INDEX

Explanations

HTML and code structure tags

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

</h4>

-2.33

 which

-1.99

 whose

-1.88

</b>

-1.88

 pertaining

-1.87

 through

-1.74

by

-1.73

 más

-1.72

 occasionally

-1.72

</em>

-1.71

POSITIVE LOGITS

桖

1.96

 花纹

1.88

 唯美

1.84

粜

1.83

媗

1.83

tures

1.80

our

1.78

having

1.77

 ใน

1.76

 素描

1.75

Activations Density 0.004%

No Known Activations