INDEX

Explanations

dates and numbers

numbers and punctuation

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

部分的

0.28

zca

0.27

RIBUTE

0.27

myY

0.27

沩

0.26

ૃતિ

0.26

")))

0.25

ydı

0.25

<unused9>

0.25

𝘷

0.25

POSITIVE LOGITS

and

0.28

 renov

0.26

<strong>

0.26

 Limited

0.25

↵↵

0.25

↵↵↵

0.24

ɖ

0.24

 While

0.24

Activations Density 0.000%