INDEX

Explanations

request access

This is a blueprint

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

祁

0.44

Gä

0.43

礻

0.43

 establecidas

0.41

書

0.40

本书

0.39

 ακό

0.39

 cargas

0.39

 Clusters

0.38

POSITIVE LOGITS

 Halloween

0.52

 halloween

0.51

NAR

0.50

ท้อง

0.50

osław

0.49

ოვნ

0.48

 我們

0.47

 Authorized

0.47

Authenticated

0.46

 Professional

0.46

Activations Density 0.001%