INDEX

Explanations

[brackets with specific headings]

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

嘅

-1.48

-1.38

 cambió

-1.35

فة

-1.32

了两

-1.31

-1.28

⁦

-1.26

 transforms

-1.24

铙

-1.22

-1.21

POSITIVE LOGITS

 they

1.94

ב

1.63

 galima

1.59

屃

1.54

Что

1.48

}());

1.45

این

1.44

Здесь

1.41

átku

1.38

itieren

1.35

Activations Density 0.124%