INDEX

Explanations

code comments and legal filings

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

 groups

-0.78

 says

-0.78

 last

-0.78

amoja

-0.75

 should

-0.74

しないと

-0.74

 mentioned

-0.73

 international

-0.72

Ers

-0.72

힌

-0.71

POSITIVE LOGITS

Nimbus

0.85

额头

0.83

 cív

0.82

 snabbt

0.81

冷漠

0.76

 Palin

0.75

pecha

0.74

Cuándo

0.73

ModuleName

0.73

 oczeki

0.73

Activations Density 0.003%