INDEX

Explanations

website navigation elements

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 eager

-0.93

 both

-0.92

tü

-0.91

 these

-0.88

 natale

-0.88

There

-0.87

 several

-0.87

Several

-0.86

 through

-0.86

 because

-0.85

POSITIVE LOGITS

FAQ

1.30

FAQ

1.15

 gift

1.12

 contact

1.09

 login

1.09

Contact

1.06

 Contact

1.05

 blog

1.01

 Frequently

0.97

 methodology

0.96

Activations Density 0.107%