INDEX

Explanations

bulleted lists of options

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

্ত

2.88

2.80

 upside

2.30

2.23

ர்த்த

2.10

 spot

2.06

ick

2.04

yer

2.04

2.01

sem

2.00

POSITIVE LOGITS

WholeProgram

3.03

迤

2.97

ostiene

2.90

 тексту

2.89

 âmbito

2.87

ifiant

2.86

യാള

2.84

 règle

2.83

蓣

2.81

 напря

2.80

Activations Density 0.258%