INDEX

Explanations

code blocks with align or curly braces

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

iesel

-0.09

VEN

-0.08

sword

-0.07

ÙĤØ§ÙĦ

-0.07

izard

-0.07

¬ģ

-0.07

ä¾ĥ

-0.07

ancis

-0.07

izzer

-0.07

Ð²ÑĤÐ¾

-0.07

POSITIVE LOGITS

pectral

0.07

uchen

0.07

Bah

0.07

ä»£ä¹ĭ

0.06

ãģĤãĤĭ

0.06

aye

0.06

å°ī

0.06

icont

0.06

ged

0.06

 Berm

0.06

Activations Density 0.001%