | -0.077 |
drowned | -0.077 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
gy | +0.406 |
bank | +0.400 |
lves | +0.374 |
drown | +0.373 |
argue | +0.363 |
Neg logit contributions | |
ad | -0.307 |
ey | -0.303 |
yogurt | -0.253 |
illy | -0.244 |
idges | -0.242 |
Token | named |
Token ID | 3706 |
Feature activation | -6.64e+00 |
Pos logit contributions | |
gy | +1.430 |
drown | +1.201 |
lves | +1.186 |
argue | +1.179 |
zag | +1.144 |
Neg logit contributions | |
ad | -0.774 |
ey | -0.704 |
honey | -0.689 |
sweet | -0.677 |
yogurt | -0.676 |
Token | Pe |
Token ID | 2631 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
gy | +0.627 |
bank | +0.518 |
lves | +0.441 |
zag | +0.424 |
station | +0.416 |
Neg logit contributions | |
ey | -0.471 |
idges | -0.394 |
yogurt | -0.391 |
honey | -0.383 |
ad | -0.383 |
Token | ppa |
Token ID | 44989 |
Feature activation | -3.86e+00 |
Pos logit contributions | |
gy | +0.251 |
bank | +0.209 |
drown | +0.174 |
lves | +0.170 |
argue | +0.168 |
Neg logit contributions | |
ey | -0.205 |
idges | -0.186 |
ad | -0.177 |
honey | -0.176 |
yogurt | -0.172 |
Token | went |
Token ID | 1816 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
gy | +0.333 |
bank | +0.295 |
lves | +0.267 |
drown | +0.260 |
argue | +0.254 |
Neg logit contributions | |
ey | -0.237 |
ad | -0.217 |
yogurt | -0.208 |
honey | -0.200 |
idges | -0.199 |
Token | for |
Token ID | 329 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
gy | +0.117 |
bank | +0.097 |
lves | +0.094 |
drowned | +0.087 |
neither | +0.086 |
Neg logit contributions | |
ey | -0.066 |
honey | -0.060 |
yogurt | -0.059 |
idges | -0.054 |
ad | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
gy | +0.087 |
bank | +0.070 |
lves | +0.056 |
zag | +0.055 |
station | +0.055 |
Neg logit contributions | |
ey | -0.085 |
idges | -0.069 |
ad | -0.066 |
yogurt | -0.065 |
honey | -0.064 |
Token | But |
Token ID | 887 |
Feature activation | -3.28e+00 |
Pos logit contributions | |
ey | +0.010 |
idges | +0.008 |
yogurt | +0.008 |
ad | +0.008 |
honey | +0.007 |
Neg logit contributions | |
gy | -0.014 |
bank | -0.012 |
lves | -0.010 |
zag | -0.010 |
station | -0.010 |
Token | when |
Token ID | 618 |
Feature activation | -9.40e-01 |
Pos logit contributions | |
gy | +0.337 |
bank | +0.280 |
lves | +0.257 |
zag | +0.244 |
drown | +0.241 |
Neg logit contributions | |
ey | -0.183 |
yogurt | -0.164 |
honey | -0.158 |
ad | -0.153 |
goodness | -0.144 |
Token | the |
Token ID | 262 |
Feature activation | +3.01e-02 |
Pos logit contributions | |
gy | +0.106 |
bank | +0.089 |
lves | +0.082 |
zag | +0.077 |
station | +0.076 |
Neg logit contributions | |
ey | -0.047 |
yogurt | -0.041 |
honey | -0.038 |
ad | -0.038 |
idges | -0.037 |
Token | pig |
Token ID | 12967 |
Feature activation | -5.51e+00 |
Pos logit contributions | |
ey | +0.003 |
idges | +0.003 |
ad | +0.002 |
yogurt | +0.002 |
ement | +0.002 |
Neg logit contributions | |
gy | -0.002 |
bank | -0.002 |
lves | -0.001 |
station | -0.001 |
zag | -0.001 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
gy | +0.354 |
bank | +0.293 |
lves | +0.263 |
drown | +0.254 |
argue | +0.237 |
Neg logit contributions | |
ey | -0.465 |
ad | -0.433 |
yogurt | -0.408 |
idges | -0.401 |
honey | -0.400 |
Token | looked |
Token ID | 3114 |
Feature activation | +4.26e-01 |
Pos logit contributions | |
gy | +1.348 |
bank | +1.037 |
lves | +0.988 |
drown | +0.975 |
zag | +0.934 |
Neg logit contributions | |
ey | -1.069 |
ad | -1.003 |
yogurt | -0.973 |
honey | -0.957 |
idges | -0.919 |
Token | around |
Token ID | 1088 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
ey | +0.039 |
idges | +0.033 |
ad | +0.033 |
yogurt | +0.032 |
mmmm | +0.031 |
Neg logit contributions | |
gy | -0.031 |
bank | -0.025 |
lves | -0.020 |
station | -0.019 |
neither | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
gy | +0.191 |
bank | +0.152 |
station | +0.119 |
fight | +0.111 |
argue | +0.111 |
Neg logit contributions | |
ey | -0.220 |
idges | -0.193 |
yogurt | -0.180 |
ad | -0.178 |
mmmm | -0.177 |
Token | he |
Token ID | 339 |
Feature activation | +8.01e-01 |
Pos logit contributions | |
ey | +0.019 |
idges | +0.016 |
yogurt | +0.015 |
ad | +0.015 |
honey | +0.015 |
Neg logit contributions | |
gy | -0.034 |
bank | -0.028 |
lves | -0.025 |
zag | -0.024 |
station | -0.024 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
ey | +0.066 |
idges | +0.057 |
yogurt | +0.057 |
ad | +0.056 |
honey | +0.055 |
Neg logit contributions | |
gy | -0.065 |
bank | -0.051 |
lves | -0.043 |
zag | -0.041 |
drown | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
gy | +0.001 |
bank | +0.001 |
station | +0.001 |
zag | +0.001 |
neither | +0.001 |
Neg logit contributions | |
ey | -0.001 |
idges | -0.001 |
honey | -0.001 |
ad | -0.001 |
ement | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
gy | +0.011 |
bank | +0.008 |
zag | +0.007 |
station | +0.007 |
lves | +0.007 |
Neg logit contributions | |
ey | -0.011 |
honey | -0.009 |
idges | -0.009 |
yogurt | -0.009 |
ement | -0.009 |
Token | " |
Token ID | 1 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
gy | +0.010 |
bank | +0.008 |
lves | +0.007 |
station | +0.007 |
zag | +0.007 |
Neg logit contributions | |
ey | -0.008 |
idges | -0.007 |
honey | -0.007 |
ement | -0.007 |
yogurt | -0.007 |
Token | The |
Token ID | 464 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
gy | +0.135 |
bank | +0.114 |
lves | +0.099 |
zag | +0.095 |
station | +0.095 |
Neg logit contributions | |
ey | -0.096 |
honey | -0.079 |
idges | -0.077 |
ement | -0.076 |
yogurt | -0.073 |
Token | pig |
Token ID | 12967 |
Feature activation | -5.27e+00 |
Pos logit contributions | |
gy | +0.035 |
bank | +0.031 |
station | +0.025 |
lves | +0.023 |
fight | +0.023 |
Neg logit contributions | |
ey | -0.035 |
idges | -0.031 |
ad | -0.029 |
ement | -0.028 |
importantly | -0.027 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
gy | +0.472 |
bank | +0.387 |
lves | +0.337 |
drown | +0.314 |
zag | +0.311 |
Neg logit contributions | |
ey | -0.385 |
idges | -0.324 |
yogurt | -0.320 |
ad | -0.320 |
honey | -0.313 |
Token | bank |
Token ID | 3331 |
Feature activation | -1.26e+01 |
Pos logit contributions | |
gy | +1.415 |
bank | +1.121 |
lves | +0.998 |
drown | +0.946 |
zag | +0.945 |
Neg logit contributions | |
ey | -1.172 |
yogurt | -1.004 |
ad | -0.996 |
idges | -0.995 |
honey | -0.980 |
Token | !" |
Token ID | 2474 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
gy | +1.124 |
bank | +0.942 |
lves | +0.748 |
station | +0.746 |
zag | +0.735 |
Neg logit contributions | |
ey | -0.961 |
idges | -0.796 |
yogurt | -0.771 |
ad | -0.759 |
honey | -0.758 |
Token | said |
Token ID | 531 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
ey | +0.017 |
yogurt | +0.014 |
idges | +0.013 |
ad | +0.013 |
honey | +0.013 |
Neg logit contributions | |
gy | -0.035 |
bank | -0.029 |
lves | -0.026 |
zag | -0.025 |
drown | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
gy | +0.058 |
bank | +0.051 |
lves | +0.047 |
zag | +0.044 |
drown | +0.043 |
Neg logit contributions | |
ey | -0.024 |
yogurt | -0.020 |
ad | -0.019 |
honey | -0.018 |
idges | -0.018 |
Token | little |
Token ID | 1310 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
gy | +0.002 |
bank | +0.002 |
station | +0.002 |
lves | +0.002 |
argue | +0.001 |
Neg logit contributions | |
ey | -0.002 |
idges | -0.002 |
ad | -0.002 |
ement | -0.002 |
yogurt | -0.002 |
Token | day |
Token ID | 1110 |
Feature activation | -2.71e+00 |
Pos logit contributions | |
gy | +0.080 |
bank | +0.064 |
lves | +0.054 |
drown | +0.049 |
zag | +0.048 |
Neg logit contributions | |
ey | -0.084 |
yogurt | -0.072 |
ad | -0.072 |
honey | -0.072 |
idges | -0.071 |
Token | onwards |
Token ID | 31719 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
gy | +0.235 |
bank | +0.194 |
lves | +0.164 |
station | +0.150 |
argue | +0.149 |
Neg logit contributions | |
ey | -0.207 |
idges | -0.183 |
ad | -0.183 |
yogurt | -0.178 |
honey | -0.177 |
Token | , |
Token ID | 11 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
gy | +0.101 |
bank | +0.083 |
lves | +0.070 |
zag | +0.067 |
drown | +0.067 |
Neg logit contributions | |
ey | -0.082 |
idges | -0.070 |
yogurt | -0.069 |
honey | -0.068 |
ad | -0.068 |
Token | the |
Token ID | 262 |
Feature activation | +3.77e-03 |
Pos logit contributions | |
ey | +0.012 |
idges | +0.010 |
yogurt | +0.010 |
honey | +0.010 |
ad | +0.010 |
Neg logit contributions | |
gy | -0.018 |
bank | -0.015 |
lves | -0.013 |
zag | -0.013 |
drown | -0.013 |
Token | pig |
Token ID | 12967 |
Feature activation | -5.38e+00 |
Pos logit contributions | |
ey | +0.000 |
idges | +0.000 |
yogurt | +0.000 |
ad | +0.000 |
ement | +0.000 |
Neg logit contributions | |
gy | -0.000 |
bank | -0.000 |
lves | -0.000 |
station | -0.000 |
zag | -0.000 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
gy | +0.470 |
bank | +0.389 |
lves | +0.340 |
drown | +0.321 |
argue | +0.316 |
Neg logit contributions | |
ey | -0.392 |
ad | -0.334 |
idges | -0.331 |
yogurt | -0.324 |
honey | -0.320 |
Token | was |
Token ID | 373 |
Feature activation | +9.87e-01 |
Pos logit contributions | |
gy | +1.458 |
bank | +1.187 |
lves | +1.037 |
drown | +0.987 |
zag | +0.974 |
Neg logit contributions | |
ey | -1.131 |
idges | -0.957 |
ad | -0.951 |
yogurt | -0.947 |
honey | -0.934 |
Token | thankful |
Token ID | 25535 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
ey | +0.080 |
idges | +0.069 |
yogurt | +0.068 |
ad | +0.068 |
honey | +0.067 |
Neg logit contributions | |
gy | -0.082 |
bank | -0.066 |
lves | -0.055 |
zag | -0.051 |
station | -0.051 |
Token | and |
Token ID | 290 |
Feature activation | -4.76e-01 |
Pos logit contributions | |
ey | +0.147 |
idges | +0.127 |
ement | +0.121 |
yogurt | +0.120 |
ad | +0.118 |
Neg logit contributions | |
gy | -0.109 |
bank | -0.083 |
neither | -0.065 |
zag | -0.062 |
station | -0.062 |
Token | more |
Token ID | 517 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
gy | +0.043 |
bank | +0.034 |
lves | +0.029 |
station | +0.027 |
zag | +0.027 |
Neg logit contributions | |
ey | -0.037 |
idges | -0.031 |
yogurt | -0.030 |
ement | -0.030 |
ad | -0.029 |
Token | careful |
Token ID | 8161 |
Feature activation | +5.17e-01 |
Pos logit contributions | |
ey | +0.120 |
idges | +0.091 |
ad | +0.090 |
mmmm | +0.089 |
yogurt | +0.088 |
Neg logit contributions | |
gy | -0.141 |
bank | -0.116 |
lves | -0.104 |
house | -0.097 |
station | -0.097 |
Token | the |
Token ID | 262 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
gy | +0.039 |
bank | +0.031 |
lves | +0.027 |
zag | +0.025 |
drown | +0.025 |
Neg logit contributions | |
ey | -0.037 |
yogurt | -0.032 |
idges | -0.031 |
honey | -0.031 |
ad | -0.031 |
Token | raced |
Token ID | 32897 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
gy | +0.014 |
bank | +0.011 |
station | +0.008 |
lves | +0.008 |
zag | +0.008 |
Neg logit contributions | |
ey | -0.018 |
idges | -0.016 |
ad | -0.016 |
ement | -0.015 |
yogurt | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
gy | +0.192 |
bank | +0.151 |
lves | +0.128 |
zag | +0.119 |
drown | +0.118 |
Neg logit contributions | |
ey | -0.189 |
idges | -0.162 |
yogurt | -0.161 |
ad | -0.160 |
honey | -0.157 |
Token | s |
Token ID | 264 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
gy | +0.016 |
bank | +0.012 |
lves | +0.010 |
zag | +0.009 |
station | +0.009 |
Neg logit contributions | |
ey | -0.020 |
idges | -0.018 |
yogurt | -0.018 |
ad | -0.017 |
honey | -0.017 |
Token | og |
Token ID | 519 |
Feature activation | -3.56e+00 |
Pos logit contributions | |
gy | +0.115 |
bank | +0.095 |
station | +0.082 |
zag | +0.081 |
house | +0.079 |
Neg logit contributions | |
ey | -0.093 |
honey | -0.079 |
sweet | -0.074 |
idges | -0.074 |
yogurt | -0.073 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
gy | +0.181 |
bank | +0.140 |
lves | +0.129 |
drown | +0.116 |
argue | +0.101 |
Neg logit contributions | |
ey | -0.333 |
ad | -0.326 |
yogurt | -0.311 |
idges | -0.309 |
honey | -0.302 |
Token | race |
Token ID | 3234 |
Feature activation | -5.49e+00 |
Pos logit contributions | |
gy | +0.856 |
lves | +0.540 |
drown | +0.463 |
neither | +0.456 |
bank | +0.443 |
Neg logit contributions | |
ey | -1.581 |
yogurt | -1.489 |
honey | -1.458 |
pasta | -1.401 |
ad | -1.397 |
Token | car |
Token ID | 1097 |
Feature activation | -6.98e+00 |
Pos logit contributions | |
gy | +0.524 |
bank | +0.422 |
lves | +0.393 |
drown | +0.363 |
zag | +0.348 |
Neg logit contributions | |
ey | -0.353 |
ad | -0.313 |
yogurt | -0.303 |
idges | -0.298 |
ement | -0.288 |
Token | around |
Token ID | 1088 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
gy | +0.687 |
bank | +0.566 |
lves | +0.554 |
drown | +0.521 |
drowned | +0.492 |
Neg logit contributions | |
ey | -0.361 |
ad | -0.347 |
yogurt | -0.319 |
idges | -0.311 |
ement | -0.303 |
Token | the |
Token ID | 262 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
gy | +0.232 |
bank | +0.189 |
lves | +0.155 |
station | +0.152 |
zag | +0.149 |
Neg logit contributions | |
ey | -0.188 |
idges | -0.157 |
yogurt | -0.151 |
honey | -0.148 |
ad | -0.146 |
Token | bowl |
Token ID | 9396 |
Feature activation | +3.59e+00 |
Pos logit contributions | |
gy | +0.022 |
bank | +0.017 |
lves | +0.014 |
zag | +0.014 |
station | +0.013 |
Neg logit contributions | |
ey | -0.021 |
idges | -0.018 |
yogurt | -0.018 |
ad | -0.018 |
honey | -0.017 |
Token | a |
Token ID | 257 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
ey | +0.077 |
idges | +0.065 |
ement | +0.061 |
ad | +0.061 |
yogurt | +0.059 |
Neg logit contributions | |
gy | -0.093 |
bank | -0.074 |
lves | -0.063 |
station | -0.060 |
zag | -0.058 |
Token | modest |
Token ID | 12949 |
Feature activation | +5.81e-02 |
Pos logit contributions | |
ey | +0.017 |
yogurt | +0.014 |
idges | +0.013 |
honey | +0.013 |
ad | +0.013 |
Neg logit contributions | |
gy | -0.028 |
bank | -0.023 |
lves | -0.020 |
zag | -0.019 |
drown | -0.019 |
Token | dog |
Token ID | 3290 |
Feature activation | -4.96e+00 |
Pos logit contributions | |
ey | +0.004 |
yogurt | +0.004 |
honey | +0.004 |
idges | +0.004 |
ad | +0.004 |
Neg logit contributions | |
gy | -0.005 |
bank | -0.004 |
lves | -0.003 |
drown | -0.003 |
zag | -0.003 |
Token | called |
Token ID | 1444 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
gy | +0.332 |
bank | +0.278 |
lves | +0.228 |
drown | +0.224 |
argue | +0.218 |
Neg logit contributions | |
ey | -0.415 |
ad | -0.399 |
yogurt | -0.397 |
honey | -0.378 |
idges | -0.377 |
Token | Dog |
Token ID | 8532 |
Feature activation | -5.28e+00 |
Pos logit contributions | |
gy | +0.160 |
bank | +0.130 |
lves | +0.110 |
zag | +0.104 |
station | +0.103 |
Neg logit contributions | |
ey | -0.131 |
idges | -0.110 |
yogurt | -0.108 |
ad | -0.107 |
honey | -0.106 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
gy | +0.204 |
bank | +0.198 |
drown | +0.151 |
lves | +0.139 |
argue | +0.131 |
Neg logit contributions | |
ad | -0.520 |
ey | -0.495 |
ango | -0.491 |
sweet | -0.488 |
honey | -0.485 |
Token | . |
Token ID | 13 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
gy | +1.529 |
bank | +1.254 |
lves | +1.198 |
drown | +1.139 |
argue | +1.092 |
Neg logit contributions | |
ey | -0.917 |
ad | -0.830 |
yogurt | -0.826 |
honey | -0.796 |
ummy | -0.768 |
Token | One |
Token ID | 1881 |
Feature activation | -5.49e+00 |
Pos logit contributions | |
ey | +0.001 |
ad | +0.001 |
honey | +0.001 |
idges | +0.001 |
sweet | +0.001 |
Neg logit contributions | |
gy | -0.001 |
bank | -0.001 |
lves | -0.001 |
drown | -0.001 |
zag | -0.001 |
Token | day |
Token ID | 1110 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
gy | +0.444 |
bank | +0.365 |
lves | +0.303 |
drown | +0.280 |
argue | +0.280 |
Neg logit contributions | |
ey | -0.451 |
yogurt | -0.394 |
ad | -0.387 |
honey | -0.385 |
idges | -0.381 |
Token | , |
Token ID | 11 |
Feature activation | +3.01e-01 |
Pos logit contributions | |
gy | +0.239 |
bank | +0.205 |
lves | +0.176 |
station | +0.165 |
drown | +0.165 |
Neg logit contributions | |
ey | -0.169 |
idges | -0.146 |
honey | -0.145 |
yogurt | -0.144 |
ad | -0.139 |
Token | Dog |
Token ID | 8532 |
Feature activation | -5.25e+00 |
Pos logit contributions | |
ey | +0.021 |
honey | +0.021 |
sweet | +0.020 |
idges | +0.020 |
importantly | +0.019 |
Neg logit contributions | |
gy | -0.024 |
bank | -0.021 |
lves | -0.017 |
drown | -0.017 |
argue | -0.016 |
Token | cookies |
Token ID | 14746 |
Feature activation | +4.69e+00 |
Pos logit contributions | |
gy | +0.548 |
bank | +0.447 |
lves | +0.380 |
zag | +0.360 |
station | +0.358 |
Neg logit contributions | |
ey | -0.437 |
idges | -0.365 |
yogurt | -0.358 |
ad | -0.353 |
honey | -0.349 |
Token | got |
Token ID | 1392 |
Feature activation | -5.25e+00 |
Pos logit contributions | |
ey | +0.335 |
idges | +0.289 |
ad | +0.285 |
yogurt | +0.284 |
honey | +0.284 |
Neg logit contributions | |
gy | -0.430 |
bank | -0.327 |
lves | -0.305 |
zag | -0.271 |
argue | -0.261 |
Token | all |
Token ID | 477 |
Feature activation | -3.91e+00 |
Pos logit contributions | |
gy | +0.455 |
bank | +0.370 |
lves | +0.315 |
zag | +0.308 |
station | +0.299 |
Neg logit contributions | |
ey | -0.387 |
honey | -0.347 |
idges | -0.343 |
yogurt | -0.339 |
ad | -0.332 |
Token | s |
Token ID | 264 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
gy | +0.340 |
bank | +0.279 |
lves | +0.241 |
zag | +0.220 |
station | +0.217 |
Neg logit contributions | |
ey | -0.293 |
yogurt | -0.254 |
honey | -0.252 |
idges | -0.251 |
ad | -0.251 |
Token | og |
Token ID | 519 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
gy | +0.114 |
bank | +0.085 |
lves | +0.074 |
zag | +0.069 |
drowned | +0.067 |
Neg logit contributions | |
ey | -0.112 |
honey | -0.098 |
yogurt | -0.095 |
importantly | -0.090 |
sweet | -0.090 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
gy | +0.015 |
bank | -0.009 |
lves | -0.035 |
drown | -0.052 |
argue | -0.059 |
Neg logit contributions | |
ey | -0.508 |
ad | -0.492 |
idges | -0.484 |
honey | -0.477 |
yogurt | -0.472 |
Token | and |
Token ID | 290 |
Feature activation | -5.00e-01 |
Pos logit contributions | |
gy | +1.479 |
bank | +1.165 |
lves | +1.153 |
drown | +1.015 |
station | +1.010 |
Neg logit contributions | |
ey | -0.996 |
honey | -0.895 |
yogurt | -0.873 |
sweet | -0.842 |
ad | -0.832 |
Token | Jerry |
Token ID | 13075 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
gy | +0.044 |
bank | +0.035 |
lves | +0.030 |
zag | +0.029 |
station | +0.028 |
Neg logit contributions | |
ey | -0.038 |
yogurt | -0.032 |
idges | -0.032 |
honey | -0.032 |
ad | -0.031 |
Token | was |
Token ID | 373 |
Feature activation | +9.66e-01 |
Pos logit contributions | |
gy | +0.114 |
bank | +0.092 |
lves | +0.083 |
station | +0.077 |
zag | +0.077 |
Neg logit contributions | |
ey | -0.085 |
ad | -0.079 |
honey | -0.077 |
yogurt | -0.075 |
idges | -0.074 |
Token | in |
Token ID | 287 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
ey | +0.066 |
honey | +0.059 |
ad | +0.058 |
idges | +0.058 |
sweet | +0.057 |
Neg logit contributions | |
gy | -0.087 |
bank | -0.072 |
lves | -0.061 |
zag | -0.058 |
station | -0.058 |
Token | a |
Token ID | 257 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
ey | +0.078 |
yogurt | +0.062 |
ad | +0.062 |
idges | +0.062 |
honey | +0.061 |
Neg logit contributions | |
gy | -0.119 |
bank | -0.098 |
lves | -0.086 |
zag | -0.083 |
station | -0.081 |
Token | there |
Token ID | 612 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
ey | +0.025 |
yogurt | +0.021 |
idges | +0.021 |
honey | +0.021 |
ad | +0.020 |
Neg logit contributions | |
gy | -0.032 |
bank | -0.028 |
lves | -0.023 |
station | -0.022 |
drown | -0.022 |
Token | was |
Token ID | 373 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
gy | +0.119 |
bank | +0.091 |
lves | +0.073 |
station | +0.072 |
zag | +0.072 |
Neg logit contributions | |
ey | -0.121 |
idges | -0.105 |
honey | -0.099 |
yogurt | -0.098 |
ement | -0.095 |
Token | a |
Token ID | 257 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
ey | +0.082 |
idges | +0.069 |
ement | +0.065 |
ad | +0.064 |
yogurt | +0.064 |
Neg logit contributions | |
gy | -0.105 |
bank | -0.084 |
lves | -0.072 |
station | -0.069 |
zag | -0.067 |
Token | generous |
Token ID | 14431 |
Feature activation | +4.81e+00 |
Pos logit contributions | |
ey | +0.024 |
yogurt | +0.019 |
honey | +0.019 |
idges | +0.019 |
sweet | +0.018 |
Neg logit contributions | |
gy | -0.043 |
bank | -0.036 |
lves | -0.031 |
zag | -0.030 |
drown | -0.030 |
Token | pig |
Token ID | 12967 |
Feature activation | -5.23e+00 |
Pos logit contributions | |
ey | +0.392 |
idges | +0.334 |
yogurt | +0.328 |
ad | +0.324 |
honey | +0.320 |
Neg logit contributions | |
gy | -0.406 |
bank | -0.324 |
lves | -0.269 |
zag | -0.254 |
station | -0.253 |
Token | gy |
Token ID | 1360 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
gy | +0.314 |
bank | +0.311 |
drown | +0.299 |
argue | +0.279 |
lves | +0.279 |
Neg logit contributions | |
ad | -0.394 |
ey | -0.349 |
ummy | -0.348 |
sweet | -0.338 |
goodness | -0.335 |
Token | . |
Token ID | 13 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
gy | +1.289 |
drown | +1.006 |
lves | +1.006 |
bank | +0.973 |
argue | +0.956 |
Neg logit contributions | |
ey | -1.001 |
ad | -0.979 |
yogurt | -0.961 |
honey | -0.939 |
ummy | -0.930 |
Token | He |
Token ID | 679 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
ey | +0.010 |
idges | +0.009 |
yogurt | +0.008 |
ement | +0.008 |
honey | +0.008 |
Neg logit contributions | |
gy | -0.013 |
bank | -0.010 |
lves | -0.008 |
station | -0.008 |
zag | -0.008 |
Token | wanted |
Token ID | 2227 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
ey | +0.203 |
yogurt | +0.170 |
idges | +0.169 |
ad | +0.169 |
honey | +0.165 |
Neg logit contributions | |
gy | -0.254 |
bank | -0.208 |
lves | -0.178 |
drown | -0.169 |
zag | -0.169 |
Token | to |
Token ID | 284 |
Feature activation | +6.57e-01 |
Pos logit contributions | |
gy | +0.088 |
bank | +0.072 |
station | +0.051 |
lves | +0.048 |
drown | +0.047 |
Neg logit contributions | |
ey | -0.129 |
ement | -0.114 |
idges | -0.112 |
ad | -0.105 |
importantly | -0.104 |
Token | give |
Token ID | 1577 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
ey | +0.055 |
idges | +0.048 |
yogurt | +0.046 |
ement | +0.046 |
ad | +0.045 |
Neg logit contributions | |
gy | -0.054 |
bank | -0.043 |
lves | -0.034 |
station | -0.033 |
zag | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | +7.60e-02 |
Pos logit contributions | |
ey | +0.226 |
idges | +0.190 |
yogurt | +0.187 |
ad | +0.184 |
honey | +0.182 |
Neg logit contributions | |
gy | -0.271 |
bank | -0.220 |
lves | -0.186 |
zag | -0.176 |
station | -0.175 |
Token |
|
Token ID | 198 |
Feature activation | +7.72e-02 |
Pos logit contributions | |
ey | +0.005 |
idges | +0.005 |
ad | +0.004 |
yogurt | +0.004 |
honey | +0.004 |
Neg logit contributions | |
gy | -0.007 |
bank | -0.006 |
lves | -0.005 |
station | -0.005 |
zag | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +7.42e-02 |
Pos logit contributions | |
ey | +0.006 |
idges | +0.005 |
yogurt | +0.005 |
honey | +0.005 |
ad | +0.005 |
Neg logit contributions | |
gy | -0.007 |
bank | -0.006 |
lves | -0.005 |
zag | -0.005 |
station | -0.005 |
Token | Mom |
Token ID | 29252 |
Feature activation | +4.02e+00 |
Pos logit contributions | |
ey | +0.004 |
idges | +0.003 |
yogurt | +0.003 |
ad | +0.003 |
ement | +0.003 |
Neg logit contributions | |
gy | -0.008 |
bank | -0.007 |
lves | -0.006 |
drown | -0.006 |
zag | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
ey | +0.351 |
idges | +0.304 |
yogurt | +0.298 |
importantly | +0.297 |
mmmm | +0.296 |
Neg logit contributions | |
gy | -0.298 |
bank | -0.233 |
station | -0.209 |
zag | -0.190 |
argue | -0.180 |
Token | Dad |
Token ID | 17415 |
Feature activation | +5.35e+00 |
Pos logit contributions | |
gy | +0.043 |
bank | +0.037 |
argue | +0.032 |
station | +0.032 |
neither | +0.032 |
Neg logit contributions | |
ey | -0.028 |
ement | -0.023 |
idges | -0.023 |
ad | -0.022 |
odic | -0.021 |
Token | hear |
Token ID | 3285 |
Feature activation | -3.24e-01 |
Pos logit contributions | |
ey | +0.424 |
idges | +0.358 |
yogurt | +0.347 |
ement | +0.342 |
importantly | +0.339 |
Neg logit contributions | |
gy | -0.457 |
bank | -0.379 |
station | -0.309 |
argue | -0.308 |
lves | -0.300 |
Token | them |
Token ID | 606 |
Feature activation | -8.74e-01 |
Pos logit contributions | |
gy | +0.027 |
bank | +0.022 |
station | +0.018 |
lves | +0.018 |
argue | +0.017 |
Neg logit contributions | |
ey | -0.027 |
idges | -0.022 |
yogurt | -0.021 |
ement | -0.021 |
ad | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
gy | +0.084 |
bank | +0.070 |
lves | +0.063 |
zag | +0.057 |
drown | +0.055 |
Neg logit contributions | |
ey | -0.058 |
ad | -0.049 |
yogurt | -0.048 |
idges | -0.048 |
honey | -0.048 |
Token | They |
Token ID | 1119 |
Feature activation | -3.90e+00 |
Pos logit contributions | |
ey | +0.012 |
idges | +0.010 |
ad | +0.010 |
yogurt | +0.010 |
honey | +0.010 |
Neg logit contributions | |
gy | -0.008 |
bank | -0.006 |
lves | -0.004 |
station | -0.004 |
zag | -0.004 |
Token | are |
Token ID | 389 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
gy | +0.328 |
bank | +0.263 |
lves | +0.216 |
station | +0.208 |
argue | +0.206 |
Neg logit contributions | |
ey | -0.317 |
idges | -0.271 |
yogurt | -0.268 |
ement | -0.261 |
honey | -0.260 |
Token | the |
Token ID | 262 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
ey | +0.086 |
idges | +0.070 |
yogurt | +0.069 |
ad | +0.068 |
honey | +0.067 |
Neg logit contributions | |
gy | -0.128 |
bank | -0.106 |
lves | -0.091 |
zag | -0.087 |
station | -0.086 |
Token | city |
Token ID | 1748 |
Feature activation | -4.46e+00 |
Pos logit contributions | |
gy | +0.002 |
bank | +0.002 |
lves | +0.001 |
zag | +0.001 |
drown | +0.001 |
Neg logit contributions | |
ey | -0.002 |
idges | -0.001 |
yogurt | -0.001 |
ad | -0.001 |
honey | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
gy | +0.452 |
bank | +0.365 |
lves | +0.343 |
drown | +0.318 |
drowned | +0.302 |
Neg logit contributions | |
ey | -0.266 |
yogurt | -0.227 |
ad | -0.226 |
honey | -0.223 |
ement | -0.223 |
Token | They |
Token ID | 1119 |
Feature activation | -3.79e+00 |
Pos logit contributions | |
ey | +0.022 |
idges | +0.020 |
ad | +0.019 |
yogurt | +0.019 |
honey | +0.019 |
Neg logit contributions | |
gy | -0.015 |
bank | -0.011 |
lves | -0.008 |
station | -0.008 |
drown | -0.008 |
Token | say |
Token ID | 910 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
gy | +0.350 |
bank | +0.275 |
lves | +0.243 |
zag | +0.222 |
station | +0.214 |
Neg logit contributions | |
ey | -0.285 |
ad | -0.237 |
yogurt | -0.234 |
idges | -0.233 |
honey | -0.230 |
Token | hello |
Token ID | 23748 |
Feature activation | +5.62e+00 |
Pos logit contributions | |
ey | +0.117 |
idges | +0.095 |
ad | +0.091 |
yogurt | +0.090 |
ement | +0.089 |
Neg logit contributions | |
gy | -0.133 |
bank | -0.110 |
station | -0.091 |
zag | -0.090 |
argue | -0.089 |
Token | to |
Token ID | 284 |
Feature activation | +6.89e-01 |
Pos logit contributions | |
ey | +0.400 |
yogurt | +0.335 |
idges | +0.334 |
honey | +0.333 |
ad | +0.331 |
Neg logit contributions | |
gy | -0.510 |
bank | -0.426 |
lves | -0.370 |
drown | -0.342 |
zag | -0.336 |
Token | the |
Token ID | 262 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
ey | +0.046 |
idges | +0.039 |
yogurt | +0.039 |
honey | +0.038 |
ad | +0.038 |
Neg logit contributions | |
gy | -0.067 |
bank | -0.055 |
lves | -0.048 |
zag | -0.046 |
drown | -0.044 |
Token | people |
Token ID | 661 |
Feature activation | +6.32e-01 |
Pos logit contributions | |
ey | +0.002 |
idges | +0.002 |
yogurt | +0.002 |
ad | +0.002 |
honey | +0.002 |
Neg logit contributions | |
gy | -0.003 |
bank | -0.003 |
lves | -0.002 |
station | -0.002 |
zag | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
ey | +0.042 |
idges | +0.036 |
yogurt | +0.036 |
ad | +0.035 |
honey | +0.035 |
Neg logit contributions | |
gy | -0.061 |
bank | -0.050 |
lves | -0.044 |
drown | -0.041 |
zag | -0.040 |
Token | the |
Token ID | 262 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
gy | +0.030 |
bank | +0.024 |
lves | +0.021 |
zag | +0.020 |
drown | +0.020 |
Neg logit contributions | |
ey | -0.021 |
yogurt | -0.017 |
idges | -0.017 |
honey | -0.017 |
ad | -0.016 |
Token | 're |
Token ID | 821 |
Feature activation | -2.38e+00 |
Pos logit contributions | |
ey | +0.181 |
ad | +0.161 |
idges | +0.156 |
honey | +0.153 |
yogurt | +0.148 |
Neg logit contributions | |
gy | -0.281 |
bank | -0.237 |
lves | -0.204 |
zag | -0.195 |
station | -0.191 |
Token | the |
Token ID | 262 |
Feature activation | +3.69e-02 |
Pos logit contributions | |
gy | +0.230 |
bank | +0.190 |
lves | +0.162 |
zag | +0.155 |
station | +0.154 |
Neg logit contributions | |
ey | -0.168 |
idges | -0.139 |
yogurt | -0.136 |
ad | -0.134 |
honey | -0.132 |
Token | best |
Token ID | 1266 |
Feature activation | -6.84e-01 |
Pos logit contributions | |
ey | +0.002 |
honey | +0.002 |
ad | +0.002 |
sweet | +0.002 |
yogurt | +0.002 |
Neg logit contributions | |
gy | -0.004 |
bank | -0.003 |
lves | -0.002 |
zag | -0.002 |
argue | -0.002 |
Token | at |
Token ID | 379 |
Feature activation | -2.99e+00 |
Pos logit contributions | |
gy | +0.062 |
bank | +0.054 |
lves | +0.047 |
banks | +0.042 |
argue | +0.042 |
Neg logit contributions | |
ey | -0.045 |
honey | -0.042 |
ummy | -0.041 |
ad | -0.041 |
yogurt | -0.041 |
Token | calling |
Token ID | 4585 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
gy | +0.285 |
bank | +0.236 |
lves | +0.205 |
station | +0.187 |
zag | +0.187 |
Neg logit contributions | |
ey | -0.209 |
idges | -0.175 |
yogurt | -0.173 |
honey | -0.172 |
ad | -0.172 |
Token | birds |
Token ID | 10087 |
Feature activation | +5.23e+00 |
Pos logit contributions | |
ey | +0.081 |
idges | +0.064 |
yogurt | +0.062 |
ement | +0.061 |
ad | +0.060 |
Neg logit contributions | |
gy | -0.107 |
bank | -0.087 |
lves | -0.073 |
station | -0.073 |
zag | -0.071 |
Token | !" |
Token ID | 2474 |
Feature activation | +4.88e-01 |
Pos logit contributions | |
ey | +0.279 |
honey | +0.265 |
idges | +0.250 |
yogurt | +0.245 |
sweet | +0.241 |
Neg logit contributions | |
gy | -0.516 |
bank | -0.440 |
lves | -0.392 |
zag | -0.349 |
station | -0.346 |
Token | Max |
Token ID | 5436 |
Feature activation | +8.54e-02 |
Pos logit contributions | |
ey | +0.036 |
idges | +0.030 |
yogurt | +0.030 |
ad | +0.029 |
honey | +0.029 |
Neg logit contributions | |
gy | -0.045 |
bank | -0.037 |
lves | -0.031 |
zag | -0.029 |
station | -0.029 |
Token | bark |
Token ID | 21405 |
Feature activation | -3.08e+00 |
Pos logit contributions | |
ey | +0.006 |
ad | +0.005 |
yogurt | +0.005 |
honey | +0.005 |
idges | +0.005 |
Neg logit contributions | |
gy | -0.008 |
bank | -0.007 |
lves | -0.006 |
drown | -0.006 |
station | -0.005 |
Token | ed |
Token ID | 276 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
gy | +0.285 |
bank | +0.237 |
lves | +0.203 |
station | +0.195 |
drown | +0.195 |
Neg logit contributions | |
ey | -0.210 |
ad | -0.182 |
idges | -0.178 |
ement | -0.177 |
yogurt | -0.177 |
Token | happily |
Token ID | 18177 |
Feature activation | +3.55e-01 |
Pos logit contributions | |
gy | +0.124 |
bank | +0.101 |
lves | +0.087 |
zag | +0.084 |
station | +0.082 |
Neg logit contributions | |
ey | -0.101 |
honey | -0.084 |
yogurt | -0.084 |
idges | -0.084 |
ad | -0.083 |
Token | looking |
Token ID | 2045 |
Feature activation | +3.32e+00 |
Pos logit contributions | |
gy | +0.426 |
bank | +0.352 |
lves | +0.302 |
zag | +0.294 |
station | +0.290 |
Neg logit contributions | |
ey | -0.286 |
yogurt | -0.246 |
honey | -0.245 |
idges | -0.239 |
ad | -0.238 |
Token | . |
Token ID | 13 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
ey | +0.256 |
idges | +0.216 |
yogurt | +0.212 |
ad | +0.209 |
honey | +0.207 |
Neg logit contributions | |
gy | -0.296 |
bank | -0.239 |
lves | -0.201 |
zag | -0.191 |
station | -0.189 |
Token | But |
Token ID | 887 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
ey | +0.013 |
idges | +0.011 |
yogurt | +0.010 |
ad | +0.010 |
honey | +0.010 |
Neg logit contributions | |
gy | -0.025 |
bank | -0.021 |
lves | -0.018 |
station | -0.017 |
zag | -0.017 |
Token | Tommy |
Token ID | 19919 |
Feature activation | +9.98e-01 |
Pos logit contributions | |
gy | +0.311 |
bank | +0.266 |
lves | +0.240 |
zag | +0.236 |
drown | +0.235 |
Neg logit contributions | |
ey | -0.162 |
yogurt | -0.150 |
ad | -0.143 |
honey | -0.139 |
himself | -0.136 |
Token | caught |
Token ID | 4978 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
ey | +0.063 |
ad | +0.058 |
yogurt | +0.056 |
honey | +0.055 |
idges | +0.055 |
Neg logit contributions | |
gy | -0.086 |
bank | -0.070 |
drown | -0.065 |
lves | -0.064 |
zag | -0.061 |
Token | him |
Token ID | 683 |
Feature activation | +6.53e+00 |
Pos logit contributions | |
gy | +0.204 |
bank | +0.167 |
lves | +0.143 |
station | +0.133 |
zag | +0.132 |
Neg logit contributions | |
ey | -0.151 |
idges | -0.124 |
yogurt | -0.119 |
ad | -0.118 |
honey | -0.116 |
Token | and |
Token ID | 290 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
ey | +0.420 |
yogurt | +0.359 |
honey | +0.352 |
ad | +0.349 |
idges | +0.345 |
Neg logit contributions | |
gy | -0.625 |
bank | -0.509 |
lves | -0.463 |
drown | -0.438 |
station | -0.427 |
Token | got |
Token ID | 1392 |
Feature activation | -5.08e+00 |
Pos logit contributions | |
gy | +0.020 |
bank | +0.016 |
lves | +0.014 |
zag | +0.014 |
drown | +0.013 |
Neg logit contributions | |
ey | -0.016 |
yogurt | -0.015 |
honey | -0.015 |
idges | -0.014 |
ad | -0.014 |
Token | very |
Token ID | 845 |
Feature activation | -7.19e-01 |
Pos logit contributions | |
gy | +0.481 |
bank | +0.381 |
zag | +0.339 |
lves | +0.333 |
station | +0.324 |
Neg logit contributions | |
ey | -0.358 |
yogurt | -0.313 |
ad | -0.302 |
honey | -0.301 |
idges | -0.299 |
Token | upset |
Token ID | 9247 |
Feature activation | -4.84e+00 |
Pos logit contributions | |
gy | +0.064 |
bank | +0.049 |
lves | +0.043 |
station | +0.042 |
zag | +0.042 |
Neg logit contributions | |
ey | -0.057 |
yogurt | -0.049 |
idges | -0.047 |
ad | -0.047 |
honey | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
gy | +0.368 |
bank | +0.292 |
neither | +0.222 |
lves | +0.214 |
station | +0.214 |
Neg logit contributions | |
ey | -0.448 |
idges | -0.375 |
ad | -0.354 |
yogurt | -0.353 |
mmmm | -0.351 |
Token | y |
Token ID | 331 |
Feature activation | +4.38e+00 |
Pos logit contributions | |
ey | +0.002 |
idges | +0.002 |
ad | +0.002 |
yogurt | +0.002 |
ement | +0.002 |
Neg logit contributions | |
gy | -0.003 |
bank | -0.003 |
station | -0.002 |
lves | -0.002 |
fight | -0.002 |
Token | ummy |
Token ID | 13513 |
Feature activation | +8.82e+00 |
Pos logit contributions | |
ey | +0.129 |
idges | +0.081 |
ad | +0.065 |
importantly | +0.060 |
yogurt | +0.060 |
Neg logit contributions | |
gy | -0.597 |
bank | -0.523 |
lves | -0.455 |
station | -0.453 |
house | -0.447 |
Token | sugar |
Token ID | 7543 |
Feature activation | +7.82e+00 |
Pos logit contributions | |
ey | +0.578 |
idges | +0.490 |
ad | +0.456 |
importantly | +0.452 |
ango | +0.439 |
Neg logit contributions | |
gy | -0.861 |
bank | -0.748 |
station | -0.624 |
fight | -0.620 |
house | -0.605 |
Token | for |
Token ID | 329 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
ey | +0.462 |
yogurt | +0.427 |
honey | +0.409 |
ummy | +0.395 |
ad | +0.380 |
Neg logit contributions | |
gy | -0.770 |
bank | -0.630 |
lves | -0.613 |
drown | -0.551 |
drowned | -0.536 |
Token | the |
Token ID | 262 |
Feature activation | +5.79e-02 |
Pos logit contributions | |
gy | +0.153 |
bank | +0.130 |
drown | +0.105 |
neither | +0.104 |
drowned | +0.104 |
Neg logit contributions | |
ey | -0.110 |
idges | -0.082 |
ement | -0.078 |
importantly | -0.072 |
ad | -0.072 |
Token | cake |
Token ID | 12187 |
Feature activation | +4.98e+00 |
Pos logit contributions | |
ey | +0.005 |
idges | +0.004 |
ad | +0.004 |
ement | +0.004 |
yogurt | +0.004 |
Neg logit contributions | |
gy | -0.005 |
bank | -0.004 |
lves | -0.003 |
station | -0.003 |
fight | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +3.49e-01 |
Pos logit contributions | |
ey | +0.326 |
yogurt | +0.283 |
honey | +0.273 |
idges | +0.270 |
ad | +0.267 |
Neg logit contributions | |
gy | -0.478 |
bank | -0.392 |
lves | -0.358 |
drown | -0.328 |
zag | -0.322 |
Token |
|
Token ID | 198 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
ey | +0.029 |
idges | +0.025 |
ement | +0.023 |
ad | +0.023 |
yogurt | +0.023 |
Neg logit contributions | |
gy | -0.031 |
bank | -0.025 |
lves | -0.019 |
zag | -0.019 |
station | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
ey | +0.010 |
idges | +0.008 |
yogurt | +0.008 |
honey | +0.008 |
ement | +0.008 |
Neg logit contributions | |
gy | -0.011 |
bank | -0.009 |
zag | -0.007 |
lves | -0.007 |
station | -0.007 |
Token | The |
Token ID | 464 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
ey | +0.009 |
ement | +0.008 |
idges | +0.008 |
yogurt | +0.008 |
honey | +0.007 |
Neg logit contributions | |
gy | -0.013 |
bank | -0.011 |
lves | -0.009 |
zag | -0.009 |
station | -0.009 |
Token | bear |
Token ID | 6842 |
Feature activation | -7.12e-01 |
Pos logit contributions | |
gy | +0.017 |
bank | +0.013 |
station | +0.010 |
fight | +0.010 |
lves | +0.010 |
Neg logit contributions | |
ey | -0.021 |
idges | -0.019 |
ad | -0.018 |
ement | -0.018 |
mmmm | -0.018 |
Token | back |
Token ID | 736 |
Feature activation | -5.46e-01 |
Pos logit contributions | |
gy | +0.410 |
bank | +0.330 |
lves | +0.300 |
drown | +0.279 |
station | +0.270 |
Neg logit contributions | |
ey | -0.347 |
yogurt | -0.314 |
honey | -0.309 |
ad | -0.294 |
idges | -0.290 |
Token | on |
Token ID | 319 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
gy | +0.043 |
bank | +0.034 |
lves | +0.027 |
zag | +0.026 |
station | +0.025 |
Neg logit contributions | |
ey | -0.048 |
idges | -0.041 |
yogurt | -0.040 |
ad | -0.040 |
honey | -0.040 |
Token | the |
Token ID | 262 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
gy | +0.116 |
bank | +0.096 |
station | +0.077 |
lves | +0.076 |
zag | +0.075 |
Neg logit contributions | |
ey | -0.093 |
idges | -0.080 |
yogurt | -0.072 |
ement | -0.072 |
importantly | -0.072 |
Token | scale |
Token ID | 5046 |
Feature activation | +2.99e+00 |
Pos logit contributions | |
ey | +0.011 |
idges | +0.010 |
ement | +0.009 |
importantly | +0.009 |
ad | +0.009 |
Neg logit contributions | |
gy | -0.012 |
bank | -0.010 |
station | -0.008 |
lves | -0.007 |
zag | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
ey | +0.200 |
yogurt | +0.168 |
ad | +0.165 |
honey | +0.162 |
ement | +0.161 |
Neg logit contributions | |
gy | -0.284 |
bank | -0.227 |
lves | -0.217 |
drown | -0.198 |
drowned | -0.191 |
Token | She |
Token ID | 1375 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
ey | +0.013 |
idges | +0.012 |
ad | +0.011 |
ement | +0.011 |
honey | +0.011 |
Neg logit contributions | |
gy | -0.013 |
bank | -0.010 |
station | -0.008 |
lves | -0.008 |
house | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
ey | +0.120 |
idges | +0.101 |
yogurt | +0.099 |
ad | +0.097 |
honey | +0.096 |
Neg logit contributions | |
gy | -0.141 |
bank | -0.114 |
lves | -0.096 |
zag | -0.092 |
station | -0.091 |
Token | her |
Token ID | 607 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
gy | +0.029 |
bank | +0.024 |
lves | +0.020 |
station | +0.020 |
argue | +0.020 |
Neg logit contributions | |
ey | -0.018 |
idges | -0.015 |
ement | -0.014 |
ad | -0.014 |
yogurt | -0.013 |
Token | Daddy |
Token ID | 26926 |
Feature activation | +4.99e+00 |
Pos logit contributions | |
ey | +0.096 |
idges | +0.086 |
mmmm | +0.084 |
ad | +0.080 |
ention | +0.078 |
Neg logit contributions | |
gy | -0.079 |
bank | -0.062 |
jet | -0.047 |
station | -0.046 |
lves | -0.045 |
Token | smiled |
Token ID | 13541 |
Feature activation | +6.18e+00 |
Pos logit contributions | |
ey | +0.393 |
idges | +0.341 |
yogurt | +0.340 |
ad | +0.340 |
honey | +0.331 |
Neg logit contributions | |
gy | -0.415 |
bank | -0.330 |
lves | -0.281 |
drown | -0.264 |
zag | -0.263 |
Token | at |
Token ID | 379 |
Feature activation | -3.21e+00 |
Pos logit contributions | |
ey | +0.537 |
idges | +0.460 |
yogurt | +0.445 |
ad | +0.434 |
honey | +0.423 |
Neg logit contributions | |
gy | -0.512 |
bank | -0.392 |
lves | -0.315 |
station | -0.311 |
zag | -0.306 |
Token | He |
Token ID | 679 |
Feature activation | +2.93e+00 |
Pos logit contributions | |
ey | +0.020 |
idges | +0.019 |
ement | +0.018 |
honey | +0.017 |
illy | +0.016 |
Neg logit contributions | |
gy | -0.024 |
bank | -0.019 |
station | -0.016 |
house | -0.015 |
zag | -0.015 |
Token | found |
Token ID | 1043 |
Feature activation | -3.09e+00 |
Pos logit contributions | |
ey | +0.227 |
yogurt | +0.192 |
idges | +0.190 |
ad | +0.188 |
honey | +0.184 |
Neg logit contributions | |
gy | -0.256 |
bank | -0.206 |
lves | -0.175 |
zag | -0.165 |
drown | -0.164 |
Token | a |
Token ID | 257 |
Feature activation | +5.36e-01 |
Pos logit contributions | |
gy | +0.202 |
bank | +0.160 |
lves | +0.121 |
crashed | +0.119 |
neither | +0.118 |
Neg logit contributions | |
ey | -0.291 |
idges | -0.262 |
ement | -0.261 |
mmmm | -0.259 |
mmm | -0.254 |
Token | big |
Token ID | 1263 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
ey | +0.040 |
idges | +0.036 |
ad | +0.035 |
ement | +0.034 |
mmmm | +0.032 |
Neg logit contributions | |
gy | -0.047 |
bank | -0.039 |
station | -0.032 |
lves | -0.032 |
zag | -0.032 |
Token | bowl |
Token ID | 9396 |
Feature activation | +3.73e+00 |
Pos logit contributions | |
ey | +0.096 |
idges | +0.088 |
ad | +0.082 |
ement | +0.081 |
importantly | +0.080 |
Neg logit contributions | |
gy | -0.096 |
bank | -0.077 |
station | -0.061 |
lves | -0.059 |
fight | -0.058 |
Token | on |
Token ID | 319 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
ey | +0.244 |
idges | +0.201 |
yogurt | +0.187 |
honey | +0.187 |
ad | +0.187 |
Neg logit contributions | |
gy | -0.378 |
bank | -0.314 |
lves | -0.263 |
station | -0.260 |
zag | -0.260 |
Token | the |
Token ID | 262 |
Feature activation | +3.14e-01 |
Pos logit contributions | |
gy | +0.093 |
bank | +0.078 |
lves | +0.062 |
station | +0.061 |
argue | +0.058 |
Neg logit contributions | |
ey | -0.077 |
idges | -0.066 |
ement | -0.062 |
importantly | -0.062 |
ad | -0.060 |
Token | floor |
Token ID | 4314 |
Feature activation | +7.22e-01 |
Pos logit contributions | |
ey | +0.025 |
idges | +0.022 |
ad | +0.020 |
ement | +0.020 |
importantly | +0.020 |
Neg logit contributions | |
gy | -0.028 |
bank | -0.023 |
station | -0.019 |
lves | -0.018 |
zag | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
ey | +0.057 |
idges | +0.048 |
yogurt | +0.046 |
honey | +0.046 |
ad | +0.046 |
Neg logit contributions | |
gy | -0.064 |
bank | -0.052 |
lves | -0.042 |
zag | -0.041 |
station | -0.041 |
Token | It |
Token ID | 632 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
ey | +0.024 |
idges | +0.023 |
ement | +0.022 |
illy | +0.021 |
ocol | +0.021 |
Neg logit contributions | |
gy | -0.026 |
bank | -0.021 |
station | -0.016 |
house | -0.016 |
lves | -0.015 |
Token | was |
Token ID | 373 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
ey | +0.111 |
idges | +0.094 |
yogurt | +0.089 |
ad | +0.089 |
ement | +0.089 |
Neg logit contributions | |
gy | -0.127 |
bank | -0.102 |
lves | -0.083 |
zag | -0.081 |
station | -0.080 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
gy | +0.018 |
bank | +0.014 |
lves | +0.013 |
drown | +0.012 |
zag | +0.012 |
Neg logit contributions | |
ey | -0.019 |
yogurt | -0.017 |
ad | -0.016 |
idges | -0.016 |
honey | -0.016 |
Token | a |
Token ID | 257 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
gy | +0.097 |
bank | +0.079 |
station | +0.063 |
lves | +0.063 |
neither | +0.062 |
Neg logit contributions | |
ey | -0.097 |
idges | -0.081 |
ement | -0.079 |
mmmm | -0.075 |
importantly | -0.075 |
Token | small |
Token ID | 1402 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
ey | +0.022 |
idges | +0.019 |
ad | +0.019 |
ement | +0.018 |
yogurt | +0.017 |
Neg logit contributions | |
gy | -0.029 |
bank | -0.025 |
lves | -0.021 |
station | -0.021 |
zag | -0.020 |
Token | black |
Token ID | 2042 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
ey | +0.024 |
idges | +0.020 |
ad | +0.019 |
yogurt | +0.019 |
ement | +0.019 |
Neg logit contributions | |
gy | -0.026 |
bank | -0.021 |
lves | -0.017 |
station | -0.017 |
zag | -0.016 |
Token | thing |
Token ID | 1517 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
ey | +0.026 |
idges | +0.024 |
importantly | +0.023 |
ement | +0.022 |
ad | +0.021 |
Neg logit contributions | |
gy | -0.025 |
bank | -0.020 |
station | -0.016 |
zag | -0.015 |
house | -0.015 |
Token | on |
Token ID | 319 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
gy | +0.213 |
bank | +0.166 |
zag | +0.132 |
station | +0.131 |
lves | +0.127 |
Neg logit contributions | |
ey | -0.247 |
idges | -0.215 |
ad | -0.203 |
ement | -0.202 |
mmmm | -0.200 |
Token | the |
Token ID | 262 |
Feature activation | +5.26e-02 |
Pos logit contributions | |
gy | +0.112 |
bank | +0.093 |
lves | +0.080 |
zag | +0.077 |
station | +0.076 |
Neg logit contributions | |
ey | -0.074 |
idges | -0.061 |
yogurt | -0.059 |
ad | -0.058 |
honey | -0.057 |
Token | ground |
Token ID | 2323 |
Feature activation | +9.10e-01 |
Pos logit contributions | |
ey | +0.004 |
idges | +0.004 |
ement | +0.004 |
importantly | +0.004 |
ad | +0.004 |
Neg logit contributions | |
gy | -0.004 |
bank | -0.004 |
station | -0.003 |
lves | -0.003 |
zag | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
ey | +0.076 |
idges | +0.066 |
honey | +0.063 |
yogurt | +0.062 |
ement | +0.061 |
Neg logit contributions | |
gy | -0.075 |
bank | -0.061 |
station | -0.048 |
zag | -0.048 |
lves | -0.045 |
Token | It |
Token ID | 632 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
ey | +0.013 |
idges | +0.012 |
ad | +0.011 |
ement | +0.011 |
importantly | +0.011 |
Neg logit contributions | |
gy | -0.016 |
bank | -0.013 |
house | -0.011 |
lves | -0.010 |
station | -0.010 |
Token | had |
Token ID | 550 |
Feature activation | -9.24e-01 |
Pos logit contributions | |
ey | +0.106 |
idges | +0.092 |
ement | +0.088 |
yogurt | +0.087 |
honey | +0.086 |
Neg logit contributions | |
gy | -0.109 |
bank | -0.085 |
lves | -0.069 |
zag | -0.067 |
station | -0.066 |
Token | slide |
Token ID | 10649 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
ey | +0.046 |
yogurt | +0.040 |
ad | +0.039 |
idges | +0.039 |
honey | +0.038 |
Neg logit contributions | |
gy | -0.040 |
bank | -0.031 |
lves | -0.027 |
zag | -0.024 |
drown | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | +5.99e-02 |
Pos logit contributions | |
gy | +0.175 |
bank | +0.145 |
drown | +0.138 |
lves | +0.133 |
drowned | +0.125 |
Neg logit contributions | |
ey | -0.093 |
yogurt | -0.086 |
sweet | -0.079 |
honey | -0.079 |
ement | -0.076 |
Token | swing |
Token ID | 9628 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
ey | +0.005 |
yogurt | +0.004 |
idges | +0.004 |
honey | +0.004 |
ad | +0.004 |
Neg logit contributions | |
gy | -0.005 |
bank | -0.004 |
lves | -0.003 |
zag | -0.003 |
drown | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | -5.32e-01 |
Pos logit contributions | |
gy | +0.152 |
bank | +0.124 |
lves | +0.116 |
drown | +0.112 |
argue | +0.105 |
Neg logit contributions | |
ey | -0.095 |
yogurt | -0.080 |
honey | -0.079 |
ad | -0.078 |
idges | -0.077 |
Token | run |
Token ID | 1057 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
gy | +0.044 |
bank | +0.033 |
lves | +0.030 |
zag | +0.028 |
station | +0.026 |
Neg logit contributions | |
ey | -0.044 |
yogurt | -0.039 |
honey | -0.037 |
ad | -0.037 |
idges | -0.036 |
Token | on |
Token ID | 319 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
gy | +0.181 |
lves | +0.146 |
bank | +0.143 |
drown | +0.140 |
station | +0.134 |
Neg logit contributions | |
ey | -0.060 |
sweet | -0.057 |
yogurt | -0.056 |
ad | -0.055 |
himself | -0.051 |
Token | the |
Token ID | 262 |
Feature activation | -6.71e-02 |
Pos logit contributions | |
gy | +0.119 |
bank | +0.100 |
lves | +0.082 |
station | +0.080 |
argue | +0.079 |
Neg logit contributions | |
ey | -0.093 |
idges | -0.078 |
ad | -0.074 |
yogurt | -0.073 |
ement | -0.073 |
Token | grass |
Token ID | 8701 |
Feature activation | +4.15e+00 |
Pos logit contributions | |
gy | +0.006 |
bank | +0.005 |
lves | +0.004 |
station | +0.004 |
zag | +0.004 |
Neg logit contributions | |
ey | -0.005 |
idges | -0.005 |
importantly | -0.005 |
ad | -0.005 |
ement | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -8.53e-02 |
Pos logit contributions | |
ey | +0.189 |
yogurt | +0.176 |
ad | +0.170 |
ummy | +0.165 |
honey | +0.163 |
Neg logit contributions | |
gy | -0.440 |
lves | -0.372 |
bank | -0.359 |
drown | -0.338 |
drowned | -0.333 |
Token | It |
Token ID | 632 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
gy | +0.008 |
bank | +0.007 |
lves | +0.006 |
zag | +0.006 |
drown | +0.006 |
Neg logit contributions | |
ey | -0.005 |
yogurt | -0.005 |
honey | -0.004 |
idges | -0.004 |
ummy | -0.004 |
Token | was |
Token ID | 373 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
ey | +0.054 |
yogurt | +0.045 |
ad | +0.044 |
honey | +0.041 |
sweet | +0.041 |
Neg logit contributions | |
gy | -0.132 |
bank | -0.111 |
lves | -0.105 |
drown | -0.100 |
argue | -0.098 |
Token | little |
Token ID | 1310 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
ey | +0.013 |
yogurt | +0.010 |
idges | +0.010 |
honey | +0.010 |
ad | +0.010 |
Neg logit contributions | |
gy | -0.020 |
bank | -0.017 |
lves | -0.015 |
zag | -0.014 |
drown | -0.014 |
Token | girl |
Token ID | 2576 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ey | +0.075 |
yogurt | +0.064 |
honey | +0.064 |
ummy | +0.060 |
sweet | +0.060 |
Neg logit contributions | |
gy | -0.095 |
bank | -0.079 |
lves | -0.066 |
zag | -0.065 |
drown | -0.064 |
Token | named |
Token ID | 3706 |
Feature activation | -6.66e+00 |
Pos logit contributions | |
gy | +0.080 |
bank | +0.068 |
lves | +0.054 |
drown | +0.052 |
zag | +0.050 |
Neg logit contributions | |
ey | -0.115 |
yogurt | -0.104 |
ad | -0.104 |
honey | -0.101 |
sweet | -0.100 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
gy | +0.648 |
bank | +0.547 |
lves | +0.464 |
zag | +0.445 |
drown | +0.441 |
Neg logit contributions | |
ey | -0.449 |
yogurt | -0.385 |
honey | -0.378 |
idges | -0.376 |
ad | -0.356 |
Token | . |
Token ID | 13 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
gy | +0.196 |
bank | +0.166 |
lves | +0.143 |
zag | +0.136 |
drown | +0.136 |
Neg logit contributions | |
ey | -0.140 |
ad | -0.116 |
yogurt | -0.115 |
idges | -0.114 |
honey | -0.111 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
gy | +0.002 |
bank | +0.001 |
house | +0.001 |
argue | +0.001 |
station | +0.001 |
Neg logit contributions | |
ey | -0.001 |
idges | -0.001 |
ement | -0.001 |
ad | -0.001 |
ango | -0.001 |
Token | loved |
Token ID | 6151 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
gy | +0.197 |
bank | +0.164 |
lves | +0.141 |
drown | +0.134 |
zag | +0.134 |
Neg logit contributions | |
ey | -0.146 |
yogurt | -0.120 |
idges | -0.120 |
ad | -0.118 |
honey | -0.116 |
Token | to |
Token ID | 284 |
Feature activation | +4.93e-01 |
Pos logit contributions | |
ey | +0.221 |
ement | +0.209 |
idges | +0.194 |
ad | +0.188 |
importantly | +0.183 |
Neg logit contributions | |
gy | -0.164 |
bank | -0.132 |
fight | -0.098 |
station | -0.097 |
lves | -0.096 |
Token | pick |
Token ID | 2298 |
Feature activation | +8.74e-01 |
Pos logit contributions | |
ey | +0.049 |
ement | +0.046 |
idges | +0.044 |
importantly | +0.043 |
ummy | +0.041 |
Neg logit contributions | |
gy | -0.033 |
bank | -0.024 |
fight | -0.019 |
argue | -0.018 |
station | -0.017 |
Token | flowers |
Token ID | 12734 |
Feature activation | +3.26e+00 |
Pos logit contributions | |
ey | +0.060 |
idges | +0.050 |
ad | +0.047 |
ement | +0.047 |
importantly | +0.047 |
Neg logit contributions | |
gy | -0.084 |
bank | -0.071 |
lves | -0.061 |
station | -0.056 |
zag | -0.055 |
Token | in |
Token ID | 287 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
ey | +0.226 |
yogurt | +0.192 |
idges | +0.190 |
honey | +0.189 |
ad | +0.184 |
Neg logit contributions | |
gy | -0.311 |
bank | -0.252 |
lves | -0.220 |
drown | -0.215 |
zag | -0.210 |
Token | . |
Token ID | 13 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
gy | +0.068 |
bank | +0.056 |
lves | +0.053 |
drown | +0.049 |
zag | +0.047 |
Neg logit contributions | |
ey | -0.039 |
yogurt | -0.035 |
honey | -0.035 |
ad | -0.034 |
goodness | -0.033 |
Token | He |
Token ID | 679 |
Feature activation | +3.01e+00 |
Pos logit contributions | |
ey | +0.027 |
idges | +0.023 |
ad | +0.022 |
yogurt | +0.022 |
ement | +0.021 |
Neg logit contributions | |
gy | -0.037 |
bank | -0.030 |
lves | -0.026 |
station | -0.025 |
zag | -0.024 |
Token | will |
Token ID | 481 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
ey | +0.229 |
idges | +0.193 |
yogurt | +0.192 |
ad | +0.189 |
honey | +0.188 |
Neg logit contributions | |
gy | -0.267 |
bank | -0.214 |
lves | -0.183 |
zag | -0.172 |
drown | -0.170 |
Token | look |
Token ID | 804 |
Feature activation | -2.39e-01 |
Pos logit contributions | |
ey | +0.093 |
idges | +0.079 |
ad | +0.078 |
yogurt | +0.078 |
honey | +0.077 |
Neg logit contributions | |
gy | -0.099 |
bank | -0.079 |
lves | -0.067 |
zag | -0.062 |
station | -0.062 |
Token | at |
Token ID | 379 |
Feature activation | -3.01e+00 |
Pos logit contributions | |
gy | +0.019 |
bank | +0.015 |
lves | +0.013 |
station | +0.012 |
drown | +0.011 |
Neg logit contributions | |
ey | -0.021 |
idges | -0.017 |
ad | -0.017 |
yogurt | -0.017 |
ement | -0.017 |
Token | their |
Token ID | 511 |
Feature activation | -5.33e+00 |
Pos logit contributions | |
gy | +0.279 |
bank | +0.230 |
lves | +0.194 |
station | +0.183 |
neither | +0.179 |
Neg logit contributions | |
ey | -0.224 |
idges | -0.182 |
ement | -0.176 |
ad | -0.174 |
yogurt | -0.171 |
Token | eyes |
Token ID | 2951 |
Feature activation | +8.64e+00 |
Pos logit contributions | |
gy | +0.468 |
bank | +0.379 |
lves | +0.316 |
zag | +0.300 |
station | +0.299 |
Neg logit contributions | |
ey | -0.416 |
idges | -0.352 |
yogurt | -0.344 |
ad | -0.341 |
honey | -0.335 |
Token | and |
Token ID | 290 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
ey | +0.715 |
idges | +0.600 |
ad | +0.578 |
yogurt | +0.578 |
honey | +0.564 |
Neg logit contributions | |
gy | -0.731 |
bank | -0.589 |
lves | -0.469 |
zag | -0.468 |
station | -0.467 |
Token | give |
Token ID | 1577 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
gy | +0.012 |
bank | +0.010 |
lves | +0.008 |
drown | +0.008 |
station | +0.008 |
Neg logit contributions | |
ey | -0.012 |
idges | -0.010 |
yogurt | -0.010 |
ement | -0.010 |
ad | -0.010 |
Token | them |
Token ID | 606 |
Feature activation | -7.38e-01 |
Pos logit contributions | |
gy | +0.129 |
bank | +0.108 |
lves | +0.092 |
station | +0.087 |
drown | +0.086 |
Neg logit contributions | |
ey | -0.091 |
idges | -0.074 |
yogurt | -0.070 |
ad | -0.070 |
ement | -0.070 |
Token | stickers |
Token ID | 28568 |
Feature activation | +4.19e+00 |
Pos logit contributions | |
gy | +0.069 |
bank | +0.057 |
lves | +0.048 |
station | +0.046 |
zag | +0.046 |
Neg logit contributions | |
ey | -0.053 |
idges | -0.044 |
yogurt | -0.043 |
ad | -0.042 |
ement | -0.042 |
Token | Mom |
Token ID | 11254 |
Feature activation | +4.00e+00 |
Pos logit contributions | |
gy | +0.035 |
bank | +0.030 |
lves | +0.026 |
zag | +0.025 |
station | +0.025 |
Neg logit contributions | |
ey | -0.019 |
idges | -0.015 |
yogurt | -0.015 |
ad | -0.015 |
honey | -0.015 |
Token | my |
Token ID | 1820 |
Feature activation | +3.19e+00 |
Pos logit contributions | |
ey | +0.351 |
idges | +0.288 |
mmmm | +0.278 |
yogurt | +0.273 |
importantly | +0.269 |
Neg logit contributions | |
gy | -0.356 |
bank | -0.250 |
station | -0.212 |
zag | -0.211 |
argue | -0.202 |
Token | . |
Token ID | 13 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
ey | +0.276 |
idges | +0.231 |
yogurt | +0.224 |
mmmm | +0.215 |
honey | +0.214 |
Neg logit contributions | |
gy | -0.273 |
bank | -0.207 |
zag | -0.165 |
station | -0.162 |
lves | -0.162 |
Token | " |
Token ID | 366 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
ey | +0.010 |
idges | +0.008 |
ad | +0.008 |
ement | +0.008 |
honey | +0.008 |
Neg logit contributions | |
gy | -0.010 |
bank | -0.008 |
station | -0.007 |
zag | -0.007 |
neither | -0.006 |
Token | Let |
Token ID | 5756 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
ey | +0.148 |
idges | +0.121 |
honey | +0.119 |
yogurt | +0.117 |
ad | +0.116 |
Neg logit contributions | |
gy | -0.213 |
bank | -0.177 |
lves | -0.160 |
drown | -0.148 |
neither | -0.148 |
Token | 's |
Token ID | 338 |
Feature activation | -6.04e+00 |
Pos logit contributions | |
ey | +0.211 |
idges | +0.189 |
honey | +0.186 |
ad | +0.185 |
yogurt | +0.185 |
Neg logit contributions | |
gy | -0.199 |
bank | -0.154 |
lves | -0.128 |
zag | -0.124 |
station | -0.117 |
Token | go |
Token ID | 467 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
gy | +0.500 |
bank | +0.400 |
lves | +0.362 |
zag | +0.325 |
station | +0.311 |
Neg logit contributions | |
ey | -0.473 |
yogurt | -0.411 |
idges | -0.407 |
ad | -0.405 |
honey | -0.399 |
Token | and |
Token ID | 290 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
ey | +0.016 |
yogurt | +0.015 |
ad | +0.015 |
honey | +0.014 |
idges | +0.014 |
Neg logit contributions | |
gy | -0.022 |
bank | -0.018 |
lves | -0.017 |
sank | -0.016 |
neither | -0.016 |
Token | buy |
Token ID | 2822 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
gy | +0.009 |
bank | +0.007 |
lves | +0.007 |
zag | +0.006 |
sank | +0.006 |
Neg logit contributions | |
ey | -0.008 |
yogurt | -0.007 |
honey | -0.007 |
idges | -0.006 |
ad | -0.006 |
Token | another |
Token ID | 1194 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
gy | +0.147 |
bank | +0.126 |
lves | +0.099 |
neither | +0.094 |
station | +0.093 |
Neg logit contributions | |
ey | -0.161 |
idges | -0.130 |
ad | -0.125 |
ement | -0.125 |
importantly | -0.124 |
Token | one |
Token ID | 530 |
Feature activation | -9.11e+00 |
Pos logit contributions | |
gy | +0.215 |
bank | +0.173 |
lves | +0.152 |
drown | +0.142 |
zag | +0.142 |
Neg logit contributions | |
ey | -0.163 |
yogurt | -0.138 |
honey | -0.134 |
idges | -0.133 |
ad | -0.130 |
Token | L |
Token ID | 43 |
Feature activation | -3.75e+00 |
Pos logit contributions | |
ey | +0.015 |
idges | +0.013 |
ement | +0.013 |
yogurt | +0.013 |
ad | +0.013 |
Neg logit contributions | |
gy | -0.017 |
bank | -0.014 |
lves | -0.012 |
station | -0.011 |
zag | -0.011 |
Token | ily |
Token ID | 813 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
gy | +0.369 |
bank | +0.282 |
zag | +0.267 |
lves | +0.261 |
station | +0.255 |
Neg logit contributions | |
ey | -0.233 |
idges | -0.184 |
yogurt | -0.179 |
importantly | -0.179 |
creamy | -0.177 |
Token | and |
Token ID | 290 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
ey | +0.211 |
idges | +0.200 |
importantly | +0.197 |
ocol | +0.194 |
mmmm | +0.192 |
Neg logit contributions | |
gy | -0.174 |
bank | -0.138 |
station | -0.105 |
zag | -0.104 |
argue | -0.102 |
Token | Ben |
Token ID | 3932 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
gy | +0.020 |
bank | +0.016 |
station | +0.014 |
lves | +0.013 |
argue | +0.013 |
Neg logit contributions | |
ey | -0.014 |
idges | -0.012 |
ement | -0.012 |
importantly | -0.011 |
ad | -0.011 |
Token | look |
Token ID | 804 |
Feature activation | -4.16e-01 |
Pos logit contributions | |
ey | +0.180 |
idges | +0.165 |
importantly | +0.164 |
ummy | +0.157 |
ement | +0.155 |
Neg logit contributions | |
gy | -0.175 |
bank | -0.150 |
argue | -0.124 |
fight | -0.116 |
station | -0.114 |
Token | at |
Token ID | 379 |
Feature activation | -3.15e+00 |
Pos logit contributions | |
gy | +0.034 |
bank | +0.027 |
lves | +0.022 |
zag | +0.021 |
station | +0.021 |
Neg logit contributions | |
ey | -0.035 |
idges | -0.030 |
yogurt | -0.029 |
ad | -0.029 |
honey | -0.029 |
Token | each |
Token ID | 1123 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
gy | +0.281 |
bank | +0.227 |
lves | +0.190 |
station | +0.180 |
zag | +0.179 |
Neg logit contributions | |
ey | -0.245 |
idges | -0.207 |
yogurt | -0.203 |
ad | -0.200 |
honey | -0.198 |
Token | other |
Token ID | 584 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
gy | +0.278 |
bank | +0.221 |
lves | +0.198 |
zag | +0.185 |
drown | +0.184 |
Neg logit contributions | |
ey | -0.186 |
yogurt | -0.161 |
idges | -0.154 |
honey | -0.154 |
ad | -0.151 |
Token | . |
Token ID | 13 |
Feature activation | +4.27e-01 |
Pos logit contributions | |
gy | +0.132 |
bank | +0.107 |
lves | +0.094 |
zag | +0.089 |
drown | +0.087 |
Neg logit contributions | |
ey | -0.099 |
yogurt | -0.083 |
idges | -0.082 |
honey | -0.081 |
ad | -0.081 |
Token | They |
Token ID | 1119 |
Feature activation | -3.70e+00 |
Pos logit contributions | |
ey | +0.043 |
idges | +0.038 |
ad | +0.037 |
yogurt | +0.037 |
honey | +0.036 |
Neg logit contributions | |
gy | -0.028 |
bank | -0.021 |
lves | -0.016 |
station | -0.015 |
zag | -0.014 |
Token | feel |
Token ID | 1254 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
gy | +0.264 |
bank | +0.218 |
argue | +0.175 |
fight | +0.167 |
drown | +0.162 |
Neg logit contributions | |
ey | -0.322 |
idges | -0.298 |
ement | -0.293 |
ummy | -0.292 |
importantly | -0.290 |
Token | to |
Token ID | 284 |
Feature activation | +4.62e-01 |
Pos logit contributions | |
gy | +0.084 |
bank | +0.066 |
lves | +0.052 |
zag | +0.052 |
drown | +0.050 |
Neg logit contributions | |
ey | -0.089 |
idges | -0.077 |
yogurt | -0.074 |
ad | -0.074 |
ement | -0.074 |
Token | land |
Token ID | 1956 |
Feature activation | -4.68e+00 |
Pos logit contributions | |
ey | +0.042 |
idges | +0.036 |
yogurt | +0.035 |
ad | +0.035 |
honey | +0.035 |
Neg logit contributions | |
gy | -0.035 |
bank | -0.027 |
lves | -0.022 |
zag | -0.021 |
station | -0.020 |
Token | but |
Token ID | 475 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
gy | +0.380 |
bank | +0.295 |
zag | +0.235 |
neither | +0.228 |
lves | +0.223 |
Neg logit contributions | |
ey | -0.407 |
idges | -0.349 |
importantly | -0.335 |
yogurt | -0.335 |
ummy | -0.327 |
Token | he |
Token ID | 339 |
Feature activation | +8.60e-01 |
Pos logit contributions | |
gy | +0.169 |
bank | +0.142 |
lves | +0.127 |
station | +0.121 |
zag | +0.120 |
Neg logit contributions | |
ey | -0.086 |
yogurt | -0.069 |
honey | -0.067 |
idges | -0.067 |
ad | -0.066 |
Token | couldn |
Token ID | 3521 |
Feature activation | -2.55e-02 |
Pos logit contributions | |
ey | +0.064 |
ad | +0.055 |
idges | +0.055 |
yogurt | +0.055 |
honey | +0.054 |
Neg logit contributions | |
gy | -0.075 |
bank | -0.061 |
lves | -0.052 |
station | -0.050 |
argue | -0.048 |
Token | 't |
Token ID | 470 |
Feature activation | -4.66e+00 |
Pos logit contributions | |
gy | +0.002 |
bank | +0.002 |
lves | +0.001 |
station | +0.001 |
zag | +0.001 |
Neg logit contributions | |
ey | -0.002 |
idges | -0.002 |
ad | -0.002 |
yogurt | -0.002 |
ement | -0.002 |
Token | reach |
Token ID | 3151 |
Feature activation | -4.89e-01 |
Pos logit contributions | |
gy | +0.397 |
bank | +0.317 |
lves | +0.279 |
station | +0.258 |
zag | +0.249 |
Neg logit contributions | |
ey | -0.355 |
ad | -0.324 |
idges | -0.303 |
honey | -0.303 |
yogurt | -0.299 |
Token | it |
Token ID | 340 |
Feature activation | -8.46e-02 |
Pos logit contributions | |
gy | +0.045 |
bank | +0.037 |
lves | +0.031 |
drown | +0.030 |
zag | +0.030 |
Neg logit contributions | |
ey | -0.037 |
idges | -0.030 |
yogurt | -0.029 |
ad | -0.029 |
ement | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
gy | +0.008 |
bank | +0.007 |
lves | +0.006 |
station | +0.005 |
drown | +0.005 |
Neg logit contributions | |
ey | -0.006 |
ad | -0.005 |
yogurt | -0.005 |
honey | -0.005 |
idges | -0.005 |
Token | He |
Token ID | 679 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
ey | +0.008 |
idges | +0.006 |
yogurt | +0.006 |
ad | +0.006 |
honey | +0.006 |
Neg logit contributions | |
gy | -0.018 |
bank | -0.016 |
lves | -0.014 |
zag | -0.013 |
station | -0.013 |
Token | got |
Token ID | 1392 |
Feature activation | -5.27e+00 |
Pos logit contributions | |
ey | +0.194 |
ad | +0.173 |
honey | +0.169 |
yogurt | +0.165 |
idges | +0.161 |
Neg logit contributions | |
gy | -0.245 |
bank | -0.192 |
lves | -0.171 |
station | -0.167 |
drown | -0.165 |
Token | horse |
Token ID | 8223 |
Feature activation | -4.56e+00 |
Pos logit contributions | |
gy | +0.017 |
bank | +0.013 |
lves | +0.011 |
station | +0.010 |
zag | +0.010 |
Neg logit contributions | |
ey | -0.019 |
idges | -0.016 |
ad | -0.016 |
yogurt | -0.016 |
ement | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
gy | +0.476 |
lves | +0.408 |
bank | +0.393 |
ene | +0.356 |
zag | +0.355 |
Neg logit contributions | |
honey | -0.192 |
yogurt | -0.186 |
sweet | -0.174 |
ey | -0.171 |
ad | -0.164 |
Token | She |
Token ID | 1375 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
ey | +0.015 |
idges | +0.014 |
yogurt | +0.012 |
ad | +0.012 |
ement | +0.012 |
Neg logit contributions | |
gy | -0.022 |
bank | -0.018 |
lves | -0.015 |
neither | -0.015 |
station | -0.015 |
Token | did |
Token ID | 750 |
Feature activation | +2.89e+00 |
Pos logit contributions | |
ey | +0.122 |
yogurt | +0.103 |
idges | +0.102 |
ad | +0.102 |
honey | +0.101 |
Neg logit contributions | |
gy | -0.139 |
bank | -0.110 |
lves | -0.094 |
zag | -0.091 |
drown | -0.089 |
Token | not |
Token ID | 407 |
Feature activation | -3.01e+00 |
Pos logit contributions | |
ey | +0.192 |
yogurt | +0.170 |
idges | +0.169 |
honey | +0.168 |
sweet | +0.167 |
Neg logit contributions | |
gy | -0.276 |
bank | -0.228 |
lves | -0.200 |
station | -0.189 |
zag | -0.178 |
Token | want |
Token ID | 765 |
Feature activation | -3.10e+00 |
Pos logit contributions | |
gy | +0.265 |
bank | +0.217 |
lves | +0.188 |
zag | +0.174 |
station | +0.171 |
Neg logit contributions | |
ey | -0.228 |
ad | -0.200 |
yogurt | -0.194 |
honey | -0.194 |
idges | -0.193 |
Token | to |
Token ID | 284 |
Feature activation | +5.60e-01 |
Pos logit contributions | |
gy | +0.237 |
bank | +0.190 |
drown | +0.145 |
station | +0.144 |
lves | +0.144 |
Neg logit contributions | |
ey | -0.279 |
idges | -0.237 |
ement | -0.233 |
ad | -0.223 |
importantly | -0.223 |
Token | be |
Token ID | 307 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
ey | +0.042 |
ad | +0.039 |
yogurt | +0.037 |
honey | +0.036 |
idges | +0.036 |
Neg logit contributions | |
gy | -0.047 |
bank | -0.037 |
lves | -0.035 |
zag | -0.031 |
station | -0.030 |
Token | punished |
Token ID | 16851 |
Feature activation | -4.17e+00 |
Pos logit contributions | |
ey | +0.016 |
sweet | +0.016 |
honey | +0.015 |
yogurt | +0.015 |
ad | +0.015 |
Neg logit contributions | |
gy | -0.023 |
bank | -0.019 |
lves | -0.016 |
zag | -0.016 |
station | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
gy | +0.383 |
bank | +0.312 |
lves | +0.266 |
zag | +0.252 |
station | +0.249 |
Neg logit contributions | |
ey | -0.308 |
idges | -0.258 |
yogurt | -0.255 |
ad | -0.252 |
honey | -0.249 |
Token | She |
Token ID | 1375 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
ey | +0.009 |
idges | +0.008 |
ad | +0.007 |
yogurt | +0.007 |
ement | +0.007 |
Neg logit contributions | |
gy | -0.012 |
bank | -0.010 |
lves | -0.008 |
neither | -0.008 |
drown | -0.008 |
Token | that |
Token ID | 326 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
gy | +0.076 |
bank | +0.064 |
neither | +0.053 |
crashed | +0.052 |
station | +0.052 |
Neg logit contributions | |
ey | -0.076 |
idges | -0.060 |
ad | -0.058 |
ement | -0.057 |
ango | -0.056 |
Token | the |
Token ID | 262 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
gy | +0.106 |
bank | +0.089 |
station | +0.072 |
lves | +0.072 |
neither | +0.071 |
Neg logit contributions | |
ey | -0.091 |
idges | -0.076 |
ement | -0.074 |
ad | -0.072 |
ummy | -0.070 |
Token | people |
Token ID | 661 |
Feature activation | +5.81e-01 |
Pos logit contributions | |
gy | +0.012 |
bank | +0.008 |
lves | +0.007 |
zag | +0.006 |
drown | +0.006 |
Neg logit contributions | |
ey | -0.019 |
ement | -0.017 |
ad | -0.017 |
idges | -0.016 |
sweet | -0.016 |
Token | in |
Token ID | 287 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
ey | +0.047 |
idges | +0.040 |
yogurt | +0.039 |
ad | +0.039 |
honey | +0.038 |
Neg logit contributions | |
gy | -0.049 |
bank | -0.039 |
lves | -0.033 |
zag | -0.031 |
station | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
ey | +0.080 |
idges | +0.067 |
yogurt | +0.065 |
ad | +0.064 |
honey | +0.063 |
Neg logit contributions | |
gy | -0.107 |
bank | -0.088 |
lves | -0.075 |
station | -0.071 |
zag | -0.071 |
Token | jet |
Token ID | 12644 |
Feature activation | -1.13e+01 |
Pos logit contributions | |
gy | +0.019 |
bank | +0.015 |
lves | +0.013 |
zag | +0.013 |
drown | +0.013 |
Neg logit contributions | |
ey | -0.013 |
yogurt | -0.010 |
ad | -0.010 |
idges | -0.010 |
honey | -0.010 |
Token | must |
Token ID | 1276 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
gy | +0.975 |
bank | +0.775 |
lves | +0.691 |
drown | +0.635 |
zag | +0.623 |
Neg logit contributions | |
ey | -0.841 |
idges | -0.725 |
ad | -0.725 |
yogurt | -0.716 |
honey | -0.711 |
Token | feel |
Token ID | 1254 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
gy | +0.051 |
bank | +0.040 |
lves | +0.033 |
zag | +0.031 |
station | +0.030 |
Neg logit contributions | |
ey | -0.054 |
idges | -0.047 |
yogurt | -0.046 |
ad | -0.046 |
honey | -0.045 |
Token | proud |
Token ID | 6613 |
Feature activation | +6.82e-01 |
Pos logit contributions | |
gy | +0.177 |
bank | +0.150 |
lves | +0.119 |
drown | +0.119 |
neither | +0.113 |
Neg logit contributions | |
ey | -0.198 |
idges | -0.169 |
ad | -0.163 |
ango | -0.161 |
ement | -0.159 |
Token | too |
Token ID | 1165 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
ey | +0.062 |
idges | +0.051 |
mmmm | +0.049 |
ango | +0.049 |
ad | +0.048 |
Neg logit contributions | |
gy | -0.046 |
bank | -0.038 |
neither | -0.032 |
station | -0.031 |
fight | -0.029 |
Token | because |
Token ID | 780 |
Feature activation | -6.68e+00 |
Pos logit contributions | |
gy | +0.095 |
bank | +0.077 |
lves | +0.065 |
zag | +0.062 |
station | +0.061 |
Neg logit contributions | |
ey | -0.079 |
idges | -0.066 |
yogurt | -0.065 |
ad | -0.064 |
honey | -0.063 |
Token | rabbit |
Token ID | 22746 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
ey | +0.007 |
yogurt | +0.006 |
idges | +0.006 |
ad | +0.005 |
honey | +0.005 |
Neg logit contributions | |
gy | -0.007 |
bank | -0.005 |
lves | -0.004 |
zag | -0.004 |
drown | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
gy | +0.199 |
bank | +0.163 |
lves | +0.141 |
station | +0.133 |
zag | +0.133 |
Neg logit contributions | |
ey | -0.153 |
idges | -0.128 |
yogurt | -0.127 |
ad | -0.126 |
honey | -0.125 |
Token | The |
Token ID | 383 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
ey | +0.006 |
yogurt | +0.005 |
ad | +0.005 |
idges | +0.005 |
honey | +0.005 |
Neg logit contributions | |
gy | -0.010 |
bank | -0.009 |
lves | -0.008 |
zag | -0.007 |
drown | -0.007 |
Token | rh |
Token ID | 9529 |
Feature activation | -6.23e+00 |
Pos logit contributions | |
gy | +0.215 |
bank | +0.169 |
lves | +0.142 |
zag | +0.137 |
drown | +0.134 |
Neg logit contributions | |
ey | -0.209 |
yogurt | -0.176 |
idges | -0.175 |
honey | -0.174 |
ad | -0.174 |
Token | in |
Token ID | 259 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
gy | +0.720 |
bank | +0.613 |
lves | +0.544 |
zag | +0.524 |
station | +0.520 |
Neg logit contributions | |
ey | -0.315 |
idges | -0.238 |
yogurt | -0.232 |
ad | -0.225 |
honey | -0.223 |
Token | oc |
Token ID | 420 |
Feature activation | -1.08e+01 |
Pos logit contributions | |
ey | +0.231 |
ad | +0.206 |
idges | +0.203 |
honey | +0.201 |
yogurt | +0.200 |
Neg logit contributions | |
gy | -0.154 |
bank | -0.118 |
lves | -0.095 |
drown | -0.094 |
argue | -0.087 |
Token | eros |
Token ID | 27498 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
gy | +0.932 |
bank | +0.736 |
lves | +0.597 |
zag | +0.554 |
station | +0.532 |
Neg logit contributions | |
ey | -0.900 |
idges | -0.768 |
yogurt | -0.763 |
importantly | -0.747 |
honey | -0.737 |
Token | explained |
Token ID | 4893 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
gy | +0.242 |
bank | +0.199 |
lves | +0.173 |
drown | +0.164 |
zag | +0.164 |
Neg logit contributions | |
ey | -0.168 |
idges | -0.140 |
yogurt | -0.138 |
ad | -0.137 |
honey | -0.134 |
Token | his |
Token ID | 465 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
gy | +0.152 |
bank | +0.123 |
lves | +0.097 |
neither | +0.096 |
argue | +0.094 |
Neg logit contributions | |
ey | -0.152 |
idges | -0.124 |
yogurt | -0.120 |
importantly | -0.117 |
ement | -0.116 |
Token | problem |
Token ID | 1917 |
Feature activation | -4.69e+00 |
Pos logit contributions | |
ey | +0.078 |
idges | +0.067 |
yogurt | +0.063 |
ad | +0.062 |
ement | +0.061 |
Neg logit contributions | |
gy | -0.103 |
bank | -0.085 |
zag | -0.069 |
lves | -0.069 |
drown | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
gy | +0.408 |
bank | +0.326 |
zag | +0.268 |
station | +0.257 |
neither | +0.254 |
Neg logit contributions | |
ey | -0.377 |
idges | -0.311 |
yogurt | -0.300 |
ement | -0.297 |
honey | -0.296 |
Token |
|
Token ID | 198 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
ey | +0.012 |
idges | +0.010 |
ement | +0.010 |
yogurt | +0.010 |
honey | +0.010 |
Neg logit contributions | |
gy | -0.013 |
bank | -0.011 |
lves | -0.009 |
zag | -0.009 |
drown | -0.008 |
Token | She |
Token ID | 3347 |
Feature activation | +2.97e+00 |
Pos logit contributions | |
ey | +0.009 |
idges | +0.007 |
yogurt | +0.007 |
ad | +0.007 |
honey | +0.006 |
Neg logit contributions | |
gy | -0.018 |
bank | -0.015 |
lves | -0.013 |
zag | -0.013 |
station | -0.013 |
Token | pushed |
Token ID | 7121 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
ey | +0.205 |
yogurt | +0.183 |
ad | +0.177 |
honey | +0.167 |
idges | +0.166 |
Neg logit contributions | |
gy | -0.275 |
bank | -0.210 |
lves | -0.195 |
drown | -0.189 |
zag | -0.179 |
Token | the |
Token ID | 262 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
gy | +0.228 |
bank | +0.189 |
lves | +0.163 |
zag | +0.156 |
station | +0.155 |
Neg logit contributions | |
ey | -0.148 |
idges | -0.121 |
yogurt | -0.119 |
ad | -0.117 |
honey | -0.116 |
Token | red |
Token ID | 2266 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
gy | +0.020 |
bank | +0.014 |
lves | +0.013 |
drown | +0.012 |
drowned | +0.011 |
Neg logit contributions | |
ey | -0.023 |
honey | -0.019 |
sweet | -0.019 |
yogurt | -0.019 |
ad | -0.019 |
Token | train |
Token ID | 4512 |
Feature activation | -9.83e+00 |
Pos logit contributions | |
ey | +0.099 |
ad | +0.089 |
yogurt | +0.086 |
honey | +0.085 |
idges | +0.083 |
Neg logit contributions | |
gy | -0.068 |
bank | -0.047 |
lves | -0.046 |
drown | -0.044 |
drowned | -0.041 |
Token | and |
Token ID | 290 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
gy | +0.920 |
lves | +0.756 |
bank | +0.701 |
drown | +0.664 |
drowned | +0.654 |
Neg logit contributions | |
ey | -0.576 |
ad | -0.528 |
honey | -0.516 |
sweet | -0.494 |
yogurt | -0.488 |
Token | made |
Token ID | 925 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
gy | +0.023 |
bank | +0.018 |
lves | +0.016 |
zag | +0.015 |
drown | +0.015 |
Neg logit contributions | |
ey | -0.017 |
yogurt | -0.015 |
honey | -0.015 |
sweet | -0.014 |
ad | -0.014 |
Token | it |
Token ID | 340 |
Feature activation | -9.78e-03 |
Pos logit contributions | |
gy | +0.127 |
bank | +0.105 |
lves | +0.087 |
station | +0.084 |
drown | +0.083 |
Neg logit contributions | |
ey | -0.107 |
idges | -0.089 |
ement | -0.085 |
yogurt | -0.084 |
ummy | -0.083 |
Token | go |
Token ID | 467 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
gy | +0.001 |
bank | +0.001 |
lves | +0.001 |
station | +0.001 |
drowned | +0.001 |
Neg logit contributions | |
ey | -0.001 |
honey | -0.001 |
yogurt | -0.001 |
ad | -0.001 |
sweet | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
sweet | +0.008 |
ad | +0.008 |
ey | +0.007 |
himself | +0.007 |
honey | +0.007 |
Neg logit contributions | |
gy | -0.017 |
bank | -0.014 |
lves | -0.014 |
drown | -0.013 |
drowned | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | -3.72e-01 |
Pos logit contributions | |
gy | +0.134 |
bank | +0.106 |
lves | +0.087 |
station | +0.082 |
zag | +0.082 |
Neg logit contributions | |
ey | -0.129 |
idges | -0.110 |
yogurt | -0.106 |
ad | -0.104 |
ement | -0.102 |
Token | bark |
Token ID | 21405 |
Feature activation | -3.29e+00 |
Pos logit contributions | |
gy | +0.034 |
bank | +0.028 |
lves | +0.025 |
station | +0.023 |
zag | +0.023 |
Neg logit contributions | |
ey | -0.026 |
honey | -0.022 |
yogurt | -0.022 |
ad | -0.021 |
idges | -0.021 |
Token | ed |
Token ID | 276 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
gy | +0.301 |
bank | +0.247 |
lves | +0.213 |
station | +0.203 |
drown | +0.202 |
Neg logit contributions | |
ey | -0.227 |
idges | -0.196 |
ad | -0.195 |
yogurt | -0.193 |
honey | -0.192 |
Token | , |
Token ID | 11 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
gy | +0.145 |
bank | +0.116 |
lves | +0.095 |
zag | +0.090 |
station | +0.089 |
Neg logit contributions | |
ey | -0.138 |
idges | -0.116 |
yogurt | -0.113 |
ad | -0.112 |
ement | -0.109 |
Token | but |
Token ID | 475 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
ey | +0.026 |
idges | +0.022 |
yogurt | +0.022 |
ad | +0.021 |
ement | +0.021 |
Neg logit contributions | |
gy | -0.033 |
bank | -0.027 |
lves | -0.022 |
zag | -0.021 |
station | -0.021 |
Token | Wh |
Token ID | 854 |
Feature activation | -9.62e+00 |
Pos logit contributions | |
gy | +0.151 |
bank | +0.128 |
lves | +0.112 |
zag | +0.106 |
drown | +0.105 |
Neg logit contributions | |
ey | -0.091 |
yogurt | -0.078 |
honey | -0.077 |
idges | -0.076 |
ad | -0.074 |
Token | isk |
Token ID | 1984 |
Feature activation | -8.01e+00 |
Pos logit contributions | |
gy | +0.601 |
bank | +0.406 |
lves | +0.300 |
zag | +0.287 |
station | +0.270 |
Neg logit contributions | |
ey | -1.032 |
yogurt | -0.906 |
idges | -0.905 |
honey | -0.887 |
ad | -0.877 |
Token | ers |
Token ID | 364 |
Feature activation | +3.63e+00 |
Pos logit contributions | |
gy | +0.855 |
bank | +0.689 |
zag | +0.607 |
lves | +0.601 |
station | +0.567 |
Neg logit contributions | |
ey | -0.528 |
yogurt | -0.404 |
ummy | -0.390 |
importantly | -0.384 |
idges | -0.383 |
Token | did |
Token ID | 750 |
Feature activation | +2.93e+00 |
Pos logit contributions | |
ey | +0.234 |
ad | +0.212 |
idges | +0.206 |
yogurt | +0.206 |
honey | +0.198 |
Neg logit contributions | |
gy | -0.334 |
bank | -0.258 |
lves | -0.241 |
drown | -0.223 |
zag | -0.223 |
Token | not |
Token ID | 407 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
ey | +0.243 |
idges | +0.213 |
yogurt | +0.210 |
ad | +0.208 |
honey | +0.206 |
Neg logit contributions | |
gy | -0.239 |
bank | -0.189 |
lves | -0.160 |
station | -0.147 |
zag | -0.145 |
Token | care |
Token ID | 1337 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
gy | +0.278 |
bank | +0.227 |
lves | +0.194 |
zag | +0.181 |
station | +0.180 |
Neg logit contributions | |
ey | -0.227 |
idges | -0.194 |
ad | -0.193 |
yogurt | -0.192 |
honey | -0.192 |
Token | , |
Token ID | 11 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
gy | +0.014 |
bank | +0.011 |
lves | +0.010 |
zag | +0.009 |
station | +0.009 |
Neg logit contributions | |
ey | -0.012 |
idges | -0.010 |
yogurt | -0.010 |
ad | -0.010 |
honey | -0.009 |
Token | she |
Token ID | 673 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
ey | +0.012 |
idges | +0.010 |
yogurt | +0.009 |
ad | +0.009 |
ement | +0.009 |
Neg logit contributions | |
gy | -0.014 |
bank | -0.012 |
lves | -0.010 |
station | -0.010 |
neither | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
gy | +0.040 |
bank | +0.032 |
lves | +0.026 |
zag | +0.025 |
station | +0.024 |
Neg logit contributions | |
ey | -0.043 |
idges | -0.036 |
yogurt | -0.036 |
ad | -0.035 |
honey | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
ey | +0.083 |
idges | +0.070 |
yogurt | +0.067 |
ad | +0.067 |
ement | +0.066 |
Neg logit contributions | |
gy | -0.085 |
bank | -0.067 |
lves | -0.055 |
zag | -0.053 |
drown | -0.053 |
Token | only |
Token ID | 691 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
ey | +0.002 |
idges | +0.002 |
yogurt | +0.002 |
ad | +0.002 |
honey | +0.002 |
Neg logit contributions | |
gy | -0.002 |
bank | -0.002 |
lves | -0.001 |
station | -0.001 |
zag | -0.001 |
Token | one |
Token ID | 530 |
Feature activation | -9.32e+00 |
Pos logit contributions | |
gy | +0.208 |
bank | +0.162 |
lves | +0.131 |
zag | +0.125 |
station | +0.124 |
Neg logit contributions | |
ey | -0.232 |
idges | -0.201 |
yogurt | -0.196 |
ad | -0.194 |
honey | -0.191 |
Token | rocking |
Token ID | 38059 |
Feature activation | +5.66e+00 |
Pos logit contributions | |
gy | +0.814 |
bank | +0.666 |
lves | +0.555 |
station | +0.543 |
zag | +0.541 |
Neg logit contributions | |
ey | -0.732 |
idges | -0.606 |
ad | -0.585 |
yogurt | -0.578 |
honey | -0.566 |
Token | . |
Token ID | 13 |
Feature activation | +9.29e-02 |
Pos logit contributions | |
ey | +0.457 |
idges | +0.388 |
yogurt | +0.384 |
ad | +0.379 |
honey | +0.374 |
Neg logit contributions | |
gy | -0.479 |
bank | -0.382 |
lves | -0.321 |
zag | -0.299 |
drown | -0.297 |
Token | Everyone |
Token ID | 11075 |
Feature activation | -5.53e-01 |
Pos logit contributions | |
ey | +0.007 |
idges | +0.006 |
honey | +0.006 |
yogurt | +0.006 |
ad | +0.006 |
Neg logit contributions | |
gy | -0.008 |
bank | -0.006 |
lves | -0.005 |
station | -0.005 |
zag | -0.005 |
Token | around |
Token ID | 1088 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
gy | +0.047 |
bank | +0.039 |
lves | +0.032 |
station | +0.031 |
zag | +0.030 |
Neg logit contributions | |
ey | -0.044 |
idges | -0.037 |
yogurt | -0.036 |
honey | -0.036 |
ad | -0.035 |
Token | her |
Token ID | 607 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
gy | +0.218 |
bank | +0.177 |
lves | +0.148 |
station | +0.141 |
zag | +0.141 |
Neg logit contributions | |
ey | -0.186 |
idges | -0.157 |
yogurt | -0.153 |
ad | -0.150 |
honey | -0.149 |