INDEX

Explanations

PMID or doi following punctuation

learning rate

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

स्‍त

0.44

椁

0.42

িলারা

0.41

柠檬

0.40

="#"><

0.38

доро

0.38

僮

0.38

েই

0.37

ಗಳಲ್ಲಿ

0.37

身上的

0.37

POSITIVE LOGITS

[!

0.43

ini

0.41

substack

0.41

pin

0.38

**

0.37

height

0.37

groupby

0.37

([],

0.37

hspace

0.36

","",

0.36

Activations Density 0.020%