INDEX

Explanations

code/module names

This neuron activates on identifiers that are part of module or package import paths (e.g. library/module names in import/require statements).

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

who

-1.16

 named

-1.13

 joined

-1.12

\'

-1.09

也

-1.09

 hadapan

-1.08

 oznacza

-1.07

 Name

-1.05

 told

-1.05

POSITIVE LOGITS

توضیحات

1.45

Ingredi

1.43

معرفی

1.42

Terug

1.41

Wady

1.41

Kleur

1.40

 作り方

1.36

مقدمه

1.34

Materiaal

1.32

Arbe

1.31

Activations Density 0.023%