INDEX

Explanations

base

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 discounts

-0.28

.Tick

-0.28

ë¶Ī

-0.27

 Discounts

-0.26

 Tick

-0.26

è´§æºĲ

-0.26

ervo

-0.25

ä¸¤åľ°

-0.25

UNU

-0.25

è½®åĽŀ

-0.25

POSITIVE LOGITS

less

0.37

plate

0.36

 plate

0.31

of

0.30

ball

0.29

äºļ

0.29

lead

0.29

æ¨ª

0.29

board

0.28

lessly

0.28

Activations Density 0.028%