Token | with |
Token ID | 351 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
unison | +0.015 |
slices | +0.014 |
seventeen | +0.014 |
amaz | +0.014 |
aroma | +0.014 |
Neg logit contributions | |
wear | -0.009 |
Woo | -0.009 |
bath | -0.008 |
Mitt | -0.008 |
day | -0.008 |
Token | lots |
Token ID | 6041 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
slices | +0.033 |
unison | +0.033 |
amaz | +0.033 |
seventeen | +0.032 |
seconds | +0.032 |
Neg logit contributions | |
wear | -0.019 |
Woo | -0.018 |
bath | -0.017 |
burn | -0.016 |
skipping | -0.016 |
Token | of |
Token ID | 286 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
slices | +0.040 |
unison | +0.039 |
seconds | +0.039 |
amaz | +0.038 |
seventeen | +0.038 |
Neg logit contributions | |
wear | -0.024 |
Woo | -0.023 |
burn | -0.021 |
bath | -0.021 |
Mitt | -0.020 |
Token | other |
Token ID | 584 |
Feature activation | -7.72e-03 |
Pos logit contributions | |
wear | +0.024 |
burn | +0.024 |
Woo | +0.023 |
bath | +0.021 |
stub | +0.021 |
Neg logit contributions | |
slices | -0.023 |
amaz | -0.022 |
aroma | -0.022 |
seconds | -0.022 |
unison | -0.022 |
Token | b |
Token ID | 275 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
slices | +0.013 |
unison | +0.013 |
amaz | +0.013 |
aroma | +0.013 |
fires | +0.013 |
Neg logit contributions | |
wear | -0.012 |
Woo | -0.011 |
bath | -0.011 |
burn | -0.011 |
Lucky | -0.010 |
Token | unn |
Token ID | 20935 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
wear | +0.030 |
Woo | +0.028 |
bath | +0.026 |
Lucky | +0.025 |
Mitt | +0.024 |
Neg logit contributions | |
slices | -0.069 |
pulp | -0.065 |
seventeen | -0.065 |
fires | -0.064 |
pter | -0.064 |
Token | ies |
Token ID | 444 |
Feature activation | +3.02e-02 |
Pos logit contributions | |
wear | +0.159 |
Woo | +0.152 |
bath | +0.147 |
Lucky | +0.137 |
Mitt | +0.137 |
Neg logit contributions | |
slices | -0.195 |
pulp | -0.184 |
pter | -0.184 |
fires | -0.184 |
unison | -0.184 |
Token | . |
Token ID | 13 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
wear | +0.047 |
Woo | +0.045 |
bath | +0.043 |
burn | +0.043 |
Mitt | +0.042 |
Neg logit contributions | |
slices | -0.050 |
unison | -0.049 |
fires | -0.048 |
seconds | -0.047 |
pter | -0.047 |
Token | George |
Token ID | 4502 |
Feature activation | +2.83e-02 |
Pos logit contributions | |
wear | +0.018 |
Woo | +0.016 |
bath | +0.016 |
Mitt | +0.015 |
Lucky | +0.014 |
Neg logit contributions | |
unison | -0.036 |
slices | -0.036 |
seventeen | -0.036 |
amaz | -0.035 |
aroma | -0.035 |
Token | was |
Token ID | 373 |
Feature activation | +9.30e-03 |
Pos logit contributions | |
Woo | +0.043 |
wear | +0.042 |
burn | +0.042 |
bath | +0.041 |
Mitt | +0.039 |
Neg logit contributions | |
unison | -0.047 |
slices | -0.046 |
fires | -0.046 |
pter | -0.045 |
seconds | -0.045 |
Token | always |
Token ID | 1464 |
Feature activation | +3.11e-02 |
Pos logit contributions | |
burn | +0.016 |
wear | +0.016 |
ibi | +0.016 |
ogging | +0.016 |
ongo | +0.016 |
Neg logit contributions | |
seconds | -0.012 |
amaz | -0.011 |
seventeen | -0.011 |
thirteen | -0.011 |
fires | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | +6.98e-03 |
Pos logit contributions | |
wear | +0.049 |
Woo | +0.048 |
burn | +0.046 |
bath | +0.044 |
important | +0.043 |
Neg logit contributions | |
amaz | -0.047 |
unison | -0.046 |
slices | -0.046 |
seconds | -0.044 |
aroma | -0.043 |
Token | a |
Token ID | 257 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
wear | +0.013 |
burn | +0.012 |
ongo | +0.011 |
Woo | +0.011 |
jog | +0.011 |
Neg logit contributions | |
slices | -0.011 |
unison | -0.010 |
conf | -0.010 |
amaz | -0.009 |
seconds | -0.009 |
Token | family |
Token ID | 1641 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
ained | +0.029 |
wear | +0.029 |
Woo | +0.027 |
burn | +0.027 |
scar | +0.026 |
Neg logit contributions | |
unison | -0.035 |
aroma | -0.033 |
slices | -0.033 |
amaz | -0.033 |
pulp | -0.032 |
Token | of |
Token ID | 286 |
Feature activation | +4.28e-02 |
Pos logit contributions | |
slices | +0.040 |
unison | +0.038 |
pter | +0.037 |
seconds | +0.036 |
candles | +0.035 |
Neg logit contributions | |
Lucky | -0.046 |
Woo | -0.045 |
scar | -0.043 |
ibi | -0.043 |
Nut | -0.043 |
Token | b |
Token ID | 275 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
wear | +0.070 |
Woo | +0.064 |
bath | +0.063 |
burn | +0.063 |
scar | +0.062 |
Neg logit contributions | |
unison | -0.067 |
slices | -0.066 |
seconds | -0.066 |
pulp | -0.065 |
conf | -0.064 |
Token | unn |
Token ID | 20935 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
wear | +0.044 |
Woo | +0.041 |
bath | +0.039 |
Lucky | +0.039 |
Mitt | +0.037 |
Neg logit contributions | |
slices | -0.083 |
pulp | -0.079 |
pter | -0.077 |
als | -0.077 |
seventeen | -0.076 |
Token | ies |
Token ID | 444 |
Feature activation | +3.60e-02 |
Pos logit contributions | |
wear | +0.166 |
Woo | +0.155 |
bath | +0.153 |
Lucky | +0.144 |
Mitt | +0.142 |
Neg logit contributions | |
slices | -0.184 |
pter | -0.176 |
pulp | -0.175 |
unison | -0.175 |
fires | -0.175 |
Token | . |
Token ID | 13 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
wear | +0.061 |
Woo | +0.058 |
ibi | +0.057 |
bath | +0.057 |
scar | +0.056 |
Neg logit contributions | |
unison | -0.052 |
slices | -0.049 |
fires | -0.048 |
seconds | -0.047 |
pter | -0.047 |
Token | The |
Token ID | 383 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
wear | +0.026 |
Woo | +0.024 |
important | +0.024 |
burn | +0.024 |
bath | +0.024 |
Neg logit contributions | |
unison | -0.020 |
slices | -0.019 |
seventeen | -0.019 |
amaz | -0.018 |
aroma | -0.018 |
Token | b |
Token ID | 275 |
Feature activation | +3.03e-02 |
Pos logit contributions | |
wear | +0.032 |
scar | +0.031 |
ogging | +0.030 |
Razor | +0.029 |
ongo | +0.029 |
Neg logit contributions | |
aroma | -0.038 |
slices | -0.038 |
unison | -0.037 |
pulp | -0.037 |
amaz | -0.036 |
Token | unn |
Token ID | 20935 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
wear | +0.037 |
bath | +0.035 |
Woo | +0.034 |
jog | +0.034 |
Lucky | +0.034 |
Neg logit contributions | |
als | -0.060 |
slices | -0.059 |
pulp | -0.058 |
edly | -0.057 |
pter | -0.056 |
Token | my |
Token ID | 1820 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
slices | +0.052 |
aroma | +0.050 |
amaz | +0.049 |
unison | +0.049 |
gasped | +0.049 |
Neg logit contributions | |
Woo | -0.029 |
bath | -0.029 |
wear | -0.028 |
jog | -0.025 |
exercise | -0.024 |
Token | told |
Token ID | 1297 |
Feature activation | -1.75e-03 |
Pos logit contributions | |
bath | +0.026 |
wear | +0.025 |
Woo | +0.025 |
jog | +0.023 |
ained | +0.023 |
Neg logit contributions | |
slices | -0.024 |
aroma | -0.023 |
unison | -0.023 |
seconds | -0.023 |
gasped | -0.022 |
Token | her |
Token ID | 607 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
unison | +0.004 |
slices | +0.004 |
seventeen | +0.004 |
amaz | +0.004 |
fires | +0.004 |
Neg logit contributions | |
wear | -0.001 |
Woo | -0.001 |
bath | -0.001 |
Mitt | -0.001 |
Lucky | -0.001 |
Token | that |
Token ID | 326 |
Feature activation | +5.34e-02 |
Pos logit contributions | |
wear | +0.035 |
Woo | +0.033 |
bath | +0.032 |
burn | +0.030 |
Mitt | +0.030 |
Neg logit contributions | |
slices | -0.045 |
unison | -0.044 |
amaz | -0.044 |
seconds | -0.044 |
seventeen | -0.044 |
Token | b |
Token ID | 275 |
Feature activation | +2.44e-02 |
Pos logit contributions | |
wear | +0.082 |
bath | +0.074 |
ibi | +0.074 |
ongo | +0.074 |
Woo | +0.073 |
Neg logit contributions | |
slices | -0.091 |
seconds | -0.086 |
pulp | -0.086 |
amaz | -0.086 |
conf | -0.085 |
Token | unn |
Token ID | 20935 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
wear | +0.028 |
bath | +0.025 |
Woo | +0.025 |
jog | +0.024 |
skipping | +0.023 |
Neg logit contributions | |
slices | -0.053 |
als | -0.051 |
pulp | -0.050 |
rs | -0.050 |
seconds | -0.050 |
Token | ies |
Token ID | 444 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
wear | +0.156 |
Woo | +0.148 |
bath | +0.144 |
Mitt | +0.134 |
skipping | +0.134 |
Neg logit contributions | |
slices | -0.190 |
unison | -0.185 |
pulp | -0.184 |
pter | -0.182 |
aroma | -0.181 |
Token | can |
Token ID | 460 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
wear | +0.057 |
Woo | +0.057 |
bath | +0.055 |
burn | +0.053 |
Mitt | +0.053 |
Neg logit contributions | |
slices | -0.039 |
unison | -0.039 |
fires | -0.038 |
aroma | -0.038 |
amaz | -0.037 |
Token | be |
Token ID | 307 |
Feature activation | +4.11e-03 |
Pos logit contributions | |
Woo | +0.031 |
important | +0.030 |
scar | +0.030 |
bath | +0.029 |
day | +0.029 |
Neg logit contributions | |
unison | -0.032 |
seventeen | -0.032 |
als | -0.031 |
slices | -0.031 |
aroma | -0.030 |
Token | scared |
Token ID | 12008 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
wear | +0.005 |
bath | +0.005 |
scar | +0.005 |
ibi | +0.005 |
burn | +0.005 |
Neg logit contributions | |
slices | -0.007 |
seconds | -0.007 |
unison | -0.007 |
pulp | -0.007 |
aroma | -0.007 |
Token | easily |
Token ID | 3538 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
wear | +0.056 |
Woo | +0.052 |
burn | +0.049 |
bath | +0.049 |
scar | +0.048 |
Neg logit contributions | |
slices | -0.078 |
seconds | -0.077 |
unison | -0.076 |
amaz | -0.074 |
fires | -0.074 |
Token | there |
Token ID | 612 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
wear | +0.032 |
Woo | +0.031 |
bath | +0.029 |
burn | +0.029 |
Mitt | +0.028 |
Neg logit contributions | |
slices | -0.035 |
unison | -0.035 |
seventeen | -0.034 |
amaz | -0.033 |
seconds | -0.033 |
Token | were |
Token ID | 547 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
wear | +0.053 |
Woo | +0.052 |
burn | +0.050 |
bath | +0.048 |
important | +0.046 |
Neg logit contributions | |
amaz | -0.055 |
slices | -0.054 |
unison | -0.054 |
seconds | -0.052 |
seventeen | -0.051 |
Token | two |
Token ID | 734 |
Feature activation | +3.09e-02 |
Pos logit contributions | |
slices | +0.035 |
conf | +0.032 |
seventeen | +0.032 |
pulp | +0.031 |
unison | +0.031 |
Neg logit contributions | |
wear | -0.033 |
Woo | -0.028 |
jog | -0.027 |
burn | -0.027 |
ibi | -0.027 |
Token | little |
Token ID | 1310 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
wear | +0.036 |
Woo | +0.034 |
bath | +0.031 |
burn | +0.029 |
Lucky | +0.028 |
Neg logit contributions | |
slices | -0.067 |
unison | -0.066 |
seventeen | -0.064 |
amaz | -0.064 |
seconds | -0.063 |
Token | b |
Token ID | 275 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
wear | +0.056 |
Woo | +0.052 |
bath | +0.052 |
Mitt | +0.049 |
Lucky | +0.048 |
Neg logit contributions | |
unison | -0.058 |
slices | -0.058 |
seventeen | -0.058 |
amaz | -0.057 |
fires | -0.056 |
Token | unn |
Token ID | 20935 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
wear | +0.042 |
Woo | +0.039 |
bath | +0.038 |
Lucky | +0.035 |
jog | +0.035 |
Neg logit contributions | |
slices | -0.077 |
pulp | -0.073 |
seventeen | -0.071 |
pter | -0.071 |
als | -0.069 |
Token | ies |
Token ID | 444 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
wear | +0.171 |
Woo | +0.163 |
bath | +0.160 |
jog | +0.150 |
Lucky | +0.147 |
Neg logit contributions | |
slices | -0.172 |
pter | -0.163 |
pulp | -0.163 |
fires | -0.158 |
seconds | -0.157 |
Token | . |
Token ID | 13 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
wear | +0.054 |
Woo | +0.052 |
ibi | +0.050 |
bath | +0.050 |
scar | +0.048 |
Neg logit contributions | |
unison | -0.048 |
slices | -0.047 |
pter | -0.044 |
seconds | -0.043 |
ah | -0.042 |
Token | They |
Token ID | 1119 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
wear | +0.025 |
important | +0.023 |
burn | +0.022 |
bath | +0.022 |
stub | +0.022 |
Neg logit contributions | |
unison | -0.031 |
slices | -0.030 |
pter | -0.029 |
seventeen | -0.029 |
pulp | -0.028 |
Token | were |
Token ID | 547 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
wear | +0.020 |
Woo | +0.019 |
bath | +0.018 |
Mitt | +0.017 |
Lucky | +0.017 |
Neg logit contributions | |
unison | -0.020 |
slices | -0.020 |
seventeen | -0.020 |
amaz | -0.019 |
aroma | -0.019 |
Token | best |
Token ID | 1266 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
slices | +0.030 |
ah | +0.029 |
tur | +0.029 |
candles | +0.029 |
pulp | +0.028 |
Neg logit contributions | |
wear | -0.028 |
ogging | -0.028 |
ongo | -0.027 |
jog | -0.027 |
ibi | -0.027 |
Token | Do |
Token ID | 5211 |
Feature activation | +3.03e-03 |
Pos logit contributions | |
amaz | +0.062 |
unison | +0.062 |
slices | +0.061 |
aroma | +0.061 |
seventeen | +0.060 |
Neg logit contributions | |
wear | -0.058 |
Woo | -0.056 |
jog | -0.052 |
bath | -0.051 |
day | -0.051 |
Token | you |
Token ID | 345 |
Feature activation | -3.37e-03 |
Pos logit contributions | |
wear | +0.004 |
Woo | +0.004 |
bath | +0.004 |
Mitt | +0.004 |
burn | +0.004 |
Neg logit contributions | |
unison | -0.005 |
slices | -0.005 |
seventeen | -0.005 |
amaz | -0.005 |
seconds | -0.005 |
Token | know |
Token ID | 760 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
unison | +0.005 |
seventeen | +0.005 |
seconds | +0.005 |
amaz | +0.005 |
slices | +0.005 |
Neg logit contributions | |
Woo | -0.006 |
bath | -0.005 |
burn | -0.005 |
wear | -0.005 |
Mitt | -0.005 |
Token | what |
Token ID | 644 |
Feature activation | +8.61e-04 |
Pos logit contributions | |
wear | +0.039 |
Woo | +0.038 |
bath | +0.037 |
burn | +0.035 |
Mitt | +0.034 |
Neg logit contributions | |
unison | -0.029 |
slices | -0.029 |
seconds | -0.028 |
seventeen | -0.028 |
amaz | -0.028 |
Token | b |
Token ID | 275 |
Feature activation | +2.31e-02 |
Pos logit contributions | |
wear | +0.001 |
Woo | +0.001 |
bath | +0.001 |
burn | +0.001 |
Mitt | +0.001 |
Neg logit contributions | |
slices | -0.001 |
unison | -0.001 |
seconds | -0.001 |
seventeen | -0.001 |
amaz | -0.001 |
Token | unn |
Token ID | 20935 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
bath | +0.026 |
Woo | +0.025 |
wear | +0.025 |
skipping | +0.024 |
stub | +0.024 |
Neg logit contributions | |
als | -0.047 |
slices | -0.046 |
rs | -0.045 |
teen | -0.045 |
aroma | -0.044 |
Token | ies |
Token ID | 444 |
Feature activation | +4.85e-02 |
Pos logit contributions | |
wear | +0.149 |
Woo | +0.148 |
bath | +0.145 |
Mitt | +0.137 |
skipping | +0.135 |
Neg logit contributions | |
slices | -0.177 |
aroma | -0.175 |
pulp | -0.172 |
fires | -0.170 |
pter | -0.170 |
Token | eat |
Token ID | 4483 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
wear | +0.065 |
Woo | +0.065 |
bath | +0.062 |
Mitt | +0.060 |
burn | +0.059 |
Neg logit contributions | |
slices | -0.088 |
unison | -0.087 |
seventeen | -0.086 |
aroma | -0.086 |
fires | -0.085 |
Token | ?" |
Token ID | 1701 |
Feature activation | +7.29e-03 |
Pos logit contributions | |
wear | +0.026 |
Woo | +0.025 |
bath | +0.023 |
Mitt | +0.022 |
burn | +0.022 |
Neg logit contributions | |
slices | -0.055 |
unison | -0.053 |
seventeen | -0.052 |
aroma | -0.052 |
fires | -0.052 |
Token |
|
Token ID | 198 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
burn | +0.012 |
wear | +0.011 |
important | +0.011 |
bath | +0.010 |
scar | +0.010 |
Neg logit contributions | |
slices | -0.012 |
seconds | -0.012 |
amaz | -0.011 |
unison | -0.011 |
pulp | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | +3.91e-03 |
Pos logit contributions | |
wear | +0.025 |
burn | +0.024 |
important | +0.022 |
Woo | +0.022 |
bath | +0.022 |
Neg logit contributions | |
slices | -0.022 |
unison | -0.022 |
seconds | -0.021 |
seventeen | -0.020 |
fires | -0.020 |
Token | ummy |
Token ID | 13513 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
slices | +0.050 |
unison | +0.048 |
seventeen | +0.048 |
aroma | +0.048 |
amaz | +0.047 |
Neg logit contributions | |
wear | -0.029 |
Woo | -0.027 |
bath | -0.025 |
Lucky | -0.024 |
Mitt | -0.024 |
Token | food |
Token ID | 2057 |
Feature activation | +9.76e-03 |
Pos logit contributions | |
unison | +0.022 |
seventeen | +0.021 |
atively | +0.021 |
fires | +0.020 |
anish | +0.020 |
Neg logit contributions | |
wear | -0.017 |
bath | -0.016 |
day | -0.015 |
exercise | -0.015 |
important | -0.014 |
Token | for |
Token ID | 329 |
Feature activation | -3.84e-02 |
Pos logit contributions | |
wear | +0.011 |
bath | +0.010 |
Woo | +0.010 |
day | +0.009 |
Lucky | +0.009 |
Neg logit contributions | |
unison | -0.021 |
seventeen | -0.021 |
amaz | -0.021 |
slices | -0.021 |
seconds | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
seventeen | +0.085 |
amaz | +0.081 |
unison | +0.080 |
pter | +0.079 |
slices | +0.079 |
Neg logit contributions | |
wear | -0.045 |
bath | -0.041 |
Woo | -0.041 |
day | -0.039 |
Mitt | -0.038 |
Token | b |
Token ID | 275 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
wear | +0.025 |
Woo | +0.023 |
bath | +0.023 |
Mitt | +0.021 |
Lucky | +0.021 |
Neg logit contributions | |
unison | -0.040 |
slices | -0.040 |
seventeen | -0.039 |
amaz | -0.039 |
aroma | -0.039 |
Token | unn |
Token ID | 20935 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
wear | +0.029 |
Woo | +0.027 |
bath | +0.026 |
skipping | +0.024 |
Mitt | +0.024 |
Neg logit contributions | |
slices | -0.046 |
seventeen | -0.044 |
unison | -0.044 |
pulp | -0.043 |
seconds | -0.043 |
Token | ies |
Token ID | 444 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
wear | +0.124 |
Woo | +0.112 |
bath | +0.110 |
day | +0.103 |
important | +0.101 |
Neg logit contributions | |
unison | -0.225 |
seventeen | -0.222 |
amaz | -0.220 |
slices | -0.217 |
aroma | -0.213 |
Token | . |
Token ID | 13 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
wear | +0.040 |
Woo | +0.036 |
bath | +0.035 |
Mitt | +0.033 |
Lucky | +0.033 |
Neg logit contributions | |
unison | -0.065 |
slices | -0.065 |
seventeen | -0.064 |
amaz | -0.064 |
aroma | -0.063 |
Token | It |
Token ID | 632 |
Feature activation | +6.81e-03 |
Pos logit contributions | |
wear | +0.020 |
Woo | +0.019 |
bath | +0.018 |
day | +0.017 |
Mitt | +0.017 |
Neg logit contributions | |
slices | -0.035 |
unison | -0.035 |
seventeen | -0.035 |
amaz | -0.035 |
aroma | -0.034 |
Token | started |
Token ID | 2067 |
Feature activation | +5.14e-02 |
Pos logit contributions | |
wear | +0.008 |
Woo | +0.007 |
bath | +0.007 |
Mitt | +0.007 |
day | +0.007 |
Neg logit contributions | |
unison | -0.014 |
slices | -0.014 |
seventeen | -0.014 |
amaz | -0.014 |
aroma | -0.014 |
Token | with |
Token ID | 351 |
Feature activation | -6.32e-03 |
Pos logit contributions | |
wear | +0.063 |
Woo | +0.058 |
bath | +0.053 |
Lucky | +0.051 |
burn | +0.049 |
Neg logit contributions | |
slices | -0.108 |
seventeen | -0.103 |
seconds | -0.102 |
aroma | -0.102 |
amaz | -0.102 |
Token | was |
Token ID | 373 |
Feature activation | +7.63e-03 |
Pos logit contributions | |
wear | +0.052 |
Woo | +0.052 |
burn | +0.050 |
bath | +0.048 |
important | +0.046 |
Neg logit contributions | |
amaz | -0.052 |
slices | -0.051 |
unison | -0.050 |
seconds | -0.048 |
aroma | -0.048 |
Token | a |
Token ID | 257 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
wear | +0.014 |
burn | +0.013 |
ongo | +0.013 |
ibi | +0.013 |
Woo | +0.013 |
Neg logit contributions | |
slices | -0.011 |
unison | -0.010 |
conf | -0.010 |
seconds | -0.009 |
amaz | -0.009 |
Token | family |
Token ID | 1641 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
wear | +0.032 |
ained | +0.031 |
burn | +0.029 |
Woo | +0.029 |
scar | +0.027 |
Neg logit contributions | |
unison | -0.039 |
slices | -0.038 |
aroma | -0.037 |
seconds | -0.037 |
amaz | -0.036 |
Token | of |
Token ID | 286 |
Feature activation | +4.13e-02 |
Pos logit contributions | |
slices | +0.043 |
unison | +0.040 |
pter | +0.038 |
seconds | +0.037 |
candles | +0.036 |
Neg logit contributions | |
Woo | -0.052 |
Lucky | -0.051 |
wear | -0.050 |
scar | -0.050 |
Nut | -0.049 |
Token | b |
Token ID | 275 |
Feature activation | +3.65e-02 |
Pos logit contributions | |
wear | +0.067 |
Woo | +0.062 |
bath | +0.061 |
burn | +0.059 |
important | +0.058 |
Neg logit contributions | |
unison | -0.068 |
slices | -0.067 |
seconds | -0.065 |
amaz | -0.064 |
pulp | -0.064 |
Token | unn |
Token ID | 20935 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
wear | +0.042 |
Woo | +0.039 |
bath | +0.038 |
Lucky | +0.038 |
Mitt | +0.035 |
Neg logit contributions | |
slices | -0.079 |
pulp | -0.075 |
pter | -0.073 |
als | -0.072 |
seventeen | -0.072 |
Token | ies |
Token ID | 444 |
Feature activation | +3.59e-02 |
Pos logit contributions | |
wear | +0.157 |
Woo | +0.146 |
bath | +0.144 |
Lucky | +0.134 |
Mitt | +0.133 |
Neg logit contributions | |
slices | -0.182 |
unison | -0.175 |
pulp | -0.173 |
pter | -0.173 |
fires | -0.173 |
Token | . |
Token ID | 13 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
wear | +0.060 |
Woo | +0.057 |
bath | +0.056 |
ibi | +0.055 |
burn | +0.054 |
Neg logit contributions | |
unison | -0.054 |
slices | -0.051 |
fires | -0.049 |
seconds | -0.049 |
pter | -0.049 |
Token | They |
Token ID | 1119 |
Feature activation | +7.84e-03 |
Pos logit contributions | |
wear | +0.029 |
Woo | +0.027 |
bath | +0.027 |
important | +0.026 |
burn | +0.026 |
Neg logit contributions | |
unison | -0.026 |
slices | -0.026 |
seventeen | -0.026 |
amaz | -0.025 |
aroma | -0.025 |
Token | loved |
Token ID | 6151 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
wear | +0.012 |
Woo | +0.012 |
bath | +0.012 |
burn | +0.011 |
Mitt | +0.011 |
Neg logit contributions | |
unison | -0.013 |
slices | -0.013 |
pulp | -0.012 |
seventeen | -0.012 |
fires | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
wear | +0.015 |
Woo | +0.014 |
bath | +0.013 |
Lucky | +0.012 |
Mitt | +0.012 |
Neg logit contributions | |
slices | -0.030 |
unison | -0.030 |
amaz | -0.029 |
seventeen | -0.029 |
seconds | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | +2.55e-03 |
Pos logit contributions | |
wear | +0.006 |
Woo | +0.006 |
bath | +0.006 |
day | +0.006 |
Mitt | +0.005 |
Neg logit contributions | |
seventeen | -0.013 |
unison | -0.013 |
slices | -0.013 |
pulp | -0.013 |
amaz | -0.013 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.70e-02 |
Pos logit contributions | |
wear | +0.004 |
bath | +0.004 |
Woo | +0.004 |
burn | +0.004 |
important | +0.004 |
Neg logit contributions | |
unison | -0.004 |
amaz | -0.004 |
seventeen | -0.004 |
slices | -0.004 |
seconds | -0.004 |
Token | Lisa |
Token ID | 44203 |
Feature activation | +1.95e-03 |
Pos logit contributions | |
slices | +0.086 |
unison | +0.085 |
amaz | +0.082 |
seventeen | +0.082 |
fires | +0.082 |
Neg logit contributions | |
wear | -0.067 |
Woo | -0.062 |
bath | -0.059 |
important | -0.059 |
burn | -0.059 |
Token | loved |
Token ID | 6151 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
Woo | +0.003 |
wear | +0.003 |
bath | +0.003 |
burn | +0.003 |
Mitt | +0.003 |
Neg logit contributions | |
slices | -0.003 |
unison | -0.003 |
seconds | -0.003 |
amaz | -0.003 |
seventeen | -0.003 |
Token | her |
Token ID | 607 |
Feature activation | +6.59e-02 |
Pos logit contributions | |
wear | +0.028 |
scar | +0.025 |
Woo | +0.025 |
Lucky | +0.025 |
opsy | +0.025 |
Neg logit contributions | |
slices | -0.040 |
conf | -0.038 |
pulp | -0.038 |
amaz | -0.036 |
seconds | -0.036 |
Token | velvet |
Token ID | 47750 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
Woo | +0.091 |
scar | +0.090 |
ained | +0.090 |
tto | +0.087 |
ongo | +0.086 |
Neg logit contributions | |
unison | -0.120 |
amaz | -0.119 |
slices | -0.118 |
conf | -0.114 |
ah | -0.108 |
Token | bear |
Token ID | 6842 |
Feature activation | +6.20e-02 |
Pos logit contributions | |
wear | +0.154 |
Woo | +0.145 |
bath | +0.136 |
important | +0.133 |
Mitt | +0.131 |
Neg logit contributions | |
unison | -0.184 |
slices | -0.183 |
aroma | -0.175 |
amaz | -0.175 |
seventeen | -0.174 |
Token | very |
Token ID | 845 |
Feature activation | +9.35e-03 |
Pos logit contributions | |
wear | +0.098 |
Woo | +0.097 |
stub | +0.093 |
scar | +0.091 |
ibi | +0.090 |
Neg logit contributions | |
slices | -0.100 |
unison | -0.097 |
pter | -0.087 |
als | -0.086 |
aroma | -0.084 |
Token | much |
Token ID | 881 |
Feature activation | +2.80e-02 |
Pos logit contributions | |
wear | +0.013 |
Woo | +0.013 |
bath | +0.012 |
burn | +0.012 |
Lucky | +0.012 |
Neg logit contributions | |
unison | -0.017 |
slices | -0.017 |
seconds | -0.016 |
seventeen | -0.016 |
amaz | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
wear | +0.033 |
Woo | +0.030 |
bath | +0.029 |
day | +0.028 |
important | +0.028 |
Neg logit contributions | |
seventeen | -0.060 |
unison | -0.059 |
slices | -0.059 |
amaz | -0.059 |
pulp | -0.058 |
Token | She |
Token ID | 1375 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
wear | +0.029 |
Woo | +0.024 |
day | +0.024 |
bath | +0.023 |
Mitt | +0.023 |
Neg logit contributions | |
slices | -0.063 |
fires | -0.062 |
pulp | -0.062 |
seventeen | -0.062 |
unison | -0.062 |
Token | choose |
Token ID | 3853 |
Feature activation | +1.80e-03 |
Pos logit contributions | |
Woo | +0.034 |
wear | +0.033 |
bath | +0.032 |
burn | +0.032 |
important | +0.030 |
Neg logit contributions | |
unison | -0.060 |
seventeen | -0.057 |
amaz | -0.057 |
aroma | -0.056 |
slices | -0.056 |
Token | one |
Token ID | 530 |
Feature activation | +3.02e-02 |
Pos logit contributions | |
wear | +0.003 |
Woo | +0.002 |
bath | +0.002 |
Mitt | +0.002 |
Lucky | +0.002 |
Neg logit contributions | |
unison | -0.003 |
slices | -0.003 |
seventeen | -0.003 |
amaz | -0.003 |
seconds | -0.003 |
Token | of |
Token ID | 286 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
wear | +0.047 |
Woo | +0.046 |
bath | +0.044 |
Mitt | +0.042 |
Lucky | +0.042 |
Neg logit contributions | |
slices | -0.052 |
unison | -0.050 |
seventeen | -0.050 |
seconds | -0.049 |
amaz | -0.049 |
Token | these |
Token ID | 777 |
Feature activation | -2.18e-03 |
Pos logit contributions | |
wear | +0.052 |
Woo | +0.049 |
bath | +0.048 |
Mitt | +0.045 |
Lucky | +0.044 |
Neg logit contributions | |
unison | -0.071 |
slices | -0.070 |
seventeen | -0.069 |
amaz | -0.069 |
seconds | -0.067 |
Token | b |
Token ID | 275 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
slices | +0.004 |
unison | +0.004 |
seventeen | +0.004 |
amaz | +0.004 |
aroma | +0.004 |
Neg logit contributions | |
wear | -0.003 |
Woo | -0.003 |
bath | -0.003 |
Mitt | -0.003 |
burn | -0.003 |
Token | unn |
Token ID | 20935 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
wear | +0.019 |
Woo | +0.017 |
bath | +0.017 |
Mitt | +0.015 |
Lucky | +0.015 |
Neg logit contributions | |
slices | -0.050 |
seventeen | -0.049 |
unison | -0.048 |
pulp | -0.048 |
seconds | -0.048 |
Token | ies |
Token ID | 444 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
wear | +0.132 |
Woo | +0.120 |
bath | +0.118 |
Mitt | +0.110 |
day | +0.109 |
Neg logit contributions | |
unison | -0.209 |
slices | -0.206 |
seventeen | -0.205 |
amaz | -0.204 |
aroma | -0.200 |
Token | to |
Token ID | 284 |
Feature activation | -2.98e-03 |
Pos logit contributions | |
wear | +0.043 |
Woo | +0.037 |
bath | +0.037 |
day | +0.036 |
important | +0.036 |
Neg logit contributions | |
unison | -0.059 |
amaz | -0.057 |
seventeen | -0.056 |
slices | -0.055 |
aroma | -0.055 |
Token | come |
Token ID | 1282 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
unison | +0.006 |
slices | +0.006 |
seventeen | +0.006 |
amaz | +0.006 |
aroma | +0.006 |
Neg logit contributions | |
wear | -0.003 |
Woo | -0.003 |
bath | -0.003 |
Mitt | -0.003 |
burn | -0.003 |
Token | with |
Token ID | 351 |
Feature activation | -2.72e-03 |
Pos logit contributions | |
slices | +0.036 |
unison | +0.035 |
seventeen | +0.035 |
amaz | +0.034 |
fires | +0.034 |
Neg logit contributions | |
wear | -0.024 |
Woo | -0.022 |
bath | -0.022 |
Mitt | -0.021 |
Lucky | -0.020 |
Token | us |
Token ID | 514 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
unison | +0.005 |
seventeen | +0.005 |
amaz | +0.005 |
slices | +0.005 |
fires | +0.005 |
Neg logit contributions | |
wear | -0.004 |
Woo | -0.004 |
bath | -0.004 |
day | -0.004 |
important | -0.004 |
Token | you |
Token ID | 345 |
Feature activation | -3.65e-02 |
Pos logit contributions | |
wear | +0.024 |
Woo | +0.023 |
bath | +0.022 |
Mitt | +0.021 |
important | +0.021 |
Neg logit contributions | |
seventeen | -0.043 |
fires | -0.042 |
atively | -0.042 |
aroma | -0.041 |
slices | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
unison | +0.067 |
slices | +0.066 |
seventeen | +0.065 |
amaz | +0.065 |
aroma | +0.064 |
Neg logit contributions | |
wear | -0.053 |
Woo | -0.050 |
bath | -0.049 |
Mitt | -0.046 |
Lucky | -0.045 |
Token | You |
Token ID | 921 |
Feature activation | -5.44e-02 |
Pos logit contributions | |
wear | +0.041 |
bath | +0.038 |
Woo | +0.038 |
day | +0.036 |
burn | +0.036 |
Neg logit contributions | |
unison | -0.026 |
seventeen | -0.026 |
slices | -0.026 |
amaz | -0.025 |
seconds | -0.025 |
Token | love |
Token ID | 1842 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
slices | +0.092 |
seventeen | +0.092 |
unison | +0.092 |
amaz | +0.090 |
aroma | +0.089 |
Neg logit contributions | |
wear | -0.093 |
Woo | -0.082 |
bath | -0.082 |
always | -0.078 |
Lucky | -0.077 |
Token | b |
Token ID | 275 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
slices | +0.042 |
unison | +0.041 |
seconds | +0.040 |
seventeen | +0.040 |
amaz | +0.040 |
Neg logit contributions | |
wear | -0.032 |
Woo | -0.030 |
bath | -0.029 |
burn | -0.028 |
Lucky | -0.027 |
Token | unn |
Token ID | 20935 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
wear | +0.021 |
Woo | +0.019 |
bath | +0.019 |
important | +0.016 |
Lucky | +0.016 |
Neg logit contributions | |
slices | -0.056 |
unison | -0.055 |
seventeen | -0.054 |
amaz | -0.054 |
seconds | -0.053 |
Token | ies |
Token ID | 444 |
Feature activation | +5.34e-02 |
Pos logit contributions | |
wear | +0.120 |
Woo | +0.110 |
bath | +0.104 |
Mitt | +0.100 |
Lucky | +0.098 |
Neg logit contributions | |
unison | -0.225 |
seventeen | -0.223 |
amaz | -0.222 |
slices | -0.219 |
seconds | -0.213 |
Token | , |
Token ID | 11 |
Feature activation | +8.94e-03 |
Pos logit contributions | |
wear | +0.071 |
Woo | +0.065 |
bath | +0.062 |
day | +0.060 |
important | +0.060 |
Neg logit contributions | |
amaz | -0.108 |
seventeen | -0.108 |
unison | -0.108 |
pulp | -0.105 |
slices | -0.105 |
Token | right |
Token ID | 826 |
Feature activation | -4.93e-02 |
Pos logit contributions | |
wear | +0.012 |
Woo | +0.011 |
bath | +0.011 |
burn | +0.010 |
Mitt | +0.010 |
Neg logit contributions | |
slices | -0.017 |
unison | -0.017 |
amaz | -0.016 |
seventeen | -0.016 |
seconds | -0.016 |
Token | ?" |
Token ID | 1701 |
Feature activation | -2.60e-03 |
Pos logit contributions | |
seventeen | +0.113 |
slices | +0.112 |
unison | +0.111 |
amaz | +0.108 |
pulp | +0.107 |
Neg logit contributions | |
wear | -0.054 |
Woo | -0.047 |
day | -0.047 |
important | -0.045 |
bath | -0.044 |
Token | Lily |
Token ID | 20037 |
Feature activation | +4.60e-03 |
Pos logit contributions | |
slices | +0.005 |
unison | +0.005 |
amaz | +0.005 |
pulp | +0.005 |
seconds | +0.005 |
Neg logit contributions | |
wear | -0.003 |
burn | -0.003 |
important | -0.003 |
bath | -0.003 |
Woo | -0.003 |
Token | there |
Token ID | 612 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
wear | +0.029 |
Woo | +0.028 |
bath | +0.027 |
burn | +0.025 |
Mitt | +0.025 |
Neg logit contributions | |
unison | -0.035 |
slices | -0.035 |
seventeen | -0.034 |
amaz | -0.034 |
seconds | -0.033 |
Token | were |
Token ID | 547 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
wear | +0.052 |
Woo | +0.050 |
bath | +0.047 |
burn | +0.047 |
Mitt | +0.045 |
Neg logit contributions | |
unison | -0.061 |
amaz | -0.061 |
slices | -0.060 |
seventeen | -0.058 |
seconds | -0.058 |
Token | two |
Token ID | 734 |
Feature activation | +3.14e-02 |
Pos logit contributions | |
slices | +0.035 |
conf | +0.032 |
unison | +0.032 |
seventeen | +0.032 |
pulp | +0.032 |
Neg logit contributions | |
wear | -0.034 |
Woo | -0.029 |
jog | -0.028 |
burn | -0.028 |
ibi | -0.028 |
Token | little |
Token ID | 1310 |
Feature activation | +3.37e-02 |
Pos logit contributions | |
wear | +0.035 |
Woo | +0.033 |
bath | +0.030 |
burn | +0.028 |
Lucky | +0.028 |
Neg logit contributions | |
slices | -0.069 |
unison | -0.068 |
seventeen | -0.067 |
amaz | -0.066 |
seconds | -0.066 |
Token | b |
Token ID | 275 |
Feature activation | +3.66e-02 |
Pos logit contributions | |
wear | +0.053 |
Woo | +0.050 |
bath | +0.049 |
Mitt | +0.047 |
Lucky | +0.046 |
Neg logit contributions | |
unison | -0.058 |
seventeen | -0.057 |
slices | -0.057 |
amaz | -0.056 |
fires | -0.055 |
Token | unn |
Token ID | 20935 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
wear | +0.043 |
Woo | +0.041 |
bath | +0.040 |
Lucky | +0.037 |
Mitt | +0.036 |
Neg logit contributions | |
slices | -0.079 |
pulp | -0.075 |
seventeen | -0.075 |
pter | -0.074 |
als | -0.073 |
Token | ies |
Token ID | 444 |
Feature activation | +3.25e-02 |
Pos logit contributions | |
wear | +0.163 |
Woo | +0.155 |
bath | +0.153 |
jog | +0.142 |
Lucky | +0.140 |
Neg logit contributions | |
slices | -0.172 |
pter | -0.164 |
pulp | -0.163 |
fires | -0.161 |
unison | -0.161 |
Token | called |
Token ID | 1444 |
Feature activation | +4.09e-02 |
Pos logit contributions | |
wear | +0.058 |
Woo | +0.057 |
bath | +0.055 |
ibi | +0.054 |
burn | +0.053 |
Neg logit contributions | |
unison | -0.046 |
slices | -0.045 |
pter | -0.042 |
seconds | -0.041 |
fires | -0.039 |
Token | Bob |
Token ID | 5811 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
wear | +0.050 |
important | +0.044 |
stub | +0.042 |
burn | +0.041 |
bath | +0.041 |
Neg logit contributions | |
unison | -0.084 |
slices | -0.082 |
seventeen | -0.079 |
ah | -0.078 |
amaz | -0.078 |
Token | and |
Token ID | 290 |
Feature activation | +3.78e-02 |
Pos logit contributions | |
wear | +0.037 |
Woo | +0.035 |
bath | +0.035 |
skipping | +0.032 |
stub | +0.032 |
Neg logit contributions | |
slices | -0.046 |
unison | -0.045 |
pter | -0.043 |
seventeen | -0.043 |
seconds | -0.042 |
Token | Su |
Token ID | 1778 |
Feature activation | +3.00e-02 |
Pos logit contributions | |
wear | +0.046 |
bath | +0.043 |
Woo | +0.040 |
important | +0.040 |
burn | +0.039 |
Neg logit contributions | |
unison | -0.075 |
slices | -0.074 |
seconds | -0.073 |
seventeen | -0.072 |
amaz | -0.072 |
Token | my |
Token ID | 1820 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
slices | +0.034 |
unison | +0.033 |
seventeen | +0.033 |
amaz | +0.033 |
aroma | +0.033 |
Neg logit contributions | |
wear | -0.012 |
Woo | -0.011 |
bath | -0.011 |
important | -0.010 |
Lucky | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -8.23e-03 |
Pos logit contributions | |
wear | +0.018 |
Woo | +0.017 |
bath | +0.017 |
important | +0.015 |
Lucky | +0.015 |
Neg logit contributions | |
slices | -0.030 |
unison | -0.030 |
seventeen | -0.030 |
amaz | -0.030 |
aroma | -0.030 |
Token | what |
Token ID | 644 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
unison | +0.014 |
slices | +0.014 |
seventeen | +0.014 |
fires | +0.014 |
aroma | +0.014 |
Neg logit contributions | |
wear | -0.013 |
Woo | -0.012 |
bath | -0.012 |
Mitt | -0.012 |
day | -0.011 |
Token | do |
Token ID | 466 |
Feature activation | +4.19e-03 |
Pos logit contributions | |
slices | +0.036 |
amaz | +0.036 |
unison | +0.036 |
seventeen | +0.034 |
seconds | +0.034 |
Neg logit contributions | |
Woo | -0.042 |
wear | -0.041 |
bath | -0.039 |
Mitt | -0.038 |
burn | -0.038 |
Token | b |
Token ID | 275 |
Feature activation | +2.44e-02 |
Pos logit contributions | |
wear | +0.007 |
Woo | +0.007 |
bath | +0.007 |
burn | +0.006 |
Mitt | +0.006 |
Neg logit contributions | |
slices | -0.006 |
unison | -0.006 |
seventeen | -0.006 |
seconds | -0.006 |
amaz | -0.006 |
Token | unn |
Token ID | 20935 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
bath | +0.029 |
stub | +0.029 |
wear | +0.029 |
important | +0.028 |
burn | +0.028 |
Neg logit contributions | |
als | -0.046 |
goodness | -0.044 |
slices | -0.044 |
aroma | -0.043 |
teen | -0.042 |
Token | ies |
Token ID | 444 |
Feature activation | +4.27e-02 |
Pos logit contributions | |
wear | +0.145 |
Woo | +0.141 |
bath | +0.137 |
Mitt | +0.129 |
important | +0.128 |
Neg logit contributions | |
slices | -0.179 |
aroma | -0.175 |
pulp | -0.172 |
ah | -0.172 |
unison | -0.172 |
Token | do |
Token ID | 466 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
Woo | +0.053 |
wear | +0.052 |
bath | +0.049 |
Mitt | +0.048 |
burn | +0.047 |
Neg logit contributions | |
slices | -0.081 |
unison | -0.081 |
aroma | -0.080 |
fires | -0.080 |
seventeen | -0.080 |
Token | ?" |
Token ID | 1701 |
Feature activation | -8.13e-03 |
Pos logit contributions | |
wear | +0.012 |
Woo | +0.012 |
bath | +0.011 |
Mitt | +0.010 |
burn | +0.010 |
Neg logit contributions | |
slices | -0.025 |
unison | -0.025 |
seventeen | -0.024 |
amaz | -0.024 |
fires | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | -4.47e-04 |
Pos logit contributions | |
unison | +0.016 |
slices | +0.015 |
amaz | +0.015 |
seventeen | +0.015 |
seconds | +0.015 |
Neg logit contributions | |
wear | -0.011 |
Woo | -0.010 |
important | -0.010 |
bath | -0.009 |
burn | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
unison | +0.001 |
slices | +0.001 |
seventeen | +0.001 |
seconds | +0.001 |
amaz | +0.001 |
Neg logit contributions | |
wear | -0.001 |
working | -0.001 |
Woo | -0.001 |
burn | -0.001 |
exercise | -0.001 |
Token | unn |
Token ID | 20935 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
wear | +0.044 |
Woo | +0.041 |
bath | +0.039 |
Lucky | +0.039 |
Mitt | +0.037 |
Neg logit contributions | |
slices | -0.083 |
pulp | -0.079 |
pter | -0.077 |
als | -0.077 |
seventeen | -0.076 |
Token | ies |
Token ID | 444 |
Feature activation | +3.60e-02 |
Pos logit contributions | |
wear | +0.166 |
Woo | +0.155 |
bath | +0.153 |
Lucky | +0.144 |
Mitt | +0.142 |
Neg logit contributions | |
slices | -0.184 |
pter | -0.176 |
pulp | -0.175 |
unison | -0.175 |
fires | -0.175 |
Token | . |
Token ID | 13 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
wear | +0.061 |
Woo | +0.058 |
ibi | +0.057 |
bath | +0.057 |
scar | +0.056 |
Neg logit contributions | |
unison | -0.052 |
slices | -0.049 |
fires | -0.048 |
seconds | -0.047 |
pter | -0.047 |
Token | The |
Token ID | 383 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
wear | +0.026 |
Woo | +0.024 |
important | +0.024 |
burn | +0.024 |
bath | +0.024 |
Neg logit contributions | |
unison | -0.020 |
slices | -0.019 |
seventeen | -0.019 |
amaz | -0.018 |
aroma | -0.018 |
Token | b |
Token ID | 275 |
Feature activation | +3.03e-02 |
Pos logit contributions | |
wear | +0.032 |
scar | +0.031 |
ogging | +0.030 |
Razor | +0.029 |
ongo | +0.029 |
Neg logit contributions | |
aroma | -0.038 |
slices | -0.038 |
unison | -0.037 |
pulp | -0.037 |
amaz | -0.036 |
Token | unn |
Token ID | 20935 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
wear | +0.037 |
bath | +0.035 |
Woo | +0.034 |
jog | +0.034 |
Lucky | +0.034 |
Neg logit contributions | |
als | -0.060 |
slices | -0.059 |
pulp | -0.058 |
edly | -0.057 |
pter | -0.056 |
Token | ies |
Token ID | 444 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
wear | +0.167 |
Woo | +0.159 |
bath | +0.157 |
jog | +0.150 |
Lucky | +0.149 |
Neg logit contributions | |
slices | -0.157 |
pulp | -0.156 |
pter | -0.152 |
fires | -0.152 |
als | -0.150 |
Token | lived |
Token ID | 5615 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
Woo | +0.046 |
burn | +0.044 |
Mitt | +0.043 |
wear | +0.043 |
bath | +0.043 |
Neg logit contributions | |
slices | -0.046 |
unison | -0.046 |
fires | -0.045 |
pulp | -0.045 |
aroma | -0.044 |
Token | in |
Token ID | 287 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
wear | +0.017 |
bath | +0.016 |
Woo | +0.016 |
ibi | +0.015 |
burn | +0.015 |
Neg logit contributions | |
slices | -0.018 |
fires | -0.017 |
seconds | -0.016 |
unison | -0.016 |
pulp | -0.016 |
Token | a |
Token ID | 257 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
unison | +0.021 |
slices | +0.021 |
amaz | +0.021 |
seconds | +0.020 |
seventeen | +0.020 |
Neg logit contributions | |
wear | -0.016 |
Woo | -0.015 |
bath | -0.014 |
burn | -0.014 |
skipping | -0.013 |
Token | cozy |
Token ID | 37438 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
unison | +0.019 |
slices | +0.019 |
pulp | +0.019 |
amaz | +0.018 |
fires | +0.018 |
Neg logit contributions | |
wear | -0.014 |
ibi | -0.013 |
burn | -0.012 |
scar | -0.012 |
ido | -0.012 |
Token | Do |
Token ID | 2141 |
Feature activation | -5.65e-03 |
Pos logit contributions | |
wear | +0.015 |
Woo | +0.015 |
bath | +0.014 |
Mitt | +0.014 |
Lucky | +0.013 |
Neg logit contributions | |
slices | -0.013 |
unison | -0.012 |
aroma | -0.012 |
seventeen | -0.012 |
amaz | -0.012 |
Token | you |
Token ID | 345 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
amaz | +0.011 |
unison | +0.011 |
slices | +0.011 |
seventeen | +0.011 |
aroma | +0.011 |
Neg logit contributions | |
wear | -0.008 |
Woo | -0.008 |
bath | -0.007 |
day | -0.007 |
Mitt | -0.007 |
Token | like |
Token ID | 588 |
Feature activation | +9.45e-03 |
Pos logit contributions | |
unison | +0.023 |
slices | +0.023 |
seventeen | +0.023 |
amaz | +0.022 |
seconds | +0.022 |
Neg logit contributions | |
Woo | -0.022 |
wear | -0.021 |
bath | -0.021 |
burn | -0.021 |
Mitt | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
wear | +0.015 |
ibi | +0.014 |
Woo | +0.014 |
stub | +0.014 |
scar | +0.014 |
Neg logit contributions | |
slices | -0.016 |
seventeen | -0.014 |
seconds | -0.014 |
unison | -0.014 |
conf | -0.014 |
Token | b |
Token ID | 275 |
Feature activation | +2.27e-02 |
Pos logit contributions | |
wear | +0.035 |
Woo | +0.033 |
bath | +0.032 |
Mitt | +0.030 |
Lucky | +0.030 |
Neg logit contributions | |
slices | -0.040 |
unison | -0.040 |
seventeen | -0.039 |
amaz | -0.039 |
aroma | -0.039 |
Token | unn |
Token ID | 20935 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
wear | +0.022 |
bath | +0.021 |
Woo | +0.020 |
skipping | +0.019 |
Lucky | +0.019 |
Neg logit contributions | |
slices | -0.052 |
seventeen | -0.051 |
als | -0.050 |
pter | -0.050 |
pulp | -0.049 |
Token | ies |
Token ID | 444 |
Feature activation | +3.97e-02 |
Pos logit contributions | |
wear | +0.147 |
Woo | +0.139 |
bath | +0.136 |
Mitt | +0.128 |
Lucky | +0.127 |
Neg logit contributions | |
slices | -0.185 |
pulp | -0.178 |
unison | -0.177 |
aroma | -0.177 |
seventeen | -0.177 |
Token | ?" |
Token ID | 1701 |
Feature activation | +9.98e-04 |
Pos logit contributions | |
Woo | +0.042 |
wear | +0.041 |
Mitt | +0.039 |
bath | +0.038 |
Lucky | +0.037 |
Neg logit contributions | |
slices | -0.084 |
unison | -0.081 |
aroma | -0.080 |
fires | -0.079 |
seventeen | -0.079 |
Token | the |
Token ID | 262 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
burn | +0.001 |
wear | +0.001 |
bath | +0.001 |
important | +0.001 |
Woo | +0.001 |
Neg logit contributions | |
slices | -0.002 |
seconds | -0.002 |
amaz | -0.002 |
unison | -0.002 |
pulp | -0.002 |
Token | lady |
Token ID | 10846 |
Feature activation | -4.66e-02 |
Pos logit contributions | |
slices | +0.020 |
unison | +0.020 |
aroma | +0.020 |
pulp | +0.020 |
seconds | +0.020 |
Neg logit contributions | |
wear | -0.016 |
Woo | -0.015 |
bath | -0.014 |
important | -0.014 |
burn | -0.014 |
Token | asks |
Token ID | 7893 |
Feature activation | -4.19e-03 |
Pos logit contributions | |
slices | +0.075 |
unison | +0.074 |
amaz | +0.074 |
seventeen | +0.074 |
seconds | +0.072 |
Neg logit contributions | |
wear | -0.076 |
Woo | -0.075 |
bath | -0.072 |
Mitt | -0.069 |
important | -0.068 |
Token | Fred |
Token ID | 30847 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
unison | +0.022 |
slices | +0.022 |
amaz | +0.022 |
seventeen | +0.021 |
aroma | +0.021 |
Neg logit contributions | |
wear | -0.018 |
Woo | -0.017 |
bath | -0.016 |
Mitt | -0.016 |
Lucky | -0.015 |
Token | played |
Token ID | 2826 |
Feature activation | +4.45e-02 |
Pos logit contributions | |
wear | +0.015 |
Woo | +0.014 |
bath | +0.013 |
burn | +0.012 |
Mitt | +0.012 |
Neg logit contributions | |
unison | -0.020 |
slices | -0.020 |
seventeen | -0.020 |
amaz | -0.020 |
seconds | -0.019 |
Token | with |
Token ID | 351 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
wear | +0.050 |
Woo | +0.046 |
bath | +0.044 |
Lucky | +0.042 |
skipping | +0.041 |
Neg logit contributions | |
seventeen | -0.098 |
unison | -0.098 |
slices | -0.096 |
amaz | -0.096 |
pulp | -0.096 |
Token | the |
Token ID | 262 |
Feature activation | +5.02e-03 |
Pos logit contributions | |
slices | +0.024 |
unison | +0.024 |
amaz | +0.023 |
seconds | +0.023 |
seventeen | +0.023 |
Neg logit contributions | |
wear | -0.017 |
Woo | -0.015 |
bath | -0.015 |
burn | -0.014 |
important | -0.014 |
Token | b |
Token ID | 275 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
wear | +0.006 |
Woo | +0.005 |
bath | +0.005 |
Mitt | +0.005 |
Lucky | +0.005 |
Neg logit contributions | |
slices | -0.010 |
aroma | -0.010 |
amaz | -0.010 |
pulp | -0.010 |
unison | -0.010 |
Token | unn |
Token ID | 20935 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
wear | +0.020 |
Woo | +0.018 |
bath | +0.017 |
Lucky | +0.016 |
skipping | +0.016 |
Neg logit contributions | |
slices | -0.042 |
als | -0.041 |
seventeen | -0.041 |
pulp | -0.040 |
aroma | -0.040 |
Token | ies |
Token ID | 444 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
wear | +0.161 |
Woo | +0.154 |
bath | +0.150 |
Mitt | +0.140 |
burn | +0.139 |
Neg logit contributions | |
slices | -0.170 |
pulp | -0.163 |
pter | -0.159 |
aroma | -0.158 |
fires | -0.158 |
Token | all |
Token ID | 477 |
Feature activation | +2.61e-03 |
Pos logit contributions | |
wear | +0.023 |
Woo | +0.022 |
bath | +0.021 |
Mitt | +0.020 |
burn | +0.019 |
Neg logit contributions | |
slices | -0.036 |
unison | -0.036 |
amaz | -0.035 |
seventeen | -0.035 |
seconds | -0.034 |
Token | day |
Token ID | 1110 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
wear | +0.002 |
Woo | +0.001 |
bath | +0.001 |
Mitt | +0.001 |
burn | +0.001 |
Neg logit contributions | |
slices | -0.007 |
unison | -0.007 |
seventeen | -0.007 |
amaz | -0.007 |
pter | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | -5.47e-03 |
Pos logit contributions | |
slices | +0.022 |
unison | +0.022 |
seventeen | +0.021 |
amaz | +0.021 |
seconds | +0.021 |
Neg logit contributions | |
wear | -0.014 |
Woo | -0.014 |
bath | -0.013 |
burn | -0.012 |
Mitt | -0.012 |
Token | hopping |
Token ID | 47153 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
slices | +0.011 |
unison | +0.011 |
seventeen | +0.011 |
amaz | +0.011 |
seconds | +0.011 |
Neg logit contributions | |
wear | -0.007 |
Woo | -0.006 |
bath | -0.006 |
burn | -0.006 |
important | -0.006 |
Token | little |
Token ID | 1310 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
ained | +0.030 |
wear | +0.030 |
Woo | +0.028 |
burn | +0.028 |
ibi | +0.026 |
Neg logit contributions | |
unison | -0.035 |
aroma | -0.033 |
slices | -0.033 |
pulp | -0.033 |
seconds | -0.032 |
Token | girl |
Token ID | 2576 |
Feature activation | -7.29e-03 |
Pos logit contributions | |
wear | +0.041 |
Woo | +0.037 |
important | +0.035 |
bath | +0.035 |
ained | +0.035 |
Neg logit contributions | |
slices | -0.032 |
aroma | -0.031 |
unison | -0.031 |
pulp | -0.031 |
conf | -0.029 |
Token | who |
Token ID | 508 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
pter | +0.014 |
seconds | +0.013 |
unison | +0.013 |
slices | +0.013 |
als | +0.012 |
Neg logit contributions | |
Woo | -0.009 |
ibi | -0.008 |
ained | -0.008 |
ongo | -0.008 |
burn | -0.008 |
Token | loved |
Token ID | 6151 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
Woo | +0.057 |
burn | +0.055 |
wear | +0.054 |
bath | +0.053 |
skipping | +0.051 |
Neg logit contributions | |
unison | -0.065 |
fires | -0.065 |
aroma | -0.064 |
pter | -0.064 |
amaz | -0.063 |
Token | her |
Token ID | 607 |
Feature activation | +6.35e-02 |
Pos logit contributions | |
wear | +0.025 |
Woo | +0.023 |
Lucky | +0.021 |
scar | +0.021 |
bath | +0.021 |
Neg logit contributions | |
slices | -0.041 |
pulp | -0.039 |
conf | -0.038 |
seconds | -0.038 |
amaz | -0.038 |
Token | rubber |
Token ID | 14239 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
scar | +0.092 |
Woo | +0.091 |
tto | +0.086 |
ained | +0.084 |
umper | +0.083 |
Neg logit contributions | |
slices | -0.113 |
amaz | -0.111 |
unison | -0.111 |
aroma | -0.107 |
conf | -0.107 |
Token | . |
Token ID | 13 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
wear | +0.133 |
Woo | +0.123 |
bath | +0.118 |
Mitt | +0.111 |
important | +0.111 |
Neg logit contributions | |
slices | -0.200 |
unison | -0.199 |
seventeen | -0.194 |
amaz | -0.193 |
aroma | -0.191 |
Token | Every |
Token ID | 3887 |
Feature activation | +5.48e-02 |
Pos logit contributions | |
wear | +0.035 |
Woo | +0.034 |
burn | +0.033 |
bath | +0.033 |
important | +0.032 |
Neg logit contributions | |
amaz | -0.036 |
unison | -0.036 |
slices | -0.036 |
seventeen | -0.035 |
aroma | -0.034 |
Token | day |
Token ID | 1110 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
Woo | +0.050 |
wear | +0.050 |
ongo | +0.047 |
Nut | +0.046 |
Mitt | +0.046 |
Neg logit contributions | |
seconds | -0.117 |
moments | -0.116 |
unison | -0.116 |
aroma | -0.112 |
slices | -0.110 |
Token | she |
Token ID | 673 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
slices | +0.019 |
als | +0.019 |
seconds | +0.019 |
aroma | +0.019 |
amaz | +0.019 |
Neg logit contributions | |
Woo | -0.014 |
wear | -0.014 |
burn | -0.014 |
scar | -0.014 |
iny | -0.014 |
Token | found |
Token ID | 1043 |
Feature activation | +9.91e-04 |
Pos logit contributions | |
Woo | +0.019 |
ibi | +0.019 |
skipping | +0.018 |
burn | +0.018 |
Mitt | +0.018 |
Neg logit contributions | |
seconds | -0.020 |
pter | -0.019 |
slices | -0.019 |
pulp | -0.019 |
ts | -0.019 |
Token | of |
Token ID | 286 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
slices | +0.015 |
unison | +0.015 |
fires | +0.014 |
amaz | +0.014 |
aroma | +0.014 |
Neg logit contributions | |
wear | -0.009 |
Woo | -0.009 |
bath | -0.008 |
Lucky | -0.008 |
burn | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | +4.86e-03 |
Pos logit contributions | |
wear | +0.055 |
Woo | +0.051 |
bath | +0.049 |
burn | +0.048 |
Lucky | +0.047 |
Neg logit contributions | |
slices | -0.077 |
unison | -0.075 |
amaz | -0.075 |
seconds | -0.074 |
pulp | -0.073 |
Token | family |
Token ID | 1641 |
Feature activation | -3.30e-02 |
Pos logit contributions | |
wear | +0.006 |
Woo | +0.005 |
burn | +0.005 |
bath | +0.005 |
Lucky | +0.005 |
Neg logit contributions | |
unison | -0.010 |
slices | -0.010 |
amaz | -0.010 |
aroma | -0.010 |
seconds | -0.010 |
Token | of |
Token ID | 286 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
slices | +0.050 |
unison | +0.049 |
pter | +0.048 |
fires | +0.047 |
amaz | +0.046 |
Neg logit contributions | |
Woo | -0.054 |
burn | -0.051 |
Mitt | -0.051 |
wear | -0.051 |
Lucky | -0.050 |
Token | b |
Token ID | 275 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
wear | +0.054 |
Woo | +0.051 |
bath | +0.049 |
burn | +0.048 |
Mitt | +0.047 |
Neg logit contributions | |
slices | -0.057 |
unison | -0.056 |
amaz | -0.055 |
seconds | -0.055 |
pulp | -0.054 |
Token | unn |
Token ID | 20935 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
wear | +0.029 |
Woo | +0.027 |
bath | +0.026 |
Lucky | +0.025 |
Mitt | +0.024 |
Neg logit contributions | |
slices | -0.061 |
pulp | -0.058 |
als | -0.057 |
pter | -0.056 |
aroma | -0.056 |
Token | ies |
Token ID | 444 |
Feature activation | +2.91e-02 |
Pos logit contributions | |
wear | +0.153 |
Woo | +0.144 |
bath | +0.141 |
Lucky | +0.131 |
Mitt | +0.131 |
Neg logit contributions | |
slices | -0.179 |
pulp | -0.171 |
unison | -0.171 |
fires | -0.170 |
pter | -0.170 |
Token | . |
Token ID | 13 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
wear | +0.045 |
Woo | +0.043 |
bath | +0.042 |
burn | +0.042 |
Mitt | +0.041 |
Neg logit contributions | |
unison | -0.047 |
slices | -0.047 |
fires | -0.045 |
seconds | -0.044 |
pter | -0.044 |
Token | One |
Token ID | 1881 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
wear | +0.016 |
Woo | +0.015 |
bath | +0.015 |
Mitt | +0.014 |
Lucky | +0.014 |
Neg logit contributions | |
slices | -0.021 |
seventeen | -0.021 |
unison | -0.021 |
amaz | -0.021 |
aroma | -0.020 |
Token | day |
Token ID | 1110 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
Woo | +0.033 |
umper | +0.028 |
ongo | +0.027 |
tying | +0.027 |
Barkley | +0.027 |
Neg logit contributions | |
aroma | -0.049 |
slices | -0.047 |
moments | -0.046 |
seconds | -0.043 |
amaz | -0.043 |
Token | one |
Token ID | 530 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
unison | +0.055 |
slices | +0.055 |
seventeen | +0.054 |
amaz | +0.054 |
aroma | +0.053 |
Neg logit contributions | |
wear | -0.026 |
Woo | -0.024 |
bath | -0.023 |
Mitt | -0.021 |
day | -0.021 |
Token | my |
Token ID | 1820 |
Feature activation | +9.52e-03 |
Pos logit contributions | |
unison | +0.062 |
slices | +0.061 |
seventeen | +0.061 |
amaz | +0.060 |
fires | +0.060 |
Neg logit contributions | |
wear | -0.008 |
Woo | -0.005 |
bath | -0.005 |
Mitt | -0.004 |
day | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
wear | +0.013 |
Woo | +0.012 |
bath | +0.012 |
Mitt | +0.011 |
Lucky | +0.011 |
Neg logit contributions | |
unison | -0.018 |
slices | -0.018 |
seventeen | -0.018 |
amaz | -0.018 |
fires | -0.017 |
Token | do |
Token ID | 466 |
Feature activation | -6.46e-03 |
Pos logit contributions | |
unison | +0.021 |
seventeen | +0.021 |
slices | +0.021 |
fires | +0.020 |
edly | +0.020 |
Neg logit contributions | |
wear | -0.016 |
bath | -0.014 |
Woo | -0.014 |
day | -0.013 |
always | -0.013 |
Token | these |
Token ID | 777 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
unison | +0.013 |
seventeen | +0.013 |
amaz | +0.013 |
slices | +0.013 |
pulp | +0.012 |
Neg logit contributions | |
wear | -0.009 |
Woo | -0.008 |
important | -0.008 |
bath | -0.008 |
Mitt | -0.008 |
Token | b |
Token ID | 275 |
Feature activation | +2.44e-02 |
Pos logit contributions | |
wear | +0.022 |
Woo | +0.021 |
bath | +0.020 |
Mitt | +0.019 |
Lucky | +0.019 |
Neg logit contributions | |
slices | -0.033 |
unison | -0.033 |
seventeen | -0.032 |
amaz | -0.032 |
aroma | -0.032 |
Token | unn |
Token ID | 20935 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
wear | +0.024 |
Woo | +0.022 |
bath | +0.021 |
Lucky | +0.020 |
Mitt | +0.019 |
Neg logit contributions | |
slices | -0.056 |
seventeen | -0.055 |
aroma | -0.054 |
seconds | -0.054 |
pulp | -0.053 |
Token | ies |
Token ID | 444 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
wear | +0.132 |
Woo | +0.122 |
bath | +0.119 |
Mitt | +0.111 |
Lucky | +0.110 |
Neg logit contributions | |
unison | -0.202 |
slices | -0.201 |
seventeen | -0.199 |
amaz | -0.197 |
aroma | -0.194 |
Token | live |
Token ID | 2107 |
Feature activation | +4.27e-02 |
Pos logit contributions | |
wear | +0.048 |
bath | +0.041 |
Woo | +0.040 |
day | +0.038 |
important | +0.037 |
Neg logit contributions | |
unison | -0.086 |
seventeen | -0.085 |
slices | -0.084 |
amaz | -0.084 |
seconds | -0.082 |
Token | here |
Token ID | 994 |
Feature activation | +3.14e-03 |
Pos logit contributions | |
wear | +0.050 |
Woo | +0.045 |
bath | +0.045 |
day | +0.042 |
important | +0.041 |
Neg logit contributions | |
unison | -0.091 |
seventeen | -0.091 |
slices | -0.090 |
amaz | -0.089 |
aroma | -0.087 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
wear | +0.004 |
Woo | +0.004 |
bath | +0.004 |
Mitt | +0.003 |
day | +0.003 |
Neg logit contributions | |
unison | -0.006 |
slices | -0.006 |
seventeen | -0.006 |
amaz | -0.006 |
aroma | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -4.67e-03 |
Pos logit contributions | |
slices | +0.025 |
unison | +0.024 |
amaz | +0.024 |
seconds | +0.024 |
seventeen | +0.024 |
Neg logit contributions | |
wear | -0.016 |
Woo | -0.015 |
bath | -0.015 |
burn | -0.014 |
important | -0.014 |
Token | was |
Token ID | 373 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
unison | +0.079 |
slices | +0.079 |
seventeen | +0.078 |
amaz | +0.077 |
aroma | +0.076 |
Neg logit contributions | |
wear | -0.055 |
Woo | -0.051 |
bath | -0.050 |
Mitt | -0.047 |
day | -0.047 |
Token | ready |
Token ID | 3492 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
Woo | +0.022 |
wear | +0.022 |
bath | +0.020 |
Lucky | +0.020 |
Mitt | +0.019 |
Neg logit contributions | |
seventeen | -0.040 |
slices | -0.040 |
unison | -0.039 |
aroma | -0.039 |
seconds | -0.039 |
Token | , |
Token ID | 11 |
Feature activation | +5.96e-03 |
Pos logit contributions | |
unison | +0.035 |
seventeen | +0.035 |
slices | +0.034 |
amaz | +0.034 |
pulp | +0.034 |
Neg logit contributions | |
wear | -0.014 |
Woo | -0.013 |
bath | -0.013 |
day | -0.012 |
Mitt | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
wear | +0.007 |
Woo | +0.007 |
bath | +0.006 |
Mitt | +0.006 |
Lucky | +0.006 |
Neg logit contributions | |
seventeen | -0.013 |
unison | -0.012 |
slices | -0.012 |
amaz | -0.012 |
pulp | -0.012 |
Token | b |
Token ID | 275 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
wear | +0.017 |
Woo | +0.016 |
bath | +0.015 |
Mitt | +0.014 |
Lucky | +0.014 |
Neg logit contributions | |
unison | -0.026 |
slices | -0.026 |
seventeen | -0.026 |
amaz | -0.025 |
aroma | -0.025 |
Token | unn |
Token ID | 20935 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
wear | +0.023 |
Woo | +0.021 |
bath | +0.020 |
skipping | +0.019 |
Lucky | +0.018 |
Neg logit contributions | |
slices | -0.057 |
seventeen | -0.056 |
unison | -0.055 |
aroma | -0.055 |
als | -0.054 |
Token | ies |
Token ID | 444 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
wear | +0.146 |
Woo | +0.139 |
bath | +0.135 |
Mitt | +0.126 |
Lucky | +0.126 |
Neg logit contributions | |
slices | -0.184 |
unison | -0.177 |
pulp | -0.176 |
aroma | -0.175 |
fires | -0.175 |
Token | hopped |
Token ID | 45230 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
wear | +0.029 |
Woo | +0.026 |
bath | +0.026 |
day | +0.024 |
skipping | +0.024 |
Neg logit contributions | |
unison | -0.043 |
seventeen | -0.043 |
slices | -0.042 |
amaz | -0.042 |
aroma | -0.041 |
Token | over |
Token ID | 625 |
Feature activation | -3.96e-02 |
Pos logit contributions | |
wear | +0.030 |
Woo | +0.028 |
bath | +0.027 |
Mitt | +0.025 |
Lucky | +0.025 |
Neg logit contributions | |
unison | -0.048 |
slices | -0.048 |
seventeen | -0.047 |
amaz | -0.047 |
aroma | -0.046 |
Token | to |
Token ID | 284 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
seventeen | +0.097 |
unison | +0.096 |
aroma | +0.094 |
pulp | +0.093 |
fires | +0.093 |
Neg logit contributions | |
wear | -0.034 |
bath | -0.032 |
Woo | -0.031 |
day | -0.029 |
skipping | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
unison | +0.060 |
slices | +0.060 |
seventeen | +0.059 |
amaz | +0.059 |
aroma | +0.058 |
Neg logit contributions | |
wear | -0.035 |
Woo | -0.032 |
bath | -0.031 |
Mitt | -0.029 |
Lucky | -0.028 |
Token | He |
Token ID | 679 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
unison | +0.002 |
seventeen | +0.002 |
amaz | +0.002 |
slices | +0.002 |
seconds | +0.002 |
Neg logit contributions | |
wear | -0.002 |
bath | -0.002 |
Woo | -0.002 |
burn | -0.002 |
important | -0.002 |
Token | had |
Token ID | 550 |
Feature activation | +9.25e-04 |
Pos logit contributions | |
unison | +0.028 |
slices | +0.027 |
amaz | +0.027 |
seconds | +0.027 |
seventeen | +0.027 |
Neg logit contributions | |
wear | -0.025 |
Woo | -0.024 |
bath | -0.024 |
burn | -0.023 |
Mitt | -0.022 |
Token | taught |
Token ID | 7817 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
wear | +0.001 |
Woo | +0.001 |
bath | +0.001 |
Mitt | +0.001 |
Lucky | +0.001 |
Neg logit contributions | |
unison | -0.002 |
slices | -0.002 |
seventeen | -0.002 |
amaz | -0.002 |
aroma | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
wear | +0.019 |
burn | +0.017 |
stub | +0.017 |
Woo | +0.017 |
bath | +0.017 |
Neg logit contributions | |
unison | -0.021 |
slices | -0.021 |
amaz | -0.021 |
seconds | -0.021 |
ah | -0.020 |
Token | b |
Token ID | 275 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
wear | +0.032 |
Woo | +0.029 |
burn | +0.027 |
ongo | +0.027 |
ibi | +0.026 |
Neg logit contributions | |
slices | -0.048 |
seconds | -0.048 |
unison | -0.048 |
aroma | -0.048 |
amaz | -0.047 |
Token | unn |
Token ID | 20935 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
wear | +0.022 |
Woo | +0.020 |
bath | +0.020 |
burn | +0.018 |
skipping | +0.018 |
Neg logit contributions | |
slices | -0.052 |
seventeen | -0.050 |
seconds | -0.050 |
fires | -0.049 |
unison | -0.049 |
Token | ies |
Token ID | 444 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
wear | +0.136 |
Woo | +0.127 |
bath | +0.124 |
Mitt | +0.116 |
Lucky | +0.114 |
Neg logit contributions | |
slices | -0.194 |
unison | -0.193 |
seventeen | -0.190 |
amaz | -0.188 |
aroma | -0.187 |
Token | to |
Token ID | 284 |
Feature activation | +8.54e-04 |
Pos logit contributions | |
wear | +0.032 |
bath | +0.028 |
Woo | +0.027 |
important | +0.027 |
day | +0.027 |
Neg logit contributions | |
seventeen | -0.049 |
unison | -0.048 |
slices | -0.048 |
pulp | -0.047 |
aroma | -0.047 |
Token | be |
Token ID | 307 |
Feature activation | -6.89e-03 |
Pos logit contributions | |
Woo | +0.001 |
wear | +0.001 |
bath | +0.001 |
burn | +0.001 |
important | +0.001 |
Neg logit contributions | |
unison | -0.002 |
amaz | -0.002 |
slices | -0.002 |
aroma | -0.002 |
seventeen | -0.002 |
Token | friendly |
Token ID | 8030 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
unison | +0.015 |
slices | +0.015 |
seventeen | +0.014 |
amaz | +0.014 |
aroma | +0.014 |
Neg logit contributions | |
wear | -0.008 |
Woo | -0.007 |
bath | -0.007 |
Mitt | -0.007 |
Lucky | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -7.78e-03 |
Pos logit contributions | |
seventeen | +0.028 |
pulp | +0.027 |
slices | +0.027 |
unison | +0.026 |
aroma | +0.026 |
Neg logit contributions | |
wear | -0.012 |
bath | -0.010 |
Woo | -0.010 |
day | -0.010 |
Lucky | -0.010 |
Token | listen |
Token ID | 6004 |
Feature activation | +1.99e-03 |
Pos logit contributions | |
wear | +0.022 |
Woo | +0.021 |
bath | +0.020 |
Mitt | +0.019 |
Lucky | +0.019 |
Neg logit contributions | |
unison | -0.029 |
slices | -0.029 |
seventeen | -0.029 |
amaz | -0.028 |
aroma | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | +3.24e-03 |
Pos logit contributions | |
wear | +0.002 |
Woo | +0.002 |
bath | +0.002 |
Mitt | +0.002 |
important | +0.002 |
Neg logit contributions | |
unison | -0.005 |
slices | -0.004 |
seventeen | -0.004 |
amaz | -0.004 |
aroma | -0.004 |
Token | She |
Token ID | 1375 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
wear | +0.005 |
bath | +0.004 |
burn | +0.004 |
Woo | +0.004 |
day | +0.004 |
Neg logit contributions | |
amaz | -0.006 |
slices | -0.006 |
unison | -0.006 |
seventeen | -0.005 |
seconds | -0.005 |
Token | threw |
Token ID | 9617 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
unison | +0.029 |
slices | +0.028 |
seconds | +0.028 |
seventeen | +0.027 |
amaz | +0.027 |
Neg logit contributions | |
wear | -0.027 |
Woo | -0.026 |
bath | -0.026 |
skipping | -0.024 |
burn | -0.024 |
Token | a |
Token ID | 257 |
Feature activation | -8.82e-03 |
Pos logit contributions | |
wear | +0.028 |
Woo | +0.027 |
bath | +0.026 |
Mitt | +0.024 |
Lucky | +0.024 |
Neg logit contributions | |
unison | -0.057 |
fires | -0.055 |
seventeen | -0.055 |
amaz | -0.054 |
slices | -0.054 |
Token | tant |
Token ID | 24246 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
unison | +0.012 |
amaz | +0.012 |
slices | +0.012 |
seventeen | +0.012 |
fires | +0.012 |
Neg logit contributions | |
wear | -0.017 |
bath | -0.016 |
Woo | -0.016 |
day | -0.015 |
Lucky | -0.015 |
Token | rum |
Token ID | 6582 |
Feature activation | -5.48e-02 |
Pos logit contributions | |
unison | +0.169 |
amaz | +0.162 |
slices | +0.161 |
seventeen | +0.160 |
fires | +0.155 |
Neg logit contributions | |
wear | -0.275 |
Woo | -0.259 |
bath | -0.253 |
Mitt | -0.246 |
important | -0.246 |
Token | in |
Token ID | 287 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
seventeen | +0.133 |
unison | +0.132 |
amaz | +0.130 |
slices | +0.128 |
aroma | +0.128 |
Neg logit contributions | |
wear | -0.050 |
Woo | -0.047 |
day | -0.044 |
skipping | -0.043 |
always | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | -3.67e-02 |
Pos logit contributions | |
unison | +0.055 |
seventeen | +0.055 |
slices | +0.055 |
amaz | +0.054 |
fires | +0.054 |
Neg logit contributions | |
wear | -0.031 |
Woo | -0.029 |
bath | -0.028 |
Mitt | -0.026 |
day | -0.026 |
Token | store |
Token ID | 3650 |
Feature activation | -1.32e-03 |
Pos logit contributions | |
unison | +0.079 |
slices | +0.078 |
seventeen | +0.077 |
amaz | +0.077 |
aroma | +0.076 |
Neg logit contributions | |
wear | -0.042 |
Woo | -0.038 |
bath | -0.035 |
Mitt | -0.034 |
burn | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -8.14e-03 |
Pos logit contributions | |
seventeen | +0.003 |
unison | +0.003 |
amaz | +0.003 |
pulp | +0.003 |
aroma | +0.003 |
Neg logit contributions | |
wear | -0.001 |
Woo | -0.001 |
bath | -0.001 |
day | -0.001 |
Mitt | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +1.10e-03 |
Pos logit contributions | |
wear | +0.007 |
Woo | +0.007 |
bath | +0.006 |
day | +0.006 |
important | +0.006 |
Neg logit contributions | |
unison | -0.013 |
seventeen | -0.013 |
slices | -0.013 |
amaz | -0.013 |
pulp | -0.013 |
Token | He |
Token ID | 679 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
wear | +0.001 |
Woo | +0.001 |
Mitt | +0.001 |
bath | +0.001 |
day | +0.001 |
Neg logit contributions | |
unison | -0.002 |
seventeen | -0.002 |
slices | -0.002 |
aroma | -0.002 |
pulp | -0.002 |
Token | kept |
Token ID | 4030 |
Feature activation | +3.67e-02 |
Pos logit contributions | |
slices | +0.043 |
unison | +0.043 |
seventeen | +0.043 |
amaz | +0.042 |
pulp | +0.041 |
Neg logit contributions | |
wear | -0.038 |
Woo | -0.035 |
bath | -0.034 |
Mitt | -0.032 |
Lucky | -0.032 |
Token | throwing |
Token ID | 9644 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
wear | +0.044 |
Woo | +0.040 |
bath | +0.040 |
important | +0.037 |
Lucky | +0.037 |
Neg logit contributions | |
slices | -0.076 |
seconds | -0.074 |
amaz | -0.073 |
unison | -0.073 |
seventeen | -0.072 |
Token | a |
Token ID | 257 |
Feature activation | -6.42e-03 |
Pos logit contributions | |
wear | +0.024 |
bath | +0.022 |
Woo | +0.021 |
Mitt | +0.020 |
day | +0.020 |
Neg logit contributions | |
unison | -0.046 |
fires | -0.045 |
seventeen | -0.044 |
atively | -0.043 |
MacDonald | -0.043 |
Token | tant |
Token ID | 24246 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
unison | +0.013 |
fires | +0.013 |
pter | +0.013 |
amaz | +0.013 |
seventeen | +0.013 |
Neg logit contributions | |
wear | -0.008 |
bath | -0.008 |
Woo | -0.007 |
day | -0.007 |
important | -0.007 |
Token | rum |
Token ID | 6582 |
Feature activation | -4.88e-02 |
Pos logit contributions | |
unison | +0.182 |
amaz | +0.174 |
slices | +0.167 |
seventeen | +0.166 |
fires | +0.166 |
Neg logit contributions | |
wear | -0.259 |
Woo | -0.240 |
important | -0.232 |
bath | -0.232 |
burn | -0.231 |
Token | . |
Token ID | 13 |
Feature activation | -5.87e-03 |
Pos logit contributions | |
seventeen | +0.111 |
unison | +0.110 |
amaz | +0.109 |
aroma | +0.108 |
pulp | +0.108 |
Neg logit contributions | |
wear | -0.050 |
Woo | -0.046 |
day | -0.044 |
bath | -0.043 |
important | -0.042 |
Token | His |
Token ID | 2399 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
unison | +0.011 |
slices | +0.011 |
seventeen | +0.011 |
aroma | +0.011 |
amaz | +0.011 |
Neg logit contributions | |
wear | -0.008 |
Woo | -0.008 |
bath | -0.008 |
Mitt | -0.007 |
Lucky | -0.007 |
Token | mom |
Token ID | 1995 |
Feature activation | -3.98e-02 |
Pos logit contributions | |
wear | +0.026 |
Woo | +0.023 |
bath | +0.023 |
Mitt | +0.020 |
day | +0.020 |
Neg logit contributions | |
unison | -0.061 |
slices | -0.060 |
seventeen | -0.060 |
amaz | -0.060 |
fires | -0.059 |
Token | was |
Token ID | 373 |
Feature activation | -5.13e-03 |
Pos logit contributions | |
unison | +0.075 |
seventeen | +0.075 |
slices | +0.074 |
amaz | +0.074 |
pulp | +0.073 |
Neg logit contributions | |
wear | -0.056 |
Woo | -0.051 |
bath | -0.050 |
Mitt | -0.048 |
Lucky | -0.047 |
Token | her |
Token ID | 607 |
Feature activation | +7.41e-03 |
Pos logit contributions | |
seventeen | +0.065 |
unison | +0.065 |
fires | +0.064 |
slices | +0.062 |
pulp | +0.062 |
Neg logit contributions | |
wear | -0.018 |
Woo | -0.015 |
Mitt | -0.015 |
bath | -0.015 |
Lucky | -0.013 |
Token | feet |
Token ID | 3625 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
wear | +0.013 |
bath | +0.012 |
Woo | +0.012 |
day | +0.012 |
skipping | +0.012 |
Neg logit contributions | |
seventeen | -0.011 |
unison | -0.011 |
slices | -0.011 |
amaz | -0.011 |
fires | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | -2.12e-03 |
Pos logit contributions | |
wear | +0.014 |
Woo | +0.013 |
bath | +0.012 |
Lucky | +0.012 |
Mitt | +0.012 |
Neg logit contributions | |
slices | -0.020 |
unison | -0.019 |
seventeen | -0.019 |
amaz | -0.019 |
seconds | -0.019 |
Token | threw |
Token ID | 9617 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
unison | +0.004 |
seventeen | +0.004 |
fires | +0.004 |
amaz | +0.004 |
slices | +0.004 |
Neg logit contributions | |
wear | -0.003 |
Woo | -0.003 |
Mitt | -0.003 |
Lucky | -0.003 |
bath | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | -9.90e-03 |
Pos logit contributions | |
wear | +0.024 |
Woo | +0.023 |
bath | +0.022 |
Mitt | +0.022 |
Lucky | +0.020 |
Neg logit contributions | |
fires | -0.055 |
unison | -0.054 |
seventeen | -0.053 |
amaz | -0.052 |
aroma | -0.051 |
Token | tant |
Token ID | 24246 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
unison | +0.013 |
fires | +0.012 |
amaz | +0.012 |
pter | +0.012 |
seventeen | +0.011 |
Neg logit contributions | |
wear | -0.021 |
bath | -0.020 |
Woo | -0.019 |
day | -0.019 |
rubber | -0.018 |
Token | rum |
Token ID | 6582 |
Feature activation | -4.81e-02 |
Pos logit contributions | |
unison | +0.149 |
slices | +0.147 |
seventeen | +0.144 |
amaz | +0.143 |
aroma | +0.138 |
Neg logit contributions | |
wear | -0.283 |
Woo | -0.269 |
bath | -0.264 |
Mitt | -0.255 |
day | -0.253 |
Token | . |
Token ID | 13 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
seventeen | +0.105 |
unison | +0.104 |
slices | +0.103 |
amaz | +0.103 |
aroma | +0.101 |
Neg logit contributions | |
wear | -0.054 |
Woo | -0.050 |
bath | -0.047 |
skipping | -0.046 |
day | -0.045 |
Token | She |
Token ID | 1375 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
slices | +0.019 |
unison | +0.019 |
seventeen | +0.018 |
amaz | +0.018 |
seconds | +0.018 |
Neg logit contributions | |
wear | -0.014 |
Woo | -0.012 |
bath | -0.012 |
burn | -0.012 |
important | -0.012 |
Token | grabbed |
Token ID | 13176 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
unison | +0.037 |
slices | +0.037 |
seventeen | +0.037 |
amaz | +0.037 |
aroma | +0.036 |
Neg logit contributions | |
wear | -0.033 |
Woo | -0.030 |
bath | -0.030 |
Mitt | -0.028 |
Lucky | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | -2.68e-03 |
Pos logit contributions | |
slices | +0.053 |
unison | +0.053 |
seventeen | +0.052 |
amaz | +0.052 |
seconds | +0.051 |
Neg logit contributions | |
wear | -0.030 |
Woo | -0.027 |
bath | -0.027 |
important | -0.025 |
Mitt | -0.024 |
Token | She |
Token ID | 1375 |
Feature activation | -8.85e-03 |
Pos logit contributions | |
wear | +0.005 |
Woo | +0.004 |
bath | +0.004 |
burn | +0.004 |
jog | +0.004 |
Neg logit contributions | |
slices | -0.005 |
amaz | -0.005 |
unison | -0.005 |
seventeen | -0.005 |
als | -0.005 |
Token | cried |
Token ID | 16896 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
seconds | +0.014 |
unison | +0.013 |
slices | +0.013 |
gasped | +0.013 |
amaz | +0.012 |
Neg logit contributions | |
wear | -0.014 |
Woo | -0.014 |
bath | -0.014 |
burn | -0.014 |
skipping | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | +7.14e-03 |
Pos logit contributions | |
wear | +0.017 |
Woo | +0.015 |
bath | +0.015 |
Mitt | +0.014 |
Lucky | +0.014 |
Neg logit contributions | |
unison | -0.030 |
slices | -0.030 |
seventeen | -0.030 |
amaz | -0.030 |
aroma | -0.029 |
Token | threw |
Token ID | 9617 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
wear | +0.010 |
Woo | +0.010 |
bath | +0.010 |
Mitt | +0.009 |
Lucky | +0.009 |
Neg logit contributions | |
unison | -0.013 |
slices | -0.013 |
seventeen | -0.013 |
amaz | -0.013 |
aroma | -0.012 |
Token | a |
Token ID | 257 |
Feature activation | -5.17e-03 |
Pos logit contributions | |
wear | +0.019 |
bath | +0.016 |
Woo | +0.016 |
Mitt | +0.016 |
day | +0.015 |
Neg logit contributions | |
fires | -0.075 |
unison | -0.075 |
seventeen | -0.074 |
amaz | -0.071 |
pulp | -0.071 |
Token | tant |
Token ID | 24246 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
unison | +0.008 |
seventeen | +0.008 |
slices | +0.008 |
amaz | +0.007 |
fires | +0.007 |
Neg logit contributions | |
wear | -0.009 |
bath | -0.009 |
Woo | -0.009 |
day | -0.008 |
Mitt | -0.008 |
Token | rum |
Token ID | 6582 |
Feature activation | -4.63e-02 |
Pos logit contributions | |
unison | +0.152 |
slices | +0.149 |
seventeen | +0.147 |
amaz | +0.146 |
fires | +0.140 |
Neg logit contributions | |
wear | -0.273 |
Woo | -0.259 |
bath | -0.254 |
Mitt | -0.246 |
day | -0.244 |
Token | . |
Token ID | 13 |
Feature activation | -9.61e-03 |
Pos logit contributions | |
seventeen | +0.106 |
unison | +0.104 |
slices | +0.103 |
amaz | +0.103 |
pulp | +0.101 |
Neg logit contributions | |
wear | -0.049 |
Woo | -0.046 |
bath | -0.043 |
day | -0.042 |
Mitt | -0.042 |
Token | Her |
Token ID | 2332 |
Feature activation | +4.41e-02 |
Pos logit contributions | |
slices | +0.017 |
amaz | +0.017 |
unison | +0.016 |
seventeen | +0.016 |
pulp | +0.016 |
Neg logit contributions | |
wear | -0.014 |
burn | -0.012 |
Woo | -0.012 |
bath | -0.012 |
jog | -0.012 |
Token | mom |
Token ID | 1995 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
wear | +0.052 |
Woo | +0.052 |
bath | +0.044 |
Lucky | +0.044 |
burn | +0.044 |
Neg logit contributions | |
slices | -0.092 |
amaz | -0.090 |
unison | -0.088 |
seconds | -0.087 |
pulp | -0.087 |
Token | tried |
Token ID | 3088 |
Feature activation | +5.71e-02 |
Pos logit contributions | |
slices | +0.060 |
unison | +0.060 |
seconds | +0.059 |
amaz | +0.058 |
seventeen | +0.058 |
Neg logit contributions | |
wear | -0.055 |
Woo | -0.052 |
bath | -0.051 |
skipping | -0.048 |
important | -0.047 |
Token | watched |
Token ID | 7342 |
Feature activation | -7.76e-03 |
Pos logit contributions | |
slices | +0.024 |
unison | +0.024 |
seventeen | +0.024 |
fires | +0.024 |
aroma | +0.024 |
Neg logit contributions | |
wear | -0.017 |
Woo | -0.015 |
bath | -0.015 |
skipping | -0.014 |
day | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
seventeen | +0.016 |
unison | +0.016 |
slices | +0.016 |
fires | +0.016 |
pulp | +0.016 |
Neg logit contributions | |
wear | -0.010 |
Woo | -0.008 |
bath | -0.008 |
skipping | -0.008 |
day | -0.008 |
Token | colorful |
Token ID | 20239 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
seventeen | +0.029 |
slices | +0.029 |
pter | +0.028 |
unison | +0.028 |
fires | +0.028 |
Neg logit contributions | |
wear | -0.018 |
day | -0.018 |
bath | -0.016 |
skipping | -0.016 |
important | -0.016 |
Token | lights |
Token ID | 7588 |
Feature activation | +3.33e-03 |
Pos logit contributions | |
unison | +0.025 |
seventeen | +0.025 |
fires | +0.025 |
slices | +0.025 |
pter | +0.024 |
Neg logit contributions | |
wear | -0.016 |
day | -0.015 |
skipping | -0.015 |
bath | -0.014 |
jog | -0.013 |
Token | in |
Token ID | 287 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
wear | +0.004 |
bath | +0.003 |
Woo | +0.003 |
day | +0.003 |
skipping | +0.003 |
Neg logit contributions | |
unison | -0.007 |
slices | -0.007 |
seventeen | -0.007 |
pulp | -0.007 |
seconds | -0.007 |
Token | amaz |
Token ID | 40642 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
seventeen | +0.066 |
fires | +0.065 |
pulp | +0.062 |
pter | +0.061 |
slices | +0.061 |
Neg logit contributions | |
wear | -0.025 |
day | -0.021 |
bath | -0.021 |
skipping | -0.020 |
Woo | -0.019 |
Token | ement |
Token ID | 972 |
Feature activation | -4.62e-02 |
Pos logit contributions | |
seventeen | +0.168 |
pulp | +0.161 |
fires | +0.157 |
unison | +0.153 |
slices | +0.145 |
Neg logit contributions | |
wear | -0.248 |
ie | -0.236 |
day | -0.235 |
bath | -0.233 |
important | -0.231 |
Token | , |
Token ID | 11 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
seventeen | +0.098 |
unison | +0.097 |
slices | +0.097 |
pulp | +0.095 |
seconds | +0.094 |
Neg logit contributions | |
wear | -0.056 |
bath | -0.048 |
day | -0.046 |
Woo | -0.046 |
skipping | -0.046 |
Token | feeling |
Token ID | 4203 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
unison | +0.042 |
slices | +0.041 |
seventeen | +0.041 |
anish | +0.041 |
fires | +0.040 |
Neg logit contributions | |
wear | -0.022 |
skipping | -0.019 |
always | -0.018 |
Woo | -0.018 |
day | -0.017 |
Token | truly |
Token ID | 4988 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
slices | +0.052 |
seventeen | +0.052 |
unison | +0.051 |
pulp | +0.050 |
anish | +0.050 |
Neg logit contributions | |
wear | -0.028 |
Woo | -0.025 |
important | -0.024 |
bath | -0.024 |
skipping | -0.024 |
Token | lucky |
Token ID | 9670 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
seventeen | +0.036 |
slices | +0.036 |
unison | +0.035 |
anish | +0.035 |
pulp | +0.035 |
Neg logit contributions | |
wear | -0.021 |
important | -0.019 |
skipping | -0.018 |
day | -0.018 |
always | -0.017 |
Token | to |
Token ID | 284 |
Feature activation | -1.01e-03 |
Pos logit contributions | |
ongo | +0.091 |
Woo | +0.088 |
wear | +0.087 |
opsy | +0.087 |
scar | +0.087 |
Neg logit contributions | |
slices | -0.102 |
amaz | -0.102 |
seconds | -0.101 |
faces | -0.096 |
moments | -0.092 |
Token | cry |
Token ID | 3960 |
Feature activation | +2.00e-03 |
Pos logit contributions | |
amaz | +0.002 |
unison | +0.002 |
aroma | +0.002 |
gasp | +0.002 |
slices | +0.001 |
Neg logit contributions | |
Woo | -0.002 |
ip | -0.002 |
important | -0.002 |
scar | -0.002 |
ained | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
wear | +0.003 |
Woo | +0.002 |
bath | +0.002 |
skipping | +0.002 |
burn | +0.002 |
Neg logit contributions | |
unison | -0.004 |
slices | -0.004 |
amaz | -0.004 |
seconds | -0.004 |
aroma | -0.004 |
Token | threw |
Token ID | 9617 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
wear | +0.015 |
Woo | +0.014 |
bath | +0.013 |
burn | +0.013 |
important | +0.012 |
Neg logit contributions | |
unison | -0.023 |
amaz | -0.023 |
slices | -0.023 |
seconds | -0.023 |
aroma | -0.023 |
Token | a |
Token ID | 257 |
Feature activation | -5.29e-03 |
Pos logit contributions | |
wear | +0.022 |
Woo | +0.020 |
bath | +0.019 |
Mitt | +0.018 |
day | +0.017 |
Neg logit contributions | |
unison | -0.063 |
seventeen | -0.062 |
fires | -0.061 |
slices | -0.060 |
amaz | -0.060 |
Token | tant |
Token ID | 24246 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
slices | +0.008 |
unison | +0.008 |
seventeen | +0.008 |
amaz | +0.008 |
aroma | +0.008 |
Neg logit contributions | |
wear | -0.009 |
Woo | -0.008 |
bath | -0.008 |
Mitt | -0.008 |
Lucky | -0.008 |
Token | rum |
Token ID | 6582 |
Feature activation | -4.14e-02 |
Pos logit contributions | |
slices | +0.135 |
unison | +0.133 |
seventeen | +0.131 |
amaz | +0.128 |
seconds | +0.126 |
Neg logit contributions | |
wear | -0.282 |
Woo | -0.271 |
bath | -0.268 |
Mitt | -0.257 |
Lucky | -0.256 |
Token | . |
Token ID | 13 |
Feature activation | -5.66e-03 |
Pos logit contributions | |
unison | +0.081 |
slices | +0.081 |
seventeen | +0.080 |
amaz | +0.079 |
aroma | +0.078 |
Neg logit contributions | |
wear | -0.055 |
Woo | -0.051 |
bath | -0.050 |
Mitt | -0.046 |
Lucky | -0.046 |
Token | She |
Token ID | 1375 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
slices | +0.011 |
amaz | +0.011 |
unison | +0.011 |
seventeen | +0.010 |
aroma | +0.010 |
Neg logit contributions | |
wear | -0.007 |
Woo | -0.007 |
bath | -0.007 |
burn | -0.007 |
day | -0.006 |
Token | didn |
Token ID | 1422 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
seconds | +0.021 |
unison | +0.020 |
slices | +0.020 |
gasped | +0.020 |
amaz | +0.019 |
Neg logit contributions | |
wear | -0.023 |
Woo | -0.023 |
bath | -0.023 |
burn | -0.023 |
skipping | -0.022 |
Token | 't |
Token ID | 470 |
Feature activation | +5.74e-02 |
Pos logit contributions | |
slices | +0.012 |
oor | +0.012 |
seventeen | +0.012 |
seconds | +0.011 |
unison | +0.011 |
Neg logit contributions | |
ety | -0.022 |
opsy | -0.022 |
wear | -0.022 |
day | -0.021 |
important | -0.021 |
Token | kept |
Token ID | 4030 |
Feature activation | +4.64e-02 |
Pos logit contributions | |
seconds | +0.005 |
unison | +0.004 |
gasped | +0.004 |
slices | +0.004 |
aroma | +0.004 |
Neg logit contributions | |
wear | -0.005 |
Woo | -0.005 |
bath | -0.005 |
burn | -0.004 |
skipping | -0.004 |
Token | crying |
Token ID | 13774 |
Feature activation | -6.83e-03 |
Pos logit contributions | |
wear | +0.061 |
ongo | +0.055 |
Woo | +0.055 |
bath | +0.054 |
ibi | +0.054 |
Neg logit contributions | |
slices | -0.088 |
seconds | -0.087 |
amaz | -0.084 |
aroma | -0.083 |
unison | -0.082 |
Token | and |
Token ID | 290 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
seventeen | +0.015 |
unison | +0.015 |
slices | +0.015 |
amaz | +0.015 |
pulp | +0.015 |
Neg logit contributions | |
wear | -0.008 |
Woo | -0.007 |
bath | -0.007 |
important | -0.006 |
Mitt | -0.006 |
Token | throwing |
Token ID | 9644 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
wear | +0.013 |
Woo | +0.012 |
bath | +0.012 |
Mitt | +0.011 |
Lucky | +0.011 |
Neg logit contributions | |
unison | -0.023 |
slices | -0.023 |
seventeen | -0.023 |
amaz | -0.023 |
seconds | -0.022 |
Token | a |
Token ID | 257 |
Feature activation | -9.37e-04 |
Pos logit contributions | |
wear | +0.019 |
Woo | +0.018 |
bath | +0.017 |
Mitt | +0.017 |
Lucky | +0.016 |
Neg logit contributions | |
unison | -0.047 |
seventeen | -0.047 |
fires | -0.046 |
amaz | -0.046 |
pter | -0.045 |
Token | tant |
Token ID | 24246 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
unison | +0.001 |
slices | +0.001 |
seventeen | +0.001 |
amaz | +0.001 |
fires | +0.001 |
Neg logit contributions | |
wear | -0.002 |
Woo | -0.002 |
bath | -0.002 |
Mitt | -0.002 |
day | -0.002 |
Token | rum |
Token ID | 6582 |
Feature activation | -3.73e-02 |
Pos logit contributions | |
slices | +0.134 |
unison | +0.132 |
seventeen | +0.130 |
amaz | +0.127 |
seconds | +0.126 |
Neg logit contributions | |
wear | -0.280 |
Woo | -0.269 |
bath | -0.266 |
Mitt | -0.255 |
Lucky | -0.254 |
Token | . |
Token ID | 13 |
Feature activation | +1.33e-04 |
Pos logit contributions | |
unison | +0.077 |
seventeen | +0.076 |
slices | +0.076 |
amaz | +0.075 |
aroma | +0.074 |
Neg logit contributions | |
wear | -0.046 |
Woo | -0.043 |
bath | -0.041 |
Mitt | -0.039 |
day | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | +3.38e-04 |
Pos logit contributions | |
wear | +0.000 |
Woo | +0.000 |
burn | +0.000 |
bath | +0.000 |
jog | +0.000 |
Neg logit contributions | |
slices | -0.000 |
unison | -0.000 |
amaz | -0.000 |
seventeen | -0.000 |
seconds | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
burn | +0.001 |
wear | +0.001 |
Woo | +0.001 |
bath | +0.001 |
exercise | +0.001 |
Neg logit contributions | |
slices | -0.000 |
seconds | -0.000 |
unison | -0.000 |
gasped | -0.000 |
fires | -0.000 |
Token | Finally |
Token ID | 11158 |
Feature activation | -3.52e-02 |
Pos logit contributions | |
unison | +0.017 |
seconds | +0.016 |
slices | +0.015 |
fires | +0.015 |
seventeen | +0.015 |
Neg logit contributions | |
wear | -0.023 |
burn | -0.023 |
Woo | -0.023 |
bath | -0.022 |
Nut | -0.021 |
Token | sad |
Token ID | 6507 |
Feature activation | -6.63e-03 |
Pos logit contributions | |
wear | +0.034 |
bath | +0.033 |
Woo | +0.032 |
burn | +0.032 |
opsy | +0.029 |
Neg logit contributions | |
slices | -0.039 |
unison | -0.039 |
aroma | -0.039 |
seconds | -0.039 |
amaz | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
seventeen | +0.016 |
unison | +0.016 |
slices | +0.015 |
pulp | +0.015 |
amaz | +0.015 |
Neg logit contributions | |
wear | -0.007 |
Woo | -0.006 |
bath | -0.006 |
day | -0.006 |
important | -0.006 |
Token | She |
Token ID | 1375 |
Feature activation | -4.83e-03 |
Pos logit contributions | |
wear | +0.027 |
bath | +0.025 |
Woo | +0.025 |
burn | +0.024 |
important | +0.022 |
Neg logit contributions | |
slices | -0.030 |
amaz | -0.030 |
seventeen | -0.029 |
unison | -0.029 |
seconds | -0.028 |
Token | threw |
Token ID | 9617 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
unison | +0.008 |
seconds | +0.008 |
slices | +0.008 |
seventeen | +0.008 |
amaz | +0.008 |
Neg logit contributions | |
wear | -0.007 |
bath | -0.007 |
Woo | -0.007 |
burn | -0.007 |
skipping | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | +7.80e-04 |
Pos logit contributions | |
wear | +0.029 |
Woo | +0.027 |
bath | +0.025 |
Mitt | +0.025 |
day | +0.024 |
Neg logit contributions | |
unison | -0.070 |
fires | -0.069 |
seventeen | -0.069 |
pulp | -0.066 |
amaz | -0.066 |
Token | tant |
Token ID | 24246 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
wear | +0.001 |
Woo | +0.001 |
bath | +0.001 |
Mitt | +0.001 |
day | +0.001 |
Neg logit contributions | |
unison | -0.001 |
slices | -0.001 |
seventeen | -0.001 |
amaz | -0.001 |
fires | -0.001 |
Token | rum |
Token ID | 6582 |
Feature activation | -4.43e-02 |
Pos logit contributions | |
unison | +0.143 |
slices | +0.141 |
seventeen | +0.138 |
amaz | +0.137 |
aroma | +0.132 |
Neg logit contributions | |
wear | -0.272 |
Woo | -0.259 |
bath | -0.254 |
Mitt | -0.245 |
day | -0.243 |
Token | and |
Token ID | 290 |
Feature activation | +2.28e-04 |
Pos logit contributions | |
seventeen | +0.098 |
unison | +0.098 |
slices | +0.098 |
amaz | +0.096 |
aroma | +0.095 |
Neg logit contributions | |
wear | -0.048 |
Woo | -0.045 |
bath | -0.042 |
skipping | -0.041 |
day | -0.040 |
Token | cried |
Token ID | 16896 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
wear | +0.000 |
Woo | +0.000 |
bath | +0.000 |
Mitt | +0.000 |
Lucky | +0.000 |
Neg logit contributions | |
unison | -0.000 |
slices | -0.000 |
seventeen | -0.000 |
amaz | -0.000 |
aroma | -0.000 |
Token | loudly |
Token ID | 23112 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
wear | +0.017 |
Woo | +0.015 |
bath | +0.015 |
burn | +0.014 |
day | +0.014 |
Neg logit contributions | |
slices | -0.024 |
unison | -0.024 |
seventeen | -0.023 |
amaz | -0.023 |
seconds | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +2.59e-03 |
Pos logit contributions | |
unison | +0.099 |
pulp | +0.098 |
seventeen | +0.097 |
slices | +0.097 |
amaz | +0.097 |
Neg logit contributions | |
wear | -0.041 |
Woo | -0.038 |
bath | -0.036 |
Mitt | -0.035 |
day | -0.035 |
Token | and |
Token ID | 290 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
unison | +0.079 |
slices | +0.078 |
seventeen | +0.078 |
amaz | +0.077 |
aroma | +0.076 |
Neg logit contributions | |
wear | -0.044 |
Woo | -0.040 |
bath | -0.039 |
Mitt | -0.036 |
Lucky | -0.036 |
Token | started |
Token ID | 2067 |
Feature activation | +5.47e-02 |
Pos logit contributions | |
wear | +0.028 |
bath | +0.026 |
Woo | +0.026 |
burn | +0.025 |
day | +0.024 |
Neg logit contributions | |
slices | -0.028 |
aroma | -0.028 |
seconds | -0.028 |
pulp | -0.027 |
unison | -0.027 |
Token | to |
Token ID | 284 |
Feature activation | -3.15e-03 |
Pos logit contributions | |
wear | +0.071 |
Woo | +0.066 |
bath | +0.060 |
ongo | +0.060 |
Lucky | +0.059 |
Neg logit contributions | |
slices | -0.109 |
seconds | -0.104 |
aroma | -0.101 |
pulp | -0.100 |
amaz | -0.100 |
Token | throw |
Token ID | 3714 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
aroma | +0.005 |
slices | +0.004 |
amaz | +0.004 |
unison | +0.004 |
pulp | +0.004 |
Neg logit contributions | |
Woo | -0.005 |
wear | -0.005 |
important | -0.005 |
un | -0.005 |
bath | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | +1.45e-03 |
Pos logit contributions | |
wear | +0.030 |
Woo | +0.029 |
bath | +0.029 |
Mitt | +0.028 |
Lucky | +0.026 |
Neg logit contributions | |
unison | -0.082 |
fires | -0.081 |
seventeen | -0.080 |
amaz | -0.080 |
folk | -0.077 |
Token | tant |
Token ID | 24246 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
wear | +0.002 |
bath | +0.002 |
Woo | +0.002 |
day | +0.002 |
Lucky | +0.002 |
Neg logit contributions | |
unison | -0.003 |
fires | -0.002 |
amaz | -0.002 |
seventeen | -0.002 |
folk | -0.002 |
Token | rum |
Token ID | 6582 |
Feature activation | -4.16e-02 |
Pos logit contributions | |
slices | +0.137 |
unison | +0.137 |
seventeen | +0.134 |
amaz | +0.131 |
aroma | +0.128 |
Neg logit contributions | |
wear | -0.275 |
Woo | -0.263 |
bath | -0.260 |
Mitt | -0.249 |
Lucky | -0.248 |
Token | . |
Token ID | 13 |
Feature activation | -6.12e-04 |
Pos logit contributions | |
unison | +0.082 |
slices | +0.082 |
seventeen | +0.081 |
amaz | +0.080 |
aroma | +0.079 |
Neg logit contributions | |
wear | -0.055 |
Woo | -0.051 |
bath | -0.050 |
Mitt | -0.046 |
Lucky | -0.046 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
slices | +0.001 |
amaz | +0.001 |
unison | +0.001 |
seventeen | +0.001 |
aroma | +0.001 |
Neg logit contributions | |
wear | -0.001 |
Woo | -0.001 |
bath | -0.001 |
burn | -0.001 |
important | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
wear | +0.022 |
Woo | +0.021 |
bath | +0.020 |
burn | +0.019 |
important | +0.019 |
Neg logit contributions | |
slices | -0.035 |
unison | -0.034 |
amaz | -0.034 |
seconds | -0.034 |
seventeen | -0.034 |
Token | she |
Token ID | 673 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
slices | +0.040 |
unison | +0.040 |
seventeen | +0.039 |
amaz | +0.039 |
aroma | +0.039 |
Neg logit contributions | |
wear | -0.017 |
Woo | -0.015 |
bath | -0.015 |
Mitt | -0.013 |
Lucky | -0.013 |
Token | ered |
Token ID | 1068 |
Feature activation | +2.71e-03 |
Pos logit contributions | |
wear | +0.009 |
Woo | +0.009 |
bath | +0.009 |
Mitt | +0.008 |
Lucky | +0.008 |
Neg logit contributions | |
unison | -0.007 |
slices | -0.007 |
seventeen | -0.007 |
amaz | -0.007 |
aroma | -0.007 |
Token | in |
Token ID | 287 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
wear | +0.003 |
Woo | +0.003 |
bath | +0.003 |
skipping | +0.002 |
important | +0.002 |
Neg logit contributions | |
unison | -0.006 |
seventeen | -0.006 |
slices | -0.006 |
aroma | -0.006 |
amaz | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -8.09e-03 |
Pos logit contributions | |
seventeen | +0.058 |
pulp | +0.057 |
fires | +0.057 |
unison | +0.055 |
aroma | +0.055 |
Neg logit contributions | |
wear | -0.025 |
bath | -0.021 |
skipping | -0.021 |
day | -0.020 |
Woo | -0.019 |
Token | blinking |
Token ID | 43196 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
unison | +0.017 |
slices | +0.016 |
seventeen | +0.016 |
amaz | +0.016 |
aroma | +0.016 |
Neg logit contributions | |
wear | -0.010 |
Woo | -0.009 |
bath | -0.009 |
skipping | -0.009 |
Mitt | -0.009 |
Token | in |
Token ID | 287 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
seventeen | +0.037 |
unison | +0.037 |
slices | +0.035 |
pulp | +0.035 |
anish | +0.035 |
Neg logit contributions | |
wear | -0.014 |
important | -0.012 |
skipping | -0.012 |
bath | -0.012 |
Lucky | -0.011 |
Token | amaz |
Token ID | 40642 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
seventeen | +0.070 |
fires | +0.070 |
pulp | +0.068 |
slices | +0.066 |
aroma | +0.066 |
Neg logit contributions | |
wear | -0.030 |
bath | -0.027 |
skipping | -0.026 |
day | -0.026 |
Mitt | -0.023 |
Token | ement |
Token ID | 972 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
pulp | +0.166 |
seventeen | +0.166 |
unison | +0.164 |
fires | +0.156 |
slices | +0.145 |
Neg logit contributions | |
important | -0.228 |
wear | -0.226 |
ie | -0.223 |
day | -0.215 |
idy | -0.214 |
Token | . |
Token ID | 13 |
Feature activation | +9.49e-04 |
Pos logit contributions | |
seventeen | +0.062 |
unison | +0.062 |
pulp | +0.060 |
slices | +0.060 |
amaz | +0.059 |
Neg logit contributions | |
wear | -0.031 |
bath | -0.026 |
day | -0.026 |
skipping | -0.026 |
Woo | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | -4.39e-03 |
Pos logit contributions | |
wear | +0.001 |
Woo | +0.001 |
bath | +0.001 |
Mitt | +0.001 |
Lucky | +0.001 |
Neg logit contributions | |
unison | -0.002 |
slices | -0.002 |
seventeen | -0.002 |
amaz | -0.002 |
aroma | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
unison | +0.009 |
slices | +0.009 |
seventeen | +0.009 |
amaz | +0.008 |
seconds | +0.008 |
Neg logit contributions | |
wear | -0.006 |
Woo | -0.005 |
bath | -0.005 |
burn | -0.005 |
Mitt | -0.005 |
Token | But |
Token ID | 1537 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
unison | +0.028 |
slices | +0.028 |
seventeen | +0.028 |
amaz | +0.028 |
aroma | +0.027 |
Neg logit contributions | |
wear | -0.027 |
Woo | -0.025 |
bath | -0.024 |
Mitt | -0.023 |
Lucky | -0.023 |
Token | surprise |
Token ID | 5975 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
wear | +0.012 |
bath | +0.012 |
day | +0.011 |
important | +0.011 |
Woo | +0.011 |
Neg logit contributions | |
unison | -0.015 |
seventeen | -0.015 |
slices | -0.014 |
fires | -0.014 |
amaz | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | +6.20e-03 |
Pos logit contributions | |
unison | +0.044 |
seventeen | +0.044 |
amaz | +0.043 |
slices | +0.043 |
pulp | +0.042 |
Neg logit contributions | |
wear | -0.027 |
bath | -0.024 |
Woo | -0.024 |
day | -0.023 |
important | -0.022 |
Token | He |
Token ID | 679 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
wear | +0.008 |
Woo | +0.008 |
bath | +0.007 |
Mitt | +0.007 |
Lucky | +0.007 |
Neg logit contributions | |
unison | -0.012 |
slices | -0.012 |
aroma | -0.012 |
seventeen | -0.012 |
pulp | -0.012 |
Token | threw |
Token ID | 9617 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
unison | +0.042 |
slices | +0.042 |
seventeen | +0.042 |
amaz | +0.041 |
aroma | +0.040 |
Neg logit contributions | |
wear | -0.038 |
Woo | -0.035 |
bath | -0.034 |
Mitt | -0.033 |
Lucky | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
Woo | +0.020 |
bath | +0.020 |
wear | +0.020 |
Mitt | +0.019 |
day | +0.019 |
Neg logit contributions | |
fires | -0.072 |
unison | -0.071 |
MacDonald | -0.067 |
edly | -0.066 |
folk | -0.066 |
Token | tant |
Token ID | 24246 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
unison | +0.025 |
fires | +0.025 |
amaz | +0.024 |
pter | +0.024 |
MacDonald | +0.024 |
Neg logit contributions | |
wear | -0.018 |
bath | -0.018 |
Woo | -0.016 |
day | -0.016 |
Lucky | -0.015 |
Token | rum |
Token ID | 6582 |
Feature activation | -5.24e-02 |
Pos logit contributions | |
unison | +0.167 |
amaz | +0.161 |
seventeen | +0.155 |
slices | +0.154 |
fires | +0.152 |
Neg logit contributions | |
wear | -0.247 |
Woo | -0.231 |
bath | -0.223 |
important | -0.222 |
day | -0.220 |
Token | . |
Token ID | 13 |
Feature activation | -4.13e-03 |
Pos logit contributions | |
seventeen | +0.125 |
unison | +0.125 |
amaz | +0.124 |
aroma | +0.123 |
pulp | +0.121 |
Neg logit contributions | |
wear | -0.048 |
Woo | -0.044 |
day | -0.043 |
bath | -0.041 |
always | -0.040 |
Token | He |
Token ID | 679 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
unison | +0.008 |
slices | +0.008 |
aroma | +0.008 |
seventeen | +0.008 |
amaz | +0.008 |
Neg logit contributions | |
wear | -0.006 |
Woo | -0.005 |
bath | -0.005 |
Mitt | -0.005 |
Lucky | -0.005 |
Token | kicked |
Token ID | 12165 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
unison | +0.064 |
amaz | +0.064 |
seventeen | +0.063 |
slices | +0.063 |
fires | +0.062 |
Neg logit contributions | |
wear | -0.050 |
Woo | -0.047 |
bath | -0.045 |
Mitt | -0.043 |
Lucky | -0.043 |
Token | and |
Token ID | 290 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
Woo | +0.012 |
wear | +0.012 |
Mitt | +0.012 |
bath | +0.011 |
Lucky | +0.011 |
Neg logit contributions | |
fires | -0.047 |
unison | -0.046 |
pter | -0.044 |
amaz | -0.044 |
aroma | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | -7.27e-03 |
Pos logit contributions | |
seventeen | +0.029 |
pulp | +0.029 |
slices | +0.028 |
unison | +0.028 |
seconds | +0.027 |
Neg logit contributions | |
wear | -0.012 |
bath | -0.011 |
day | -0.011 |
always | -0.010 |
skipping | -0.010 |
Token | The |
Token ID | 383 |
Feature activation | +2.70e-03 |
Pos logit contributions | |
unison | +0.015 |
slices | +0.015 |
aroma | +0.015 |
seventeen | +0.015 |
amaz | +0.015 |
Neg logit contributions | |
wear | -0.009 |
Woo | -0.009 |
Mitt | -0.008 |
bath | -0.008 |
Lucky | -0.008 |
Token | deer |
Token ID | 20096 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
wear | +0.003 |
Woo | +0.003 |
bath | +0.003 |
day | +0.003 |
important | +0.003 |
Neg logit contributions | |
slices | -0.006 |
seventeen | -0.006 |
unison | -0.006 |
amaz | -0.005 |
pulp | -0.005 |
Token | stared |
Token ID | 18484 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
wear | +0.031 |
Woo | +0.028 |
bath | +0.027 |
day | +0.026 |
skipping | +0.026 |
Neg logit contributions | |
slices | -0.037 |
unison | -0.037 |
seventeen | -0.037 |
amaz | -0.036 |
aroma | -0.036 |
Token | in |
Token ID | 287 |
Feature activation | -2.42e-02 |
Pos logit contributions | |
wear | +0.014 |
Woo | +0.013 |
bath | +0.012 |
Mitt | +0.012 |
Lucky | +0.011 |
Neg logit contributions | |
unison | -0.023 |
slices | -0.023 |
seventeen | -0.022 |
amaz | -0.022 |
aroma | -0.022 |
Token | amaz |
Token ID | 40642 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
fires | +0.037 |
seventeen | +0.034 |
folk | +0.034 |
pulp | +0.034 |
pter | +0.034 |
Neg logit contributions | |
wear | -0.047 |
skipping | -0.045 |
bath | -0.044 |
day | -0.044 |
jog | -0.042 |
Token | ement |
Token ID | 972 |
Feature activation | -4.44e-02 |
Pos logit contributions | |
seventeen | +0.159 |
pulp | +0.154 |
unison | +0.151 |
fires | +0.148 |
slices | +0.143 |
Neg logit contributions | |
wear | -0.243 |
ie | -0.232 |
important | -0.230 |
always | -0.225 |
bath | -0.221 |
Token | . |
Token ID | 13 |
Feature activation | -7.53e-03 |
Pos logit contributions | |
seventeen | +0.103 |
unison | +0.100 |
slices | +0.100 |
pulp | +0.099 |
fires | +0.097 |
Neg logit contributions | |
wear | -0.046 |
day | -0.039 |
bath | -0.039 |
skipping | -0.038 |
Woo | -0.037 |
Token | He |
Token ID | 679 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
unison | +0.015 |
slices | +0.014 |
aroma | +0.014 |
seventeen | +0.014 |
amaz | +0.014 |
Neg logit contributions | |
wear | -0.010 |
Woo | -0.010 |
bath | -0.009 |
Mitt | -0.009 |
Lucky | -0.009 |
Token | had |
Token ID | 550 |
Feature activation | -9.95e-03 |
Pos logit contributions | |
slices | +0.051 |
seventeen | +0.051 |
unison | +0.050 |
aroma | +0.049 |
pulp | +0.049 |
Neg logit contributions | |
wear | -0.039 |
Woo | -0.034 |
always | -0.034 |
bath | -0.033 |
Lucky | -0.033 |
Token | never |
Token ID | 1239 |
Feature activation | +7.76e-04 |
Pos logit contributions | |
unison | +0.022 |
slices | +0.022 |
seventeen | +0.021 |
aroma | +0.021 |
fires | +0.021 |
Neg logit contributions | |
wear | -0.011 |
Woo | -0.010 |
always | -0.010 |
bath | -0.010 |
skipping | -0.010 |
Token | seen |
Token ID | 1775 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
unison | +0.082 |
amaz | +0.078 |
seventeen | +0.078 |
atively | +0.076 |
fires | +0.076 |
Neg logit contributions | |
wear | -0.070 |
day | -0.064 |
bath | -0.062 |
important | -0.061 |
always | -0.060 |
Token | ! |
Token ID | 0 |
Feature activation | -8.47e-03 |
Pos logit contributions | |
wear | +0.012 |
Woo | +0.011 |
bath | +0.011 |
day | +0.011 |
skipping | +0.011 |
Neg logit contributions | |
seventeen | -0.022 |
unison | -0.022 |
amaz | -0.021 |
slices | -0.021 |
pulp | -0.021 |
Token | She |
Token ID | 1375 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
slices | +0.014 |
unison | +0.014 |
amaz | +0.014 |
seventeen | +0.013 |
ah | +0.013 |
Neg logit contributions | |
wear | -0.013 |
burn | -0.012 |
important | -0.012 |
bath | -0.012 |
Woo | -0.012 |
Token | gasped |
Token ID | 45236 |
Feature activation | -3.64e-02 |
Pos logit contributions | |
unison | +0.040 |
slices | +0.039 |
seventeen | +0.039 |
amaz | +0.039 |
aroma | +0.038 |
Neg logit contributions | |
wear | -0.028 |
Woo | -0.025 |
bath | -0.025 |
Mitt | -0.023 |
Lucky | -0.023 |
Token | in |
Token ID | 287 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
unison | +0.081 |
seventeen | +0.081 |
slices | +0.080 |
pulp | +0.078 |
fires | +0.078 |
Neg logit contributions | |
wear | -0.039 |
Woo | -0.036 |
bath | -0.033 |
Mitt | -0.033 |
skipping | -0.033 |
Token | amaz |
Token ID | 40642 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
fires | +0.043 |
seventeen | +0.038 |
works | +0.036 |
ymes | +0.036 |
folk | +0.036 |
Neg logit contributions | |
wear | -0.062 |
skipping | -0.057 |
bath | -0.056 |
day | -0.056 |
Mitt | -0.054 |
Token | ement |
Token ID | 972 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
fires | +0.170 |
seventeen | +0.170 |
unison | +0.169 |
pulp | +0.165 |
slices | +0.149 |
Neg logit contributions | |
important | -0.224 |
wear | -0.224 |
ie | -0.222 |
always | -0.218 |
idy | -0.209 |
Token | , |
Token ID | 11 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
seventeen | +0.072 |
unison | +0.072 |
slices | +0.069 |
pulp | +0.069 |
amaz | +0.069 |
Neg logit contributions | |
wear | -0.038 |
skipping | -0.031 |
Woo | -0.031 |
bath | -0.031 |
day | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
unison | +0.024 |
slices | +0.024 |
seventeen | +0.024 |
fires | +0.023 |
pter | +0.023 |
Neg logit contributions | |
wear | -0.013 |
Woo | -0.011 |
skipping | -0.011 |
bath | -0.010 |
Mitt | -0.010 |
Token | her |
Token ID | 607 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
wear | +0.015 |
Woo | +0.014 |
Mitt | +0.013 |
bath | +0.013 |
skipping | +0.013 |
Neg logit contributions | |
unison | -0.026 |
seventeen | -0.025 |
slices | -0.025 |
fires | -0.024 |
pter | -0.024 |
Token | tears |
Token ID | 10953 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
wear | +0.032 |
day | +0.031 |
bath | +0.030 |
important | +0.029 |
skipping | +0.027 |
Neg logit contributions | |
seventeen | -0.052 |
fires | -0.051 |
unison | -0.050 |
slices | -0.049 |
anish | -0.049 |
Token | ! |
Token ID | 0 |
Feature activation | -6.05e-03 |
Pos logit contributions | |
seventeen | +0.013 |
unison | +0.013 |
slices | +0.012 |
pulp | +0.012 |
fires | +0.012 |
Neg logit contributions | |
wear | -0.012 |
bath | -0.011 |
important | -0.011 |
Woo | -0.011 |
day | -0.011 |
Token | They |
Token ID | 1119 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
slices | +0.011 |
seventeen | +0.011 |
unison | +0.011 |
aroma | +0.011 |
amaz | +0.011 |
Neg logit contributions | |
wear | -0.008 |
Woo | -0.008 |
bath | -0.008 |
Mitt | -0.007 |
Lucky | -0.007 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.86e-03 |
Pos logit contributions | |
seventeen | +0.030 |
slices | +0.029 |
aroma | +0.028 |
MacDonald | +0.028 |
unison | +0.028 |
Neg logit contributions | |
wear | -0.021 |
always | -0.019 |
also | -0.017 |
day | -0.017 |
bath | -0.016 |
Token | around |
Token ID | 1088 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
wear | +0.002 |
Woo | +0.002 |
bath | +0.002 |
important | +0.002 |
day | +0.002 |
Neg logit contributions | |
unison | -0.004 |
slices | -0.004 |
seventeen | -0.004 |
pter | -0.004 |
pulp | -0.004 |
Token | in |
Token ID | 287 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
seventeen | +0.095 |
bridge | +0.092 |
pulp | +0.092 |
unison | +0.091 |
pter | +0.090 |
Neg logit contributions | |
wear | -0.025 |
day | -0.025 |
skipping | -0.022 |
always | -0.022 |
bath | -0.022 |
Token | amaz |
Token ID | 40642 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
seventeen | +0.040 |
bridge | +0.039 |
fires | +0.038 |
pulp | +0.038 |
folk | +0.037 |
Neg logit contributions | |
wear | -0.039 |
day | -0.038 |
bath | -0.037 |
skipping | -0.036 |
important | -0.034 |
Token | ement |
Token ID | 972 |
Feature activation | -4.43e-02 |
Pos logit contributions | |
seventeen | +0.165 |
pulp | +0.154 |
slices | +0.144 |
unison | +0.142 |
fires | +0.141 |
Neg logit contributions | |
wear | -0.241 |
important | -0.234 |
always | -0.228 |
idy | -0.223 |
ie | -0.222 |
Token | and |
Token ID | 290 |
Feature activation | +6.07e-03 |
Pos logit contributions | |
seventeen | +0.109 |
unison | +0.104 |
slices | +0.104 |
pulp | +0.103 |
bridge | +0.102 |
Neg logit contributions | |
wear | -0.041 |
skipping | -0.034 |
day | -0.034 |
bath | -0.033 |
always | -0.032 |
Token | quickly |
Token ID | 2952 |
Feature activation | -9.53e-03 |
Pos logit contributions | |
wear | +0.008 |
always | +0.007 |
day | +0.007 |
skipping | +0.007 |
Lucky | +0.007 |
Neg logit contributions | |
seventeen | -0.013 |
slices | -0.013 |
unison | -0.013 |
anish | -0.012 |
bridge | -0.012 |
Token | closed |
Token ID | 4838 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
seventeen | +0.019 |
unison | +0.019 |
slices | +0.019 |
anish | +0.019 |
fires | +0.019 |
Neg logit contributions | |
wear | -0.013 |
bath | -0.011 |
always | -0.010 |
Woo | -0.010 |
day | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
wear | +0.016 |
Woo | +0.015 |
bath | +0.014 |
skipping | +0.013 |
Lucky | +0.013 |
Neg logit contributions | |
seventeen | -0.033 |
slices | -0.032 |
unison | -0.031 |
fires | -0.031 |
pulp | -0.031 |
Token | curtains |
Token ID | 41160 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
seventeen | +0.060 |
slices | +0.060 |
unison | +0.059 |
pulp | +0.057 |
fires | +0.057 |
Neg logit contributions | |
wear | -0.056 |
bath | -0.056 |
day | -0.051 |
important | -0.050 |
skipping | -0.049 |
Token | , |
Token ID | 11 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
seventeen | +0.071 |
unison | +0.071 |
pulp | +0.069 |
slices | +0.068 |
fires | +0.068 |
Neg logit contributions | |
wear | -0.027 |
bath | -0.024 |
skipping | -0.022 |
Woo | -0.022 |
day | -0.021 |
Token | she |
Token ID | 673 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
slices | +0.025 |
unison | +0.025 |
amaz | +0.025 |
seconds | +0.024 |
seventeen | +0.024 |
Neg logit contributions | |
wear | -0.011 |
Woo | -0.010 |
bath | -0.010 |
burn | -0.009 |
important | -0.009 |
Token | gasped |
Token ID | 45236 |
Feature activation | -3.77e-02 |
Pos logit contributions | |
seventeen | +0.031 |
unison | +0.031 |
slices | +0.030 |
aroma | +0.029 |
anish | +0.029 |
Neg logit contributions | |
wear | -0.027 |
bath | -0.024 |
stub | -0.023 |
always | -0.023 |
skipping | -0.023 |
Token | in |
Token ID | 287 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
seventeen | +0.089 |
unison | +0.086 |
slices | +0.085 |
pulp | +0.084 |
pter | +0.083 |
Neg logit contributions | |
wear | -0.038 |
Woo | -0.034 |
bath | -0.033 |
skipping | -0.032 |
Mitt | -0.032 |
Token | amaz |
Token ID | 40642 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
fires | +0.042 |
seventeen | +0.042 |
pulp | +0.039 |
slices | +0.039 |
folk | +0.039 |
Neg logit contributions | |
wear | -0.062 |
bath | -0.057 |
skipping | -0.057 |
jog | -0.056 |
day | -0.054 |
Token | ement |
Token ID | 972 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
seventeen | +0.152 |
pulp | +0.139 |
unison | +0.137 |
fires | +0.134 |
slices | +0.132 |
Neg logit contributions | |
wear | -0.262 |
important | -0.243 |
always | -0.239 |
bath | -0.239 |
ie | -0.237 |
Token | . |
Token ID | 13 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
seventeen | +0.098 |
unison | +0.094 |
slices | +0.094 |
pulp | +0.093 |
anish | +0.091 |
Neg logit contributions | |
wear | -0.045 |
bath | -0.038 |
skipping | -0.037 |
Woo | -0.036 |
day | -0.036 |
Token | It |
Token ID | 632 |
Feature activation | +1.11e-05 |
Pos logit contributions | |
amaz | +0.022 |
unison | +0.022 |
slices | +0.022 |
seventeen | +0.022 |
seconds | +0.021 |
Neg logit contributions | |
wear | -0.021 |
Woo | -0.020 |
bath | -0.019 |
burn | -0.019 |
important | -0.018 |
Token | was |
Token ID | 373 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
wear | +0.000 |
bath | +0.000 |
always | +0.000 |
important | +0.000 |
day | +0.000 |
Neg logit contributions | |
slices | -0.000 |
seventeen | -0.000 |
unison | -0.000 |
pulp | -0.000 |
pter | -0.000 |
Token | snow |
Token ID | 6729 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
unison | +0.030 |
slices | +0.030 |
amaz | +0.030 |
seventeen | +0.030 |
seconds | +0.029 |
Neg logit contributions | |
wear | -0.021 |
Woo | -0.019 |
bath | -0.018 |
burn | -0.017 |
Lucky | -0.017 |
Token | looked |
Token ID | 3114 |
Feature activation | +2.59e-03 |
Pos logit contributions | |
seventeen | +0.031 |
unison | +0.031 |
slices | +0.030 |
aroma | +0.030 |
amaz | +0.029 |
Neg logit contributions | |
wear | -0.025 |
bath | -0.022 |
Woo | -0.022 |
day | -0.021 |
skipping | -0.021 |
Token | at |
Token ID | 379 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
wear | +0.003 |
Woo | +0.002 |
bath | +0.002 |
important | +0.002 |
day | +0.002 |
Neg logit contributions | |
unison | -0.006 |
slices | -0.006 |
seventeen | -0.006 |
seconds | -0.006 |
fires | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
seventeen | +0.074 |
fires | +0.071 |
slices | +0.071 |
unison | +0.071 |
aroma | +0.070 |
Neg logit contributions | |
wear | -0.030 |
day | -0.028 |
Woo | -0.026 |
bath | -0.026 |
Mitt | -0.025 |
Token | rainbow |
Token ID | 27223 |
Feature activation | +2.85e-03 |
Pos logit contributions | |
unison | +0.033 |
seventeen | +0.033 |
slices | +0.032 |
pter | +0.032 |
fires | +0.031 |
Neg logit contributions | |
day | -0.022 |
bath | -0.021 |
wear | -0.020 |
important | -0.019 |
always | -0.018 |
Token | in |
Token ID | 287 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
wear | +0.003 |
bath | +0.003 |
Woo | +0.003 |
day | +0.003 |
skipping | +0.003 |
Neg logit contributions | |
unison | -0.006 |
seventeen | -0.006 |
slices | -0.006 |
amaz | -0.006 |
seconds | -0.006 |
Token | amaz |
Token ID | 40642 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
seventeen | +0.060 |
fires | +0.059 |
bridge | +0.058 |
pter | +0.058 |
edly | +0.057 |
Neg logit contributions | |
wear | -0.019 |
day | -0.017 |
bath | -0.017 |
Woo | -0.015 |
skipping | -0.015 |
Token | ement |
Token ID | 972 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
seventeen | +0.156 |
pulp | +0.145 |
unison | +0.142 |
fires | +0.140 |
slices | +0.132 |
Neg logit contributions | |
wear | -0.245 |
bath | -0.233 |
day | -0.232 |
important | -0.231 |
ie | -0.229 |
Token | and |
Token ID | 290 |
Feature activation | +5.32e-03 |
Pos logit contributions | |
seventeen | +0.093 |
unison | +0.090 |
slices | +0.089 |
pulp | +0.088 |
amaz | +0.088 |
Neg logit contributions | |
wear | -0.042 |
day | -0.036 |
bath | -0.035 |
skipping | -0.034 |
Woo | -0.033 |
Token | laughed |
Token ID | 13818 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
wear | +0.007 |
always | +0.006 |
day | +0.006 |
skipping | +0.006 |
also | +0.006 |
Neg logit contributions | |
bridge | -0.012 |
seventeen | -0.012 |
slices | -0.011 |
unison | -0.011 |
anish | -0.011 |
Token | with |
Token ID | 351 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
seventeen | +0.039 |
pulp | +0.038 |
slices | +0.038 |
pter | +0.037 |
unison | +0.037 |
Neg logit contributions | |
wear | -0.016 |
Woo | -0.014 |
bath | -0.014 |
important | -0.014 |
day | -0.013 |
Token | joy |
Token ID | 8716 |
Feature activation | -5.06e-02 |
Pos logit contributions | |
fires | +0.049 |
pter | +0.047 |
seventeen | +0.047 |
pulp | +0.045 |
boards | +0.045 |
Neg logit contributions | |
wear | -0.047 |
day | -0.044 |
bath | -0.042 |
skipping | -0.041 |
important | -0.041 |
Token | looked |
Token ID | 3114 |
Feature activation | +2.59e-03 |
Pos logit contributions | |
seventeen | +0.031 |
unison | +0.031 |
slices | +0.030 |
aroma | +0.030 |
amaz | +0.029 |
Neg logit contributions | |
wear | -0.025 |
bath | -0.022 |
Woo | -0.022 |
day | -0.021 |
skipping | -0.021 |
Token | at |
Token ID | 379 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
wear | +0.003 |
Woo | +0.002 |
bath | +0.002 |
important | +0.002 |
day | +0.002 |
Neg logit contributions | |
unison | -0.006 |
slices | -0.006 |
seventeen | -0.006 |
seconds | -0.006 |
fires | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
seventeen | +0.074 |
fires | +0.071 |
slices | +0.071 |
unison | +0.071 |
aroma | +0.070 |
Neg logit contributions | |
wear | -0.030 |
day | -0.028 |
Woo | -0.026 |
bath | -0.026 |
Mitt | -0.025 |
Token | rainbow |
Token ID | 27223 |
Feature activation | +2.85e-03 |
Pos logit contributions | |
unison | +0.033 |
seventeen | +0.033 |
slices | +0.032 |
pter | +0.032 |
fires | +0.031 |
Neg logit contributions | |
day | -0.022 |
bath | -0.021 |
wear | -0.020 |
important | -0.019 |
always | -0.018 |
Token | in |
Token ID | 287 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
wear | +0.003 |
bath | +0.003 |
Woo | +0.003 |
day | +0.003 |
skipping | +0.003 |
Neg logit contributions | |
unison | -0.006 |
seventeen | -0.006 |
slices | -0.006 |
amaz | -0.006 |
seconds | -0.006 |
Token | amaz |
Token ID | 40642 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
seventeen | +0.060 |
fires | +0.059 |
bridge | +0.058 |
pter | +0.058 |
edly | +0.057 |
Neg logit contributions | |
wear | -0.019 |
day | -0.017 |
bath | -0.017 |
Woo | -0.015 |
skipping | -0.015 |
Token | ement |
Token ID | 972 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
seventeen | +0.156 |
pulp | +0.145 |
unison | +0.142 |
fires | +0.140 |
slices | +0.132 |
Neg logit contributions | |
wear | -0.245 |
bath | -0.233 |
day | -0.232 |
important | -0.231 |
ie | -0.229 |
Token | and |
Token ID | 290 |
Feature activation | +5.32e-03 |
Pos logit contributions | |
seventeen | +0.093 |
unison | +0.090 |
slices | +0.089 |
pulp | +0.088 |
amaz | +0.088 |
Neg logit contributions | |
wear | -0.042 |
day | -0.036 |
bath | -0.035 |
skipping | -0.034 |
Woo | -0.033 |
Token | laughed |
Token ID | 13818 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
wear | +0.007 |
always | +0.006 |
day | +0.006 |
skipping | +0.006 |
also | +0.006 |
Neg logit contributions | |
bridge | -0.012 |
seventeen | -0.012 |
slices | -0.011 |
unison | -0.011 |
anish | -0.011 |
Token | with |
Token ID | 351 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
seventeen | +0.039 |
pulp | +0.038 |
slices | +0.038 |
pter | +0.037 |
unison | +0.037 |
Neg logit contributions | |
wear | -0.016 |
Woo | -0.014 |
bath | -0.014 |
important | -0.014 |
day | -0.013 |
Token | joy |
Token ID | 8716 |
Feature activation | -5.06e-02 |
Pos logit contributions | |
fires | +0.049 |
pter | +0.047 |
seventeen | +0.047 |
pulp | +0.045 |
boards | +0.045 |
Neg logit contributions | |
wear | -0.047 |
day | -0.044 |
bath | -0.042 |
skipping | -0.041 |
important | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +1.61e-03 |
Pos logit contributions | |
seventeen | +0.121 |
pulp | +0.120 |
slices | +0.118 |
unison | +0.116 |
amaz | +0.113 |
Neg logit contributions | |
wear | -0.050 |
Woo | -0.045 |
day | -0.044 |
important | -0.042 |
bath | -0.042 |
Token | Tiger |
Token ID | 17030 |
Feature activation | +9.36e-03 |
Pos logit contributions | |
wear | +0.002 |
Woo | +0.002 |
bath | +0.002 |
Mitt | +0.002 |
Lucky | +0.002 |
Neg logit contributions | |
slices | -0.003 |
unison | -0.003 |
seventeen | -0.003 |
aroma | -0.003 |
amaz | -0.003 |
Token | looked |
Token ID | 3114 |
Feature activation | +6.20e-03 |
Pos logit contributions | |
wear | +0.013 |
Woo | +0.012 |
bath | +0.011 |
Lucky | +0.011 |
Mitt | +0.011 |
Neg logit contributions | |
slices | -0.018 |
unison | -0.018 |
seventeen | -0.018 |
amaz | -0.018 |
pulp | -0.017 |
Token | around |
Token ID | 1088 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
wear | +0.006 |
Woo | +0.005 |
bath | +0.005 |
important | +0.005 |
day | +0.005 |
Neg logit contributions | |
slices | -0.015 |
unison | -0.015 |
seventeen | -0.015 |
pulp | -0.014 |
seconds | -0.014 |
Token | in |
Token ID | 287 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
seventeen | +0.096 |
pulp | +0.096 |
slices | +0.093 |
unison | +0.092 |
bridge | +0.091 |
Neg logit contributions | |
day | -0.025 |
wear | -0.025 |
skipping | -0.022 |
always | -0.021 |
bath | -0.021 |
Token | amaz |
Token ID | 40642 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
seventeen | +0.031 |
pulp | +0.031 |
fires | +0.030 |
anish | +0.029 |
slices | +0.029 |
Neg logit contributions | |
wear | -0.033 |
day | -0.031 |
bath | -0.031 |
skipping | -0.031 |
jog | -0.030 |
Token | ement |
Token ID | 972 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
pulp | +0.169 |
seventeen | +0.169 |
slices | +0.155 |
unison | +0.150 |
fires | +0.147 |
Neg logit contributions | |
wear | -0.226 |
important | -0.223 |
always | -0.220 |
ie | -0.214 |
day | -0.210 |
Token | and |
Token ID | 290 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
seventeen | +0.087 |
unison | +0.085 |
slices | +0.085 |
pulp | +0.084 |
aroma | +0.082 |
Neg logit contributions | |
wear | -0.036 |
day | -0.030 |
bath | -0.030 |
skipping | -0.030 |
Woo | -0.029 |
Token | he |
Token ID | 339 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
wear | +0.021 |
Woo | +0.019 |
day | +0.018 |
Lucky | +0.018 |
bath | +0.018 |
Neg logit contributions | |
unison | -0.032 |
slices | -0.032 |
seventeen | -0.031 |
pulp | -0.030 |
anish | -0.030 |
Token | knew |
Token ID | 2993 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
seventeen | +0.022 |
slices | +0.022 |
unison | +0.022 |
aroma | +0.021 |
pulp | +0.021 |
Neg logit contributions | |
wear | -0.018 |
Woo | -0.016 |
bath | -0.016 |
always | -0.015 |
Lucky | -0.015 |
Token | he |
Token ID | 339 |
Feature activation | -6.68e-03 |
Pos logit contributions | |
amaz | +0.061 |
unison | +0.060 |
seconds | +0.060 |
slices | +0.059 |
fires | +0.058 |
Neg logit contributions | |
wear | -0.048 |
Woo | -0.046 |
bath | -0.045 |
burn | -0.044 |
Mitt | -0.042 |
Token | ice |
Token ID | 4771 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
unison | +0.026 |
slices | +0.025 |
seventeen | +0.025 |
fires | +0.024 |
amaz | +0.024 |
Neg logit contributions | |
wear | -0.016 |
bath | -0.015 |
Woo | -0.015 |
day | -0.014 |
important | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -8.43e-03 |
Pos logit contributions | |
wear | +0.025 |
Woo | +0.023 |
bath | +0.023 |
day | +0.022 |
important | +0.022 |
Neg logit contributions | |
unison | -0.025 |
seventeen | -0.025 |
amaz | -0.025 |
slices | -0.025 |
seconds | -0.024 |
Token | They |
Token ID | 1119 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
pter | +0.017 |
aroma | +0.017 |
seventeen | +0.017 |
slices | +0.017 |
bridge | +0.016 |
Neg logit contributions | |
bath | -0.010 |
Lucky | -0.010 |
wear | -0.010 |
Woo | -0.010 |
D | -0.010 |
Token | stared |
Token ID | 18484 |
Feature activation | +8.17e-03 |
Pos logit contributions | |
unison | +0.026 |
seventeen | +0.026 |
slices | +0.025 |
aroma | +0.025 |
amaz | +0.025 |
Neg logit contributions | |
wear | -0.024 |
Woo | -0.021 |
bath | -0.021 |
Mitt | -0.020 |
skipping | -0.020 |
Token | in |
Token ID | 287 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
wear | +0.009 |
Woo | +0.008 |
bath | +0.008 |
skipping | +0.008 |
important | +0.008 |
Neg logit contributions | |
unison | -0.018 |
seventeen | -0.018 |
slices | -0.018 |
pulp | -0.017 |
aroma | -0.017 |
Token | amaz |
Token ID | 40642 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
fires | +0.043 |
bridge | +0.043 |
seventeen | +0.042 |
anish | +0.042 |
pulp | +0.042 |
Neg logit contributions | |
wear | -0.032 |
skipping | -0.030 |
bath | -0.030 |
day | -0.029 |
Woo | -0.028 |
Token | ement |
Token ID | 972 |
Feature activation | -4.87e-02 |
Pos logit contributions | |
seventeen | +0.167 |
pulp | +0.165 |
unison | +0.154 |
fires | +0.150 |
slices | +0.149 |
Neg logit contributions | |
wear | -0.231 |
important | -0.223 |
ie | -0.217 |
always | -0.212 |
idy | -0.212 |
Token | . |
Token ID | 13 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
seventeen | +0.117 |
pulp | +0.112 |
unison | +0.112 |
slices | +0.112 |
amaz | +0.110 |
Neg logit contributions | |
wear | -0.048 |
day | -0.041 |
bath | -0.039 |
skipping | -0.038 |
important | -0.037 |
Token | They |
Token ID | 1119 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
pter | +0.025 |
aroma | +0.024 |
seventeen | +0.024 |
slices | +0.024 |
ymes | +0.024 |
Neg logit contributions | |
bath | -0.012 |
Lucky | -0.012 |
wear | -0.012 |
Woo | -0.012 |
Mitt | -0.012 |
Token | wanted |
Token ID | 2227 |
Feature activation | -2.93e-03 |
Pos logit contributions | |
seventeen | +0.029 |
slices | +0.029 |
unison | +0.029 |
fires | +0.028 |
aroma | +0.028 |
Neg logit contributions | |
wear | -0.026 |
Woo | -0.023 |
bath | -0.022 |
always | -0.022 |
skipping | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | -7.35e-03 |
Pos logit contributions | |
unison | +0.006 |
slices | +0.006 |
seventeen | +0.006 |
amaz | +0.006 |
seconds | +0.006 |
Neg logit contributions | |
wear | -0.003 |
Woo | -0.003 |
bath | -0.003 |
Mitt | -0.003 |
Lucky | -0.003 |
Token | She |
Token ID | 1375 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
slices | +0.024 |
unison | +0.024 |
seventeen | +0.024 |
amaz | +0.023 |
aroma | +0.023 |
Neg logit contributions | |
wear | -0.015 |
Woo | -0.014 |
bath | -0.014 |
Mitt | -0.013 |
day | -0.013 |
Token | opened |
Token ID | 4721 |
Feature activation | +2.28e-03 |
Pos logit contributions | |
slices | +0.033 |
seventeen | +0.032 |
unison | +0.032 |
aroma | +0.031 |
pulp | +0.031 |
Neg logit contributions | |
wear | -0.024 |
bath | -0.021 |
Woo | -0.021 |
day | -0.020 |
important | -0.020 |
Token | her |
Token ID | 607 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
wear | +0.002 |
Woo | +0.002 |
bath | +0.002 |
skipping | +0.002 |
day | +0.001 |
Neg logit contributions | |
seventeen | -0.006 |
unison | -0.006 |
fires | -0.006 |
slices | -0.006 |
pulp | -0.005 |
Token | mouth |
Token ID | 5422 |
Feature activation | -4.91e-03 |
Pos logit contributions | |
seventeen | +0.016 |
slices | +0.015 |
unison | +0.015 |
fires | +0.015 |
pulp | +0.014 |
Neg logit contributions | |
wear | -0.033 |
bath | -0.032 |
skipping | -0.030 |
day | -0.030 |
important | -0.029 |
Token | in |
Token ID | 287 |
Feature activation | -4.10e-02 |
Pos logit contributions | |
unison | +0.010 |
slices | +0.010 |
seventeen | +0.010 |
amaz | +0.010 |
pulp | +0.010 |
Neg logit contributions | |
wear | -0.006 |
Woo | -0.005 |
bath | -0.005 |
Mitt | -0.005 |
skipping | -0.005 |
Token | amaz |
Token ID | 40642 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
fires | +0.099 |
seventeen | +0.098 |
pulp | +0.095 |
pter | +0.093 |
folk | +0.093 |
Neg logit contributions | |
wear | -0.044 |
bath | -0.036 |
skipping | -0.036 |
day | -0.035 |
Woo | -0.032 |
Token | ement |
Token ID | 972 |
Feature activation | -4.08e-02 |
Pos logit contributions | |
seventeen | +0.147 |
fires | +0.139 |
pulp | +0.138 |
unison | +0.136 |
slices | +0.131 |
Neg logit contributions | |
wear | -0.260 |
important | -0.238 |
bath | -0.238 |
skipping | -0.236 |
always | -0.234 |
Token | . |
Token ID | 13 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
seventeen | +0.087 |
unison | +0.086 |
slices | +0.086 |
pulp | +0.084 |
seconds | +0.084 |
Neg logit contributions | |
wear | -0.049 |
bath | -0.042 |
Woo | -0.041 |
skipping | -0.040 |
day | -0.040 |
Token | She |
Token ID | 1375 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
slices | +0.033 |
unison | +0.033 |
seventeen | +0.033 |
amaz | +0.032 |
aroma | +0.032 |
Neg logit contributions | |
wear | -0.028 |
Woo | -0.026 |
bath | -0.026 |
Mitt | -0.024 |
skipping | -0.024 |
Token | had |
Token ID | 550 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
slices | +0.043 |
seventeen | +0.042 |
unison | +0.042 |
fires | +0.041 |
aroma | +0.041 |
Neg logit contributions | |
wear | -0.033 |
bath | -0.028 |
always | -0.028 |
skipping | -0.027 |
Woo | -0.027 |
Token | never |
Token ID | 1239 |
Feature activation | -1.48e-03 |
Pos logit contributions | |
unison | +0.023 |
seventeen | +0.023 |
slices | +0.023 |
fires | +0.023 |
aroma | +0.023 |
Neg logit contributions | |
wear | -0.014 |
always | -0.012 |
bath | -0.012 |
Woo | -0.012 |
skipping | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | +8.35e-03 |
Pos logit contributions | |
wear | +0.007 |
Woo | +0.006 |
burn | +0.006 |
ibi | +0.006 |
stub | +0.006 |
Neg logit contributions | |
seconds | -0.009 |
slices | -0.009 |
fires | -0.008 |
unison | -0.008 |
amaz | -0.008 |
Token | be |
Token ID | 307 |
Feature activation | +2.95e-03 |
Pos logit contributions | |
ip | +0.012 |
Woo | +0.012 |
scar | +0.011 |
Mitt | +0.011 |
ibi | +0.011 |
Neg logit contributions | |
unison | -0.014 |
click | -0.013 |
amaz | -0.013 |
seconds | -0.013 |
aroma | -0.013 |
Token | careful |
Token ID | 8161 |
Feature activation | +3.72e-02 |
Pos logit contributions | |
wear | +0.004 |
bath | +0.004 |
Woo | +0.004 |
ibi | +0.004 |
Lucky | +0.004 |
Neg logit contributions | |
seconds | -0.005 |
slices | -0.005 |
fires | -0.005 |
unison | -0.005 |
pulp | -0.005 |
Token | because |
Token ID | 780 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
Woo | +0.050 |
wear | +0.049 |
bath | +0.047 |
Lucky | +0.045 |
opsy | +0.044 |
Neg logit contributions | |
slices | -0.069 |
fires | -0.068 |
seconds | -0.066 |
unison | -0.065 |
pter | -0.065 |
Token | his |
Token ID | 465 |
Feature activation | +3.82e-02 |
Pos logit contributions | |
ongo | +0.081 |
wear | +0.080 |
ibi | +0.079 |
burn | +0.078 |
opsy | +0.077 |
Neg logit contributions | |
slices | -0.057 |
seconds | -0.057 |
lava | -0.057 |
statues | -0.056 |
pulp | -0.056 |
Token | wings |
Token ID | 12098 |
Feature activation | +6.93e-02 |
Pos logit contributions | |
wear | +0.049 |
Woo | +0.048 |
Lucky | +0.043 |
Mitt | +0.042 |
bath | +0.042 |
Neg logit contributions | |
slices | -0.074 |
pulp | -0.073 |
amaz | -0.073 |
seconds | -0.072 |
unison | -0.072 |
Token | were |
Token ID | 547 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
wear | +0.130 |
Woo | +0.121 |
bath | +0.120 |
day | +0.114 |
important | +0.114 |
Neg logit contributions | |
unison | -0.099 |
slices | -0.098 |
seventeen | -0.097 |
amaz | -0.096 |
seconds | -0.094 |
Token | fragile |
Token ID | 21049 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
slices | +0.027 |
seconds | +0.027 |
seventeen | +0.026 |
pulp | +0.026 |
aroma | +0.025 |
Neg logit contributions | |
wear | -0.020 |
Woo | -0.019 |
bath | -0.019 |
ibi | -0.018 |
scar | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
wear | +0.051 |
Woo | +0.048 |
bath | +0.046 |
Mitt | +0.044 |
burn | +0.044 |
Neg logit contributions | |
slices | -0.063 |
unison | -0.062 |
seconds | -0.061 |
aroma | -0.060 |
amaz | -0.060 |
Token | could |
Token ID | 714 |
Feature activation | +3.77e-02 |
Pos logit contributions | |
wear | +0.027 |
Woo | +0.026 |
bath | +0.025 |
burn | +0.024 |
skipping | +0.023 |
Neg logit contributions | |
slices | -0.040 |
seventeen | -0.040 |
amaz | -0.039 |
seconds | -0.039 |
unison | -0.039 |
Token | break |
Token ID | 2270 |
Feature activation | +3.33e-02 |
Pos logit contributions | |
Woo | +0.047 |
wear | +0.044 |
scar | +0.044 |
bath | +0.044 |
ido | +0.043 |
Neg logit contributions | |
slices | -0.069 |
seventeen | -0.068 |
seconds | -0.067 |
pulp | -0.066 |
unison | -0.066 |
Token | winning |
Token ID | 5442 |
Feature activation | -6.11e-04 |
Pos logit contributions | |
wear | +0.048 |
bath | +0.045 |
Woo | +0.044 |
opsy | +0.042 |
Mitt | +0.041 |
Neg logit contributions | |
unison | -0.074 |
slices | -0.074 |
seconds | -0.073 |
amaz | -0.073 |
seventeen | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | +1.43e-03 |
Pos logit contributions | |
seventeen | +0.001 |
slices | +0.001 |
pulp | +0.001 |
unison | +0.001 |
aroma | +0.001 |
Neg logit contributions | |
wear | -0.001 |
Woo | -0.001 |
bath | -0.001 |
important | -0.001 |
day | -0.001 |
Token | The |
Token ID | 383 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
wear | +0.002 |
Woo | +0.002 |
Mitt | +0.001 |
Lucky | +0.001 |
bath | +0.001 |
Neg logit contributions | |
aroma | -0.003 |
slices | -0.003 |
unison | -0.003 |
amaz | -0.003 |
seventeen | -0.003 |
Token | persistent |
Token ID | 16218 |
Feature activation | -4.48e-04 |
Pos logit contributions | |
wear | +0.020 |
Woo | +0.019 |
bath | +0.018 |
burn | +0.017 |
Mitt | +0.017 |
Neg logit contributions | |
unison | -0.032 |
slices | -0.032 |
seventeen | -0.032 |
amaz | -0.031 |
seconds | -0.031 |
Token | friend |
Token ID | 1545 |
Feature activation | +5.41e-03 |
Pos logit contributions | |
unison | +0.001 |
slices | +0.001 |
seconds | +0.001 |
seventeen | +0.001 |
amaz | +0.001 |
Neg logit contributions | |
wear | -0.000 |
Woo | -0.000 |
bath | -0.000 |
Mitt | -0.000 |
burn | -0.000 |
Token | tried |
Token ID | 3088 |
Feature activation | +5.52e-02 |
Pos logit contributions | |
wear | +0.008 |
Woo | +0.007 |
bath | +0.007 |
Mitt | +0.007 |
Lucky | +0.007 |
Neg logit contributions | |
unison | -0.010 |
slices | -0.010 |
seventeen | -0.010 |
amaz | -0.010 |
seconds | -0.010 |
Token | their |
Token ID | 511 |
Feature activation | -3.44e-03 |
Pos logit contributions | |
wear | +0.080 |
opsy | +0.075 |
bath | +0.073 |
Woo | +0.072 |
burn | +0.072 |
Neg logit contributions | |
slices | -0.095 |
amaz | -0.094 |
seconds | -0.094 |
unison | -0.092 |
seventeen | -0.090 |
Token | best |
Token ID | 1266 |
Feature activation | +3.79e-03 |
Pos logit contributions | |
slices | +0.007 |
unison | +0.007 |
seconds | +0.007 |
amaz | +0.007 |
seventeen | +0.007 |
Neg logit contributions | |
wear | -0.004 |
Woo | -0.004 |
burn | -0.004 |
bath | -0.004 |
opsy | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -2.00e-03 |
Pos logit contributions | |
wear | +0.005 |
Woo | +0.005 |
bath | +0.005 |
Mitt | +0.004 |
burn | +0.004 |
Neg logit contributions | |
unison | -0.007 |
slices | -0.007 |
amaz | -0.007 |
seventeen | -0.007 |
seconds | -0.007 |
Token | but |
Token ID | 475 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
slices | +0.003 |
unison | +0.003 |
amaz | +0.003 |
seventeen | +0.003 |
seconds | +0.003 |
Neg logit contributions | |
wear | -0.003 |
Woo | -0.003 |
bath | -0.003 |
burn | -0.003 |
Mitt | -0.003 |
Token | they |
Token ID | 484 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
wear | +0.016 |
Woo | +0.015 |
bath | +0.015 |
burn | +0.014 |
Mitt | +0.014 |
Neg logit contributions | |
slices | -0.020 |
unison | -0.020 |
amaz | -0.020 |
seventeen | -0.020 |
seconds | -0.020 |
Token | He |
Token ID | 679 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
wear | +0.012 |
Woo | +0.011 |
bath | +0.011 |
burn | +0.011 |
important | +0.011 |
Neg logit contributions | |
unison | -0.015 |
slices | -0.015 |
seventeen | -0.015 |
amaz | -0.015 |
seconds | -0.014 |
Token | felt |
Token ID | 2936 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
unison | +0.029 |
seconds | +0.029 |
slices | +0.029 |
amaz | +0.028 |
seventeen | +0.028 |
Neg logit contributions | |
wear | -0.028 |
Woo | -0.027 |
bath | -0.026 |
burn | -0.025 |
Mitt | -0.025 |
Token | like |
Token ID | 588 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
wear | +0.015 |
burn | +0.014 |
Woo | +0.014 |
bath | +0.013 |
ibi | +0.013 |
Neg logit contributions | |
unison | -0.023 |
amaz | -0.023 |
seconds | -0.022 |
slices | -0.022 |
seventeen | -0.022 |
Token | he |
Token ID | 339 |
Feature activation | +6.13e-03 |
Pos logit contributions | |
unison | +0.029 |
slices | +0.028 |
seventeen | +0.028 |
amaz | +0.028 |
aroma | +0.027 |
Neg logit contributions | |
wear | -0.016 |
Woo | -0.015 |
bath | -0.014 |
Mitt | -0.013 |
day | -0.013 |
Token | didn |
Token ID | 1422 |
Feature activation | -4.31e-03 |
Pos logit contributions | |
wear | +0.009 |
Woo | +0.009 |
bath | +0.009 |
burn | +0.008 |
Mitt | +0.008 |
Neg logit contributions | |
unison | -0.011 |
slices | -0.010 |
amaz | -0.010 |
seconds | -0.010 |
fires | -0.010 |
Token | 't |
Token ID | 470 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
seconds | +0.006 |
slices | +0.005 |
seventeen | +0.005 |
oor | +0.005 |
thirteen | +0.005 |
Neg logit contributions | |
ety | -0.008 |
opsy | -0.007 |
wear | -0.007 |
burn | -0.007 |
Woo | -0.007 |
Token | belong |
Token ID | 5594 |
Feature activation | +3.36e-03 |
Pos logit contributions | |
Woo | +0.083 |
burn | +0.081 |
bath | +0.080 |
scar | +0.079 |
opsy | +0.078 |
Neg logit contributions | |
seconds | -0.080 |
fires | -0.073 |
slices | -0.071 |
amaz | -0.071 |
unison | -0.071 |
Token | . |
Token ID | 13 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
wear | +0.004 |
Woo | +0.004 |
bath | +0.004 |
Mitt | +0.004 |
burn | +0.004 |
Neg logit contributions | |
slices | -0.007 |
fires | -0.006 |
seconds | -0.006 |
unison | -0.006 |
amaz | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +7.44e-03 |
Pos logit contributions | |
wear | +0.019 |
Woo | +0.018 |
bath | +0.017 |
burn | +0.017 |
skipping | +0.016 |
Neg logit contributions | |
unison | -0.024 |
slices | -0.024 |
amaz | -0.023 |
seventeen | -0.023 |
seconds | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | -6.72e-03 |
Pos logit contributions | |
wear | +0.012 |
burn | +0.012 |
Woo | +0.011 |
bath | +0.011 |
skipping | +0.010 |
Neg logit contributions | |
unison | -0.012 |
slices | -0.012 |
seconds | -0.012 |
fires | -0.011 |
seventeen | -0.011 |
Token | M |
Token ID | 44 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
unison | +0.010 |
slices | +0.010 |
seconds | +0.010 |
seventeen | +0.009 |
fires | +0.009 |
Neg logit contributions | |
wear | -0.012 |
Woo | -0.012 |
burn | -0.011 |
bath | -0.011 |
skipping | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | +2.09e-03 |
Pos logit contributions | |
slices | +0.043 |
amaz | +0.043 |
aroma | +0.042 |
fires | +0.041 |
unison | +0.041 |
Neg logit contributions | |
wear | -0.030 |
Woo | -0.030 |
bath | -0.027 |
burn | -0.027 |
Lucky | -0.027 |
Token | Lucy |
Token ID | 22162 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
wear | +0.003 |
Woo | +0.002 |
bath | +0.002 |
day | +0.002 |
Mitt | +0.002 |
Neg logit contributions | |
seventeen | -0.004 |
unison | -0.004 |
slices | -0.004 |
amaz | -0.004 |
fires | -0.004 |
Token | took |
Token ID | 1718 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
Woo | +0.037 |
wear | +0.037 |
bath | +0.035 |
burn | +0.034 |
important | +0.033 |
Neg logit contributions | |
slices | -0.048 |
amaz | -0.046 |
unison | -0.046 |
seconds | -0.045 |
aroma | -0.045 |
Token | out |
Token ID | 503 |
Feature activation | -3.65e-02 |
Pos logit contributions | |
slices | +0.031 |
unison | +0.030 |
amaz | +0.030 |
aroma | +0.030 |
seconds | +0.030 |
Neg logit contributions | |
wear | -0.020 |
Woo | -0.019 |
bath | -0.018 |
important | -0.017 |
skipping | -0.017 |
Token | her |
Token ID | 607 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
slices | +0.075 |
unison | +0.075 |
amaz | +0.073 |
seventeen | +0.073 |
seconds | +0.072 |
Neg logit contributions | |
wear | -0.045 |
Woo | -0.041 |
bath | -0.040 |
Mitt | -0.037 |
Lucky | -0.037 |
Token | sewing |
Token ID | 35356 |
Feature activation | +4.83e-02 |
Pos logit contributions | |
Woo | +0.027 |
scar | +0.027 |
ido | +0.027 |
tto | +0.026 |
burn | +0.025 |
Neg logit contributions | |
slices | -0.038 |
amaz | -0.035 |
seconds | -0.033 |
conf | -0.033 |
unison | -0.033 |
Token | machine |
Token ID | 4572 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
wear | +0.080 |
Woo | +0.075 |
bath | +0.074 |
Mitt | +0.070 |
Lucky | +0.070 |
Neg logit contributions | |
unison | -0.079 |
slices | -0.078 |
seventeen | -0.077 |
amaz | -0.076 |
aroma | -0.075 |
Token | . |
Token ID | 13 |
Feature activation | +2.87e-03 |
Pos logit contributions | |
slices | +0.031 |
unison | +0.030 |
amaz | +0.030 |
seventeen | +0.029 |
aroma | +0.029 |
Neg logit contributions | |
wear | -0.019 |
Woo | -0.018 |
bath | -0.017 |
Mitt | -0.016 |
Lucky | -0.016 |
Token | She |
Token ID | 1375 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
wear | +0.004 |
Woo | +0.004 |
bath | +0.003 |
Mitt | +0.003 |
day | +0.003 |
Neg logit contributions | |
unison | -0.006 |
slices | -0.006 |
seventeen | -0.006 |
amaz | -0.005 |
aroma | -0.005 |
Token | carefully |
Token ID | 7773 |
Feature activation | -7.81e-03 |
Pos logit contributions | |
slices | +0.025 |
unison | +0.025 |
amaz | +0.024 |
seconds | +0.024 |
seventeen | +0.024 |
Neg logit contributions | |
wear | -0.016 |
Woo | -0.016 |
bath | -0.015 |
burn | -0.015 |
day | -0.014 |
Token | loaded |
Token ID | 9639 |
Feature activation | -2.75e-03 |
Pos logit contributions | |
slices | +0.015 |
seconds | +0.014 |
amaz | +0.014 |
unison | +0.014 |
seventeen | +0.014 |
Neg logit contributions | |
Woo | -0.010 |
wear | -0.010 |
important | -0.010 |
day | -0.010 |
bath | -0.010 |
Token | for |
Token ID | 329 |
Feature activation | -4.11e-02 |
Pos logit contributions | |
wear | +0.010 |
Woo | +0.009 |
bath | +0.009 |
day | +0.009 |
important | +0.009 |
Neg logit contributions | |
seventeen | -0.022 |
unison | -0.022 |
slices | -0.022 |
pulp | -0.022 |
amaz | -0.022 |
Token | you |
Token ID | 345 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
seventeen | +0.080 |
slices | +0.080 |
unison | +0.079 |
amaz | +0.078 |
aroma | +0.078 |
Neg logit contributions | |
wear | -0.057 |
Woo | -0.053 |
bath | -0.052 |
Mitt | -0.049 |
skipping | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
unison | +0.051 |
slices | +0.050 |
seventeen | +0.050 |
amaz | +0.050 |
pulp | +0.049 |
Neg logit contributions | |
wear | -0.031 |
Woo | -0.028 |
bath | -0.027 |
day | -0.025 |
Mitt | -0.025 |
Token | You |
Token ID | 921 |
Feature activation | -5.70e-02 |
Pos logit contributions | |
wear | +0.032 |
bath | +0.030 |
Woo | +0.030 |
burn | +0.029 |
important | +0.029 |
Neg logit contributions | |
unison | -0.022 |
amaz | -0.022 |
seventeen | -0.022 |
seconds | -0.021 |
slices | -0.021 |
Token | don |
Token ID | 836 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
slices | +0.100 |
seventeen | +0.099 |
unison | +0.097 |
amaz | +0.095 |
pulp | +0.095 |
Neg logit contributions | |
wear | -0.098 |
bath | -0.086 |
Woo | -0.084 |
always | -0.084 |
Lucky | -0.080 |
Token | 't |
Token ID | 470 |
Feature activation | +6.04e-02 |
Pos logit contributions | |
slices | +0.037 |
unison | +0.037 |
seventeen | +0.036 |
anish | +0.034 |
folk | +0.034 |
Neg logit contributions | |
wear | -0.036 |
bath | -0.033 |
Woo | -0.033 |
day | -0.033 |
opsy | -0.033 |
Token | have |
Token ID | 423 |
Feature activation | +1.95e-03 |
Pos logit contributions | |
Woo | +0.076 |
wear | +0.074 |
bath | +0.072 |
Mitt | +0.070 |
skipping | +0.070 |
Neg logit contributions | |
unison | -0.109 |
seventeen | -0.106 |
aroma | -0.105 |
seconds | -0.105 |
amaz | -0.104 |
Token | to |
Token ID | 284 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
wear | +0.002 |
Woo | +0.002 |
bath | +0.002 |
Mitt | +0.002 |
Lucky | +0.002 |
Neg logit contributions | |
slices | -0.004 |
unison | -0.004 |
seventeen | -0.004 |
amaz | -0.004 |
seconds | -0.004 |
Token | be |
Token ID | 307 |
Feature activation | -7.49e-03 |
Pos logit contributions | |
unison | +0.025 |
aroma | +0.023 |
amaz | +0.023 |
seconds | +0.022 |
rs | +0.022 |
Neg logit contributions | |
ip | -0.019 |
scar | -0.018 |
Woo | -0.018 |
Lucky | -0.017 |
important | -0.017 |
Token | sad |
Token ID | 6507 |
Feature activation | +7.19e-03 |
Pos logit contributions | |
unison | +0.015 |
slices | +0.014 |
seventeen | +0.014 |
amaz | +0.014 |
seconds | +0.014 |
Neg logit contributions | |
wear | -0.010 |
Woo | -0.009 |
bath | -0.009 |
Mitt | -0.008 |
burn | -0.008 |
Token | or |
Token ID | 393 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
wear | +0.010 |
Woo | +0.009 |
bath | +0.009 |
Mitt | +0.008 |
important | +0.008 |
Neg logit contributions | |
unison | -0.014 |
slices | -0.014 |
seventeen | -0.014 |
amaz | -0.014 |
aroma | -0.014 |
Token | on |
Token ID | 319 |
Feature activation | +5.47e-03 |
Pos logit contributions | |
wear | +0.059 |
Woo | +0.059 |
ibi | +0.059 |
ety | +0.058 |
ongo | +0.057 |
Neg logit contributions | |
moments | -0.073 |
anish | -0.071 |
slices | -0.070 |
seconds | -0.070 |
seventeen | -0.068 |
Token | , |
Token ID | 11 |
Feature activation | +6.26e-03 |
Pos logit contributions | |
wear | +0.008 |
burn | +0.008 |
Woo | +0.008 |
ibi | +0.008 |
bath | +0.007 |
Neg logit contributions | |
slices | -0.009 |
seconds | -0.009 |
aroma | -0.009 |
moments | -0.009 |
anish | -0.008 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.72e-02 |
Pos logit contributions | |
scar | +0.008 |
ido | +0.008 |
ongo | +0.008 |
ibi | +0.008 |
burn | +0.008 |
Neg logit contributions | |
slices | -0.012 |
aroma | -0.011 |
amaz | -0.011 |
ah | -0.011 |
pulp | -0.011 |
Token | always |
Token ID | 1464 |
Feature activation | +6.98e-03 |
Pos logit contributions | |
Woo | +0.055 |
wear | +0.054 |
bath | +0.053 |
ido | +0.049 |
Lucky | +0.049 |
Neg logit contributions | |
amaz | -0.061 |
seconds | -0.061 |
slices | -0.060 |
moments | -0.058 |
aroma | -0.058 |
Token | picked |
Token ID | 6497 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
wear | +0.011 |
Woo | +0.010 |
bath | +0.010 |
Lucky | +0.009 |
Mitt | +0.009 |
Neg logit contributions | |
amaz | -0.012 |
unison | -0.012 |
seconds | -0.012 |
slices | -0.012 |
seventeen | -0.011 |
Token | herbs |
Token ID | 25411 |
Feature activation | +6.78e-03 |
Pos logit contributions | |
wear | +0.024 |
Woo | +0.022 |
bath | +0.021 |
burn | +0.019 |
stub | +0.019 |
Neg logit contributions | |
slices | -0.045 |
amaz | -0.043 |
aroma | -0.043 |
pulp | -0.042 |
seconds | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
wear | +0.008 |
Woo | +0.007 |
bath | +0.007 |
Mitt | +0.007 |
Lucky | +0.006 |
Neg logit contributions | |
slices | -0.014 |
unison | -0.014 |
seventeen | -0.014 |
amaz | -0.014 |
aroma | -0.014 |
Token | covered |
Token ID | 5017 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
wear | +0.030 |
Woo | +0.028 |
bath | +0.027 |
Mitt | +0.025 |
Lucky | +0.025 |
Neg logit contributions | |
slices | -0.042 |
unison | -0.042 |
amaz | -0.041 |
seventeen | -0.041 |
seconds | -0.041 |
Token | her |
Token ID | 607 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
wear | +0.021 |
Woo | +0.019 |
bath | +0.019 |
Mitt | +0.018 |
Lucky | +0.018 |
Neg logit contributions | |
unison | -0.027 |
slices | -0.026 |
seventeen | -0.026 |
amaz | -0.026 |
seconds | -0.025 |
Token | room |
Token ID | 2119 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
wear | +0.016 |
bath | +0.015 |
important | +0.014 |
Mitt | +0.014 |
Woo | +0.014 |
Neg logit contributions | |
seventeen | -0.025 |
unison | -0.024 |
seconds | -0.023 |
folk | -0.023 |
bridge | -0.023 |
Token | with |
Token ID | 351 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
wear | +0.013 |
bath | +0.011 |
Woo | +0.011 |
important | +0.010 |
day | +0.010 |
Neg logit contributions | |
unison | -0.024 |
seventeen | -0.024 |
slices | -0.023 |
pulp | -0.023 |
amaz | -0.023 |
Token | safe |
Token ID | 3338 |
Feature activation | +6.50e-03 |
Pos logit contributions | |
wear | +0.069 |
Woo | +0.064 |
bath | +0.062 |
Mitt | +0.057 |
Lucky | +0.057 |
Neg logit contributions | |
unison | -0.116 |
slices | -0.115 |
seventeen | -0.114 |
amaz | -0.113 |
aroma | -0.111 |
Token | . |
Token ID | 13 |
Feature activation | +6.41e-04 |
Pos logit contributions | |
wear | +0.008 |
Woo | +0.008 |
bath | +0.007 |
Mitt | +0.007 |
day | +0.007 |
Neg logit contributions | |
unison | -0.013 |
slices | -0.013 |
seventeen | -0.013 |
amaz | -0.013 |
aroma | -0.013 |
Token | She |
Token ID | 1375 |
Feature activation | -7.33e-03 |
Pos logit contributions | |
wear | +0.001 |
bath | +0.001 |
Woo | +0.001 |
Mitt | +0.001 |
day | +0.001 |
Neg logit contributions | |
seventeen | -0.001 |
unison | -0.001 |
slices | -0.001 |
pulp | -0.001 |
fires | -0.001 |
Token | quickly |
Token ID | 2952 |
Feature activation | +2.93e-03 |
Pos logit contributions | |
seconds | +0.012 |
slices | +0.012 |
unison | +0.012 |
amaz | +0.012 |
gasped | +0.011 |
Neg logit contributions | |
Woo | -0.011 |
wear | -0.011 |
burn | -0.011 |
bath | -0.010 |
skipping | -0.010 |
Token | grabbed |
Token ID | 13176 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
Woo | +0.004 |
wear | +0.004 |
burn | +0.004 |
bath | +0.004 |
day | +0.004 |
Neg logit contributions | |
slices | -0.005 |
unison | -0.005 |
seconds | -0.005 |
amaz | -0.005 |
pulp | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | -3.39e-03 |
Pos logit contributions | |
unison | +0.051 |
seventeen | +0.051 |
slices | +0.050 |
fires | +0.049 |
amaz | +0.049 |
Neg logit contributions | |
wear | -0.032 |
Woo | -0.029 |
bath | -0.028 |
Mitt | -0.026 |
day | -0.026 |
Token | laser |
Token ID | 12855 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
unison | +0.008 |
slices | +0.008 |
seventeen | +0.007 |
amaz | +0.007 |
seconds | +0.007 |
Neg logit contributions | |
wear | -0.004 |
Woo | -0.003 |
bath | -0.003 |
Mitt | -0.003 |
day | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +2.60e-03 |
Pos logit contributions | |
wear | +0.034 |
Woo | +0.031 |
bath | +0.030 |
Mitt | +0.028 |
Lucky | +0.028 |
Neg logit contributions | |
unison | -0.059 |
slices | -0.058 |
seventeen | -0.058 |
amaz | -0.057 |
aroma | -0.056 |
Token | hid |
Token ID | 24519 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
wear | +0.003 |
Woo | +0.003 |
bath | +0.003 |
Mitt | +0.003 |
Lucky | +0.003 |
Neg logit contributions | |
unison | -0.005 |
slices | -0.005 |
seventeen | -0.005 |
amaz | -0.005 |
aroma | -0.005 |
Token | it |
Token ID | 340 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
wear | +0.022 |
Woo | +0.019 |
skipping | +0.019 |
day | +0.018 |
bath | +0.017 |
Neg logit contributions | |
seventeen | -0.057 |
unison | -0.052 |
aroma | -0.049 |
teen | -0.049 |
pulp | -0.049 |
Token | away |
Token ID | 1497 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
slices | +0.023 |
unison | +0.023 |
seventeen | +0.023 |
amaz | +0.023 |
seconds | +0.022 |
Neg logit contributions | |
wear | -0.016 |
Woo | -0.015 |
bath | -0.014 |
Mitt | -0.014 |
burn | -0.013 |
Token | know |
Token ID | 760 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
Woo | +0.106 |
wear | +0.103 |
bath | +0.102 |
burn | +0.099 |
Mitt | +0.098 |
Neg logit contributions | |
seconds | -0.114 |
unison | -0.112 |
amaz | -0.111 |
aroma | -0.111 |
slices | -0.109 |
Token | , |
Token ID | 11 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
wear | +0.023 |
Woo | +0.022 |
bath | +0.021 |
Mitt | +0.020 |
Lucky | +0.020 |
Neg logit contributions | |
unison | -0.034 |
slices | -0.033 |
seventeen | -0.033 |
amaz | -0.033 |
aroma | -0.032 |
Token | Rover |
Token ID | 36718 |
Feature activation | +2.99e-02 |
Pos logit contributions | |
wear | +0.015 |
Woo | +0.014 |
bath | +0.014 |
burn | +0.013 |
day | +0.013 |
Neg logit contributions | |
unison | -0.019 |
slices | -0.018 |
amaz | -0.018 |
seventeen | -0.018 |
seconds | -0.018 |
Token | ," |
Token ID | 553 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
wear | +0.042 |
Woo | +0.040 |
bath | +0.039 |
Mitt | +0.036 |
skipping | +0.036 |
Neg logit contributions | |
unison | -0.055 |
slices | -0.055 |
seventeen | -0.054 |
amaz | -0.054 |
aroma | -0.053 |
Token | she |
Token ID | 673 |
Feature activation | -9.86e-03 |
Pos logit contributions | |
unison | +0.065 |
amaz | +0.064 |
slices | +0.064 |
seconds | +0.063 |
aroma | +0.061 |
Neg logit contributions | |
wear | -0.040 |
bath | -0.037 |
Woo | -0.037 |
burn | -0.037 |
important | -0.035 |
Token | said |
Token ID | 531 |
Feature activation | +6.71e-03 |
Pos logit contributions | |
unison | +0.017 |
slices | +0.017 |
seventeen | +0.017 |
amaz | +0.017 |
seconds | +0.017 |
Neg logit contributions | |
wear | -0.015 |
Woo | -0.014 |
bath | -0.014 |
burn | -0.013 |
Mitt | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
wear | +0.009 |
Woo | +0.008 |
bath | +0.008 |
Mitt | +0.008 |
Lucky | +0.007 |
Neg logit contributions | |
unison | -0.013 |
slices | -0.013 |
seventeen | -0.013 |
amaz | -0.013 |
seconds | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -3.64e-03 |
Pos logit contributions | |
unison | +0.029 |
slices | +0.029 |
seventeen | +0.028 |
amaz | +0.028 |
seconds | +0.028 |
Neg logit contributions | |
wear | -0.020 |
Woo | -0.019 |
bath | -0.018 |
Mitt | -0.017 |
day | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
unison | +0.006 |
slices | +0.006 |
seconds | +0.006 |
fires | +0.005 |
amaz | +0.005 |
Neg logit contributions | |
wear | -0.006 |
burn | -0.005 |
bath | -0.005 |
important | -0.005 |
Woo | -0.005 |
Token | That |
Token ID | 2504 |
Feature activation | -3.76e-02 |
Pos logit contributions | |
unison | +0.019 |
slices | +0.019 |
seventeen | +0.018 |
amaz | +0.018 |
seconds | +0.018 |
Neg logit contributions | |
wear | -0.021 |
bath | -0.019 |
Woo | -0.019 |
day | -0.018 |
important | -0.018 |
Token | night |
Token ID | 1755 |
Feature activation | -6.22e-03 |
Pos logit contributions | |
amaz | +0.077 |
unison | +0.076 |
aroma | +0.076 |
slices | +0.075 |
seconds | +0.074 |
Neg logit contributions | |
wear | -0.045 |
Woo | -0.044 |
bath | -0.040 |
Mitt | -0.040 |
skipping | -0.038 |
Token | he |
Token ID | 339 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
wear | +0.018 |
Woo | +0.016 |
bath | +0.016 |
burn | +0.015 |
skipping | +0.015 |
Neg logit contributions | |
slices | -0.027 |
unison | -0.026 |
seconds | -0.026 |
amaz | -0.026 |
seventeen | -0.026 |
Token | didn |
Token ID | 1422 |
Feature activation | -3.53e-03 |
Pos logit contributions | |
seconds | +0.025 |
slices | +0.024 |
aroma | +0.023 |
unison | +0.022 |
fires | +0.022 |
Neg logit contributions | |
Woo | -0.029 |
wear | -0.028 |
bath | -0.027 |
skipping | -0.027 |
burn | -0.027 |
Token | 't |
Token ID | 470 |
Feature activation | +6.30e-02 |
Pos logit contributions | |
slices | +0.004 |
seventeen | +0.004 |
unison | +0.004 |
seconds | +0.004 |
fires | +0.004 |
Neg logit contributions | |
ety | -0.006 |
opsy | -0.006 |
wear | -0.006 |
ibi | -0.006 |
important | -0.006 |
Token | behave |
Token ID | 17438 |
Feature activation | -4.22e-03 |
Pos logit contributions | |
scar | +0.100 |
opsy | +0.097 |
tto | +0.093 |
Woo | +0.093 |
Lucky | +0.092 |
Neg logit contributions | |
aroma | -0.090 |
click | -0.090 |
rs | -0.088 |
fires | -0.088 |
seconds | -0.086 |
Token | , |
Token ID | 11 |
Feature activation | +8.45e-03 |
Pos logit contributions | |
unison | +0.008 |
slices | +0.008 |
fires | +0.008 |
seconds | +0.007 |
aroma | +0.007 |
Neg logit contributions | |
wear | -0.006 |
Woo | -0.005 |
bath | -0.005 |
burn | -0.005 |
skipping | -0.005 |
Token | he |
Token ID | 339 |
Feature activation | +1.49e-03 |
Pos logit contributions | |
wear | +0.010 |
Woo | +0.009 |
bath | +0.009 |
burn | +0.008 |
Lucky | +0.008 |
Neg logit contributions | |
slices | -0.017 |
unison | -0.017 |
amaz | -0.017 |
seventeen | -0.017 |
seconds | -0.017 |
Token | wouldn |
Token ID | 3636 |
Feature activation | -5.42e-03 |
Pos logit contributions | |
Woo | +0.002 |
bath | +0.002 |
skipping | +0.002 |
wear | +0.002 |
burn | +0.002 |
Neg logit contributions | |
seconds | -0.002 |
fires | -0.002 |
slices | -0.002 |
aroma | -0.002 |
unison | -0.002 |
Token | 't |
Token ID | 470 |
Feature activation | +4.87e-02 |
Pos logit contributions | |
ah | +0.006 |
seconds | +0.005 |
slices | +0.005 |
als | +0.005 |
click | +0.005 |
Neg logit contributions | |
important | -0.010 |
day | -0.010 |
opsy | -0.009 |
ety | -0.009 |
wear | -0.009 |
Token | receive |
Token ID | 3328 |
Feature activation | +2.29e-03 |
Pos logit contributions | |
scar | +0.076 |
opsy | +0.076 |
ip | +0.072 |
Woo | +0.071 |
tto | +0.070 |
Neg logit contributions | |
click | -0.071 |
aroma | -0.066 |
seconds | -0.066 |
faces | -0.065 |
fires | -0.064 |
Token | any |
Token ID | 597 |
Feature activation | +5.29e-03 |
Pos logit contributions | |
wear | +0.003 |
Woo | +0.003 |
bath | +0.003 |
Mitt | +0.003 |
important | +0.003 |
Neg logit contributions | |
unison | -0.004 |
slices | -0.004 |
seventeen | -0.004 |
amaz | -0.004 |
fires | -0.004 |
Token | presents |
Token ID | 10969 |
Feature activation | -2.97e-02 |
Pos logit contributions | |
wear | +0.007 |
bath | +0.007 |
Woo | +0.007 |
important | +0.006 |
day | +0.006 |
Neg logit contributions | |
unison | -0.010 |
seventeen | -0.010 |
slices | -0.010 |
amaz | -0.010 |
fires | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
Woo | +0.044 |
wear | +0.043 |
ibi | +0.041 |
opsy | +0.040 |
stub | +0.040 |
Neg logit contributions | |
slices | -0.041 |
aroma | -0.034 |
unison | -0.034 |
pter | -0.033 |
fires | -0.032 |
Token | Clara |
Token ID | 27443 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
wear | +0.021 |
Woo | +0.020 |
bath | +0.019 |
important | +0.019 |
burn | +0.018 |
Neg logit contributions | |
unison | -0.028 |
slices | -0.028 |
amaz | -0.028 |
seventeen | -0.027 |
aroma | -0.027 |
Token | asked |
Token ID | 1965 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
wear | +0.024 |
Woo | +0.023 |
bath | +0.022 |
burn | +0.021 |
Mitt | +0.021 |
Neg logit contributions | |
slices | -0.028 |
unison | -0.028 |
amaz | -0.027 |
seventeen | -0.027 |
seconds | -0.027 |
Token | her |
Token ID | 607 |
Feature activation | +9.98e-03 |
Pos logit contributions | |
wear | +0.020 |
Woo | +0.018 |
bath | +0.018 |
burn | +0.017 |
Lucky | +0.016 |
Neg logit contributions | |
slices | -0.040 |
unison | -0.039 |
amaz | -0.039 |
seconds | -0.038 |
aroma | -0.038 |
Token | mom |
Token ID | 1995 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
Woo | +0.015 |
wear | +0.014 |
ained | +0.014 |
scar | +0.014 |
burn | +0.013 |
Neg logit contributions | |
slices | -0.018 |
amaz | -0.018 |
seconds | -0.017 |
unison | -0.016 |
aroma | -0.016 |
Token | my |
Token ID | 1820 |
Feature activation | +1.75e-03 |
Pos logit contributions | |
slices | +0.076 |
amaz | +0.070 |
aroma | +0.070 |
seventeen | +0.070 |
pter | +0.069 |
Neg logit contributions | |
Woo | -0.036 |
stub | -0.032 |
wear | -0.032 |
bath | -0.031 |
opsy | -0.030 |
Token | what |
Token ID | 644 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
Woo | +0.002 |
wear | +0.002 |
bath | +0.002 |
burn | +0.002 |
stub | +0.002 |
Neg logit contributions | |
slices | -0.003 |
amaz | -0.003 |
seventeen | -0.003 |
aroma | -0.003 |
teen | -0.003 |
Token | they |
Token ID | 484 |
Feature activation | -9.94e-03 |
Pos logit contributions | |
slices | +0.036 |
statues | +0.033 |
conf | +0.033 |
seconds | +0.032 |
amaz | +0.032 |
Neg logit contributions | |
wear | -0.035 |
Woo | -0.035 |
ongo | -0.035 |
umper | -0.034 |
burn | -0.034 |
Token | were |
Token ID | 547 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
slices | +0.012 |
unison | +0.012 |
amaz | +0.012 |
seventeen | +0.012 |
aroma | +0.012 |
Neg logit contributions | |
wear | -0.020 |
Woo | -0.019 |
bath | -0.019 |
Mitt | -0.018 |
Lucky | -0.018 |
Token | going |
Token ID | 1016 |
Feature activation | +9.32e-03 |
Pos logit contributions | |
slices | +0.022 |
aroma | +0.022 |
fires | +0.021 |
pulp | +0.021 |
statues | +0.020 |
Neg logit contributions | |
Woo | -0.012 |
Lucky | -0.012 |
wear | -0.012 |
bath | -0.011 |
Mitt | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | -5.33e-04 |
Pos logit contributions | |
opsy | +0.015 |
Woo | +0.014 |
scar | +0.014 |
stub | +0.014 |
ibi | +0.014 |
Neg logit contributions | |
slices | -0.014 |
fires | -0.013 |
als | -0.012 |
pter | -0.012 |
anish | -0.012 |
Token | Are |
Token ID | 8491 |
Feature activation | -7.03e-03 |
Pos logit contributions | |
unison | +0.013 |
slices | +0.012 |
seventeen | +0.012 |
amaz | +0.012 |
seconds | +0.012 |
Neg logit contributions | |
wear | -0.018 |
bath | -0.017 |
Woo | -0.017 |
Lucky | -0.017 |
Mitt | -0.016 |
Token | you |
Token ID | 345 |
Feature activation | -7.15e-03 |
Pos logit contributions | |
ah | +0.008 |
seconds | +0.008 |
thirteen | +0.008 |
seventeen | +0.008 |
fires | +0.008 |
Neg logit contributions | |
wear | -0.013 |
burn | -0.013 |
bath | -0.013 |
iny | -0.013 |
ibi | -0.012 |
Token | ready |
Token ID | 3492 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
unison | +0.014 |
amaz | +0.014 |
slices | +0.014 |
aroma | +0.014 |
fires | +0.013 |
Neg logit contributions | |
wear | -0.008 |
Woo | -0.008 |
bath | -0.008 |
Lucky | -0.007 |
Mitt | -0.007 |
Token | for |
Token ID | 329 |
Feature activation | -3.26e-02 |
Pos logit contributions | |
amaz | +0.019 |
unison | +0.019 |
seconds | +0.019 |
slices | +0.019 |
teen | +0.019 |
Neg logit contributions | |
Woo | -0.013 |
wear | -0.012 |
important | -0.012 |
Mitt | -0.012 |
burn | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
unison | +0.060 |
seconds | +0.060 |
slices | +0.059 |
amaz | +0.059 |
seventeen | +0.057 |
Neg logit contributions | |
wear | -0.044 |
Woo | -0.042 |
burn | -0.040 |
bath | -0.040 |
Mitt | -0.039 |
Token | meeting |
Token ID | 3249 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
wear | +0.033 |
Woo | +0.031 |
burn | +0.029 |
bath | +0.028 |
Mitt | +0.028 |
Neg logit contributions | |
unison | -0.053 |
slices | -0.053 |
aroma | -0.052 |
fires | -0.052 |
seconds | -0.052 |
Token | ? |
Token ID | 30 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
slices | +0.038 |
unison | +0.038 |
amaz | +0.037 |
seventeen | +0.037 |
seconds | +0.037 |
Neg logit contributions | |
wear | -0.017 |
Woo | -0.017 |
bath | -0.015 |
Mitt | -0.014 |
burn | -0.014 |
Token | â |
Token ID | 22940 |
Feature activation | -4.13e-03 |
Pos logit contributions | |
wear | +0.017 |
important | +0.016 |
burn | +0.016 |
day | +0.016 |
bath | +0.016 |
Neg logit contributions | |
unison | -0.016 |
amaz | -0.015 |
slices | -0.015 |
fires | -0.014 |
seventeen | -0.014 |
Token | € |
Token ID | 26391 |
Feature activation | +5.90e-03 |
Pos logit contributions | |
unison | +0.007 |
seventeen | +0.007 |
fires | +0.007 |
aroma | +0.007 |
amaz | +0.007 |
Neg logit contributions | |
wear | -0.006 |
Woo | -0.006 |
bath | -0.006 |
Mitt | -0.005 |
burn | -0.005 |
Token | M |
Token ID | 337 |
Feature activation | +2.96e-02 |
Pos logit contributions | |
wear | +0.010 |
Woo | +0.009 |
bath | +0.009 |
Mitt | +0.009 |
skipping | +0.008 |
Neg logit contributions | |
seventeen | -0.010 |
slices | -0.010 |
unison | -0.010 |
pulp | -0.010 |
aroma | -0.010 |
Token | addy |
Token ID | 13218 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
Woo | +0.040 |
wear | +0.039 |
Nut | +0.036 |
bath | +0.036 |
Lucky | +0.036 |
Neg logit contributions | |
amaz | -0.054 |
unison | -0.053 |
seconds | -0.053 |
aroma | -0.052 |
teen | -0.052 |
Token | with |
Token ID | 351 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
wear | +0.043 |
bath | +0.040 |
Woo | +0.039 |
skipping | +0.038 |
jog | +0.038 |
Neg logit contributions | |
slices | -0.044 |
unison | -0.043 |
als | -0.042 |
fires | -0.041 |
pter | -0.041 |
Token | soft |
Token ID | 2705 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
slices | +0.034 |
unison | +0.034 |
seventeen | +0.034 |
amaz | +0.034 |
aroma | +0.033 |
Neg logit contributions | |
wear | -0.022 |
Woo | -0.021 |
bath | -0.020 |
burn | -0.019 |
Mitt | -0.019 |
Token | fur |
Token ID | 9230 |
Feature activation | +5.47e-02 |
Pos logit contributions | |
wear | +0.020 |
Woo | +0.019 |
bath | +0.018 |
Mitt | +0.018 |
important | +0.018 |
Neg logit contributions | |
slices | -0.023 |
unison | -0.023 |
seventeen | -0.023 |
amaz | -0.023 |
aroma | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
wear | +0.075 |
Woo | +0.068 |
bath | +0.067 |
Mitt | +0.062 |
important | +0.062 |
Neg logit contributions | |
slices | -0.105 |
unison | -0.104 |
seventeen | -0.102 |
aroma | -0.101 |
amaz | -0.100 |
Token | big |
Token ID | 1263 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
wear | +0.016 |
Woo | +0.015 |
bath | +0.015 |
burn | +0.014 |
ido | +0.014 |
Neg logit contributions | |
slices | -0.023 |
amaz | -0.022 |
unison | -0.022 |
seconds | -0.022 |
seventeen | -0.022 |
Token | eyes |
Token ID | 2951 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
fires | +0.031 |
unison | +0.030 |
seventeen | +0.028 |
slices | +0.028 |
anish | +0.028 |
Neg logit contributions | |
wear | -0.043 |
bath | -0.043 |
Woo | -0.040 |
Mitt | -0.040 |
skipping | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
slices | +0.027 |
unison | +0.027 |
seventeen | +0.026 |
amaz | +0.026 |
aroma | +0.026 |
Neg logit contributions | |
wear | -0.019 |
Woo | -0.018 |
bath | -0.017 |
Mitt | -0.016 |
Lucky | -0.016 |
Token | One |
Token ID | 1881 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
wear | +0.023 |
burn | +0.022 |
Woo | +0.021 |
important | +0.021 |
scar | +0.021 |
Neg logit contributions | |
slices | -0.027 |
unison | -0.027 |
amaz | -0.026 |
seventeen | -0.026 |
teen | -0.025 |
Token | day |
Token ID | 1110 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
Woo | +0.037 |
Cle | +0.033 |
Barkley | +0.031 |
Buster | +0.031 |
Nut | +0.030 |
Neg logit contributions | |
moments | -0.045 |
amaz | -0.045 |
slices | -0.045 |
aroma | -0.045 |
slice | -0.043 |
Token | , |
Token ID | 11 |
Feature activation | -3.94e-03 |
Pos logit contributions | |
slices | +0.048 |
unison | +0.046 |
seventeen | +0.046 |
seconds | +0.046 |
aroma | +0.046 |
Neg logit contributions | |
wear | -0.033 |
Woo | -0.031 |
burn | -0.029 |
important | -0.029 |
bath | -0.029 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
unison | +0.009 |
slices | +0.009 |
seventeen | +0.009 |
amaz | +0.009 |
aroma | +0.009 |
Neg logit contributions | |
wear | -0.004 |
Woo | -0.003 |
bath | -0.003 |
Mitt | -0.003 |
important | -0.003 |
Token | had |
Token ID | 550 |
Feature activation | -6.24e-03 |
Pos logit contributions | |
unison | +0.033 |
slices | +0.032 |
seventeen | +0.032 |
amaz | +0.032 |
aroma | +0.031 |
Neg logit contributions | |
wear | -0.029 |
Woo | -0.027 |
bath | -0.027 |
Mitt | -0.025 |
Lucky | -0.025 |
Token | just |
Token ID | 655 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
fires | +0.013 |
aroma | +0.013 |
unison | +0.013 |
seventeen | +0.013 |
slices | +0.013 |
Neg logit contributions | |
wear | -0.007 |
Woo | -0.007 |
bath | -0.007 |
always | -0.007 |
important | -0.006 |
Token | seen |
Token ID | 1775 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
slices | +0.041 |
unison | +0.041 |
seventeen | +0.040 |
fires | +0.040 |
aroma | +0.039 |
Neg logit contributions | |
wear | -0.031 |
day | -0.027 |
bath | -0.027 |
Woo | -0.027 |
always | -0.026 |
Token | something |
Token ID | 1223 |
Feature activation | -2.79e-03 |
Pos logit contributions | |
wear | +0.026 |
Woo | +0.024 |
bath | +0.024 |
day | +0.023 |
Mitt | +0.022 |
Neg logit contributions | |
seventeen | -0.049 |
unison | -0.048 |
slices | -0.048 |
fires | -0.047 |
pulp | -0.047 |
Token | truly |
Token ID | 4988 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
unison | +0.006 |
slices | +0.006 |
amaz | +0.006 |
seventeen | +0.006 |
seconds | +0.006 |
Neg logit contributions | |
wear | -0.003 |
Woo | -0.003 |
bath | -0.003 |
burn | -0.003 |
Mitt | -0.003 |
Token | magical |
Token ID | 10883 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
slices | +0.043 |
unison | +0.043 |
seventeen | +0.043 |
pulp | +0.041 |
amaz | +0.041 |
Neg logit contributions | |
wear | -0.020 |
important | -0.018 |
day | -0.017 |
bath | -0.017 |
Woo | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -5.08e-05 |
Pos logit contributions | |
seventeen | +0.034 |
slices | +0.033 |
unison | +0.033 |
pulp | +0.033 |
amaz | +0.032 |
Neg logit contributions | |
wear | -0.019 |
day | -0.017 |
bath | -0.016 |
important | -0.016 |
Woo | -0.016 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.59e-02 |
Pos logit contributions | |
unison | +0.000 |
slices | +0.000 |
amaz | +0.000 |
seventeen | +0.000 |
seconds | +0.000 |
Neg logit contributions | |
wear | -0.000 |
Woo | -0.000 |
bath | -0.000 |
burn | -0.000 |
important | -0.000 |
Token | Once |
Token ID | 7454 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
slices | +0.078 |
unison | +0.078 |
amaz | +0.075 |
seventeen | +0.075 |
fires | +0.074 |
Neg logit contributions | |
wear | -0.072 |
Woo | -0.067 |
bath | -0.065 |
important | -0.063 |
Mitt | -0.062 |
Token | upon |
Token ID | 2402 |
Feature activation | -5.66e-02 |
Pos logit contributions | |
Woo | +0.040 |
wear | +0.039 |
burn | +0.039 |
bath | +0.036 |
Mitt | +0.035 |
Neg logit contributions | |
slices | -0.037 |
amaz | -0.037 |
unison | -0.036 |
pter | -0.035 |
seconds | -0.035 |
Token | a |
Token ID | 257 |
Feature activation | +4.38e-02 |
Pos logit contributions | |
unison | +0.113 |
slices | +0.112 |
seventeen | +0.111 |
amaz | +0.110 |
aroma | +0.108 |
Neg logit contributions | |
wear | -0.073 |
Woo | -0.067 |
bath | -0.066 |
Mitt | -0.061 |
Lucky | -0.060 |
Token | out |
Token ID | 503 |
Feature activation | -3.94e-02 |
Pos logit contributions | |
unison | +0.028 |
pter | +0.027 |
seventeen | +0.027 |
fires | +0.026 |
slices | +0.026 |
Neg logit contributions | |
wear | -0.013 |
Mitt | -0.012 |
Woo | -0.012 |
bath | -0.012 |
Lucky | -0.012 |
Token | came |
Token ID | 1625 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
seventeen | +0.074 |
unison | +0.074 |
slices | +0.074 |
pulp | +0.073 |
pter | +0.072 |
Neg logit contributions | |
wear | -0.056 |
bath | -0.050 |
day | -0.048 |
skipping | -0.047 |
always | -0.047 |
Token | a |
Token ID | 257 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
slices | +0.026 |
unison | +0.026 |
amaz | +0.026 |
seventeen | +0.026 |
seconds | +0.025 |
Neg logit contributions | |
wear | -0.020 |
Woo | -0.018 |
bath | -0.018 |
day | -0.016 |
Mitt | -0.016 |
Token | big |
Token ID | 1263 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
pter | +0.030 |
unison | +0.030 |
seventeen | +0.030 |
slices | +0.029 |
fires | +0.029 |
Neg logit contributions | |
bath | -0.015 |
wear | -0.015 |
day | -0.013 |
Woo | -0.013 |
important | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
fires | +0.091 |
unison | +0.089 |
pter | +0.089 |
seventeen | +0.088 |
slices | +0.088 |
Neg logit contributions | |
wear | -0.049 |
bath | -0.049 |
day | -0.046 |
Woo | -0.044 |
Mitt | -0.042 |
Token | foolish |
Token ID | 19538 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
unison | +0.022 |
slices | +0.022 |
seventeen | +0.022 |
amaz | +0.022 |
fires | +0.022 |
Neg logit contributions | |
wear | -0.012 |
Woo | -0.011 |
bath | -0.011 |
important | -0.010 |
Mitt | -0.010 |
Token | snake |
Token ID | 17522 |
Feature activation | +6.69e-03 |
Pos logit contributions | |
unison | +0.029 |
slices | +0.029 |
seventeen | +0.029 |
amaz | +0.028 |
fires | +0.028 |
Neg logit contributions | |
wear | -0.021 |
Woo | -0.020 |
bath | -0.020 |
day | -0.018 |
Mitt | -0.018 |
Token | ! |
Token ID | 0 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
wear | +0.009 |
bath | +0.008 |
Woo | +0.008 |
skipping | +0.008 |
Mitt | +0.008 |
Neg logit contributions | |
seventeen | -0.013 |
unison | -0.013 |
slices | -0.013 |
amaz | -0.013 |
aroma | -0.013 |
Token | The |
Token ID | 383 |
Feature activation | +1.66e-03 |
Pos logit contributions | |
unison | +0.042 |
pter | +0.041 |
seventeen | +0.041 |
amaz | +0.041 |
aroma | +0.041 |
Neg logit contributions | |
Woo | -0.024 |
Mitt | -0.023 |
wear | -0.022 |
bath | -0.022 |
Lucky | -0.022 |
Token | snake |
Token ID | 17522 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
wear | +0.002 |
bath | +0.002 |
Woo | +0.002 |
day | +0.002 |
important | +0.002 |
Neg logit contributions | |
unison | -0.003 |
seventeen | -0.003 |
slices | -0.003 |
amaz | -0.003 |
fires | -0.003 |
Token | was |
Token ID | 373 |
Feature activation | -1.90e-03 |
Pos logit contributions | |
wear | +0.028 |
bath | +0.025 |
always | +0.025 |
Mitt | +0.024 |
skipping | +0.024 |
Neg logit contributions | |
unison | -0.045 |
seventeen | -0.045 |
slices | -0.044 |
fires | -0.044 |
amaz | -0.044 |
Token | of |
Token ID | 286 |
Feature activation | +9.52e-03 |
Pos logit contributions | |
slices | +0.021 |
unison | +0.020 |
seventeen | +0.020 |
amaz | +0.020 |
seconds | +0.019 |
Neg logit contributions | |
wear | -0.013 |
Woo | -0.013 |
bath | -0.012 |
Mitt | -0.011 |
burn | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
wear | +0.010 |
Woo | +0.010 |
bath | +0.009 |
Mitt | +0.009 |
day | +0.009 |
Neg logit contributions | |
unison | -0.021 |
seventeen | -0.021 |
fires | -0.021 |
slices | -0.020 |
amaz | -0.020 |
Token | garden |
Token ID | 11376 |
Feature activation | +3.71e-03 |
Pos logit contributions | |
slices | +0.048 |
unison | +0.046 |
seventeen | +0.046 |
amaz | +0.046 |
pulp | +0.046 |
Neg logit contributions | |
wear | -0.036 |
Woo | -0.034 |
bath | -0.033 |
burn | -0.031 |
Mitt | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
wear | +0.004 |
day | +0.004 |
bath | +0.004 |
Woo | +0.004 |
skipping | +0.003 |
Neg logit contributions | |
unison | -0.008 |
seventeen | -0.008 |
amaz | -0.008 |
aroma | -0.008 |
pulp | -0.008 |
Token | But |
Token ID | 887 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
wear | +0.021 |
Woo | +0.020 |
Mitt | +0.020 |
bath | +0.019 |
Lucky | +0.019 |
Neg logit contributions | |
slices | -0.018 |
unison | -0.018 |
aroma | -0.018 |
seventeen | -0.017 |
pulp | -0.017 |
Token | they |
Token ID | 484 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
wear | +0.015 |
Woo | +0.014 |
bath | +0.014 |
Mitt | +0.013 |
day | +0.013 |
Neg logit contributions | |
unison | -0.020 |
seventeen | -0.020 |
slices | -0.020 |
amaz | -0.019 |
fires | -0.019 |
Token | are |
Token ID | 389 |
Feature activation | -4.08e-02 |
Pos logit contributions | |
unison | +0.034 |
seventeen | +0.033 |
slices | +0.033 |
amaz | +0.032 |
seconds | +0.032 |
Neg logit contributions | |
wear | -0.032 |
bath | -0.028 |
Woo | -0.028 |
day | -0.027 |
always | -0.027 |
Token | too |
Token ID | 1165 |
Feature activation | +8.31e-03 |
Pos logit contributions | |
slices | +0.083 |
unison | +0.082 |
seventeen | +0.081 |
amaz | +0.081 |
seconds | +0.080 |
Neg logit contributions | |
wear | -0.051 |
Woo | -0.048 |
bath | -0.047 |
burn | -0.043 |
Mitt | -0.043 |
Token | high |
Token ID | 1029 |
Feature activation | +1.39e-03 |
Pos logit contributions | |
wear | +0.011 |
Woo | +0.010 |
bath | +0.010 |
burn | +0.010 |
Lucky | +0.009 |
Neg logit contributions | |
slices | -0.017 |
seconds | -0.016 |
pulp | -0.016 |
seventeen | -0.016 |
amaz | -0.016 |
Token | for |
Token ID | 329 |
Feature activation | -4.19e-02 |
Pos logit contributions | |
wear | +0.002 |
Woo | +0.002 |
bath | +0.001 |
Mitt | +0.001 |
Lucky | +0.001 |
Neg logit contributions | |
unison | -0.003 |
slices | -0.003 |
seventeen | -0.003 |
amaz | -0.003 |
aroma | -0.003 |
Token | you |
Token ID | 345 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
unison | +0.077 |
seventeen | +0.076 |
amaz | +0.075 |
slices | +0.075 |
fires | +0.073 |
Neg logit contributions | |
wear | -0.062 |
Woo | -0.058 |
bath | -0.057 |
Mitt | -0.054 |
skipping | -0.053 |
Token | went |
Token ID | 1816 |
Feature activation | +7.45e-03 |
Pos logit contributions | |
unison | +0.072 |
slices | +0.072 |
seventeen | +0.071 |
amaz | +0.070 |
seconds | +0.069 |
Neg logit contributions | |
wear | -0.057 |
Woo | -0.053 |
bath | -0.052 |
Mitt | -0.049 |
Lucky | -0.048 |
Token | to |
Token ID | 284 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
wear | +0.011 |
Woo | +0.010 |
opsy | +0.010 |
scar | +0.010 |
Mitt | +0.010 |
Neg logit contributions | |
slices | -0.012 |
fires | -0.012 |
amaz | -0.011 |
seconds | -0.011 |
unison | -0.011 |
Token | bed |
Token ID | 3996 |
Feature activation | -6.61e-03 |
Pos logit contributions | |
amaz | +0.065 |
unison | +0.064 |
slices | +0.063 |
seconds | +0.062 |
seventeen | +0.061 |
Neg logit contributions | |
wear | -0.036 |
Woo | -0.035 |
burn | -0.032 |
important | -0.032 |
bath | -0.032 |
Token | with |
Token ID | 351 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
unison | +0.015 |
seventeen | +0.015 |
pulp | +0.015 |
slices | +0.015 |
amaz | +0.015 |
Neg logit contributions | |
wear | -0.007 |
Woo | -0.006 |
bath | -0.006 |
day | -0.006 |
Mitt | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
unison | +0.042 |
seventeen | +0.042 |
slices | +0.042 |
fires | +0.041 |
aroma | +0.041 |
Neg logit contributions | |
wear | -0.022 |
bath | -0.020 |
Woo | -0.020 |
Mitt | -0.018 |
day | -0.018 |
Token | smile |
Token ID | 8212 |
Feature activation | -7.58e-02 |
Pos logit contributions | |
unison | +0.026 |
slices | +0.026 |
seventeen | +0.026 |
fires | +0.026 |
seconds | +0.026 |
Neg logit contributions | |
wear | -0.015 |
bath | -0.014 |
Woo | -0.013 |
day | -0.012 |
Mitt | -0.012 |
Token | on |
Token ID | 319 |
Feature activation | -2.30e-02 |
Pos logit contributions | |
slices | +0.159 |
unison | +0.158 |
seventeen | +0.156 |
amaz | +0.156 |
aroma | +0.155 |
Neg logit contributions | |
wear | -0.087 |
Woo | -0.082 |
bath | -0.077 |
Mitt | -0.074 |
important | -0.073 |
Token | his |
Token ID | 465 |
Feature activation | -9.62e-04 |
Pos logit contributions | |
fires | +0.060 |
unison | +0.059 |
seventeen | +0.059 |
slices | +0.059 |
pulp | +0.058 |
Neg logit contributions | |
wear | -0.019 |
bath | -0.015 |
Woo | -0.015 |
Mitt | -0.013 |
day | -0.013 |
Token | face |
Token ID | 1986 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
seventeen | +0.002 |
slices | +0.002 |
teen | +0.002 |
fires | +0.002 |
pter | +0.002 |
Neg logit contributions | |
wear | -0.002 |
bath | -0.001 |
important | -0.001 |
day | -0.001 |
Lucky | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
unison | +0.066 |
slices | +0.066 |
seventeen | +0.065 |
amaz | +0.065 |
aroma | +0.063 |
Neg logit contributions | |
wear | -0.041 |
Woo | -0.038 |
bath | -0.037 |
Mitt | -0.035 |
day | -0.034 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.45e-02 |
Pos logit contributions | |
unison | +0.019 |
slices | +0.019 |
amaz | +0.019 |
seventeen | +0.019 |
seconds | +0.018 |
Neg logit contributions | |
wear | -0.014 |
Woo | -0.013 |
bath | -0.013 |
burn | -0.012 |
important | -0.012 |
Token | rode |
Token ID | 22075 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
wear | +0.010 |
Woo | +0.010 |
bath | +0.010 |
important | +0.009 |
opsy | +0.009 |
Neg logit contributions | |
slices | -0.014 |
amaz | -0.014 |
unison | -0.014 |
seconds | -0.014 |
seventeen | -0.013 |
Token | away |
Token ID | 1497 |
Feature activation | -4.59e-02 |
Pos logit contributions | |
wear | +0.040 |
Woo | +0.037 |
bath | +0.035 |
day | +0.034 |
Lucky | +0.033 |
Neg logit contributions | |
unison | -0.081 |
pulp | -0.080 |
slices | -0.080 |
seventeen | -0.080 |
aroma | -0.079 |
Token | with |
Token ID | 351 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
seventeen | +0.107 |
unison | +0.106 |
pulp | +0.105 |
slices | +0.104 |
aroma | +0.103 |
Neg logit contributions | |
wear | -0.049 |
Woo | -0.042 |
bath | -0.041 |
day | -0.041 |
skipping | -0.039 |
Token | a |
Token ID | 257 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
unison | +0.041 |
slices | +0.041 |
seventeen | +0.040 |
amaz | +0.040 |
aroma | +0.039 |
Neg logit contributions | |
wear | -0.023 |
Woo | -0.021 |
bath | -0.021 |
Mitt | -0.019 |
Lucky | -0.019 |
Token | big |
Token ID | 1263 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
unison | +0.028 |
slices | +0.028 |
seventeen | +0.027 |
amaz | +0.027 |
aroma | +0.027 |
Neg logit contributions | |
wear | -0.017 |
Woo | -0.016 |
bath | -0.016 |
Mitt | -0.015 |
Lucky | -0.014 |
Token | smile |
Token ID | 8212 |
Feature activation | -7.96e-02 |
Pos logit contributions | |
slices | +0.064 |
fires | +0.064 |
unison | +0.064 |
seventeen | +0.063 |
seconds | +0.061 |
Neg logit contributions | |
wear | -0.055 |
bath | -0.052 |
Woo | -0.050 |
day | -0.048 |
skipping | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | -5.61e-03 |
Pos logit contributions | |
slices | +0.164 |
unison | +0.163 |
seventeen | +0.160 |
amaz | +0.159 |
pter | +0.158 |
Neg logit contributions | |
wear | -0.094 |
Woo | -0.088 |
bath | -0.084 |
Mitt | -0.081 |
important | -0.079 |
Token | He |
Token ID | 679 |
Feature activation | -3.26e-02 |
Pos logit contributions | |
unison | +0.009 |
amaz | +0.009 |
slices | +0.009 |
seconds | +0.008 |
seventeen | +0.008 |
Neg logit contributions | |
wear | -0.010 |
burn | -0.009 |
bath | -0.009 |
Woo | -0.009 |
skipping | -0.008 |
Token | felt |
Token ID | 2936 |
Feature activation | +4.85e-03 |
Pos logit contributions | |
unison | +0.053 |
amaz | +0.052 |
slices | +0.052 |
seconds | +0.051 |
seventeen | +0.051 |
Neg logit contributions | |
wear | -0.052 |
Woo | -0.050 |
bath | -0.050 |
Mitt | -0.047 |
burn | -0.047 |
Token | happy |
Token ID | 3772 |
Feature activation | -5.69e-02 |
Pos logit contributions | |
wear | +0.006 |
burn | +0.006 |
Woo | +0.006 |
bath | +0.006 |
Mitt | +0.006 |
Neg logit contributions | |
amaz | -0.009 |
unison | -0.009 |
slices | -0.009 |
seconds | -0.009 |
seventeen | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
seventeen | +0.159 |
unison | +0.151 |
pulp | +0.148 |
teen | +0.147 |
aroma | +0.147 |
Neg logit contributions | |
wear | -0.045 |
day | -0.041 |
bath | -0.038 |
always | -0.037 |
skipping | -0.037 |
Token | ride |
Token ID | 6594 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
Woo | +0.022 |
wear | +0.020 |
important | +0.020 |
scar | +0.020 |
burn | +0.020 |
Neg logit contributions | |
amaz | -0.041 |
unison | -0.039 |
slices | -0.038 |
seconds | -0.038 |
ah | -0.038 |
Token | my |
Token ID | 616 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
wear | +0.041 |
Woo | +0.038 |
bath | +0.036 |
day | +0.034 |
Lucky | +0.034 |
Neg logit contributions | |
unison | -0.068 |
amaz | -0.067 |
seventeen | -0.066 |
slices | -0.066 |
pulp | -0.065 |
Token | bike |
Token ID | 7161 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
wear | +0.014 |
Woo | +0.013 |
bath | +0.013 |
Mitt | +0.012 |
Lucky | +0.012 |
Neg logit contributions | |
unison | -0.023 |
slices | -0.023 |
seventeen | -0.022 |
amaz | -0.022 |
aroma | -0.022 |
Token | now |
Token ID | 783 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
wear | +0.011 |
Woo | +0.010 |
bath | +0.009 |
Mitt | +0.009 |
Lucky | +0.008 |
Neg logit contributions | |
slices | -0.022 |
unison | -0.022 |
seventeen | -0.022 |
amaz | -0.022 |
fires | -0.022 |
Token | ? |
Token ID | 30 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
wear | +0.026 |
Woo | +0.025 |
Mitt | +0.024 |
bath | +0.024 |
Lucky | +0.023 |
Neg logit contributions | |
slices | -0.046 |
aroma | -0.046 |
amaz | -0.046 |
seventeen | -0.045 |
unison | -0.045 |
Token | Please |
Token ID | 4222 |
Feature activation | -7.83e-02 |
Pos logit contributions | |
wear | +0.022 |
Woo | +0.021 |
bath | +0.021 |
Mitt | +0.020 |
day | +0.020 |
Neg logit contributions | |
unison | -0.015 |
slices | -0.015 |
seventeen | -0.015 |
amaz | -0.015 |
seconds | -0.014 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
unison | +0.150 |
amaz | +0.146 |
seconds | +0.146 |
slices | +0.145 |
seventeen | +0.145 |
Neg logit contributions | |
Woo | -0.095 |
wear | -0.095 |
bath | -0.091 |
important | -0.088 |
Mitt | -0.086 |
Token |
|
Token ID | 198 |
Feature activation | -3.13e-03 |
Pos logit contributions | |
unison | +0.021 |
slices | +0.021 |
seventeen | +0.021 |
amaz | +0.021 |
aroma | +0.021 |
Neg logit contributions | |
wear | -0.018 |
Woo | -0.017 |
bath | -0.017 |
Mitt | -0.016 |
Lucky | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
slices | +0.006 |
unison | +0.005 |
seconds | +0.005 |
amaz | +0.005 |
seventeen | +0.005 |
Neg logit contributions | |
wear | -0.005 |
Woo | -0.004 |
burn | -0.004 |
important | -0.004 |
bath | -0.004 |
Token | M |
Token ID | 44 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
unison | +0.021 |
slices | +0.021 |
seventeen | +0.021 |
amaz | +0.020 |
aroma | +0.020 |
Neg logit contributions | |
wear | -0.021 |
Woo | -0.019 |
bath | -0.019 |
Mitt | -0.018 |
Lucky | -0.018 |
Token | ummy |
Token ID | 13513 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
slices | +0.049 |
unison | +0.048 |
aroma | +0.047 |
amaz | +0.047 |
seconds | +0.047 |
Neg logit contributions | |
wear | -0.033 |
Woo | -0.032 |
bath | -0.028 |
Mitt | -0.028 |
burn | -0.028 |
Token | granted |
Token ID | 7520 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
wear | +0.008 |
Woo | +0.007 |
bath | +0.007 |
burn | +0.007 |
Mitt | +0.006 |
Neg logit contributions | |
unison | -0.011 |
slices | -0.011 |
amaz | -0.010 |
seventeen | -0.010 |
seconds | -0.010 |
Token | ! |
Token ID | 0 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
wear | +0.015 |
Woo | +0.014 |
bath | +0.013 |
Mitt | +0.012 |
Lucky | +0.012 |
Neg logit contributions | |
unison | -0.024 |
slices | -0.024 |
seventeen | -0.023 |
amaz | -0.023 |
aroma | -0.023 |
Token | The |
Token ID | 383 |
Feature activation | +7.09e-03 |
Pos logit contributions | |
unison | +0.022 |
slices | +0.022 |
amaz | +0.021 |
seconds | +0.021 |
seventeen | +0.021 |
Neg logit contributions | |
wear | -0.018 |
bath | -0.016 |
burn | -0.016 |
Woo | -0.016 |
important | -0.016 |
Token | star |
Token ID | 3491 |
Feature activation | +6.91e-03 |
Pos logit contributions | |
wear | +0.009 |
Woo | +0.008 |
bath | +0.008 |
Mitt | +0.008 |
day | +0.008 |
Neg logit contributions | |
unison | -0.014 |
slices | -0.014 |
seventeen | -0.014 |
amaz | -0.014 |
aroma | -0.014 |
Token | shone |
Token ID | 44193 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
wear | +0.009 |
bath | +0.008 |
day | +0.008 |
Woo | +0.007 |
important | +0.007 |
Neg logit contributions | |
unison | -0.014 |
seventeen | -0.014 |
slices | -0.014 |
amaz | -0.014 |
fires | -0.013 |
Token | brightly |
Token ID | 35254 |
Feature activation | -7.38e-02 |
Pos logit contributions | |
wear | +0.025 |
important | +0.024 |
Woo | +0.023 |
always | +0.023 |
day | +0.023 |
Neg logit contributions | |
unison | -0.041 |
seventeen | -0.040 |
pulp | -0.040 |
pter | -0.040 |
slices | -0.039 |
Token | and |
Token ID | 290 |
Feature activation | -5.02e-04 |
Pos logit contributions | |
unison | +0.180 |
seventeen | +0.177 |
slices | +0.176 |
pulp | +0.176 |
amaz | +0.172 |
Neg logit contributions | |
wear | -0.066 |
day | -0.059 |
bath | -0.053 |
Woo | -0.049 |
important | -0.049 |
Token | lit |
Token ID | 6578 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
unison | +0.001 |
seventeen | +0.001 |
slices | +0.001 |
pter | +0.001 |
fires | +0.001 |
Neg logit contributions | |
wear | -0.001 |
Woo | -0.001 |
day | -0.001 |
bath | -0.001 |
Mitt | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
slices | +0.078 |
unison | +0.077 |
seventeen | +0.076 |
pulp | +0.076 |
fires | +0.075 |
Neg logit contributions | |
wear | -0.021 |
Woo | -0.017 |
bath | -0.016 |
important | -0.016 |
day | -0.015 |
Token | night |
Token ID | 1755 |
Feature activation | -8.44e-03 |
Pos logit contributions | |
slices | +0.022 |
seventeen | +0.022 |
unison | +0.022 |
seconds | +0.021 |
fires | +0.021 |
Neg logit contributions | |
wear | -0.017 |
bath | -0.017 |
day | -0.016 |
important | -0.016 |
Woo | -0.015 |
Token | sky |
Token ID | 6766 |
Feature activation | +8.54e-03 |
Pos logit contributions | |
seventeen | +0.017 |
unison | +0.017 |
pulp | +0.017 |
slices | +0.016 |
anish | +0.016 |
Neg logit contributions | |
wear | -0.011 |
day | -0.011 |
bath | -0.011 |
always | -0.009 |
important | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +1.61e-03 |
Pos logit contributions | |
seventeen | +0.121 |
pulp | +0.120 |
slices | +0.118 |
unison | +0.116 |
amaz | +0.113 |
Neg logit contributions | |
wear | -0.050 |
Woo | -0.045 |
day | -0.044 |
important | -0.042 |
bath | -0.042 |
Token | Tiger |
Token ID | 17030 |
Feature activation | +9.36e-03 |
Pos logit contributions | |
wear | +0.002 |
Woo | +0.002 |
bath | +0.002 |
Mitt | +0.002 |
Lucky | +0.002 |
Neg logit contributions | |
slices | -0.003 |
unison | -0.003 |
seventeen | -0.003 |
aroma | -0.003 |
amaz | -0.003 |
Token | looked |
Token ID | 3114 |
Feature activation | +6.20e-03 |
Pos logit contributions | |
wear | +0.013 |
Woo | +0.012 |
bath | +0.011 |
Lucky | +0.011 |
Mitt | +0.011 |
Neg logit contributions | |
slices | -0.018 |
unison | -0.018 |
seventeen | -0.018 |
amaz | -0.018 |
pulp | -0.017 |
Token | around |
Token ID | 1088 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
wear | +0.006 |
Woo | +0.005 |
bath | +0.005 |
important | +0.005 |
day | +0.005 |
Neg logit contributions | |
slices | -0.015 |
unison | -0.015 |
seventeen | -0.015 |
pulp | -0.014 |
seconds | -0.014 |
Token | in |
Token ID | 287 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
seventeen | +0.096 |
pulp | +0.096 |
slices | +0.093 |
unison | +0.092 |
bridge | +0.091 |
Neg logit contributions | |
day | -0.025 |
wear | -0.025 |
skipping | -0.022 |
always | -0.021 |
bath | -0.021 |
Token | amaz |
Token ID | 40642 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
seventeen | +0.031 |
pulp | +0.031 |
fires | +0.030 |
anish | +0.029 |
slices | +0.029 |
Neg logit contributions | |
wear | -0.033 |
day | -0.031 |
bath | -0.031 |
skipping | -0.031 |
jog | -0.030 |
Token | ement |
Token ID | 972 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
pulp | +0.169 |
seventeen | +0.169 |
slices | +0.155 |
unison | +0.150 |
fires | +0.147 |
Neg logit contributions | |
wear | -0.226 |
important | -0.223 |
always | -0.220 |
ie | -0.214 |
day | -0.210 |
Token | and |
Token ID | 290 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
seventeen | +0.087 |
unison | +0.085 |
slices | +0.085 |
pulp | +0.084 |
aroma | +0.082 |
Neg logit contributions | |
wear | -0.036 |
day | -0.030 |
bath | -0.030 |
skipping | -0.030 |
Woo | -0.029 |
Token | he |
Token ID | 339 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
wear | +0.021 |
Woo | +0.019 |
day | +0.018 |
Lucky | +0.018 |
bath | +0.018 |
Neg logit contributions | |
unison | -0.032 |
slices | -0.032 |
seventeen | -0.031 |
pulp | -0.030 |
anish | -0.030 |
Token | knew |
Token ID | 2993 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
seventeen | +0.022 |
slices | +0.022 |
unison | +0.022 |
aroma | +0.021 |
pulp | +0.021 |
Neg logit contributions | |
wear | -0.018 |
Woo | -0.016 |
bath | -0.016 |
always | -0.015 |
Lucky | -0.015 |
Token | he |
Token ID | 339 |
Feature activation | -6.68e-03 |
Pos logit contributions | |
amaz | +0.061 |
unison | +0.060 |
seconds | +0.060 |
slices | +0.059 |
fires | +0.058 |
Neg logit contributions | |
wear | -0.048 |
Woo | -0.046 |
bath | -0.045 |
burn | -0.044 |
Mitt | -0.042 |