INDEX

Explanations

This neuron responds to content-bearing, domain-specific nouns or proper names (i.e. relatively rare, informative terms).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 আর

0.78

 மட்டுமல்ல

0.74

𒋛

0.73

 және

0.72

чатку

0.71

এবং

0.70

 वगैरह

0.70

그리고

0.70

 multitudes

0.70

 diccionario

0.70

POSITIVE LOGITS

3.77

（

3.11

（

2.77

(=

2.59

2.58

(\

2.36

 (“

2.30

((

2.25

("

2.23

($

2.16

Activations Density 1.611%