INDEX

Explanations

numbers and specific items

based on your instructions

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 قیام

0.56

adería

0.51

 नींव

0.48

ızı

0.47

 ურთიერთ

0.47

renamiento

0.46

 الجمع

0.46

 الات

0.45

ủy

0.45

 பயிற்சிய

0.45

POSITIVE LOGITS

0.45

Oct

0.44

 selects

0.44

 organes

0.43

選

0.42

OMA

0.42

OKE

0.42

 executive

0.40

$~

0.40

Activations Density 0.000%