INDEX

Explanations

closing, fill-in-the-blank, code

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

Images

2.22

Frames

2.13

'^

2.12

más

2.10

'~

2.08

Idea

2.07

Prototype

1.99

Frame

1.98

tiktok

1.96

Offers

1.95

POSITIVE LOGITS

য়ে

1.98

…”

1.97

င်း

1.95

 Hong

1.90

ر

1.89

 می‌توانید

1.87

ంగ్

1.87

ayah

1.86

몰

1.86

ه‌ها

1.84

Activations Density 0.003%