INDEX

Explanations

length or duration descriptors

specific continuations after keywords

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

 instru

0.51

 filets

0.50

 ساف

0.49

 Kapitel

0.47

 eure

0.45

 gelt

0.45

:*

0.44

AID

0.44

 רי

0.43

CHEC

0.43

POSITIVE LOGITS

ων

0.46

 принима

0.45

 बेहद

0.45

 accept

0.43

λο

0.43

πα

0.42

Tao

0.42

ᱩ

0.42

 통해

0.42

த்தோ

0.42

Activations Density 0.001%