Token | got |
Token ID | 1392 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
joy | +0.123 |
acts | +0.120 |
ones | +0.119 |
folk | +0.118 |
argued | +0.117 |
Neg logit contributions | |
paste | -0.178 |
prise | -0.174 |
web | -0.171 |
iders | -0.163 |
izz | -0.156 |
Token | four |
Token ID | 1440 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
paste | +0.074 |
web | +0.072 |
prise | +0.069 |
iders | +0.069 |
aid | +0.063 |
Neg logit contributions | |
ones | -0.049 |
joy | -0.047 |
wanted | -0.046 |
yacht | -0.046 |
challenged | -0.046 |
Token | sides |
Token ID | 5389 |
Feature activation | +3.61e-03 |
Pos logit contributions | |
joy | +0.030 |
folk | +0.029 |
acts | +0.029 |
argued | +0.028 |
ones | +0.028 |
Neg logit contributions | |
paste | -0.043 |
prise | -0.043 |
web | -0.042 |
iders | -0.040 |
izz | -0.038 |
Token | ?" |
Token ID | 1701 |
Feature activation | +3.31e-03 |
Pos logit contributions | |
joy | +0.007 |
folk | +0.006 |
acts | +0.006 |
argued | +0.006 |
ones | +0.006 |
Neg logit contributions | |
paste | -0.006 |
prise | -0.006 |
web | -0.006 |
iders | -0.005 |
izz | -0.005 |
Token | Joe |
Token ID | 5689 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
joy | +0.005 |
folk | +0.005 |
acts | +0.005 |
argued | +0.004 |
uniforms | +0.004 |
Neg logit contributions | |
prise | -0.006 |
web | -0.006 |
paste | -0.006 |
iders | -0.006 |
izz | -0.006 |
Token | nodded |
Token ID | 14464 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
joy | +0.014 |
folk | +0.013 |
country | +0.012 |
ones | +0.012 |
yacht | +0.012 |
Neg logit contributions | |
web | -0.033 |
paste | -0.033 |
prise | -0.032 |
iders | -0.030 |
aid | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
folk | +0.321 |
acts | +0.317 |
seals | +0.305 |
goats | +0.302 |
clubs | +0.297 |
Neg logit contributions | |
paste | -0.389 |
web | -0.384 |
prise | -0.367 |
izz | -0.343 |
ores | -0.333 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
joy | +0.041 |
folk | +0.039 |
acts | +0.037 |
uniforms | +0.037 |
ones | +0.035 |
Neg logit contributions | |
paste | -0.064 |
prise | -0.063 |
web | -0.063 |
iders | -0.057 |
aid | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
folk | +0.234 |
acts | +0.230 |
argued | +0.219 |
uniforms | +0.213 |
wandered | +0.212 |
Neg logit contributions | |
paste | -0.259 |
web | -0.249 |
prise | -0.243 |
ores | -0.217 |
aid | -0.215 |
Token |
|
Token ID | 198 |
Feature activation | -2.68e-03 |
Pos logit contributions | |
joy | +0.036 |
acts | +0.036 |
folk | +0.036 |
argued | +0.036 |
uniforms | +0.035 |
Neg logit contributions | |
paste | -0.061 |
web | -0.061 |
prise | -0.061 |
iders | -0.058 |
izz | -0.057 |
Token |
|
Token ID | 198 |
Feature activation | -4.02e-03 |
Pos logit contributions | |
paste | +0.007 |
web | +0.007 |
prise | +0.007 |
iders | +0.006 |
potion | +0.006 |
Neg logit contributions | |
joy | -0.002 |
folk | -0.002 |
challenged | -0.002 |
walked | -0.002 |
wandered | -0.002 |
Token | my |
Token ID | 616 |
Feature activation | -4.24e-03 |
Pos logit contributions | |
joy | +0.074 |
folk | +0.071 |
acts | +0.071 |
argued | +0.068 |
ones | +0.067 |
Neg logit contributions | |
prise | -0.106 |
paste | -0.106 |
web | -0.105 |
iders | -0.097 |
izz | -0.094 |
Token | muscles |
Token ID | 12749 |
Feature activation | +7.79e-02 |
Pos logit contributions | |
paste | +0.008 |
web | +0.008 |
prise | +0.008 |
iders | +0.008 |
potion | +0.007 |
Neg logit contributions | |
joy | -0.006 |
folk | -0.006 |
argued | -0.006 |
acts | -0.005 |
ened | -0.005 |
Token | !" |
Token ID | 2474 |
Feature activation | +3.17e-02 |
Pos logit contributions | |
joy | +0.103 |
folk | +0.100 |
acts | +0.099 |
argued | +0.097 |
ones | +0.094 |
Neg logit contributions | |
prise | -0.158 |
paste | -0.157 |
web | -0.156 |
iders | -0.145 |
izz | -0.141 |
Token | His |
Token ID | 2399 |
Feature activation | +6.25e-02 |
Pos logit contributions | |
folk | +0.034 |
acts | +0.032 |
hi | +0.031 |
argued | +0.031 |
joy | +0.030 |
Neg logit contributions | |
paste | -0.069 |
prise | -0.063 |
web | -0.061 |
izz | -0.061 |
aid | -0.061 |
Token | dad |
Token ID | 9955 |
Feature activation | +4.15e-02 |
Pos logit contributions | |
folk | +0.129 |
hi | +0.116 |
travelled | +0.116 |
traveled | +0.115 |
ones | +0.114 |
Neg logit contributions | |
paste | -0.076 |
prise | -0.074 |
web | -0.070 |
x | -0.069 |
iders | -0.060 |
Token | nodded |
Token ID | 14464 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
joy | +0.035 |
folk | +0.032 |
acts | +0.032 |
argued | +0.030 |
ones | +0.029 |
Neg logit contributions | |
paste | -0.105 |
web | -0.105 |
prise | -0.104 |
iders | -0.098 |
izz | -0.095 |
Token | and |
Token ID | 290 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
folk | +0.300 |
acts | +0.298 |
seals | +0.291 |
goats | +0.285 |
stalls | +0.285 |
Neg logit contributions | |
paste | -0.404 |
web | -0.391 |
prise | -0.384 |
aid | -0.347 |
izz | -0.346 |
Token | lied |
Token ID | 19837 |
Feature activation | +7.99e-03 |
Pos logit contributions | |
joy | +0.033 |
folk | +0.032 |
acts | +0.030 |
uniforms | +0.029 |
ones | +0.029 |
Neg logit contributions | |
paste | -0.061 |
prise | -0.061 |
web | -0.061 |
iders | -0.055 |
aid | -0.055 |
Token | down |
Token ID | 866 |
Feature activation | +7.56e-02 |
Pos logit contributions | |
folk | +0.011 |
acts | +0.011 |
joy | +0.011 |
goats | +0.010 |
country | +0.010 |
Neg logit contributions | |
paste | -0.016 |
web | -0.015 |
prise | -0.015 |
iders | -0.014 |
aid | -0.013 |
Token | on |
Token ID | 319 |
Feature activation | +7.32e-02 |
Pos logit contributions | |
folk | +0.113 |
hi | +0.112 |
argued | +0.112 |
acts | +0.112 |
joy | +0.109 |
Neg logit contributions | |
paste | -0.135 |
web | -0.131 |
prise | -0.124 |
iders | -0.117 |
ores | -0.113 |
Token | the |
Token ID | 262 |
Feature activation | +4.17e-02 |
Pos logit contributions | |
folk | +0.098 |
acts | +0.094 |
argued | +0.093 |
hi | +0.093 |
joy | +0.092 |
Neg logit contributions | |
paste | -0.143 |
web | -0.139 |
prise | -0.139 |
iders | -0.128 |
potion | -0.126 |
Token | things |
Token ID | 1243 |
Feature activation | +4.12e-02 |
Pos logit contributions | |
paste | +0.023 |
prise | +0.023 |
web | +0.023 |
iders | +0.021 |
aid | +0.019 |
Neg logit contributions | |
joy | -0.016 |
folk | -0.015 |
argued | -0.015 |
acts | -0.015 |
hi | -0.014 |
Token | with |
Token ID | 351 |
Feature activation | +6.55e-02 |
Pos logit contributions | |
joy | +0.063 |
folk | +0.062 |
acts | +0.061 |
ones | +0.059 |
argued | +0.059 |
Neg logit contributions | |
prise | -0.074 |
paste | -0.074 |
web | -0.074 |
iders | -0.068 |
izz | -0.067 |
Token | math |
Token ID | 10688 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
acts | +0.097 |
joy | +0.097 |
folk | +0.096 |
ones | +0.095 |
uniforms | +0.094 |
Neg logit contributions | |
paste | -0.125 |
web | -0.120 |
prise | -0.119 |
iders | -0.113 |
izz | -0.110 |
Token | !" |
Token ID | 2474 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
prise | +0.048 |
paste | +0.048 |
web | +0.048 |
iders | +0.044 |
izz | +0.043 |
Neg logit contributions | |
joy | -0.042 |
folk | -0.040 |
acts | -0.040 |
argued | -0.039 |
ones | -0.038 |
Token | Joe |
Token ID | 5689 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
joy | +0.044 |
acts | +0.042 |
folk | +0.042 |
argued | +0.041 |
uniforms | +0.041 |
Neg logit contributions | |
prise | -0.057 |
web | -0.057 |
paste | -0.056 |
iders | -0.053 |
izz | -0.051 |
Token | nodded |
Token ID | 14464 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
joy | +0.018 |
folk | +0.015 |
ones | +0.014 |
uniforms | +0.014 |
acts | +0.014 |
Neg logit contributions | |
web | -0.039 |
paste | -0.039 |
prise | -0.038 |
iders | -0.036 |
aid | -0.035 |
Token | and |
Token ID | 290 |
Feature activation | +3.01e-02 |
Pos logit contributions | |
folk | +0.277 |
acts | +0.277 |
joy | +0.259 |
goats | +0.259 |
stalls | +0.256 |
Neg logit contributions | |
paste | -0.425 |
web | -0.424 |
prise | -0.410 |
ores | -0.368 |
iders | -0.367 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
joy | +0.036 |
folk | +0.033 |
acts | +0.031 |
uniforms | +0.031 |
ones | +0.030 |
Neg logit contributions | |
paste | -0.066 |
web | -0.066 |
prise | -0.065 |
iders | -0.059 |
aid | -0.059 |
Token | , |
Token ID | 11 |
Feature activation | +4.41e-02 |
Pos logit contributions | |
folk | +0.214 |
acts | +0.213 |
argued | +0.206 |
joy | +0.197 |
wandered | +0.197 |
Neg logit contributions | |
paste | -0.268 |
web | -0.259 |
prise | -0.248 |
ores | -0.225 |
iders | -0.224 |
Token | he |
Token ID | 339 |
Feature activation | +4.95e-02 |
Pos logit contributions | |
acts | +0.062 |
folk | +0.060 |
uniforms | +0.059 |
ones | +0.057 |
joy | +0.057 |
Neg logit contributions | |
paste | -0.083 |
web | -0.082 |
prise | -0.077 |
aid | -0.073 |
iders | -0.071 |
Token | was |
Token ID | 373 |
Feature activation | +5.28e-02 |
Pos logit contributions | |
joy | +0.049 |
folk | +0.045 |
acts | +0.043 |
argued | +0.041 |
uniforms | +0.041 |
Neg logit contributions | |
paste | -0.120 |
web | -0.119 |
prise | -0.117 |
iders | -0.110 |
izz | -0.106 |
Token |
|
Token ID | 198 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
paste | +0.018 |
web | +0.018 |
prise | +0.018 |
iders | +0.016 |
izz | +0.016 |
Neg logit contributions | |
argued | -0.013 |
acts | -0.013 |
joy | -0.013 |
folk | -0.013 |
ened | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
paste | +0.030 |
web | +0.029 |
prise | +0.029 |
iders | +0.028 |
potion | +0.026 |
Neg logit contributions | |
joy | -0.008 |
| -0.007 |
folk | -0.006 |
walked | -0.006 |
nodded | -0.006 |
Token | A |
Token ID | 32 |
Feature activation | -6.52e-02 |
Pos logit contributions | |
paste | +0.022 |
web | +0.021 |
prise | +0.021 |
iders | +0.020 |
izz | +0.018 |
Neg logit contributions | |
joy | -0.020 |
folk | -0.019 |
acts | -0.018 |
argued | -0.018 |
hi | -0.018 |
Token | lic |
Token ID | 677 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
paste | +0.141 |
prise | +0.139 |
web | +0.139 |
iders | +0.131 |
izz | +0.125 |
Neg logit contributions | |
joy | -0.082 |
folk | -0.078 |
acts | -0.076 |
challenged | -0.072 |
argued | -0.072 |
Token | ia |
Token ID | 544 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
web | +0.060 |
paste | +0.057 |
prise | +0.056 |
iders | +0.054 |
izz | +0.053 |
Neg logit contributions | |
joy | -0.050 |
ones | -0.045 |
acts | -0.044 |
folk | -0.044 |
uniforms | -0.043 |
Token | nodded |
Token ID | 14464 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
joy | +0.027 |
folk | +0.024 |
acts | +0.023 |
ones | +0.023 |
attitude | +0.022 |
Neg logit contributions | |
web | -0.060 |
prise | -0.059 |
paste | -0.058 |
izz | -0.053 |
iders | -0.053 |
Token | and |
Token ID | 290 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
folk | +0.305 |
acts | +0.298 |
bank | +0.296 |
seals | +0.294 |
joy | +0.293 |
Neg logit contributions | |
web | -0.405 |
paste | -0.388 |
prise | -0.360 |
aid | -0.326 |
izz | -0.325 |
Token | said |
Token ID | 531 |
Feature activation | +5.81e-02 |
Pos logit contributions | |
joy | +0.031 |
acts | +0.030 |
folk | +0.030 |
attitude | +0.029 |
ones | +0.028 |
Neg logit contributions | |
paste | -0.046 |
web | -0.045 |
prise | -0.045 |
aid | -0.041 |
izz | -0.041 |
Token | , |
Token ID | 11 |
Feature activation | +4.01e-02 |
Pos logit contributions | |
joy | +0.072 |
folk | +0.072 |
argued | +0.071 |
acts | +0.070 |
uniforms | +0.070 |
Neg logit contributions | |
paste | -0.122 |
prise | -0.118 |
web | -0.116 |
iders | -0.106 |
aid | -0.105 |
Token | " |
Token ID | 366 |
Feature activation | +5.30e-02 |
Pos logit contributions | |
acts | +0.049 |
argued | +0.044 |
joy | +0.044 |
uniforms | +0.044 |
folk | +0.044 |
Neg logit contributions | |
paste | -0.087 |
web | -0.082 |
prise | -0.079 |
iders | -0.077 |
Chick | -0.075 |
Token | I |
Token ID | 40 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
joy | +0.087 |
acts | +0.085 |
argued | +0.083 |
ened | +0.083 |
folk | +0.083 |
Neg logit contributions | |
prise | -0.089 |
paste | -0.089 |
web | -0.088 |
iders | -0.080 |
aid | -0.079 |
Token | man |
Token ID | 582 |
Feature activation | +3.17e-02 |
Pos logit contributions | |
prise | +0.008 |
paste | +0.008 |
web | +0.008 |
aid | +0.007 |
x | +0.007 |
Neg logit contributions | |
joy | -0.005 |
wandered | -0.005 |
argued | -0.005 |
hi | -0.005 |
folk | -0.005 |
Token | some |
Token ID | 617 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
joy | +0.046 |
folk | +0.044 |
acts | +0.043 |
argued | +0.043 |
ened | +0.041 |
Neg logit contributions | |
paste | -0.061 |
prise | -0.061 |
web | -0.060 |
iders | -0.056 |
izz | -0.053 |
Token | advice |
Token ID | 5608 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
acts | +0.023 |
joy | +0.022 |
folk | +0.022 |
ones | +0.022 |
uniforms | +0.021 |
Neg logit contributions | |
web | -0.034 |
paste | -0.033 |
prise | -0.033 |
iders | -0.032 |
izz | -0.030 |
Token | ?". |
Token ID | 43634 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
joy | +0.178 |
folk | +0.171 |
acts | +0.169 |
argued | +0.166 |
uniforms | +0.160 |
Neg logit contributions | |
paste | -0.224 |
prise | -0.223 |
web | -0.223 |
iders | -0.206 |
izz | -0.196 |
Token | Tom |
Token ID | 4186 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
acts | +0.042 |
folk | +0.041 |
joy | +0.041 |
argued | +0.041 |
ones | +0.039 |
Neg logit contributions | |
paste | -0.064 |
prise | -0.063 |
web | -0.062 |
iders | -0.058 |
izz | -0.057 |
Token | nodded |
Token ID | 14464 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
joy | +0.018 |
folk | +0.017 |
ones | +0.016 |
acts | +0.015 |
uniforms | +0.015 |
Neg logit contributions | |
web | -0.035 |
paste | -0.035 |
prise | -0.034 |
iders | -0.032 |
izz | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
folk | +0.287 |
acts | +0.286 |
stalls | +0.267 |
seals | +0.266 |
joy | +0.265 |
Neg logit contributions | |
paste | -0.422 |
web | -0.415 |
prise | -0.402 |
izz | -0.370 |
aid | -0.366 |
Token | The |
Token ID | 383 |
Feature activation | +4.58e-02 |
Pos logit contributions | |
acts | +0.032 |
argued | +0.031 |
folk | +0.030 |
joy | +0.030 |
ened | +0.029 |
Neg logit contributions | |
paste | -0.064 |
prise | -0.062 |
web | -0.062 |
iders | -0.060 |
izz | -0.058 |
Token | old |
Token ID | 1468 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
argued | +0.089 |
folk | +0.083 |
joy | +0.082 |
ened | +0.081 |
wandered | +0.080 |
Neg logit contributions | |
web | -0.062 |
x | -0.060 |
paste | -0.058 |
prise | -0.058 |
aid | -0.055 |
Token | man |
Token ID | 582 |
Feature activation | +8.20e-02 |
Pos logit contributions | |
argued | +0.050 |
anced | +0.047 |
ened | +0.046 |
folk | +0.046 |
ashed | +0.046 |
Neg logit contributions | |
web | -0.034 |
x | -0.031 |
aid | -0.029 |
paste | -0.027 |
gen | -0.027 |
Token | said |
Token ID | 531 |
Feature activation | +5.67e-02 |
Pos logit contributions | |
joy | +0.092 |
folk | +0.082 |
ones | +0.080 |
acts | +0.080 |
uniforms | +0.078 |
Neg logit contributions | |
web | -0.187 |
paste | -0.187 |
prise | -0.181 |
aid | -0.168 |
iders | -0.165 |
Token | ? |
Token ID | 30 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
web | +0.012 |
paste | +0.012 |
prise | +0.012 |
iders | +0.011 |
izz | +0.010 |
Neg logit contributions | |
joy | -0.019 |
folk | -0.018 |
acts | -0.018 |
argued | -0.018 |
uniforms | -0.018 |
Token | Need |
Token ID | 10664 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
joy | +0.025 |
folk | +0.025 |
acts | +0.025 |
argued | +0.023 |
challenged | +0.023 |
Neg logit contributions | |
prise | -0.046 |
web | -0.046 |
paste | -0.046 |
iders | -0.043 |
aid | -0.041 |
Token | help |
Token ID | 1037 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
acts | +0.042 |
folk | +0.041 |
joy | +0.041 |
argued | +0.039 |
ened | +0.038 |
Neg logit contributions | |
prise | -0.062 |
paste | -0.062 |
web | -0.061 |
iders | -0.058 |
izz | -0.055 |
Token | ?" |
Token ID | 1701 |
Feature activation | +2.25e-03 |
Pos logit contributions | |
acts | +0.041 |
's | +0.040 |
wanted | +0.039 |
folk | +0.039 |
challenged | +0.038 |
Neg logit contributions | |
prise | -0.080 |
paste | -0.079 |
web | -0.075 |
iders | -0.071 |
aid | -0.071 |
Token | Joe |
Token ID | 5689 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
joy | +0.003 |
acts | +0.003 |
folk | +0.003 |
argued | +0.003 |
uniforms | +0.003 |
Neg logit contributions | |
web | -0.005 |
prise | -0.005 |
paste | -0.004 |
iders | -0.004 |
izz | -0.004 |
Token | nodded |
Token ID | 14464 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
joy | +0.012 |
folk | +0.010 |
ones | +0.010 |
acts | +0.010 |
uniforms | +0.009 |
Neg logit contributions | |
web | -0.025 |
paste | -0.025 |
prise | -0.024 |
iders | -0.023 |
izz | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | +4.40e-02 |
Pos logit contributions | |
folk | +0.307 |
acts | +0.299 |
joy | +0.282 |
seals | +0.282 |
stalls | +0.282 |
Neg logit contributions | |
paste | -0.410 |
web | -0.405 |
prise | -0.383 |
izz | -0.352 |
ores | -0.351 |
Token | so |
Token ID | 523 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
acts | +0.054 |
folk | +0.053 |
joy | +0.053 |
uniforms | +0.050 |
ones | +0.050 |
Neg logit contributions | |
paste | -0.092 |
web | -0.091 |
prise | -0.088 |
iders | -0.083 |
aid | -0.082 |
Token | the |
Token ID | 262 |
Feature activation | +7.88e-02 |
Pos logit contributions | |
joy | +0.027 |
acts | +0.026 |
folk | +0.026 |
argued | +0.025 |
uniforms | +0.024 |
Neg logit contributions | |
paste | -0.045 |
web | -0.045 |
prise | -0.045 |
iders | -0.042 |
izz | -0.040 |
Token | bird |
Token ID | 6512 |
Feature activation | +4.18e-02 |
Pos logit contributions | |
joy | +0.092 |
folk | +0.088 |
acts | +0.087 |
argued | +0.085 |
wandered | +0.084 |
Neg logit contributions | |
paste | -0.175 |
prise | -0.175 |
web | -0.170 |
iders | -0.167 |
izz | -0.160 |
Token | said |
Token ID | 531 |
Feature activation | +6.48e-02 |
Pos logit contributions | |
joy | +0.049 |
folk | +0.045 |
acts | +0.044 |
argued | +0.042 |
uniforms | +0.042 |
Neg logit contributions | |
web | -0.093 |
paste | -0.093 |
prise | -0.092 |
iders | -0.086 |
aid | -0.083 |
Token | you |
Token ID | 345 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
acts | +0.009 |
ones | +0.009 |
's | +0.009 |
joy | +0.009 |
one | +0.009 |
Neg logit contributions | |
paste | -0.014 |
web | -0.013 |
iders | -0.013 |
prise | -0.013 |
aid | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | -5.92e-02 |
Pos logit contributions | |
paste | +0.054 |
web | +0.054 |
prise | +0.053 |
iders | +0.051 |
izz | +0.048 |
Neg logit contributions | |
joy | -0.037 |
acts | -0.035 |
folk | -0.034 |
argued | -0.034 |
ones | -0.033 |
Token | sandwich |
Token ID | 20433 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
paste | +0.125 |
iders | +0.117 |
web | +0.117 |
izz | +0.115 |
prise | +0.113 |
Neg logit contributions | |
joy | -0.082 |
yacht | -0.078 |
country | -0.078 |
acts | -0.075 |
ones | -0.074 |
Token | ?". |
Token ID | 43634 |
Feature activation | +3.08e-02 |
Pos logit contributions | |
ones | +0.021 |
market | +0.019 |
yacht | +0.019 |
wanted | +0.018 |
joy | +0.018 |
Neg logit contributions | |
web | -0.036 |
prise | -0.036 |
paste | -0.036 |
izz | -0.035 |
iders | -0.035 |
Token | Jill |
Token ID | 22772 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
folk | +0.061 |
seals | +0.059 |
goats | +0.059 |
yacht | +0.059 |
acts | +0.057 |
Neg logit contributions | |
prise | -0.040 |
web | -0.039 |
paste | -0.038 |
iders | -0.037 |
izz | -0.035 |
Token | nodded |
Token ID | 14464 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
joy | +0.023 |
folk | +0.021 |
ones | +0.019 |
goats | +0.019 |
seals | +0.019 |
Neg logit contributions | |
web | -0.043 |
paste | -0.041 |
prise | -0.041 |
iders | -0.038 |
izz | -0.037 |
Token | with |
Token ID | 351 |
Feature activation | +8.59e-02 |
Pos logit contributions | |
folk | +0.274 |
acts | +0.263 |
joy | +0.253 |
seals | +0.248 |
argued | +0.245 |
Neg logit contributions | |
web | -0.447 |
paste | -0.445 |
prise | -0.436 |
izz | -0.400 |
aid | -0.395 |
Token | excitement |
Token ID | 14067 |
Feature activation | +8.41e-02 |
Pos logit contributions | |
folk | +0.133 |
joy | +0.132 |
argued | +0.128 |
acts | +0.126 |
challenged | +0.125 |
Neg logit contributions | |
prise | -0.154 |
paste | -0.149 |
web | -0.147 |
aid | -0.133 |
izz | -0.132 |
Token | and |
Token ID | 290 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
acts | +0.103 |
folk | +0.102 |
joy | +0.101 |
hi | +0.095 |
wandered | +0.095 |
Neg logit contributions | |
paste | -0.178 |
web | -0.175 |
prise | -0.168 |
iders | -0.158 |
izz | -0.151 |
Token | watched |
Token ID | 7342 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
joy | +0.030 |
folk | +0.030 |
uniforms | +0.027 |
acts | +0.027 |
ones | +0.026 |
Neg logit contributions | |
paste | -0.052 |
prise | -0.050 |
web | -0.050 |
aid | -0.046 |
izz | -0.046 |
Token | as |
Token ID | 355 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
joy | +0.023 |
ened | +0.022 |
acts | +0.020 |
yacht | +0.020 |
argued | +0.020 |
Neg logit contributions | |
prise | -0.049 |
web | -0.049 |
paste | -0.048 |
iders | -0.047 |
aid | -0.044 |
Token | for |
Token ID | 329 |
Feature activation | -5.96e-02 |
Pos logit contributions | |
prise | +0.043 |
web | +0.043 |
paste | +0.043 |
izz | +0.040 |
iders | +0.040 |
Neg logit contributions | |
ones | -0.027 |
wanted | -0.026 |
joy | -0.025 |
argued | -0.025 |
travelled | -0.025 |
Token | dinner |
Token ID | 8073 |
Feature activation | -5.76e-02 |
Pos logit contributions | |
paste | +0.131 |
web | +0.127 |
iders | +0.121 |
aid | +0.116 |
prise | +0.108 |
Neg logit contributions | |
joy | -0.072 |
ones | -0.072 |
goats | -0.071 |
uniforms | -0.070 |
market | -0.068 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.19e-02 |
Pos logit contributions | |
web | +0.094 |
paste | +0.094 |
prise | +0.094 |
iders | +0.085 |
aid | +0.083 |
Neg logit contributions | |
joy | -0.098 |
folk | -0.094 |
ones | -0.093 |
argued | -0.092 |
acts | -0.092 |
Token | Her |
Token ID | 2332 |
Feature activation | +2.07e-02 |
Pos logit contributions | |
joy | +0.023 |
folk | +0.023 |
argued | +0.021 |
acts | +0.021 |
ones | +0.020 |
Neg logit contributions | |
paste | -0.050 |
web | -0.049 |
prise | -0.049 |
iders | -0.046 |
aid | -0.044 |
Token | mother |
Token ID | 2802 |
Feature activation | +2.55e-02 |
Pos logit contributions | |
folk | +0.035 |
joy | +0.035 |
acts | +0.033 |
argued | +0.033 |
uniforms | +0.032 |
Neg logit contributions | |
web | -0.035 |
paste | -0.035 |
prise | -0.034 |
iders | -0.031 |
izz | -0.029 |
Token | nodded |
Token ID | 14464 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
joy | +0.018 |
folk | +0.017 |
acts | +0.016 |
argued | +0.016 |
nodded | +0.015 |
Neg logit contributions | |
prise | -0.068 |
paste | -0.067 |
web | -0.067 |
iders | -0.064 |
izz | -0.061 |
Token | and |
Token ID | 290 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
folk | +0.298 |
acts | +0.287 |
seals | +0.286 |
bank | +0.274 |
stalls | +0.272 |
Neg logit contributions | |
paste | -0.410 |
web | -0.405 |
prise | -0.402 |
izz | -0.365 |
aid | -0.354 |
Token | together |
Token ID | 1978 |
Feature activation | +7.49e-02 |
Pos logit contributions | |
joy | +0.022 |
folk | +0.021 |
acts | +0.021 |
argued | +0.020 |
wandered | +0.019 |
Neg logit contributions | |
paste | -0.046 |
web | -0.046 |
prise | -0.046 |
iders | -0.043 |
izz | -0.042 |
Token | they |
Token ID | 484 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
joy | +0.083 |
folk | +0.079 |
acts | +0.079 |
argued | +0.073 |
uniforms | +0.072 |
Neg logit contributions | |
paste | -0.171 |
prise | -0.168 |
web | -0.168 |
iders | -0.156 |
izz | -0.153 |
Token | went |
Token ID | 1816 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
joy | +0.097 |
folk | +0.091 |
acts | +0.089 |
argued | +0.088 |
wandered | +0.082 |
Neg logit contributions | |
paste | -0.249 |
web | -0.249 |
prise | -0.249 |
iders | -0.235 |
izz | -0.225 |
Token | inside |
Token ID | 2641 |
Feature activation | -8.47e-02 |
Pos logit contributions | |
paste | +0.057 |
prise | +0.057 |
web | +0.056 |
iders | +0.052 |
izz | +0.051 |
Neg logit contributions | |
joy | -0.030 |
folk | -0.030 |
acts | -0.029 |
argued | -0.028 |
ened | -0.027 |
Token | ol |
Token ID | 25776 |
Feature activation | -5.04e-02 |
Pos logit contributions | |
ones | +0.018 |
wanted | +0.015 |
acts | +0.015 |
blue | +0.015 |
one | +0.014 |
Neg logit contributions | |
web | -0.036 |
paste | -0.036 |
prise | -0.035 |
iders | -0.035 |
plug | -0.032 |
Token | ives |
Token ID | 1083 |
Feature activation | -2.61e-03 |
Pos logit contributions | |
web | +0.109 |
prise | +0.101 |
paste | +0.095 |
izz | +0.092 |
iders | +0.091 |
Neg logit contributions | |
folk | -0.059 |
ones | -0.058 |
clubs | -0.056 |
acts | -0.055 |
joy | -0.054 |
Token | ?" |
Token ID | 1701 |
Feature activation | -4.78e-03 |
Pos logit contributions | |
web | +0.005 |
paste | +0.005 |
prise | +0.005 |
iders | +0.005 |
izz | +0.005 |
Neg logit contributions | |
ones | -0.004 |
folk | -0.004 |
yacht | -0.003 |
joy | -0.003 |
uniforms | -0.003 |
Token | His |
Token ID | 2399 |
Feature activation | +5.65e-02 |
Pos logit contributions | |
paste | +0.010 |
prise | +0.009 |
izz | +0.009 |
Chick | +0.009 |
web | +0.009 |
Neg logit contributions | |
folk | -0.006 |
acts | -0.005 |
clubs | -0.005 |
joy | -0.005 |
argued | -0.005 |
Token | mum |
Token ID | 25682 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
folk | +0.106 |
joy | +0.100 |
argued | +0.098 |
acts | +0.098 |
ened | +0.095 |
Neg logit contributions | |
prise | -0.086 |
paste | -0.083 |
web | -0.083 |
iders | -0.076 |
izz | -0.072 |
Token | nodded |
Token ID | 14464 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
joy | +0.041 |
folk | +0.038 |
acts | +0.037 |
uniforms | +0.035 |
ones | +0.034 |
Neg logit contributions | |
paste | -0.090 |
web | -0.090 |
prise | -0.089 |
iders | -0.083 |
izz | -0.081 |
Token | and |
Token ID | 290 |
Feature activation | +2.80e-02 |
Pos logit contributions | |
folk | +0.307 |
acts | +0.302 |
seals | +0.288 |
bank | +0.287 |
habitats | +0.280 |
Neg logit contributions | |
paste | -0.401 |
web | -0.396 |
prise | -0.388 |
izz | -0.359 |
aid | -0.352 |
Token | said |
Token ID | 531 |
Feature activation | +6.33e-02 |
Pos logit contributions | |
joy | +0.038 |
folk | +0.036 |
acts | +0.034 |
uniforms | +0.033 |
ones | +0.033 |
Neg logit contributions | |
prise | -0.057 |
paste | -0.057 |
web | -0.056 |
iders | -0.051 |
izz | -0.051 |
Token | , |
Token ID | 11 |
Feature activation | +4.36e-02 |
Pos logit contributions | |
joy | +0.077 |
folk | +0.076 |
acts | +0.074 |
uniforms | +0.073 |
argued | +0.073 |
Neg logit contributions | |
paste | -0.137 |
prise | -0.133 |
web | -0.131 |
iders | -0.121 |
aid | -0.118 |
Token | " |
Token ID | 366 |
Feature activation | +5.85e-02 |
Pos logit contributions | |
acts | +0.049 |
folk | +0.045 |
joy | +0.045 |
uniforms | +0.044 |
country | +0.042 |
Neg logit contributions | |
paste | -0.097 |
web | -0.094 |
prise | -0.091 |
iders | -0.090 |
izz | -0.086 |
Token | Of |
Token ID | 5189 |
Feature activation | +5.02e-02 |
Pos logit contributions | |
joy | +0.093 |
folk | +0.089 |
acts | +0.088 |
argued | +0.087 |
ened | +0.084 |
Neg logit contributions | |
paste | -0.105 |
prise | -0.105 |
web | -0.104 |
iders | -0.096 |
izz | -0.091 |
Token | in |
Token ID | 287 |
Feature activation | +8.71e-02 |
Pos logit contributions | |
joy | +0.084 |
folk | +0.081 |
yacht | +0.079 |
uniforms | +0.077 |
ones | +0.077 |
Neg logit contributions | |
paste | -0.090 |
prise | -0.087 |
web | -0.084 |
aid | -0.080 |
iders | -0.079 |
Token | birthday |
Token ID | 10955 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
joy | +0.131 |
acts | +0.124 |
uniforms | +0.122 |
folk | +0.120 |
ones | +0.118 |
Neg logit contributions | |
paste | -0.171 |
web | -0.164 |
prise | -0.158 |
iders | -0.156 |
aid | -0.147 |
Token | surprises |
Token ID | 24072 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
web | +0.034 |
paste | +0.033 |
prise | +0.033 |
iders | +0.031 |
aid | +0.030 |
Neg logit contributions | |
joy | -0.024 |
folk | -0.022 |
acts | -0.022 |
argued | -0.022 |
uniforms | -0.022 |
Token | ?" |
Token ID | 1701 |
Feature activation | -5.19e-03 |
Pos logit contributions | |
joy | +0.044 |
folk | +0.044 |
acts | +0.043 |
uniforms | +0.042 |
ones | +0.042 |
Neg logit contributions | |
paste | -0.040 |
web | -0.039 |
prise | -0.039 |
iders | -0.036 |
aid | -0.034 |
Token | Jack |
Token ID | 3619 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
prise | +0.008 |
paste | +0.008 |
izz | +0.008 |
web | +0.007 |
ores | +0.007 |
Neg logit contributions | |
acts | -0.008 |
argued | -0.008 |
folk | -0.008 |
ened | -0.008 |
traveled | -0.007 |
Token | nodded |
Token ID | 14464 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
joy | +0.040 |
folk | +0.038 |
acts | +0.035 |
ones | +0.035 |
country | +0.034 |
Neg logit contributions | |
paste | -0.066 |
web | -0.065 |
prise | -0.062 |
iders | -0.059 |
izz | -0.057 |
Token | his |
Token ID | 465 |
Feature activation | +5.79e-02 |
Pos logit contributions | |
acts | +0.315 |
folk | +0.315 |
seals | +0.294 |
clubs | +0.291 |
goats | +0.289 |
Neg logit contributions | |
paste | -0.386 |
web | -0.379 |
prise | -0.375 |
izz | -0.344 |
aid | -0.330 |
Token | head |
Token ID | 1182 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
folk | +0.068 |
joy | +0.066 |
acts | +0.062 |
argued | +0.060 |
challenged | +0.058 |
Neg logit contributions | |
paste | -0.126 |
web | -0.125 |
prise | -0.121 |
iders | -0.116 |
x | -0.111 |
Token | and |
Token ID | 290 |
Feature activation | +3.25e-02 |
Pos logit contributions | |
acts | +0.196 |
folk | +0.183 |
seals | +0.181 |
uniforms | +0.172 |
ones | +0.171 |
Neg logit contributions | |
paste | -0.256 |
web | -0.239 |
prise | -0.223 |
aid | -0.214 |
izz | -0.209 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
joy | +0.041 |
folk | +0.039 |
acts | +0.038 |
uniforms | +0.035 |
ones | +0.035 |
Neg logit contributions | |
paste | -0.069 |
prise | -0.068 |
web | -0.068 |
iders | -0.062 |
izz | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
folk | +0.235 |
acts | +0.233 |
argued | +0.221 |
clubs | +0.213 |
wandered | +0.211 |
Neg logit contributions | |
paste | -0.234 |
web | -0.223 |
prise | -0.220 |
ores | -0.198 |
izz | -0.191 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
paste | +0.140 |
prise | +0.138 |
web | +0.135 |
iders | +0.125 |
izz | +0.121 |
Neg logit contributions | |
joy | -0.166 |
folk | -0.157 |
acts | -0.156 |
ones | -0.155 |
uniforms | -0.154 |
Token |
|
Token ID | 198 |
Feature activation | -1.29e-03 |
Pos logit contributions | |
web | +0.023 |
paste | +0.023 |
prise | +0.023 |
iders | +0.021 |
izz | +0.021 |
Neg logit contributions | |
joy | -0.016 |
acts | -0.015 |
argued | -0.015 |
folk | -0.015 |
ened | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -3.21e-03 |
Pos logit contributions | |
paste | +0.003 |
prise | +0.003 |
web | +0.003 |
iders | +0.003 |
aid | +0.003 |
Neg logit contributions | |
joy | -0.001 |
folk | -0.001 |
acts | -0.001 |
challenged | -0.001 |
uniforms | -0.001 |
Token | Luc |
Token ID | 22946 |
Feature activation | -3.94e-03 |
Pos logit contributions | |
paste | +0.005 |
prise | +0.005 |
web | +0.005 |
iders | +0.004 |
izz | +0.004 |
Neg logit contributions | |
joy | -0.006 |
folk | -0.006 |
acts | -0.006 |
argued | -0.006 |
ened | -0.005 |
Token | as |
Token ID | 292 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
paste | +0.007 |
web | +0.006 |
prise | +0.006 |
iders | +0.006 |
aid | +0.005 |
Neg logit contributions | |
folk | -0.007 |
joy | -0.007 |
acts | -0.007 |
argued | -0.006 |
ened | -0.006 |
Token | nodded |
Token ID | 14464 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
joy | +0.041 |
acts | +0.035 |
ones | +0.035 |
goats | +0.034 |
stalls | +0.034 |
Neg logit contributions | |
web | -0.072 |
prise | -0.071 |
paste | -0.070 |
izz | -0.067 |
iders | -0.067 |
Token | his |
Token ID | 465 |
Feature activation | +5.11e-02 |
Pos logit contributions | |
acts | +0.319 |
folk | +0.316 |
seals | +0.308 |
goats | +0.300 |
stalls | +0.299 |
Neg logit contributions | |
web | -0.380 |
paste | -0.374 |
prise | -0.353 |
izz | -0.325 |
bly | -0.318 |
Token | head |
Token ID | 1182 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
joy | +0.048 |
folk | +0.046 |
acts | +0.044 |
argued | +0.043 |
ened | +0.040 |
Neg logit contributions | |
web | -0.125 |
paste | -0.124 |
prise | -0.123 |
iders | -0.116 |
izz | -0.111 |
Token | and |
Token ID | 290 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
acts | +0.211 |
seals | +0.203 |
folk | +0.201 |
uniforms | +0.194 |
stalls | +0.194 |
Neg logit contributions | |
paste | -0.237 |
web | -0.221 |
aid | -0.197 |
prise | -0.197 |
izz | -0.186 |
Token | said |
Token ID | 531 |
Feature activation | +5.62e-02 |
Pos logit contributions | |
joy | +0.037 |
folk | +0.035 |
acts | +0.033 |
uniforms | +0.032 |
ones | +0.032 |
Neg logit contributions | |
paste | -0.055 |
web | -0.052 |
prise | -0.052 |
aid | -0.048 |
iders | -0.047 |
Token | , |
Token ID | 11 |
Feature activation | +4.20e-02 |
Pos logit contributions | |
joy | +0.069 |
acts | +0.067 |
folk | +0.067 |
argued | +0.066 |
uniforms | +0.065 |
Neg logit contributions | |
paste | -0.120 |
prise | -0.115 |
web | -0.114 |
iders | -0.106 |
aid | -0.103 |
Token | look |
Token ID | 804 |
Feature activation | -3.00e-04 |
Pos logit contributions | |
acts | +0.038 |
argued | +0.037 |
folk | +0.036 |
ones | +0.036 |
joy | +0.036 |
Neg logit contributions | |
web | -0.068 |
paste | -0.068 |
prise | -0.066 |
iders | -0.065 |
izz | -0.060 |
Token | for |
Token ID | 329 |
Feature activation | -6.35e-02 |
Pos logit contributions | |
web | +0.001 |
prise | +0.001 |
paste | +0.001 |
iders | +0.001 |
izz | +0.001 |
Neg logit contributions | |
joy | -0.000 |
acts | -0.000 |
folk | -0.000 |
ones | -0.000 |
ened | -0.000 |
Token | something |
Token ID | 1223 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
web | +0.129 |
paste | +0.127 |
iders | +0.125 |
prise | +0.125 |
izz | +0.115 |
Neg logit contributions | |
joy | -0.089 |
acts | -0.087 |
ones | -0.086 |
folk | -0.081 |
uniforms | -0.080 |
Token | ." |
Token ID | 526 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
web | +0.066 |
prise | +0.065 |
paste | +0.065 |
iders | +0.062 |
izz | +0.060 |
Neg logit contributions | |
joy | -0.042 |
acts | -0.042 |
folk | -0.041 |
ones | -0.040 |
argued | -0.039 |
Token | Lucy |
Token ID | 22162 |
Feature activation | -1.61e-03 |
Pos logit contributions | |
joy | +0.051 |
folk | +0.048 |
argued | +0.047 |
acts | +0.046 |
wandered | +0.046 |
Neg logit contributions | |
prise | -0.081 |
web | -0.081 |
paste | -0.080 |
iders | -0.075 |
aid | -0.071 |
Token | nodded |
Token ID | 14464 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
web | +0.004 |
paste | +0.004 |
prise | +0.003 |
iders | +0.003 |
izz | +0.003 |
Neg logit contributions | |
joy | -0.002 |
folk | -0.002 |
ones | -0.002 |
goats | -0.002 |
yacht | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
folk | +0.283 |
acts | +0.274 |
joy | +0.271 |
argued | +0.261 |
seals | +0.258 |
Neg logit contributions | |
paste | -0.430 |
web | -0.428 |
prise | -0.421 |
iders | -0.383 |
izz | -0.378 |
Token | said |
Token ID | 531 |
Feature activation | +6.30e-02 |
Pos logit contributions | |
joy | +0.033 |
folk | +0.032 |
acts | +0.029 |
uniforms | +0.029 |
ones | +0.028 |
Neg logit contributions | |
prise | -0.059 |
paste | -0.058 |
web | -0.058 |
iders | -0.052 |
aid | -0.052 |
Token | , |
Token ID | 11 |
Feature activation | +4.78e-02 |
Pos logit contributions | |
joy | +0.078 |
folk | +0.077 |
uniforms | +0.076 |
acts | +0.075 |
argued | +0.075 |
Neg logit contributions | |
paste | -0.134 |
prise | -0.129 |
web | -0.127 |
iders | -0.119 |
aid | -0.116 |
Token | " |
Token ID | 366 |
Feature activation | +6.70e-02 |
Pos logit contributions | |
acts | +0.052 |
joy | +0.048 |
uniforms | +0.047 |
folk | +0.047 |
ened | +0.044 |
Neg logit contributions | |
paste | -0.108 |
web | -0.104 |
prise | -0.102 |
iders | -0.100 |
ores | -0.094 |
Token | Like |
Token ID | 7594 |
Feature activation | -6.34e-02 |
Pos logit contributions | |
joy | +0.095 |
folk | +0.092 |
acts | +0.089 |
argued | +0.087 |
ones | +0.087 |
Neg logit contributions | |
prise | -0.132 |
web | -0.132 |
paste | -0.131 |
iders | -0.121 |
izz | -0.116 |
Token | homeless |
Token ID | 10463 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
joy | +0.032 |
folk | +0.031 |
acts | +0.030 |
argued | +0.030 |
uniforms | +0.028 |
Neg logit contributions | |
paste | -0.057 |
web | -0.057 |
prise | -0.057 |
iders | -0.053 |
izz | -0.051 |
Token | man |
Token ID | 582 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
web | +0.026 |
prise | +0.025 |
paste | +0.025 |
iders | +0.024 |
aid | +0.022 |
Neg logit contributions | |
joy | -0.017 |
argued | -0.015 |
folk | -0.015 |
ened | -0.015 |
wandered | -0.015 |
Token | ?" |
Token ID | 1701 |
Feature activation | -5.70e-03 |
Pos logit contributions | |
joy | +0.053 |
folk | +0.052 |
acts | +0.052 |
argued | +0.051 |
challenged | +0.050 |
Neg logit contributions | |
paste | -0.072 |
web | -0.071 |
prise | -0.070 |
iders | -0.067 |
izz | -0.063 |
Token | His |
Token ID | 2399 |
Feature activation | +5.74e-02 |
Pos logit contributions | |
prise | +0.012 |
paste | +0.012 |
web | +0.012 |
izz | +0.011 |
Chick | +0.011 |
Neg logit contributions | |
folk | -0.006 |
acts | -0.006 |
argued | -0.006 |
joy | -0.006 |
ones | -0.006 |
Token | dad |
Token ID | 9955 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
folk | +0.125 |
hi | +0.118 |
travelled | +0.117 |
argued | +0.116 |
ones | +0.116 |
Neg logit contributions | |
prise | -0.060 |
paste | -0.057 |
web | -0.053 |
x | -0.052 |
iders | -0.047 |
Token | nodded |
Token ID | 14464 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
joy | +0.034 |
folk | +0.031 |
acts | +0.030 |
ones | +0.029 |
uniforms | +0.028 |
Neg logit contributions | |
paste | -0.083 |
web | -0.083 |
prise | -0.082 |
iders | -0.076 |
izz | -0.075 |
Token | , |
Token ID | 11 |
Feature activation | +3.54e-02 |
Pos logit contributions | |
folk | +0.308 |
acts | +0.305 |
seals | +0.299 |
clubs | +0.296 |
habitats | +0.295 |
Neg logit contributions | |
paste | -0.381 |
web | -0.375 |
prise | -0.369 |
izz | -0.336 |
bly | -0.332 |
Token | " |
Token ID | 366 |
Feature activation | +5.33e-02 |
Pos logit contributions | |
acts | +0.045 |
folk | +0.044 |
joy | +0.042 |
uniforms | +0.041 |
ones | +0.041 |
Neg logit contributions | |
paste | -0.074 |
web | -0.072 |
prise | -0.070 |
aid | -0.066 |
iders | -0.065 |
Token | Yes |
Token ID | 5297 |
Feature activation | +8.35e-02 |
Pos logit contributions | |
joy | +0.069 |
folk | +0.068 |
acts | +0.065 |
argued | +0.063 |
hi | +0.063 |
Neg logit contributions | |
web | -0.111 |
paste | -0.110 |
prise | -0.108 |
iders | -0.104 |
izz | -0.098 |
Token | , |
Token ID | 11 |
Feature activation | +4.68e-02 |
Pos logit contributions | |
joy | +0.075 |
folk | +0.070 |
acts | +0.068 |
argued | +0.065 |
ened | +0.064 |
Neg logit contributions | |
web | -0.209 |
prise | -0.206 |
paste | -0.205 |
iders | -0.196 |
aid | -0.188 |
Token | Toby |
Token ID | 37578 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
joy | +0.071 |
folk | +0.069 |
acts | +0.069 |
argued | +0.067 |
uniforms | +0.065 |
Neg logit contributions | |
web | -0.086 |
prise | -0.086 |
paste | -0.086 |
iders | -0.079 |
izz | -0.075 |
Token | said |
Token ID | 531 |
Feature activation | +5.09e-02 |
Pos logit contributions | |
joy | +0.023 |
folk | +0.022 |
acts | +0.022 |
argued | +0.021 |
uniforms | +0.021 |
Neg logit contributions | |
paste | -0.029 |
web | -0.029 |
prise | -0.029 |
iders | -0.027 |
izz | -0.026 |
Token | yes |
Token ID | 3763 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
joy | +0.063 |
folk | +0.061 |
acts | +0.060 |
argued | +0.059 |
uniforms | +0.056 |
Neg logit contributions | |
paste | -0.108 |
prise | -0.107 |
web | -0.107 |
iders | -0.099 |
izz | -0.095 |
Token | with |
Token ID | 351 |
Feature activation | +8.59e-02 |
Pos logit contributions | |
joy | +0.167 |
folk | +0.158 |
acts | +0.157 |
argued | +0.155 |
ones | +0.149 |
Neg logit contributions | |
web | -0.274 |
prise | -0.273 |
paste | -0.273 |
iders | -0.255 |
izz | -0.246 |
Token | a |
Token ID | 257 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
joy | +0.121 |
folk | +0.114 |
acts | +0.114 |
argued | +0.110 |
uniforms | +0.107 |
Neg logit contributions | |
paste | -0.171 |
web | -0.171 |
prise | -0.170 |
iders | -0.160 |
izz | -0.152 |
Token | happy |
Token ID | 3772 |
Feature activation | +5.84e-02 |
Pos logit contributions | |
web | +0.045 |
paste | +0.045 |
prise | +0.044 |
iders | +0.043 |
izz | +0.041 |
Neg logit contributions | |
joy | -0.023 |
acts | -0.021 |
folk | -0.021 |
argued | -0.020 |
ones | -0.020 |
Token | nod |
Token ID | 18666 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
folk | +0.070 |
joy | +0.069 |
argued | +0.067 |
acts | +0.066 |
uniforms | +0.063 |
Neg logit contributions | |
prise | -0.124 |
paste | -0.123 |
web | -0.123 |
iders | -0.111 |
aid | -0.110 |
Token | . |
Token ID | 13 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
joy | +0.333 |
acts | +0.311 |
folk | +0.310 |
argued | +0.307 |
challenged | +0.304 |
Neg logit contributions | |
paste | -0.395 |
web | -0.391 |
prise | -0.353 |
potion | -0.321 |
iders | -0.318 |
Token |
|
Token ID | 198 |
Feature activation | -7.34e-03 |
Pos logit contributions | |
folk | +0.037 |
acts | +0.037 |
argued | +0.036 |
yacht | +0.036 |
uniforms | +0.036 |
Neg logit contributions | |
prise | -0.056 |
paste | -0.055 |
web | -0.052 |
iders | -0.052 |
izz | -0.052 |
Token |
|
Token ID | 198 |
Feature activation | -8.21e-03 |
Pos logit contributions | |
paste | +0.018 |
web | +0.018 |
prise | +0.017 |
iders | +0.016 |
aid | +0.015 |
Neg logit contributions | |
joy | -0.007 |
folk | -0.007 |
acts | -0.006 |
wandered | -0.006 |
argued | -0.006 |
Token | The |
Token ID | 464 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
paste | +0.013 |
web | +0.013 |
prise | +0.013 |
iders | +0.012 |
izz | +0.011 |
Neg logit contributions | |
joy | -0.014 |
folk | -0.014 |
acts | -0.014 |
argued | -0.014 |
uniforms | -0.013 |
Token | Princess |
Token ID | 8449 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
joy | +0.055 |
acts | +0.053 |
folk | +0.053 |
ones | +0.051 |
argued | +0.050 |
Neg logit contributions | |
paste | -0.076 |
web | -0.075 |
prise | -0.075 |
iders | -0.071 |
izz | -0.068 |
Token | peek |
Token ID | 27185 |
Feature activation | -4.17e-02 |
Pos logit contributions | |
ones | +0.015 |
acts | +0.015 |
yacht | +0.015 |
wanted | +0.015 |
argued | +0.014 |
Neg logit contributions | |
web | -0.036 |
paste | -0.036 |
iders | -0.035 |
prise | -0.035 |
aid | -0.032 |
Token | too |
Token ID | 1165 |
Feature activation | -4.80e-02 |
Pos logit contributions | |
prise | +0.092 |
paste | +0.087 |
web | +0.087 |
iders | +0.086 |
aid | +0.082 |
Neg logit contributions | |
joy | -0.051 |
ened | -0.047 |
ones | -0.046 |
uniforms | -0.046 |
yacht | -0.045 |
Token | ?" |
Token ID | 1701 |
Feature activation | -6.90e-03 |
Pos logit contributions | |
paste | +0.099 |
prise | +0.087 |
web | +0.087 |
iders | +0.083 |
aid | +0.083 |
Neg logit contributions | |
country | -0.072 |
ones | -0.070 |
wanted | -0.069 |
field | -0.069 |
market | -0.068 |
Token | Her |
Token ID | 2332 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
paste | +0.012 |
web | +0.012 |
prise | +0.012 |
iders | +0.011 |
izz | +0.010 |
Neg logit contributions | |
joy | -0.011 |
folk | -0.011 |
acts | -0.011 |
argued | -0.011 |
ened | -0.010 |
Token | brother |
Token ID | 3956 |
Feature activation | +4.94e-02 |
Pos logit contributions | |
joy | +0.014 |
acts | +0.013 |
folk | +0.012 |
argued | +0.012 |
uniforms | +0.012 |
Neg logit contributions | |
prise | -0.031 |
paste | -0.031 |
web | -0.031 |
iders | -0.029 |
izz | -0.028 |
Token | nodded |
Token ID | 14464 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
joy | +0.056 |
folk | +0.056 |
acts | +0.052 |
country | +0.052 |
ones | +0.050 |
Neg logit contributions | |
paste | -0.107 |
web | -0.103 |
prise | -0.102 |
iders | -0.095 |
aid | -0.094 |
Token | eagerly |
Token ID | 30130 |
Feature activation | +6.25e-02 |
Pos logit contributions | |
folk | +0.281 |
acts | +0.277 |
seals | +0.255 |
stalls | +0.255 |
traveled | +0.255 |
Neg logit contributions | |
paste | -0.425 |
web | -0.418 |
prise | -0.413 |
aid | -0.374 |
izz | -0.372 |
Token | and |
Token ID | 290 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
folk | +0.093 |
acts | +0.092 |
uniforms | +0.090 |
traveled | +0.089 |
seals | +0.088 |
Neg logit contributions | |
paste | -0.111 |
prise | -0.104 |
web | -0.102 |
aid | -0.094 |
heads | -0.093 |
Token | joined |
Token ID | 5399 |
Feature activation | +9.71e-02 |
Pos logit contributions | |
joy | +0.031 |
folk | +0.029 |
acts | +0.028 |
argued | +0.027 |
uniforms | +0.026 |
Neg logit contributions | |
paste | -0.059 |
prise | -0.059 |
web | -0.059 |
iders | -0.055 |
izz | -0.053 |
Token | them |
Token ID | 606 |
Feature activation | +6.30e-03 |
Pos logit contributions | |
joy | +0.137 |
folk | +0.131 |
acts | +0.129 |
argued | +0.128 |
ened | +0.123 |
Neg logit contributions | |
paste | -0.192 |
prise | -0.191 |
web | -0.191 |
iders | -0.178 |
izz | -0.169 |
Token | . |
Token ID | 13 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
joy | +0.008 |
folk | +0.007 |
acts | +0.007 |
argued | +0.007 |
ones | +0.007 |
Neg logit contributions | |
paste | -0.014 |
web | -0.014 |
prise | -0.014 |
iders | -0.013 |
izz | -0.012 |
Token | keep |
Token ID | 1394 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
web | +0.062 |
paste | +0.061 |
iders | +0.060 |
prise | +0.057 |
izz | +0.053 |
Neg logit contributions | |
wanted | -0.026 |
wandered | -0.025 |
acts | -0.025 |
walked | -0.025 |
hi | -0.025 |
Token | it |
Token ID | 340 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
paste | +0.042 |
iders | +0.041 |
web | +0.041 |
prise | +0.041 |
aid | +0.038 |
Neg logit contributions | |
joy | -0.027 |
ones | -0.025 |
uniforms | -0.025 |
yacht | -0.025 |
acts | -0.024 |
Token | ?" |
Token ID | 1701 |
Feature activation | -9.51e-04 |
Pos logit contributions | |
paste | +0.090 |
iders | +0.086 |
web | +0.084 |
prise | +0.084 |
izz | +0.081 |
Neg logit contributions | |
ones | -0.061 |
yacht | -0.059 |
one | -0.057 |
visited | -0.057 |
joy | -0.056 |
Token | Her |
Token ID | 2332 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
paste | +0.002 |
prise | +0.002 |
web | +0.002 |
iders | +0.002 |
izz | +0.002 |
Neg logit contributions | |
joy | -0.001 |
folk | -0.001 |
acts | -0.001 |
argued | -0.001 |
uniforms | -0.001 |
Token | Mom |
Token ID | 11254 |
Feature activation | -4.68e-02 |
Pos logit contributions | |
joy | +0.030 |
acts | +0.028 |
folk | +0.028 |
argued | +0.028 |
uniforms | +0.027 |
Neg logit contributions | |
paste | -0.024 |
web | -0.024 |
prise | -0.024 |
iders | -0.022 |
izz | -0.021 |
Token | nodded |
Token ID | 14464 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
paste | +0.119 |
web | +0.119 |
prise | +0.118 |
iders | +0.111 |
izz | +0.109 |
Neg logit contributions | |
joy | -0.039 |
folk | -0.036 |
acts | -0.035 |
argued | -0.032 |
ones | -0.032 |
Token | and |
Token ID | 290 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
folk | +0.286 |
acts | +0.279 |
seals | +0.273 |
bank | +0.263 |
joy | +0.259 |
Neg logit contributions | |
paste | -0.425 |
web | -0.420 |
prise | -0.409 |
izz | -0.379 |
aid | -0.372 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
joy | +0.029 |
folk | +0.027 |
acts | +0.027 |
argued | +0.026 |
uniforms | +0.025 |
Neg logit contributions | |
paste | -0.059 |
prise | -0.059 |
web | -0.059 |
iders | -0.055 |
izz | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
folk | +0.221 |
acts | +0.214 |
argued | +0.209 |
seals | +0.198 |
uniforms | +0.198 |
Neg logit contributions | |
paste | -0.254 |
web | -0.246 |
prise | -0.240 |
aid | -0.214 |
ores | -0.212 |
Token | |
Token ID | 220 |
Feature activation | +8.73e-02 |
Pos logit contributions | |
joy | +0.030 |
acts | +0.029 |
folk | +0.029 |
argued | +0.028 |
uniforms | +0.026 |
Neg logit contributions | |
paste | -0.060 |
web | -0.059 |
prise | -0.059 |
iders | -0.055 |
izz | -0.053 |
Token |
|
Token ID | 198 |
Feature activation | -5.19e-03 |
Pos logit contributions | |
joy | +0.075 |
folk | +0.066 |
acts | +0.063 |
wandered | +0.062 |
walked | +0.061 |
Neg logit contributions | |
prise | -0.222 |
web | -0.221 |
paste | -0.220 |
iders | -0.203 |
aid | -0.196 |
Token | l |
Token ID | 300 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
joy | +0.094 |
acts | +0.092 |
folk | +0.091 |
argued | +0.089 |
ones | +0.088 |
Neg logit contributions | |
web | -0.135 |
paste | -0.134 |
prise | -0.134 |
iders | -0.127 |
izz | -0.120 |
Token | aces |
Token ID | 2114 |
Feature activation | +7.53e-02 |
Pos logit contributions | |
joy | +0.047 |
folk | +0.046 |
acts | +0.045 |
wandered | +0.040 |
uniforms | +0.040 |
Neg logit contributions | |
prise | -0.073 |
web | -0.072 |
paste | -0.068 |
iders | -0.065 |
Chick | -0.062 |
Token | ." |
Token ID | 526 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
joy | +0.101 |
folk | +0.096 |
acts | +0.095 |
argued | +0.093 |
wandered | +0.089 |
Neg logit contributions | |
paste | -0.153 |
web | -0.152 |
prise | -0.152 |
iders | -0.141 |
izz | -0.135 |
Token | Her |
Token ID | 2332 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
joy | +0.023 |
folk | +0.021 |
acts | +0.019 |
argued | +0.019 |
ones | +0.018 |
Neg logit contributions | |
web | -0.060 |
paste | -0.059 |
prise | -0.059 |
iders | -0.056 |
izz | -0.053 |
Token | son |
Token ID | 3367 |
Feature activation | +5.71e-02 |
Pos logit contributions | |
joy | +0.033 |
acts | +0.030 |
folk | +0.030 |
argued | +0.030 |
uniforms | +0.029 |
Neg logit contributions | |
prise | -0.036 |
paste | -0.036 |
web | -0.036 |
iders | -0.034 |
izz | -0.032 |
Token | nodded |
Token ID | 14464 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
joy | +0.053 |
folk | +0.050 |
acts | +0.048 |
ones | +0.045 |
uniforms | +0.045 |
Neg logit contributions | |
paste | -0.139 |
web | -0.138 |
prise | -0.136 |
iders | -0.127 |
aid | -0.125 |
Token | and |
Token ID | 290 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
acts | +0.263 |
folk | +0.262 |
seals | +0.244 |
argued | +0.241 |
joy | +0.240 |
Neg logit contributions | |
paste | -0.438 |
web | -0.434 |
prise | -0.428 |
izz | -0.382 |
aid | -0.382 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
joy | +0.023 |
folk | +0.022 |
acts | +0.021 |
uniforms | +0.021 |
ones | +0.020 |
Neg logit contributions | |
paste | -0.039 |
prise | -0.038 |
web | -0.038 |
aid | -0.035 |
iders | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
folk | +0.205 |
acts | +0.204 |
argued | +0.197 |
joy | +0.187 |
wandered | +0.187 |
Neg logit contributions | |
paste | -0.256 |
web | -0.247 |
prise | -0.240 |
iders | -0.215 |
aid | -0.214 |
Token |
|
Token ID | 198 |
Feature activation | -9.86e-03 |
Pos logit contributions | |
joy | +0.020 |
folk | +0.018 |
acts | +0.018 |
argued | +0.017 |
wandered | +0.016 |
Neg logit contributions | |
prise | -0.048 |
web | -0.048 |
paste | -0.047 |
iders | -0.045 |
izz | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
paste | +0.029 |
web | +0.028 |
prise | +0.027 |
iders | +0.027 |
potion | +0.025 |
Neg logit contributions | |
| -0.006 |
joy | -0.005 |
walked | -0.005 |
They | -0.004 |
wanted | -0.004 |
Token | Grand |
Token ID | 5675 |
Feature activation | -5.23e-02 |
Pos logit contributions | |
web | +0.022 |
paste | +0.022 |
iders | +0.021 |
prise | +0.021 |
izz | +0.019 |
Neg logit contributions | |
folk | -0.013 |
joy | -0.013 |
acts | -0.013 |
argued | -0.013 |
nodded | -0.013 |
Token | ma |
Token ID | 2611 |
Feature activation | -3.27e-03 |
Pos logit contributions | |
paste | +0.091 |
web | +0.089 |
prise | +0.088 |
iders | +0.083 |
aid | +0.076 |
Neg logit contributions | |
joy | -0.087 |
folk | -0.085 |
acts | -0.083 |
argued | -0.082 |
uniforms | -0.079 |
Token | said |
Token ID | 531 |
Feature activation | +7.12e-02 |
Pos logit contributions | |
prise | +0.008 |
paste | +0.008 |
web | +0.008 |
iders | +0.007 |
izz | +0.007 |
Neg logit contributions | |
joy | -0.003 |
folk | -0.003 |
acts | -0.003 |
argued | -0.003 |
ened | -0.003 |
Token | with |
Token ID | 351 |
Feature activation | +8.94e-02 |
Pos logit contributions | |
joy | +0.089 |
acts | +0.085 |
folk | +0.085 |
argued | +0.082 |
uniforms | +0.080 |
Neg logit contributions | |
paste | -0.151 |
prise | -0.149 |
web | -0.149 |
iders | -0.138 |
izz | -0.134 |
Token | a |
Token ID | 257 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
joy | +0.124 |
acts | +0.118 |
folk | +0.117 |
uniforms | +0.116 |
argued | +0.113 |
Neg logit contributions | |
web | -0.185 |
prise | -0.178 |
paste | -0.178 |
iders | -0.176 |
izz | -0.163 |
Token | nod |
Token ID | 18666 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
web | +0.044 |
paste | +0.044 |
prise | +0.043 |
iders | +0.042 |
izz | +0.040 |
Neg logit contributions | |
joy | -0.021 |
acts | -0.019 |
folk | -0.019 |
argued | -0.019 |
uniforms | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
joy | +0.320 |
acts | +0.313 |
folk | +0.306 |
seals | +0.302 |
stalls | +0.296 |
Neg logit contributions | |
paste | -0.395 |
web | -0.392 |
prise | -0.359 |
izz | -0.333 |
aid | -0.329 |
Token | " |
Token ID | 366 |
Feature activation | +5.66e-02 |
Pos logit contributions | |
joy | +0.021 |
folk | +0.020 |
acts | +0.019 |
argued | +0.018 |
uniforms | +0.017 |
Neg logit contributions | |
paste | -0.071 |
prise | -0.071 |
web | -0.071 |
iders | -0.067 |
izz | -0.065 |
Token | You |
Token ID | 1639 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
joy | +0.081 |
folk | +0.079 |
acts | +0.074 |
hi | +0.074 |
argued | +0.073 |
Neg logit contributions | |
paste | -0.110 |
web | -0.110 |
prise | -0.110 |
iders | -0.102 |
izz | -0.096 |
Token | can |
Token ID | 460 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
acts | +0.042 |
folk | +0.042 |
argued | +0.041 |
joy | +0.041 |
wanted | +0.039 |
Neg logit contributions | |
web | -0.088 |
paste | -0.087 |
prise | -0.085 |
iders | -0.083 |
izz | -0.078 |
Token | 't |
Token ID | 470 |
Feature activation | -5.55e-02 |
Pos logit contributions | |
paste | +0.052 |
web | +0.052 |
prise | +0.050 |
iders | +0.049 |
aid | +0.044 |
Neg logit contributions | |
acts | -0.026 |
hi | -0.026 |
argued | -0.026 |
folk | -0.026 |
walked | -0.026 |
Token | the |
Token ID | 262 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
prise | +0.080 |
web | +0.072 |
paste | +0.072 |
aid | +0.071 |
iders | +0.071 |
Neg logit contributions | |
ones | -0.040 |
ened | -0.039 |
acts | -0.037 |
folk | -0.036 |
uniforms | -0.035 |
Token | cloth |
Token ID | 16270 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
joy | +0.036 |
acts | +0.035 |
ones | +0.034 |
folk | +0.033 |
uniforms | +0.033 |
Neg logit contributions | |
prise | -0.047 |
web | -0.046 |
paste | -0.045 |
iders | -0.044 |
izz | -0.042 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
prise | +0.061 |
paste | +0.059 |
web | +0.059 |
iders | +0.056 |
izz | +0.054 |
Neg logit contributions | |
joy | -0.034 |
folk | -0.034 |
acts | -0.033 |
argued | -0.033 |
ened | -0.032 |
Token | His |
Token ID | 2399 |
Feature activation | +5.87e-02 |
Pos logit contributions | |
joy | +0.019 |
folk | +0.017 |
acts | +0.017 |
argued | +0.016 |
uniforms | +0.015 |
Neg logit contributions | |
paste | -0.059 |
prise | -0.059 |
web | -0.059 |
iders | -0.055 |
izz | -0.054 |
Token | mom |
Token ID | 1995 |
Feature activation | +2.41e-03 |
Pos logit contributions | |
joy | +0.090 |
acts | +0.083 |
uniforms | +0.081 |
argued | +0.081 |
nodded | +0.080 |
Neg logit contributions | |
prise | -0.114 |
paste | -0.113 |
web | -0.111 |
iders | -0.106 |
izz | -0.103 |
Token | nodded |
Token ID | 14464 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
joy | +0.002 |
argued | +0.002 |
folk | +0.002 |
acts | +0.001 |
nodded | +0.001 |
Neg logit contributions | |
prise | -0.006 |
paste | -0.006 |
web | -0.006 |
iders | -0.006 |
izz | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
folk | +0.269 |
acts | +0.265 |
seals | +0.257 |
goats | +0.249 |
stalls | +0.247 |
Neg logit contributions | |
paste | -0.429 |
web | -0.425 |
prise | -0.421 |
izz | -0.381 |
aid | -0.379 |
Token | said |
Token ID | 531 |
Feature activation | +5.71e-02 |
Pos logit contributions | |
joy | +0.020 |
folk | +0.018 |
acts | +0.018 |
argued | +0.018 |
wandered | +0.017 |
Neg logit contributions | |
paste | -0.044 |
web | -0.044 |
prise | -0.044 |
iders | -0.041 |
izz | -0.040 |
Token | , |
Token ID | 11 |
Feature activation | +4.01e-02 |
Pos logit contributions | |
joy | +0.061 |
folk | +0.059 |
acts | +0.058 |
argued | +0.056 |
uniforms | +0.055 |
Neg logit contributions | |
paste | -0.132 |
web | -0.129 |
prise | -0.129 |
iders | -0.121 |
aid | -0.116 |
Token | " |
Token ID | 366 |
Feature activation | +5.59e-02 |
Pos logit contributions | |
acts | +0.037 |
joy | +0.036 |
folk | +0.034 |
uniforms | +0.034 |
argued | +0.033 |
Neg logit contributions | |
paste | -0.098 |
web | -0.095 |
prise | -0.093 |
iders | -0.090 |
izz | -0.087 |
Token | Very |
Token ID | 16371 |
Feature activation | -8.71e-03 |
Pos logit contributions | |
joy | +0.076 |
folk | +0.072 |
acts | +0.071 |
argued | +0.069 |
ones | +0.068 |
Neg logit contributions | |
prise | -0.114 |
paste | -0.113 |
web | -0.112 |
iders | -0.105 |
izz | -0.100 |
Token | ?' |
Token ID | 8348 |
Feature activation | +7.91e-03 |
Pos logit contributions | |
prise | +0.077 |
web | +0.077 |
paste | +0.077 |
iders | +0.071 |
izz | +0.066 |
Neg logit contributions | |
joy | -0.059 |
folk | -0.058 |
argued | -0.056 |
acts | -0.056 |
uniforms | -0.054 |
Token |
|
Token ID | 198 |
Feature activation | +7.40e-03 |
Pos logit contributions | |
joy | +0.010 |
folk | +0.010 |
acts | +0.010 |
argued | +0.010 |
uniforms | +0.009 |
Neg logit contributions | |
paste | -0.016 |
prise | -0.016 |
web | -0.016 |
iders | -0.015 |
izz | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | +1.95e-03 |
Pos logit contributions | |
joy | +0.006 |
nodded | +0.005 |
folk | +0.005 |
| +0.005 |
wandered | +0.005 |
Neg logit contributions | |
web | -0.020 |
paste | -0.020 |
prise | -0.019 |
iders | -0.019 |
potion | -0.017 |
Token | Her |
Token ID | 9360 |
Feature activation | +6.74e-03 |
Pos logit contributions | |
joy | +0.003 |
folk | +0.003 |
nodded | +0.003 |
wandered | +0.003 |
argued | +0.003 |
Neg logit contributions | |
paste | -0.003 |
web | -0.003 |
prise | -0.003 |
iders | -0.003 |
potion | -0.003 |
Token | daughter |
Token ID | 4957 |
Feature activation | +5.70e-02 |
Pos logit contributions | |
folk | +0.012 |
argued | +0.011 |
joy | +0.011 |
travelled | +0.011 |
traveled | +0.011 |
Neg logit contributions | |
prise | -0.011 |
paste | -0.011 |
web | -0.011 |
iders | -0.010 |
izz | -0.010 |
Token | nodded |
Token ID | 14464 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
joy | +0.054 |
folk | +0.050 |
acts | +0.047 |
uniforms | +0.046 |
ones | +0.046 |
Neg logit contributions | |
paste | -0.137 |
web | -0.135 |
prise | -0.132 |
iders | -0.124 |
aid | -0.122 |
Token | , |
Token ID | 11 |
Feature activation | +3.59e-02 |
Pos logit contributions | |
folk | +0.304 |
acts | +0.300 |
seals | +0.289 |
joy | +0.287 |
argued | +0.283 |
Neg logit contributions | |
paste | -0.394 |
web | -0.394 |
prise | -0.388 |
izz | -0.351 |
iders | -0.342 |
Token | smiling |
Token ID | 16755 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
acts | +0.049 |
folk | +0.049 |
uniforms | +0.048 |
joy | +0.047 |
ones | +0.047 |
Neg logit contributions | |
paste | -0.068 |
web | -0.067 |
prise | -0.066 |
aid | -0.060 |
iders | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
acts | +0.150 |
folk | +0.150 |
joy | +0.143 |
argued | +0.141 |
uniforms | +0.140 |
Neg logit contributions | |
paste | -0.227 |
web | -0.219 |
prise | -0.211 |
iders | -0.197 |
aid | -0.190 |
Token | ' |
Token ID | 705 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
uniforms | +0.022 |
folk | +0.022 |
argued | +0.022 |
acts | +0.022 |
yacht | +0.021 |
Neg logit contributions | |
paste | -0.032 |
prise | -0.032 |
web | -0.030 |
iders | -0.030 |
izz | -0.030 |
Token | Yes |
Token ID | 5297 |
Feature activation | +7.80e-02 |
Pos logit contributions | |
joy | +0.032 |
acts | +0.032 |
folk | +0.032 |
argued | +0.031 |
uniforms | +0.031 |
Neg logit contributions | |
prise | -0.038 |
paste | -0.036 |
web | -0.036 |
iders | -0.033 |
aid | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
paste | +0.090 |
web | +0.089 |
prise | +0.089 |
iders | +0.087 |
potion | +0.080 |
Neg logit contributions | |
| -0.020 |
joy | -0.017 |
walked | -0.016 |
wandered | -0.016 |
wanted | -0.015 |
Token | Jack |
Token ID | 14295 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
prise | +0.050 |
paste | +0.050 |
web | +0.050 |
iders | +0.046 |
potion | +0.045 |
Neg logit contributions | |
joy | -0.048 |
wandered | -0.047 |
walked | -0.047 |
wanted | -0.047 |
challenged | -0.046 |
Token | had |
Token ID | 550 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
web | +0.057 |
paste | +0.057 |
prise | +0.056 |
iders | +0.052 |
izz | +0.051 |
Neg logit contributions | |
joy | -0.026 |
folk | -0.025 |
acts | -0.023 |
ones | -0.023 |
uniforms | -0.022 |
Token | brought |
Token ID | 3181 |
Feature activation | -2.50e-03 |
Pos logit contributions | |
folk | +0.030 |
joy | +0.029 |
acts | +0.029 |
yacht | +0.027 |
hi | +0.026 |
Neg logit contributions | |
paste | -0.038 |
prise | -0.037 |
web | -0.037 |
aid | -0.032 |
izz | -0.032 |
Token | L |
Token ID | 406 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
web | +0.006 |
prise | +0.006 |
paste | +0.005 |
iders | +0.005 |
izz | +0.005 |
Neg logit contributions | |
joy | -0.003 |
ones | -0.003 |
acts | -0.003 |
uniforms | -0.003 |
ened | -0.003 |
Token | izz |
Token ID | 6457 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
paste | +0.085 |
prise | +0.079 |
web | +0.075 |
iders | +0.058 |
heads | +0.058 |
Neg logit contributions | |
joy | -0.243 |
acts | -0.237 |
wanted | -0.232 |
wandered | -0.229 |
disagreed | -0.227 |
Token | ie |
Token ID | 494 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
prise | +0.387 |
paste | +0.382 |
web | +0.380 |
iders | +0.354 |
izz | +0.326 |
Neg logit contributions | |
joy | -0.317 |
acts | -0.306 |
folk | -0.304 |
argued | -0.298 |
ened | -0.295 |
Token | a |
Token ID | 257 |
Feature activation | -6.95e-02 |
Pos logit contributions | |
paste | +0.267 |
web | +0.266 |
prise | +0.266 |
iders | +0.249 |
izz | +0.238 |
Neg logit contributions | |
joy | -0.173 |
folk | -0.164 |
acts | -0.162 |
argued | -0.159 |
ones | -0.153 |
Token | present |
Token ID | 1944 |
Feature activation | +7.26e-03 |
Pos logit contributions | |
paste | +0.143 |
web | +0.142 |
iders | +0.141 |
prise | +0.136 |
izz | +0.135 |
Neg logit contributions | |
joy | -0.095 |
yacht | -0.088 |
ones | -0.084 |
acts | -0.084 |
hi | -0.083 |
Token | . |
Token ID | 13 |
Feature activation | +6.89e-03 |
Pos logit contributions | |
joy | +0.010 |
folk | +0.009 |
acts | +0.009 |
argued | +0.009 |
ones | +0.009 |
Neg logit contributions | |
prise | -0.015 |
paste | -0.015 |
web | -0.015 |
iders | -0.014 |
izz | -0.013 |
Token | It |
Token ID | 632 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
joy | +0.006 |
folk | +0.005 |
acts | +0.005 |
wandered | +0.005 |
ened | +0.004 |
Neg logit contributions | |
prise | -0.018 |
web | -0.018 |
paste | -0.018 |
iders | -0.017 |
aid | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -4.71e-02 |
Pos logit contributions | |
prise | +0.031 |
web | +0.031 |
paste | +0.031 |
iders | +0.029 |
izz | +0.028 |
Neg logit contributions | |
joy | -0.011 |
folk | -0.010 |
acts | -0.010 |
argued | -0.009 |
uniforms | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -4.42e-02 |
Pos logit contributions | |
paste | +0.137 |
web | +0.134 |
prise | +0.132 |
iders | +0.129 |
potion | +0.122 |
Neg logit contributions | |
| -0.026 |
They | -0.025 |
wanted | -0.024 |
joy | -0.023 |
walked | -0.023 |
Token | " |
Token ID | 1 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
paste | +0.091 |
web | +0.089 |
prise | +0.088 |
iders | +0.084 |
potion | +0.081 |
Neg logit contributions | |
walked | -0.056 |
wanted | -0.056 |
joy | -0.055 |
uniforms | -0.055 |
wandered | -0.055 |
Token | RE |
Token ID | 2200 |
Feature activation | -3.98e-02 |
Pos logit contributions | |
joy | +0.038 |
folk | +0.038 |
acts | +0.037 |
ones | +0.037 |
argued | +0.036 |
Neg logit contributions | |
prise | -0.035 |
paste | -0.035 |
web | -0.035 |
iders | -0.032 |
izz | -0.030 |
Token | VERS |
Token ID | 28884 |
Feature activation | -5.10e-02 |
Pos logit contributions | |
prise | +0.080 |
web | +0.078 |
paste | +0.077 |
aid | +0.071 |
iders | +0.070 |
Neg logit contributions | |
joy | -0.058 |
acts | -0.052 |
uniforms | -0.050 |
argued | -0.048 |
ened | -0.048 |
Token | E |
Token ID | 36 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
paste | +0.074 |
prise | +0.073 |
web | +0.068 |
iders | +0.063 |
potion | +0.059 |
Neg logit contributions | |
joy | -0.092 |
argued | -0.092 |
folk | -0.090 |
challenged | -0.090 |
uniforms | -0.089 |
Token | , |
Token ID | 11 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
prise | +0.359 |
paste | +0.357 |
web | +0.355 |
iders | +0.322 |
izz | +0.309 |
Neg logit contributions | |
joy | -0.325 |
acts | -0.317 |
folk | -0.315 |
argued | -0.306 |
ones | -0.303 |
Token | huh |
Token ID | 24926 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
prise | +0.031 |
web | +0.030 |
iders | +0.029 |
paste | +0.029 |
izz | +0.026 |
Neg logit contributions | |
wanted | -0.020 |
joy | -0.020 |
folk | -0.019 |
acts | -0.019 |
waved | -0.019 |
Token | ? |
Token ID | 30 |
Feature activation | -9.70e-03 |
Pos logit contributions | |
paste | +0.045 |
prise | +0.042 |
web | +0.042 |
iders | +0.041 |
potion | +0.039 |
Neg logit contributions | |
wanted | -0.034 |
ones | -0.033 |
nodded | -0.033 |
acts | -0.032 |
nods | -0.032 |
Token | That |
Token ID | 1320 |
Feature activation | -2.86e-03 |
Pos logit contributions | |
web | +0.022 |
prise | +0.021 |
paste | +0.021 |
iders | +0.020 |
izz | +0.019 |
Neg logit contributions | |
joy | -0.011 |
acts | -0.011 |
folk | -0.011 |
wanted | -0.010 |
ones | -0.010 |
Token | 's |
Token ID | 338 |
Feature activation | +5.99e-02 |
Pos logit contributions | |
web | +0.007 |
paste | +0.007 |
prise | +0.007 |
iders | +0.007 |
izz | +0.006 |
Neg logit contributions | |
joy | -0.003 |
folk | -0.003 |
argued | -0.002 |
acts | -0.002 |
ened | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | -4.50e-03 |
Pos logit contributions | |
joy | +0.005 |
acts | +0.005 |
folk | +0.004 |
nods | +0.004 |
ened | +0.004 |
Neg logit contributions | |
web | -0.007 |
paste | -0.007 |
iders | -0.006 |
prise | -0.006 |
aid | -0.006 |
Token | buses |
Token ID | 16893 |
Feature activation | +5.83e-02 |
Pos logit contributions | |
web | +0.009 |
paste | +0.009 |
prise | +0.009 |
iders | +0.008 |
aid | +0.008 |
Neg logit contributions | |
joy | -0.006 |
acts | -0.006 |
folk | -0.006 |
argued | -0.006 |
ened | -0.005 |
Token | that |
Token ID | 326 |
Feature activation | -3.46e-02 |
Pos logit contributions | |
joy | +0.110 |
acts | +0.096 |
nods | +0.091 |
folk | +0.088 |
attitude | +0.087 |
Neg logit contributions | |
paste | -0.091 |
web | -0.090 |
aid | -0.078 |
iders | -0.077 |
prise | -0.076 |
Token | went |
Token ID | 1816 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
paste | +0.083 |
web | +0.083 |
prise | +0.082 |
iders | +0.078 |
izz | +0.075 |
Neg logit contributions | |
joy | -0.035 |
folk | -0.030 |
acts | -0.030 |
argued | -0.029 |
ened | -0.027 |
Token | wh |
Token ID | 348 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
web | +0.034 |
paste | +0.031 |
prise | +0.030 |
ores | +0.028 |
iders | +0.028 |
Neg logit contributions | |
acts | -0.045 |
joy | -0.044 |
folk | -0.043 |
nods | -0.042 |
nodded | -0.042 |
Token | izz |
Token ID | 6457 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
paste | +0.014 |
prise | +0.014 |
web | +0.014 |
iders | +0.012 |
aid | +0.011 |
Neg logit contributions | |
joy | -0.023 |
folk | -0.022 |
argued | -0.022 |
acts | -0.022 |
uniforms | -0.021 |
Token | ing |
Token ID | 278 |
Feature activation | +2.63e-02 |
Pos logit contributions | |
web | +0.369 |
paste | +0.367 |
iders | +0.336 |
prise | +0.329 |
izz | +0.313 |
Neg logit contributions | |
joy | -0.332 |
acts | -0.309 |
uniforms | -0.297 |
folk | -0.295 |
argued | -0.295 |
Token | by |
Token ID | 416 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
joy | +0.036 |
acts | +0.034 |
folk | +0.034 |
argued | +0.033 |
wandered | +0.032 |
Neg logit contributions | |
web | -0.053 |
paste | -0.052 |
prise | -0.050 |
iders | -0.048 |
izz | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | +1.05e-03 |
Pos logit contributions | |
joy | +0.029 |
acts | +0.027 |
folk | +0.027 |
argued | +0.027 |
ones | +0.026 |
Neg logit contributions | |
prise | -0.060 |
paste | -0.060 |
web | -0.058 |
iders | -0.055 |
izz | -0.054 |
Token |
|
Token ID | 198 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
joy | +0.002 |
argued | +0.002 |
acts | +0.002 |
folk | +0.001 |
ened | +0.001 |
Neg logit contributions | |
web | -0.002 |
prise | -0.002 |
paste | -0.002 |
iders | -0.002 |
x | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
paste | +0.075 |
web | +0.073 |
prise | +0.073 |
iders | +0.069 |
aid | +0.066 |
Neg logit contributions | |
joy | -0.028 |
folk | -0.026 |
acts | -0.025 |
challenged | -0.024 |
uniforms | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
paste | +0.008 |
web | +0.008 |
prise | +0.008 |
iders | +0.007 |
aid | +0.007 |
Neg logit contributions | |
joy | -0.005 |
folk | -0.005 |
acts | -0.005 |
uniforms | -0.004 |
argued | -0.004 |
Token | went |
Token ID | 1816 |
Feature activation | -4.55e-02 |
Pos logit contributions | |
prise | +0.026 |
paste | +0.026 |
web | +0.026 |
iders | +0.024 |
izz | +0.023 |
Neg logit contributions | |
joy | -0.014 |
folk | -0.013 |
acts | -0.013 |
argued | -0.013 |
ones | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | -5.13e-02 |
Pos logit contributions | |
web | +0.108 |
paste | +0.108 |
prise | +0.107 |
iders | +0.102 |
izz | +0.097 |
Neg logit contributions | |
joy | -0.046 |
folk | -0.042 |
acts | -0.041 |
argued | -0.040 |
ones | -0.040 |
Token | the |
Token ID | 262 |
Feature activation | -9.61e-03 |
Pos logit contributions | |
paste | +0.136 |
web | +0.132 |
iders | +0.127 |
prise | +0.125 |
aid | +0.118 |
Neg logit contributions | |
joy | -0.048 |
argued | -0.042 |
acts | -0.042 |
market | -0.041 |
walked | -0.041 |
Token | to |
Token ID | 284 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
prise | +0.021 |
paste | +0.021 |
web | +0.020 |
iders | +0.020 |
izz | +0.020 |
Neg logit contributions | |
market | -0.012 |
joy | -0.012 |
ones | -0.011 |
bank | -0.011 |
yacht | -0.011 |
Token | aster |
Token ID | 1603 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
paste | +0.058 |
web | +0.058 |
iders | +0.056 |
prise | +0.052 |
aid | +0.049 |
Neg logit contributions | |
market | -0.022 |
walked | -0.022 |
joy | -0.021 |
yacht | -0.021 |
acts | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
paste | +0.431 |
web | +0.427 |
prise | +0.425 |
iders | +0.396 |
aid | +0.376 |
Neg logit contributions | |
joy | -0.258 |
folk | -0.251 |
acts | -0.243 |
argued | -0.237 |
ened | -0.229 |
Token | She |
Token ID | 1375 |
Feature activation | -5.68e-02 |
Pos logit contributions | |
web | +0.043 |
prise | +0.042 |
paste | +0.040 |
iders | +0.039 |
potion | +0.037 |
Neg logit contributions | |
joy | -0.009 |
They | -0.008 |
| -0.007 |
wanted | -0.006 |
His | -0.006 |
Token | saw |
Token ID | 2497 |
Feature activation | -7.22e-02 |
Pos logit contributions | |
prise | +0.127 |
web | +0.125 |
paste | +0.125 |
iders | +0.119 |
izz | +0.112 |
Neg logit contributions | |
joy | -0.063 |
acts | -0.061 |
argued | -0.061 |
folk | -0.060 |
wandered | -0.059 |
Token | that |
Token ID | 326 |
Feature activation | -6.67e-02 |
Pos logit contributions | |
prise | +0.182 |
web | +0.174 |
iders | +0.172 |
paste | +0.164 |
heads | +0.157 |
Neg logit contributions | |
ones | -0.075 |
joy | -0.069 |
acts | -0.065 |
wanted | -0.063 |
one | -0.063 |
Token | the |
Token ID | 262 |
Feature activation | +3.37e-02 |
Pos logit contributions | |
prise | +0.135 |
paste | +0.135 |
web | +0.134 |
iders | +0.125 |
izz | +0.120 |
Neg logit contributions | |
joy | -0.092 |
folk | -0.086 |
acts | -0.086 |
argued | -0.084 |
ones | -0.082 |
Token | . |
Token ID | 13 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
paste | +0.151 |
web | +0.151 |
prise | +0.148 |
iders | +0.139 |
izz | +0.133 |
Neg logit contributions | |
joy | -0.105 |
folk | -0.099 |
acts | -0.098 |
argued | -0.096 |
uniforms | -0.092 |
Token | She |
Token ID | 1375 |
Feature activation | -6.04e-02 |
Pos logit contributions | |
prise | +0.049 |
web | +0.049 |
paste | +0.047 |
iders | +0.045 |
aid | +0.043 |
Neg logit contributions | |
joy | -0.016 |
acts | -0.013 |
folk | -0.012 |
nodded | -0.012 |
wanted | -0.012 |
Token | grabbed |
Token ID | 13176 |
Feature activation | -4.16e-02 |
Pos logit contributions | |
prise | +0.169 |
paste | +0.165 |
iders | +0.163 |
web | +0.162 |
ores | +0.151 |
Neg logit contributions | |
wanted | -0.040 |
walked | -0.040 |
waved | -0.037 |
nodded | -0.037 |
wandered | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | -6.24e-03 |
Pos logit contributions | |
web | +0.116 |
prise | +0.114 |
iders | +0.110 |
paste | +0.107 |
heads | +0.100 |
Neg logit contributions | |
ened | -0.031 |
joy | -0.031 |
ones | -0.030 |
uniforms | -0.029 |
one | -0.028 |
Token | to |
Token ID | 284 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
prise | +0.012 |
web | +0.012 |
paste | +0.011 |
iders | +0.011 |
izz | +0.011 |
Neg logit contributions | |
joy | -0.009 |
acts | -0.009 |
folk | -0.009 |
ones | -0.009 |
argued | -0.009 |
Token | aster |
Token ID | 1603 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
web | +0.047 |
paste | +0.047 |
iders | +0.045 |
prise | +0.044 |
aid | +0.041 |
Neg logit contributions | |
joy | -0.009 |
walked | -0.009 |
| -0.008 |
one | -0.008 |
ones | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | -8.79e-03 |
Pos logit contributions | |
prise | +0.505 |
web | +0.502 |
paste | +0.501 |
iders | +0.474 |
izz | +0.460 |
Neg logit contributions | |
joy | -0.178 |
acts | -0.165 |
folk | -0.165 |
argued | -0.159 |
ones | -0.149 |
Token | threw |
Token ID | 9617 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
prise | +0.022 |
web | +0.021 |
iders | +0.021 |
paste | +0.021 |
izz | +0.019 |
Neg logit contributions | |
walked | -0.008 |
argued | -0.008 |
wanted | -0.008 |
waved | -0.008 |
nodded | -0.008 |
Token | it |
Token ID | 340 |
Feature activation | -6.84e-02 |
Pos logit contributions | |
prise | +0.061 |
web | +0.059 |
iders | +0.057 |
paste | +0.055 |
izz | +0.053 |
Neg logit contributions | |
ones | -0.024 |
joy | -0.022 |
ened | -0.021 |
argued | -0.021 |
uniforms | -0.021 |
Token | in |
Token ID | 287 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
prise | +0.164 |
paste | +0.155 |
web | +0.155 |
iders | +0.149 |
izz | +0.148 |
Neg logit contributions | |
argued | -0.068 |
joy | -0.068 |
ones | -0.066 |
ened | -0.066 |
walked | -0.065 |
Token | the |
Token ID | 262 |
Feature activation | +6.22e-03 |
Pos logit contributions | |
joy | +0.013 |
ones | +0.012 |
uniforms | +0.011 |
acts | +0.010 |
argued | +0.009 |
Neg logit contributions | |
prise | -0.049 |
paste | -0.049 |
iders | -0.049 |
web | -0.049 |
aid | -0.045 |
Token | and |
Token ID | 290 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
paste | +0.033 |
web | +0.033 |
prise | +0.031 |
iders | +0.030 |
potion | +0.029 |
Neg logit contributions | |
joy | -0.023 |
folk | -0.020 |
acts | -0.019 |
argued | -0.019 |
challenged | -0.019 |
Token | made |
Token ID | 925 |
Feature activation | +8.32e-03 |
Pos logit contributions | |
paste | +0.030 |
prise | +0.030 |
web | +0.030 |
iders | +0.027 |
aid | +0.027 |
Neg logit contributions | |
joy | -0.014 |
folk | -0.013 |
acts | -0.013 |
hi | -0.012 |
uniforms | -0.011 |
Token | faces |
Token ID | 6698 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
ones | +0.010 |
joy | +0.010 |
argued | +0.010 |
acts | +0.010 |
wanted | +0.010 |
Neg logit contributions | |
web | -0.018 |
prise | -0.018 |
paste | -0.017 |
iders | -0.017 |
izz | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -9.17e-03 |
Pos logit contributions | |
joy | +0.015 |
folk | +0.014 |
acts | +0.014 |
argued | +0.014 |
wandered | +0.013 |
Neg logit contributions | |
paste | -0.026 |
web | -0.025 |
prise | -0.025 |
iders | -0.023 |
izz | -0.023 |
Token | " |
Token ID | 366 |
Feature activation | +2.63e-02 |
Pos logit contributions | |
paste | +0.025 |
web | +0.025 |
prise | +0.025 |
iders | +0.024 |
izz | +0.023 |
Neg logit contributions | |
joy | -0.006 |
folk | -0.005 |
acts | -0.005 |
argued | -0.005 |
ened | -0.005 |
Token | E |
Token ID | 36 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
folk | +0.033 |
ones | +0.033 |
argued | +0.032 |
hi | +0.032 |
joy | +0.032 |
Neg logit contributions | |
web | -0.055 |
paste | -0.053 |
prise | -0.053 |
iders | -0.051 |
potion | -0.048 |
Token | w |
Token ID | 86 |
Feature activation | -6.56e-02 |
Pos logit contributions | |
paste | +0.265 |
web | +0.261 |
prise | +0.255 |
iders | +0.226 |
izz | +0.212 |
Neg logit contributions | |
joy | -0.411 |
folk | -0.403 |
acts | -0.403 |
argued | -0.398 |
ones | -0.388 |
Token | , |
Token ID | 11 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
prise | +0.142 |
web | +0.140 |
paste | +0.138 |
iders | +0.129 |
aid | +0.126 |
Neg logit contributions | |
argued | -0.077 |
joy | -0.075 |
folk | -0.073 |
ened | -0.072 |
acts | -0.072 |
Token | this |
Token ID | 428 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
yacht | +0.020 |
wanted | +0.019 |
were | +0.019 |
argued | +0.018 |
acts | +0.018 |
Neg logit contributions | |
web | -0.034 |
iders | -0.031 |
paste | -0.031 |
prise | -0.031 |
aid | -0.029 |
Token | is |
Token ID | 318 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
joy | +0.031 |
folk | +0.029 |
argued | +0.028 |
acts | +0.027 |
uniforms | +0.026 |
Neg logit contributions | |
web | -0.040 |
paste | -0.040 |
prise | -0.039 |
iders | -0.036 |
izz | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | +3.49e-02 |
Pos logit contributions | |
wanted | +0.036 |
ones | +0.035 |
argued | +0.035 |
goats | +0.034 |
yacht | +0.034 |
Neg logit contributions | |
web | -0.064 |
prise | -0.062 |
paste | -0.062 |
aid | -0.061 |
iders | -0.060 |
Token | a |
Token ID | 257 |
Feature activation | -9.48e-02 |
Pos logit contributions | |
ones | +0.004 |
one | +0.004 |
ened | +0.003 |
wanted | +0.003 |
argued | +0.003 |
Neg logit contributions | |
web | -0.011 |
prise | -0.011 |
iders | -0.011 |
paste | -0.009 |
jar | -0.009 |
Token | funny |
Token ID | 8258 |
Feature activation | -3.84e-02 |
Pos logit contributions | |
web | +0.223 |
iders | +0.220 |
izz | +0.211 |
paste | +0.209 |
prise | +0.201 |
Neg logit contributions | |
joy | -0.108 |
nodded | -0.099 |
yacht | -0.097 |
wanted | -0.096 |
way | -0.095 |
Token | face |
Token ID | 1986 |
Feature activation | +9.38e-03 |
Pos logit contributions | |
web | +0.090 |
prise | +0.089 |
paste | +0.088 |
iders | +0.085 |
izz | +0.085 |
Neg logit contributions | |
joy | -0.042 |
ones | -0.038 |
acts | -0.038 |
wanted | -0.036 |
argued | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
joy | +0.009 |
argued | +0.008 |
acts | +0.008 |
ones | +0.008 |
folk | +0.008 |
Neg logit contributions | |
prise | -0.023 |
web | -0.022 |
paste | -0.021 |
iders | -0.021 |
izz | -0.020 |
Token | " |
Token ID | 366 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
prise | +0.051 |
web | +0.051 |
paste | +0.048 |
iders | +0.046 |
izz | +0.045 |
Neg logit contributions | |
joy | -0.014 |
They | -0.012 |
nodded | -0.011 |
ones | -0.011 |
wanted | -0.011 |
Token | E |
Token ID | 36 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
ones | +0.033 |
hi | +0.032 |
wanted | +0.031 |
nodded | +0.030 |
acts | +0.030 |
Neg logit contributions | |
web | -0.050 |
prise | -0.048 |
iders | -0.046 |
paste | -0.045 |
potion | -0.042 |
Token | w |
Token ID | 86 |
Feature activation | -6.72e-02 |
Pos logit contributions | |
paste | +0.291 |
web | +0.284 |
prise | +0.270 |
iders | +0.244 |
izz | +0.231 |
Neg logit contributions | |
acts | -0.375 |
joy | -0.363 |
ones | -0.361 |
argued | -0.356 |
folk | -0.355 |
Token | , |
Token ID | 11 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
prise | +0.191 |
web | +0.184 |
paste | +0.179 |
iders | +0.176 |
jar | +0.176 |
Neg logit contributions | |
ened | -0.035 |
joy | -0.033 |
argued | -0.033 |
nodded | -0.032 |
wanted | -0.030 |
Token | this |
Token ID | 428 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
yacht | +0.018 |
were | +0.018 |
wanted | +0.018 |
sw | +0.018 |
joy | +0.017 |
Neg logit contributions | |
web | -0.029 |
iders | -0.028 |
prise | -0.026 |
izz | -0.025 |
paste | -0.025 |
Token | is |
Token ID | 318 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
joy | +0.024 |
acts | +0.023 |
folk | +0.022 |
ones | +0.022 |
argued | +0.022 |
Neg logit contributions | |
prise | -0.053 |
web | -0.053 |
paste | -0.053 |
iders | -0.050 |
izz | -0.048 |
Token | sour |
Token ID | 11348 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
wanted | +0.027 |
ones | +0.026 |
blue | +0.026 |
one | +0.026 |
sw | +0.026 |
Neg logit contributions | |
prise | -0.052 |
web | -0.052 |
iders | -0.051 |
aid | -0.049 |
izz | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
prise | +0.225 |
web | +0.223 |
paste | +0.222 |
iders | +0.210 |
izz | +0.204 |
Neg logit contributions | |
joy | -0.106 |
folk | -0.100 |
acts | -0.099 |
argued | -0.096 |
ones | -0.093 |
Token | They |
Token ID | 1119 |
Feature activation | +3.35e-02 |
Pos logit contributions | |
paste | +0.054 |
prise | +0.054 |
web | +0.054 |
iders | +0.051 |
izz | +0.050 |
Neg logit contributions | |
joy | -0.010 |
folk | -0.009 |
acts | -0.009 |
argued | -0.008 |
ened | -0.007 |
Token | said |
Token ID | 531 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
joy | +0.033 |
folk | +0.032 |
acts | +0.032 |
argued | +0.031 |
wandered | +0.029 |
Neg logit contributions | |
web | -0.079 |
prise | -0.079 |
paste | -0.079 |
iders | -0.075 |
izz | -0.071 |
Token | , |
Token ID | 11 |
Feature activation | +7.47e-03 |
Pos logit contributions | |
folk | +0.016 |
joy | +0.016 |
argued | +0.015 |
acts | +0.015 |
uniforms | +0.015 |
Neg logit contributions | |
paste | -0.024 |
prise | -0.023 |
web | -0.022 |
iders | -0.021 |
aid | -0.020 |
Token | " |
Token ID | 366 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
joy | +0.006 |
folk | +0.006 |
acts | +0.006 |
argued | +0.006 |
ened | +0.005 |
Neg logit contributions | |
paste | -0.019 |
web | -0.019 |
prise | -0.019 |
iders | -0.018 |
aid | -0.017 |
Token | E |
Token ID | 36 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
joy | +0.052 |
argued | +0.050 |
acts | +0.050 |
folk | +0.050 |
ones | +0.049 |
Neg logit contributions | |
prise | -0.067 |
paste | -0.066 |
web | -0.066 |
iders | -0.062 |
izz | -0.058 |
Token | w |
Token ID | 86 |
Feature activation | -6.79e-02 |
Pos logit contributions | |
paste | +0.272 |
web | +0.265 |
prise | +0.264 |
iders | +0.233 |
izz | +0.221 |
Neg logit contributions | |
joy | -0.399 |
acts | -0.395 |
folk | -0.389 |
argued | -0.382 |
ones | -0.379 |
Token | , |
Token ID | 11 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
prise | +0.192 |
paste | +0.186 |
web | +0.186 |
aid | +0.181 |
jar | +0.178 |
Neg logit contributions | |
joy | -0.034 |
ened | -0.028 |
argued | -0.028 |
's | -0.028 |
acts | -0.027 |
Token | gross |
Token ID | 10319 |
Feature activation | -8.83e-02 |
Pos logit contributions | |
yacht | +0.020 |
acts | +0.020 |
sw | +0.020 |
joy | +0.019 |
wanted | +0.019 |
Neg logit contributions | |
web | -0.025 |
iders | -0.023 |
paste | -0.023 |
prise | -0.023 |
izz | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
prise | +0.169 |
web | +0.165 |
paste | +0.164 |
iders | +0.150 |
izz | +0.147 |
Neg logit contributions | |
joy | -0.129 |
acts | -0.128 |
ones | -0.126 |
wanted | -0.121 |
argued | -0.120 |
Token | Y |
Token ID | 575 |
Feature activation | -6.97e-04 |
Pos logit contributions | |
web | +0.041 |
prise | +0.039 |
paste | +0.039 |
iders | +0.038 |
izz | +0.036 |
Neg logit contributions | |
acts | -0.017 |
joy | -0.016 |
folk | -0.015 |
argued | -0.015 |
ones | -0.015 |
Token | that |
Token ID | 326 |
Feature activation | -4.12e-02 |
Pos logit contributions | |
joy | +0.124 |
acts | +0.117 |
folk | +0.117 |
argued | +0.116 |
ened | +0.112 |
Neg logit contributions | |
web | -0.079 |
paste | -0.079 |
prise | -0.076 |
iders | -0.068 |
aid | -0.063 |
Token | said |
Token ID | 531 |
Feature activation | +3.66e-02 |
Pos logit contributions | |
prise | +0.095 |
paste | +0.094 |
web | +0.094 |
iders | +0.088 |
izz | +0.085 |
Neg logit contributions | |
joy | -0.043 |
folk | -0.042 |
acts | -0.042 |
argued | -0.041 |
wandered | -0.039 |
Token | " |
Token ID | 366 |
Feature activation | +4.85e-02 |
Pos logit contributions | |
joy | +0.028 |
folk | +0.026 |
acts | +0.025 |
argued | +0.025 |
ened | +0.022 |
Neg logit contributions | |
paste | -0.096 |
web | -0.095 |
prise | -0.095 |
iders | -0.090 |
izz | -0.087 |
Token | RE |
Token ID | 2200 |
Feature activation | -2.97e-02 |
Pos logit contributions | |
joy | +0.071 |
folk | +0.070 |
argued | +0.069 |
acts | +0.068 |
hi | +0.067 |
Neg logit contributions | |
paste | -0.091 |
prise | -0.091 |
web | -0.091 |
iders | -0.084 |
izz | -0.079 |
Token | VERS |
Token ID | 28884 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
prise | +0.060 |
web | +0.060 |
paste | +0.059 |
iders | +0.054 |
aid | +0.054 |
Neg logit contributions | |
joy | -0.041 |
acts | -0.038 |
folk | -0.038 |
uniforms | -0.037 |
argued | -0.036 |
Token | E |
Token ID | 36 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
prise | +0.056 |
paste | +0.056 |
web | +0.052 |
iders | +0.047 |
potion | +0.046 |
Neg logit contributions | |
folk | -0.071 |
argued | -0.071 |
challenged | -0.070 |
joy | -0.069 |
uniforms | -0.068 |
Token | ". |
Token ID | 1911 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
paste | +0.326 |
prise | +0.326 |
web | +0.324 |
iders | +0.294 |
izz | +0.279 |
Neg logit contributions | |
joy | -0.346 |
folk | -0.335 |
acts | -0.332 |
argued | -0.326 |
ones | -0.317 |
Token | L |
Token ID | 406 |
Feature activation | -4.47e-02 |
Pos logit contributions | |
joy | +0.035 |
folk | +0.033 |
acts | +0.033 |
argued | +0.032 |
uniforms | +0.030 |
Neg logit contributions | |
prise | -0.073 |
paste | -0.073 |
web | -0.073 |
iders | -0.068 |
izz | -0.066 |
Token | ila |
Token ID | 10102 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
paste | +0.053 |
prise | +0.052 |
web | +0.051 |
iders | +0.045 |
aid | +0.041 |
Neg logit contributions | |
joy | -0.099 |
folk | -0.096 |
acts | -0.096 |
argued | -0.094 |
wandered | -0.092 |
Token | and |
Token ID | 290 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
web | +0.071 |
paste | +0.070 |
prise | +0.065 |
iders | +0.065 |
izz | +0.061 |
Neg logit contributions | |
joy | -0.041 |
folk | -0.038 |
ones | -0.037 |
yacht | -0.036 |
acts | -0.035 |
Token | Tom |
Token ID | 4186 |
Feature activation | +2.44e-03 |
Pos logit contributions | |
folk | +0.047 |
joy | +0.046 |
hi | +0.045 |
ened | +0.044 |
travelled | +0.044 |
Neg logit contributions | |
paste | -0.028 |
web | -0.027 |
prise | -0.027 |
iders | -0.025 |
Rat | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
joy | +0.044 |
acts | +0.040 |
folk | +0.040 |
argued | +0.040 |
ones | +0.038 |
Neg logit contributions | |
paste | -0.083 |
prise | -0.083 |
web | -0.082 |
iders | -0.079 |
izz | -0.073 |
Token | disgusting |
Token ID | 23374 |
Feature activation | -7.91e-02 |
Pos logit contributions | |
joy | +0.045 |
folk | +0.043 |
acts | +0.043 |
argued | +0.042 |
ened | +0.041 |
Neg logit contributions | |
web | -0.052 |
prise | -0.052 |
paste | -0.051 |
aid | -0.046 |
iders | -0.045 |
Token | mud |
Token ID | 17492 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
web | +0.134 |
paste | +0.132 |
prise | +0.126 |
iders | +0.114 |
potion | +0.112 |
Neg logit contributions | |
joy | -0.131 |
folk | -0.131 |
argued | -0.125 |
acts | -0.124 |
challenged | -0.123 |
Token | . |
Token ID | 13 |
Feature activation | -2.35e-03 |
Pos logit contributions | |
web | +0.237 |
paste | +0.237 |
prise | +0.229 |
iders | +0.218 |
izz | +0.206 |
Neg logit contributions | |
joy | -0.149 |
folk | -0.141 |
acts | -0.139 |
argued | -0.132 |
uniforms | -0.130 |
Token | " |
Token ID | 366 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
paste | +0.006 |
prise | +0.006 |
web | +0.006 |
iders | +0.005 |
izz | +0.005 |
Neg logit contributions | |
joy | -0.002 |
folk | -0.002 |
acts | -0.002 |
argued | -0.002 |
uniforms | -0.002 |
Token | E |
Token ID | 36 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
wanted | +0.038 |
argued | +0.037 |
Let | +0.037 |
hi | +0.035 |
wandered | +0.035 |
Neg logit contributions | |
prise | -0.060 |
paste | -0.056 |
web | -0.056 |
iders | -0.054 |
potion | -0.054 |
Token | w |
Token ID | 86 |
Feature activation | -6.65e-02 |
Pos logit contributions | |
paste | +0.274 |
web | +0.263 |
prise | +0.257 |
iders | +0.227 |
izz | +0.214 |
Neg logit contributions | |
joy | -0.392 |
acts | -0.388 |
argued | -0.383 |
folk | -0.379 |
ones | -0.373 |
Token | , |
Token ID | 11 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
prise | +0.161 |
web | +0.152 |
paste | +0.152 |
iders | +0.145 |
ores | +0.141 |
Neg logit contributions | |
argued | -0.064 |
ened | -0.064 |
wanted | -0.061 |
joy | -0.061 |
challenged | -0.059 |
Token | y |
Token ID | 331 |
Feature activation | -7.54e-03 |
Pos logit contributions | |
wanted | +0.020 |
yacht | +0.020 |
were | +0.020 |
sw | +0.018 |
's | +0.018 |
Neg logit contributions | |
iders | -0.033 |
web | -0.033 |
prise | -0.032 |
izz | -0.030 |
paste | -0.029 |
Token | ucky |
Token ID | 5309 |
Feature activation | -9.59e-02 |
Pos logit contributions | |
web | +0.014 |
paste | +0.014 |
prise | +0.013 |
iders | +0.012 |
aid | +0.012 |
Neg logit contributions | |
joy | -0.011 |
argued | -0.011 |
ened | -0.011 |
acts | -0.011 |
hi | -0.011 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
prise | +0.165 |
paste | +0.159 |
web | +0.158 |
izz | +0.151 |
iders | +0.149 |
Neg logit contributions | |
wanted | -0.152 |
ones | -0.151 |
argued | -0.151 |
joy | -0.151 |
acts | -0.149 |
Token | . |
Token ID | 13 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
paste | +0.046 |
web | +0.045 |
prise | +0.045 |
iders | +0.042 |
aid | +0.039 |
Neg logit contributions | |
joy | -0.030 |
folk | -0.029 |
argued | -0.028 |
acts | -0.027 |
hi | -0.027 |
Token | He |
Token ID | 679 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
prise | +0.045 |
web | +0.044 |
paste | +0.043 |
iders | +0.041 |
izz | +0.040 |
Neg logit contributions | |
joy | -0.007 |
wanted | -0.006 |
They | -0.006 |
acts | -0.006 |
ones | -0.006 |
Token | says |
Token ID | 1139 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
paste | +0.036 |
web | +0.036 |
prise | +0.036 |
iders | +0.034 |
aid | +0.032 |
Neg logit contributions | |
joy | -0.019 |
folk | -0.018 |
argued | -0.015 |
hi | -0.015 |
ened | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | +8.81e-03 |
Pos logit contributions | |
joy | +0.011 |
folk | +0.010 |
acts | +0.010 |
argued | +0.010 |
ones | +0.010 |
Neg logit contributions | |
web | -0.032 |
prise | -0.032 |
paste | -0.032 |
iders | -0.031 |
izz | -0.030 |
Token | " |
Token ID | 366 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
joy | +0.006 |
folk | +0.005 |
acts | +0.005 |
argued | +0.005 |
uniforms | +0.004 |
Neg logit contributions | |
paste | -0.024 |
prise | -0.024 |
web | -0.024 |
iders | -0.023 |
izz | -0.022 |
Token | E |
Token ID | 36 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
joy | +0.062 |
folk | +0.060 |
acts | +0.059 |
argued | +0.058 |
uniforms | +0.057 |
Neg logit contributions | |
paste | -0.058 |
prise | -0.058 |
web | -0.058 |
iders | -0.052 |
izz | -0.050 |
Token | w |
Token ID | 86 |
Feature activation | -6.10e-02 |
Pos logit contributions | |
paste | +0.302 |
prise | +0.297 |
web | +0.296 |
iders | +0.263 |
izz | +0.253 |
Neg logit contributions | |
joy | -0.361 |
acts | -0.358 |
folk | -0.352 |
ones | -0.343 |
argued | -0.343 |
Token | , |
Token ID | 11 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
prise | +0.166 |
web | +0.158 |
paste | +0.155 |
aid | +0.148 |
iders | +0.148 |
Neg logit contributions | |
acts | -0.039 |
joy | -0.039 |
ones | -0.038 |
ened | -0.038 |
argued | -0.037 |
Token | a |
Token ID | 257 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
acts | +0.022 |
wanted | +0.022 |
seals | +0.022 |
sw | +0.021 |
yacht | +0.021 |
Neg logit contributions | |
web | -0.029 |
prise | -0.029 |
paste | -0.027 |
izz | -0.027 |
iders | -0.027 |
Token | lizard |
Token ID | 42406 |
Feature activation | +3.20e-02 |
Pos logit contributions | |
prise | +0.038 |
web | +0.037 |
paste | +0.037 |
iders | +0.037 |
izz | +0.036 |
Neg logit contributions | |
joy | -0.024 |
ones | -0.023 |
acts | -0.023 |
wanted | -0.022 |
yacht | -0.022 |
Token | ! |
Token ID | 0 |
Feature activation | -8.06e-03 |
Pos logit contributions | |
acts | +0.032 |
wanted | +0.030 |
joy | +0.029 |
ones | +0.029 |
argued | +0.028 |
Neg logit contributions | |
prise | -0.074 |
paste | -0.073 |
web | -0.072 |
izz | -0.070 |
iders | -0.069 |
Token | . |
Token ID | 13 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
joy | +0.008 |
folk | +0.008 |
acts | +0.008 |
argued | +0.008 |
ones | +0.007 |
Neg logit contributions | |
prise | -0.015 |
paste | -0.015 |
web | -0.015 |
iders | -0.014 |
izz | -0.013 |
Token | They |
Token ID | 1119 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
paste | +0.030 |
prise | +0.030 |
web | +0.030 |
iders | +0.029 |
izz | +0.028 |
Neg logit contributions | |
joy | -0.008 |
folk | -0.007 |
acts | -0.007 |
argued | -0.007 |
ened | -0.006 |
Token | said |
Token ID | 531 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
joy | +0.042 |
folk | +0.039 |
acts | +0.037 |
ones | +0.035 |
ened | +0.034 |
Neg logit contributions | |
paste | -0.093 |
prise | -0.092 |
web | -0.091 |
iders | -0.085 |
izz | -0.083 |
Token | , |
Token ID | 11 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
folk | +0.021 |
joy | +0.021 |
acts | +0.020 |
argued | +0.020 |
hi | +0.019 |
Neg logit contributions | |
paste | -0.038 |
prise | -0.036 |
web | -0.035 |
iders | -0.033 |
aid | -0.033 |
Token | " |
Token ID | 366 |
Feature activation | +3.76e-02 |
Pos logit contributions | |
acts | +0.010 |
joy | +0.010 |
folk | +0.010 |
argued | +0.010 |
ened | +0.009 |
Neg logit contributions | |
paste | -0.030 |
web | -0.029 |
prise | -0.028 |
iders | -0.027 |
izz | -0.026 |
Token | E |
Token ID | 36 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
joy | +0.053 |
argued | +0.053 |
folk | +0.053 |
acts | +0.052 |
ones | +0.052 |
Neg logit contributions | |
web | -0.072 |
prise | -0.072 |
paste | -0.071 |
iders | -0.068 |
izz | -0.062 |
Token | w |
Token ID | 86 |
Feature activation | -6.22e-02 |
Pos logit contributions | |
paste | +0.265 |
web | +0.260 |
prise | +0.257 |
iders | +0.228 |
izz | +0.213 |
Neg logit contributions | |
joy | -0.399 |
acts | -0.392 |
folk | -0.390 |
argued | -0.383 |
ones | -0.377 |
Token | , |
Token ID | 11 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
prise | +0.171 |
web | +0.166 |
paste | +0.166 |
aid | +0.158 |
iders | +0.157 |
Neg logit contributions | |
joy | -0.037 |
argued | -0.035 |
ened | -0.033 |
nodded | -0.030 |
acts | -0.030 |
Token | disgusting |
Token ID | 23374 |
Feature activation | -6.63e-02 |
Pos logit contributions | |
yacht | +0.024 |
sw | +0.022 |
were | +0.021 |
wanted | +0.020 |
seals | +0.020 |
Neg logit contributions | |
web | -0.037 |
iders | -0.036 |
prise | -0.035 |
paste | -0.035 |
aid | -0.033 |
Token | !" |
Token ID | 2474 |
Feature activation | +5.36e-03 |
Pos logit contributions | |
prise | +0.133 |
paste | +0.130 |
web | +0.130 |
iders | +0.122 |
izz | +0.117 |
Neg logit contributions | |
joy | -0.090 |
acts | -0.089 |
ones | -0.087 |
folk | -0.086 |
argued | -0.085 |
Token |
|
Token ID | 198 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
joy | +0.005 |
folk | +0.005 |
acts | +0.005 |
argued | +0.004 |
ened | +0.004 |
Neg logit contributions | |
paste | -0.013 |
web | -0.013 |
prise | -0.013 |
iders | -0.012 |
izz | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
prise | +0.010 |
paste | +0.009 |
web | +0.009 |
iders | +0.009 |
izz | +0.009 |
Neg logit contributions | |
joy | -0.003 |
ones | -0.002 |
acts | -0.002 |
wanted | -0.002 |
folk | -0.002 |
Token | She |
Token ID | 1375 |
Feature activation | -6.28e-02 |
Pos logit contributions | |
paste | +0.039 |
prise | +0.039 |
web | +0.038 |
iders | +0.037 |
izz | +0.035 |
Neg logit contributions | |
folk | -0.024 |
joy | -0.023 |
argued | -0.023 |
acts | -0.022 |
ened | -0.022 |
Token | says |
Token ID | 1139 |
Feature activation | +8.37e-03 |
Pos logit contributions | |
web | +0.156 |
prise | +0.156 |
paste | +0.155 |
iders | +0.146 |
izz | +0.141 |
Neg logit contributions | |
acts | -0.056 |
joy | -0.055 |
argued | -0.052 |
wanted | -0.050 |
folk | -0.050 |
Token | , |
Token ID | 11 |
Feature activation | +7.33e-03 |
Pos logit contributions | |
joy | +0.008 |
folk | +0.007 |
acts | +0.007 |
argued | +0.007 |
ones | +0.006 |
Neg logit contributions | |
web | -0.021 |
prise | -0.020 |
paste | -0.020 |
iders | -0.019 |
izz | -0.019 |
Token | " |
Token ID | 366 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
acts | +0.007 |
folk | +0.006 |
joy | +0.006 |
argued | +0.006 |
uniforms | +0.006 |
Neg logit contributions | |
paste | -0.018 |
web | -0.017 |
prise | -0.017 |
iders | -0.016 |
aid | -0.016 |
Token | E |
Token ID | 36 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
joy | +0.058 |
folk | +0.055 |
acts | +0.055 |
argued | +0.055 |
ones | +0.053 |
Neg logit contributions | |
paste | -0.055 |
web | -0.055 |
prise | -0.054 |
iders | -0.050 |
izz | -0.047 |
Token | ek |
Token ID | 988 |
Feature activation | +3.67e-02 |
Pos logit contributions | |
paste | +0.274 |
web | +0.266 |
prise | +0.262 |
iders | +0.233 |
izz | +0.220 |
Neg logit contributions | |
joy | -0.390 |
acts | -0.385 |
folk | -0.376 |
ones | -0.371 |
argued | -0.370 |
Token | , |
Token ID | 11 |
Feature activation | +6.98e-03 |
Pos logit contributions | |
joy | +0.039 |
acts | +0.038 |
ened | +0.034 |
argued | +0.034 |
ones | +0.033 |
Neg logit contributions | |
web | -0.085 |
paste | -0.085 |
prise | -0.085 |
iders | -0.080 |
izz | -0.078 |
Token | Dad |
Token ID | 17415 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
sw | +0.009 |
yacht | +0.008 |
wanted | +0.008 |
acts | +0.008 |
joy | +0.008 |
Neg logit contributions | |
web | -0.016 |
iders | -0.015 |
paste | -0.014 |
prise | -0.014 |
izz | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
paste | +0.053 |
web | +0.053 |
prise | +0.053 |
iders | +0.049 |
izz | +0.047 |
Neg logit contributions | |
joy | -0.028 |
folk | -0.027 |
acts | -0.027 |
argued | -0.026 |
uniforms | -0.025 |
Token | I |
Token ID | 314 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
wanted | +0.016 |
sw | +0.015 |
yacht | +0.015 |
were | +0.015 |
acts | +0.015 |
Neg logit contributions | |
web | -0.027 |
iders | -0.026 |
paste | -0.025 |
izz | -0.024 |
prise | -0.023 |
Token | on |
Token ID | 319 |
Feature activation | +7.30e-02 |
Pos logit contributions | |
joy | +0.029 |
folk | +0.027 |
acts | +0.027 |
argued | +0.027 |
wandered | +0.026 |
Neg logit contributions | |
prise | -0.041 |
paste | -0.041 |
web | -0.040 |
iders | -0.038 |
izz | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
ones | +0.049 |
market | +0.045 |
uniforms | +0.043 |
field | +0.042 |
yacht | +0.041 |
Neg logit contributions | |
prise | -0.213 |
paste | -0.205 |
web | -0.194 |
iders | -0.192 |
aid | -0.188 |
Token | ground |
Token ID | 2323 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
joy | +0.053 |
folk | +0.051 |
acts | +0.050 |
argued | +0.049 |
ones | +0.047 |
Neg logit contributions | |
prise | -0.101 |
paste | -0.101 |
web | -0.100 |
iders | -0.094 |
izz | -0.090 |
Token | . |
Token ID | 13 |
Feature activation | +6.22e-03 |
Pos logit contributions | |
web | +0.027 |
paste | +0.025 |
prise | +0.024 |
iders | +0.024 |
izz | +0.022 |
Neg logit contributions | |
joy | -0.020 |
folk | -0.020 |
acts | -0.019 |
argued | -0.018 |
uniforms | -0.018 |
Token | " |
Token ID | 366 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
joy | +0.007 |
folk | +0.007 |
acts | +0.007 |
argued | +0.006 |
uniforms | +0.006 |
Neg logit contributions | |
paste | -0.014 |
web | -0.014 |
prise | -0.014 |
iders | -0.013 |
izz | -0.013 |
Token | E |
Token ID | 36 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
joy | +0.043 |
argued | +0.042 |
hi | +0.042 |
wandered | +0.042 |
wanted | +0.041 |
Neg logit contributions | |
prise | -0.072 |
paste | -0.072 |
web | -0.071 |
iders | -0.068 |
potion | -0.062 |
Token | w |
Token ID | 86 |
Feature activation | -6.20e-02 |
Pos logit contributions | |
paste | +0.261 |
web | +0.250 |
prise | +0.244 |
iders | +0.217 |
izz | +0.201 |
Neg logit contributions | |
joy | -0.404 |
acts | -0.394 |
folk | -0.388 |
argued | -0.384 |
ones | -0.382 |
Token | , |
Token ID | 11 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
prise | +0.159 |
paste | +0.152 |
web | +0.151 |
iders | +0.145 |
aid | +0.140 |
Neg logit contributions | |
joy | -0.049 |
ened | -0.049 |
argued | -0.049 |
acts | -0.046 |
challenged | -0.046 |
Token | this |
Token ID | 428 |
Feature activation | +2.44e-02 |
Pos logit contributions | |
yacht | +0.028 |
wanted | +0.028 |
joy | +0.026 |
acts | +0.026 |
sw | +0.025 |
Neg logit contributions | |
web | -0.048 |
iders | -0.048 |
prise | -0.047 |
paste | -0.047 |
izz | -0.044 |
Token | glove |
Token ID | 29144 |
Feature activation | +3.59e-02 |
Pos logit contributions | |
folk | +0.045 |
joy | +0.043 |
argued | +0.041 |
hi | +0.038 |
visited | +0.038 |
Neg logit contributions | |
paste | -0.041 |
web | -0.040 |
prise | -0.037 |
iders | -0.035 |
potion | -0.034 |
Token | is |
Token ID | 318 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
acts | +0.033 |
wanted | +0.031 |
argued | +0.030 |
ones | +0.030 |
ened | +0.029 |
Neg logit contributions | |
prise | -0.088 |
paste | -0.085 |
web | -0.084 |
iders | -0.081 |
izz | -0.081 |
Token | on |
Token ID | 319 |
Feature activation | +8.37e-02 |
Pos logit contributions | |
prise | +0.101 |
paste | +0.100 |
web | +0.100 |
iders | +0.093 |
izz | +0.089 |
Neg logit contributions | |
joy | -0.066 |
folk | -0.063 |
acts | -0.062 |
argued | -0.061 |
ened | -0.059 |
Token | the |
Token ID | 262 |
Feature activation | +4.69e-02 |
Pos logit contributions | |
ones | +0.067 |
uniforms | +0.067 |
joy | +0.067 |
acts | +0.063 |
folk | +0.063 |
Neg logit contributions | |
prise | -0.221 |
paste | -0.215 |
web | -0.209 |
iders | -0.203 |
aid | -0.197 |
Token | ground |
Token ID | 2323 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
joy | +0.066 |
folk | +0.062 |
argued | +0.061 |
acts | +0.060 |
ened | +0.059 |
Neg logit contributions | |
web | -0.093 |
paste | -0.092 |
prise | -0.091 |
iders | -0.085 |
izz | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | +4.12e-03 |
Pos logit contributions | |
web | +0.034 |
paste | +0.034 |
prise | +0.034 |
iders | +0.032 |
izz | +0.030 |
Neg logit contributions | |
joy | -0.021 |
folk | -0.020 |
acts | -0.020 |
argued | -0.019 |
uniforms | -0.018 |
Token | " |
Token ID | 366 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
joy | +0.003 |
folk | +0.003 |
acts | +0.003 |
argued | +0.003 |
wandered | +0.003 |
Neg logit contributions | |
prise | -0.011 |
web | -0.011 |
paste | -0.011 |
iders | -0.010 |
izz | -0.010 |
Token | E |
Token ID | 36 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
argued | +0.045 |
wandered | +0.043 |
wanted | +0.042 |
hi | +0.042 |
challenged | +0.042 |
Neg logit contributions | |
prise | -0.061 |
paste | -0.060 |
web | -0.058 |
iders | -0.056 |
potion | -0.053 |
Token | w |
Token ID | 86 |
Feature activation | -6.21e-02 |
Pos logit contributions | |
paste | +0.259 |
web | +0.249 |
prise | +0.244 |
iders | +0.215 |
izz | +0.206 |
Neg logit contributions | |
joy | -0.402 |
acts | -0.395 |
folk | -0.385 |
argued | -0.383 |
ones | -0.377 |
Token | , |
Token ID | 11 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
prise | +0.157 |
paste | +0.147 |
web | +0.144 |
aid | +0.141 |
jar | +0.140 |
Neg logit contributions | |
joy | -0.054 |
ened | -0.052 |
argued | -0.052 |
challenged | -0.049 |
wanted | -0.048 |
Token | what |
Token ID | 644 |
Feature activation | -6.01e-02 |
Pos logit contributions | |
wanted | +0.029 |
sw | +0.027 |
yacht | +0.027 |
were | +0.027 |
's | +0.026 |
Neg logit contributions | |
iders | -0.043 |
web | -0.042 |
prise | -0.041 |
izz | -0.041 |
paste | -0.041 |
Token | is |
Token ID | 318 |
Feature activation | +5.11e-02 |
Pos logit contributions | |
paste | +0.139 |
prise | +0.139 |
web | +0.138 |
iders | +0.129 |
izz | +0.124 |
Neg logit contributions | |
joy | -0.067 |
folk | -0.063 |
argued | -0.058 |
acts | -0.058 |
hi | -0.056 |
Token | that |
Token ID | 326 |
Feature activation | -4.60e-02 |
Pos logit contributions | |
yacht | +0.080 |
joy | +0.078 |
acts | +0.078 |
argued | +0.077 |
ones | +0.077 |
Neg logit contributions | |
paste | -0.090 |
prise | -0.089 |
web | -0.086 |
iders | -0.082 |
aid | -0.080 |
Token | in |
Token ID | 287 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
joy | +0.019 |
folk | +0.018 |
acts | +0.018 |
argued | +0.017 |
hi | +0.017 |
Neg logit contributions | |
web | -0.021 |
paste | -0.020 |
prise | -0.020 |
iders | -0.019 |
izz | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
joy | +0.024 |
folk | +0.022 |
acts | +0.022 |
argued | +0.021 |
ones | +0.021 |
Neg logit contributions | |
paste | -0.050 |
web | -0.050 |
prise | -0.050 |
iders | -0.047 |
izz | -0.044 |
Token | water |
Token ID | 1660 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
joy | +0.041 |
folk | +0.040 |
argued | +0.039 |
acts | +0.039 |
ened | +0.037 |
Neg logit contributions | |
web | -0.041 |
paste | -0.040 |
prise | -0.040 |
izz | -0.036 |
iders | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | +2.82e-03 |
Pos logit contributions | |
web | +0.049 |
paste | +0.046 |
prise | +0.044 |
iders | +0.043 |
ores | +0.041 |
Neg logit contributions | |
joy | -0.050 |
acts | -0.047 |
folk | -0.047 |
nods | -0.045 |
argued | -0.044 |
Token | " |
Token ID | 366 |
Feature activation | +3.25e-02 |
Pos logit contributions | |
folk | +0.004 |
languages | +0.004 |
acts | +0.004 |
argued | +0.004 |
hi | +0.004 |
Neg logit contributions | |
Chick | -0.005 |
paste | -0.005 |
izz | -0.004 |
ores | -0.004 |
prise | -0.004 |
Token | E |
Token ID | 36 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
joy | +0.055 |
folk | +0.053 |
acts | +0.052 |
argued | +0.052 |
uniforms | +0.050 |
Neg logit contributions | |
paste | -0.055 |
web | -0.055 |
prise | -0.055 |
iders | -0.050 |
izz | -0.047 |
Token | w |
Token ID | 86 |
Feature activation | -6.30e-02 |
Pos logit contributions | |
prise | +0.217 |
web | +0.205 |
paste | +0.202 |
iders | +0.184 |
izz | +0.176 |
Neg logit contributions | |
joy | -0.454 |
folk | -0.447 |
uniforms | -0.435 |
acts | -0.432 |
argued | -0.429 |
Token | , |
Token ID | 11 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
prise | +0.153 |
paste | +0.149 |
web | +0.148 |
iders | +0.140 |
aid | +0.136 |
Neg logit contributions | |
joy | -0.060 |
argued | -0.058 |
ened | -0.056 |
acts | -0.055 |
challenged | -0.055 |
Token | that |
Token ID | 326 |
Feature activation | -3.30e-02 |
Pos logit contributions | |
yacht | +0.031 |
wanted | +0.029 |
joy | +0.029 |
acts | +0.029 |
argued | +0.028 |
Neg logit contributions | |
web | -0.039 |
paste | -0.038 |
iders | -0.036 |
prise | -0.036 |
izz | -0.034 |
Token | looks |
Token ID | 3073 |
Feature activation | +5.46e-03 |
Pos logit contributions | |
paste | +0.073 |
web | +0.073 |
prise | +0.071 |
iders | +0.068 |
potion | +0.064 |
Neg logit contributions | |
joy | -0.040 |
folk | -0.038 |
uniforms | -0.035 |
argued | -0.033 |
ened | -0.033 |
Token | gross |
Token ID | 10319 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
wanted | +0.008 |
joy | +0.008 |
yacht | +0.008 |
argued | +0.007 |
ones | +0.007 |
Neg logit contributions | |
web | -0.011 |
iders | -0.011 |
paste | -0.010 |
prise | -0.010 |
izz | -0.010 |
Token | flames |
Token ID | 16209 |
Feature activation | -9.84e-02 |
Pos logit contributions | |
prise | +0.049 |
web | +0.048 |
paste | +0.046 |
iders | +0.046 |
izz | +0.044 |
Neg logit contributions | |
acts | -0.048 |
wanted | -0.046 |
argued | -0.046 |
joy | -0.046 |
walked | -0.045 |
Token | coming |
Token ID | 2406 |
Feature activation | -5.60e-02 |
Pos logit contributions | |
paste | +0.204 |
prise | +0.204 |
web | +0.203 |
iders | +0.189 |
izz | +0.182 |
Neg logit contributions | |
joy | -0.126 |
folk | -0.121 |
acts | -0.119 |
argued | -0.117 |
uniforms | -0.112 |
Token | from |
Token ID | 422 |
Feature activation | -3.16e-02 |
Pos logit contributions | |
paste | +0.147 |
prise | +0.147 |
aid | +0.138 |
web | +0.138 |
iders | +0.137 |
Neg logit contributions | |
ones | -0.048 |
They | -0.046 |
wanted | -0.044 |
yacht | -0.043 |
were | -0.043 |
Token | the |
Token ID | 262 |
Feature activation | +5.50e-03 |
Pos logit contributions | |
paste | +0.086 |
prise | +0.084 |
aid | +0.079 |
iders | +0.079 |
web | +0.077 |
Neg logit contributions | |
ones | -0.028 |
uniforms | -0.026 |
joy | -0.026 |
market | -0.025 |
field | -0.025 |
Token | to |
Token ID | 284 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
ones | +0.007 |
joy | +0.007 |
market | +0.007 |
uniforms | +0.007 |
yacht | +0.007 |
Neg logit contributions | |
prise | -0.011 |
web | -0.011 |
paste | -0.011 |
iders | -0.011 |
izz | -0.010 |
Token | aster |
Token ID | 1603 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
paste | +0.078 |
iders | +0.073 |
web | +0.073 |
prise | +0.071 |
aid | +0.069 |
Neg logit contributions | |
ones | -0.016 |
market | -0.016 |
walked | -0.015 |
country | -0.015 |
wandered | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -9.21e-03 |
Pos logit contributions | |
paste | +0.400 |
web | +0.398 |
prise | +0.394 |
iders | +0.370 |
izz | +0.347 |
Neg logit contributions | |
joy | -0.262 |
folk | -0.253 |
acts | -0.246 |
argued | -0.238 |
ened | -0.233 |
Token | Their |
Token ID | 5334 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
prise | +0.024 |
web | +0.024 |
paste | +0.024 |
iders | +0.022 |
izz | +0.022 |
Neg logit contributions | |
joy | -0.007 |
folk | -0.007 |
acts | -0.007 |
argued | -0.006 |
uniforms | -0.006 |
Token | mom |
Token ID | 1995 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
joy | +0.018 |
argued | +0.017 |
acts | +0.017 |
uniforms | +0.017 |
ones | +0.017 |
Neg logit contributions | |
web | -0.022 |
prise | -0.021 |
paste | -0.021 |
iders | -0.020 |
izz | -0.019 |
Token | is |
Token ID | 318 |
Feature activation | +2.57e-02 |
Pos logit contributions | |
paste | +0.077 |
web | +0.076 |
prise | +0.075 |
iders | +0.071 |
aid | +0.069 |
Neg logit contributions | |
acts | -0.027 |
joy | -0.027 |
folk | -0.026 |
argued | -0.025 |
uniforms | -0.024 |
Token | trying |
Token ID | 2111 |
Feature activation | +3.89e-03 |
Pos logit contributions | |
joy | +0.035 |
folk | +0.033 |
uniforms | +0.032 |
argued | +0.032 |
walked | +0.031 |
Neg logit contributions | |
paste | -0.052 |
prise | -0.051 |
web | -0.050 |
iders | -0.049 |
aid | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | +7.89e-03 |
Pos logit contributions | |
web | +0.034 |
paste | +0.030 |
iders | +0.029 |
prise | +0.028 |
potion | +0.027 |
Neg logit contributions | |
joy | -0.033 |
acts | -0.031 |
ened | -0.031 |
argued | -0.031 |
folk | -0.030 |
Token | He |
Token ID | 679 |
Feature activation | -2.85e-03 |
Pos logit contributions | |
joy | +0.008 |
folk | +0.007 |
acts | +0.007 |
argued | +0.007 |
uniforms | +0.007 |
Neg logit contributions | |
paste | -0.019 |
web | -0.018 |
prise | -0.017 |
iders | -0.017 |
izz | -0.016 |
Token | shouted |
Token ID | 17293 |
Feature activation | +3.25e-02 |
Pos logit contributions | |
web | +0.007 |
paste | +0.007 |
prise | +0.007 |
iders | +0.006 |
izz | +0.006 |
Neg logit contributions | |
joy | -0.003 |
folk | -0.003 |
acts | -0.002 |
uniforms | -0.002 |
ones | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
joy | +0.034 |
folk | +0.033 |
acts | +0.033 |
argued | +0.032 |
uniforms | +0.030 |
Neg logit contributions | |
paste | -0.075 |
prise | -0.074 |
web | -0.074 |
iders | -0.069 |
izz | -0.067 |
Token | " |
Token ID | 366 |
Feature activation | +3.73e-02 |
Pos logit contributions | |
acts | +0.015 |
folk | +0.015 |
joy | +0.014 |
uniforms | +0.013 |
ones | +0.013 |
Neg logit contributions | |
paste | -0.032 |
web | -0.031 |
prise | -0.030 |
iders | -0.029 |
Chick | -0.028 |
Token | E |
Token ID | 36 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
joy | +0.050 |
folk | +0.048 |
acts | +0.047 |
argued | +0.047 |
hi | +0.046 |
Neg logit contributions | |
paste | -0.076 |
web | -0.076 |
prise | -0.076 |
iders | -0.071 |
izz | -0.067 |
Token | ure |
Token ID | 495 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
paste | +0.265 |
web | +0.257 |
prise | +0.253 |
iders | +0.226 |
izz | +0.214 |
Neg logit contributions | |
joy | -0.401 |
acts | -0.390 |
folk | -0.387 |
argued | -0.381 |
ones | -0.375 |
Token | ka |
Token ID | 4914 |
Feature activation | +3.85e-02 |
Pos logit contributions | |
joy | +0.071 |
folk | +0.069 |
acts | +0.068 |
argued | +0.067 |
uniforms | +0.066 |
Neg logit contributions | |
web | -0.061 |
prise | -0.061 |
paste | -0.061 |
iders | -0.055 |
izz | -0.052 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
joy | +0.073 |
folk | +0.072 |
acts | +0.071 |
argued | +0.071 |
uniforms | +0.070 |
Neg logit contributions | |
paste | -0.056 |
prise | -0.055 |
web | -0.054 |
iders | -0.048 |
izz | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | +2.58e-02 |
Pos logit contributions | |
joy | +0.014 |
folk | +0.013 |
acts | +0.012 |
argued | +0.012 |
wandered | +0.011 |
Neg logit contributions | |
prise | -0.045 |
web | -0.045 |
paste | -0.044 |
iders | -0.042 |
izz | -0.041 |
Token | grabbed |
Token ID | 13176 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
joy | +0.034 |
acts | +0.032 |
folk | +0.031 |
uniforms | +0.030 |
attitude | +0.028 |
Neg logit contributions | |
paste | -0.053 |
web | -0.053 |
prise | -0.049 |
iders | -0.047 |
aid | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
joy | +0.051 |
folk | +0.049 |
acts | +0.048 |
argued | +0.046 |
uniforms | +0.044 |
Neg logit contributions | |
paste | -0.105 |
web | -0.104 |
prise | -0.102 |
iders | -0.096 |
izz | -0.092 |
Token | he |
Token ID | 339 |
Feature activation | +4.22e-02 |
Pos logit contributions | |
joy | +0.012 |
folk | +0.012 |
acts | +0.012 |
argued | +0.011 |
uniforms | +0.011 |
Neg logit contributions | |
paste | -0.025 |
web | -0.025 |
prise | -0.024 |
iders | -0.023 |
izz | -0.022 |
Token | was |
Token ID | 373 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
joy | +0.042 |
folk | +0.040 |
acts | +0.038 |
argued | +0.036 |
ones | +0.036 |
Neg logit contributions | |
paste | -0.100 |
web | -0.100 |
prise | -0.099 |
iders | -0.093 |
izz | -0.090 |
Token | carrying |
Token ID | 6872 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
joy | +0.051 |
folk | +0.048 |
acts | +0.047 |
argued | +0.047 |
uniforms | +0.045 |
Neg logit contributions | |
paste | -0.074 |
prise | -0.074 |
web | -0.074 |
iders | -0.069 |
izz | -0.065 |
Token | L |
Token ID | 406 |
Feature activation | -9.85e-02 |
Pos logit contributions | |
prise | +0.035 |
web | +0.035 |
paste | +0.034 |
iders | +0.034 |
izz | +0.031 |
Neg logit contributions | |
joy | -0.018 |
uniforms | -0.016 |
ones | -0.016 |
ened | -0.016 |
acts | -0.015 |
Token | izz |
Token ID | 6457 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
paste | +0.078 |
prise | +0.072 |
web | +0.067 |
iders | +0.051 |
potion | +0.050 |
Neg logit contributions | |
joy | -0.248 |
acts | -0.235 |
wandered | -0.233 |
argued | -0.231 |
wanted | -0.231 |
Token | ie |
Token ID | 494 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
prise | +0.355 |
paste | +0.351 |
web | +0.346 |
iders | +0.323 |
aid | +0.299 |
Neg logit contributions | |
joy | -0.310 |
folk | -0.297 |
acts | -0.296 |
argued | -0.292 |
ened | -0.287 |
Token | â |
Token ID | 22940 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
paste | +0.238 |
prise | +0.237 |
web | +0.236 |
iders | +0.221 |
izz | +0.212 |
Neg logit contributions | |
joy | -0.149 |
folk | -0.141 |
acts | -0.140 |
argued | -0.137 |
uniforms | -0.131 |
Token | € |
Token ID | 26391 |
Feature activation | +7.70e-03 |
Pos logit contributions | |
prise | +0.031 |
paste | +0.031 |
web | +0.031 |
iders | +0.028 |
izz | +0.028 |
Neg logit contributions | |
joy | -0.023 |
folk | -0.022 |
acts | -0.022 |
argued | -0.021 |
ones | -0.021 |
Token | â„¢ |
Token ID | 8151 |
Feature activation | +2.58e-02 |
Pos logit contributions | |
acts | +0.013 |
joy | +0.013 |
folk | +0.013 |
argued | +0.013 |
ened | +0.012 |
Neg logit contributions | |
paste | -0.013 |
web | -0.013 |
prise | -0.013 |
iders | -0.012 |
izz | -0.011 |
Token | s |
Token ID | 82 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
acts | +0.042 |
joy | +0.041 |
folk | +0.040 |
argued | +0.039 |
ened | +0.039 |
Neg logit contributions | |
prise | -0.045 |
paste | -0.045 |
web | -0.044 |
iders | -0.041 |
izz | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
joy | +0.011 |
folk | +0.010 |
acts | +0.010 |
argued | +0.009 |
hi | +0.009 |
Neg logit contributions | |
paste | -0.016 |
web | -0.016 |
prise | -0.015 |
iders | -0.014 |
aid | -0.014 |
Token | She |
Token ID | 1375 |
Feature activation | -5.31e-02 |
Pos logit contributions | |
web | +0.026 |
prise | +0.026 |
paste | +0.026 |
iders | +0.024 |
aid | +0.023 |
Neg logit contributions | |
joy | -0.009 |
acts | -0.009 |
folk | -0.008 |
ones | -0.008 |
argued | -0.008 |
Token | says |
Token ID | 1139 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
paste | +0.125 |
prise | +0.123 |
web | +0.123 |
iders | +0.116 |
izz | +0.112 |
Neg logit contributions | |
joy | -0.055 |
folk | -0.053 |
argued | -0.047 |
acts | -0.047 |
ones | -0.046 |
Token | , |
Token ID | 11 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
joy | +0.015 |
folk | +0.014 |
acts | +0.014 |
argued | +0.013 |
uniforms | +0.013 |
Neg logit contributions | |
paste | -0.037 |
prise | -0.037 |
web | -0.037 |
iders | -0.034 |
izz | -0.033 |
Token | " |
Token ID | 366 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
joy | +0.007 |
folk | +0.007 |
acts | +0.007 |
argued | +0.007 |
uniforms | +0.006 |
Neg logit contributions | |
paste | -0.027 |
web | -0.027 |
prise | -0.026 |
iders | -0.025 |
izz | -0.024 |
Token | E |
Token ID | 36 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
joy | +0.062 |
folk | +0.060 |
acts | +0.060 |
argued | +0.060 |
ones | +0.059 |
Neg logit contributions | |
web | -0.077 |
paste | -0.076 |
prise | -0.076 |
iders | -0.071 |
izz | -0.067 |
Token | w |
Token ID | 86 |
Feature activation | -6.45e-02 |
Pos logit contributions | |
paste | +0.299 |
web | +0.287 |
prise | +0.282 |
iders | +0.250 |
izz | +0.242 |
Neg logit contributions | |
acts | -0.357 |
joy | -0.356 |
ones | -0.346 |
folk | -0.342 |
argued | -0.338 |
Token | , |
Token ID | 11 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
prise | +0.177 |
paste | +0.174 |
web | +0.174 |
aid | +0.166 |
iders | +0.165 |
Neg logit contributions | |
joy | -0.038 |
argued | -0.034 |
ened | -0.033 |
acts | -0.031 |
nodded | -0.031 |
Token | what |
Token ID | 644 |
Feature activation | -6.22e-02 |
Pos logit contributions | |
yacht | +0.019 |
wanted | +0.019 |
were | +0.018 |
acts | +0.018 |
goodbye | +0.018 |
Neg logit contributions | |
web | -0.033 |
iders | -0.031 |
prise | -0.031 |
paste | -0.030 |
aid | -0.029 |
Token | is |
Token ID | 318 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
prise | +0.144 |
paste | +0.144 |
web | +0.143 |
iders | +0.134 |
izz | +0.129 |
Neg logit contributions | |
joy | -0.067 |
folk | -0.064 |
acts | -0.061 |
argued | -0.060 |
ened | -0.057 |
Token | this |
Token ID | 428 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
uniforms | +0.055 |
ones | +0.055 |
yacht | +0.055 |
argued | +0.053 |
acts | +0.053 |
Neg logit contributions | |
paste | -0.075 |
prise | -0.074 |
web | -0.073 |
iders | -0.070 |
aid | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | -8.48e-03 |
Pos logit contributions | |
paste | +0.012 |
prise | +0.012 |
web | +0.012 |
iders | +0.011 |
izz | +0.011 |
Neg logit contributions | |
joy | -0.008 |
folk | -0.008 |
acts | -0.008 |
argued | -0.007 |
ened | -0.007 |
Token | Spot |
Token ID | 15899 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
paste | +0.020 |
prise | +0.020 |
web | +0.020 |
iders | +0.019 |
izz | +0.018 |
Neg logit contributions | |
joy | -0.009 |
folk | -0.008 |
acts | -0.008 |
argued | -0.008 |
uniforms | -0.007 |
Token | w |
Token ID | 266 |
Feature activation | +7.40e-03 |
Pos logit contributions | |
folk | +0.030 |
joy | +0.029 |
argued | +0.027 |
ened | +0.026 |
yacht | +0.025 |
Neg logit contributions | |
paste | -0.047 |
web | -0.046 |
prise | -0.045 |
iders | -0.044 |
x | -0.040 |
Token | ags |
Token ID | 3775 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
joy | +0.008 |
acts | +0.008 |
folk | +0.007 |
uniforms | +0.007 |
attitude | +0.007 |
Neg logit contributions | |
web | -0.017 |
paste | -0.017 |
prise | -0.016 |
iders | -0.015 |
Chick | -0.015 |
Token | his |
Token ID | 465 |
Feature activation | +5.82e-02 |
Pos logit contributions | |
joy | +0.118 |
folk | +0.113 |
acts | +0.112 |
argued | +0.108 |
ened | +0.101 |
Neg logit contributions | |
web | -0.248 |
paste | -0.247 |
prise | -0.245 |
iders | -0.230 |
izz | -0.222 |
Token | tail |
Token ID | 7894 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
folk | +0.075 |
argued | +0.073 |
acts | +0.071 |
challenged | +0.070 |
ened | +0.069 |
Neg logit contributions | |
web | -0.119 |
paste | -0.112 |
prise | -0.110 |
x | -0.106 |
iders | -0.100 |
Token | and |
Token ID | 290 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
folk | +0.150 |
joy | +0.145 |
argued | +0.141 |
acts | +0.138 |
yacht | +0.134 |
Neg logit contributions | |
paste | -0.256 |
web | -0.253 |
prise | -0.231 |
aid | -0.224 |
iders | -0.222 |
Token | bar |
Token ID | 2318 |
Feature activation | +8.97e-03 |
Pos logit contributions | |
joy | +0.028 |
folk | +0.027 |
argued | +0.025 |
hi | +0.023 |
ened | +0.023 |
Neg logit contributions | |
paste | -0.035 |
web | -0.033 |
prise | -0.032 |
aid | -0.030 |
iders | -0.030 |
Token | ks |
Token ID | 591 |
Feature activation | +7.74e-02 |
Pos logit contributions | |
folk | +0.013 |
joy | +0.013 |
challenged | +0.013 |
yacht | +0.013 |
argued | +0.013 |
Neg logit contributions | |
prise | -0.016 |
paste | -0.016 |
web | -0.015 |
aid | -0.015 |
iders | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -9.39e-03 |
Pos logit contributions | |
joy | +0.103 |
folk | +0.102 |
acts | +0.101 |
argued | +0.100 |
wandered | +0.096 |
Neg logit contributions | |
paste | -0.156 |
web | -0.149 |
prise | -0.148 |
iders | -0.137 |
aid | -0.135 |
Token | He |
Token ID | 679 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
prise | +0.025 |
web | +0.025 |
paste | +0.024 |
iders | +0.023 |
izz | +0.022 |
Neg logit contributions | |
joy | -0.008 |
folk | -0.007 |
acts | -0.007 |
uniforms | -0.006 |
ones | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
joy | +0.012 |
folk | +0.011 |
acts | +0.011 |
argued | +0.010 |
wandered | +0.010 |
Neg logit contributions | |
paste | -0.031 |
web | -0.031 |
prise | -0.031 |
iders | -0.029 |
izz | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
paste | +0.054 |
prise | +0.051 |
web | +0.051 |
iders | +0.049 |
potion | +0.046 |
Neg logit contributions | |
joy | -0.015 |
| -0.015 |
wandered | -0.013 |
walked | -0.013 |
wanted | -0.013 |
Token | The |
Token ID | 464 |
Feature activation | +3.22e-02 |
Pos logit contributions | |
paste | +0.040 |
prise | +0.038 |
web | +0.037 |
iders | +0.035 |
potion | +0.034 |
Neg logit contributions | |
wandered | -0.021 |
folk | -0.021 |
joy | -0.021 |
hi | -0.021 |
challenged | -0.021 |
Token | little |
Token ID | 1310 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
acts | +0.039 |
joy | +0.039 |
ones | +0.038 |
wanted | +0.036 |
yacht | +0.035 |
Neg logit contributions | |
paste | -0.072 |
prise | -0.070 |
iders | -0.067 |
izz | -0.066 |
web | -0.066 |
Token | girl |
Token ID | 2576 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
paste | +0.027 |
prise | +0.027 |
web | +0.027 |
iders | +0.026 |
izz | +0.025 |
Neg logit contributions | |
joy | -0.016 |
acts | -0.015 |
folk | -0.015 |
ones | -0.014 |
argued | -0.014 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
joy | +0.037 |
folk | +0.034 |
ones | +0.032 |
acts | +0.032 |
uniforms | +0.032 |
Neg logit contributions | |
web | -0.076 |
paste | -0.074 |
prise | -0.073 |
iders | -0.067 |
izz | -0.066 |
Token | and |
Token ID | 290 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
folk | +0.175 |
argued | +0.168 |
acts | +0.167 |
joy | +0.162 |
uniforms | +0.158 |
Neg logit contributions | |
paste | -0.294 |
web | -0.293 |
prise | -0.279 |
iders | -0.261 |
aid | -0.252 |
Token | kept |
Token ID | 4030 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
joy | +0.025 |
folk | +0.024 |
acts | +0.023 |
uniforms | +0.023 |
ones | +0.022 |
Neg logit contributions | |
paste | -0.047 |
web | -0.046 |
prise | -0.046 |
aid | -0.042 |
izz | -0.042 |
Token | walking |
Token ID | 6155 |
Feature activation | +3.23e-02 |
Pos logit contributions | |
joy | +0.025 |
folk | +0.024 |
acts | +0.024 |
argued | +0.023 |
uniforms | +0.022 |
Neg logit contributions | |
paste | -0.048 |
prise | -0.048 |
web | -0.048 |
iders | -0.045 |
izz | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | -1.58e-03 |
Pos logit contributions | |
joy | +0.043 |
acts | +0.042 |
folk | +0.042 |
argued | +0.041 |
challenged | +0.039 |
Neg logit contributions | |
web | -0.065 |
paste | -0.064 |
prise | -0.063 |
iders | -0.059 |
izz | -0.056 |
Token | Although |
Token ID | 4900 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
paste | +0.003 |
web | +0.003 |
prise | +0.003 |
iders | +0.003 |
izz | +0.003 |
Neg logit contributions | |
joy | -0.002 |
folk | -0.002 |
acts | -0.002 |
argued | -0.002 |
uniforms | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +2.29e-03 |
Pos logit contributions | |
web | +0.065 |
paste | +0.064 |
prise | +0.062 |
iders | +0.059 |
potion | +0.054 |
Neg logit contributions | |
joy | -0.046 |
folk | -0.044 |
acts | -0.043 |
argued | -0.043 |
wandered | -0.041 |
Token | Snow |
Token ID | 7967 |
Feature activation | -5.29e-02 |
Pos logit contributions | |
joy | +0.003 |
folk | +0.003 |
acts | +0.003 |
argued | +0.002 |
ened | +0.002 |
Neg logit contributions | |
paste | -0.005 |
web | -0.005 |
prise | -0.005 |
iders | -0.005 |
izz | -0.005 |
Token | y |
Token ID | 88 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
prise | +0.147 |
paste | +0.144 |
web | +0.141 |
iders | +0.136 |
aid | +0.133 |
Neg logit contributions | |
joy | -0.033 |
folk | -0.029 |
acts | -0.027 |
argued | -0.027 |
ened | -0.026 |
Token | was |
Token ID | 373 |
Feature activation | +3.65e-02 |
Pos logit contributions | |
joy | +0.025 |
folk | +0.024 |
acts | +0.022 |
ones | +0.021 |
uniforms | +0.021 |
Neg logit contributions | |
web | -0.052 |
paste | -0.051 |
prise | -0.049 |
iders | -0.047 |
izz | -0.044 |
Token | over |
Token ID | 625 |
Feature activation | +2.99e-02 |
Pos logit contributions | |
joy | +0.051 |
folk | +0.049 |
acts | +0.048 |
argued | +0.047 |
uniforms | +0.045 |
Neg logit contributions | |
paste | -0.073 |
prise | -0.073 |
web | -0.072 |
iders | -0.067 |
izz | -0.064 |
Token | joy |
Token ID | 2633 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
folk | +0.012 |
argued | +0.011 |
acts | +0.009 |
yacht | +0.009 |
hi | +0.008 |
Neg logit contributions | |
web | -0.087 |
prise | -0.086 |
paste | -0.085 |
izz | -0.079 |
iders | -0.079 |
Token | ed |
Token ID | 276 |
Feature activation | +8.04e-02 |
Pos logit contributions | |
yacht | +0.235 |
market | +0.224 |
bank | +0.216 |
stalls | +0.208 |
uniforms | +0.206 |
Neg logit contributions | |
paste | -0.311 |
web | -0.299 |
prise | -0.287 |
aid | -0.273 |
izz | -0.257 |
Token | that |
Token ID | 326 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
folk | +0.091 |
joy | +0.088 |
acts | +0.086 |
argued | +0.085 |
uniforms | +0.081 |
Neg logit contributions | |
web | -0.177 |
paste | -0.177 |
prise | -0.172 |
iders | -0.162 |
izz | -0.154 |
Token | he |
Token ID | 339 |
Feature activation | +4.08e-02 |
Pos logit contributions | |
paste | +0.054 |
web | +0.053 |
prise | +0.051 |
Chick | +0.046 |
izz | +0.046 |
Neg logit contributions | |
joy | -0.044 |
folk | -0.043 |
argued | -0.041 |
acts | -0.040 |
hi | -0.039 |
Token | was |
Token ID | 373 |
Feature activation | +4.24e-02 |
Pos logit contributions | |
joy | +0.040 |
folk | +0.037 |
acts | +0.036 |
argued | +0.034 |
ones | +0.033 |
Neg logit contributions | |
web | -0.100 |
paste | -0.099 |
prise | -0.098 |
iders | -0.091 |
izz | -0.089 |
Token | not |
Token ID | 407 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
acts | +0.076 |
joy | +0.075 |
folk | +0.074 |
argued | +0.072 |
uniforms | +0.071 |
Neg logit contributions | |
web | -0.067 |
prise | -0.067 |
paste | -0.066 |
izz | -0.058 |
iders | -0.057 |
Token | different |
Token ID | 1180 |
Feature activation | -2.60e-02 |
Pos logit contributions | |
paste | +0.165 |
web | +0.165 |
prise | +0.164 |
iders | +0.155 |
izz | +0.148 |
Neg logit contributions | |
joy | -0.078 |
folk | -0.071 |
acts | -0.071 |
argued | -0.069 |
ones | -0.068 |
Token | things |
Token ID | 1243 |
Feature activation | +2.96e-02 |
Pos logit contributions | |
web | +0.042 |
prise | +0.042 |
paste | +0.041 |
potion | +0.038 |
iders | +0.037 |
Neg logit contributions | |
joy | -0.041 |
folk | -0.038 |
rode | -0.037 |
traded | -0.037 |
wandered | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
joy | +0.039 |
folk | +0.037 |
acts | +0.036 |
argued | +0.036 |
wandered | +0.034 |
Neg logit contributions | |
paste | -0.061 |
web | -0.061 |
prise | -0.061 |
iders | -0.056 |
aid | -0.053 |
Token | The |
Token ID | 383 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
prise | +0.034 |
web | +0.034 |
paste | +0.033 |
iders | +0.031 |
izz | +0.030 |
Neg logit contributions | |
joy | -0.015 |
folk | -0.014 |
acts | -0.014 |
argued | -0.013 |
ones | -0.013 |
Token | dolphin |
Token ID | 44136 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
joy | +0.026 |
folk | +0.026 |
argued | +0.026 |
ened | +0.025 |
acts | +0.024 |
Neg logit contributions | |
web | -0.044 |
prise | -0.043 |
paste | -0.043 |
iders | -0.040 |
potion | -0.039 |
Token | nods |
Token ID | 34833 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
joy | +0.052 |
folk | +0.048 |
argued | +0.043 |
ened | +0.042 |
hi | +0.042 |
Neg logit contributions | |
paste | -0.082 |
web | -0.081 |
prise | -0.080 |
iders | -0.075 |
potion | -0.071 |
Token | or |
Token ID | 393 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
folk | +0.257 |
joy | +0.249 |
acts | +0.238 |
argued | +0.237 |
challenged | +0.226 |
Neg logit contributions | |
paste | -0.366 |
web | -0.365 |
prise | -0.348 |
iders | -0.325 |
ores | -0.313 |
Token | shakes |
Token ID | 29959 |
Feature activation | +6.36e-02 |
Pos logit contributions | |
paste | +0.065 |
prise | +0.065 |
web | +0.064 |
iders | +0.058 |
aid | +0.056 |
Neg logit contributions | |
joy | -0.048 |
folk | -0.047 |
argued | -0.043 |
acts | -0.043 |
ened | -0.042 |
Token | its |
Token ID | 663 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
folk | +0.084 |
joy | +0.082 |
acts | +0.078 |
argued | +0.078 |
wandered | +0.077 |
Neg logit contributions | |
paste | -0.126 |
web | -0.125 |
prise | -0.122 |
iders | -0.113 |
aid | -0.108 |
Token | head |
Token ID | 1182 |
Feature activation | +8.99e-02 |
Pos logit contributions | |
joy | +0.012 |
folk | +0.012 |
argued | +0.011 |
acts | +0.011 |
wandered | +0.011 |
Neg logit contributions | |
web | -0.026 |
paste | -0.026 |
prise | -0.026 |
iders | -0.024 |
potion | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -6.60e-03 |
Pos logit contributions | |
joy | +0.135 |
folk | +0.129 |
acts | +0.128 |
argued | +0.125 |
wandered | +0.123 |
Neg logit contributions | |
paste | -0.173 |
web | -0.165 |
prise | -0.160 |
iders | -0.150 |
aid | -0.144 |
Token | . |
Token ID | 13 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
joy | +0.085 |
folk | +0.085 |
acts | +0.082 |
uniforms | +0.081 |
argued | +0.081 |
Neg logit contributions | |
paste | -0.114 |
web | -0.110 |
prise | -0.108 |
iders | -0.101 |
aid | -0.097 |
Token | She |
Token ID | 1375 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
joy | +0.018 |
folk | +0.017 |
acts | +0.016 |
argued | +0.016 |
uniforms | +0.015 |
Neg logit contributions | |
prise | -0.050 |
paste | -0.050 |
web | -0.049 |
iders | -0.047 |
izz | -0.045 |
Token | cheered |
Token ID | 34767 |
Feature activation | +9.50e-02 |
Pos logit contributions | |
paste | +0.063 |
web | +0.063 |
prise | +0.060 |
iders | +0.057 |
izz | +0.056 |
Neg logit contributions | |
joy | -0.030 |
folk | -0.030 |
acts | -0.026 |
ones | -0.026 |
argued | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
joy | +0.097 |
folk | +0.084 |
acts | +0.083 |
ones | +0.083 |
argued | +0.081 |
Neg logit contributions | |
prise | -0.233 |
paste | -0.230 |
web | -0.228 |
iders | -0.216 |
izz | -0.214 |
Token | cl |
Token ID | 537 |
Feature activation | +8.67e-02 |
Pos logit contributions | |
paste | +0.036 |
web | +0.035 |
prise | +0.035 |
aid | +0.033 |
izz | +0.032 |
Neg logit contributions | |
folk | -0.021 |
joy | -0.021 |
acts | -0.020 |
uniforms | -0.019 |
ones | -0.019 |
Token | apped |
Token ID | 6320 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
folk | +0.057 |
acts | +0.057 |
joy | +0.056 |
ones | +0.053 |
market | +0.052 |
Neg logit contributions | |
prise | -0.237 |
paste | -0.233 |
web | -0.229 |
izz | -0.220 |
iders | -0.216 |
Token | . |
Token ID | 13 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
joy | +0.140 |
acts | +0.128 |
folk | +0.125 |
argued | +0.123 |
ones | +0.123 |
Neg logit contributions | |
prise | -0.282 |
paste | -0.280 |
web | -0.279 |
iders | -0.265 |
izz | -0.256 |
Token | " |
Token ID | 366 |
Feature activation | +3.28e-02 |
Pos logit contributions | |
prise | +0.025 |
web | +0.025 |
paste | +0.025 |
iders | +0.023 |
izz | +0.023 |
Neg logit contributions | |
joy | -0.010 |
folk | -0.009 |
acts | -0.009 |
argued | -0.009 |
ones | -0.008 |
Token | You |
Token ID | 1639 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
joy | +0.043 |
folk | +0.042 |
acts | +0.041 |
ones | +0.040 |
hi | +0.040 |
Neg logit contributions | |
prise | -0.067 |
paste | -0.066 |
web | -0.066 |
iders | -0.062 |
izz | -0.059 |
Token | 're |
Token ID | 821 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
joy | +0.021 |
folk | +0.020 |
acts | +0.019 |
argued | +0.019 |
wandered | +0.018 |
Neg logit contributions | |
web | -0.057 |
paste | -0.057 |
prise | -0.057 |
iders | -0.054 |
izz | -0.052 |
Token | awesome |
Token ID | 7427 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
joy | +0.013 |
folk | +0.012 |
ones | +0.012 |
waved | +0.012 |
challenged | +0.012 |
Neg logit contributions | |
paste | -0.027 |
web | -0.026 |
iders | -0.025 |
prise | -0.025 |
aid | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
paste | +0.032 |
prise | +0.032 |
web | +0.032 |
iders | +0.029 |
izz | +0.028 |
Neg logit contributions | |
joy | -0.016 |
folk | -0.016 |
acts | -0.015 |
argued | -0.014 |
ened | -0.014 |
Token | play |
Token ID | 711 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
joy | +0.028 |
acts | +0.027 |
folk | +0.027 |
argued | +0.026 |
wandered | +0.026 |
Neg logit contributions | |
web | -0.057 |
paste | -0.056 |
prise | -0.056 |
iders | -0.053 |
izz | -0.050 |
Token | outside |
Token ID | 2354 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
joy | +0.018 |
uniforms | +0.018 |
ones | +0.017 |
yacht | +0.017 |
walked | +0.017 |
Neg logit contributions | |
prise | -0.028 |
paste | -0.028 |
web | -0.027 |
iders | -0.026 |
izz | -0.025 |
Token | with |
Token ID | 351 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
web | +0.081 |
paste | +0.076 |
prise | +0.074 |
iders | +0.071 |
izz | +0.065 |
Neg logit contributions | |
joy | -0.085 |
folk | -0.082 |
acts | -0.081 |
argued | -0.080 |
challenged | -0.077 |
Token | her |
Token ID | 607 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
ones | +0.046 |
uniforms | +0.045 |
walked | +0.042 |
wandered | +0.041 |
yacht | +0.041 |
Neg logit contributions | |
paste | -0.111 |
prise | -0.110 |
web | -0.107 |
iders | -0.105 |
izz | -0.095 |
Token | friends |
Token ID | 2460 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
paste | +0.062 |
prise | +0.061 |
web | +0.061 |
iders | +0.057 |
izz | +0.055 |
Neg logit contributions | |
joy | -0.028 |
acts | -0.026 |
folk | -0.026 |
ones | -0.025 |
argued | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
joy | +0.199 |
folk | +0.195 |
acts | +0.184 |
argued | +0.181 |
ened | +0.177 |
Neg logit contributions | |
web | -0.163 |
paste | -0.155 |
prise | -0.153 |
iders | -0.139 |
Chick | -0.132 |
Token | collect |
Token ID | 2824 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
joy | +0.034 |
folk | +0.032 |
ened | +0.031 |
acts | +0.029 |
argued | +0.027 |
Neg logit contributions | |
prise | -0.031 |
paste | -0.029 |
web | -0.029 |
Chick | -0.026 |
potion | -0.026 |
Token | pretty |
Token ID | 2495 |
Feature activation | -7.70e-02 |
Pos logit contributions | |
paste | +0.051 |
prise | +0.051 |
web | +0.051 |
iders | +0.046 |
izz | +0.045 |
Neg logit contributions | |
joy | -0.039 |
folk | -0.038 |
argued | -0.037 |
acts | -0.036 |
hi | -0.034 |
Token | rocks |
Token ID | 12586 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
paste | +0.156 |
web | +0.154 |
prise | +0.152 |
iders | +0.137 |
potion | +0.136 |
Neg logit contributions | |
joy | -0.104 |
folk | -0.102 |
argued | -0.097 |
acts | -0.094 |
challenged | -0.090 |
Token | . |
Token ID | 13 |
Feature activation | +6.13e-03 |
Pos logit contributions | |
paste | +0.042 |
web | +0.041 |
prise | +0.040 |
iders | +0.038 |
aid | +0.036 |
Neg logit contributions | |
joy | -0.032 |
folk | -0.029 |
acts | -0.029 |
argued | -0.029 |
ened | -0.027 |
Token | He |
Token ID | 679 |
Feature activation | +8.90e-04 |
Pos logit contributions | |
argued | +0.007 |
acts | +0.007 |
joy | +0.007 |
folk | +0.007 |
ened | +0.006 |
Neg logit contributions | |
paste | -0.017 |
prise | -0.017 |
web | -0.017 |
izz | -0.016 |
iders | -0.016 |
Token | opened |
Token ID | 4721 |
Feature activation | -3.63e-02 |
Pos logit contributions | |
joy | +0.001 |
folk | +0.001 |
acts | +0.001 |
argued | +0.001 |
ones | +0.001 |
Neg logit contributions | |
web | -0.002 |
paste | -0.002 |
prise | -0.002 |
iders | -0.002 |
izz | -0.002 |
Token | each |
Token ID | 1123 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
web | +0.073 |
paste | +0.072 |
prise | +0.071 |
iders | +0.064 |
izz | +0.064 |
Neg logit contributions | |
folk | -0.047 |
joy | -0.046 |
acts | -0.045 |
argued | -0.045 |
hi | -0.044 |
Token | book |
Token ID | 1492 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
joy | +0.035 |
folk | +0.034 |
acts | +0.034 |
argued | +0.033 |
ones | +0.032 |
Neg logit contributions | |
paste | -0.042 |
prise | -0.042 |
web | -0.042 |
iders | -0.038 |
izz | -0.037 |
Token | one |
Token ID | 530 |
Feature activation | +5.82e-02 |
Pos logit contributions | |
paste | +0.034 |
web | +0.034 |
prise | +0.034 |
iders | +0.032 |
izz | +0.030 |
Neg logit contributions | |
joy | -0.015 |
folk | -0.014 |
acts | -0.014 |
argued | -0.014 |
wandered | -0.013 |
Token | by |
Token ID | 416 |
Feature activation | +8.79e-02 |
Pos logit contributions | |
joy | +0.072 |
acts | +0.071 |
folk | +0.070 |
ones | +0.067 |
argued | +0.065 |
Neg logit contributions | |
paste | -0.124 |
prise | -0.123 |
web | -0.120 |
iders | -0.116 |
izz | -0.114 |
Token | one |
Token ID | 530 |
Feature activation | +8.32e-02 |
Pos logit contributions | |
joy | +0.094 |
folk | +0.094 |
argued | +0.084 |
yacht | +0.084 |
hi | +0.084 |
Neg logit contributions | |
web | -0.195 |
prise | -0.189 |
paste | -0.185 |
iders | -0.172 |
x | -0.171 |
Token | and |
Token ID | 290 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
joy | +0.091 |
acts | +0.088 |
folk | +0.086 |
ones | +0.081 |
argued | +0.081 |
Neg logit contributions | |
paste | -0.191 |
prise | -0.189 |
web | -0.188 |
iders | -0.180 |
izz | -0.175 |
Token | carefully |
Token ID | 7773 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
joy | +0.029 |
folk | +0.027 |
acts | +0.027 |
argued | +0.026 |
uniforms | +0.025 |
Neg logit contributions | |
paste | -0.048 |
prise | -0.048 |
web | -0.048 |
iders | -0.044 |
izz | -0.042 |
Token | turned |
Token ID | 2900 |
Feature activation | -4.02e-02 |
Pos logit contributions | |
paste | +0.058 |
web | +0.057 |
prise | +0.057 |
iders | +0.052 |
izz | +0.051 |
Neg logit contributions | |
joy | -0.033 |
folk | -0.032 |
acts | -0.030 |
argued | -0.030 |
uniforms | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
paste | +0.082 |
web | +0.081 |
prise | +0.081 |
iders | +0.074 |
izz | +0.072 |
Neg logit contributions | |
joy | -0.053 |
folk | -0.051 |
acts | -0.051 |
argued | -0.050 |
wandered | -0.048 |
Token | and |
Token ID | 290 |
Feature activation | -3.76e-02 |
Pos logit contributions | |
folk | +0.020 |
argued | +0.018 |
hi | +0.018 |
acts | +0.018 |
challenged | +0.018 |
Neg logit contributions | |
prise | -0.026 |
web | -0.026 |
paste | -0.024 |
izz | -0.023 |
iders | -0.022 |
Token | D |
Token ID | 360 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
paste | +0.083 |
prise | +0.082 |
web | +0.082 |
izz | +0.076 |
iders | +0.074 |
Neg logit contributions | |
joy | -0.044 |
acts | -0.041 |
folk | -0.041 |
uniforms | -0.038 |
attitude | -0.038 |
Token | ucky |
Token ID | 5309 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
prise | +0.014 |
paste | +0.014 |
web | +0.013 |
iders | +0.012 |
izz | +0.011 |
Neg logit contributions | |
joy | -0.025 |
folk | -0.024 |
acts | -0.024 |
argued | -0.024 |
uniforms | -0.023 |
Token | row |
Token ID | 5752 |
Feature activation | +8.95e-03 |
Pos logit contributions | |
paste | +0.236 |
web | +0.232 |
prise | +0.219 |
iders | +0.204 |
Chick | +0.204 |
Neg logit contributions | |
joy | -0.129 |
folk | -0.123 |
acts | -0.110 |
attitude | -0.109 |
languages | -0.107 |
Token | ed |
Token ID | 276 |
Feature activation | +2.64e-02 |
Pos logit contributions | |
joy | +0.011 |
acts | +0.011 |
market | +0.010 |
folk | +0.010 |
uniforms | +0.010 |
Neg logit contributions | |
web | -0.019 |
paste | -0.019 |
prise | -0.017 |
izz | -0.017 |
iders | -0.017 |
Token | them |
Token ID | 606 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
folk | +0.036 |
joy | +0.036 |
acts | +0.035 |
argued | +0.034 |
challenged | +0.033 |
Neg logit contributions | |
paste | -0.051 |
web | -0.050 |
prise | -0.050 |
iders | -0.045 |
izz | -0.044 |
Token | to |
Token ID | 284 |
Feature activation | +7.12e-03 |
Pos logit contributions | |
joy | +0.025 |
acts | +0.024 |
folk | +0.024 |
uniforms | +0.022 |
goats | +0.022 |
Neg logit contributions | |
web | -0.024 |
paste | -0.024 |
prise | -0.022 |
iders | -0.021 |
aid | -0.020 |
Token | safety |
Token ID | 3747 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
joy | +0.010 |
folk | +0.009 |
acts | +0.009 |
argued | +0.009 |
ened | +0.008 |
Neg logit contributions | |
prise | -0.014 |
web | -0.014 |
paste | -0.014 |
iders | -0.013 |
izz | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +7.67e-04 |
Pos logit contributions | |
web | +0.049 |
paste | +0.048 |
prise | +0.046 |
iders | +0.043 |
izz | +0.041 |
Neg logit contributions | |
joy | -0.039 |
folk | -0.037 |
acts | -0.037 |
argued | -0.036 |
hi | -0.035 |
Token | The |
Token ID | 383 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
joy | +0.001 |
folk | +0.001 |
acts | +0.001 |
argued | +0.001 |
uniforms | +0.001 |
Neg logit contributions | |
paste | -0.002 |
prise | -0.002 |
web | -0.002 |
izz | -0.002 |
iders | -0.002 |
Token | cat |
Token ID | 3797 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
joy | +0.035 |
folk | +0.033 |
argued | +0.033 |
acts | +0.032 |
ened | +0.031 |
Neg logit contributions | |
web | -0.057 |
paste | -0.056 |
prise | -0.056 |
potion | -0.050 |
izz | -0.050 |
Token | They |
Token ID | 1119 |
Feature activation | +4.91e-02 |
Pos logit contributions | |
folk | +0.012 |
joy | +0.012 |
argued | +0.011 |
acts | +0.011 |
ened | +0.011 |
Neg logit contributions | |
paste | -0.037 |
prise | -0.037 |
web | -0.035 |
iders | -0.035 |
izz | -0.034 |
Token | like |
Token ID | 588 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
joy | +0.090 |
folk | +0.079 |
ened | +0.076 |
acts | +0.075 |
argued | +0.073 |
Neg logit contributions | |
prise | -0.080 |
paste | -0.073 |
web | -0.071 |
aid | -0.070 |
izz | -0.069 |
Token | to |
Token ID | 284 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
paste | +0.021 |
prise | +0.020 |
web | +0.020 |
aid | +0.020 |
Chick | +0.018 |
Neg logit contributions | |
folk | -0.025 |
joy | -0.025 |
ened | -0.024 |
argued | -0.024 |
attitude | -0.022 |
Token | play |
Token ID | 711 |
Feature activation | +8.00e-03 |
Pos logit contributions | |
joy | +0.028 |
folk | +0.026 |
ened | +0.024 |
acts | +0.024 |
argued | +0.023 |
Neg logit contributions | |
prise | -0.036 |
paste | -0.035 |
web | -0.034 |
iders | -0.032 |
aid | -0.031 |
Token | in |
Token ID | 287 |
Feature activation | +3.66e-02 |
Pos logit contributions | |
joy | +0.012 |
folk | +0.012 |
acts | +0.011 |
argued | +0.011 |
ened | +0.011 |
Neg logit contributions | |
web | -0.015 |
paste | -0.015 |
prise | -0.015 |
iders | -0.014 |
izz | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | +3.32e-02 |
Pos logit contributions | |
folk | +0.055 |
joy | +0.054 |
argued | +0.053 |
acts | +0.051 |
nodded | +0.051 |
Neg logit contributions | |
web | -0.068 |
prise | -0.067 |
paste | -0.061 |
iders | -0.059 |
izz | -0.059 |
Token | lab |
Token ID | 2248 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
joy | +0.033 |
folk | +0.032 |
argued | +0.031 |
acts | +0.031 |
ened | +0.029 |
Neg logit contributions | |
web | -0.079 |
prise | -0.079 |
paste | -0.077 |
iders | -0.073 |
izz | -0.071 |
Token | . |
Token ID | 13 |
Feature activation | -5.78e-04 |
Pos logit contributions | |
joy | +0.018 |
folk | +0.016 |
ened | +0.016 |
hi | +0.015 |
argued | +0.015 |
Neg logit contributions | |
web | -0.019 |
prise | -0.018 |
paste | -0.017 |
iders | -0.017 |
potion | -0.015 |
Token | The |
Token ID | 383 |
Feature activation | +2.15e-02 |
Pos logit contributions | |
paste | +0.001 |
prise | +0.001 |
web | +0.001 |
iders | +0.001 |
ores | +0.001 |
Neg logit contributions | |
joy | -0.001 |
folk | -0.001 |
argued | -0.001 |
acts | -0.001 |
ened | -0.001 |
Token | lab |
Token ID | 2248 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
joy | +0.030 |
folk | +0.028 |
acts | +0.028 |
argued | +0.028 |
ened | +0.026 |
Neg logit contributions | |
paste | -0.043 |
web | -0.043 |
prise | -0.043 |
iders | -0.040 |
izz | -0.038 |
Token | has |
Token ID | 468 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
joy | +0.047 |
folk | +0.040 |
hi | +0.040 |
ened | +0.040 |
argued | +0.039 |
Neg logit contributions | |
web | -0.054 |
prise | -0.053 |
paste | -0.052 |
iders | -0.051 |
x | -0.046 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
joy | +0.064 |
acts | +0.064 |
argued | +0.063 |
wanted | +0.062 |
folk | +0.062 |
Neg logit contributions | |
paste | -0.065 |
prise | -0.064 |
web | -0.064 |
iders | -0.059 |
izz | -0.058 |
Token | they |
Token ID | 484 |
Feature activation | +9.60e-02 |
Pos logit contributions | |
web | +0.031 |
prise | +0.031 |
paste | +0.030 |
iders | +0.029 |
izz | +0.027 |
Neg logit contributions | |
joy | -0.017 |
folk | -0.017 |
argued | -0.017 |
walked | -0.016 |
wanted | -0.016 |
Token | shouted |
Token ID | 17293 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
joy | +0.121 |
folk | +0.120 |
argued | +0.119 |
acts | +0.118 |
challenged | +0.111 |
Neg logit contributions | |
web | -0.200 |
prise | -0.199 |
paste | -0.199 |
iders | -0.186 |
izz | -0.177 |
Token | . |
Token ID | 13 |
Feature activation | +5.80e-03 |
Pos logit contributions | |
joy | +0.051 |
folk | +0.050 |
acts | +0.050 |
ones | +0.047 |
argued | +0.047 |
Neg logit contributions | |
paste | -0.078 |
prise | -0.076 |
web | -0.076 |
iders | -0.070 |
izz | -0.068 |
Token | " |
Token ID | 366 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
joy | +0.007 |
acts | +0.007 |
folk | +0.007 |
argued | +0.007 |
ened | +0.007 |
Neg logit contributions | |
paste | -0.012 |
web | -0.012 |
prise | -0.012 |
iders | -0.011 |
izz | -0.011 |
Token | You |
Token ID | 1639 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
folk | +0.052 |
joy | +0.052 |
argued | +0.051 |
acts | +0.049 |
hi | +0.048 |
Neg logit contributions | |
prise | -0.082 |
paste | -0.081 |
web | -0.080 |
iders | -0.073 |
izz | -0.071 |
Token | ate |
Token ID | 15063 |
Feature activation | -5.95e-02 |
Pos logit contributions | |
argued | +0.028 |
acts | +0.028 |
folk | +0.027 |
wanted | +0.027 |
wandered | +0.027 |
Neg logit contributions | |
web | -0.055 |
paste | -0.055 |
prise | -0.054 |
iders | -0.052 |
izz | -0.049 |
Token | our |
Token ID | 674 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
web | +0.135 |
prise | +0.135 |
paste | +0.134 |
izz | +0.130 |
iders | +0.127 |
Neg logit contributions | |
ones | -0.071 |
uniforms | -0.065 |
argued | -0.065 |
yacht | -0.064 |
goats | -0.064 |
Token | cookies |
Token ID | 14746 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
paste | +0.043 |
prise | +0.043 |
web | +0.043 |
iders | +0.040 |
izz | +0.038 |
Neg logit contributions | |
joy | -0.039 |
folk | -0.038 |
acts | -0.037 |
argued | -0.037 |
uniforms | -0.035 |
Token | ! |
Token ID | 0 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
prise | +0.021 |
web | +0.021 |
paste | +0.021 |
izz | +0.019 |
iders | +0.019 |
Neg logit contributions | |
argued | -0.015 |
joy | -0.015 |
ones | -0.015 |
folk | -0.014 |
yacht | -0.014 |
Token | You |
Token ID | 921 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
web | +0.031 |
prise | +0.031 |
paste | +0.031 |
iders | +0.029 |
izz | +0.028 |
Neg logit contributions | |
joy | -0.017 |
folk | -0.016 |
acts | -0.016 |
argued | -0.015 |
uniforms | -0.015 |
Token | I |
Token ID | 314 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
web | +0.041 |
prise | +0.041 |
paste | +0.040 |
iders | +0.038 |
izz | +0.037 |
Neg logit contributions | |
joy | -0.020 |
acts | -0.020 |
folk | -0.019 |
ones | -0.018 |
uniforms | -0.018 |
Token | don |
Token ID | 836 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
web | +0.053 |
paste | +0.052 |
iders | +0.051 |
prise | +0.050 |
izz | +0.046 |
Neg logit contributions | |
acts | -0.027 |
wanted | -0.027 |
argued | -0.026 |
joy | -0.025 |
wandered | -0.025 |
Token | 't |
Token ID | 470 |
Feature activation | -3.97e-02 |
Pos logit contributions | |
prise | +0.048 |
paste | +0.048 |
web | +0.047 |
iders | +0.042 |
aid | +0.040 |
Neg logit contributions | |
folk | -0.046 |
acts | -0.044 |
joy | -0.044 |
argued | -0.043 |
challenged | -0.043 |
Token | mind |
Token ID | 2000 |
Feature activation | +5.15e-02 |
Pos logit contributions | |
paste | +0.100 |
web | +0.100 |
iders | +0.095 |
prise | +0.094 |
izz | +0.087 |
Neg logit contributions | |
wanted | -0.040 |
argued | -0.040 |
acts | -0.035 |
ones | -0.034 |
attitude | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
joy | +0.070 |
folk | +0.069 |
acts | +0.069 |
ones | +0.068 |
uniforms | +0.067 |
Neg logit contributions | |
prise | -0.104 |
paste | -0.103 |
web | -0.100 |
iders | -0.094 |
aid | -0.092 |
Token | I |
Token ID | 314 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
web | +0.039 |
prise | +0.039 |
paste | +0.038 |
iders | +0.036 |
izz | +0.035 |
Neg logit contributions | |
acts | -0.019 |
joy | -0.019 |
folk | -0.018 |
ones | -0.017 |
uniforms | -0.017 |
Token | love |
Token ID | 1842 |
Feature activation | -4.27e-02 |
Pos logit contributions | |
web | +0.049 |
paste | +0.048 |
prise | +0.047 |
iders | +0.046 |
izz | +0.042 |
Neg logit contributions | |
acts | -0.028 |
argued | -0.027 |
wanted | -0.027 |
joy | -0.026 |
wandered | -0.025 |
Token | you |
Token ID | 345 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
paste | +0.074 |
prise | +0.071 |
web | +0.071 |
iders | +0.067 |
izz | +0.064 |
Neg logit contributions | |
ones | -0.076 |
joy | -0.072 |
uniforms | -0.071 |
acts | -0.071 |
yacht | -0.067 |
Token | , |
Token ID | 11 |
Feature activation | +8.68e-04 |
Pos logit contributions | |
joy | +0.016 |
folk | +0.015 |
acts | +0.015 |
argued | +0.015 |
ones | +0.014 |
Neg logit contributions | |
web | -0.037 |
prise | -0.037 |
paste | -0.037 |
iders | -0.034 |
izz | -0.034 |
Token | Jazz |
Token ID | 21406 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
joy | +0.001 |
acts | +0.001 |
folk | +0.001 |
argued | +0.001 |
uniforms | +0.001 |
Neg logit contributions | |
paste | -0.002 |
prise | -0.002 |
web | -0.002 |
iders | -0.001 |
izz | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
joy | +0.018 |
folk | +0.017 |
argued | +0.017 |
acts | +0.017 |
wandered | +0.017 |
Neg logit contributions | |
paste | -0.025 |
web | -0.024 |
prise | -0.023 |
iders | -0.022 |
aid | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
paste | +0.015 |
prise | +0.015 |
web | +0.015 |
iders | +0.014 |
izz | +0.014 |
Neg logit contributions | |
joy | -0.006 |
folk | -0.006 |
acts | -0.005 |
uniforms | -0.005 |
ones | -0.005 |
Token | people |
Token ID | 661 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
joy | +0.017 |
folk | +0.016 |
acts | +0.015 |
argued | +0.015 |
ones | +0.014 |
Neg logit contributions | |
paste | -0.039 |
prise | -0.039 |
web | -0.039 |
iders | -0.037 |
izz | -0.036 |
Token | in |
Token ID | 287 |
Feature activation | +5.48e-02 |
Pos logit contributions | |
joy | +0.058 |
acts | +0.053 |
folk | +0.051 |
argued | +0.051 |
ened | +0.050 |
Neg logit contributions | |
prise | -0.064 |
web | -0.063 |
paste | -0.063 |
iders | -0.056 |
izz | -0.055 |
Token | his |
Token ID | 465 |
Feature activation | +1.45e-02 |
Pos logit contributions | |
hi | +0.078 |
argued | +0.078 |
folk | +0.077 |
goats | +0.074 |
joy | +0.074 |
Neg logit contributions | |
prise | -0.096 |
web | -0.095 |
paste | -0.089 |
x | -0.084 |
izz | -0.084 |
Token | neighborhood |
Token ID | 6232 |
Feature activation | +5.95e-02 |
Pos logit contributions | |
folk | +0.022 |
argued | +0.021 |
ened | +0.020 |
joy | +0.020 |
hi | +0.020 |
Neg logit contributions | |
prise | -0.026 |
web | -0.026 |
paste | -0.024 |
x | -0.023 |
iders | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +3.09e-03 |
Pos logit contributions | |
joy | +0.105 |
acts | +0.101 |
folk | +0.098 |
nods | +0.097 |
ened | +0.095 |
Neg logit contributions | |
web | -0.086 |
paste | -0.081 |
prise | -0.080 |
else | -0.070 |
iders | -0.070 |
Token | He |
Token ID | 679 |
Feature activation | -1.85e-03 |
Pos logit contributions | |
hi | +0.004 |
mun | +0.004 |
joy | +0.004 |
goats | +0.004 |
acts | +0.004 |
Neg logit contributions | |
paste | -0.005 |
ores | -0.005 |
x | -0.005 |
iders | -0.005 |
prise | -0.004 |
Token | thought |
Token ID | 1807 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
web | +0.004 |
paste | +0.004 |
prise | +0.004 |
iders | +0.004 |
izz | +0.004 |
Neg logit contributions | |
joy | -0.002 |
folk | -0.002 |
acts | -0.002 |
attitude | -0.002 |
ones | -0.002 |
Token | this |
Token ID | 428 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
joy | +0.050 |
folk | +0.048 |
acts | +0.047 |
argued | +0.047 |
uniforms | +0.045 |
Neg logit contributions | |
paste | -0.066 |
web | -0.066 |
prise | -0.066 |
iders | -0.061 |
izz | -0.058 |
Token | would |
Token ID | 561 |
Feature activation | +9.07e-02 |
Pos logit contributions | |
joy | +0.065 |
folk | +0.055 |
argued | +0.052 |
hi | +0.049 |
ened | +0.047 |
Neg logit contributions | |
web | -0.103 |
paste | -0.095 |
prise | -0.090 |
potion | -0.088 |
x | -0.087 |
Token | make |
Token ID | 787 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
acts | +0.090 |
argued | +0.088 |
folk | +0.087 |
ones | +0.087 |
joy | +0.086 |
Neg logit contributions | |
paste | -0.216 |
web | -0.213 |
prise | -0.212 |
iders | -0.205 |
izz | -0.197 |
Token | not |
Token ID | 407 |
Feature activation | -4.44e-02 |
Pos logit contributions | |
joy | +0.070 |
argued | +0.068 |
wanted | +0.066 |
acts | +0.066 |
wandered | +0.066 |
Neg logit contributions | |
prise | -0.078 |
web | -0.077 |
paste | -0.077 |
iders | -0.072 |
izz | -0.069 |
Token | survive |
Token ID | 7866 |
Feature activation | -9.82e-02 |
Pos logit contributions | |
web | +0.109 |
iders | +0.103 |
paste | +0.101 |
prise | +0.098 |
izz | +0.093 |
Neg logit contributions | |
wanted | -0.049 |
argued | -0.046 |
acts | -0.045 |
wandered | -0.044 |
challenged | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
prise | +0.239 |
paste | +0.234 |
web | +0.228 |
iders | +0.222 |
aid | +0.215 |
Neg logit contributions | |
joy | -0.099 |
acts | -0.090 |
ones | -0.087 |
folk | -0.087 |
ened | -0.086 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -7.48e-02 |
Pos logit contributions | |
web | +0.048 |
prise | +0.048 |
paste | +0.047 |
iders | +0.045 |
potion | +0.043 |
Neg logit contributions | |
joy | -0.013 |
They | -0.010 |
wanted | -0.010 |
acts | -0.010 |
folk | -0.009 |
Token | L |
Token ID | 43 |
Feature activation | -8.89e-02 |
Pos logit contributions | |
web | +0.147 |
paste | +0.146 |
prise | +0.146 |
iders | +0.135 |
izz | +0.131 |
Neg logit contributions | |
joy | -0.103 |
folk | -0.101 |
argued | -0.100 |
acts | -0.099 |
ened | -0.095 |
Token | ila |
Token ID | 10102 |
Feature activation | -6.05e-02 |
Pos logit contributions | |
paste | +0.131 |
web | +0.127 |
prise | +0.122 |
iders | +0.118 |
aid | +0.105 |
Neg logit contributions | |
joy | -0.171 |
acts | -0.164 |
argued | -0.162 |
folk | -0.162 |
wandered | -0.161 |
Token | and |
Token ID | 290 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
web | +0.131 |
paste | +0.129 |
prise | +0.123 |
iders | +0.118 |
izz | +0.116 |
Neg logit contributions | |
joy | -0.073 |
folk | -0.071 |
argued | -0.062 |
yacht | -0.062 |
uniforms | -0.061 |
Token | Ben |
Token ID | 3932 |
Feature activation | +1.23e-05 |
Pos logit contributions | |
joy | +0.028 |
folk | +0.026 |
acts | +0.025 |
ened | +0.025 |
ones | +0.025 |
Neg logit contributions | |
prise | -0.027 |
paste | -0.026 |
web | -0.025 |
aid | -0.023 |
iders | -0.023 |
Token | liked |
Token ID | 8288 |
Feature activation | +2.77e-02 |
Pos logit contributions | |
joy | +0.000 |
folk | +0.000 |
ones | +0.000 |
attitude | +0.000 |
acts | +0.000 |
Neg logit contributions | |
prise | -0.000 |
paste | -0.000 |
web | -0.000 |
izz | -0.000 |
iders | -0.000 |
Token | to |
Token ID | 284 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
folk | +0.033 |
joy | +0.032 |
acts | +0.031 |
argued | +0.030 |
ened | +0.029 |
Neg logit contributions | |
paste | -0.060 |
prise | -0.060 |
web | -0.059 |
iders | -0.054 |
aid | -0.052 |
Token | play |
Token ID | 711 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
joy | +0.025 |
folk | +0.023 |
acts | +0.023 |
argued | +0.022 |
ened | +0.021 |
Neg logit contributions | |
paste | -0.043 |
prise | -0.043 |
web | -0.043 |
iders | -0.040 |
izz | -0.039 |
Token | walks |
Token ID | 11114 |
Feature activation | +6.04e-02 |
Pos logit contributions | |
paste | +0.145 |
prise | +0.144 |
web | +0.143 |
iders | +0.134 |
izz | +0.128 |
Neg logit contributions | |
joy | -0.085 |
folk | -0.083 |
argued | -0.077 |
hi | -0.074 |
ened | -0.072 |
Token | up |
Token ID | 510 |
Feature activation | +3.54e-02 |
Pos logit contributions | |
acts | +0.079 |
joy | +0.079 |
folk | +0.077 |
argued | +0.074 |
yacht | +0.072 |
Neg logit contributions | |
web | -0.121 |
paste | -0.120 |
prise | -0.118 |
iders | -0.111 |
aid | -0.107 |
Token | the |
Token ID | 262 |
Feature activation | +2.41e-02 |
Pos logit contributions | |
joy | +0.042 |
folk | +0.042 |
acts | +0.041 |
argued | +0.040 |
hi | +0.039 |
Neg logit contributions | |
web | -0.075 |
paste | -0.075 |
prise | -0.074 |
iders | -0.070 |
izz | -0.067 |
Token | hill |
Token ID | 12788 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
joy | +0.023 |
folk | +0.021 |
acts | +0.021 |
ones | +0.019 |
uniforms | +0.019 |
Neg logit contributions | |
prise | -0.060 |
paste | -0.060 |
web | -0.059 |
iders | -0.056 |
izz | -0.055 |
Token | with |
Token ID | 351 |
Feature activation | +4.13e-02 |
Pos logit contributions | |
goats | +0.054 |
joy | +0.054 |
folk | +0.053 |
acts | +0.053 |
argued | +0.053 |
Neg logit contributions | |
web | -0.069 |
iders | -0.062 |
paste | -0.061 |
prise | -0.059 |
ores | -0.058 |
Token | L |
Token ID | 406 |
Feature activation | -7.74e-02 |
Pos logit contributions | |
joy | +0.054 |
folk | +0.050 |
acts | +0.048 |
argued | +0.048 |
ones | +0.048 |
Neg logit contributions | |
prise | -0.089 |
web | -0.089 |
paste | -0.089 |
iders | -0.084 |
izz | -0.078 |
Token | ila |
Token ID | 10102 |
Feature activation | -6.98e-02 |
Pos logit contributions | |
paste | +0.096 |
prise | +0.094 |
web | +0.094 |
iders | +0.081 |
aid | +0.076 |
Neg logit contributions | |
joy | -0.167 |
acts | -0.160 |
folk | -0.159 |
argued | -0.156 |
wandered | -0.153 |
Token | and |
Token ID | 290 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
web | +0.149 |
paste | +0.147 |
prise | +0.140 |
iders | +0.135 |
aid | +0.124 |
Neg logit contributions | |
folk | -0.083 |
joy | -0.081 |
yacht | -0.076 |
argued | -0.076 |
uniforms | -0.076 |
Token | Tom |
Token ID | 4186 |
Feature activation | -7.50e-03 |
Pos logit contributions | |
joy | +0.022 |
folk | +0.021 |
acts | +0.020 |
argued | +0.020 |
hi | +0.020 |
Neg logit contributions | |
paste | -0.025 |
prise | -0.025 |
web | -0.025 |
iders | -0.023 |
izz | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -1.03e-03 |
Pos logit contributions | |
prise | +0.014 |
paste | +0.014 |
web | +0.013 |
iders | +0.012 |
aid | +0.012 |
Neg logit contributions | |
joy | -0.011 |
folk | -0.011 |
acts | -0.010 |
ened | -0.010 |
uniforms | -0.010 |
Token | They |
Token ID | 1119 |
Feature activation | +4.31e-02 |
Pos logit contributions | |
prise | +0.003 |
web | +0.003 |
paste | +0.003 |
iders | +0.003 |
aid | +0.003 |
Neg logit contributions | |
joy | -0.001 |
acts | -0.001 |
folk | -0.001 |
ones | -0.001 |
uniforms | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
paste | +0.012 |
web | +0.012 |
prise | +0.012 |
iders | +0.011 |
aid | +0.010 |
Neg logit contributions | |
joy | -0.006 |
folk | -0.006 |
acts | -0.006 |
argued | -0.006 |
hi | -0.006 |
Token | keep |
Token ID | 1394 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
acts | +0.030 |
folk | +0.030 |
argued | +0.030 |
joy | +0.029 |
wandered | +0.028 |
Neg logit contributions | |
paste | -0.053 |
web | -0.052 |
prise | -0.051 |
iders | -0.049 |
izz | -0.045 |
Token | practicing |
Token ID | 18207 |
Feature activation | +3.92e-02 |
Pos logit contributions | |
paste | +0.040 |
prise | +0.040 |
web | +0.040 |
iders | +0.037 |
izz | +0.035 |
Neg logit contributions | |
joy | -0.027 |
folk | -0.026 |
acts | -0.026 |
argued | -0.025 |
ones | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
joy | +0.046 |
folk | +0.044 |
acts | +0.043 |
argued | +0.043 |
wandered | +0.041 |
Neg logit contributions | |
paste | -0.085 |
web | -0.085 |
prise | -0.085 |
iders | -0.079 |
izz | -0.075 |
Token |
|
Token ID | 198 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
joy | +0.011 |
folk | +0.010 |
acts | +0.010 |
argued | +0.009 |
ones | +0.009 |
Neg logit contributions | |
web | -0.026 |
prise | -0.026 |
paste | -0.026 |
iders | -0.024 |
izz | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
paste | +0.050 |
web | +0.048 |
prise | +0.048 |
iders | +0.047 |
potion | +0.043 |
Neg logit contributions | |
| -0.017 |
joy | -0.016 |
walked | -0.014 |
wandered | -0.014 |
were | -0.013 |
Token | So |
Token ID | 2396 |
Feature activation | +5.07e-02 |
Pos logit contributions | |
paste | +0.038 |
prise | +0.036 |
web | +0.035 |
iders | +0.034 |
potion | +0.033 |
Neg logit contributions | |
folk | -0.022 |
joy | -0.022 |
wandered | -0.022 |
walked | -0.021 |
challenged | -0.021 |
Token | Ellie |
Token ID | 44801 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
joy | +0.062 |
folk | +0.059 |
acts | +0.059 |
argued | +0.057 |
ones | +0.055 |
Neg logit contributions | |
paste | -0.108 |
web | -0.108 |
prise | -0.108 |
iders | -0.101 |
izz | -0.097 |
Token | kept |
Token ID | 4030 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
paste | +0.028 |
web | +0.028 |
prise | +0.026 |
aid | +0.025 |
izz | +0.024 |
Neg logit contributions | |
joy | -0.015 |
folk | -0.014 |
ones | -0.012 |
acts | -0.012 |
uniforms | -0.012 |
Token | practicing |
Token ID | 18207 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
folk | +0.036 |
acts | +0.034 |
argued | +0.034 |
joy | +0.033 |
hi | +0.033 |
Neg logit contributions | |
web | -0.043 |
prise | -0.042 |
paste | -0.042 |
iders | -0.037 |
izz | -0.036 |
Token | , |
Token ID | 11 |
Feature activation | +7.55e-03 |
Pos logit contributions | |
folk | +0.037 |
joy | +0.037 |
argued | +0.036 |
acts | +0.036 |
hi | +0.035 |
Neg logit contributions | |
web | -0.070 |
paste | -0.070 |
prise | -0.067 |
iders | -0.064 |
potion | -0.061 |
Token | put |
Token ID | 1234 |
Feature activation | -5.64e-02 |
Pos logit contributions | |
web | +0.028 |
prise | +0.028 |
paste | +0.027 |
iders | +0.027 |
izz | +0.025 |
Neg logit contributions | |
acts | -0.018 |
argued | -0.018 |
walked | -0.017 |
nodded | -0.017 |
joy | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
prise | +0.126 |
web | +0.124 |
paste | +0.122 |
iders | +0.121 |
aid | +0.107 |
Neg logit contributions | |
joy | -0.066 |
ones | -0.066 |
ened | -0.062 |
folk | -0.062 |
acts | -0.061 |
Token | pan |
Token ID | 3425 |
Feature activation | -7.85e-02 |
Pos logit contributions | |
iders | +0.041 |
web | +0.040 |
paste | +0.040 |
prise | +0.039 |
izz | +0.036 |
Neg logit contributions | |
ones | -0.031 |
joy | -0.027 |
yacht | -0.027 |
market | -0.026 |
acts | -0.026 |
Token | in |
Token ID | 287 |
Feature activation | +2.55e-02 |
Pos logit contributions | |
prise | +0.181 |
web | +0.173 |
paste | +0.170 |
iders | +0.170 |
izz | +0.157 |
Neg logit contributions | |
ones | -0.080 |
walked | -0.079 |
argued | -0.078 |
folk | -0.077 |
wandered | -0.076 |
Token | the |
Token ID | 262 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
joy | +0.027 |
ones | +0.025 |
acts | +0.024 |
folk | +0.024 |
argued | +0.024 |
Neg logit contributions | |
paste | -0.060 |
prise | -0.060 |
web | -0.059 |
iders | -0.057 |
aid | -0.053 |
Token | oven |
Token ID | 14361 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
joy | +0.029 |
acts | +0.028 |
folk | +0.028 |
ones | +0.028 |
argued | +0.027 |
Neg logit contributions | |
web | -0.029 |
prise | -0.029 |
paste | -0.029 |
iders | -0.027 |
izz | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -9.98e-03 |
Pos logit contributions | |
prise | +0.241 |
web | +0.240 |
paste | +0.239 |
iders | +0.225 |
izz | +0.216 |
Neg logit contributions | |
joy | -0.123 |
folk | -0.118 |
acts | -0.116 |
argued | -0.114 |
ones | -0.110 |
Token | They |
Token ID | 1119 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
prise | +0.028 |
web | +0.028 |
paste | +0.027 |
iders | +0.026 |
izz | +0.025 |
Neg logit contributions | |
joy | -0.006 |
folk | -0.005 |
acts | -0.005 |
argued | -0.005 |
ones | -0.005 |
Token | set |
Token ID | 900 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
acts | +0.045 |
argued | +0.044 |
folk | +0.043 |
walked | +0.042 |
nodded | +0.042 |
Neg logit contributions | |
web | -0.076 |
paste | -0.073 |
prise | -0.073 |
iders | -0.070 |
izz | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
ones | +0.018 |
joy | +0.018 |
ened | +0.017 |
folk | +0.017 |
argued | +0.017 |
Neg logit contributions | |
prise | -0.057 |
web | -0.056 |
paste | -0.056 |
iders | -0.054 |
aid | -0.050 |
Token | timer |
Token ID | 19781 |
Feature activation | -5.83e-02 |
Pos logit contributions | |
paste | +0.026 |
web | +0.025 |
iders | +0.025 |
prise | +0.025 |
izz | +0.023 |
Neg logit contributions | |
ones | -0.020 |
market | -0.019 |
yacht | -0.019 |
joy | -0.018 |
stalls | -0.018 |
Token | helped |
Token ID | 4193 |
Feature activation | +5.04e-02 |
Pos logit contributions | |
joy | +0.114 |
ones | +0.098 |
acts | +0.097 |
folk | +0.094 |
stalls | +0.093 |
Neg logit contributions | |
prise | -0.173 |
web | -0.164 |
else | -0.162 |
paste | -0.159 |
izz | -0.149 |
Token | him |
Token ID | 683 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
joy | +0.054 |
argued | +0.054 |
acts | +0.052 |
folk | +0.052 |
uniforms | +0.051 |
Neg logit contributions | |
web | -0.116 |
prise | -0.116 |
paste | -0.115 |
iders | -0.108 |
izz | -0.103 |
Token | clean |
Token ID | 3424 |
Feature activation | -5.31e-02 |
Pos logit contributions | |
prise | +0.062 |
web | +0.062 |
paste | +0.062 |
iders | +0.058 |
izz | +0.056 |
Neg logit contributions | |
acts | -0.034 |
joy | -0.033 |
argued | -0.033 |
challenged | -0.032 |
wandered | -0.032 |
Token | up |
Token ID | 510 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
paste | +0.115 |
web | +0.115 |
prise | +0.115 |
iders | +0.107 |
izz | +0.103 |
Neg logit contributions | |
joy | -0.064 |
folk | -0.061 |
acts | -0.060 |
argued | -0.059 |
wandered | -0.056 |
Token | the |
Token ID | 262 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
joy | +0.039 |
folk | +0.036 |
acts | +0.036 |
argued | +0.035 |
ones | +0.035 |
Neg logit contributions | |
prise | -0.091 |
paste | -0.091 |
web | -0.090 |
iders | -0.085 |
izz | -0.082 |
Token | mess |
Token ID | 2085 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
joy | +0.038 |
acts | +0.037 |
folk | +0.036 |
ones | +0.036 |
yacht | +0.035 |
Neg logit contributions | |
prise | -0.053 |
paste | -0.052 |
web | -0.051 |
iders | -0.049 |
izz | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | +1.97e-04 |
Pos logit contributions | |
prise | +0.312 |
web | +0.303 |
paste | +0.303 |
iders | +0.287 |
izz | +0.278 |
Neg logit contributions | |
folk | -0.131 |
joy | -0.129 |
acts | -0.124 |
argued | -0.122 |
ones | -0.119 |
Token | They |
Token ID | 1119 |
Feature activation | +5.06e-02 |
Pos logit contributions | |
joy | +0.000 |
acts | +0.000 |
folk | +0.000 |
argued | +0.000 |
ened | +0.000 |
Neg logit contributions | |
paste | -0.000 |
prise | -0.000 |
web | -0.000 |
iders | -0.000 |
izz | -0.000 |
Token | used |
Token ID | 973 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
joy | +0.056 |
folk | +0.051 |
acts | +0.050 |
ones | +0.046 |
uniforms | +0.046 |
Neg logit contributions | |
prise | -0.117 |
paste | -0.117 |
web | -0.116 |
iders | -0.108 |
izz | -0.106 |
Token | a |
Token ID | 257 |
Feature activation | -6.96e-02 |
Pos logit contributions | |
prise | +0.036 |
web | +0.036 |
paste | +0.036 |
iders | +0.034 |
izz | +0.032 |
Neg logit contributions | |
joy | -0.022 |
folk | -0.021 |
acts | -0.020 |
argued | -0.020 |
uniforms | -0.020 |
Token | bucket |
Token ID | 19236 |
Feature activation | -2.35e-03 |
Pos logit contributions | |
web | +0.144 |
iders | +0.142 |
paste | +0.140 |
prise | +0.138 |
izz | +0.133 |
Neg logit contributions | |
joy | -0.095 |
yacht | -0.087 |
ones | -0.084 |
uniforms | -0.084 |
way | -0.084 |
Token | 's |
Token ID | 338 |
Feature activation | +9.20e-02 |
Pos logit contributions | |
joy | +0.008 |
folk | +0.006 |
argued | +0.005 |
uniforms | +0.005 |
acts | +0.005 |
Neg logit contributions | |
web | -0.022 |
paste | -0.022 |
prise | -0.022 |
iders | -0.021 |
izz | -0.020 |
Token | time |
Token ID | 640 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
joy | +0.088 |
ones | +0.082 |
argued | +0.080 |
uniforms | +0.080 |
wandered | +0.080 |
Neg logit contributions | |
paste | -0.221 |
prise | -0.220 |
web | -0.219 |
iders | -0.209 |
aid | -0.201 |
Token | to |
Token ID | 284 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
joy | +0.030 |
folk | +0.028 |
acts | +0.028 |
argued | +0.027 |
uniforms | +0.026 |
Neg logit contributions | |
paste | -0.067 |
web | -0.067 |
prise | -0.067 |
iders | -0.063 |
izz | -0.060 |
Token | go |
Token ID | 467 |
Feature activation | -5.44e-02 |
Pos logit contributions | |
argued | +0.017 |
acts | +0.017 |
joy | +0.017 |
uniforms | +0.016 |
wandered | +0.016 |
Neg logit contributions | |
web | -0.035 |
paste | -0.035 |
prise | -0.034 |
iders | -0.033 |
aid | -0.031 |
Token | pot |
Token ID | 1787 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
paste | +0.134 |
web | +0.130 |
iders | +0.129 |
prise | +0.127 |
aid | +0.116 |
Neg logit contributions | |
wandered | -0.053 |
ones | -0.052 |
walked | -0.052 |
joy | -0.051 |
uniforms | -0.050 |
Token | ty |
Token ID | 774 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
prise | +0.226 |
web | +0.226 |
paste | +0.221 |
iders | +0.206 |
aid | +0.199 |
Neg logit contributions | |
yacht | -0.103 |
ones | -0.101 |
bank | -0.100 |
uniforms | -0.100 |
challenged | -0.100 |
Token | . |
Token ID | 13 |
Feature activation | -1.00e-03 |
Pos logit contributions | |
prise | +0.225 |
paste | +0.224 |
web | +0.224 |
iders | +0.208 |
izz | +0.200 |
Neg logit contributions | |
joy | -0.150 |
folk | -0.145 |
acts | -0.144 |
argued | -0.142 |
ones | -0.137 |
Token | Let |
Token ID | 3914 |
Feature activation | +3.92e-02 |
Pos logit contributions | |
prise | +0.002 |
web | +0.002 |
paste | +0.002 |
iders | +0.002 |
aid | +0.002 |
Neg logit contributions | |
joy | -0.001 |
folk | -0.001 |
acts | -0.001 |
uniforms | -0.001 |
argued | -0.001 |
Token | 's |
Token ID | 338 |
Feature activation | +4.20e-02 |
Pos logit contributions | |
joy | +0.032 |
folk | +0.027 |
acts | +0.026 |
argued | +0.023 |
uniforms | +0.022 |
Neg logit contributions | |
paste | -0.102 |
prise | -0.102 |
web | -0.101 |
izz | -0.096 |
iders | -0.095 |
Token | go |
Token ID | 467 |
Feature activation | -4.92e-02 |
Pos logit contributions | |
yacht | +0.034 |
argued | +0.032 |
wandered | +0.031 |
bank | +0.031 |
walked | +0.031 |
Neg logit contributions | |
web | -0.111 |
paste | -0.110 |
iders | -0.108 |
prise | -0.104 |
aid | -0.098 |
Token | to |
Token ID | 284 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
paste | +0.109 |
web | +0.109 |
iders | +0.106 |
prise | +0.105 |
aid | +0.094 |
Neg logit contributions | |
joy | -0.054 |
wandered | -0.053 |
yacht | -0.052 |
walked | -0.052 |
ones | -0.052 |
Token | burning |
Token ID | 9482 |
Feature activation | -4.60e-02 |
Pos logit contributions | |
prise | +0.067 |
web | +0.063 |
paste | +0.061 |
iders | +0.061 |
izz | +0.057 |
Neg logit contributions | |
wanted | -0.050 |
argued | -0.048 |
nodded | -0.047 |
acts | -0.047 |
ones | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
prise | +0.102 |
paste | +0.100 |
web | +0.100 |
iders | +0.094 |
izz | +0.090 |
Neg logit contributions | |
joy | -0.054 |
folk | -0.052 |
acts | -0.051 |
argued | -0.051 |
ened | -0.049 |
Token | She |
Token ID | 1375 |
Feature activation | -5.25e-02 |
Pos logit contributions | |
web | +0.070 |
prise | +0.070 |
paste | +0.067 |
iders | +0.064 |
potion | +0.062 |
Neg logit contributions | |
joy | -0.017 |
They | -0.014 |
wanted | -0.013 |
folk | -0.012 |
acts | -0.012 |
Token | opened |
Token ID | 4721 |
Feature activation | -5.46e-02 |
Pos logit contributions | |
prise | +0.141 |
paste | +0.140 |
web | +0.138 |
iders | +0.137 |
ores | +0.127 |
Neg logit contributions | |
wanted | -0.043 |
walked | -0.041 |
nodded | -0.037 |
argued | -0.037 |
waved | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
prise | +0.134 |
paste | +0.132 |
web | +0.131 |
iders | +0.126 |
aid | +0.119 |
Neg logit contributions | |
joy | -0.056 |
ened | -0.050 |
ones | -0.049 |
folk | -0.049 |
argued | -0.048 |
Token | oven |
Token ID | 14361 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
web | +0.027 |
iders | +0.027 |
prise | +0.027 |
paste | +0.026 |
izz | +0.024 |
Neg logit contributions | |
joy | -0.018 |
ones | -0.018 |
market | -0.018 |
uniforms | -0.017 |
bank | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | -4.58e-02 |
Pos logit contributions | |
paste | +0.296 |
web | +0.294 |
prise | +0.291 |
iders | +0.274 |
izz | +0.262 |
Neg logit contributions | |
joy | -0.156 |
folk | -0.150 |
acts | -0.146 |
argued | -0.142 |
hi | -0.137 |
Token | saw |
Token ID | 2497 |
Feature activation | -8.15e-02 |
Pos logit contributions | |
web | +0.126 |
iders | +0.125 |
paste | +0.125 |
prise | +0.125 |
mmm | +0.116 |
Neg logit contributions | |
walked | -0.033 |
wanted | -0.028 |
waved | -0.028 |
nodded | -0.026 |
argued | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | +5.65e-03 |
Pos logit contributions | |
prise | +0.213 |
web | +0.210 |
paste | +0.203 |
iders | +0.198 |
heads | +0.198 |
Neg logit contributions | |
were | -0.067 |
ones | -0.066 |
seals | -0.060 |
folk | -0.059 |
acts | -0.058 |
Token | cake |
Token ID | 12187 |
Feature activation | -7.92e-02 |
Pos logit contributions | |
joy | +0.009 |
ones | +0.009 |
market | +0.009 |
uniforms | +0.009 |
yacht | +0.009 |
Neg logit contributions | |
prise | -0.010 |
web | -0.010 |
paste | -0.010 |
iders | -0.009 |
izz | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
prise | +0.182 |
paste | +0.174 |
web | +0.174 |
aid | +0.167 |
iders | +0.166 |
Neg logit contributions | |
ones | -0.079 |
argued | -0.078 |
joy | -0.078 |
folk | -0.077 |
ened | -0.077 |
Token | said |
Token ID | 531 |
Feature activation | +6.15e-02 |
Pos logit contributions | |
paste | +0.046 |
web | +0.044 |
prise | +0.044 |
iders | +0.041 |
izz | +0.039 |
Neg logit contributions | |
joy | -0.033 |
folk | -0.032 |
acts | -0.030 |
ones | -0.029 |
yacht | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
joy | +0.086 |
folk | +0.084 |
acts | +0.083 |
argued | +0.080 |
uniforms | +0.079 |
Neg logit contributions | |
paste | -0.122 |
web | -0.119 |
prise | -0.117 |
iders | -0.109 |
aid | -0.105 |
Token |
|
Token ID | 198 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
joy | +0.027 |
folk | +0.025 |
acts | +0.025 |
argued | +0.024 |
ones | +0.023 |
Neg logit contributions | |
paste | -0.054 |
prise | -0.054 |
web | -0.054 |
iders | -0.050 |
izz | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
paste | +0.042 |
web | +0.040 |
prise | +0.040 |
iders | +0.038 |
potion | +0.036 |
Neg logit contributions | |
joy | -0.010 |
| -0.008 |
They | -0.008 |
uniforms | -0.008 |
walked | -0.008 |
Token | " |
Token ID | 1 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
paste | +0.034 |
web | +0.033 |
prise | +0.032 |
iders | +0.030 |
aid | +0.029 |
Neg logit contributions | |
joy | -0.025 |
uniforms | -0.024 |
folk | -0.023 |
acts | -0.023 |
nodded | -0.023 |
Token | Daddy |
Token ID | 48280 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
nodded | +0.041 |
ones | +0.040 |
folk | +0.040 |
joy | +0.040 |
wanted | +0.040 |
Neg logit contributions | |
prise | -0.063 |
web | -0.061 |
paste | -0.061 |
iders | -0.057 |
izz | -0.056 |
Token | , |
Token ID | 11 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
web | +0.259 |
paste | +0.258 |
prise | +0.257 |
iders | +0.245 |
izz | +0.236 |
Neg logit contributions | |
joy | -0.096 |
acts | -0.092 |
folk | -0.089 |
argued | -0.087 |
nodded | -0.082 |
Token | Lily |
Token ID | 20037 |
Feature activation | -3.97e-02 |
Pos logit contributions | |
wanted | +0.022 |
yacht | +0.020 |
acts | +0.020 |
ones | +0.020 |
nodded | +0.020 |
Neg logit contributions | |
web | -0.039 |
iders | -0.039 |
paste | -0.036 |
izz | -0.036 |
prise | -0.036 |
Token | was |
Token ID | 373 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
paste | +0.086 |
web | +0.086 |
prise | +0.085 |
iders | +0.079 |
izz | +0.076 |
Neg logit contributions | |
joy | -0.047 |
folk | -0.046 |
argued | -0.043 |
acts | -0.043 |
uniforms | -0.041 |
Token | trying |
Token ID | 2111 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
joy | +0.026 |
folk | +0.022 |
ones | +0.022 |
challenged | +0.021 |
uniforms | +0.021 |
Neg logit contributions | |
prise | -0.043 |
paste | -0.042 |
web | -0.042 |
iders | -0.041 |
aid | -0.039 |
Token | to |
Token ID | 284 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
prise | +0.033 |
paste | +0.032 |
web | +0.032 |
izz | +0.032 |
iders | +0.032 |
Neg logit contributions | |
joy | -0.006 |
ones | -0.006 |
uniforms | -0.005 |
acts | -0.005 |
wanted | -0.005 |