INDEX

Explanations

technical names and identifiers

The neuron fires on capitalized names and acronyms—especially software project, product, or company identifiers.

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

鑷

-1.32

ÉM

-1.30

꿔

-1.28

ÍO

-1.24

 เรา

-1.24

ofern

-1.23

")

-1.22

’，

-1.22

ريكية

-1.21

 子ども

-1.21

POSITIVE LOGITS

 alemanes

1.52

<bos>

1.44

OUTUBE

1.42

to

1.40

1.29

 there

1.29

早已

1.24

1.20

hurst

1.16

irited

1.16

Activations Density 0.009%