INDEX

Explanations

not to be confused with

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

klare

-1.77

 нашей

-1.76

 désolés

-1.54

 dát

-1.53

 нашего

-1.52

 depois

-1.49

έα

-1.48

ומית

-1.48

 rápidas

-1.48

 этих

-1.48

POSITIVE LOGITS

as

2.81

for

2.48

 from

2.47

any

2.44

2.41

on

2.31

one

2.22

 during

2.11

 that

2.09

to

2.03

Activations Density 0.164%