INDEX

Explanations

specific words and punctuation

open source software

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

醉

0.55

สนามกีฬา

0.50

 भरने

0.48

媟

0.46

achite

0.46

 текста

0.45

 어려운

0.45

ገኛ

0.44

За

0.44

 मुश्किल

0.43

POSITIVE LOGITS

 ANSI

0.49

 Emma

0.45

 Canadian

0.45

 hjel

0.44

 compiler

0.43

 Model

0.42

 SORT

0.42

BIS

0.42

ROM

0.41

 Kind

0.40

Activations Density 0.001%