INDEX

Explanations

as a reference

details, reward, innovate, gods

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

idle

0.44

rouge

0.43

bold

0.43

 बेच

0.42

idea

0.42

"]);

0.42

↵↵↵↵↵

0.41

Ale

0.41

ادل

0.41

त्मा

0.40

POSITIVE LOGITS

 dispositivi

0.52

 இவற்ற

0.52

 முன்ன

0.50

 detain

0.50

ຢູ່ໃນ

0.50

 ప్రస్తుతం

0.50

 frequ

0.49

ංග

0.49

𝐽

0.49

 immunos

0.49

Activations Density 0.001%