INDEX

Explanations

mysterious ways

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

UIActions

0.54

 Ấn

0.51

 destinado

0.51

 शिक्षकों

0.50

 Thì

0.48

 láp

0.48

ployed

0.47

úst

0.47

 Doubles

0.46

vió

0.46

POSITIVE LOGITS

 rearrangement

0.52

AND

0.47

 relocation

0.46

 wheels

0.46

OR

0.44

ZA

0.44

 evasion

0.43

 inflection

0.43

 multifamily

0.43

Activations Density 0.000%