INDEX

Explanations

technical terms and actions

understand and generate

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

 తెలు

0.57



0.57

ôn

0.57

 publicó

0.53

 escuchar

0.53

 станда

0.52

 automakers

0.52

 habló

0.52

 drunkenness

0.51

 తాగ

0.51

POSITIVE LOGITS

AV

0.57

Vis

0.50

Word

0.48

Fib

0.47

Gauge

0.47

BV

0.46

Three

0.46

in

0.46

Text

0.45

Activations Density 0.001%