INDEX

Explanations

press release identifiers

code elements 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

 नाइट्र

0.27

িগুণ

0.27

marshalN

0.26

 শ্বাস

0.25

𒇽

0.25

<unused40>

0.24

ೊಂಡು

0.24

 স্থির

0.24

URNIZOR

0.24

 ক্ষীণ

0.24

POSITIVE LOGITS

0.33

0.32

↵↵

0.30

↵

0.29

。

0.27

.”

0.26

↵↵↵

0.26

."

0.26

,"

0.25

Activations Density 0.002%