INDEX

Explanations

knows what

designated as confidential

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

 strait

0.48

のみ

0.48

Joshua

0.47

 probabilmente

0.46

Ronald

0.46

 richiest

0.46

Josh

0.45

 ומ

0.45

ཝ

0.45

 complejos

0.45

POSITIVE LOGITS

小组

0.46

រួម

0.45

iebel

0.42

opaedic

0.42

╗

0.42

㇇

0.42

вшего

0.41

idän

0.41

LoopBlend

0.41

$(*

0.41

Activations Density 0.001%