INDEX

Explanations

non-English words and concepts

using transformer models

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

0.46

ToShow

0.46

 Manufacturers

0.45

は

0.44

 vrlo

0.44

är

0.43

 ヴィンテージ

0.43

 tulee

0.43

 calles

0.43

 manufacturers

0.43

POSITIVE LOGITS

 momentous

0.52

GALAD

0.48

GoObject

0.43

 کشتی

0.42

 επίπε

0.42

邬

0.41

 restful

0.41

鼾

0.41

澶

0.41

 réc

0.40

Activations Density 0.001%