INDEX

Explanations

numbers followed by separators

version: "3.x"

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

ులు

0.43

برها

0.43

 längst

0.42

<unused84>

0.42

 गरम

0.42

 mated

0.42

 बेटर

0.41

 veik

0.40

ুল্লাহ

0.40

￺

0.39

POSITIVE LOGITS

Ф

0.46

0.44

 Xiao

0.44

 William

0.42

Ф

0.41

وی

0.40

 Frederick

0.40

Л

0.40

 الأبيض

0.39

ј

0.39

Activations Density 0.004%