INDEX

Explanations

header files

This neuron fires on programming identifiers and symbols in code (e.g. module names, file names, macros).

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

 разных

-1.31

 mismas

-1.20

 remé

-1.20

 verfügb

-1.20

 разные

-1.19

Nim

-1.19

 están

-1.19

 ökolog

-1.16

γει

-1.16

オリーブ

-1.15

POSITIVE LOGITS

 such

1.61

of

1.60

蔟

1.54

}\

1.51

1.45

";

1.44

his

1.36

 into

1.32

 anadolu

1.30

dling

1.30

Activations Density 0.010%