INDEX

Explanations

numbered lists (e.g., 8. 9. 10. 11. 12.)

The neuron fires on numeric tokens—years, list/item numbers, and other multi‐digit numbers in the text.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

하는

1.91

ت

1.86

या

1.74

다

1.73

ι

1.68

س

1.66

تان

1.63

क

1.60

я

1.59

ní

1.57

POSITIVE LOGITS

மார்க்

1.31

 нашу

1.30

FBSDKGraph

1.26

 sedative

1.24

ёв

1.24

 мою

1.20

 ندا

1.19

뜻

1.18

 ƯCLN

1.16

 spôsob

1.16

Activations Density 5.124%