INDEX

Explanations

Code/Programming

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 reduction

-0.56

Reduction

-0.54

 Reduction

-0.53

reduction

-0.52

MECHAN

-0.51

 mechanical

-0.51

 REDUCTION

-0.47

Mechan

-0.47

InputBorder

-0.47

 Petit

-0.46

POSITIVE LOGITS

ScopeManager

0.62

uxxxx

0.58

ForRow

0.53

砌

0.52

 AssemblyProduct

0.52

XmlType

0.51

مقاله

0.51

 systematically

0.49

Hosp

0.49

iyle

0.48

Activations Density 0.001%