INDEX

Explanations

I cant

hate and offensive language

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

toFixed

0.40

ischem

0.38

 সানডে

0.38

Arqu

0.37

 حلقه

0.37

 Penc

0.37

ाइस

0.36

Eis

0.36

 quelconque

0.36

emis

0.36

POSITIVE LOGITS

 notes

0.54

track

0.42

 realize

0.42

notes

0.42

 offend

0.42

 cheat

0.41

ノート

0.41

 note

0.40

最小

0.40

 access

0.40

Activations Density 0.000%