INDEX

Explanations

let's tackle or break down

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

또한

1.57

 বাহিরে

1.56

 देखील

1.56

非常的

1.50

 jedoch

1.49

utilizzo

1.48

 Utilizing

1.41

 permasalahan

1.40

 אשר

1.39

不断的

1.36

POSITIVE LOGITS

 doubtless

1.42

 lousy

1.31

 वगैरह

1.30

 cheerfully

1.30

 variously

1.22

 luxuri

1.21

 needn

1.17

 heft

1.15

 esche

1.12

 darn

1.11

Activations Density 2.370%