INDEX

Explanations

code structures and data references

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 руб

-0.80

SAV

-0.69

IM

-0.67

任

-0.66

Jar

-0.65

 اطلاع

-0.65

isMethod

-0.65

 Coin

-0.65

clu

-0.65

Beginning

-0.64

POSITIVE LOGITS

onik

0.69

 anhydride

0.67

 planning

0.67

 بات

0.67

дьба

0.66

 treaty

0.66

Bing

0.66

 battles

0.66

erah

0.65

litis

0.65

Activations Density 0.072%