INDEX

Explanations

formatting list items

This neuron detects ingredient names (especially natural additives and extracts) in product or recipe ingredient lists.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 技术

0.93

 项目

0.90

Cdc

0.90

Psp

0.89

 关闭

0.86

Cbd

0.86

 点击

0.84

 公司

0.83

 产品

0.82

 设计

0.79

POSITIVE LOGITS

 WHAT

1.70

THE

1.66

THE

1.63

NEW

1.63

 ANOTHER

1.59

DEFAULT

1.58

HOW

1.56

 OTHER

1.55

 LOCAL

1.52

 FOUR

1.52

Activations Density 0.381%