INDEX

Explanations

induce quick settlement

specific topics 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

0.58

 Ashok

0.52

 Iain

0.52

 Orwell

0.51

 Representations

0.49

 Wilfred

0.49

 Kevin

0.48

 Eight

0.48

Ash

0.47

POSITIVE LOGITS

ರಿನ

0.53

晚餐

0.50

ピアノ

0.50

प्ताहिक

0.49

መም

0.48

噜

0.48

 condensing

0.48

ел

0.47

 זמן

0.46

櫸

0.46

Activations Density 0.001%