INDEX

Explanations

technical documentation and code structure

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

५

-0.82

 Seulement

-0.80

覚

-0.79

၎

-0.79

gav

-0.78

 victoire

-0.77

 chồng

-0.76

싸

-0.76

 Vaticano

-0.74

 посмотре

-0.74

POSITIVE LOGITS

人都

0.93

 hochwertige

0.92

 Gets

0.91

 tornou

0.91

{\"

0.89

 mögliche

0.86

 nuisance

0.85

LEVANT

0.83

 controls

0.83

 spezielle

0.80

Activations Density 0.019%