INDEX

Explanations

new

The neuron detects uses of the adjective “new.”

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

магазин

-1.73

<bos>

-1.73

onymes

-1.47

cún

-1.41

is

-1.39

 hermosas

-1.29

商人

-1.26

магази

-1.24

普遍

-1.23

tanque

-1.20

POSITIVE LOGITS

we

1.59

不由得

1.54

ER

1.46

メルカリ

1.45

EDOM

1.40

跸

1.38

大満足

1.36

ブログ村

1.35

 satisfacción

1.34

 mortos

1.34

Activations Density 0.041%