INDEX

Explanations

numbers dates email addresses

reply all to this email

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

Burning

0.47

Unlock

0.46

Dump

0.46

Github

0.45

Honestly

0.45

Video

0.44

Epoch

0.44

indruck

0.44

github

0.43

 Мін

0.43

POSITIVE LOGITS

ရေ

0.45

 glau

0.44

 diast

0.44

 peritoneal

0.44

 folklor

0.44

 breat

0.43

 folate

0.43

سته

0.43

 heme

0.43

 poesia

0.42

Activations Density 0.034%