INDEX

Explanations

slightly larger

answer/output with specifics

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

kef

0.43

өз

0.41

 районов

0.41

違

0.40

浬

0.40

 conquests

0.40

 شرطونه

0.39

ponsorship

0.39

 acuer

0.39

 ಸ್ವಾ

0.39

POSITIVE LOGITS

 Document

0.43

 Alles

0.42

Ghost

0.41

 Intelli

0.41

kiyor

0.41

volving

0.41

是

0.41

峤

0.41

 unterstützt

0.40

 Beispiel

0.40

Activations Density 0.001%