INDEX

Explanations

chain of thought prompting

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 pouze

0.95

 Dirac

0.92

 relativamente

0.92

ندوق

0.89

 espacial

0.89

 térm

0.88

 Redist

0.87

আন্তর্জাতিক

0.87

 такого

0.86

ૌ

0.86

POSITIVE LOGITS

[[

1.19

[[

1.14

1.10

1.06

[_

1.02

[(

0.96

[[[[

0.95

[[[

0.90

[(

0.90

[<

0.89

Activations Density 0.002%