Token | clown |
Token ID | 25573 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
makeup | +0.079 |
vel | +0.078 |
face | +0.078 |
spit | +0.076 |
blush | +0.070 |
Neg logit contributions | |
rob | -0.176 |
zag | -0.166 |
Hipp | -0.164 |
Hero | -0.157 |
girls | -0.156 |
Token | s |
Token ID | 82 |
Feature activation | +3.02e-02 |
Pos logit contributions | |
rob | +0.160 |
Hipp | +0.142 |
zag | +0.142 |
steep | +0.135 |
Hero | +0.133 |
Neg logit contributions | |
vel | -0.181 |
face | -0.180 |
makeup | -0.179 |
spit | -0.176 |
blush | -0.169 |
Token | juggling |
Token ID | 48097 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
rob | +0.070 |
zag | +0.065 |
Hipp | +0.065 |
Hero | +0.062 |
girls | +0.062 |
Neg logit contributions | |
face | -0.032 |
makeup | -0.032 |
vel | -0.032 |
spit | -0.031 |
blush | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
rob | +0.410 |
zag | +0.384 |
Hipp | +0.381 |
Hero | +0.363 |
girls | +0.359 |
Neg logit contributions | |
vel | -0.227 |
makeup | -0.227 |
face | -0.225 |
spit | -0.220 |
blush | -0.206 |
Token | ac |
Token ID | 936 |
Feature activation | +3.59e-02 |
Pos logit contributions | |
rob | +0.026 |
zag | +0.024 |
steep | +0.023 |
Hipp | +0.022 |
girls | +0.022 |
Neg logit contributions | |
face | -0.015 |
makeup | -0.015 |
blush | -0.014 |
her | -0.014 |
spit | -0.014 |
Token | rob |
Token ID | 22609 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
Hipp | +0.010 |
zag | +0.007 |
rob | +0.005 |
steep | +0.005 |
Hero | +0.004 |
Neg logit contributions | |
vel | -0.106 |
makeup | -0.105 |
She | -0.102 |
face | -0.101 |
spit | -0.101 |
Token | ats |
Token ID | 1381 |
Feature activation | +6.75e-02 |
Pos logit contributions | |
rob | +0.676 |
zag | +0.611 |
Hipp | +0.581 |
girls | +0.556 |
Hero | +0.552 |
Neg logit contributions | |
makeup | -0.726 |
face | -0.725 |
vel | -0.720 |
spit | -0.716 |
blush | -0.682 |
Token | doing |
Token ID | 1804 |
Feature activation | +5.84e-03 |
Pos logit contributions | |
rob | +0.160 |
zag | +0.150 |
Hipp | +0.149 |
Hero | +0.143 |
steep | +0.142 |
Neg logit contributions | |
vel | -0.067 |
makeup | -0.067 |
face | -0.066 |
spit | -0.064 |
blush | -0.060 |
Token | flips |
Token ID | 45971 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
rob | +0.012 |
zag | +0.011 |
Hipp | +0.011 |
girls | +0.011 |
Hero | +0.011 |
Neg logit contributions | |
vel | -0.008 |
makeup | -0.007 |
face | -0.007 |
spit | -0.007 |
her | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -8.51e-03 |
Pos logit contributions | |
rob | +0.306 |
zag | +0.283 |
Hipp | +0.281 |
girls | +0.269 |
Hero | +0.267 |
Neg logit contributions | |
makeup | -0.126 |
vel | -0.125 |
face | -0.123 |
spit | -0.116 |
her | -0.111 |
Token | But |
Token ID | 887 |
Feature activation | +3.28e-02 |
Pos logit contributions | |
makeup | +0.009 |
her | +0.009 |
face | +0.009 |
She | +0.008 |
Her | +0.008 |
Neg logit contributions | |
rob | -0.021 |
zag | -0.020 |
steep | -0.019 |
girls | -0.019 |
Hipp | -0.018 |
Token | of |
Token ID | 286 |
Feature activation | -8.57e-02 |
Pos logit contributions | |
makeup | +0.041 |
vel | +0.038 |
face | +0.037 |
spit | +0.034 |
blush | +0.027 |
Neg logit contributions | |
rob | -0.338 |
zag | -0.322 |
Hipp | -0.321 |
Hero | -0.309 |
girls | -0.309 |
Token | fun |
Token ID | 1257 |
Feature activation | +5.14e-02 |
Pos logit contributions | |
makeup | +0.116 |
face | +0.112 |
spit | +0.107 |
vel | +0.102 |
prise | +0.100 |
Neg logit contributions | |
zag | -0.182 |
rob | -0.173 |
Hipp | -0.170 |
field | -0.159 |
Hero | -0.159 |
Token | watching |
Token ID | 4964 |
Feature activation | +5.00e-02 |
Pos logit contributions | |
rob | +0.128 |
Hipp | +0.122 |
zag | +0.122 |
Hero | +0.118 |
steep | +0.117 |
Neg logit contributions | |
vel | -0.045 |
makeup | -0.045 |
face | -0.044 |
spit | -0.042 |
blush | -0.037 |
Token | the |
Token ID | 262 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
rob | +0.104 |
zag | +0.097 |
Hipp | +0.095 |
steep | +0.090 |
Hero | +0.089 |
Neg logit contributions | |
makeup | -0.066 |
face | -0.066 |
vel | -0.065 |
spit | -0.063 |
blush | -0.060 |
Token | ac |
Token ID | 936 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
rob | +0.043 |
zag | +0.042 |
Hipp | +0.040 |
steep | +0.038 |
Zig | +0.037 |
Neg logit contributions | |
makeup | -0.055 |
face | -0.055 |
vel | -0.054 |
spit | -0.053 |
prise | -0.051 |
Token | rob |
Token ID | 22609 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
Hipp | +0.008 |
steep | +0.006 |
zag | +0.005 |
Hero | +0.004 |
girls | +0.004 |
Neg logit contributions | |
vel | -0.082 |
makeup | -0.081 |
She | -0.079 |
face | -0.079 |
on | -0.079 |
Token | ats |
Token ID | 1381 |
Feature activation | +6.06e-02 |
Pos logit contributions | |
rob | +0.695 |
Hipp | +0.661 |
zag | +0.642 |
Hero | +0.619 |
steep | +0.616 |
Neg logit contributions | |
vel | -0.666 |
makeup | -0.654 |
face | -0.649 |
spit | -0.640 |
prise | -0.608 |
Token | and |
Token ID | 290 |
Feature activation | +3.14e-03 |
Pos logit contributions | |
rob | +0.140 |
zag | +0.132 |
Hipp | +0.132 |
Hero | +0.126 |
girls | +0.125 |
Neg logit contributions | |
vel | -0.063 |
makeup | -0.062 |
face | -0.061 |
spit | -0.060 |
blush | -0.055 |
Token | clown |
Token ID | 25573 |
Feature activation | +9.10e-02 |
Pos logit contributions | |
rob | +0.006 |
zag | +0.006 |
Hipp | +0.006 |
steep | +0.006 |
Hero | +0.005 |
Neg logit contributions | |
face | -0.004 |
makeup | -0.004 |
spit | -0.004 |
her | -0.004 |
blush | -0.004 |
Token | s |
Token ID | 82 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
rob | +0.154 |
Hipp | +0.137 |
zag | +0.135 |
steep | +0.132 |
Hero | +0.128 |
Neg logit contributions | |
vel | -0.153 |
face | -0.153 |
makeup | -0.150 |
spit | -0.148 |
blush | -0.142 |
Token | . |
Token ID | 13 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
rob | +0.051 |
Hipp | +0.047 |
zag | +0.047 |
Hero | +0.045 |
steep | +0.044 |
Neg logit contributions | |
face | -0.019 |
makeup | -0.018 |
vel | -0.018 |
spit | -0.018 |
blush | -0.016 |
Token | thought |
Token ID | 1807 |
Feature activation | -4.87e-02 |
Pos logit contributions | |
face | +0.051 |
vel | +0.050 |
spit | +0.050 |
makeup | +0.049 |
blush | +0.045 |
Neg logit contributions | |
rob | -0.112 |
Hipp | -0.105 |
zag | -0.101 |
Hero | -0.099 |
girls | -0.099 |
Token | they |
Token ID | 484 |
Feature activation | +7.63e-02 |
Pos logit contributions | |
vel | +0.064 |
makeup | +0.064 |
face | +0.063 |
spit | +0.062 |
blush | +0.058 |
Neg logit contributions | |
rob | -0.100 |
zag | -0.093 |
Hipp | -0.092 |
Hero | -0.088 |
steep | -0.087 |
Token | looked |
Token ID | 3114 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
rob | +0.125 |
zag | +0.121 |
Hipp | +0.118 |
steep | +0.115 |
Hero | +0.111 |
Neg logit contributions | |
vel | -0.125 |
makeup | -0.121 |
face | -0.118 |
prise | -0.117 |
spit | -0.115 |
Token | like |
Token ID | 588 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
vel | +0.018 |
makeup | +0.017 |
face | +0.017 |
spit | +0.017 |
blush | +0.016 |
Neg logit contributions | |
rob | -0.040 |
zag | -0.037 |
Hipp | -0.037 |
Hero | -0.036 |
girls | -0.035 |
Token | ac |
Token ID | 936 |
Feature activation | +2.54e-02 |
Pos logit contributions | |
makeup | +0.021 |
face | +0.021 |
vel | +0.020 |
spit | +0.020 |
blush | +0.019 |
Neg logit contributions | |
rob | -0.045 |
zag | -0.043 |
Hipp | -0.042 |
steep | -0.040 |
Hero | -0.040 |
Token | rob |
Token ID | 22609 |
Feature activation | +4.02e-01 |
Pos logit contributions | |
rob | +0.013 |
Hipp | +0.013 |
zag | +0.012 |
Hero | +0.010 |
steep | +0.009 |
Neg logit contributions | |
vel | -0.070 |
makeup | -0.070 |
face | -0.068 |
spit | -0.067 |
blush | -0.067 |
Token | ats |
Token ID | 1381 |
Feature activation | +5.87e-02 |
Pos logit contributions | |
rob | +0.670 |
zag | +0.596 |
Hipp | +0.544 |
girls | +0.542 |
Hero | +0.524 |
Neg logit contributions | |
face | -0.700 |
spit | -0.687 |
makeup | -0.685 |
vel | -0.676 |
She | -0.648 |
Token | . |
Token ID | 13 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
rob | +0.143 |
zag | +0.134 |
Hipp | +0.133 |
steep | +0.127 |
Hero | +0.127 |
Neg logit contributions | |
makeup | -0.056 |
face | -0.054 |
vel | -0.053 |
spit | -0.050 |
blush | -0.049 |
Token | She |
Token ID | 1375 |
Feature activation | -4.73e-02 |
Pos logit contributions | |
makeup | +0.022 |
face | +0.021 |
spit | +0.019 |
vel | +0.019 |
blush | +0.019 |
Neg logit contributions | |
rob | -0.059 |
zag | -0.053 |
Hipp | -0.052 |
girls | -0.050 |
Hero | -0.050 |
Token | enjoyed |
Token ID | 8359 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
spit | +0.049 |
face | +0.047 |
makeup | +0.044 |
vel | +0.043 |
cut | +0.042 |
Neg logit contributions | |
rob | -0.119 |
Hipp | -0.109 |
zag | -0.107 |
girls | -0.107 |
Hero | -0.104 |
Token | the |
Token ID | 262 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
rob | +0.042 |
zag | +0.040 |
Hipp | +0.039 |
steep | +0.039 |
Hero | +0.038 |
Neg logit contributions | |
makeup | -0.017 |
vel | -0.016 |
her | -0.015 |
face | -0.015 |
spit | -0.015 |
Token | clown |
Token ID | 25573 |
Feature activation | +9.25e-02 |
Pos logit contributions | |
rob | +0.064 |
zag | +0.059 |
Hipp | +0.059 |
Hero | +0.056 |
steep | +0.055 |
Neg logit contributions | |
vel | -0.056 |
makeup | -0.055 |
face | -0.054 |
spit | -0.054 |
blush | -0.052 |
Token | s |
Token ID | 82 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
rob | +0.194 |
zag | +0.187 |
Hipp | +0.184 |
Hero | +0.177 |
girls | +0.175 |
Neg logit contributions | |
vel | -0.114 |
makeup | -0.113 |
spit | -0.108 |
face | -0.108 |
blush | -0.102 |
Token | and |
Token ID | 290 |
Feature activation | +6.16e-03 |
Pos logit contributions | |
rob | +0.043 |
zag | +0.042 |
Hipp | +0.041 |
steep | +0.040 |
Hero | +0.040 |
Neg logit contributions | |
vel | -0.023 |
makeup | -0.022 |
spit | -0.021 |
prise | -0.021 |
face | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
rob | +0.014 |
zag | +0.013 |
Hipp | +0.013 |
steep | +0.012 |
girls | +0.012 |
Neg logit contributions | |
face | -0.007 |
spit | -0.007 |
makeup | -0.007 |
her | -0.007 |
blush | -0.007 |
Token | ac |
Token ID | 936 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
rob | +0.065 |
zag | +0.060 |
Hipp | +0.060 |
Hero | +0.057 |
steep | +0.056 |
Neg logit contributions | |
vel | -0.051 |
makeup | -0.050 |
face | -0.050 |
spit | -0.049 |
blush | -0.047 |
Token | rob |
Token ID | 22609 |
Feature activation | +4.02e-01 |
Pos logit contributions | |
rob | +0.010 |
Hipp | +0.007 |
zag | +0.006 |
Hero | +0.004 |
steep | +0.003 |
Neg logit contributions | |
vel | -0.087 |
makeup | -0.087 |
face | -0.086 |
spit | -0.085 |
blush | -0.083 |
Token | ats |
Token ID | 1381 |
Feature activation | +5.73e-02 |
Pos logit contributions | |
rob | +0.652 |
zag | +0.596 |
Hipp | +0.587 |
Hero | +0.552 |
girls | +0.546 |
Neg logit contributions | |
vel | -0.703 |
makeup | -0.702 |
face | -0.695 |
spit | -0.688 |
blush | -0.657 |
Token | and |
Token ID | 290 |
Feature activation | +4.92e-03 |
Pos logit contributions | |
rob | +0.124 |
zag | +0.120 |
Hipp | +0.119 |
Hero | +0.115 |
Zig | +0.113 |
Neg logit contributions | |
vel | -0.066 |
makeup | -0.065 |
prise | -0.061 |
spit | -0.061 |
face | -0.060 |
Token | the |
Token ID | 262 |
Feature activation | +3.32e-02 |
Pos logit contributions | |
rob | +0.012 |
zag | +0.011 |
Hipp | +0.011 |
steep | +0.010 |
girls | +0.010 |
Neg logit contributions | |
face | -0.005 |
spit | -0.005 |
makeup | -0.005 |
her | -0.005 |
blush | -0.005 |
Token | lions |
Token ID | 31420 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
rob | +0.063 |
Hipp | +0.058 |
zag | +0.058 |
Hero | +0.055 |
steep | +0.054 |
Neg logit contributions | |
vel | -0.050 |
makeup | -0.048 |
spit | -0.047 |
face | -0.047 |
blush | -0.045 |
Token | and |
Token ID | 290 |
Feature activation | +4.62e-03 |
Pos logit contributions | |
rob | +0.449 |
zag | +0.426 |
Hipp | +0.425 |
girls | +0.410 |
Hero | +0.409 |
Neg logit contributions | |
vel | -0.221 |
makeup | -0.220 |
spit | -0.205 |
face | -0.205 |
prise | -0.200 |
Token | elephants |
Token ID | 25368 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
makeup | +0.041 |
face | +0.041 |
vel | +0.040 |
spit | +0.040 |
blush | +0.038 |
Neg logit contributions | |
rob | -0.065 |
zag | -0.061 |
Hipp | -0.060 |
Hero | -0.057 |
steep | -0.057 |
Token | , |
Token ID | 11 |
Feature activation | -3.18e-02 |
Pos logit contributions | |
rob | +0.479 |
Hipp | +0.455 |
zag | +0.452 |
girls | +0.435 |
Hero | +0.435 |
Neg logit contributions | |
vel | -0.261 |
makeup | -0.253 |
spit | -0.248 |
face | -0.248 |
prise | -0.231 |
Token | lions |
Token ID | 31420 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
vel | +0.038 |
makeup | +0.037 |
face | +0.036 |
spit | +0.036 |
blush | +0.033 |
Neg logit contributions | |
rob | -0.070 |
zag | -0.065 |
Hipp | -0.065 |
Hero | -0.063 |
girls | -0.062 |
Token | and |
Token ID | 290 |
Feature activation | +5.91e-03 |
Pos logit contributions | |
rob | +0.447 |
Hipp | +0.424 |
zag | +0.423 |
Hero | +0.407 |
girls | +0.405 |
Neg logit contributions | |
vel | -0.230 |
makeup | -0.226 |
spit | -0.221 |
face | -0.220 |
blush | -0.205 |
Token | ac |
Token ID | 936 |
Feature activation | +2.77e-02 |
Pos logit contributions | |
rob | +0.010 |
zag | +0.010 |
Hipp | +0.009 |
Hero | +0.009 |
steep | +0.009 |
Neg logit contributions | |
makeup | -0.009 |
face | -0.009 |
spit | -0.009 |
vel | -0.009 |
blush | -0.009 |
Token | rob |
Token ID | 22609 |
Feature activation | +4.02e-01 |
Pos logit contributions | |
zag | +0.006 |
Hipp | +0.006 |
rob | +0.005 |
steep | +0.004 |
Hero | +0.003 |
Neg logit contributions | |
makeup | -0.082 |
vel | -0.081 |
face | -0.079 |
spit | -0.078 |
She | -0.078 |
Token | ats |
Token ID | 1381 |
Feature activation | +5.61e-02 |
Pos logit contributions | |
rob | +0.635 |
zag | +0.576 |
Hipp | +0.561 |
Hero | +0.528 |
girls | +0.528 |
Neg logit contributions | |
makeup | -0.719 |
vel | -0.717 |
face | -0.714 |
spit | -0.706 |
blush | -0.675 |
Token | . |
Token ID | 13 |
Feature activation | -1.65e-02 |
Pos logit contributions | |
rob | +0.139 |
zag | +0.131 |
Hipp | +0.129 |
steep | +0.125 |
Hero | +0.125 |
Neg logit contributions | |
vel | -0.049 |
makeup | -0.049 |
face | -0.049 |
spit | -0.047 |
blush | -0.043 |
Token | They |
Token ID | 1119 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
face | +0.023 |
makeup | +0.021 |
spit | +0.021 |
her | +0.020 |
L | +0.020 |
Neg logit contributions | |
rob | -0.039 |
zag | -0.037 |
field | -0.033 |
aboard | -0.030 |
xy | -0.030 |
Token | are |
Token ID | 389 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
rob | +0.033 |
zag | +0.031 |
Hipp | +0.031 |
Hero | +0.030 |
steep | +0.029 |
Neg logit contributions | |
makeup | -0.010 |
vel | -0.010 |
face | -0.010 |
spit | -0.010 |
blush | -0.009 |
Token | amazed |
Token ID | 24562 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
face | +0.115 |
spit | +0.109 |
makeup | +0.107 |
vel | +0.103 |
her | +0.099 |
Neg logit contributions | |
rob | -0.281 |
zag | -0.255 |
Hipp | -0.245 |
Hero | -0.238 |
steep | -0.233 |
Token | the |
Token ID | 262 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
face | +0.097 |
makeup | +0.096 |
vel | +0.091 |
spit | +0.090 |
her | +0.089 |
Neg logit contributions | |
zag | -0.185 |
rob | -0.183 |
Hipp | -0.171 |
girls | -0.166 |
steep | -0.164 |
Token | animals |
Token ID | 4695 |
Feature activation | +3.67e-02 |
Pos logit contributions | |
zag | +0.053 |
rob | +0.052 |
Hipp | +0.048 |
Zig | +0.047 |
girls | +0.046 |
Neg logit contributions | |
face | -0.039 |
makeup | -0.038 |
vel | -0.036 |
spit | -0.036 |
goodness | -0.034 |
Token | , |
Token ID | 11 |
Feature activation | -3.91e-02 |
Pos logit contributions | |
rob | +0.090 |
zag | +0.084 |
Hipp | +0.083 |
Hero | +0.080 |
steep | +0.080 |
Neg logit contributions | |
vel | -0.034 |
face | -0.034 |
makeup | -0.033 |
spit | -0.032 |
blush | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
face | +0.036 |
makeup | +0.034 |
spit | +0.033 |
her | +0.031 |
blush | +0.029 |
Neg logit contributions | |
rob | -0.103 |
zag | -0.098 |
Hipp | -0.090 |
girls | -0.087 |
aboard | -0.087 |
Token | ac |
Token ID | 936 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
rob | +0.054 |
zag | +0.053 |
Hipp | +0.048 |
Zig | +0.047 |
Hero | +0.046 |
Neg logit contributions | |
face | -0.042 |
makeup | -0.041 |
spit | -0.039 |
vel | -0.038 |
mirror | -0.037 |
Token | rob |
Token ID | 22609 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
Hipp | +0.007 |
steep | +0.005 |
zag | +0.005 |
Hero | +0.004 |
Zig | +0.004 |
Neg logit contributions | |
makeup | -0.068 |
vel | -0.068 |
on | -0.067 |
She | -0.066 |
face | -0.066 |
Token | ats |
Token ID | 1381 |
Feature activation | +5.46e-02 |
Pos logit contributions | |
rob | +0.668 |
Hipp | +0.629 |
zag | +0.617 |
Hero | +0.593 |
steep | +0.587 |
Neg logit contributions | |
vel | -0.667 |
makeup | -0.660 |
face | -0.652 |
spit | -0.646 |
prise | -0.611 |
Token | and |
Token ID | 290 |
Feature activation | -3.05e-04 |
Pos logit contributions | |
rob | +0.131 |
zag | +0.123 |
Hipp | +0.121 |
steep | +0.117 |
Hero | +0.117 |
Neg logit contributions | |
face | -0.052 |
makeup | -0.051 |
vel | -0.051 |
spit | -0.050 |
blush | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
face | +0.000 |
makeup | +0.000 |
spit | +0.000 |
her | +0.000 |
blush | +0.000 |
Neg logit contributions | |
rob | -0.001 |
zag | -0.001 |
Hipp | -0.001 |
steep | -0.001 |
Hero | -0.001 |
Token | clown |
Token ID | 25573 |
Feature activation | +8.63e-02 |
Pos logit contributions | |
rob | +0.051 |
zag | +0.048 |
Hipp | +0.047 |
Hero | +0.044 |
Zig | +0.044 |
Neg logit contributions | |
makeup | -0.040 |
face | -0.040 |
vel | -0.040 |
spit | -0.039 |
blush | -0.037 |
Token | s |
Token ID | 82 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
rob | +0.152 |
steep | +0.134 |
zag | +0.134 |
Hipp | +0.132 |
Hero | +0.127 |
Neg logit contributions | |
face | -0.138 |
makeup | -0.135 |
vel | -0.135 |
spit | -0.135 |
blush | -0.128 |
Token | saw |
Token ID | 2497 |
Feature activation | -9.39e-02 |
Pos logit contributions | |
spit | +0.024 |
face | +0.023 |
makeup | +0.022 |
vel | +0.021 |
blush | +0.020 |
Neg logit contributions | |
rob | -0.055 |
zag | -0.052 |
Hipp | -0.050 |
girls | -0.048 |
steep | -0.048 |
Token | clown |
Token ID | 25573 |
Feature activation | +8.31e-02 |
Pos logit contributions | |
makeup | +0.120 |
face | +0.119 |
vel | +0.118 |
spit | +0.117 |
blush | +0.109 |
Neg logit contributions | |
rob | -0.197 |
zag | -0.185 |
Hipp | -0.181 |
Hero | -0.172 |
steep | -0.172 |
Token | s |
Token ID | 82 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
rob | +0.125 |
zag | +0.113 |
Hipp | +0.112 |
Hero | +0.105 |
steep | +0.103 |
Neg logit contributions | |
vel | -0.155 |
makeup | -0.155 |
face | -0.153 |
spit | -0.152 |
blush | -0.145 |
Token | and |
Token ID | 290 |
Feature activation | -2.47e-03 |
Pos logit contributions | |
rob | +0.025 |
zag | +0.024 |
Hipp | +0.024 |
Hero | +0.023 |
steep | +0.022 |
Neg logit contributions | |
vel | -0.012 |
makeup | -0.012 |
face | -0.011 |
spit | -0.011 |
blush | -0.011 |
Token | ac |
Token ID | 936 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
face | +0.004 |
makeup | +0.004 |
spit | +0.004 |
blush | +0.003 |
her | +0.003 |
Neg logit contributions | |
rob | -0.005 |
zag | -0.005 |
Hipp | -0.004 |
steep | -0.004 |
girls | -0.004 |
Token | rob |
Token ID | 22609 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
rob | +0.008 |
Hipp | +0.007 |
zag | +0.007 |
steep | +0.005 |
Hero | +0.005 |
Neg logit contributions | |
vel | -0.061 |
makeup | -0.061 |
face | -0.059 |
spit | -0.059 |
blush | -0.058 |
Token | ats |
Token ID | 1381 |
Feature activation | +5.00e-02 |
Pos logit contributions | |
rob | +0.618 |
zag | +0.560 |
Hipp | +0.544 |
girls | +0.513 |
Hero | +0.511 |
Neg logit contributions | |
makeup | -0.701 |
vel | -0.699 |
face | -0.697 |
spit | -0.688 |
blush | -0.659 |
Token | and |
Token ID | 290 |
Feature activation | -3.20e-03 |
Pos logit contributions | |
rob | +0.117 |
zag | +0.110 |
Hipp | +0.110 |
Hero | +0.105 |
girls | +0.104 |
Neg logit contributions | |
vel | -0.051 |
makeup | -0.050 |
face | -0.050 |
spit | -0.049 |
blush | -0.045 |
Token | lions |
Token ID | 31420 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
face | +0.005 |
makeup | +0.005 |
spit | +0.005 |
blush | +0.005 |
dust | +0.004 |
Neg logit contributions | |
rob | -0.006 |
zag | -0.006 |
Hipp | -0.005 |
steep | -0.005 |
Hero | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | -3.55e-03 |
Pos logit contributions | |
rob | +0.440 |
zag | +0.416 |
Hipp | +0.415 |
girls | +0.401 |
Hero | +0.400 |
Neg logit contributions | |
vel | -0.205 |
makeup | -0.203 |
face | -0.196 |
spit | -0.195 |
prise | -0.182 |
Token | elephants |
Token ID | 25368 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
makeup | +0.006 |
face | +0.006 |
spit | +0.006 |
vel | +0.006 |
blush | +0.006 |
Neg logit contributions | |
rob | -0.006 |
zag | -0.006 |
Hipp | -0.005 |
steep | -0.005 |
Hero | -0.005 |
Token | finally |
Token ID | 3443 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
rob | +0.271 |
zag | +0.250 |
Hipp | +0.247 |
Hero | +0.236 |
girls | +0.236 |
Neg logit contributions | |
makeup | -0.189 |
vel | -0.189 |
face | -0.187 |
spit | -0.184 |
blush | -0.174 |
Token | , |
Token ID | 11 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
rob | +0.262 |
zag | +0.244 |
Hipp | +0.243 |
girls | +0.235 |
Hero | +0.234 |
Neg logit contributions | |
makeup | -0.107 |
vel | -0.107 |
face | -0.104 |
spit | -0.102 |
blush | -0.094 |
Token | the |
Token ID | 262 |
Feature activation | +3.77e-02 |
Pos logit contributions | |
makeup | +0.033 |
face | +0.032 |
vel | +0.032 |
spit | +0.031 |
blush | +0.029 |
Neg logit contributions | |
rob | -0.059 |
zag | -0.055 |
Hipp | -0.054 |
girls | -0.052 |
Hero | -0.052 |
Token | z |
Token ID | 1976 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
rob | +0.077 |
Hipp | +0.074 |
zag | +0.073 |
Hero | +0.070 |
steep | +0.069 |
Neg logit contributions | |
vel | -0.049 |
makeup | -0.046 |
spit | -0.046 |
face | -0.045 |
blush | -0.043 |
Token | ig |
Token ID | 328 |
Feature activation | +8.49e-02 |
Pos logit contributions | |
rob | +0.204 |
Hipp | +0.190 |
zag | +0.182 |
Hero | +0.178 |
steep | +0.178 |
Neg logit contributions | |
vel | -0.221 |
face | -0.218 |
makeup | -0.216 |
spit | -0.211 |
prise | -0.203 |
Token | zag |
Token ID | 50183 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
rob | +0.179 |
Hipp | +0.166 |
Hero | +0.157 |
zag | +0.156 |
girls | +0.155 |
Neg logit contributions | |
vel | -0.110 |
face | -0.110 |
makeup | -0.107 |
spit | -0.106 |
blush | -0.098 |
Token | had |
Token ID | 550 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
rob | +0.741 |
zag | +0.687 |
Hipp | +0.685 |
Hero | +0.652 |
girls | +0.644 |
Neg logit contributions | |
vel | -0.493 |
makeup | -0.491 |
face | -0.486 |
spit | -0.479 |
blush | -0.450 |
Token | an |
Token ID | 281 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
rob | +0.042 |
zag | +0.039 |
Hipp | +0.039 |
Hero | +0.037 |
girls | +0.037 |
Neg logit contributions | |
makeup | -0.023 |
vel | -0.023 |
face | -0.022 |
spit | -0.022 |
blush | -0.020 |
Token | idea |
Token ID | 2126 |
Feature activation | +4.28e-03 |
Pos logit contributions | |
rob | +0.061 |
steep | +0.059 |
Hipp | +0.059 |
zag | +0.059 |
Hero | +0.055 |
Neg logit contributions | |
vel | -0.065 |
makeup | -0.065 |
blush | -0.063 |
spit | -0.063 |
prise | -0.062 |
Token | . |
Token ID | 13 |
Feature activation | -5.92e-03 |
Pos logit contributions | |
rob | +0.009 |
zag | +0.009 |
Hipp | +0.009 |
Zig | +0.008 |
Hero | +0.008 |
Neg logit contributions | |
vel | -0.005 |
makeup | -0.005 |
spit | -0.005 |
face | -0.005 |
blush | -0.004 |
Token | It |
Token ID | 632 |
Feature activation | -4.49e-02 |
Pos logit contributions | |
makeup | +0.008 |
face | +0.008 |
vel | +0.008 |
spit | +0.007 |
She | +0.007 |
Neg logit contributions | |
rob | -0.013 |
zag | -0.011 |
steep | -0.011 |
Hipp | -0.011 |
girls | -0.010 |
Token | ement |
Token ID | 972 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
vel | +0.031 |
makeup | +0.031 |
face | +0.031 |
spit | +0.030 |
blush | +0.029 |
Neg logit contributions | |
rob | -0.030 |
Hipp | -0.028 |
zag | -0.028 |
girls | -0.026 |
steep | -0.026 |
Token | as |
Token ID | 355 |
Feature activation | +8.13e-02 |
Pos logit contributions | |
rob | +0.326 |
Hipp | +0.302 |
zag | +0.301 |
Hero | +0.290 |
girls | +0.287 |
Neg logit contributions | |
face | -0.134 |
vel | -0.134 |
makeup | -0.133 |
spit | -0.127 |
blush | -0.117 |
Token | the |
Token ID | 262 |
Feature activation | +4.33e-02 |
Pos logit contributions | |
rob | +0.161 |
zag | +0.149 |
Hipp | +0.145 |
girls | +0.139 |
Hero | +0.137 |
Neg logit contributions | |
makeup | -0.116 |
face | -0.114 |
vel | -0.113 |
spit | -0.112 |
blush | -0.105 |
Token | z |
Token ID | 1976 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
rob | +0.086 |
Hipp | +0.081 |
zag | +0.080 |
Hero | +0.076 |
steep | +0.076 |
Neg logit contributions | |
vel | -0.060 |
makeup | -0.058 |
spit | -0.057 |
face | -0.057 |
blush | -0.053 |
Token | ig |
Token ID | 328 |
Feature activation | +8.47e-02 |
Pos logit contributions | |
rob | +0.240 |
Hipp | +0.227 |
zag | +0.218 |
Hero | +0.217 |
steep | +0.216 |
Neg logit contributions | |
vel | -0.202 |
makeup | -0.199 |
face | -0.197 |
spit | -0.192 |
blush | -0.185 |
Token | zag |
Token ID | 50183 |
Feature activation | +3.67e-01 |
Pos logit contributions | |
rob | +0.161 |
Hipp | +0.147 |
Hero | +0.141 |
girls | +0.138 |
steep | +0.136 |
Neg logit contributions | |
face | -0.126 |
vel | -0.126 |
makeup | -0.123 |
spit | -0.120 |
blush | -0.115 |
Token | soared |
Token ID | 35292 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
rob | +0.770 |
Hipp | +0.713 |
zag | +0.709 |
Hero | +0.681 |
girls | +0.673 |
Neg logit contributions | |
vel | -0.466 |
makeup | -0.464 |
face | -0.464 |
spit | -0.454 |
blush | -0.425 |
Token | in |
Token ID | 287 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
rob | +0.421 |
Hipp | +0.383 |
girls | +0.370 |
zag | +0.368 |
Hero | +0.367 |
Neg logit contributions | |
face | -0.194 |
vel | -0.191 |
makeup | -0.186 |
her | -0.179 |
She | -0.175 |
Token | the |
Token ID | 262 |
Feature activation | +4.58e-02 |
Pos logit contributions | |
rob | +0.039 |
zag | +0.036 |
Hipp | +0.036 |
Hero | +0.035 |
girls | +0.034 |
Neg logit contributions | |
makeup | -0.021 |
vel | -0.021 |
face | -0.021 |
spit | -0.021 |
blush | -0.019 |
Token | sky |
Token ID | 6766 |
Feature activation | +8.31e-02 |
Pos logit contributions | |
rob | +0.079 |
zag | +0.073 |
Hipp | +0.072 |
steep | +0.068 |
Hero | +0.068 |
Neg logit contributions | |
vel | -0.076 |
makeup | -0.075 |
spit | -0.074 |
face | -0.073 |
blush | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | -4.98e-03 |
Pos logit contributions | |
rob | +0.217 |
Hipp | +0.201 |
zag | +0.198 |
Hero | +0.195 |
girls | +0.193 |
Neg logit contributions | |
face | -0.067 |
vel | -0.066 |
makeup | -0.064 |
spit | -0.061 |
blush | -0.055 |
Token | hill |
Token ID | 12788 |
Feature activation | +8.04e-02 |
Pos logit contributions | |
rob | +0.095 |
zag | +0.089 |
Hipp | +0.088 |
steep | +0.085 |
Hero | +0.084 |
Neg logit contributions | |
vel | -0.070 |
spit | -0.069 |
makeup | -0.069 |
face | -0.067 |
blush | -0.065 |
Token | in |
Token ID | 287 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
rob | +0.207 |
Hipp | +0.190 |
girls | +0.182 |
Hero | +0.180 |
zag | +0.178 |
Neg logit contributions | |
face | -0.071 |
vel | -0.066 |
makeup | -0.065 |
her | -0.063 |
hand | -0.059 |
Token | a |
Token ID | 257 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
rob | +0.032 |
zag | +0.030 |
Hipp | +0.030 |
Hero | +0.029 |
steep | +0.029 |
Neg logit contributions | |
vel | -0.023 |
makeup | -0.022 |
spit | -0.022 |
face | -0.022 |
blush | -0.021 |
Token | z |
Token ID | 1976 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
makeup | +0.014 |
vel | +0.014 |
spit | +0.014 |
face | +0.014 |
blush | +0.014 |
Neg logit contributions | |
rob | -0.022 |
zag | -0.021 |
steep | -0.021 |
Hipp | -0.021 |
Hero | -0.020 |
Token | ig |
Token ID | 328 |
Feature activation | +8.62e-02 |
Pos logit contributions | |
rob | +0.230 |
zag | +0.222 |
Hipp | +0.202 |
Zig | +0.196 |
steep | +0.193 |
Neg logit contributions | |
spit | -0.219 |
makeup | -0.219 |
vel | -0.212 |
face | -0.212 |
blush | -0.204 |
Token | zag |
Token ID | 50183 |
Feature activation | +3.67e-01 |
Pos logit contributions | |
rob | +0.047 |
Hipp | +0.038 |
Hero | +0.030 |
Skip | +0.024 |
girls | +0.024 |
Neg logit contributions | |
vel | -0.245 |
face | -0.245 |
makeup | -0.237 |
She | -0.233 |
spit | -0.232 |
Token | pattern |
Token ID | 3912 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
rob | +0.757 |
Hipp | +0.711 |
Hero | +0.670 |
girls | +0.661 |
zag | +0.660 |
Neg logit contributions | |
face | -0.502 |
vel | -0.479 |
makeup | -0.475 |
spit | -0.465 |
blush | -0.435 |
Token | . |
Token ID | 13 |
Feature activation | -6.39e-03 |
Pos logit contributions | |
rob | +0.069 |
zag | +0.066 |
Hipp | +0.065 |
Hero | +0.062 |
steep | +0.062 |
Neg logit contributions | |
vel | -0.029 |
makeup | -0.029 |
face | -0.028 |
spit | -0.028 |
blush | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +9.92e-03 |
Pos logit contributions | |
makeup | +0.008 |
face | +0.008 |
vel | +0.008 |
She | +0.008 |
spit | +0.008 |
Neg logit contributions | |
rob | -0.014 |
zag | -0.013 |
steep | -0.012 |
girls | -0.012 |
Hipp | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
rob | +0.020 |
zag | +0.019 |
Hipp | +0.018 |
Hero | +0.018 |
girls | +0.017 |
Neg logit contributions | |
vel | -0.013 |
makeup | -0.013 |
face | -0.013 |
spit | -0.013 |
blush | -0.012 |
Token | When |
Token ID | 2215 |
Feature activation | -6.01e-03 |
Pos logit contributions | |
rob | +0.022 |
zag | +0.021 |
Hipp | +0.021 |
Zig | +0.020 |
Hero | +0.020 |
Neg logit contributions | |
face | -0.014 |
makeup | -0.014 |
spit | -0.013 |
vel | -0.013 |
blush | -0.012 |
Token | fly |
Token ID | 6129 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
rob | +0.074 |
zag | +0.070 |
Hipp | +0.069 |
Hero | +0.067 |
steep | +0.066 |
Neg logit contributions | |
vel | -0.036 |
makeup | -0.035 |
face | -0.035 |
spit | -0.034 |
blush | -0.032 |
Token | like |
Token ID | 588 |
Feature activation | -1.31e-03 |
Pos logit contributions | |
rob | +0.254 |
Hipp | +0.232 |
girls | +0.224 |
zag | +0.223 |
Hero | +0.218 |
Neg logit contributions | |
her | -0.099 |
face | -0.097 |
makeup | -0.095 |
vel | -0.095 |
She | -0.086 |
Token | the |
Token ID | 262 |
Feature activation | +4.62e-02 |
Pos logit contributions | |
makeup | +0.002 |
face | +0.001 |
vel | +0.001 |
spit | +0.001 |
her | +0.001 |
Neg logit contributions | |
rob | -0.003 |
zag | -0.003 |
Hipp | -0.003 |
girls | -0.003 |
steep | -0.003 |
Token | z |
Token ID | 1976 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
rob | +0.090 |
zag | +0.084 |
Hipp | +0.083 |
Hero | +0.079 |
steep | +0.078 |
Neg logit contributions | |
vel | -0.065 |
makeup | -0.065 |
face | -0.064 |
spit | -0.063 |
blush | -0.060 |
Token | ig |
Token ID | 328 |
Feature activation | +8.70e-02 |
Pos logit contributions | |
rob | +0.262 |
Hipp | +0.242 |
zag | +0.242 |
Hero | +0.231 |
girls | +0.228 |
Neg logit contributions | |
vel | -0.191 |
makeup | -0.190 |
face | -0.187 |
spit | -0.184 |
blush | -0.175 |
Token | zag |
Token ID | 50183 |
Feature activation | +3.67e-01 |
Pos logit contributions | |
rob | +0.051 |
Hipp | +0.037 |
Hero | +0.031 |
girls | +0.027 |
Skip | +0.024 |
Neg logit contributions | |
vel | -0.246 |
face | -0.243 |
makeup | -0.239 |
She | -0.232 |
her | -0.232 |
Token | if |
Token ID | 611 |
Feature activation | +4.53e-02 |
Pos logit contributions | |
rob | +0.781 |
Hipp | +0.719 |
zag | +0.687 |
Hero | +0.686 |
girls | +0.677 |
Neg logit contributions | |
face | -0.474 |
vel | -0.467 |
makeup | -0.460 |
spit | -0.445 |
blush | -0.418 |
Token | he |
Token ID | 339 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
rob | +0.095 |
zag | +0.084 |
girls | +0.081 |
Hipp | +0.078 |
Hero | +0.075 |
Neg logit contributions | |
makeup | -0.062 |
her | -0.061 |
face | -0.061 |
spit | -0.059 |
vel | -0.058 |
Token | just |
Token ID | 655 |
Feature activation | +9.68e-02 |
Pos logit contributions | |
rob | +0.083 |
Hipp | +0.075 |
zag | +0.074 |
Hero | +0.072 |
girls | +0.071 |
Neg logit contributions | |
spit | -0.046 |
vel | -0.044 |
face | -0.044 |
makeup | -0.042 |
dust | -0.040 |
Token | opened |
Token ID | 4721 |
Feature activation | +3.82e-02 |
Pos logit contributions | |
rob | +0.214 |
zag | +0.194 |
Hipp | +0.194 |
girls | +0.186 |
Hero | +0.185 |
Neg logit contributions | |
face | -0.120 |
spit | -0.118 |
makeup | -0.117 |
vel | -0.116 |
blush | -0.108 |
Token | his |
Token ID | 465 |
Feature activation | -8.98e-03 |
Pos logit contributions | |
rob | +0.098 |
zag | +0.090 |
Hipp | +0.088 |
girls | +0.087 |
Hero | +0.083 |
Neg logit contributions | |
her | -0.038 |
makeup | -0.037 |
face | -0.035 |
vel | -0.034 |
dust | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | -8.54e-03 |
Pos logit contributions | |
face | +0.022 |
makeup | +0.022 |
vel | +0.021 |
her | +0.019 |
blush | +0.018 |
Neg logit contributions | |
rob | -0.066 |
Hipp | -0.062 |
Hero | -0.060 |
zag | -0.060 |
steep | -0.059 |
Token | One |
Token ID | 1881 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
She | +0.009 |
her | +0.008 |
She | +0.008 |
Her | +0.007 |
L | +0.007 |
Neg logit contributions | |
rob | -0.020 |
roles | -0.018 |
lights | -0.018 |
zag | -0.018 |
common | -0.017 |
Token | day |
Token ID | 1110 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
rob | +0.285 |
Hipp | +0.255 |
girls | +0.252 |
zag | +0.251 |
Hero | +0.247 |
Neg logit contributions | |
face | -0.206 |
makeup | -0.203 |
vel | -0.200 |
spit | -0.195 |
her | -0.191 |
Token | , |
Token ID | 11 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
rob | +0.028 |
zag | +0.026 |
girls | +0.025 |
steep | +0.024 |
habitats | +0.024 |
Neg logit contributions | |
her | -0.010 |
vel | -0.009 |
face | -0.009 |
makeup | -0.009 |
spit | -0.009 |
Token | Zig |
Token ID | 24992 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
her | +0.050 |
She | +0.047 |
face | +0.046 |
little | +0.045 |
spit | +0.045 |
Neg logit contributions | |
rob | -0.035 |
zag | -0.031 |
girls | -0.028 |
habitats | -0.027 |
lights | -0.027 |
Token | zag |
Token ID | 50183 |
Feature activation | +3.66e-01 |
Pos logit contributions | |
rob | +0.324 |
Hipp | +0.231 |
Skip | +0.209 |
Hero | +0.205 |
steep | +0.202 |
Neg logit contributions | |
vel | -0.682 |
face | -0.675 |
She | -0.660 |
ime | -0.656 |
e | -0.645 |
Token | 's |
Token ID | 338 |
Feature activation | -7.45e-03 |
Pos logit contributions | |
rob | +0.781 |
Hipp | +0.722 |
zag | +0.718 |
Hero | +0.690 |
girls | +0.684 |
Neg logit contributions | |
vel | -0.450 |
face | -0.450 |
makeup | -0.448 |
spit | -0.441 |
blush | -0.407 |
Token | owner |
Token ID | 4870 |
Feature activation | +5.52e-02 |
Pos logit contributions | |
vel | +0.012 |
makeup | +0.011 |
spit | +0.011 |
face | +0.011 |
blush | +0.011 |
Neg logit contributions | |
rob | -0.013 |
zag | -0.013 |
Hipp | -0.013 |
Hero | -0.012 |
Zig | -0.012 |
Token | wanted |
Token ID | 2227 |
Feature activation | -7.15e-02 |
Pos logit contributions | |
rob | +0.105 |
zag | +0.101 |
Hipp | +0.101 |
Hero | +0.097 |
Zig | +0.095 |
Neg logit contributions | |
vel | -0.078 |
makeup | -0.076 |
face | -0.073 |
spit | -0.071 |
blush | -0.070 |
Token | to |
Token ID | 284 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
makeup | +0.085 |
her | +0.082 |
face | +0.079 |
spit | +0.074 |
dust | +0.073 |
Neg logit contributions | |
rob | -0.167 |
zag | -0.155 |
Hipp | -0.145 |
girls | -0.143 |
steep | -0.142 |
Token | take |
Token ID | 1011 |
Feature activation | -9.65e-02 |
Pos logit contributions | |
zag | +0.031 |
Hipp | +0.031 |
rob | +0.031 |
Zig | +0.030 |
Hero | +0.029 |
Neg logit contributions | |
vel | -0.019 |
makeup | -0.016 |
prise | -0.015 |
blush | -0.015 |
spit | -0.015 |
Token | had |
Token ID | 550 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
rob | +0.020 |
zag | +0.019 |
Hipp | +0.018 |
girls | +0.018 |
Hero | +0.018 |
Neg logit contributions | |
face | -0.010 |
spit | -0.010 |
makeup | -0.010 |
vel | -0.010 |
blush | -0.009 |
Token | their |
Token ID | 511 |
Feature activation | +4.28e-02 |
Pos logit contributions | |
rob | +0.043 |
zag | +0.041 |
Hipp | +0.040 |
Hero | +0.038 |
girls | +0.037 |
Neg logit contributions | |
makeup | -0.022 |
face | -0.021 |
spit | -0.021 |
vel | -0.021 |
her | -0.019 |
Token | own |
Token ID | 898 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
rob | +0.084 |
zag | +0.075 |
Hipp | +0.075 |
Hero | +0.071 |
girls | +0.070 |
Neg logit contributions | |
makeup | -0.064 |
face | -0.063 |
spit | -0.061 |
vel | -0.060 |
blush | -0.057 |
Token | z |
Token ID | 1976 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
rob | +0.083 |
Hipp | +0.075 |
zag | +0.074 |
Hero | +0.072 |
girls | +0.071 |
Neg logit contributions | |
makeup | -0.054 |
face | -0.053 |
vel | -0.052 |
spit | -0.051 |
blush | -0.048 |
Token | ig |
Token ID | 328 |
Feature activation | +8.22e-02 |
Pos logit contributions | |
rob | +0.251 |
Hipp | +0.232 |
zag | +0.231 |
Hero | +0.221 |
steep | +0.218 |
Neg logit contributions | |
vel | -0.190 |
makeup | -0.188 |
face | -0.185 |
spit | -0.182 |
blush | -0.173 |
Token | zag |
Token ID | 50183 |
Feature activation | +3.65e-01 |
Pos logit contributions | |
rob | +0.055 |
Hipp | +0.044 |
Hero | +0.035 |
steep | +0.032 |
girls | +0.031 |
Neg logit contributions | |
vel | -0.224 |
face | -0.222 |
makeup | -0.221 |
spit | -0.214 |
blush | -0.211 |
Token | . |
Token ID | 13 |
Feature activation | -3.49e-03 |
Pos logit contributions | |
rob | +0.747 |
Hipp | +0.701 |
Hero | +0.654 |
zag | +0.644 |
girls | +0.644 |
Neg logit contributions | |
face | -0.505 |
makeup | -0.494 |
vel | -0.487 |
spit | -0.469 |
blush | -0.443 |
Token |
|
Token ID | 198 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
She | +0.003 |
her | +0.003 |
hand | +0.002 |
face | +0.002 |
Her | +0.002 |
Neg logit contributions | |
rob | -0.009 |
zag | -0.009 |
steep | -0.008 |
aboard | -0.008 |
lights | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
rob | +0.030 |
zag | +0.028 |
Hipp | +0.027 |
steep | +0.027 |
girls | +0.027 |
Neg logit contributions | |
She | -0.007 |
vel | -0.007 |
face | -0.007 |
her | -0.007 |
Her | -0.007 |
Token | The |
Token ID | 464 |
Feature activation | +3.71e-02 |
Pos logit contributions | |
rob | +0.025 |
zag | +0.023 |
Hipp | +0.022 |
girls | +0.022 |
steep | +0.022 |
Neg logit contributions | |
vel | -0.014 |
She | -0.013 |
spit | -0.013 |
Her | -0.013 |
makeup | -0.013 |
Token | two |
Token ID | 734 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
rob | +0.074 |
zag | +0.068 |
Hipp | +0.067 |
Hero | +0.063 |
girls | +0.063 |
Neg logit contributions | |
makeup | -0.052 |
face | -0.052 |
vel | -0.051 |
spit | -0.051 |
blush | -0.048 |
Token | can |
Token ID | 460 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
rob | +0.053 |
zag | +0.049 |
Hipp | +0.049 |
Hero | +0.046 |
girls | +0.046 |
Neg logit contributions | |
vel | -0.045 |
makeup | -0.045 |
face | -0.044 |
spit | -0.043 |
blush | -0.041 |
Token | make |
Token ID | 787 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
rob | +0.027 |
zag | +0.026 |
Hipp | +0.026 |
Hero | +0.024 |
girls | +0.024 |
Neg logit contributions | |
vel | -0.011 |
makeup | -0.011 |
spit | -0.011 |
face | -0.011 |
blush | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | +4.65e-02 |
Pos logit contributions | |
vel | +0.034 |
makeup | +0.034 |
spit | +0.034 |
face | +0.034 |
blush | +0.031 |
Neg logit contributions | |
rob | -0.074 |
Hipp | -0.069 |
zag | -0.069 |
Hero | -0.066 |
Zig | -0.065 |
Token | z |
Token ID | 1976 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
rob | +0.091 |
zag | +0.085 |
Hipp | +0.085 |
Hero | +0.081 |
steep | +0.080 |
Neg logit contributions | |
vel | -0.064 |
makeup | -0.062 |
spit | -0.061 |
face | -0.061 |
blush | -0.058 |
Token | ig |
Token ID | 328 |
Feature activation | +8.40e-02 |
Pos logit contributions | |
rob | +0.214 |
Hipp | +0.206 |
Hero | +0.197 |
steep | +0.197 |
girls | +0.192 |
Neg logit contributions | |
vel | -0.212 |
face | -0.204 |
makeup | -0.198 |
spit | -0.195 |
prise | -0.193 |
Token | zag |
Token ID | 50183 |
Feature activation | +3.65e-01 |
Pos logit contributions | |
rob | +0.158 |
Hipp | +0.145 |
Hero | +0.138 |
girls | +0.134 |
Skip | +0.132 |
Neg logit contributions | |
face | -0.129 |
vel | -0.129 |
makeup | -0.126 |
spit | -0.122 |
blush | -0.117 |
Token | another |
Token ID | 1194 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
rob | +0.762 |
Hipp | +0.704 |
zag | +0.686 |
Hero | +0.672 |
girls | +0.661 |
Neg logit contributions | |
vel | -0.475 |
face | -0.474 |
makeup | -0.466 |
spit | -0.457 |
blush | -0.427 |
Token | day |
Token ID | 1110 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
rob | +0.253 |
Hipp | +0.231 |
zag | +0.227 |
Hero | +0.223 |
girls | +0.222 |
Neg logit contributions | |
face | -0.164 |
vel | -0.163 |
makeup | -0.162 |
spit | -0.157 |
blush | -0.147 |
Token | . |
Token ID | 13 |
Feature activation | -5.39e-03 |
Pos logit contributions | |
rob | +0.031 |
zag | +0.029 |
Hipp | +0.029 |
Hero | +0.027 |
Zig | +0.027 |
Neg logit contributions | |
vel | -0.019 |
makeup | -0.019 |
face | -0.019 |
spit | -0.018 |
blush | -0.017 |
Token | I |
Token ID | 314 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
face | +0.006 |
vel | +0.006 |
She | +0.006 |
makeup | +0.006 |
hand | +0.006 |
Neg logit contributions | |
rob | -0.013 |
zag | -0.011 |
steep | -0.010 |
ahs | -0.010 |
Hipp | -0.010 |
Token | 'm |
Token ID | 1101 |
Feature activation | -9.71e-03 |
Pos logit contributions | |
vel | +0.034 |
makeup | +0.033 |
face | +0.033 |
spit | +0.033 |
blush | +0.031 |
Neg logit contributions | |
rob | -0.046 |
Hipp | -0.042 |
zag | -0.042 |
Hero | -0.040 |
girls | -0.039 |
Token | hanging |
Token ID | 10938 |
Feature activation | +3.60e-02 |
Pos logit contributions | |
rob | +0.070 |
Hipp | +0.066 |
zag | +0.065 |
Hero | +0.063 |
Zig | +0.062 |
Neg logit contributions | |
vel | -0.043 |
makeup | -0.042 |
spit | -0.042 |
face | -0.042 |
prise | -0.040 |
Token | from |
Token ID | 422 |
Feature activation | +4.70e-02 |
Pos logit contributions | |
rob | +0.078 |
zag | +0.073 |
Hipp | +0.072 |
Hero | +0.069 |
steep | +0.068 |
Neg logit contributions | |
vel | -0.043 |
makeup | -0.043 |
face | -0.043 |
spit | -0.042 |
blush | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | +4.41e-02 |
Pos logit contributions | |
rob | +0.109 |
zag | +0.103 |
Hipp | +0.101 |
girls | +0.097 |
Hero | +0.096 |
Neg logit contributions | |
makeup | -0.051 |
face | -0.049 |
vel | -0.048 |
spit | -0.048 |
her | -0.046 |
Token | z |
Token ID | 1976 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
rob | +0.081 |
Hipp | +0.078 |
zag | +0.077 |
steep | +0.075 |
Hero | +0.074 |
Neg logit contributions | |
vel | -0.068 |
spit | -0.062 |
makeup | -0.062 |
face | -0.060 |
prise | -0.060 |
Token | ig |
Token ID | 328 |
Feature activation | +8.31e-02 |
Pos logit contributions | |
rob | +0.248 |
Hipp | +0.230 |
zag | +0.229 |
Hero | +0.218 |
girls | +0.215 |
Neg logit contributions | |
vel | -0.191 |
makeup | -0.189 |
face | -0.188 |
spit | -0.184 |
blush | -0.174 |
Token | zag |
Token ID | 50183 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
rob | +0.145 |
zag | +0.133 |
Hipp | +0.132 |
Hero | +0.125 |
girls | +0.123 |
Neg logit contributions | |
vel | -0.134 |
makeup | -0.134 |
face | -0.133 |
spit | -0.131 |
blush | -0.125 |
Token | . |
Token ID | 13 |
Feature activation | -6.40e-03 |
Pos logit contributions | |
rob | +0.718 |
zag | +0.691 |
Hipp | +0.666 |
steep | +0.642 |
Hero | +0.636 |
Neg logit contributions | |
vel | -0.497 |
makeup | -0.493 |
spit | -0.484 |
face | -0.476 |
blush | -0.452 |
Token | M |
Token ID | 337 |
Feature activation | -7.25e-03 |
Pos logit contributions | |
face | +0.010 |
spit | +0.010 |
vel | +0.009 |
makeup | +0.009 |
She | +0.009 |
Neg logit contributions | |
rob | -0.013 |
zag | -0.011 |
Hipp | -0.011 |
girls | -0.011 |
steep | -0.010 |
Token | omm |
Token ID | 2002 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
vel | +0.009 |
makeup | +0.009 |
face | +0.009 |
spit | +0.009 |
blush | +0.009 |
Neg logit contributions | |
rob | -0.015 |
zag | -0.014 |
Hipp | -0.014 |
Zig | -0.013 |
steep | -0.013 |
Token | a |
Token ID | 64 |
Feature activation | +4.08e-02 |
Pos logit contributions | |
zag | +0.085 |
rob | +0.085 |
Hipp | +0.080 |
girls | +0.077 |
Zig | +0.076 |
Neg logit contributions | |
makeup | -0.049 |
vel | -0.044 |
face | -0.044 |
blush | -0.043 |
spit | -0.043 |
Token | repeated |
Token ID | 5100 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
rob | +0.082 |
zag | +0.077 |
Hipp | +0.076 |
Hero | +0.073 |
girls | +0.072 |
Neg logit contributions | |
vel | -0.054 |
makeup | -0.054 |
face | -0.053 |
spit | -0.052 |
blush | -0.050 |
Token | side |
Token ID | 1735 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
rob | +0.047 |
zag | +0.044 |
Hipp | +0.044 |
Hero | +0.042 |
girls | +0.041 |
Neg logit contributions | |
vel | -0.023 |
makeup | -0.023 |
face | -0.023 |
spit | -0.022 |
blush | -0.021 |
Token | of |
Token ID | 286 |
Feature activation | -7.13e-02 |
Pos logit contributions | |
vel | +0.043 |
spit | +0.042 |
makeup | +0.042 |
face | +0.041 |
blush | +0.039 |
Neg logit contributions | |
rob | -0.076 |
zag | -0.073 |
Hipp | -0.073 |
Hero | -0.070 |
Zig | -0.070 |
Token | the |
Token ID | 262 |
Feature activation | +4.89e-02 |
Pos logit contributions | |
vel | +0.093 |
makeup | +0.090 |
face | +0.090 |
spit | +0.089 |
blush | +0.084 |
Neg logit contributions | |
rob | -0.149 |
Hipp | -0.139 |
zag | -0.138 |
Hero | -0.133 |
steep | -0.131 |
Token | z |
Token ID | 1976 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
rob | +0.081 |
Hipp | +0.076 |
zag | +0.074 |
steep | +0.073 |
Hero | +0.070 |
Neg logit contributions | |
vel | -0.087 |
makeup | -0.082 |
spit | -0.082 |
face | -0.081 |
blush | -0.079 |
Token | ig |
Token ID | 328 |
Feature activation | +8.26e-02 |
Pos logit contributions | |
rob | +0.218 |
zag | +0.203 |
Hipp | +0.193 |
Zig | +0.180 |
girls | +0.180 |
Neg logit contributions | |
makeup | -0.230 |
spit | -0.228 |
vel | -0.226 |
face | -0.225 |
blush | -0.216 |
Token | zag |
Token ID | 50183 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
rob | +0.174 |
Hipp | +0.161 |
Hero | +0.153 |
zag | +0.152 |
girls | +0.150 |
Neg logit contributions | |
vel | -0.107 |
face | -0.107 |
makeup | -0.103 |
spit | -0.100 |
blush | -0.094 |
Token | , |
Token ID | 11 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
rob | +0.729 |
zag | +0.677 |
Hipp | +0.674 |
Hero | +0.641 |
girls | +0.633 |
Neg logit contributions | |
vel | -0.494 |
makeup | -0.491 |
face | -0.486 |
spit | -0.479 |
blush | -0.450 |
Token | to |
Token ID | 284 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
vel | +0.023 |
makeup | +0.023 |
face | +0.023 |
spit | +0.022 |
blush | +0.021 |
Neg logit contributions | |
rob | -0.043 |
zag | -0.041 |
Hipp | -0.040 |
Hero | -0.039 |
girls | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | +4.75e-02 |
Pos logit contributions | |
rob | +0.043 |
zag | +0.040 |
Hipp | +0.040 |
Hero | +0.039 |
steep | +0.038 |
Neg logit contributions | |
vel | -0.021 |
makeup | -0.021 |
face | -0.021 |
spit | -0.021 |
blush | -0.019 |
Token | other |
Token ID | 584 |
Feature activation | +2.41e-02 |
Pos logit contributions | |
rob | +0.090 |
Hipp | +0.086 |
zag | +0.085 |
steep | +0.082 |
Hero | +0.081 |
Neg logit contributions | |
vel | -0.070 |
makeup | -0.068 |
spit | -0.065 |
face | -0.064 |
prise | -0.063 |
Token | side |
Token ID | 1735 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
rob | +0.043 |
Hipp | +0.043 |
Zig | +0.040 |
steep | +0.040 |
zag | +0.040 |
Neg logit contributions | |
vel | -0.039 |
makeup | -0.038 |
prise | -0.037 |
spit | -0.036 |
blush | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | +4.53e-02 |
Pos logit contributions | |
Hipp | +0.012 |
Zig | +0.012 |
aboard | +0.011 |
Hero | +0.011 |
zag | +0.011 |
Neg logit contributions | |
prise | -0.014 |
vel | -0.013 |
aky | -0.012 |
spit | -0.012 |
P | -0.012 |
Token | rain |
Token ID | 6290 |
Feature activation | +4.43e-02 |
Pos logit contributions | |
Hipp | +0.085 |
rob | +0.084 |
zag | +0.082 |
steep | +0.079 |
Zig | +0.076 |
Neg logit contributions | |
vel | -0.067 |
makeup | -0.059 |
spit | -0.059 |
prise | -0.058 |
face | -0.058 |
Token | drops |
Token ID | 49253 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
rob | +0.083 |
zag | +0.082 |
Hipp | +0.078 |
steep | +0.076 |
Zig | +0.075 |
Neg logit contributions | |
vel | -0.065 |
makeup | -0.063 |
spit | -0.062 |
prise | -0.060 |
face | -0.060 |
Token | z |
Token ID | 1976 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
zag | +0.286 |
rob | +0.280 |
steep | +0.273 |
Hipp | +0.270 |
Zig | +0.263 |
Neg logit contributions | |
vel | -0.189 |
makeup | -0.186 |
prise | -0.184 |
spit | -0.178 |
face | -0.171 |
Token | ig |
Token ID | 328 |
Feature activation | +8.38e-02 |
Pos logit contributions | |
zag | +0.229 |
rob | +0.229 |
aboard | +0.194 |
Hipp | +0.193 |
girls | +0.187 |
Neg logit contributions | |
spit | -0.203 |
face | -0.201 |
worry | -0.196 |
makeup | -0.191 |
She | -0.190 |
Token | zag |
Token ID | 50183 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
rob | +0.146 |
zag | +0.142 |
Hipp | +0.135 |
Zig | +0.128 |
steep | +0.127 |
Neg logit contributions | |
makeup | -0.133 |
vel | -0.133 |
spit | -0.130 |
face | -0.129 |
prise | -0.124 |
Token | down |
Token ID | 866 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
zag | +0.695 |
rob | +0.687 |
Hipp | +0.652 |
steep | +0.651 |
Zig | +0.633 |
Neg logit contributions | |
vel | -0.497 |
makeup | -0.492 |
spit | -0.475 |
prise | -0.469 |
face | -0.455 |
Token | the |
Token ID | 262 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
rob | +0.028 |
zag | +0.026 |
Hipp | +0.026 |
Hero | +0.025 |
steep | +0.025 |
Neg logit contributions | |
vel | -0.014 |
makeup | -0.013 |
face | -0.013 |
spit | -0.013 |
blush | -0.012 |
Token | window |
Token ID | 4324 |
Feature activation | -4.45e-02 |
Pos logit contributions | |
rob | +0.085 |
zag | +0.079 |
Hipp | +0.079 |
Hero | +0.075 |
steep | +0.074 |
Neg logit contributions | |
vel | -0.067 |
makeup | -0.066 |
face | -0.065 |
spit | -0.065 |
blush | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | -6.36e-03 |
Pos logit contributions | |
vel | +0.057 |
makeup | +0.056 |
prise | +0.055 |
spit | +0.054 |
blush | +0.049 |
Neg logit contributions | |
zag | -0.090 |
rob | -0.086 |
Hipp | -0.084 |
steep | -0.083 |
Zig | -0.080 |
Token | She |
Token ID | 1375 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
makeup | +0.007 |
face | +0.007 |
spit | +0.007 |
vel | +0.006 |
blush | +0.006 |
Neg logit contributions | |
rob | -0.015 |
zag | -0.014 |
Hipp | -0.014 |
girls | -0.013 |
Hero | -0.013 |
Token | hill |
Token ID | 12788 |
Feature activation | +7.80e-02 |
Pos logit contributions | |
rob | +0.080 |
zag | +0.074 |
Hipp | +0.074 |
Hero | +0.070 |
steep | +0.069 |
Neg logit contributions | |
vel | -0.058 |
makeup | -0.058 |
face | -0.057 |
spit | -0.057 |
blush | -0.053 |
Token | and |
Token ID | 290 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
rob | +0.212 |
Hipp | +0.195 |
girls | +0.191 |
zag | +0.185 |
Hero | +0.184 |
Neg logit contributions | |
face | -0.056 |
her | -0.049 |
makeup | -0.049 |
spit | -0.047 |
vel | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | +4.75e-02 |
Pos logit contributions | |
rob | +0.042 |
zag | +0.040 |
Hipp | +0.038 |
girls | +0.036 |
xy | +0.035 |
Neg logit contributions | |
face | -0.015 |
spit | -0.015 |
her | -0.015 |
makeup | -0.014 |
dust | -0.013 |
Token | z |
Token ID | 1976 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
rob | +0.087 |
zag | +0.081 |
Hipp | +0.079 |
Hero | +0.075 |
girls | +0.075 |
Neg logit contributions | |
makeup | -0.073 |
face | -0.072 |
vel | -0.071 |
spit | -0.071 |
blush | -0.067 |
Token | ig |
Token ID | 328 |
Feature activation | +8.19e-02 |
Pos logit contributions | |
rob | +0.262 |
Hipp | +0.247 |
zag | +0.243 |
Hero | +0.235 |
steep | +0.234 |
Neg logit contributions | |
vel | -0.171 |
makeup | -0.168 |
face | -0.167 |
spit | -0.163 |
blush | -0.152 |
Token | zag |
Token ID | 50183 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
rob | +0.102 |
Hipp | +0.090 |
girls | +0.082 |
Hero | +0.082 |
Skip | +0.079 |
Neg logit contributions | |
face | -0.177 |
vel | -0.173 |
spit | -0.165 |
makeup | -0.165 |
her | -0.164 |
Token | path |
Token ID | 3108 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
rob | +0.765 |
Hipp | +0.702 |
Hero | +0.666 |
girls | +0.661 |
Skip | +0.640 |
Neg logit contributions | |
face | -0.517 |
spit | -0.466 |
vel | -0.454 |
makeup | -0.453 |
dust | -0.433 |
Token | in |
Token ID | 287 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
rob | +0.083 |
Hipp | +0.077 |
girls | +0.075 |
zag | +0.074 |
Hero | +0.073 |
Neg logit contributions | |
face | -0.025 |
her | -0.022 |
makeup | -0.022 |
spit | -0.021 |
vel | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | +5.12e-02 |
Pos logit contributions | |
rob | +0.041 |
zag | +0.038 |
Hipp | +0.037 |
girls | +0.036 |
Hero | +0.036 |
Neg logit contributions | |
makeup | -0.019 |
face | -0.019 |
vel | -0.019 |
spit | -0.019 |
her | -0.017 |
Token | rear |
Token ID | 8286 |
Feature activation | +7.03e-02 |
Pos logit contributions | |
rob | +0.097 |
zag | +0.090 |
Hipp | +0.089 |
Hero | +0.085 |
steep | +0.084 |
Neg logit contributions | |
vel | -0.076 |
makeup | -0.075 |
face | -0.074 |
spit | -0.074 |
blush | -0.070 |
Token | view |
Token ID | 1177 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
rob | +0.152 |
Hipp | +0.144 |
Hero | +0.139 |
Skip | +0.136 |
girls | +0.135 |
Neg logit contributions | |
face | -0.089 |
vel | -0.081 |
makeup | -0.081 |
spit | -0.080 |
ime | -0.076 |
Token | he |
Token ID | 339 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
rob | +0.150 |
zag | +0.138 |
Hipp | +0.134 |
girls | +0.128 |
steep | +0.128 |
Neg logit contributions | |
face | -0.103 |
makeup | -0.103 |
vel | -0.101 |
spit | -0.100 |
blush | -0.094 |
Token | saw |
Token ID | 2497 |
Feature activation | -7.10e-02 |
Pos logit contributions | |
rob | +0.085 |
Hipp | +0.079 |
zag | +0.078 |
Hero | +0.074 |
Zig | +0.074 |
Neg logit contributions | |
spit | -0.035 |
face | -0.035 |
vel | -0.034 |
makeup | -0.033 |
blush | -0.030 |
Token | a |
Token ID | 257 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
makeup | +0.068 |
face | +0.067 |
vel | +0.066 |
spit | +0.066 |
blush | +0.059 |
Neg logit contributions | |
rob | -0.173 |
zag | -0.164 |
Hipp | -0.160 |
girls | -0.154 |
steep | -0.154 |
Token | z |
Token ID | 1976 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
vel | +0.015 |
makeup | +0.014 |
spit | +0.013 |
blush | +0.013 |
prise | +0.013 |
Neg logit contributions | |
steep | -0.022 |
zag | -0.022 |
Hipp | -0.021 |
rob | -0.021 |
Hero | -0.020 |
Token | ig |
Token ID | 328 |
Feature activation | +8.23e-02 |
Pos logit contributions | |
zag | +0.188 |
rob | +0.188 |
Hipp | +0.163 |
Zig | +0.156 |
Hero | +0.155 |
Neg logit contributions | |
makeup | -0.251 |
spit | -0.248 |
vel | -0.243 |
face | -0.239 |
blush | -0.234 |
Token | zag |
Token ID | 50183 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
rob | +0.134 |
zag | +0.129 |
Hipp | +0.121 |
steep | +0.115 |
Zig | +0.115 |
Neg logit contributions | |
makeup | -0.142 |
vel | -0.140 |
spit | -0.138 |
face | -0.137 |
blush | -0.132 |
Token | branch |
Token ID | 8478 |
Feature activation | +7.91e-02 |
Pos logit contributions | |
rob | +0.682 |
zag | +0.676 |
steep | +0.628 |
Hipp | +0.627 |
Zig | +0.604 |
Neg logit contributions | |
makeup | -0.523 |
vel | -0.519 |
spit | -0.498 |
prise | -0.492 |
face | -0.482 |
Token | and |
Token ID | 290 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
rob | +0.182 |
zag | +0.172 |
Hipp | +0.170 |
Hero | +0.163 |
steep | +0.163 |
Neg logit contributions | |
vel | -0.084 |
makeup | -0.084 |
face | -0.081 |
spit | -0.081 |
blush | -0.075 |
Token | decided |
Token ID | 3066 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
rob | +0.035 |
zag | +0.032 |
Hipp | +0.032 |
girls | +0.031 |
Hero | +0.030 |
Neg logit contributions | |
spit | -0.018 |
face | -0.018 |
makeup | -0.018 |
vel | -0.018 |
blush | -0.017 |
Token | to |
Token ID | 284 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
rob | +0.042 |
zag | +0.037 |
girls | +0.036 |
Hipp | +0.035 |
Hero | +0.033 |
Neg logit contributions | |
face | -0.018 |
her | -0.017 |
makeup | -0.016 |
vel | -0.015 |
spit | -0.015 |
Token | try |
Token ID | 1949 |
Feature activation | -5.58e-02 |
Pos logit contributions | |
zag | +0.039 |
Hipp | +0.038 |
rob | +0.038 |
Zig | +0.037 |
Hero | +0.037 |
Neg logit contributions | |
vel | -0.021 |
makeup | -0.021 |
prise | -0.019 |
face | -0.019 |
spit | -0.018 |
Token | slide |
Token ID | 10649 |
Feature activation | +5.91e-02 |
Pos logit contributions | |
vel | +0.139 |
spit | +0.130 |
makeup | +0.129 |
face | +0.123 |
blush | +0.123 |
Neg logit contributions | |
rob | -0.163 |
Hipp | -0.154 |
steep | -0.150 |
zag | -0.149 |
Hero | -0.146 |
Token | with |
Token ID | 351 |
Feature activation | -1.20e-03 |
Pos logit contributions | |
rob | +0.150 |
Hipp | +0.142 |
zag | +0.139 |
girls | +0.137 |
Hero | +0.137 |
Neg logit contributions | |
makeup | -0.050 |
face | -0.049 |
vel | -0.047 |
spit | -0.046 |
blush | -0.042 |
Token | a |
Token ID | 257 |
Feature activation | -9.68e-03 |
Pos logit contributions | |
vel | +0.001 |
makeup | +0.001 |
face | +0.001 |
spit | +0.001 |
prise | +0.001 |
Neg logit contributions | |
rob | -0.003 |
Hipp | -0.003 |
zag | -0.002 |
steep | -0.002 |
Hero | -0.002 |
Token | z |
Token ID | 1976 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
vel | +0.014 |
makeup | +0.013 |
spit | +0.013 |
face | +0.012 |
prise | +0.012 |
Neg logit contributions | |
rob | -0.019 |
zag | -0.018 |
steep | -0.018 |
Hipp | -0.018 |
Hero | -0.017 |
Token | ig |
Token ID | 328 |
Feature activation | +8.08e-02 |
Pos logit contributions | |
rob | +0.209 |
zag | +0.198 |
Hipp | +0.181 |
Zig | +0.171 |
steep | +0.170 |
Neg logit contributions | |
makeup | -0.220 |
vel | -0.218 |
spit | -0.218 |
face | -0.213 |
blush | -0.206 |
Token | zag |
Token ID | 50183 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
rob | +0.090 |
Hipp | +0.080 |
Hero | +0.074 |
girls | +0.073 |
Skip | +0.070 |
Neg logit contributions | |
face | -0.185 |
vel | -0.182 |
makeup | -0.180 |
spit | -0.177 |
blush | -0.172 |
Token | shape |
Token ID | 5485 |
Feature activation | -7.67e-02 |
Pos logit contributions | |
rob | +0.784 |
Hipp | +0.734 |
zag | +0.715 |
Hero | +0.704 |
girls | +0.696 |
Neg logit contributions | |
face | -0.443 |
makeup | -0.437 |
vel | -0.434 |
spit | -0.428 |
blush | -0.398 |
Token | . |
Token ID | 13 |
Feature activation | -2.39e-03 |
Pos logit contributions | |
vel | +0.077 |
makeup | +0.076 |
face | +0.074 |
spit | +0.073 |
prise | +0.068 |
Neg logit contributions | |
rob | -0.180 |
zag | -0.172 |
Hipp | -0.169 |
steep | -0.162 |
Hero | -0.161 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
makeup | +0.003 |
face | +0.003 |
spit | +0.003 |
vel | +0.003 |
blush | +0.003 |
Neg logit contributions | |
rob | -0.005 |
zag | -0.005 |
steep | -0.004 |
Hipp | -0.004 |
girls | -0.004 |
Token | wanted |
Token ID | 2227 |
Feature activation | -6.83e-02 |
Pos logit contributions | |
face | +0.032 |
vel | +0.031 |
spit | +0.031 |
makeup | +0.031 |
blush | +0.028 |
Neg logit contributions | |
rob | -0.066 |
Hipp | -0.061 |
zag | -0.060 |
Hero | -0.058 |
girls | -0.058 |
Token | to |
Token ID | 284 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
makeup | +0.083 |
her | +0.079 |
face | +0.074 |
clean | +0.068 |
spit | +0.068 |
Neg logit contributions | |
rob | -0.164 |
zag | -0.149 |
Hipp | -0.142 |
girls | -0.140 |
Hero | -0.138 |
Token | a |
Token ID | 257 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
makeup | +0.003 |
face | +0.003 |
spit | +0.002 |
vel | +0.002 |
her | +0.002 |
Neg logit contributions | |
rob | -0.006 |
zag | -0.006 |
Hipp | -0.006 |
girls | -0.005 |
Hero | -0.005 |
Token | big |
Token ID | 1263 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
vel | +0.033 |
makeup | +0.032 |
face | +0.032 |
spit | +0.031 |
blush | +0.029 |
Neg logit contributions | |
rob | -0.071 |
zag | -0.066 |
Hipp | -0.066 |
Hero | -0.063 |
girls | -0.063 |
Token | smile |
Token ID | 8212 |
Feature activation | -5.54e-02 |
Pos logit contributions | |
vel | +0.121 |
makeup | +0.118 |
spit | +0.115 |
face | +0.115 |
blush | +0.107 |
Neg logit contributions | |
rob | -0.244 |
zag | -0.229 |
Hipp | -0.228 |
Hero | -0.219 |
steep | -0.217 |
Token | on |
Token ID | 319 |
Feature activation | +3.35e-02 |
Pos logit contributions | |
vel | +0.062 |
makeup | +0.061 |
spit | +0.060 |
face | +0.059 |
prise | +0.057 |
Neg logit contributions | |
rob | -0.123 |
zag | -0.118 |
Hipp | -0.115 |
Hero | -0.111 |
steep | -0.109 |
Token | his |
Token ID | 465 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
rob | +0.056 |
Hipp | +0.053 |
Hero | +0.052 |
zag | +0.052 |
Skip | +0.050 |
Neg logit contributions | |
vel | -0.055 |
prise | -0.053 |
makeup | -0.053 |
spit | -0.050 |
blush | -0.049 |
Token | face |
Token ID | 1986 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
vel | +0.012 |
prise | +0.012 |
makeup | +0.010 |
She | +0.008 |
spit | +0.008 |
Neg logit contributions | |
rob | -0.086 |
zag | -0.086 |
Hipp | -0.083 |
Hero | -0.083 |
steep | -0.081 |
Token | and |
Token ID | 290 |
Feature activation | -6.00e-03 |
Pos logit contributions | |
vel | +0.264 |
makeup | +0.262 |
face | +0.255 |
spit | +0.255 |
blush | +0.233 |
Neg logit contributions | |
rob | -0.581 |
zag | -0.551 |
Hipp | -0.546 |
Hero | -0.524 |
steep | -0.519 |
Token | was |
Token ID | 373 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
face | +0.008 |
spit | +0.007 |
makeup | +0.007 |
her | +0.007 |
blush | +0.007 |
Neg logit contributions | |
rob | -0.013 |
zag | -0.012 |
Hipp | -0.012 |
steep | -0.012 |
Hero | -0.011 |
Token | carrying |
Token ID | 6872 |
Feature activation | +3.37e-03 |
Pos logit contributions | |
rob | +0.032 |
zag | +0.029 |
Hipp | +0.029 |
Hero | +0.028 |
girls | +0.027 |
Neg logit contributions | |
face | -0.014 |
vel | -0.014 |
makeup | -0.014 |
spit | -0.013 |
blush | -0.012 |
Token | his |
Token ID | 465 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
rob | +0.007 |
Hipp | +0.006 |
Hero | +0.006 |
zag | +0.006 |
Zig | +0.006 |
Neg logit contributions | |
vel | -0.004 |
spit | -0.004 |
makeup | -0.004 |
face | -0.004 |
prise | -0.004 |
Token | own |
Token ID | 898 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
face | +0.042 |
makeup | +0.040 |
spit | +0.039 |
vel | +0.037 |
hand | +0.036 |
Neg logit contributions | |
rob | -0.065 |
Hipp | -0.058 |
zag | -0.058 |
Zig | -0.056 |
girls | -0.055 |
Token | felt |
Token ID | 2936 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
spit | +0.024 |
face | +0.024 |
makeup | +0.023 |
vel | +0.023 |
blush | +0.021 |
Neg logit contributions | |
rob | -0.055 |
zag | -0.051 |
Hipp | -0.051 |
girls | -0.049 |
Hero | -0.049 |
Token | a |
Token ID | 257 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
vel | +0.046 |
makeup | +0.041 |
prise | +0.040 |
spit | +0.039 |
face | +0.039 |
Neg logit contributions | |
Hipp | -0.070 |
rob | -0.070 |
steep | -0.069 |
Zig | -0.067 |
Hero | -0.066 |
Token | breeze |
Token ID | 28633 |
Feature activation | -5.32e-02 |
Pos logit contributions | |
vel | +0.046 |
makeup | +0.044 |
ily | +0.042 |
prise | +0.042 |
on | +0.041 |
Neg logit contributions | |
steep | -0.055 |
Hipp | -0.047 |
zag | -0.047 |
field | -0.046 |
camps | -0.044 |
Token | on |
Token ID | 319 |
Feature activation | +3.52e-02 |
Pos logit contributions | |
prise | +0.070 |
makeup | +0.069 |
vel | +0.069 |
spit | +0.067 |
face | +0.062 |
Neg logit contributions | |
zag | -0.108 |
rob | -0.105 |
Hipp | -0.101 |
steep | -0.096 |
aboard | -0.095 |
Token | his |
Token ID | 465 |
Feature activation | -2.76e-02 |
Pos logit contributions | |
rob | +0.054 |
Hipp | +0.054 |
Hero | +0.053 |
Zig | +0.052 |
Zac | +0.051 |
Neg logit contributions | |
vel | -0.063 |
prise | -0.061 |
makeup | -0.058 |
spit | -0.056 |
blush | -0.054 |
Token | face |
Token ID | 1986 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
prise | +0.022 |
vel | +0.022 |
illy | +0.016 |
spit | +0.015 |
pped | +0.015 |
Neg logit contributions | |
zag | -0.069 |
Hero | -0.066 |
rob | -0.065 |
steep | -0.064 |
Zig | -0.064 |
Token | . |
Token ID | 13 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
prise | +0.332 |
vel | +0.327 |
makeup | +0.317 |
spit | +0.313 |
face | +0.294 |
Neg logit contributions | |
zag | -0.489 |
rob | -0.479 |
Hipp | -0.470 |
steep | -0.453 |
Hero | -0.445 |
Token | He |
Token ID | 679 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
vel | +0.046 |
makeup | +0.046 |
face | +0.046 |
spit | +0.045 |
blush | +0.043 |
Neg logit contributions | |
rob | -0.041 |
zag | -0.037 |
Hipp | -0.037 |
Hero | -0.034 |
girls | -0.034 |
Token | wondered |
Token ID | 14028 |
Feature activation | -3.71e-02 |
Pos logit contributions | |
spit | +0.026 |
face | +0.026 |
makeup | +0.026 |
vel | +0.026 |
blush | +0.024 |
Neg logit contributions | |
rob | -0.055 |
zag | -0.051 |
Hipp | -0.050 |
girls | -0.048 |
Hero | -0.048 |
Token | what |
Token ID | 644 |
Feature activation | -5.80e-02 |
Pos logit contributions | |
vel | +0.033 |
makeup | +0.032 |
spit | +0.031 |
face | +0.031 |
blush | +0.029 |
Neg logit contributions | |
rob | -0.090 |
Hipp | -0.086 |
zag | -0.085 |
Hero | -0.083 |
Zig | -0.083 |
Token | it |
Token ID | 340 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
makeup | +0.072 |
face | +0.071 |
spit | +0.070 |
vel | +0.068 |
her | +0.065 |
Neg logit contributions | |
rob | -0.125 |
zag | -0.120 |
Hipp | -0.115 |
girls | -0.111 |
Hero | -0.109 |
Token | with |
Token ID | 351 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
face | +0.009 |
vel | +0.008 |
makeup | +0.008 |
spit | +0.008 |
her | +0.007 |
Neg logit contributions | |
rob | -0.026 |
Hipp | -0.024 |
zag | -0.024 |
girls | -0.023 |
Hero | -0.023 |
Token | a |
Token ID | 257 |
Feature activation | -3.16e-02 |
Pos logit contributions | |
vel | +0.020 |
makeup | +0.019 |
spit | +0.018 |
face | +0.018 |
blush | +0.018 |
Neg logit contributions | |
rob | -0.035 |
Hipp | -0.033 |
zag | -0.032 |
Hero | -0.032 |
Zig | -0.032 |
Token | smile |
Token ID | 8212 |
Feature activation | -5.33e-02 |
Pos logit contributions | |
vel | +0.036 |
makeup | +0.035 |
spit | +0.034 |
face | +0.033 |
blush | +0.032 |
Neg logit contributions | |
rob | -0.070 |
zag | -0.067 |
Hipp | -0.066 |
steep | -0.064 |
Hero | -0.064 |
Token | on |
Token ID | 319 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
makeup | +0.056 |
vel | +0.055 |
spit | +0.054 |
face | +0.053 |
blush | +0.049 |
Neg logit contributions | |
rob | -0.123 |
zag | -0.117 |
Hipp | -0.116 |
Hero | -0.111 |
steep | -0.110 |
Token | his |
Token ID | 465 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
rob | +0.060 |
Hipp | +0.058 |
zag | +0.055 |
Hero | +0.055 |
Zig | +0.054 |
Neg logit contributions | |
vel | -0.054 |
makeup | -0.053 |
prise | -0.051 |
spit | -0.051 |
blush | -0.050 |
Token | face |
Token ID | 1986 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
vel | +0.015 |
makeup | +0.015 |
face | +0.014 |
spit | +0.013 |
blush | +0.011 |
Neg logit contributions | |
rob | -0.091 |
zag | -0.087 |
Hipp | -0.086 |
Hero | -0.084 |
steep | -0.083 |
Token | . |
Token ID | 13 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
makeup | +0.260 |
vel | +0.259 |
spit | +0.252 |
face | +0.244 |
prise | +0.235 |
Neg logit contributions | |
rob | -0.573 |
zag | -0.554 |
Hipp | -0.546 |
Hero | -0.522 |
steep | -0.520 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -6.12e-02 |
Pos logit contributions | |
face | +0.028 |
She | +0.027 |
vel | +0.026 |
her | +0.026 |
hand | +0.025 |
Neg logit contributions | |
rob | -0.054 |
zag | -0.050 |
girls | -0.047 |
Hipp | -0.046 |
steep | -0.045 |
Token | One |
Token ID | 3198 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
makeup | +0.086 |
vel | +0.084 |
face | +0.081 |
She | +0.079 |
her | +0.079 |
Neg logit contributions | |
rob | -0.114 |
zag | -0.111 |
Hipp | -0.106 |
steep | -0.105 |
girls | -0.103 |
Token | day |
Token ID | 1110 |
Feature activation | -7.98e-03 |
Pos logit contributions | |
rob | +0.210 |
zag | +0.197 |
Hipp | +0.194 |
steep | +0.183 |
Hero | +0.181 |
Neg logit contributions | |
vel | -0.223 |
makeup | -0.220 |
spit | -0.216 |
face | -0.215 |
blush | -0.207 |
Token | , |
Token ID | 11 |
Feature activation | -4.28e-02 |
Pos logit contributions | |
vel | +0.010 |
makeup | +0.010 |
spit | +0.010 |
face | +0.009 |
blush | +0.009 |
Neg logit contributions | |
rob | -0.017 |
Hipp | -0.016 |
zag | -0.016 |
Zig | -0.015 |
Hero | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
rob | +0.267 |
zag | +0.257 |
Hipp | +0.250 |
Hero | +0.246 |
steep | +0.244 |
Neg logit contributions | |
vel | -0.136 |
spit | -0.128 |
makeup | -0.128 |
face | -0.123 |
prise | -0.122 |
Token | She |
Token ID | 1375 |
Feature activation | -5.67e-02 |
Pos logit contributions | |
vel | +0.035 |
prise | +0.032 |
makeup | +0.029 |
spit | +0.028 |
face | +0.027 |
Neg logit contributions | |
Hipp | -0.056 |
rob | -0.054 |
Zig | -0.052 |
Hero | -0.052 |
zag | -0.052 |
Token | sees |
Token ID | 7224 |
Feature activation | -4.55e-02 |
Pos logit contributions | |
makeup | +0.095 |
face | +0.094 |
vel | +0.094 |
spit | +0.093 |
blush | +0.088 |
Neg logit contributions | |
rob | -0.097 |
zag | -0.090 |
Hipp | -0.089 |
Hero | -0.083 |
girls | -0.082 |
Token | mom |
Token ID | 1995 |
Feature activation | -4.86e-02 |
Pos logit contributions | |
vel | +0.067 |
prise | +0.057 |
face | +0.056 |
makeup | +0.056 |
spit | +0.055 |
Neg logit contributions | |
rob | -0.090 |
Hipp | -0.087 |
Hero | -0.082 |
Zig | -0.081 |
zag | -0.080 |
Token | 's |
Token ID | 338 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
vel | +0.077 |
illy | +0.074 |
P | +0.073 |
prise | +0.067 |
goodness | +0.065 |
Neg logit contributions | |
girls | -0.074 |
ousel | -0.071 |
zag | -0.069 |
rob | -0.067 |
drives | -0.066 |
Token | face |
Token ID | 1986 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
vel | +0.039 |
pped | +0.034 |
She | +0.033 |
P | +0.033 |
prise | +0.033 |
Neg logit contributions | |
zag | -0.050 |
fault | -0.047 |
girls | -0.046 |
missions | -0.045 |
Hero | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
prise | +0.396 |
vel | +0.372 |
goodness | +0.332 |
makeup | +0.326 |
illy | +0.322 |
Neg logit contributions | |
zag | -0.420 |
steep | -0.391 |
Hipp | -0.382 |
rob | -0.380 |
Hero | -0.367 |
Token | She |
Token ID | 1375 |
Feature activation | -5.74e-02 |
Pos logit contributions | |
vel | +0.037 |
prise | +0.033 |
makeup | +0.030 |
spit | +0.029 |
aming | +0.028 |
Neg logit contributions | |
Hipp | -0.057 |
rob | -0.055 |
Zig | -0.053 |
Hero | -0.052 |
zag | -0.052 |
Token | is |
Token ID | 318 |
Feature activation | +1.93e-02 |
Pos logit contributions | |
vel | +0.101 |
makeup | +0.098 |
face | +0.096 |
spit | +0.095 |
blush | +0.092 |
Neg logit contributions | |
rob | -0.093 |
Hipp | -0.084 |
zag | -0.084 |
Hero | -0.080 |
steep | -0.078 |
Token | scared |
Token ID | 12008 |
Feature activation | +4.68e-02 |
Pos logit contributions | |
zag | +0.031 |
Hipp | +0.031 |
rob | +0.030 |
Hero | +0.029 |
steep | +0.029 |
Neg logit contributions | |
vel | -0.029 |
goodness | -0.026 |
prise | -0.025 |
illy | -0.025 |
file | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
lights | +0.051 |
Hipp | +0.050 |
steep | +0.047 |
tigers | +0.045 |
Hero | +0.044 |
Neg logit contributions | |
prise | -0.096 |
vel | -0.096 |
file | -0.085 |
spit | -0.085 |
P | -0.084 |
Token | a |
Token ID | 257 |
Feature activation | -3.03e-02 |
Pos logit contributions | |
vel | +0.025 |
makeup | +0.023 |
prise | +0.023 |
spit | +0.023 |
blush | +0.022 |
Neg logit contributions | |
rob | -0.027 |
Hipp | -0.026 |
Hero | -0.026 |
Zig | -0.025 |
steep | -0.025 |
Token | big |
Token ID | 1263 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
vel | +0.039 |
makeup | +0.037 |
spit | +0.035 |
illy | +0.035 |
Her | +0.034 |
Neg logit contributions | |
steep | -0.059 |
zag | -0.058 |
rob | -0.058 |
Hipp | -0.056 |
Hero | -0.055 |
Token | smile |
Token ID | 8212 |
Feature activation | -5.40e-02 |
Pos logit contributions | |
vel | +0.142 |
makeup | +0.132 |
Her | +0.132 |
spit | +0.131 |
She | +0.130 |
Neg logit contributions | |
rob | -0.209 |
steep | -0.204 |
Hipp | -0.202 |
zag | -0.198 |
Hero | -0.197 |
Token | on |
Token ID | 319 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
makeup | +0.062 |
spit | +0.061 |
vel | +0.060 |
prise | +0.058 |
face | +0.058 |
Neg logit contributions | |
rob | -0.118 |
zag | -0.115 |
Hipp | -0.110 |
Hero | -0.107 |
steep | -0.104 |
Token | his |
Token ID | 465 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
Hipp | +0.050 |
rob | +0.049 |
Hero | +0.048 |
zag | +0.047 |
Zig | +0.046 |
Neg logit contributions | |
prise | -0.060 |
vel | -0.059 |
makeup | -0.058 |
blush | -0.054 |
spit | -0.054 |
Token | face |
Token ID | 1986 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
prise | +0.015 |
vel | +0.014 |
makeup | +0.012 |
She | +0.011 |
spit | +0.010 |
Neg logit contributions | |
zag | -0.082 |
rob | -0.081 |
Hero | -0.080 |
Hipp | -0.080 |
Zig | -0.078 |
Token | . |
Token ID | 13 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
vel | +0.272 |
makeup | +0.271 |
spit | +0.267 |
prise | +0.258 |
face | +0.254 |
Neg logit contributions | |
rob | -0.550 |
zag | -0.539 |
Hipp | -0.528 |
Hero | -0.504 |
steep | -0.501 |
Token |
|
Token ID | 198 |
Feature activation | -6.21e-03 |
Pos logit contributions | |
face | +0.032 |
spit | +0.030 |
She | +0.029 |
vel | +0.028 |
hand | +0.028 |
Neg logit contributions | |
rob | -0.052 |
zag | -0.046 |
Hipp | -0.046 |
girls | -0.044 |
Zig | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | -6.60e-03 |
Pos logit contributions | |
face | +0.007 |
vel | +0.007 |
spit | +0.007 |
She | +0.007 |
makeup | +0.007 |
Neg logit contributions | |
rob | -0.014 |
Hipp | -0.013 |
zag | -0.013 |
girls | -0.012 |
Hero | -0.012 |
Token | The |
Token ID | 464 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
vel | +0.009 |
makeup | +0.009 |
face | +0.009 |
spit | +0.009 |
blush | +0.008 |
Neg logit contributions | |
rob | -0.013 |
zag | -0.013 |
Hipp | -0.012 |
Hero | -0.012 |
girls | -0.012 |
Token | farmer |
Token ID | 18739 |
Feature activation | -4.26e-03 |
Pos logit contributions | |
rob | +0.037 |
Hipp | +0.035 |
zag | +0.034 |
Hero | +0.034 |
steep | +0.033 |
Neg logit contributions | |
vel | -0.035 |
makeup | -0.034 |
spit | -0.033 |
blush | -0.033 |
face | -0.032 |
Token | pe |
Token ID | 613 |
Feature activation | -2.33e-03 |
Pos logit contributions | |
Hipp | +0.042 |
rob | +0.041 |
zag | +0.041 |
steep | +0.040 |
Claw | +0.039 |
Neg logit contributions | |
vel | -0.041 |
makeup | -0.035 |
blush | -0.035 |
She | -0.035 |
prise | -0.034 |
Token | aches |
Token ID | 3694 |
Feature activation | +7.61e-02 |
Pos logit contributions | |
makeup | +0.004 |
face | +0.004 |
vel | +0.004 |
spit | +0.004 |
blush | +0.004 |
Neg logit contributions | |
rob | -0.004 |
zag | -0.004 |
Hipp | -0.004 |
steep | -0.003 |
Zig | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | -5.37e-03 |
Pos logit contributions | |
rob | +0.145 |
zag | +0.144 |
steep | +0.140 |
Hipp | +0.135 |
Hero | +0.132 |
Neg logit contributions | |
vel | -0.104 |
makeup | -0.101 |
spit | -0.099 |
prise | -0.099 |
face | -0.095 |
Token | made |
Token ID | 925 |
Feature activation | -2.63e-03 |
Pos logit contributions | |
vel | +0.008 |
makeup | +0.007 |
prise | +0.007 |
face | +0.007 |
ime | +0.006 |
Neg logit contributions | |
rob | -0.010 |
Hipp | -0.010 |
Hero | -0.009 |
girls | -0.009 |
Zig | -0.009 |
Token | a |
Token ID | 257 |
Feature activation | -3.17e-02 |
Pos logit contributions | |
vel | +0.003 |
makeup | +0.003 |
face | +0.003 |
spit | +0.003 |
prise | +0.003 |
Neg logit contributions | |
rob | -0.006 |
Hipp | -0.005 |
zag | -0.005 |
Zig | -0.005 |
Hero | -0.005 |
Token | face |
Token ID | 1986 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
vel | +0.043 |
makeup | +0.040 |
pped | +0.040 |
illy | +0.039 |
aming | +0.038 |
Neg logit contributions | |
steep | -0.063 |
zag | -0.061 |
Hipp | -0.058 |
Zig | -0.056 |
rob | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
prise | +0.347 |
makeup | +0.331 |
vel | +0.320 |
illy | +0.319 |
goodness | +0.315 |
Neg logit contributions | |
zag | -0.477 |
Hipp | -0.449 |
rob | -0.433 |
steep | -0.424 |
girls | -0.424 |
Token | They |
Token ID | 1119 |
Feature activation | +2.55e-03 |
Pos logit contributions | |
face | +0.045 |
spit | +0.043 |
hand | +0.039 |
her | +0.038 |
makeup | +0.037 |
Neg logit contributions | |
rob | -0.076 |
zag | -0.062 |
ousel | -0.056 |
ahs | -0.056 |
steep | -0.053 |
Token | did |
Token ID | 750 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
rob | +0.004 |
Hipp | +0.004 |
steep | +0.004 |
girls | +0.004 |
Hero | +0.004 |
Neg logit contributions | |
vel | -0.004 |
makeup | -0.003 |
prise | -0.003 |
goodness | -0.003 |
ime | -0.003 |
Token | not |
Token ID | 407 |
Feature activation | +4.62e-02 |
Pos logit contributions | |
face | +0.042 |
spit | +0.042 |
makeup | +0.041 |
vel | +0.040 |
blush | +0.038 |
Neg logit contributions | |
rob | -0.066 |
Hipp | -0.059 |
zag | -0.058 |
Hero | -0.055 |
girls | -0.055 |
Token | want |
Token ID | 765 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
steep | +0.066 |
Claw | +0.065 |
Hipp | +0.065 |
Hero | +0.062 |
common | +0.061 |
Neg logit contributions | |
vel | -0.072 |
aming | -0.068 |
goodness | -0.067 |
Her | -0.065 |
illy | -0.065 |
Token | a |
Token ID | 257 |
Feature activation | -3.18e-02 |
Pos logit contributions | |
vel | +0.021 |
makeup | +0.019 |
spit | +0.019 |
prise | +0.018 |
face | +0.018 |
Neg logit contributions | |
rob | -0.035 |
Hipp | -0.035 |
Hero | -0.033 |
Zig | -0.033 |
Skip | -0.033 |
Token | big |
Token ID | 1263 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
vel | +0.037 |
makeup | +0.037 |
spit | +0.035 |
face | +0.033 |
blush | +0.032 |
Neg logit contributions | |
rob | -0.069 |
zag | -0.067 |
Hipp | -0.065 |
steep | -0.064 |
Hero | -0.062 |
Token | smile |
Token ID | 8212 |
Feature activation | -5.41e-02 |
Pos logit contributions | |
vel | +0.125 |
makeup | +0.121 |
spit | +0.119 |
face | +0.113 |
prise | +0.109 |
Neg logit contributions | |
rob | -0.238 |
Hipp | -0.227 |
zag | -0.226 |
steep | -0.219 |
Hero | -0.216 |
Token | on |
Token ID | 319 |
Feature activation | +3.31e-02 |
Pos logit contributions | |
makeup | +0.068 |
spit | +0.068 |
prise | +0.066 |
vel | +0.065 |
face | +0.063 |
Neg logit contributions | |
zag | -0.110 |
rob | -0.110 |
Hipp | -0.104 |
Zig | -0.099 |
Hero | -0.098 |
Token | her |
Token ID | 607 |
Feature activation | -7.58e-02 |
Pos logit contributions | |
Hipp | +0.081 |
Zig | +0.078 |
rob | +0.076 |
zag | +0.076 |
Hero | +0.075 |
Neg logit contributions | |
prise | -0.028 |
vel | -0.025 |
makeup | -0.023 |
file | -0.021 |
spit | -0.020 |
Token | face |
Token ID | 1986 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
prise | +0.035 |
vel | +0.031 |
makeup | +0.030 |
spit | +0.026 |
She | +0.025 |
Neg logit contributions | |
zag | -0.212 |
rob | -0.211 |
Hipp | -0.206 |
Zig | -0.203 |
Hero | -0.202 |
Token | . |
Token ID | 13 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
vel | +0.280 |
makeup | +0.279 |
spit | +0.279 |
prise | +0.273 |
face | +0.261 |
Neg logit contributions | |
rob | -0.531 |
zag | -0.531 |
Hipp | -0.515 |
Zig | -0.488 |
steep | -0.487 |
Token | |
Token ID | 220 |
Feature activation | +5.38e-04 |
Pos logit contributions | |
vel | +0.028 |
makeup | +0.027 |
face | +0.027 |
She | +0.026 |
spit | +0.026 |
Neg logit contributions | |
rob | -0.054 |
zag | -0.051 |
Hipp | -0.049 |
steep | -0.048 |
girls | -0.048 |
Token |
|
Token ID | 628 |
Feature activation | +5.42e-02 |
Pos logit contributions | |
rob | +0.001 |
zag | +0.001 |
Hipp | +0.001 |
Hero | +0.001 |
steep | +0.001 |
Neg logit contributions | |
vel | -0.001 |
makeup | -0.001 |
face | -0.001 |
spit | -0.001 |
blush | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -8.24e-03 |
Pos logit contributions | |
rob | +0.110 |
zag | +0.103 |
Hipp | +0.101 |
steep | +0.097 |
Hero | +0.096 |
Neg logit contributions | |
vel | -0.072 |
makeup | -0.072 |
face | -0.071 |
spit | -0.070 |
blush | -0.066 |
Token |
|
Token ID | 198 |
Feature activation | -8.20e-03 |
Pos logit contributions | |
vel | +0.008 |
makeup | +0.008 |
face | +0.008 |
spit | +0.008 |
blush | +0.007 |
Neg logit contributions | |
rob | -0.020 |
zag | -0.018 |
Hipp | -0.018 |
Hero | -0.017 |
steep | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
prise | +0.101 |
vel | +0.099 |
makeup | +0.098 |
blush | +0.091 |
spit | +0.089 |
Neg logit contributions | |
zag | -0.134 |
rob | -0.128 |
steep | -0.124 |
Hipp | -0.121 |
Hero | -0.118 |
Token | He |
Token ID | 679 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
vel | +0.043 |
makeup | +0.042 |
face | +0.042 |
spit | +0.041 |
blush | +0.040 |
Neg logit contributions | |
rob | -0.034 |
Hipp | -0.031 |
zag | -0.031 |
Hero | -0.029 |
Zig | -0.028 |
Token | saw |
Token ID | 2497 |
Feature activation | -9.32e-02 |
Pos logit contributions | |
vel | +0.036 |
makeup | +0.035 |
prise | +0.032 |
blush | +0.032 |
goodness | +0.032 |
Neg logit contributions | |
rob | -0.039 |
Hipp | -0.037 |
Hero | -0.037 |
zag | -0.037 |
Zig | -0.035 |
Token | Tom |
Token ID | 4186 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
vel | +0.148 |
face | +0.132 |
makeup | +0.132 |
spit | +0.129 |
prise | +0.129 |
Neg logit contributions | |
rob | -0.177 |
Hipp | -0.165 |
Hero | -0.164 |
Zig | -0.158 |
zag | -0.154 |
Token | 's |
Token ID | 338 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
prise | +0.024 |
P | +0.023 |
illy | +0.023 |
makeup | +0.022 |
goodness | +0.021 |
Neg logit contributions | |
zag | -0.022 |
ousel | -0.021 |
Hero | -0.020 |
Zig | -0.019 |
girls | -0.018 |
Token | face |
Token ID | 1986 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
prise | +0.032 |
vel | +0.032 |
illy | +0.027 |
Her | +0.027 |
She | +0.027 |
Neg logit contributions | |
zag | -0.053 |
steep | -0.051 |
fault | -0.049 |
Hero | -0.049 |
rob | -0.048 |
Token | and |
Token ID | 290 |
Feature activation | -3.68e-03 |
Pos logit contributions | |
prise | +0.476 |
vel | +0.399 |
illy | +0.391 |
goodness | +0.388 |
makeup | +0.374 |
Neg logit contributions | |
zag | -0.408 |
steep | -0.346 |
ahs | -0.340 |
ousel | -0.336 |
ller | -0.335 |
Token | knew |
Token ID | 2993 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
vel | +0.007 |
prise | +0.007 |
makeup | +0.006 |
illy | +0.006 |
ucky | +0.006 |
Neg logit contributions | |
rob | -0.005 |
Zig | -0.005 |
Hipp | -0.005 |
Hero | -0.005 |
tigers | -0.005 |
Token | what |
Token ID | 644 |
Feature activation | -5.75e-02 |
Pos logit contributions | |
makeup | +0.113 |
face | +0.109 |
vel | +0.109 |
spit | +0.107 |
her | +0.100 |
Neg logit contributions | |
rob | -0.257 |
zag | -0.243 |
Hipp | -0.237 |
girls | -0.229 |
Hero | -0.227 |
Token | had |
Token ID | 550 |
Feature activation | -6.12e-03 |
Pos logit contributions | |
makeup | +0.084 |
vel | +0.084 |
face | +0.083 |
spit | +0.082 |
blush | +0.077 |
Neg logit contributions | |
rob | -0.111 |
zag | -0.103 |
Hipp | -0.102 |
Hero | -0.096 |
girls | -0.096 |
Token | happened |
Token ID | 3022 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
vel | +0.011 |
spit | +0.011 |
face | +0.011 |
makeup | +0.011 |
blush | +0.011 |
Neg logit contributions | |
rob | -0.010 |
zag | -0.009 |
Hipp | -0.008 |
Hero | -0.008 |
Zig | -0.008 |
Token | feels |
Token ID | 5300 |
Feature activation | -5.97e-03 |
Pos logit contributions | |
face | +0.077 |
spit | +0.076 |
her | +0.076 |
mirror | +0.076 |
cut | +0.074 |
Neg logit contributions | |
Hipp | -0.104 |
rob | -0.098 |
Claw | -0.098 |
zag | -0.097 |
Whale | -0.096 |
Token | the |
Token ID | 262 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
vel | +0.007 |
makeup | +0.007 |
spit | +0.007 |
face | +0.007 |
prise | +0.006 |
Neg logit contributions | |
rob | -0.013 |
Hipp | -0.012 |
Hero | -0.012 |
steep | -0.012 |
Zig | -0.012 |
Token | wind |
Token ID | 2344 |
Feature activation | -5.67e-02 |
Pos logit contributions | |
rob | +0.056 |
steep | +0.052 |
zag | +0.052 |
Hipp | +0.051 |
Hero | +0.050 |
Neg logit contributions | |
vel | -0.045 |
prise | -0.041 |
spit | -0.040 |
makeup | -0.040 |
cloth | -0.038 |
Token | in |
Token ID | 287 |
Feature activation | +3.22e-03 |
Pos logit contributions | |
makeup | +0.067 |
vel | +0.066 |
spit | +0.065 |
prise | +0.063 |
face | +0.063 |
Neg logit contributions | |
rob | -0.124 |
zag | -0.122 |
Hipp | -0.116 |
Hero | -0.113 |
aboard | -0.113 |
Token | his |
Token ID | 465 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
rob | +0.005 |
Hipp | +0.005 |
Hero | +0.005 |
Zig | +0.005 |
zag | +0.005 |
Neg logit contributions | |
vel | -0.006 |
prise | -0.005 |
makeup | -0.005 |
spit | -0.005 |
face | -0.005 |
Token | face |
Token ID | 1986 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
vel | +0.035 |
prise | +0.032 |
spit | +0.030 |
makeup | +0.028 |
illy | +0.028 |
Neg logit contributions | |
rob | -0.058 |
zag | -0.057 |
Hero | -0.054 |
Zig | -0.053 |
steep | -0.053 |
Token | and |
Token ID | 290 |
Feature activation | -4.19e-03 |
Pos logit contributions | |
prise | +0.334 |
vel | +0.327 |
makeup | +0.324 |
spit | +0.316 |
face | +0.297 |
Neg logit contributions | |
zag | -0.489 |
rob | -0.485 |
Hipp | -0.451 |
steep | -0.446 |
Hero | -0.445 |
Token | the |
Token ID | 262 |
Feature activation | +2.84e-02 |
Pos logit contributions | |
face | +0.005 |
makeup | +0.005 |
spit | +0.005 |
her | +0.005 |
blush | +0.005 |
Neg logit contributions | |
rob | -0.009 |
zag | -0.009 |
Hipp | -0.009 |
girls | -0.008 |
Hero | -0.008 |
Token | pedals |
Token ID | 42563 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
rob | +0.049 |
steep | +0.046 |
Hipp | +0.046 |
Hero | +0.046 |
zag | +0.045 |
Neg logit contributions | |
vel | -0.048 |
prise | -0.044 |
spit | -0.042 |
makeup | -0.041 |
cloth | -0.040 |
Token | under |
Token ID | 739 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
rob | +0.251 |
zag | +0.236 |
Hipp | +0.234 |
Hero | +0.225 |
steep | +0.223 |
Neg logit contributions | |
vel | -0.125 |
makeup | -0.124 |
face | -0.121 |
spit | -0.120 |
blush | -0.111 |
Token | his |
Token ID | 465 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
Hero | +0.046 |
rob | +0.046 |
Hipp | +0.045 |
Zig | +0.045 |
Zac | +0.043 |
Neg logit contributions | |
vel | -0.067 |
prise | -0.066 |
makeup | -0.059 |
spit | -0.058 |
blush | -0.056 |
Token |
|
Token ID | 198 |
Feature activation | -5.28e-03 |
Pos logit contributions | |
vel | +0.008 |
makeup | +0.008 |
face | +0.008 |
spit | +0.008 |
blush | +0.007 |
Neg logit contributions | |
rob | -0.010 |
zag | -0.009 |
Hipp | -0.009 |
Hero | -0.009 |
Zig | -0.009 |
Token | Her |
Token ID | 9360 |
Feature activation | -4.09e-02 |
Pos logit contributions | |
spit | +0.006 |
makeup | +0.006 |
face | +0.006 |
vel | +0.006 |
dust | +0.006 |
Neg logit contributions | |
rob | -0.011 |
zag | -0.010 |
Hero | -0.010 |
steep | -0.010 |
Skip | -0.010 |
Token | Mom |
Token ID | 11254 |
Feature activation | -8.34e-02 |
Pos logit contributions | |
vel | +0.057 |
She | +0.050 |
makeup | +0.050 |
spit | +0.048 |
prise | +0.048 |
Neg logit contributions | |
rob | -0.077 |
zag | -0.077 |
Hipp | -0.076 |
Hero | -0.072 |
steep | -0.071 |
Token | saw |
Token ID | 2497 |
Feature activation | -9.41e-02 |
Pos logit contributions | |
makeup | +0.132 |
vel | +0.130 |
face | +0.123 |
spit | +0.123 |
prise | +0.121 |
Neg logit contributions | |
rob | -0.137 |
zag | -0.137 |
Hipp | -0.135 |
Hero | -0.130 |
girls | -0.126 |
Token | her |
Token ID | 607 |
Feature activation | -7.48e-02 |
Pos logit contributions | |
vel | +0.126 |
spit | +0.124 |
prise | +0.121 |
face | +0.116 |
toe | +0.110 |
Neg logit contributions | |
rob | -0.170 |
Hero | -0.160 |
aboard | -0.156 |
Skip | -0.153 |
Claw | -0.147 |
Token | face |
Token ID | 1986 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
vel | +0.090 |
prise | +0.090 |
She | +0.087 |
P | +0.087 |
Her | +0.084 |
Neg logit contributions | |
zag | -0.143 |
Hero | -0.133 |
Hipp | -0.133 |
steep | -0.131 |
rob | -0.131 |
Token | and |
Token ID | 290 |
Feature activation | -3.62e-03 |
Pos logit contributions | |
prise | +0.372 |
vel | +0.347 |
makeup | +0.323 |
spit | +0.313 |
goodness | +0.312 |
Neg logit contributions | |
zag | -0.464 |
steep | -0.420 |
rob | -0.415 |
Hipp | -0.414 |
Hero | -0.401 |
Token | said |
Token ID | 531 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
vel | +0.006 |
makeup | +0.006 |
prise | +0.006 |
face | +0.005 |
spit | +0.005 |
Neg logit contributions | |
rob | -0.006 |
Hipp | -0.006 |
zag | -0.005 |
Hero | -0.005 |
Zig | -0.005 |
Token | " |
Token ID | 366 |
Feature activation | -1.78e-03 |
Pos logit contributions | |
vel | +0.017 |
makeup | +0.017 |
face | +0.017 |
spit | +0.017 |
blush | +0.015 |
Neg logit contributions | |
rob | -0.034 |
Hipp | -0.032 |
zag | -0.032 |
Hero | -0.031 |
Zig | -0.031 |
Token | Let |
Token ID | 5756 |
Feature activation | -5.60e-02 |
Pos logit contributions | |
face | +0.002 |
vel | +0.002 |
makeup | +0.002 |
spit | +0.002 |
prise | +0.002 |
Neg logit contributions | |
rob | -0.004 |
zag | -0.003 |
Hipp | -0.003 |
Hero | -0.003 |
steep | -0.003 |
Token | me |
Token ID | 502 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
vel | +0.074 |
makeup | +0.070 |
spit | +0.070 |
blush | +0.069 |
prise | +0.068 |
Neg logit contributions | |
rob | -0.101 |
zag | -0.098 |
Hero | -0.097 |
Hipp | -0.096 |
Skip | -0.095 |
Token | It |
Token ID | 632 |
Feature activation | -6.20e-02 |
Pos logit contributions | |
face | +0.042 |
spit | +0.041 |
makeup | +0.041 |
vel | +0.040 |
blush | +0.038 |
Neg logit contributions | |
rob | -0.051 |
zag | -0.046 |
Hipp | -0.045 |
girls | -0.043 |
Hero | -0.043 |
Token | hits |
Token ID | 7127 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
spit | +0.091 |
face | +0.086 |
dust | +0.081 |
spill | +0.076 |
cut | +0.076 |
Neg logit contributions | |
rob | -0.139 |
Hipp | -0.118 |
zag | -0.117 |
Hero | -0.116 |
lot | -0.113 |
Token | Tom |
Token ID | 4186 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
rob | +0.282 |
Hipp | +0.268 |
Zig | +0.262 |
zag | +0.262 |
Hero | +0.260 |
Neg logit contributions | |
vel | -0.209 |
makeup | -0.196 |
spit | -0.194 |
prise | -0.192 |
face | -0.190 |
Token | in |
Token ID | 287 |
Feature activation | +7.50e-03 |
Pos logit contributions | |
makeup | +0.020 |
prise | +0.020 |
illy | +0.019 |
blush | +0.018 |
spit | +0.018 |
Neg logit contributions | |
zag | -0.023 |
Zig | -0.019 |
steep | -0.019 |
aboard | -0.019 |
ousel | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
Hipp | +0.010 |
Hero | +0.010 |
steep | +0.010 |
Zig | +0.010 |
rob | +0.009 |
Neg logit contributions | |
vel | -0.015 |
prise | -0.014 |
blush | -0.013 |
makeup | -0.013 |
illy | -0.013 |
Token | face |
Token ID | 1986 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
rob | +0.065 |
steep | +0.065 |
Hipp | +0.063 |
zag | +0.062 |
Zig | +0.058 |
Neg logit contributions | |
vel | -0.052 |
prise | -0.043 |
makeup | -0.042 |
blush | -0.041 |
illy | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
prise | +0.398 |
makeup | +0.394 |
vel | +0.359 |
blush | +0.346 |
spit | +0.345 |
Neg logit contributions | |
zag | -0.437 |
steep | -0.397 |
Hipp | -0.391 |
Zac | -0.371 |
Zig | -0.371 |
Token | He |
Token ID | 679 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
vel | +0.049 |
makeup | +0.049 |
face | +0.048 |
spit | +0.047 |
blush | +0.046 |
Neg logit contributions | |
rob | -0.046 |
zag | -0.042 |
Hipp | -0.042 |
Hero | -0.040 |
Zig | -0.039 |
Token | drops |
Token ID | 10532 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
spit | +0.034 |
face | +0.034 |
makeup | +0.034 |
vel | +0.033 |
blush | +0.032 |
Neg logit contributions | |
rob | -0.042 |
zag | -0.039 |
Hipp | -0.038 |
Hero | -0.036 |
Zig | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | +3.37e-02 |
Pos logit contributions | |
vel | +0.113 |
prise | +0.102 |
makeup | +0.098 |
ivering | +0.095 |
blush | +0.094 |
Neg logit contributions | |
Zig | -0.064 |
steep | -0.062 |
Hipp | -0.062 |
aboard | -0.062 |
Hero | -0.061 |
Token | shovel |
Token ID | 33677 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
steep | +0.057 |
zag | +0.057 |
rob | +0.056 |
Hipp | +0.056 |
aboard | +0.052 |
Neg logit contributions | |
vel | -0.057 |
Her | -0.050 |
prise | -0.049 |
makeup | -0.048 |
blush | -0.047 |
Token | She |
Token ID | 1375 |
Feature activation | -5.43e-02 |
Pos logit contributions | |
vel | +0.058 |
prise | +0.055 |
spit | +0.052 |
makeup | +0.052 |
aming | +0.050 |
Neg logit contributions | |
Hipp | -0.037 |
rob | -0.036 |
Zig | -0.035 |
Hero | -0.034 |
zag | -0.032 |
Token | had |
Token ID | 550 |
Feature activation | -9.46e-03 |
Pos logit contributions | |
vel | +0.096 |
makeup | +0.087 |
prise | +0.084 |
goodness | +0.081 |
blush | +0.080 |
Neg logit contributions | |
rob | -0.090 |
Hipp | -0.086 |
Zig | -0.083 |
Hero | -0.082 |
zag | -0.081 |
Token | shampoo |
Token ID | 47259 |
Feature activation | +7.19e-02 |
Pos logit contributions | |
vel | +0.013 |
prise | +0.011 |
goodness | +0.011 |
ime | +0.011 |
face | +0.011 |
Neg logit contributions | |
rob | -0.018 |
steep | -0.017 |
Hipp | -0.017 |
Zig | -0.017 |
Hero | -0.016 |
Token | on |
Token ID | 319 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
rob | +0.148 |
zag | +0.142 |
Hipp | +0.132 |
girls | +0.131 |
steep | +0.130 |
Neg logit contributions | |
vel | -0.093 |
prise | -0.090 |
spit | -0.085 |
face | -0.084 |
makeup | -0.082 |
Token | her |
Token ID | 607 |
Feature activation | -7.03e-02 |
Pos logit contributions | |
rob | +0.069 |
Hipp | +0.066 |
Zig | +0.066 |
zag | +0.065 |
Hero | +0.064 |
Neg logit contributions | |
prise | -0.047 |
vel | -0.047 |
spit | -0.041 |
face | -0.038 |
aming | -0.038 |
Token | face |
Token ID | 1986 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
vel | +0.080 |
prise | +0.078 |
makeup | +0.068 |
spit | +0.067 |
She | +0.064 |
Neg logit contributions | |
zag | -0.152 |
rob | -0.150 |
Hipp | -0.143 |
girls | -0.142 |
Hero | -0.140 |
Token | . |
Token ID | 13 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
vel | +0.310 |
prise | +0.304 |
makeup | +0.294 |
spit | +0.292 |
face | +0.284 |
Neg logit contributions | |
zag | -0.500 |
rob | -0.497 |
Hipp | -0.476 |
steep | -0.469 |
Hero | -0.458 |
Token | She |
Token ID | 1375 |
Feature activation | -5.49e-02 |
Pos logit contributions | |
vel | +0.050 |
prise | +0.046 |
spit | +0.044 |
makeup | +0.043 |
face | +0.042 |
Neg logit contributions | |
Hipp | -0.047 |
rob | -0.046 |
Zig | -0.044 |
Hero | -0.044 |
zag | -0.042 |
Token | was |
Token ID | 373 |
Feature activation | +9.18e-03 |
Pos logit contributions | |
vel | +0.096 |
makeup | +0.086 |
prise | +0.084 |
goodness | +0.080 |
face | +0.079 |
Neg logit contributions | |
rob | -0.093 |
Hipp | -0.086 |
Hero | -0.083 |
zag | -0.082 |
Zig | -0.082 |
Token | crying |
Token ID | 13774 |
Feature activation | -5.71e-02 |
Pos logit contributions | |
Hipp | +0.017 |
steep | +0.017 |
rob | +0.016 |
zag | +0.016 |
Zig | +0.016 |
Neg logit contributions | |
vel | -0.012 |
prise | -0.011 |
file | -0.011 |
goodness | -0.011 |
makeup | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
prise | +0.105 |
makeup | +0.097 |
vel | +0.093 |
blush | +0.092 |
lipstick | +0.089 |
Neg logit contributions | |
steep | -0.082 |
rob | -0.077 |
Hipp | -0.076 |
Hero | -0.076 |
zag | -0.073 |
Token | . |
Token ID | 13 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
vel | +0.024 |
makeup | +0.024 |
face | +0.023 |
spit | +0.023 |
blush | +0.021 |
Neg logit contributions | |
rob | -0.063 |
zag | -0.059 |
Hipp | -0.059 |
Hero | -0.056 |
girls | -0.056 |
Token | She |
Token ID | 1375 |
Feature activation | -5.20e-02 |
Pos logit contributions | |
vel | +0.027 |
makeup | +0.026 |
face | +0.025 |
spit | +0.025 |
prise | +0.024 |
Neg logit contributions | |
rob | -0.042 |
Hipp | -0.041 |
zag | -0.040 |
Hero | -0.040 |
Zig | -0.039 |
Token | had |
Token ID | 550 |
Feature activation | -5.70e-03 |
Pos logit contributions | |
vel | +0.079 |
makeup | +0.076 |
prise | +0.072 |
face | +0.072 |
blush | +0.069 |
Neg logit contributions | |
rob | -0.095 |
Hipp | -0.088 |
zag | -0.087 |
Hero | -0.085 |
steep | -0.084 |
Token | a |
Token ID | 257 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
makeup | +0.006 |
vel | +0.006 |
face | +0.006 |
spit | +0.006 |
blush | +0.006 |
Neg logit contributions | |
rob | -0.013 |
zag | -0.012 |
Hipp | -0.012 |
Hero | -0.011 |
girls | -0.011 |
Token | red |
Token ID | 2266 |
Feature activation | -9.07e-03 |
Pos logit contributions | |
vel | +0.032 |
spit | +0.029 |
makeup | +0.028 |
face | +0.028 |
blush | +0.027 |
Neg logit contributions | |
rob | -0.060 |
zag | -0.059 |
Hipp | -0.057 |
steep | -0.056 |
Hero | -0.055 |
Token | face |
Token ID | 1986 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
vel | +0.012 |
spit | +0.011 |
makeup | +0.011 |
prise | +0.011 |
blush | +0.010 |
Neg logit contributions | |
rob | -0.018 |
zag | -0.018 |
Hipp | -0.018 |
Hero | -0.017 |
Zig | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | -2.79e-03 |
Pos logit contributions | |
vel | +0.311 |
prise | +0.308 |
makeup | +0.290 |
spit | +0.286 |
face | +0.281 |
Neg logit contributions | |
zag | -0.511 |
rob | -0.503 |
Hipp | -0.478 |
steep | -0.460 |
Hero | -0.458 |
Token | a |
Token ID | 257 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
vel | +0.003 |
prise | +0.003 |
makeup | +0.003 |
face | +0.003 |
spit | +0.003 |
Neg logit contributions | |
rob | -0.006 |
Hipp | -0.006 |
zag | -0.006 |
Hero | -0.006 |
steep | -0.006 |
Token | hot |
Token ID | 3024 |
Feature activation | -6.09e-02 |
Pos logit contributions | |
vel | +0.039 |
makeup | +0.034 |
prise | +0.034 |
spit | +0.034 |
ime | +0.033 |
Neg logit contributions | |
zag | -0.053 |
rob | -0.053 |
steep | -0.052 |
Hipp | -0.051 |
Hero | -0.049 |
Token | head |
Token ID | 1182 |
Feature activation | -8.91e-02 |
Pos logit contributions | |
vel | +0.085 |
makeup | +0.076 |
spit | +0.075 |
prise | +0.075 |
She | +0.073 |
Neg logit contributions | |
rob | -0.123 |
zag | -0.117 |
Hipp | -0.116 |
Hero | -0.108 |
steep | -0.107 |
Token | . |
Token ID | 13 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
vel | +0.097 |
makeup | +0.094 |
spit | +0.090 |
face | +0.090 |
prise | +0.089 |
Neg logit contributions | |
rob | -0.198 |
zag | -0.193 |
Hipp | -0.188 |
Hero | -0.179 |
steep | -0.178 |
Token | a |
Token ID | 257 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
face | +0.001 |
makeup | +0.001 |
vel | +0.001 |
her | +0.001 |
spit | +0.001 |
Neg logit contributions | |
rob | -0.003 |
zag | -0.003 |
Hipp | -0.003 |
girls | -0.003 |
steep | -0.003 |
Token | big |
Token ID | 1263 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
makeup | +0.034 |
face | +0.034 |
vel | +0.033 |
spit | +0.032 |
prise | +0.030 |
Neg logit contributions | |
rob | -0.072 |
Hipp | -0.067 |
zag | -0.067 |
girls | -0.065 |
Hero | -0.063 |
Token | smile |
Token ID | 8212 |
Feature activation | -5.31e-02 |
Pos logit contributions | |
vel | +0.122 |
makeup | +0.121 |
face | +0.119 |
spit | +0.117 |
blush | +0.109 |
Neg logit contributions | |
rob | -0.239 |
zag | -0.224 |
Hipp | -0.223 |
Hero | -0.214 |
girls | -0.211 |
Token | on |
Token ID | 319 |
Feature activation | +3.43e-02 |
Pos logit contributions | |
vel | +0.050 |
face | +0.049 |
makeup | +0.049 |
spit | +0.046 |
blush | +0.044 |
Neg logit contributions | |
rob | -0.130 |
Hipp | -0.122 |
zag | -0.120 |
Hero | -0.116 |
girls | -0.116 |
Token | his |
Token ID | 465 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
rob | +0.062 |
Hipp | +0.059 |
zag | +0.057 |
Hero | +0.056 |
Zig | +0.055 |
Neg logit contributions | |
vel | -0.052 |
makeup | -0.052 |
spit | -0.050 |
prise | -0.049 |
face | -0.048 |
Token | face |
Token ID | 1986 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
vel | +0.010 |
makeup | +0.009 |
prise | +0.008 |
spit | +0.008 |
blush | +0.006 |
Neg logit contributions | |
rob | -0.086 |
zag | -0.085 |
Hipp | -0.083 |
Hero | -0.082 |
Zig | -0.081 |
Token | . |
Token ID | 13 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
vel | +0.226 |
makeup | +0.223 |
face | +0.222 |
spit | +0.213 |
blush | +0.196 |
Neg logit contributions | |
rob | -0.625 |
zag | -0.583 |
Hipp | -0.583 |
Hero | -0.561 |
girls | -0.556 |
Token | He |
Token ID | 679 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
face | +0.029 |
her | +0.026 |
She | +0.025 |
makeup | +0.024 |
hand | +0.024 |
Neg logit contributions | |
rob | -0.054 |
girls | -0.048 |
zag | -0.047 |
Hipp | -0.044 |
steep | -0.044 |
Token | waved |
Token ID | 26834 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
face | +0.020 |
spit | +0.019 |
cut | +0.017 |
her | +0.017 |
dress | +0.015 |
Neg logit contributions | |
rob | -0.052 |
Hipp | -0.048 |
girls | -0.048 |
zag | -0.047 |
Claw | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | -3.89e-03 |
Pos logit contributions | |
rob | +0.170 |
zag | +0.153 |
Hipp | +0.148 |
girls | +0.141 |
steep | +0.136 |
Neg logit contributions | |
face | -0.193 |
makeup | -0.187 |
vel | -0.187 |
spit | -0.183 |
her | -0.177 |
Token | waved |
Token ID | 26834 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
face | +0.005 |
her | +0.005 |
spit | +0.004 |
makeup | +0.004 |
vel | +0.004 |
Neg logit contributions | |
rob | -0.009 |
zag | -0.008 |
Hipp | -0.008 |
girls | -0.008 |
Hero | -0.008 |
Token | feel |
Token ID | 1254 |
Feature activation | -8.80e-02 |
Pos logit contributions | |
face | +0.001 |
clean | +0.001 |
spit | +0.001 |
makeup | +0.001 |
wipe | +0.001 |
Neg logit contributions | |
rob | -0.004 |
Hipp | -0.004 |
zag | -0.004 |
girls | -0.004 |
Hero | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | +2.41e-02 |
Pos logit contributions | |
vel | +0.113 |
makeup | +0.112 |
face | +0.109 |
spit | +0.109 |
blush | +0.103 |
Neg logit contributions | |
rob | -0.182 |
Hipp | -0.172 |
zag | -0.169 |
Hero | -0.164 |
steep | -0.162 |
Token | wind |
Token ID | 2344 |
Feature activation | -5.70e-02 |
Pos logit contributions | |
rob | +0.051 |
Hipp | +0.047 |
zag | +0.047 |
steep | +0.046 |
Hero | +0.045 |
Neg logit contributions | |
vel | -0.032 |
makeup | -0.030 |
spit | -0.029 |
prise | -0.029 |
face | -0.028 |
Token | on |
Token ID | 319 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
makeup | +0.061 |
vel | +0.061 |
face | +0.058 |
spit | +0.058 |
prise | +0.056 |
Neg logit contributions | |
rob | -0.130 |
zag | -0.125 |
Hipp | -0.122 |
Hero | -0.117 |
steep | -0.116 |
Token | his |
Token ID | 465 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
Hipp | +0.046 |
rob | +0.045 |
Zig | +0.044 |
Hero | +0.043 |
zag | +0.042 |
Neg logit contributions | |
prise | -0.070 |
vel | -0.066 |
makeup | -0.062 |
spit | -0.061 |
blush | -0.060 |
Token | face |
Token ID | 1986 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
vel | +0.022 |
prise | +0.020 |
makeup | +0.020 |
spit | +0.019 |
blush | +0.017 |
Neg logit contributions | |
rob | -0.079 |
zag | -0.076 |
Hipp | -0.073 |
Hero | -0.073 |
steep | -0.073 |
Token | . |
Token ID | 13 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
makeup | +0.278 |
vel | +0.277 |
spit | +0.270 |
prise | +0.268 |
face | +0.262 |
Neg logit contributions | |
rob | -0.543 |
zag | -0.536 |
Hipp | -0.518 |
steep | -0.502 |
Hero | -0.500 |
Token |
|
Token ID | 198 |
Feature activation | -4.68e-03 |
Pos logit contributions | |
face | +0.035 |
makeup | +0.035 |
spit | +0.034 |
vel | +0.033 |
She | +0.032 |
Neg logit contributions | |
rob | -0.045 |
zag | -0.042 |
Hipp | -0.039 |
Hero | -0.039 |
girls | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | -5.03e-03 |
Pos logit contributions | |
vel | +0.007 |
makeup | +0.006 |
face | +0.006 |
spit | +0.006 |
blush | +0.006 |
Neg logit contributions | |
rob | -0.009 |
zag | -0.008 |
Hipp | -0.008 |
Hero | -0.008 |
girls | -0.008 |
Token | One |
Token ID | 3198 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
makeup | +0.010 |
face | +0.009 |
vel | +0.009 |
blush | +0.009 |
spit | +0.009 |
Neg logit contributions | |
rob | -0.007 |
Hipp | -0.007 |
zag | -0.007 |
steep | -0.007 |
Zig | -0.006 |
Token | day |
Token ID | 1110 |
Feature activation | -7.73e-03 |
Pos logit contributions | |
rob | +0.213 |
zag | +0.199 |
Hipp | +0.195 |
steep | +0.185 |
Hero | +0.183 |
Neg logit contributions | |
vel | -0.229 |
makeup | -0.226 |
spit | -0.221 |
face | -0.220 |
blush | -0.212 |
Token | a |
Token ID | 257 |
Feature activation | -3.07e-02 |
Pos logit contributions | |
face | +0.002 |
makeup | +0.002 |
spit | +0.002 |
vel | +0.002 |
blush | +0.001 |
Neg logit contributions | |
rob | -0.003 |
Hipp | -0.003 |
zag | -0.003 |
girls | -0.003 |
Hero | -0.003 |
Token | big |
Token ID | 1263 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
vel | +0.038 |
makeup | +0.037 |
spit | +0.036 |
face | +0.035 |
blush | +0.034 |
Neg logit contributions | |
rob | -0.064 |
zag | -0.062 |
Hipp | -0.061 |
steep | -0.059 |
Hero | -0.058 |
Token | smile |
Token ID | 8212 |
Feature activation | -5.31e-02 |
Pos logit contributions | |
vel | +0.123 |
makeup | +0.121 |
spit | +0.118 |
face | +0.116 |
blush | +0.110 |
Neg logit contributions | |
rob | -0.231 |
zag | -0.218 |
Hipp | -0.216 |
Hero | -0.208 |
steep | -0.208 |
Token | on |
Token ID | 319 |
Feature activation | +3.43e-02 |
Pos logit contributions | |
makeup | +0.063 |
vel | +0.062 |
spit | +0.062 |
prise | +0.061 |
face | +0.058 |
Neg logit contributions | |
rob | -0.111 |
zag | -0.110 |
Hipp | -0.104 |
Hero | -0.101 |
steep | -0.100 |
Token | his |
Token ID | 465 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
Hipp | +0.051 |
rob | +0.050 |
Hero | +0.050 |
Zig | +0.048 |
zag | +0.047 |
Neg logit contributions | |
prise | -0.063 |
vel | -0.059 |
makeup | -0.058 |
blush | -0.054 |
spit | -0.053 |
Token | face |
Token ID | 1986 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
prise | +0.016 |
makeup | +0.015 |
vel | +0.013 |
She | +0.012 |
blush | +0.011 |
Neg logit contributions | |
zag | -0.079 |
rob | -0.076 |
Hero | -0.076 |
Hipp | -0.075 |
Zig | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
vel | +0.282 |
makeup | +0.280 |
spit | +0.275 |
prise | +0.272 |
face | +0.264 |
Neg logit contributions | |
rob | -0.535 |
zag | -0.527 |
Hipp | -0.512 |
Hero | -0.491 |
steep | -0.489 |
Token | He |
Token ID | 679 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
face | +0.032 |
vel | +0.032 |
makeup | +0.032 |
spit | +0.031 |
blush | +0.029 |
Neg logit contributions | |
rob | -0.037 |
zag | -0.034 |
Hipp | -0.033 |
girls | -0.032 |
steep | -0.031 |
Token | remembered |
Token ID | 12086 |
Feature activation | +2.77e-03 |
Pos logit contributions | |
spit | +0.022 |
face | +0.022 |
vel | +0.020 |
makeup | +0.019 |
cut | +0.019 |
Neg logit contributions | |
rob | -0.050 |
zag | -0.045 |
Hipp | -0.045 |
girls | -0.044 |
steep | -0.043 |
Token | his |
Token ID | 465 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
rob | +0.006 |
zag | +0.006 |
Hipp | +0.006 |
Hero | +0.005 |
steep | +0.005 |
Neg logit contributions | |
vel | -0.003 |
makeup | -0.003 |
face | -0.003 |
spit | -0.003 |
blush | -0.003 |
Token | dream |
Token ID | 4320 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
vel | +0.038 |
face | +0.038 |
makeup | +0.038 |
spit | +0.037 |
blush | +0.034 |
Neg logit contributions | |
rob | -0.062 |
Hipp | -0.057 |
zag | -0.057 |
Hero | -0.054 |
girls | -0.054 |
Token | at |
Token ID | 379 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
face | +0.017 |
makeup | +0.016 |
vel | +0.016 |
spit | +0.015 |
blush | +0.014 |
Neg logit contributions | |
rob | -0.064 |
zag | -0.061 |
Hipp | -0.060 |
girls | -0.058 |
Zig | -0.058 |
Token | Lily |
Token ID | 20037 |
Feature activation | -5.34e-02 |
Pos logit contributions | |
makeup | +0.023 |
face | +0.023 |
vel | +0.023 |
spit | +0.022 |
blush | +0.021 |
Neg logit contributions | |
rob | -0.050 |
zag | -0.047 |
Hipp | -0.047 |
Hero | -0.045 |
steep | -0.045 |
Token | with |
Token ID | 351 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
prise | +0.077 |
makeup | +0.074 |
vel | +0.072 |
spit | +0.071 |
blush | +0.069 |
Neg logit contributions | |
zag | -0.097 |
rob | -0.092 |
Hero | -0.092 |
Hipp | -0.090 |
ousel | -0.088 |
Token | a |
Token ID | 257 |
Feature activation | -3.09e-02 |
Pos logit contributions | |
vel | +0.024 |
prise | +0.020 |
makeup | +0.019 |
illy | +0.019 |
ime | +0.018 |
Neg logit contributions | |
rob | -0.028 |
Hero | -0.028 |
Hipp | -0.026 |
steep | -0.026 |
Skip | -0.026 |
Token | proud |
Token ID | 6613 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
vel | +0.039 |
makeup | +0.037 |
spit | +0.035 |
prise | +0.034 |
face | +0.033 |
Neg logit contributions | |
rob | -0.063 |
zag | -0.061 |
steep | -0.060 |
Hipp | -0.058 |
Hero | -0.057 |
Token | face |
Token ID | 1986 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
zag | +0.027 |
rob | +0.027 |
steep | +0.026 |
Hipp | +0.026 |
Hero | +0.025 |
Neg logit contributions | |
vel | -0.019 |
pped | -0.017 |
prise | -0.017 |
makeup | -0.017 |
She | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
vel | +0.249 |
makeup | +0.247 |
face | +0.240 |
spit | +0.238 |
blush | +0.218 |
Neg logit contributions | |
rob | -0.593 |
zag | -0.561 |
Hipp | -0.558 |
Hero | -0.536 |
girls | -0.530 |
Token | He |
Token ID | 679 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
vel | +0.048 |
makeup | +0.048 |
face | +0.048 |
spit | +0.047 |
blush | +0.045 |
Neg logit contributions | |
rob | -0.043 |
zag | -0.039 |
Hipp | -0.038 |
Hero | -0.036 |
girls | -0.036 |
Token | kissed |
Token ID | 27946 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
spit | +0.029 |
face | +0.028 |
makeup | +0.027 |
vel | +0.027 |
blush | +0.025 |
Neg logit contributions | |
rob | -0.057 |
zag | -0.054 |
Hipp | -0.053 |
girls | -0.051 |
Zig | -0.050 |
Token | her |
Token ID | 607 |
Feature activation | -7.43e-02 |
Pos logit contributions | |
Hero | +0.066 |
rob | +0.065 |
Zig | +0.064 |
Hipp | +0.064 |
Zac | +0.062 |
Neg logit contributions | |
vel | -0.030 |
prise | -0.027 |
spit | -0.024 |
ime | -0.022 |
goodness | -0.021 |
Token | on |
Token ID | 319 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
vel | +0.081 |
makeup | +0.080 |
spit | +0.077 |
face | +0.076 |
prise | +0.072 |
Neg logit contributions | |
rob | -0.169 |
zag | -0.161 |
Hipp | -0.159 |
Hero | -0.154 |
steep | -0.151 |
Token | let |
Token ID | 1309 |
Feature activation | -4.06e-02 |
Pos logit contributions | |
vel | +0.004 |
makeup | +0.004 |
face | +0.004 |
spit | +0.004 |
blush | +0.004 |
Neg logit contributions | |
rob | -0.006 |
zag | -0.005 |
Hipp | -0.005 |
Hero | -0.005 |
steep | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
makeup | +0.047 |
vel | +0.047 |
spit | +0.043 |
prise | +0.043 |
blush | +0.043 |
Neg logit contributions | |
rob | -0.089 |
Hipp | -0.085 |
zag | -0.084 |
Hero | -0.082 |
Zig | -0.082 |
Token | sunshine |
Token ID | 34488 |
Feature activation | -4.87e-02 |
Pos logit contributions | |
rob | +0.041 |
Hipp | +0.039 |
girls | +0.036 |
Hero | +0.036 |
Zig | +0.035 |
Neg logit contributions | |
vel | -0.043 |
makeup | -0.040 |
prise | -0.040 |
cloth | -0.038 |
Her | -0.038 |
Token | warm |
Token ID | 5814 |
Feature activation | -9.35e-02 |
Pos logit contributions | |
vel | +0.054 |
makeup | +0.051 |
prise | +0.048 |
spit | +0.046 |
lipstick | +0.044 |
Neg logit contributions | |
zag | -0.107 |
rob | -0.105 |
Hipp | -0.104 |
Zig | -0.101 |
girls | -0.100 |
Token | his |
Token ID | 465 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
vel | +0.110 |
makeup | +0.106 |
spit | +0.103 |
face | +0.102 |
prise | +0.098 |
Neg logit contributions | |
rob | -0.207 |
Hipp | -0.195 |
zag | -0.192 |
Hero | -0.185 |
Zig | -0.184 |
Token | face |
Token ID | 1986 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
vel | +0.039 |
makeup | +0.036 |
prise | +0.036 |
She | +0.035 |
blush | +0.034 |
Neg logit contributions | |
Hipp | -0.059 |
zag | -0.059 |
rob | -0.058 |
Hero | -0.056 |
girls | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
vel | +0.266 |
makeup | +0.263 |
spit | +0.251 |
face | +0.251 |
prise | +0.237 |
Neg logit contributions | |
rob | -0.566 |
zag | -0.547 |
Hipp | -0.537 |
steep | -0.516 |
Hero | -0.514 |
Token | It |
Token ID | 632 |
Feature activation | -6.63e-02 |
Pos logit contributions | |
face | +0.032 |
makeup | +0.030 |
spit | +0.029 |
vel | +0.028 |
her | +0.027 |
Neg logit contributions | |
rob | -0.048 |
zag | -0.042 |
girls | -0.040 |
Hipp | -0.038 |
steep | -0.038 |
Token | felt |
Token ID | 2936 |
Feature activation | -3.30e-02 |
Pos logit contributions | |
spit | +0.076 |
face | +0.073 |
vel | +0.067 |
dust | +0.066 |
her | +0.066 |
Neg logit contributions | |
rob | -0.155 |
zag | -0.143 |
Hipp | -0.143 |
girls | -0.142 |
Hero | -0.139 |
Token | really |
Token ID | 1107 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
vel | +0.041 |
makeup | +0.039 |
prise | +0.037 |
spit | +0.036 |
blush | +0.036 |
Neg logit contributions | |
Hipp | -0.067 |
rob | -0.066 |
steep | -0.065 |
Zig | -0.063 |
Hero | -0.062 |
Token | good |
Token ID | 922 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
vel | +0.033 |
makeup | +0.033 |
face | +0.032 |
spit | +0.031 |
blush | +0.029 |
Neg logit contributions | |
rob | -0.077 |
Hipp | -0.073 |
zag | -0.072 |
steep | -0.070 |
Hero | -0.069 |
Token | a |
Token ID | 257 |
Feature activation | -3.09e-02 |
Pos logit contributions | |
vel | +0.026 |
spit | +0.022 |
ivering | +0.021 |
makeup | +0.021 |
file | +0.021 |
Neg logit contributions | |
rob | -0.029 |
Hipp | -0.029 |
Hero | -0.029 |
Skip | -0.028 |
aboard | -0.027 |
Token | big |
Token ID | 1263 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
vel | +0.037 |
spit | +0.034 |
makeup | +0.033 |
face | +0.031 |
She | +0.031 |
Neg logit contributions | |
rob | -0.065 |
zag | -0.064 |
Hipp | -0.062 |
steep | -0.062 |
Hero | -0.060 |
Token | smile |
Token ID | 8212 |
Feature activation | -5.42e-02 |
Pos logit contributions | |
vel | +0.140 |
spit | +0.128 |
makeup | +0.127 |
She | +0.125 |
Her | +0.122 |
Neg logit contributions | |
rob | -0.218 |
zag | -0.208 |
Hipp | -0.207 |
steep | -0.205 |
Hero | -0.202 |
Token | on |
Token ID | 319 |
Feature activation | +3.33e-02 |
Pos logit contributions | |
makeup | +0.060 |
spit | +0.058 |
vel | +0.058 |
face | +0.055 |
prise | +0.054 |
Neg logit contributions | |
rob | -0.121 |
zag | -0.117 |
Hipp | -0.113 |
Hero | -0.109 |
Zig | -0.107 |
Token | his |
Token ID | 465 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
Hipp | +0.046 |
rob | +0.043 |
Hero | +0.042 |
Arnold | +0.042 |
Zac | +0.042 |
Neg logit contributions | |
prise | -0.061 |
vel | -0.059 |
makeup | -0.056 |
file | -0.054 |
spit | -0.054 |
Token | face |
Token ID | 1986 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
vel | +0.016 |
prise | +0.016 |
makeup | +0.014 |
She | +0.011 |
spit | +0.011 |
Neg logit contributions | |
zag | -0.084 |
rob | -0.082 |
Hipp | -0.081 |
Hero | -0.081 |
Zig | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
vel | +0.241 |
makeup | +0.240 |
face | +0.233 |
spit | +0.232 |
blush | +0.211 |
Neg logit contributions | |
rob | -0.599 |
zag | -0.568 |
Hipp | -0.564 |
Hero | -0.541 |
steep | -0.536 |
Token | What |
Token ID | 1867 |
Feature activation | -8.89e-02 |
Pos logit contributions | |
makeup | +0.038 |
face | +0.037 |
vel | +0.036 |
spit | +0.035 |
She | +0.035 |
Neg logit contributions | |
rob | -0.056 |
zag | -0.053 |
Hipp | -0.048 |
Hero | -0.048 |
girls | -0.048 |
Token | a |
Token ID | 257 |
Feature activation | -3.19e-02 |
Pos logit contributions | |
vel | +0.084 |
makeup | +0.084 |
face | +0.082 |
spit | +0.081 |
blush | +0.074 |
Neg logit contributions | |
rob | -0.215 |
zag | -0.203 |
Hipp | -0.202 |
Hero | -0.194 |
steep | -0.192 |
Token | great |
Token ID | 1049 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
vel | +0.046 |
makeup | +0.043 |
spit | +0.043 |
blush | +0.042 |
cloth | +0.041 |
Neg logit contributions | |
rob | -0.058 |
Hipp | -0.056 |
Hero | -0.054 |
steep | -0.054 |
zag | -0.053 |
Token | day |
Token ID | 1110 |
Feature activation | -9.78e-03 |
Pos logit contributions | |
vel | +0.021 |
makeup | +0.020 |
spit | +0.020 |
face | +0.019 |
blush | +0.019 |
Neg logit contributions | |
rob | -0.031 |
zag | -0.029 |
Hipp | -0.029 |
Hero | -0.028 |
steep | -0.027 |
Token | with |
Token ID | 351 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
face | +0.011 |
vel | +0.010 |
makeup | +0.010 |
spit | +0.010 |
her | +0.009 |
Neg logit contributions | |
rob | -0.029 |
Hipp | -0.027 |
zag | -0.027 |
girls | -0.026 |
Hero | -0.026 |
Token | a |
Token ID | 257 |
Feature activation | -3.24e-02 |
Pos logit contributions | |
vel | +0.020 |
spit | +0.018 |
face | +0.017 |
makeup | +0.016 |
blush | +0.016 |
Neg logit contributions | |
rob | -0.038 |
Hipp | -0.036 |
Hero | -0.035 |
Zig | -0.035 |
Skip | -0.034 |
Token | smile |
Token ID | 8212 |
Feature activation | -5.33e-02 |
Pos logit contributions | |
vel | +0.033 |
makeup | +0.032 |
spit | +0.031 |
face | +0.030 |
blush | +0.028 |
Neg logit contributions | |
rob | -0.076 |
zag | -0.072 |
Hipp | -0.072 |
steep | -0.069 |
Hero | -0.069 |
Token | on |
Token ID | 319 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
makeup | +0.060 |
vel | +0.059 |
spit | +0.059 |
face | +0.056 |
prise | +0.054 |
Neg logit contributions | |
rob | -0.118 |
zag | -0.114 |
Hipp | -0.112 |
Hero | -0.107 |
Zig | -0.105 |
Token | her |
Token ID | 607 |
Feature activation | -7.76e-02 |
Pos logit contributions | |
Hipp | +0.088 |
rob | +0.084 |
Zig | +0.083 |
Hero | +0.082 |
Skip | +0.081 |
Neg logit contributions | |
vel | -0.025 |
prise | -0.023 |
spit | -0.021 |
makeup | -0.020 |
file | -0.019 |
Token | face |
Token ID | 1986 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
vel | +0.016 |
makeup | +0.015 |
spit | +0.011 |
prise | +0.008 |
blush | +0.007 |
Neg logit contributions | |
rob | -0.242 |
zag | -0.234 |
Hipp | -0.233 |
Hero | -0.226 |
steep | -0.225 |
Token | . |
Token ID | 13 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
vel | +0.273 |
makeup | +0.271 |
spit | +0.269 |
face | +0.257 |
prise | +0.252 |
Neg logit contributions | |
rob | -0.553 |
zag | -0.539 |
Hipp | -0.530 |
Hero | -0.504 |
steep | -0.502 |
Token |
|
Token ID | 198 |
Feature activation | -5.76e-03 |
Pos logit contributions | |
vel | +0.031 |
makeup | +0.031 |
face | +0.030 |
spit | +0.030 |
blush | +0.028 |
Neg logit contributions | |
rob | -0.039 |
zag | -0.036 |
Hipp | -0.036 |
Hero | -0.034 |
Zig | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | -6.10e-03 |
Pos logit contributions | |
makeup | +0.008 |
vel | +0.008 |
spit | +0.007 |
face | +0.007 |
blush | +0.007 |
Neg logit contributions | |
rob | -0.012 |
zag | -0.011 |
Hipp | -0.011 |
Zig | -0.011 |
Hero | -0.010 |
Token | She |
Token ID | 3347 |
Feature activation | -7.37e-02 |
Pos logit contributions | |
makeup | +0.011 |
face | +0.010 |
blush | +0.010 |
spit | +0.010 |
lipstick | +0.010 |
Neg logit contributions | |
zag | -0.010 |
rob | -0.009 |
Hipp | -0.009 |
Zig | -0.009 |
Skip | -0.009 |
Token | had |
Token ID | 550 |
Feature activation | -3.82e-03 |
Pos logit contributions | |
face | +0.069 |
her | +0.065 |
spit | +0.063 |
hand | +0.058 |
clean | +0.058 |
Neg logit contributions | |
rob | -0.184 |
Hipp | -0.171 |
girls | -0.170 |
zag | -0.170 |
steep | -0.166 |
Token | car |
Token ID | 1097 |
Feature activation | +3.69e-02 |
Pos logit contributions | |
rob | +0.071 |
zag | +0.065 |
Hipp | +0.065 |
Hero | +0.061 |
girls | +0.061 |
Neg logit contributions | |
vel | -0.062 |
makeup | -0.062 |
face | -0.061 |
spit | -0.061 |
blush | -0.057 |
Token | to |
Token ID | 284 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
rob | +0.086 |
Hipp | +0.079 |
zag | +0.077 |
girls | +0.075 |
Hero | +0.074 |
Neg logit contributions | |
makeup | -0.041 |
face | -0.041 |
vel | -0.039 |
spit | -0.038 |
her | -0.036 |
Token | zoom |
Token ID | 19792 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
rob | +0.029 |
zag | +0.028 |
Hipp | +0.028 |
Zig | +0.026 |
Hero | +0.026 |
Neg logit contributions | |
vel | -0.013 |
makeup | -0.012 |
spit | -0.011 |
face | -0.011 |
blush | -0.011 |
Token | around |
Token ID | 1088 |
Feature activation | +7.43e-02 |
Pos logit contributions | |
rob | +0.279 |
Hipp | +0.263 |
zag | +0.253 |
girls | +0.253 |
field | +0.245 |
Neg logit contributions | |
face | -0.111 |
makeup | -0.108 |
her | -0.105 |
vel | -0.101 |
lipstick | -0.090 |
Token | even |
Token ID | 772 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
rob | +0.168 |
zag | +0.157 |
Hipp | +0.156 |
Hero | +0.150 |
girls | +0.148 |
Neg logit contributions | |
vel | -0.082 |
makeup | -0.082 |
face | -0.081 |
spit | -0.079 |
blush | -0.074 |
Token | faster |
Token ID | 5443 |
Feature activation | +2.66e-01 |
Pos logit contributions | |
rob | +0.216 |
Hipp | +0.197 |
zag | +0.196 |
Zig | +0.183 |
girls | +0.183 |
Neg logit contributions | |
face | -0.199 |
vel | -0.199 |
makeup | -0.199 |
spit | -0.195 |
blush | -0.184 |
Token | . |
Token ID | 13 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
rob | +0.668 |
Hipp | +0.630 |
zag | +0.629 |
Hero | +0.600 |
girls | +0.599 |
Neg logit contributions | |
face | -0.235 |
makeup | -0.231 |
vel | -0.227 |
spit | -0.214 |
her | -0.197 |
Token | His |
Token ID | 2399 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
face | +0.020 |
spit | +0.019 |
makeup | +0.018 |
vel | +0.018 |
She | +0.017 |
Neg logit contributions | |
rob | -0.024 |
zag | -0.020 |
girls | -0.019 |
Hero | -0.019 |
Hipp | -0.019 |
Token | mom |
Token ID | 1995 |
Feature activation | -4.02e-02 |
Pos logit contributions | |
rob | +0.041 |
Hipp | +0.037 |
zag | +0.037 |
Hero | +0.036 |
girls | +0.035 |
Neg logit contributions | |
face | -0.018 |
makeup | -0.018 |
spit | -0.017 |
vel | -0.017 |
blush | -0.015 |
Token | smiled |
Token ID | 13541 |
Feature activation | +7.58e-02 |
Pos logit contributions | |
vel | +0.057 |
makeup | +0.057 |
face | +0.056 |
spit | +0.055 |
prise | +0.052 |
Neg logit contributions | |
rob | -0.076 |
zag | -0.072 |
Hipp | -0.071 |
Hero | -0.068 |
girls | -0.066 |
Token | again |
Token ID | 757 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
rob | +0.152 |
zag | +0.151 |
Hipp | +0.150 |
Hero | +0.145 |
steep | +0.143 |
Neg logit contributions | |
makeup | -0.091 |
vel | -0.090 |
spit | -0.088 |
prise | -0.087 |
blush | -0.082 |
Token | snap |
Token ID | 11495 |
Feature activation | -4.00e-04 |
Pos logit contributions | |
zag | +0.076 |
rob | +0.073 |
steep | +0.071 |
Skip | +0.069 |
Hero | +0.069 |
Neg logit contributions | |
makeup | -0.042 |
vel | -0.040 |
prise | -0.040 |
spit | -0.038 |
face | -0.035 |
Token | , |
Token ID | 11 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
vel | +0.000 |
spit | +0.000 |
makeup | +0.000 |
face | +0.000 |
prise | +0.000 |
Neg logit contributions | |
rob | -0.001 |
zag | -0.001 |
Hipp | -0.001 |
Hero | -0.001 |
Zig | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
makeup | +0.032 |
face | +0.031 |
vel | +0.030 |
spit | +0.030 |
her | +0.030 |
Neg logit contributions | |
rob | -0.060 |
zag | -0.055 |
Hipp | -0.053 |
girls | -0.052 |
Hero | -0.051 |
Token | whip |
Token ID | 22972 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
rob | +0.073 |
zag | +0.069 |
Hipp | +0.068 |
steep | +0.066 |
Hero | +0.066 |
Neg logit contributions | |
vel | -0.056 |
makeup | -0.054 |
spit | -0.053 |
prise | -0.052 |
face | -0.052 |
Token | moved |
Token ID | 3888 |
Feature activation | +5.79e-02 |
Pos logit contributions | |
rob | +0.209 |
zag | +0.198 |
Hipp | +0.194 |
Hero | +0.186 |
steep | +0.184 |
Neg logit contributions | |
vel | -0.147 |
makeup | -0.147 |
face | -0.142 |
spit | -0.141 |
blush | -0.133 |
Token | faster |
Token ID | 5443 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
rob | +0.108 |
zag | +0.105 |
Hipp | +0.104 |
steep | +0.102 |
Zig | +0.101 |
Neg logit contributions | |
spit | -0.080 |
makeup | -0.080 |
vel | -0.079 |
face | -0.078 |
prise | -0.077 |
Token | and |
Token ID | 290 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
rob | +0.585 |
zag | +0.555 |
Hipp | +0.545 |
steep | +0.527 |
Hero | +0.526 |
Neg logit contributions | |
vel | -0.312 |
makeup | -0.311 |
spit | -0.306 |
face | -0.301 |
prise | -0.285 |
Token | faster |
Token ID | 5443 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
rob | +0.019 |
zag | +0.018 |
habitats | +0.017 |
Hipp | +0.017 |
girls | +0.016 |
Neg logit contributions | |
her | -0.018 |
clean | -0.018 |
face | -0.018 |
makeup | -0.018 |
spit | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
rob | +0.593 |
zag | +0.561 |
Hipp | +0.555 |
Hero | +0.534 |
steep | +0.533 |
Neg logit contributions | |
vel | -0.303 |
makeup | -0.302 |
spit | -0.294 |
face | -0.290 |
prise | -0.278 |
Token | Now |
Token ID | 2735 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
She | +0.017 |
makeup | +0.017 |
face | +0.017 |
her | +0.017 |
vel | +0.016 |
Neg logit contributions | |
rob | -0.027 |
zag | -0.023 |
girls | -0.022 |
Hero | -0.022 |
steep | -0.022 |
Token | it |
Token ID | 340 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
vel | +0.015 |
makeup | +0.015 |
face | +0.014 |
spit | +0.014 |
blush | +0.013 |
Neg logit contributions | |
rob | -0.028 |
zag | -0.026 |
Hipp | -0.026 |
Hero | -0.025 |
steep | -0.025 |
Token | across |
Token ID | 1973 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
rob | +0.157 |
zag | +0.139 |
field | +0.135 |
Hipp | +0.134 |
girls | +0.131 |
Neg logit contributions | |
her | -0.055 |
face | -0.052 |
hand | -0.046 |
clean | -0.046 |
makeup | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
makeup | +0.021 |
face | +0.020 |
her | +0.020 |
vel | +0.019 |
spit | +0.019 |
Neg logit contributions | |
rob | -0.061 |
zag | -0.057 |
Hipp | -0.054 |
girls | -0.053 |
Hero | -0.052 |
Token | pool |
Token ID | 5933 |
Feature activation | +4.53e-02 |
Pos logit contributions | |
rob | +0.040 |
zag | +0.036 |
Hipp | +0.036 |
Hero | +0.034 |
girls | +0.033 |
Neg logit contributions | |
vel | -0.043 |
makeup | -0.043 |
face | -0.043 |
spit | -0.042 |
blush | -0.040 |
Token | , |
Token ID | 11 |
Feature activation | -4.07e-02 |
Pos logit contributions | |
rob | +0.114 |
Hipp | +0.105 |
zag | +0.103 |
girls | +0.100 |
Hero | +0.100 |
Neg logit contributions | |
face | -0.042 |
vel | -0.039 |
makeup | -0.039 |
spit | -0.038 |
her | -0.034 |
Token | swimming |
Token ID | 14899 |
Feature activation | +9.55e-02 |
Pos logit contributions | |
face | +0.043 |
hand | +0.039 |
spit | +0.038 |
clean | +0.038 |
makeup | +0.037 |
Neg logit contributions | |
rob | -0.100 |
zag | -0.089 |
Hipp | -0.088 |
girls | -0.086 |
Hero | -0.084 |
Token | faster |
Token ID | 5443 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
rob | +0.259 |
Hipp | +0.232 |
zag | +0.232 |
girls | +0.224 |
Hero | +0.223 |
Neg logit contributions | |
face | -0.076 |
makeup | -0.063 |
vel | -0.063 |
clean | -0.062 |
spit | -0.061 |
Token | and |
Token ID | 290 |
Feature activation | -2.94e-03 |
Pos logit contributions | |
rob | +0.646 |
Hipp | +0.602 |
zag | +0.598 |
Hero | +0.575 |
girls | +0.568 |
Neg logit contributions | |
face | -0.224 |
makeup | -0.205 |
vel | -0.203 |
spit | -0.192 |
her | -0.187 |
Token | faster |
Token ID | 5443 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
clean | +0.005 |
face | +0.004 |
spit | +0.004 |
dirt | +0.004 |
her | +0.004 |
Neg logit contributions | |
rob | -0.006 |
zag | -0.005 |
Hipp | -0.005 |
habitats | -0.005 |
Hero | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
rob | +0.622 |
zag | +0.580 |
Hipp | +0.573 |
Hero | +0.552 |
girls | +0.543 |
Neg logit contributions | |
face | -0.242 |
vel | -0.230 |
makeup | -0.229 |
spit | -0.223 |
her | -0.200 |
Token |
|
Token ID | 198 |
Feature activation | -4.98e-03 |
Pos logit contributions | |
face | +0.030 |
makeup | +0.029 |
spit | +0.028 |
vel | +0.028 |
hand | +0.027 |
Neg logit contributions | |
rob | -0.044 |
zag | -0.037 |
girls | -0.034 |
steep | -0.034 |
Hipp | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | -4.77e-03 |
Pos logit contributions | |
vel | +0.006 |
face | +0.006 |
makeup | +0.006 |
spit | +0.006 |
She | +0.006 |
Neg logit contributions | |
rob | -0.011 |
Hipp | -0.010 |
zag | -0.010 |
girls | -0.010 |
Hero | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +6.15e-04 |
Pos logit contributions | |
vel | +0.005 |
makeup | +0.005 |
face | +0.005 |
spit | +0.005 |
blush | +0.004 |
Neg logit contributions | |
rob | -0.008 |
zag | -0.007 |
Hipp | -0.007 |
Hero | -0.007 |
steep | -0.007 |
Token | The |
Token ID | 464 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
rob | +0.001 |
zag | +0.001 |
Hipp | +0.001 |
Zig | +0.001 |
Hero | +0.001 |
Neg logit contributions | |
face | -0.001 |
makeup | -0.001 |
spit | -0.001 |
blush | -0.001 |
vel | -0.001 |
Token | little |
Token ID | 1310 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
rob | +0.052 |
Hipp | +0.050 |
zag | +0.049 |
Hero | +0.047 |
steep | +0.046 |
Neg logit contributions | |
vel | -0.039 |
makeup | -0.038 |
spit | -0.037 |
face | -0.037 |
blush | -0.036 |
Token | squirrel |
Token ID | 33039 |
Feature activation | -4.20e-03 |
Pos logit contributions | |
vel | +0.227 |
makeup | +0.223 |
face | +0.220 |
spit | +0.218 |
blush | +0.204 |
Neg logit contributions | |
rob | -0.377 |
Hipp | -0.352 |
zag | -0.351 |
Hero | -0.335 |
steep | -0.331 |
Token | ran |
Token ID | 4966 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
vel | +0.005 |
face | +0.005 |
spit | +0.005 |
makeup | +0.005 |
blush | +0.005 |
Neg logit contributions | |
rob | -0.009 |
Hipp | -0.008 |
zag | -0.008 |
Hero | -0.008 |
steep | -0.008 |
Token | faster |
Token ID | 5443 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
face | +0.014 |
makeup | +0.014 |
vel | +0.014 |
spit | +0.013 |
her | +0.013 |
Neg logit contributions | |
rob | -0.029 |
zag | -0.026 |
Hipp | -0.026 |
girls | -0.025 |
Hero | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
rob | +0.662 |
Hipp | +0.615 |
zag | +0.613 |
Hero | +0.588 |
girls | +0.588 |
Neg logit contributions | |
face | -0.268 |
vel | -0.265 |
makeup | -0.260 |
spit | -0.250 |
her | -0.227 |
Token | but |
Token ID | 475 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
face | +0.024 |
makeup | +0.022 |
spit | +0.022 |
vel | +0.022 |
her | +0.021 |
Neg logit contributions | |
rob | -0.050 |
zag | -0.045 |
Hipp | -0.044 |
girls | -0.043 |
Hero | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | +4.55e-02 |
Pos logit contributions | |
her | +0.025 |
face | +0.025 |
vel | +0.023 |
spit | +0.023 |
makeup | +0.023 |
Neg logit contributions | |
rob | -0.046 |
zag | -0.039 |
girls | -0.038 |
Hipp | -0.037 |
habitats | -0.036 |
Token | fox |
Token ID | 21831 |
Feature activation | +6.04e-02 |
Pos logit contributions | |
rob | +0.099 |
Hipp | +0.096 |
zag | +0.094 |
Hero | +0.091 |
steep | +0.091 |
Neg logit contributions | |
vel | -0.054 |
makeup | -0.051 |
spit | -0.049 |
face | -0.048 |
prise | -0.048 |
Token | was |
Token ID | 373 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
rob | +0.134 |
Hipp | +0.120 |
zag | +0.120 |
Hero | +0.116 |
girls | +0.115 |
Neg logit contributions | |
spit | -0.073 |
face | -0.073 |
vel | -0.072 |
makeup | -0.070 |
blush | -0.064 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.74e-02 |
Pos logit contributions | |
face | +0.003 |
She | +0.003 |
spit | +0.003 |
vel | +0.003 |
makeup | +0.003 |
Neg logit contributions | |
rob | -0.009 |
zag | -0.008 |
steep | -0.007 |
Hero | -0.007 |
Hipp | -0.007 |
Token | Sam |
Token ID | 16305 |
Feature activation | -2.55e-02 |
Pos logit contributions | |
makeup | +0.051 |
vel | +0.050 |
her | +0.048 |
She | +0.048 |
face | +0.048 |
Neg logit contributions | |
rob | -0.103 |
zag | -0.099 |
Hipp | -0.096 |
steep | -0.096 |
girls | -0.093 |
Token | was |
Token ID | 373 |
Feature activation | +3.32e-02 |
Pos logit contributions | |
vel | +0.024 |
makeup | +0.022 |
face | +0.022 |
spit | +0.022 |
blush | +0.020 |
Neg logit contributions | |
rob | -0.063 |
zag | -0.059 |
Hipp | -0.058 |
Hero | -0.057 |
steep | -0.056 |
Token | picking |
Token ID | 10868 |
Feature activation | -5.75e-02 |
Pos logit contributions | |
rob | +0.072 |
Hipp | +0.068 |
zag | +0.068 |
Hero | +0.065 |
steep | +0.065 |
Neg logit contributions | |
vel | -0.038 |
makeup | -0.037 |
face | -0.036 |
spit | -0.036 |
blush | -0.034 |
Token | wild |
Token ID | 4295 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
vel | +0.083 |
spit | +0.081 |
face | +0.079 |
makeup | +0.078 |
prise | +0.076 |
Neg logit contributions | |
rob | -0.112 |
Hipp | -0.105 |
zag | -0.102 |
Hero | -0.098 |
Zig | -0.097 |
Token | fl |
Token ID | 2704 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
rob | +0.059 |
zag | +0.056 |
Hipp | +0.054 |
steep | +0.052 |
Hero | +0.052 |
Neg logit contributions | |
makeup | -0.035 |
face | -0.033 |
vel | -0.032 |
spit | -0.032 |
blush | -0.031 |
Token | owers |
Token ID | 3618 |
Feature activation | +5.19e-02 |
Pos logit contributions | |
rob | +0.416 |
zag | +0.373 |
Hipp | +0.359 |
girls | +0.338 |
Zig | +0.322 |
Neg logit contributions | |
face | -0.468 |
spit | -0.466 |
vel | -0.458 |
makeup | -0.445 |
blush | -0.428 |
Token | in |
Token ID | 287 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
rob | +0.111 |
Hipp | +0.109 |
zag | +0.109 |
steep | +0.103 |
aboard | +0.102 |
Neg logit contributions | |
vel | -0.059 |
spit | -0.058 |
makeup | -0.056 |
face | -0.056 |
prise | -0.054 |
Token | the |
Token ID | 262 |
Feature activation | +4.29e-02 |
Pos logit contributions | |
Hipp | +0.023 |
rob | +0.023 |
aboard | +0.022 |
steep | +0.021 |
Hero | +0.021 |
Neg logit contributions | |
vel | -0.026 |
spit | -0.025 |
blush | -0.024 |
face | -0.023 |
prise | -0.023 |
Token | me |
Token ID | 502 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
rob | +0.077 |
Hipp | +0.074 |
steep | +0.070 |
zag | +0.070 |
Hero | +0.068 |
Neg logit contributions | |
vel | -0.069 |
spit | -0.064 |
face | -0.059 |
blush | -0.059 |
prise | -0.058 |
Token | adow |
Token ID | 4584 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
vel | +0.162 |
spit | +0.161 |
makeup | +0.161 |
face | +0.157 |
blush | +0.152 |
Neg logit contributions | |
zag | -0.196 |
rob | -0.191 |
Hipp | -0.177 |
steep | -0.176 |
girls | -0.169 |
Token | comfortable |
Token ID | 6792 |
Feature activation | -3.24e-03 |
Pos logit contributions | |
rob | +0.142 |
zag | +0.133 |
Hipp | +0.133 |
Hero | +0.127 |
steep | +0.126 |
Neg logit contributions | |
vel | -0.077 |
makeup | -0.077 |
face | -0.075 |
spit | -0.074 |
blush | -0.070 |
Token | and |
Token ID | 290 |
Feature activation | +6.05e-04 |
Pos logit contributions | |
face | +0.004 |
makeup | +0.004 |
vel | +0.004 |
spit | +0.004 |
her | +0.003 |
Neg logit contributions | |
rob | -0.007 |
zag | -0.007 |
Hipp | -0.007 |
steep | -0.006 |
girls | -0.006 |
Token | relaxed |
Token ID | 18397 |
Feature activation | +9.43e-03 |
Pos logit contributions | |
rob | +0.001 |
zag | +0.001 |
Hipp | +0.001 |
Hero | +0.001 |
girls | +0.001 |
Neg logit contributions | |
vel | -0.001 |
makeup | -0.001 |
face | -0.001 |
spit | -0.001 |
blush | -0.001 |
Token | as |
Token ID | 355 |
Feature activation | +6.52e-02 |
Pos logit contributions | |
rob | +0.022 |
zag | +0.021 |
Hipp | +0.021 |
Hero | +0.020 |
steep | +0.020 |
Neg logit contributions | |
vel | -0.010 |
makeup | -0.009 |
spit | -0.009 |
face | -0.009 |
blush | -0.008 |
Token | he |
Token ID | 339 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
rob | +0.123 |
zag | +0.110 |
Hipp | +0.106 |
girls | +0.104 |
Hero | +0.099 |
Neg logit contributions | |
face | -0.101 |
makeup | -0.100 |
vel | -0.100 |
spit | -0.098 |
her | -0.096 |
Token | always |
Token ID | 1464 |
Feature activation | +7.97e-02 |
Pos logit contributions | |
rob | +0.042 |
zag | +0.039 |
Hipp | +0.038 |
Hero | +0.036 |
steep | +0.036 |
Neg logit contributions | |
spit | -0.026 |
vel | -0.026 |
face | -0.026 |
makeup | -0.025 |
blush | -0.024 |
Token | did |
Token ID | 750 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
rob | +0.168 |
zag | +0.157 |
Hipp | +0.157 |
Hero | +0.150 |
girls | +0.149 |
Neg logit contributions | |
vel | -0.100 |
face | -0.100 |
makeup | -0.100 |
spit | -0.099 |
blush | -0.092 |
Token | outside |
Token ID | 2354 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
makeup | +0.024 |
vel | +0.024 |
spit | +0.023 |
face | +0.023 |
blush | +0.022 |
Neg logit contributions | |
rob | -0.047 |
zag | -0.045 |
Hipp | -0.044 |
Hero | -0.042 |
steep | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
vel | +0.030 |
makeup | +0.029 |
spit | +0.029 |
face | +0.029 |
blush | +0.027 |
Neg logit contributions | |
rob | -0.060 |
zag | -0.056 |
Hipp | -0.056 |
Hero | -0.054 |
steep | -0.053 |
Token | He |
Token ID | 679 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
face | +0.032 |
spit | +0.030 |
vel | +0.030 |
makeup | +0.029 |
She | +0.029 |
Neg logit contributions | |
rob | -0.040 |
zag | -0.036 |
Hipp | -0.035 |
girls | -0.034 |
Zig | -0.032 |
Token | also |
Token ID | 635 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
spit | +0.020 |
face | +0.019 |
vel | +0.017 |
dust | +0.017 |
cut | +0.017 |
Neg logit contributions | |
rob | -0.043 |
zag | -0.039 |
Hipp | -0.039 |
girls | -0.037 |
Zig | -0.037 |
Token | stop |
Token ID | 2245 |
Feature activation | -6.49e-02 |
Pos logit contributions | |
Hipp | +0.095 |
zag | +0.092 |
Hero | +0.087 |
rob | +0.086 |
steep | +0.085 |
Neg logit contributions | |
vel | -0.084 |
makeup | -0.081 |
prise | -0.076 |
pled | -0.075 |
aming | -0.074 |
Token | smiling |
Token ID | 16755 |
Feature activation | +7.63e-02 |
Pos logit contributions | |
vel | +0.082 |
prise | +0.082 |
makeup | +0.079 |
spit | +0.077 |
face | +0.074 |
Neg logit contributions | |
Hipp | -0.122 |
rob | -0.120 |
zag | -0.116 |
aboard | -0.112 |
Hero | -0.111 |
Token | . |
Token ID | 13 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
rob | +0.159 |
Hipp | +0.155 |
zag | +0.155 |
Hero | +0.150 |
steep | +0.148 |
Neg logit contributions | |
makeup | -0.085 |
vel | -0.085 |
spit | -0.084 |
prise | -0.082 |
face | -0.078 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -5.32e-02 |
Pos logit contributions | |
face | +0.024 |
vel | +0.024 |
makeup | +0.024 |
spit | +0.023 |
blush | +0.022 |
Neg logit contributions | |
rob | -0.032 |
zag | -0.030 |
Hipp | -0.029 |
girls | -0.028 |
steep | -0.027 |
Token | He |
Token ID | 1544 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
makeup | +0.077 |
vel | +0.076 |
face | +0.074 |
spit | +0.072 |
She | +0.072 |
Neg logit contributions | |
rob | -0.098 |
zag | -0.093 |
Hipp | -0.091 |
steep | -0.089 |
girls | -0.086 |
Token | idi |
Token ID | 19830 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
vel | +0.060 |
makeup | +0.059 |
face | +0.058 |
spit | +0.056 |
blush | +0.054 |
Neg logit contributions | |
rob | -0.091 |
zag | -0.084 |
Hipp | -0.084 |
Hero | -0.080 |
steep | -0.079 |
Token | had |
Token ID | 550 |
Feature activation | +6.77e-03 |
Pos logit contributions | |
rob | +0.361 |
Hipp | +0.337 |
zag | +0.336 |
Hero | +0.322 |
girls | +0.317 |
Neg logit contributions | |
vel | -0.216 |
makeup | -0.212 |
face | -0.208 |
spit | -0.205 |
blush | -0.193 |
Token | a |
Token ID | 257 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
rob | +0.017 |
zag | +0.016 |
Hipp | +0.016 |
Hero | +0.015 |
girls | +0.015 |
Neg logit contributions | |
makeup | -0.006 |
vel | -0.006 |
face | -0.006 |
spit | -0.006 |
blush | -0.005 |
Token | big |
Token ID | 1263 |
Feature activation | -9.99e-02 |
Pos logit contributions | |
vel | +0.028 |
makeup | +0.026 |
spit | +0.026 |
face | +0.026 |
blush | +0.024 |
Neg logit contributions | |
rob | -0.052 |
zag | -0.049 |
Hipp | -0.049 |
Hero | -0.047 |
steep | -0.046 |
Token | problem |
Token ID | 1917 |
Feature activation | +6.36e-02 |
Pos logit contributions | |
vel | +0.127 |
spit | +0.121 |
makeup | +0.119 |
face | +0.116 |
blush | +0.110 |
Neg logit contributions | |
rob | -0.210 |
zag | -0.196 |
Hipp | -0.196 |
Hero | -0.188 |
steep | -0.188 |
Token | , |
Token ID | 11 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
rob | +0.139 |
zag | +0.135 |
Hipp | +0.131 |
steep | +0.125 |
Hero | +0.125 |
Neg logit contributions | |
vel | -0.070 |
spit | -0.069 |
face | -0.068 |
makeup | -0.067 |
blush | -0.064 |
Token | car |
Token ID | 1097 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
rob | +0.073 |
zag | +0.067 |
Hipp | +0.067 |
Hero | +0.064 |
steep | +0.063 |
Neg logit contributions | |
vel | -0.073 |
makeup | -0.069 |
spit | -0.069 |
face | -0.068 |
blush | -0.066 |
Token | went |
Token ID | 1816 |
Feature activation | -5.13e-02 |
Pos logit contributions | |
rob | +0.095 |
Hipp | +0.088 |
girls | +0.083 |
zag | +0.083 |
Skip | +0.082 |
Neg logit contributions | |
face | -0.042 |
spit | -0.041 |
makeup | -0.036 |
cut | -0.035 |
vel | -0.033 |
Token | too |
Token ID | 1165 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
makeup | +0.058 |
face | +0.056 |
vel | +0.054 |
spit | +0.053 |
blush | +0.050 |
Neg logit contributions | |
rob | -0.119 |
zag | -0.109 |
Hipp | -0.109 |
girls | -0.103 |
Hero | -0.102 |
Token | fast |
Token ID | 3049 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
rob | +0.276 |
Hipp | +0.255 |
zag | +0.255 |
Hero | +0.242 |
girls | +0.241 |
Neg logit contributions | |
vel | -0.179 |
makeup | -0.178 |
face | -0.177 |
spit | -0.174 |
blush | -0.162 |
Token | and |
Token ID | 290 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
rob | +0.247 |
zag | +0.234 |
Hipp | +0.230 |
Hero | +0.224 |
steep | +0.221 |
Neg logit contributions | |
vel | -0.117 |
makeup | -0.116 |
spit | -0.114 |
face | -0.113 |
blush | -0.106 |
Token | hit |
Token ID | 2277 |
Feature activation | +9.92e-02 |
Pos logit contributions | |
rob | +0.035 |
zag | +0.031 |
Hipp | +0.029 |
girls | +0.029 |
habitats | +0.028 |
Neg logit contributions | |
spit | -0.013 |
face | -0.013 |
her | -0.013 |
drop | -0.012 |
clean | -0.012 |
Token | a |
Token ID | 257 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
rob | +0.208 |
zag | +0.194 |
Hipp | +0.193 |
Hero | +0.183 |
girls | +0.182 |
Neg logit contributions | |
vel | -0.127 |
makeup | -0.127 |
face | -0.125 |
spit | -0.123 |
blush | -0.115 |
Token | wall |
Token ID | 3355 |
Feature activation | -5.53e-02 |
Pos logit contributions | |
vel | +0.021 |
makeup | +0.021 |
illy | +0.020 |
prise | +0.020 |
spit | +0.019 |
Neg logit contributions | |
steep | -0.022 |
zag | -0.021 |
rob | -0.021 |
Hipp | -0.020 |
Hero | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -7.36e-03 |
Pos logit contributions | |
makeup | +0.066 |
vel | +0.065 |
prise | +0.062 |
face | +0.061 |
spit | +0.061 |
Neg logit contributions | |
zag | -0.116 |
rob | -0.115 |
steep | -0.111 |
Hipp | -0.108 |
Hero | -0.106 |
Token | The |
Token ID | 383 |
Feature activation | +5.32e-02 |
Pos logit contributions | |
face | +0.009 |
spit | +0.009 |
makeup | +0.009 |
vel | +0.008 |
blush | +0.008 |
Neg logit contributions | |
rob | -0.016 |
zag | -0.015 |
Hipp | -0.015 |
girls | -0.014 |
Hero | -0.013 |
Token | toy |
Token ID | 13373 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
rob | +0.075 |
zag | +0.072 |
Hero | +0.070 |
steep | +0.070 |
Hipp | +0.069 |
Neg logit contributions | |
vel | -0.106 |
illy | -0.095 |
prise | -0.095 |
spit | -0.094 |
makeup | -0.093 |
Token | with |
Token ID | 351 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
spit | +0.080 |
vel | +0.080 |
prise | +0.079 |
worry | +0.079 |
blush | +0.077 |
Neg logit contributions | |
rob | -0.159 |
zag | -0.142 |
Hero | -0.141 |
Hipp | -0.138 |
Zig | -0.137 |
Token | his |
Token ID | 465 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
vel | +0.031 |
prise | +0.029 |
spit | +0.028 |
face | +0.027 |
blush | +0.027 |
Neg logit contributions | |
rob | -0.030 |
Hipp | -0.027 |
Hero | -0.026 |
zag | -0.026 |
Zig | -0.026 |
Token | toys |
Token ID | 14958 |
Feature activation | -8.15e-02 |
Pos logit contributions | |
makeup | +0.030 |
face | +0.030 |
vel | +0.030 |
spit | +0.029 |
blush | +0.026 |
Neg logit contributions | |
rob | -0.071 |
zag | -0.067 |
Hipp | -0.066 |
Hero | -0.064 |
steep | -0.063 |
Token | , |
Token ID | 11 |
Feature activation | -4.34e-02 |
Pos logit contributions | |
vel | +0.098 |
spit | +0.094 |
makeup | +0.092 |
face | +0.091 |
prise | +0.089 |
Neg logit contributions | |
rob | -0.179 |
zag | -0.170 |
Hipp | -0.164 |
Hero | -0.162 |
steep | -0.159 |
Token | but |
Token ID | 475 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
vel | +0.064 |
makeup | +0.061 |
spit | +0.059 |
face | +0.059 |
prise | +0.058 |
Neg logit contributions | |
rob | -0.084 |
Hipp | -0.078 |
zag | -0.077 |
Hero | -0.074 |
steep | -0.074 |
Token | one |
Token ID | 530 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
her | +0.049 |
makeup | +0.045 |
face | +0.043 |
worry | +0.041 |
spit | +0.040 |
Neg logit contributions | |
rob | -0.079 |
girls | -0.074 |
zag | -0.073 |
Hipp | -0.071 |
Hero | -0.068 |
Token | day |
Token ID | 1110 |
Feature activation | -9.07e-03 |
Pos logit contributions | |
rob | +0.142 |
zag | +0.139 |
Hipp | +0.137 |
steep | +0.125 |
Hero | +0.125 |
Neg logit contributions | |
vel | -0.252 |
makeup | -0.237 |
spit | -0.233 |
blush | -0.230 |
face | -0.229 |
Token | his |
Token ID | 465 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
vel | +0.017 |
spit | +0.017 |
makeup | +0.016 |
prise | +0.016 |
face | +0.016 |
Neg logit contributions | |
rob | -0.013 |
Hipp | -0.012 |
zag | -0.012 |
Zig | -0.012 |
Hero | -0.011 |
Token | favorite |
Token ID | 4004 |
Feature activation | +6.32e-02 |
Pos logit contributions | |
vel | +0.042 |
spit | +0.038 |
makeup | +0.037 |
face | +0.036 |
prise | +0.036 |
Neg logit contributions | |
rob | -0.059 |
zag | -0.058 |
Hipp | -0.057 |
Hero | -0.054 |
Zig | -0.053 |
Token | toy |
Token ID | 13373 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
rob | +0.120 |
zag | +0.114 |
Hipp | +0.111 |
Hero | +0.106 |
Zig | +0.103 |
Neg logit contributions | |
vel | -0.097 |
prise | -0.087 |
spit | -0.086 |
makeup | -0.086 |
She | -0.085 |
Token | broke |
Token ID | 6265 |
Feature activation | -4.16e-03 |
Pos logit contributions | |
vel | +0.063 |
makeup | +0.060 |
spit | +0.059 |
face | +0.059 |
prise | +0.057 |
Neg logit contributions | |
rob | -0.080 |
zag | -0.076 |
Hipp | -0.075 |
Hero | -0.071 |
Zig | -0.070 |
Token | , |
Token ID | 11 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
makeup | +0.056 |
vel | +0.055 |
face | +0.054 |
spit | +0.052 |
blush | +0.049 |
Neg logit contributions | |
rob | -0.145 |
zag | -0.136 |
Hipp | -0.135 |
Hero | -0.130 |
girls | -0.129 |
Token | she |
Token ID | 673 |
Feature activation | +4.31e-03 |
Pos logit contributions | |
vel | +0.035 |
makeup | +0.035 |
face | +0.034 |
spit | +0.034 |
blush | +0.031 |
Neg logit contributions | |
rob | -0.084 |
zag | -0.079 |
Hipp | -0.079 |
Hero | -0.076 |
steep | -0.075 |
Token | improved |
Token ID | 6596 |
Feature activation | +5.97e-02 |
Pos logit contributions | |
rob | +0.009 |
Hipp | +0.009 |
zag | +0.009 |
Hero | +0.008 |
Zig | +0.008 |
Neg logit contributions | |
spit | -0.005 |
makeup | -0.005 |
face | -0.005 |
vel | -0.005 |
blush | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
rob | +0.141 |
zag | +0.133 |
Hipp | +0.132 |
Hero | +0.128 |
steep | +0.127 |
Neg logit contributions | |
vel | -0.058 |
makeup | -0.057 |
face | -0.056 |
spit | -0.056 |
blush | -0.051 |
Token | world |
Token ID | 995 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
rob | +0.051 |
zag | +0.048 |
steep | +0.048 |
Hipp | +0.047 |
Hero | +0.046 |
Neg logit contributions | |
vel | -0.055 |
spit | -0.053 |
makeup | -0.052 |
blush | -0.050 |
face | -0.050 |
Token | one |
Token ID | 530 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
makeup | +0.045 |
vel | +0.044 |
face | +0.043 |
spit | +0.041 |
her | +0.039 |
Neg logit contributions | |
rob | -0.105 |
Hipp | -0.097 |
zag | -0.096 |
girls | -0.093 |
Hero | -0.092 |
Token | little |
Token ID | 1310 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
rob | +0.267 |
zag | +0.256 |
Hipp | +0.252 |
steep | +0.243 |
Hero | +0.242 |
Neg logit contributions | |
vel | -0.144 |
makeup | -0.141 |
spit | -0.139 |
face | -0.138 |
blush | -0.130 |
Token | bit |
Token ID | 1643 |
Feature activation | -6.98e-02 |
Pos logit contributions | |
makeup | +0.238 |
face | +0.236 |
vel | +0.235 |
spit | +0.230 |
blush | +0.216 |
Neg logit contributions | |
rob | -0.392 |
zag | -0.364 |
Hipp | -0.360 |
Hero | -0.343 |
Zig | -0.340 |
Token | at |
Token ID | 379 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
vel | +0.078 |
makeup | +0.077 |
face | +0.076 |
spit | +0.075 |
blush | +0.069 |
Neg logit contributions | |
rob | -0.155 |
Hipp | -0.147 |
zag | -0.147 |
steep | -0.141 |
Hero | -0.140 |
Token | a |
Token ID | 257 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
makeup | +0.019 |
vel | +0.019 |
face | +0.019 |
spit | +0.018 |
blush | +0.017 |
Neg logit contributions | |
rob | -0.038 |
Hipp | -0.036 |
zag | -0.036 |
Hero | -0.034 |
girls | -0.034 |
Token | time |
Token ID | 640 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
makeup | +0.037 |
spit | +0.037 |
vel | +0.036 |
blush | +0.036 |
face | +0.036 |
Neg logit contributions | |
zag | -0.046 |
rob | -0.046 |
Hipp | -0.046 |
steep | -0.045 |
Hero | -0.043 |
Token | was |
Token ID | 373 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
spit | +0.024 |
makeup | +0.024 |
face | +0.024 |
vel | +0.024 |
dust | +0.022 |
Neg logit contributions | |
rob | -0.048 |
Hipp | -0.045 |
zag | -0.044 |
girls | -0.043 |
Zig | -0.042 |
Token | his |
Token ID | 465 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
rob | +0.046 |
zag | +0.043 |
Hipp | +0.042 |
girls | +0.041 |
Hero | +0.040 |
Neg logit contributions | |
makeup | -0.017 |
face | -0.017 |
vel | -0.017 |
spit | -0.016 |
her | -0.016 |
Token | reward |
Token ID | 6721 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
face | +0.040 |
makeup | +0.040 |
vel | +0.038 |
spit | +0.038 |
blush | +0.036 |
Neg logit contributions | |
rob | -0.060 |
Hipp | -0.053 |
zag | -0.053 |
girls | -0.051 |
Hero | -0.051 |
Token | ! |
Token ID | 0 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
rob | +0.035 |
zag | +0.033 |
Hipp | +0.033 |
Hero | +0.032 |
girls | +0.031 |
Neg logit contributions | |
vel | -0.015 |
makeup | -0.015 |
face | -0.015 |
spit | -0.015 |
blush | -0.013 |
Token | Jack |
Token ID | 3619 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
vel | +0.021 |
makeup | +0.021 |
face | +0.021 |
spit | +0.021 |
blush | +0.019 |
Neg logit contributions | |
rob | -0.026 |
zag | -0.024 |
Hipp | -0.024 |
Hero | -0.023 |
Zig | -0.022 |
Token | was |
Token ID | 373 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
face | +0.045 |
vel | +0.045 |
makeup | +0.045 |
spit | +0.044 |
blush | +0.041 |
Neg logit contributions | |
rob | -0.073 |
Hipp | -0.068 |
zag | -0.067 |
Hero | -0.064 |
steep | -0.064 |
Token | so |
Token ID | 523 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
rob | +0.043 |
zag | +0.040 |
Hipp | +0.040 |
Hero | +0.038 |
steep | +0.038 |
Neg logit contributions | |
vel | -0.017 |
makeup | -0.017 |
face | -0.016 |
spit | -0.016 |
blush | -0.015 |
Token | happy |
Token ID | 3772 |
Feature activation | -4.60e-02 |
Pos logit contributions | |
Hipp | +0.043 |
zag | +0.041 |
steep | +0.041 |
rob | +0.041 |
Zig | +0.040 |
Neg logit contributions | |
makeup | -0.038 |
vel | -0.037 |
blush | -0.035 |
lipstick | -0.034 |
file | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
face | +0.050 |
makeup | +0.049 |
vel | +0.049 |
spit | +0.047 |
her | +0.043 |
Neg logit contributions | |
rob | -0.108 |
zag | -0.100 |
Hipp | -0.099 |
girls | -0.095 |
steep | -0.095 |
Token | He |
Token ID | 679 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
face | +0.031 |
makeup | +0.030 |
vel | +0.029 |
spit | +0.029 |
She | +0.028 |
Neg logit contributions | |
rob | -0.036 |
zag | -0.033 |
Hipp | -0.033 |
girls | -0.031 |
steep | -0.031 |
Token | learned |
Token ID | 4499 |
Feature activation | +2.10e-03 |
Pos logit contributions | |
face | +0.023 |
spit | +0.022 |
makeup | +0.021 |
vel | +0.020 |
her | +0.019 |
Neg logit contributions | |
rob | -0.045 |
zag | -0.042 |
Hipp | -0.042 |
girls | -0.040 |
steep | -0.040 |
Token | things |
Token ID | 1243 |
Feature activation | -6.90e-02 |
Pos logit contributions | |
vel | +0.038 |
makeup | +0.037 |
spit | +0.037 |
blush | +0.036 |
face | +0.036 |
Neg logit contributions | |
rob | -0.041 |
zag | -0.036 |
steep | -0.036 |
Hipp | -0.036 |
Hero | -0.034 |
Token | with |
Token ID | 351 |
Feature activation | -8.33e-03 |
Pos logit contributions | |
makeup | +0.069 |
vel | +0.068 |
face | +0.067 |
spit | +0.065 |
blush | +0.060 |
Neg logit contributions | |
rob | -0.165 |
Hipp | -0.154 |
zag | -0.153 |
Hero | -0.147 |
girls | -0.146 |
Token | blocks |
Token ID | 7021 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
vel | +0.011 |
makeup | +0.011 |
spit | +0.011 |
face | +0.011 |
prise | +0.010 |
Neg logit contributions | |
rob | -0.016 |
zag | -0.016 |
Hipp | -0.016 |
steep | -0.015 |
Hero | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
rob | +0.042 |
Hipp | +0.039 |
zag | +0.039 |
Hero | +0.038 |
Zig | +0.037 |
Neg logit contributions | |
makeup | -0.016 |
vel | -0.016 |
face | -0.015 |
spit | -0.015 |
blush | -0.014 |
Token | like |
Token ID | 588 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
makeup | +0.042 |
face | +0.041 |
spit | +0.040 |
vel | +0.040 |
blush | +0.038 |
Neg logit contributions | |
rob | -0.068 |
zag | -0.063 |
Hipp | -0.063 |
Hero | -0.060 |
girls | -0.059 |
Token | houses |
Token ID | 7777 |
Feature activation | +7.39e-02 |
Pos logit contributions | |
vel | +0.022 |
makeup | +0.021 |
face | +0.021 |
spit | +0.021 |
blush | +0.020 |
Neg logit contributions | |
rob | -0.027 |
Hipp | -0.025 |
zag | -0.025 |
Hero | -0.024 |
Zig | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
rob | +0.184 |
Hipp | +0.172 |
zag | +0.171 |
Hero | +0.164 |
Zig | +0.163 |
Neg logit contributions | |
makeup | -0.067 |
face | -0.066 |
vel | -0.066 |
spit | -0.063 |
blush | -0.058 |
Token | cars |
Token ID | 5006 |
Feature activation | +9.95e-02 |
Pos logit contributions | |
makeup | +0.048 |
face | +0.047 |
spit | +0.046 |
vel | +0.045 |
blush | +0.044 |
Neg logit contributions | |
rob | -0.063 |
zag | -0.059 |
Hipp | -0.059 |
Hero | -0.055 |
girls | -0.055 |
Token | , |
Token ID | 11 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
rob | +0.219 |
zag | +0.209 |
Hipp | +0.203 |
Hero | +0.198 |
girls | +0.196 |
Neg logit contributions | |
vel | -0.116 |
makeup | -0.113 |
spit | -0.112 |
face | -0.110 |
blush | -0.104 |
Token | and |
Token ID | 290 |
Feature activation | +5.28e-03 |
Pos logit contributions | |
makeup | +0.047 |
face | +0.046 |
vel | +0.046 |
spit | +0.045 |
blush | +0.043 |
Neg logit contributions | |
rob | -0.065 |
zag | -0.061 |
Hipp | -0.060 |
Hero | -0.056 |
girls | -0.056 |
Token | towers |
Token ID | 18028 |
Feature activation | +5.19e-02 |
Pos logit contributions | |
rob | +0.009 |
Hipp | +0.008 |
zag | +0.008 |
Hero | +0.008 |
steep | +0.008 |
Neg logit contributions | |
vel | -0.009 |
makeup | -0.008 |
face | -0.008 |
spit | -0.008 |
blush | -0.008 |
Token | took |
Token ID | 1718 |
Feature activation | -9.30e-02 |
Pos logit contributions | |
face | +0.021 |
spit | +0.021 |
vel | +0.021 |
makeup | +0.020 |
blush | +0.019 |
Neg logit contributions | |
rob | -0.038 |
zag | -0.034 |
Hipp | -0.034 |
girls | -0.033 |
Hero | -0.032 |
Token | his |
Token ID | 465 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
vel | +0.128 |
makeup | +0.127 |
spit | +0.126 |
face | +0.120 |
blush | +0.119 |
Neg logit contributions | |
rob | -0.179 |
Hipp | -0.169 |
Hero | -0.168 |
aboard | -0.166 |
Zig | -0.165 |
Token | treasure |
Token ID | 14068 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
vel | +0.036 |
makeup | +0.035 |
spit | +0.035 |
face | +0.035 |
blush | +0.033 |
Neg logit contributions | |
rob | -0.043 |
zag | -0.041 |
Hipp | -0.040 |
Hero | -0.039 |
steep | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | +2.82e-03 |
Pos logit contributions | |
rob | +0.038 |
zag | +0.037 |
Hipp | +0.036 |
Hero | +0.035 |
steep | +0.035 |
Neg logit contributions | |
vel | -0.022 |
makeup | -0.021 |
spit | -0.021 |
face | -0.020 |
blush | -0.020 |
Token | went |
Token ID | 1816 |
Feature activation | -6.27e-02 |
Pos logit contributions | |
rob | +0.006 |
zag | +0.005 |
Hipp | +0.005 |
Hero | +0.005 |
girls | +0.005 |
Neg logit contributions | |
vel | -0.004 |
makeup | -0.004 |
face | -0.004 |
spit | -0.004 |
blush | -0.003 |
Token | running |
Token ID | 2491 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
makeup | +0.069 |
face | +0.069 |
vel | +0.068 |
spit | +0.063 |
blush | +0.061 |
Neg logit contributions | |
rob | -0.145 |
Hipp | -0.134 |
zag | -0.132 |
girls | -0.128 |
Hero | -0.127 |
Token | back |
Token ID | 736 |
Feature activation | +8.35e-03 |
Pos logit contributions | |
rob | +0.085 |
Hipp | +0.077 |
zag | +0.076 |
girls | +0.074 |
Hero | +0.073 |
Neg logit contributions | |
face | -0.039 |
makeup | -0.038 |
vel | -0.036 |
her | -0.036 |
blush | -0.033 |
Token | home |
Token ID | 1363 |
Feature activation | -5.95e-03 |
Pos logit contributions | |
rob | +0.021 |
Hipp | +0.019 |
girls | +0.018 |
zag | +0.018 |
Hero | +0.018 |
Neg logit contributions | |
her | -0.008 |
makeup | -0.008 |
face | -0.008 |
hand | -0.007 |
clean | -0.007 |
Token | with |
Token ID | 351 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
vel | +0.006 |
makeup | +0.005 |
face | +0.005 |
spit | +0.005 |
blush | +0.005 |
Neg logit contributions | |
rob | -0.015 |
Hipp | -0.014 |
zag | -0.014 |
Hero | -0.013 |
girls | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
vel | +0.020 |
makeup | +0.019 |
spit | +0.019 |
blush | +0.018 |
face | +0.018 |
Neg logit contributions | |
rob | -0.025 |
Hipp | -0.024 |
Hero | -0.023 |
Zig | -0.023 |
steep | -0.023 |
Token | big |
Token ID | 1263 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
vel | +0.034 |
makeup | +0.032 |
spit | +0.031 |
blush | +0.029 |
She | +0.028 |
Neg logit contributions | |
zag | -0.052 |
steep | -0.051 |
rob | -0.051 |
Hipp | -0.050 |
Hero | -0.049 |
Token | , |
Token ID | 11 |
Feature activation | -3.86e-02 |
Pos logit contributions | |
makeup | +0.062 |
vel | +0.059 |
face | +0.059 |
spit | +0.055 |
her | +0.053 |
Neg logit contributions | |
rob | -0.156 |
zag | -0.147 |
Hipp | -0.145 |
Hero | -0.138 |
steep | -0.138 |
Token | the |
Token ID | 262 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
makeup | +0.043 |
vel | +0.042 |
face | +0.042 |
spit | +0.040 |
blush | +0.038 |
Neg logit contributions | |
rob | -0.088 |
zag | -0.083 |
Hipp | -0.081 |
Hero | -0.078 |
girls | -0.077 |
Token | rabbit |
Token ID | 22746 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
rob | +0.057 |
zag | +0.053 |
Hipp | +0.053 |
Hero | +0.050 |
steep | +0.050 |
Neg logit contributions | |
vel | -0.044 |
makeup | -0.043 |
spit | -0.042 |
face | -0.042 |
blush | -0.040 |
Token | was |
Token ID | 373 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
face | +0.024 |
spit | +0.024 |
makeup | +0.024 |
her | +0.024 |
vel | +0.023 |
Neg logit contributions | |
rob | -0.032 |
zag | -0.028 |
Hipp | -0.028 |
steep | -0.027 |
Hero | -0.026 |
Token | right |
Token ID | 826 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
rob | +0.043 |
Hipp | +0.039 |
zag | +0.039 |
girls | +0.038 |
Hero | +0.038 |
Neg logit contributions | |
her | -0.018 |
makeup | -0.018 |
vel | -0.017 |
face | -0.017 |
spit | -0.016 |
Token | there |
Token ID | 612 |
Feature activation | +3.22e-02 |
Pos logit contributions | |
makeup | +0.026 |
vel | +0.026 |
face | +0.026 |
spit | +0.024 |
her | +0.022 |
Neg logit contributions | |
rob | -0.066 |
Hipp | -0.061 |
zag | -0.061 |
Hero | -0.059 |
Zig | -0.058 |
Token | with |
Token ID | 351 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
rob | +0.065 |
zag | +0.062 |
Hipp | +0.062 |
Hero | +0.059 |
steep | +0.059 |
Neg logit contributions | |
vel | -0.041 |
makeup | -0.040 |
spit | -0.040 |
face | -0.040 |
blush | -0.037 |
Token | her |
Token ID | 607 |
Feature activation | -7.29e-02 |
Pos logit contributions | |
vel | +0.015 |
face | +0.014 |
spit | +0.014 |
makeup | +0.014 |
prise | +0.013 |
Neg logit contributions | |
rob | -0.033 |
Hipp | -0.031 |
zag | -0.031 |
Hero | -0.030 |
girls | -0.030 |
Token | - |
Token ID | 532 |
Feature activation | +4.01e-02 |
Pos logit contributions | |
vel | +0.093 |
makeup | +0.092 |
face | +0.091 |
spit | +0.090 |
blush | +0.085 |
Neg logit contributions | |
rob | -0.153 |
zag | -0.143 |
Hipp | -0.142 |
Hero | -0.136 |
steep | -0.134 |
Token | even |
Token ID | 772 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
rob | +0.092 |
zag | +0.086 |
Hipp | +0.083 |
girls | +0.080 |
Hero | +0.080 |
Neg logit contributions | |
makeup | -0.045 |
vel | -0.044 |
face | -0.044 |
spit | -0.042 |
her | -0.040 |
Token | when |
Token ID | 618 |
Feature activation | +7.52e-03 |
Pos logit contributions | |
rob | +0.255 |
zag | +0.239 |
Hipp | +0.238 |
Hero | +0.227 |
girls | +0.225 |
Neg logit contributions | |
vel | -0.129 |
makeup | -0.129 |
face | -0.127 |
spit | -0.125 |
blush | -0.116 |
Token | there |
Token ID | 612 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
vel | +0.043 |
makeup | +0.042 |
face | +0.042 |
spit | +0.042 |
blush | +0.040 |
Neg logit contributions | |
rob | -0.080 |
zag | -0.079 |
Hipp | -0.078 |
steep | -0.075 |
Zig | -0.075 |
Token | was |
Token ID | 373 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
rob | +0.062 |
zag | +0.059 |
Hipp | +0.058 |
Zig | +0.056 |
Hero | +0.054 |
Neg logit contributions | |
vel | -0.056 |
makeup | -0.054 |
face | -0.053 |
spit | -0.053 |
blush | -0.052 |
Token | an |
Token ID | 281 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
rob | +0.043 |
Hipp | +0.041 |
zag | +0.040 |
Zig | +0.040 |
Hero | +0.039 |
Neg logit contributions | |
vel | -0.024 |
face | -0.023 |
spit | -0.022 |
makeup | -0.022 |
blush | -0.021 |
Token | empty |
Token ID | 6565 |
Feature activation | -6.18e-02 |
Pos logit contributions | |
Hipp | +0.043 |
rob | +0.042 |
steep | +0.039 |
Hero | +0.038 |
zag | +0.038 |
Neg logit contributions | |
vel | -0.041 |
blush | -0.038 |
makeup | -0.038 |
spit | -0.037 |
prise | -0.037 |
Token | circle |
Token ID | 9197 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
vel | +0.094 |
spit | +0.090 |
makeup | +0.090 |
face | +0.089 |
blush | +0.085 |
Neg logit contributions | |
rob | -0.115 |
zag | -0.107 |
Hipp | -0.107 |
Hero | -0.101 |
steep | -0.100 |
Token | . |
Token ID | 13 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
rob | +0.047 |
zag | +0.045 |
Hipp | +0.044 |
steep | +0.042 |
Hero | +0.042 |
Neg logit contributions | |
vel | -0.022 |
spit | -0.021 |
makeup | -0.021 |
face | -0.020 |
prise | -0.020 |
Token | Nothing |
Token ID | 10528 |
Feature activation | +5.99e-02 |
Pos logit contributions | |
makeup | +0.029 |
face | +0.028 |
vel | +0.028 |
spit | +0.028 |
blush | +0.026 |
Neg logit contributions | |
rob | -0.037 |
zag | -0.034 |
Hipp | -0.032 |
girls | -0.031 |
Hero | -0.031 |
Token | was |
Token ID | 373 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
rob | +0.138 |
zag | +0.125 |
Hipp | +0.123 |
girls | +0.122 |
Hero | +0.117 |
Neg logit contributions | |
makeup | -0.070 |
face | -0.070 |
spit | -0.068 |
dust | -0.065 |
her | -0.065 |
Token | inside |
Token ID | 2641 |
Feature activation | +4.08e-02 |
Pos logit contributions | |
rob | +0.045 |
zag | +0.041 |
Hipp | +0.041 |
girls | +0.040 |
Hero | +0.039 |
Neg logit contributions | |
makeup | -0.021 |
face | -0.021 |
vel | -0.020 |
spit | -0.020 |
blush | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | +3.01e-02 |
Pos logit contributions | |
rob | +0.077 |
Hipp | +0.074 |
zag | +0.074 |
steep | +0.073 |
Hero | +0.071 |
Neg logit contributions | |
vel | -0.056 |
makeup | -0.054 |
spit | -0.054 |
blush | -0.052 |
face | -0.051 |
Token | circle |
Token ID | 9197 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
zag | +0.049 |
steep | +0.048 |
rob | +0.047 |
Hipp | +0.046 |
Hero | +0.044 |
Neg logit contributions | |
vel | -0.050 |
spit | -0.046 |
Her | -0.045 |
She | -0.043 |
makeup | -0.043 |
Token | Slow |
Token ID | 36423 |
Feature activation | +4.92e-02 |
Pos logit contributions | |
rob | +0.008 |
zag | +0.007 |
Hipp | +0.007 |
Hero | +0.007 |
Zig | +0.007 |
Neg logit contributions | |
face | -0.006 |
makeup | -0.006 |
spit | -0.005 |
vel | -0.005 |
blush | -0.005 |
Token | ly |
Token ID | 306 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
rob | +0.143 |
zag | +0.134 |
Hipp | +0.132 |
girls | +0.129 |
Hero | +0.126 |
Neg logit contributions | |
face | -0.028 |
makeup | -0.026 |
vel | -0.024 |
spit | -0.022 |
blush | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
vel | +0.154 |
makeup | +0.153 |
face | +0.149 |
spit | +0.148 |
blush | +0.138 |
Neg logit contributions | |
rob | -0.294 |
zag | -0.278 |
Hipp | -0.276 |
Hero | -0.266 |
Zig | -0.262 |
Token | the |
Token ID | 262 |
Feature activation | +3.66e-02 |
Pos logit contributions | |
face | +0.038 |
makeup | +0.037 |
vel | +0.036 |
spit | +0.036 |
blush | +0.034 |
Neg logit contributions | |
rob | -0.070 |
zag | -0.065 |
Hipp | -0.063 |
girls | -0.061 |
Hero | -0.061 |
Token | mom |
Token ID | 1995 |
Feature activation | -3.94e-02 |
Pos logit contributions | |
rob | +0.081 |
zag | +0.077 |
Hipp | +0.077 |
Hero | +0.074 |
steep | +0.073 |
Neg logit contributions | |
vel | -0.042 |
makeup | -0.040 |
face | -0.039 |
spit | -0.039 |
blush | -0.036 |
Token | my |
Token ID | 1820 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
vel | +0.036 |
makeup | +0.035 |
spit | +0.034 |
face | +0.034 |
blush | +0.030 |
Neg logit contributions | |
rob | -0.096 |
zag | -0.094 |
Hipp | -0.091 |
Hero | -0.088 |
girls | -0.088 |
Token | got |
Token ID | 1392 |
Feature activation | -7.63e-02 |
Pos logit contributions | |
face | +0.116 |
vel | +0.114 |
spit | +0.112 |
makeup | +0.109 |
dust | +0.102 |
Neg logit contributions | |
rob | -0.249 |
Hipp | -0.227 |
zag | -0.226 |
steep | -0.218 |
Hero | -0.215 |
Token | back |
Token ID | 736 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
vel | +0.100 |
makeup | +0.093 |
face | +0.092 |
spit | +0.092 |
blush | +0.087 |
Neg logit contributions | |
rob | -0.151 |
Hipp | -0.150 |
zag | -0.142 |
Hero | -0.141 |
aboard | -0.141 |
Token | to |
Token ID | 284 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
rob | +0.035 |
zag | +0.033 |
Hipp | +0.032 |
Hero | +0.031 |
girls | +0.031 |
Neg logit contributions | |
vel | -0.016 |
makeup | -0.016 |
face | -0.016 |
spit | -0.015 |
blush | -0.014 |
Token | work |
Token ID | 670 |
Feature activation | -6.25e-03 |
Pos logit contributions | |
rob | +0.023 |
Hipp | +0.022 |
zag | +0.021 |
Zig | +0.021 |
Hero | +0.021 |
Neg logit contributions | |
vel | -0.015 |
makeup | -0.014 |
spit | -0.013 |
face | -0.013 |
blush | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
vel | +0.006 |
makeup | +0.006 |
face | +0.006 |
spit | +0.006 |
blush | +0.006 |
Neg logit contributions | |
rob | -0.015 |
Hipp | -0.014 |
zag | -0.014 |
Hero | -0.013 |
girls | -0.013 |
Token | scared |
Token ID | 12008 |
Feature activation | +5.28e-02 |
Pos logit contributions | |
vel | +0.038 |
prise | +0.036 |
makeup | +0.035 |
blush | +0.033 |
ime | +0.032 |
Neg logit contributions | |
Hipp | -0.046 |
steep | -0.045 |
Hero | -0.042 |
Zig | -0.042 |
rob | -0.042 |
Token | when |
Token ID | 618 |
Feature activation | +9.54e-03 |
Pos logit contributions | |
steep | +0.082 |
Hipp | +0.082 |
rob | +0.078 |
Zig | +0.078 |
zag | +0.075 |
Neg logit contributions | |
prise | -0.093 |
spit | -0.090 |
vel | -0.090 |
makeup | -0.088 |
blush | -0.088 |
Token | he |
Token ID | 339 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
rob | +0.017 |
zag | +0.015 |
Hipp | +0.015 |
girls | +0.014 |
Hero | +0.014 |
Neg logit contributions | |
face | -0.016 |
makeup | -0.016 |
spit | -0.015 |
vel | -0.015 |
her | -0.015 |
Token | saw |
Token ID | 2497 |
Feature activation | -8.79e-02 |
Pos logit contributions | |
rob | +0.045 |
Hipp | +0.044 |
zag | +0.043 |
steep | +0.042 |
Zig | +0.042 |
Neg logit contributions | |
vel | -0.030 |
makeup | -0.029 |
face | -0.027 |
prise | -0.026 |
blush | -0.026 |
Token | a |
Token ID | 257 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
makeup | +0.089 |
face | +0.087 |
vel | +0.086 |
spit | +0.085 |
blush | +0.077 |
Neg logit contributions | |
rob | -0.209 |
zag | -0.198 |
Hipp | -0.194 |
girls | -0.186 |
Hero | -0.186 |
Token | big |
Token ID | 1263 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
vel | +0.034 |
makeup | +0.031 |
blush | +0.030 |
spit | +0.030 |
prise | +0.029 |
Neg logit contributions | |
steep | -0.053 |
Hipp | -0.053 |
zag | -0.053 |
rob | -0.052 |
Hero | -0.050 |
Token | bird |
Token ID | 6512 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
vel | +0.164 |
spit | +0.150 |
makeup | +0.149 |
prise | +0.148 |
Her | +0.147 |
Neg logit contributions | |
rob | -0.187 |
Hipp | -0.179 |
steep | -0.174 |
zag | -0.171 |
Hero | -0.168 |
Token | nearby |
Token ID | 6716 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
zag | +0.034 |
rob | +0.033 |
Hipp | +0.032 |
Zig | +0.030 |
ousel | +0.030 |
Neg logit contributions | |
makeup | -0.023 |
vel | -0.023 |
prise | -0.022 |
spit | -0.022 |
blush | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
vel | +0.031 |
makeup | +0.030 |
prise | +0.029 |
face | +0.028 |
spit | +0.028 |
Neg logit contributions | |
rob | -0.058 |
Hipp | -0.056 |
zag | -0.055 |
Zig | -0.053 |
Hero | -0.053 |
Token | |
Token ID | 220 |
Feature activation | +8.16e-03 |
Pos logit contributions | |
face | +0.025 |
spit | +0.023 |
makeup | +0.022 |
hand | +0.022 |
her | +0.021 |
Neg logit contributions | |
rob | -0.044 |
zag | -0.040 |
girls | -0.037 |
habitats | -0.036 |
steep | -0.035 |
Token |
|
Token ID | 198 |
Feature activation | -1.41e-03 |
Pos logit contributions | |
rob | +0.017 |
Hipp | +0.016 |
zag | +0.016 |
Hero | +0.015 |
girls | +0.015 |
Neg logit contributions | |
vel | -0.011 |
face | -0.010 |
makeup | -0.010 |
spit | -0.010 |
She | -0.010 |
Token | park |
Token ID | 3952 |
Feature activation | +9.39e-03 |
Pos logit contributions | |
rob | +0.055 |
zag | +0.051 |
Hipp | +0.050 |
steep | +0.049 |
Hero | +0.048 |
Neg logit contributions | |
vel | -0.050 |
spit | -0.047 |
face | -0.044 |
blush | -0.044 |
makeup | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
rob | +0.022 |
zag | +0.021 |
Hipp | +0.021 |
Hero | +0.020 |
steep | +0.020 |
Neg logit contributions | |
vel | -0.010 |
makeup | -0.009 |
spit | -0.009 |
face | -0.009 |
blush | -0.008 |
Token | They |
Token ID | 1119 |
Feature activation | +9.73e-03 |
Pos logit contributions | |
face | +0.032 |
spit | +0.031 |
her | +0.031 |
makeup | +0.030 |
She | +0.030 |
Neg logit contributions | |
rob | -0.040 |
zag | -0.036 |
girls | -0.034 |
Hipp | -0.032 |
Hero | -0.032 |
Token | saw |
Token ID | 2497 |
Feature activation | -9.08e-02 |
Pos logit contributions | |
rob | +0.020 |
Hipp | +0.019 |
zag | +0.019 |
steep | +0.018 |
Zig | +0.018 |
Neg logit contributions | |
vel | -0.013 |
makeup | -0.013 |
face | -0.012 |
spit | -0.012 |
prise | -0.012 |
Token | a |
Token ID | 257 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
vel | +0.091 |
makeup | +0.087 |
face | +0.086 |
spit | +0.085 |
blush | +0.078 |
Neg logit contributions | |
rob | -0.217 |
Hipp | -0.203 |
zag | -0.203 |
Hero | -0.196 |
Zig | -0.194 |
Token | big |
Token ID | 1263 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
vel | +0.033 |
spit | +0.029 |
makeup | +0.029 |
blush | +0.028 |
face | +0.028 |
Neg logit contributions | |
rob | -0.052 |
zag | -0.050 |
Hipp | -0.050 |
steep | -0.050 |
Hero | -0.047 |
Token | tree |
Token ID | 5509 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
vel | +0.171 |
spit | +0.162 |
makeup | +0.157 |
face | +0.154 |
prise | +0.152 |
Neg logit contributions | |
rob | -0.181 |
Hipp | -0.166 |
zag | -0.163 |
steep | -0.161 |
Hero | -0.156 |
Token | with |
Token ID | 351 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
rob | +0.031 |
zag | +0.030 |
Hipp | +0.029 |
steep | +0.029 |
Zig | +0.028 |
Neg logit contributions | |
vel | -0.014 |
makeup | -0.014 |
spit | -0.013 |
face | -0.013 |
blush | -0.012 |
Token | many |
Token ID | 867 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
vel | +0.018 |
prise | +0.016 |
face | +0.015 |
makeup | +0.015 |
spit | +0.015 |
Neg logit contributions | |
Hipp | -0.019 |
rob | -0.019 |
steep | -0.018 |
Zig | -0.018 |
tigers | -0.017 |
Token | branches |
Token ID | 13737 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
vel | +0.050 |
prise | +0.047 |
makeup | +0.047 |
spit | +0.046 |
face | +0.046 |
Neg logit contributions | |
rob | -0.048 |
steep | -0.045 |
Hipp | -0.045 |
zag | -0.043 |
Zig | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
rob | +0.026 |
zag | +0.025 |
Hipp | +0.025 |
Hero | +0.024 |
girls | +0.023 |
Neg logit contributions | |
vel | -0.010 |
makeup | -0.010 |
face | -0.010 |
spit | -0.010 |
blush | -0.009 |
Token | After |
Token ID | 2293 |
Feature activation | +4.83e-02 |
Pos logit contributions | |
face | +0.033 |
spit | +0.031 |
makeup | +0.030 |
clean | +0.027 |
blush | +0.026 |
Neg logit contributions | |
rob | -0.049 |
zag | -0.044 |
girls | -0.040 |
Hipp | -0.039 |
field | -0.039 |
Token | a |
Token ID | 257 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
rob | +0.111 |
Hipp | +0.105 |
zag | +0.104 |
Hero | +0.101 |
Zig | +0.100 |
Neg logit contributions | |
vel | -0.052 |
makeup | -0.050 |
face | -0.050 |
spit | -0.049 |
blush | -0.046 |
Token | while |
Token ID | 981 |
Feature activation | -4.96e-02 |
Pos logit contributions | |
vel | +0.038 |
makeup | +0.037 |
face | +0.036 |
spit | +0.036 |
blush | +0.034 |
Neg logit contributions | |
rob | -0.056 |
zag | -0.053 |
Hipp | -0.053 |
Hero | -0.050 |
steep | -0.050 |
Token | , |
Token ID | 11 |
Feature activation | -4.00e-02 |
Pos logit contributions | |
face | +0.042 |
makeup | +0.040 |
vel | +0.039 |
spit | +0.038 |
her | +0.035 |
Neg logit contributions | |
rob | -0.129 |
zag | -0.119 |
Hipp | -0.116 |
girls | -0.114 |
Hero | -0.113 |
Token | Tim |
Token ID | 5045 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
face | +0.047 |
her | +0.046 |
spit | +0.045 |
makeup | +0.045 |
clean | +0.040 |
Neg logit contributions | |
rob | -0.092 |
zag | -0.087 |
girls | -0.081 |
habitats | -0.079 |
Hipp | -0.078 |
Token | my |
Token ID | 1820 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
vel | +0.048 |
makeup | +0.048 |
spit | +0.046 |
face | +0.044 |
blush | +0.043 |
Neg logit contributions | |
rob | -0.113 |
zag | -0.113 |
Hero | -0.104 |
Hipp | -0.103 |
girls | -0.103 |
Token | saw |
Token ID | 2497 |
Feature activation | -9.27e-02 |
Pos logit contributions | |
vel | +0.122 |
makeup | +0.121 |
face | +0.117 |
spit | +0.115 |
blush | +0.109 |
Neg logit contributions | |
rob | -0.246 |
zag | -0.232 |
Hipp | -0.230 |
Hero | -0.221 |
Zig | -0.218 |
Token | the |
Token ID | 262 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
face | +0.071 |
makeup | +0.068 |
spit | +0.064 |
her | +0.063 |
dust | +0.055 |
Neg logit contributions | |
rob | -0.248 |
zag | -0.245 |
Hipp | -0.225 |
girls | -0.224 |
steep | -0.222 |
Token | dog |
Token ID | 3290 |
Feature activation | -3.81e-02 |
Pos logit contributions | |
rob | +0.053 |
zag | +0.049 |
Hipp | +0.048 |
Zig | +0.046 |
girls | +0.046 |
Neg logit contributions | |
face | -0.029 |
makeup | -0.029 |
spit | -0.028 |
vel | -0.028 |
blush | -0.025 |
Token | hiding |
Token ID | 11816 |
Feature activation | +5.18e-02 |
Pos logit contributions | |
vel | +0.048 |
makeup | +0.047 |
spit | +0.044 |
prise | +0.044 |
face | +0.044 |
Neg logit contributions | |
rob | -0.076 |
Hipp | -0.076 |
zag | -0.076 |
Hero | -0.073 |
Zig | -0.072 |
Token | behind |
Token ID | 2157 |
Feature activation | +6.24e-02 |
Pos logit contributions | |
rob | +0.116 |
zag | +0.109 |
Hipp | +0.108 |
Hero | +0.103 |
girls | +0.102 |
Neg logit contributions | |
vel | -0.059 |
makeup | -0.059 |
face | -0.058 |
spit | -0.057 |
blush | -0.053 |
Token | hungry |
Token ID | 14720 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
vel | +0.011 |
prise | +0.011 |
makeup | +0.010 |
blush | +0.010 |
envelope | +0.010 |
Neg logit contributions | |
steep | -0.011 |
Hipp | -0.010 |
zag | -0.010 |
Zig | -0.009 |
Hero | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -9.24e-03 |
Pos logit contributions | |
prise | +0.034 |
vel | +0.032 |
spit | +0.031 |
blush | +0.030 |
makeup | +0.029 |
Neg logit contributions | |
rob | -0.025 |
steep | -0.024 |
zag | -0.023 |
Hipp | -0.022 |
Hero | -0.022 |
Token | She |
Token ID | 1375 |
Feature activation | -3.50e-02 |
Pos logit contributions | |
vel | +0.013 |
prise | +0.013 |
spit | +0.012 |
makeup | +0.011 |
aming | +0.011 |
Neg logit contributions | |
Hipp | -0.017 |
zag | -0.017 |
Zig | -0.017 |
rob | -0.017 |
Hero | -0.016 |
Token | wanted |
Token ID | 2227 |
Feature activation | -7.18e-02 |
Pos logit contributions | |
vel | +0.050 |
makeup | +0.048 |
face | +0.045 |
prise | +0.044 |
blush | +0.044 |
Neg logit contributions | |
rob | -0.067 |
Hipp | -0.064 |
zag | -0.063 |
Hero | -0.061 |
Zig | -0.060 |
Token | to |
Token ID | 284 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
makeup | +0.099 |
face | +0.097 |
vel | +0.097 |
spit | +0.096 |
blush | +0.090 |
Neg logit contributions | |
rob | -0.145 |
zag | -0.135 |
Hipp | -0.133 |
Hero | -0.126 |
girls | -0.126 |
Token | eat |
Token ID | 4483 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
Hipp | +0.025 |
Hero | +0.023 |
Claw | +0.023 |
zag | +0.023 |
Zig | +0.023 |
Neg logit contributions | |
vel | -0.024 |
aming | -0.019 |
makeup | -0.019 |
pled | -0.019 |
prise | -0.018 |
Token | a |
Token ID | 257 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
vel | +0.141 |
prise | +0.129 |
face | +0.127 |
spit | +0.126 |
ime | +0.125 |
Neg logit contributions | |
rob | -0.223 |
Hipp | -0.216 |
Zig | -0.202 |
Hero | -0.201 |
steep | -0.194 |
Token | snack |
Token ID | 26906 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
vel | +0.022 |
makeup | +0.021 |
spit | +0.021 |
face | +0.021 |
blush | +0.020 |
Neg logit contributions | |
rob | -0.030 |
Hipp | -0.028 |
zag | -0.028 |
Hero | -0.027 |
steep | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
vel | +0.046 |
spit | +0.044 |
prise | +0.042 |
makeup | +0.042 |
face | +0.042 |
Neg logit contributions | |
rob | -0.061 |
zag | -0.060 |
steep | -0.056 |
Hipp | -0.056 |
Hero | -0.055 |
Token | but |
Token ID | 475 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
vel | +0.047 |
prise | +0.046 |
ime | +0.043 |
aming | +0.041 |
makeup | +0.041 |
Neg logit contributions | |
steep | -0.035 |
Hipp | -0.035 |
Zig | -0.035 |
rob | -0.034 |
Hero | -0.033 |
Token | she |
Token ID | 673 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
vel | +0.018 |
makeup | +0.015 |
face | +0.015 |
spit | +0.015 |
prise | +0.014 |
Neg logit contributions | |
rob | -0.050 |
Hipp | -0.049 |
Zig | -0.048 |
zag | -0.048 |
steep | -0.047 |