INDEX

Explanations

objective lens component

identify trends and annotation

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

Զ

0.52

 pracovní

0.47

স

0.46

去

0.45

Edges

0.44

Ná

0.44

Battle

0.44

ไป

0.44

鱻

0.43

Кра

0.43

POSITIVE LOGITS

lyan

0.49

ue

0.48

 judiciary

0.45

 omdat

0.44

 looted

0.44

 lawan

0.44

 yorum

0.44

 thumb

0.43

 loan

0.43

 ramai

0.43

Activations Density 0.001%