INDEX

Explanations

`/` and `login`, `hello`, `users`

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

犹如

1.59

hes

1.56

پ

1.54

狠

1.53

极其

1.50

最后

1.50

ittle

1.47

گیرد

1.46

anguage

1.45

让我

1.43

POSITIVE LOGITS

 greener

1.61

ále

1.60

вропей

1.49

ilib

1.46

 グリーン

1.44

 rurales

1.44

覧

1.44

 Collins

1.43

 지하

1.40

 아니

1.40

Activations Density 0.044%