INDEX

Explanations

conjunctions and separators

The neuron activates on numeric tokens (digits or numbers, including integers and decimals).

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

-2.17

 then

-1.98

his

-1.76

its

-1.62

our

-1.61

쨩

-1.61

⢅

-1.60

xvii

-1.59

𝑝

-1.58

POSITIVE LOGITS

,$_

1.44

larından

1.41

 plethora

1.38

سيم

1.38

 была

1.32

 there

1.30

1.29

comen

1.27

imbawa

1.27

全部

1.27

Activations Density 0.203%