INDEX

Explanations

politics, Israel, path

The neuron fires on numeric tokens (digits or numbers) 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

 then

-1.73

 these

-1.64

so

-1.48

 even

-1.38

 first

-1.30

𝗺

-1.28

 before

-1.27

 using

-1.27

 some

-1.20

此事

-1.20

POSITIVE LOGITS

 reprezent

1.33

雠

1.16

可爱

1.16

跻

1.13

 già

1.13

 already

1.13

ématique

1.13

🚐

1.13

ينج

1.12

🧫

1.11

Activations Density 0.009%