INDEX

Explanations

breads, snacks, and stuffed foods

The neuron detects individual food or dish names (e.g. “bread,” “nachos,” “pretzels,” “steaks”) in the text.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

لة

-1.61

Fitment

-1.30

 działalności

-1.24

 fotográfico

-1.23

 colorida

-1.23

вгений

-1.23

}$$

-1.23

 ample

-1.23

Ident

-1.19

ktı

-1.19

POSITIVE LOGITS

game

1.34

 including

1.32

kolan

1.31

 così

1.28

 mulheres

1.27

who

1.26

 коррек

1.24

nuovo

1.24

 også

1.23

Activations Density 0.018%