INDEX

Explanations

help

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

Portail

-0.64

 pinulongan

-0.57

 trattato

-0.52

findpost

-0.51

ByUsername

-0.48

Reducing

-0.48

Treating

-0.48

reducing

-0.46

tagHelperRunner

-0.45

 Treating

-0.45

POSITIVE LOGITS

 AttributeSet

0.71

 tell

0.65

tagext

0.64

 ISNI

0.61

kuuta

0.59

ètres

0.57

 Viscount

0.56

usermodel

0.55

észetes

0.55

]=>

0.54

Activations Density 0.003%

No Known Activations