INDEX

Explanations

research documents

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

BeginContext

-0.63

 propOrder

-0.48

withstanding

-0.45

 MainAxisSize

-0.44

 study

-0.44

Brainz

-0.42



-0.42

OPSIS

-0.42

脚注の使い方

-0.42

 समीक्षक

-0.42

POSITIVE LOGITS

 financières

0.69

 fevere

0.67

 pleaſure

0.67

Past

0.66

 Past

0.66

 houſe

0.65

 Majefty

0.65

 purpoſe

0.64

 élevées

0.63

 fondament

0.63

Activations Density 0.000%