INDEX

Explanations

greater than or less than

The neuron detects comparative terms expressing quantitative relationships (e.g. greater, less, equal, smaller, shorter).

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

𓐍

-1.36

 كورة

-1.32

 două

-1.28

 afectada

-1.20

Հ

-1.16

 exerc

-1.16

鹪

-1.16

卺

-1.16

 fizer

-1.15

 produse

-1.14

POSITIVE LOGITS

 than

2.75

of

2.19

or

2.16

to

1.89

выше

1.45

 Städten

1.30

Então

1.30

 dahulu

1.29

 seorang

1.27

 atau

1.27

Activations Density 0.020%