INDEX

Explanations

references to detector components or parameters in scientific context

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pc

-1.36

 متعلقه

-0.60

itope

-0.59

MLLoader

-0.57

 autorytatywna

-0.56

țional

-0.56

agisse

-0.56

joseph

-0.55

ior

-0.54

 Himo

-0.54

POSITIVE LOGITS

PC

1.18

PCs

0.65

pkg

0.59

が広

0.53

 sleeved

0.50

ccia

0.49

strSql

0.49

 ПК

0.49

един

0.48

 Codable

0.48

Activations Density 0.008%