INDEX

Explanations

good with descriptions

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

ness

1.96

but

1.93

ity

1.92

然而

1.83

 아니다

1.83

성은

1.82

성을

1.80

 thereof

1.78

yet

1.76

itet

1.76

POSITIVE LOGITS

 molte

1.59

many

1.44

 drugim

1.40

 muchos

1.39

 многие

1.38

燹

1.34

 многих

1.34

 many

1.29

deal

1.25

 muchas

1.22

Activations Density 0.126%