INDEX

Explanations

Code and URLs

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

parsedMessage

-0.49

mit

-0.49

-0.47

shadowRadius

-0.43

 Base

-0.40

RAL

-0.40

égio

-0.39

 Kleid

-0.39

Klar

-0.38

POSITIVE LOGITS

 GenerationType

0.68

 Comstock

0.65

InstanceState

0.63

 ब्रेकडाउन

0.63

DebuggerNonUser

0.62

 quæ

0.60

InvalidProtocol

0.59

ionView

0.58

enderror

0.56

новништво

0.56

Activations Density 0.014%