INDEX

Explanations

its

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

expandindo

-1.09

 Roskov

-1.09

 propOrder

-1.08

 ujednoznacz

-1.04

OGND

-0.99

 nakalista

-0.98

 protoimpl

-0.96

 Мексичка

-0.96

awtextra

-0.94

 resourceCulture

-0.93

POSITIVE LOGITS

 length

0.63

length

0.58

 full

0.57

 entire

0.53

0.51

LENGTH

0.48

우

0.44

Length

0.44

 LENGTH

0.44

 completo

0.43

Activations Density 0.115%