INDEX

Explanations

url, code, and enzyme context

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

nev

-0.11

 commons

-0.10

urd

-0.10

 Shed

-0.10

 Pais

-0.10

anker

-0.10

urch

-0.10

 Loft

-0.10

vig

-0.10

Wak

-0.10

POSITIVE LOGITS

 spir

0.09

 destruct

0.08

GD

0.08

 encyclopedia

0.07

bin

0.07

 wound

0.07

GD

0.07

eming

0.07

indi

0.07

erce

0.07

Activations Density 0.036%