INDEX

Explanations

citations and 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

 psychologie

-1.33

 conster

-1.07

がありますが

-1.05

 تصميم

-1.05

izational

-1.02

೫

-1.01

 vanille

-1.00

 étions

-1.00

೮

-1.00

罍

-1.00

POSITIVE LOGITS

extendable

1.05

uega

1.04

 expertly

1.03

 encompassing

0.99

sistem

0.98

 encompasses

0.98

 ucran

0.97

ciec

0.94

 eliminating

0.92

 lender

0.92

Activations Density 0.041%