INDEX

Explanations

struct definition followed by name

The neuron fires on numeric constants (particularly decimal numbers and their assignment operators) in source code.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 from

-2.48

 their

-2.33

who

-2.11

Its

-2.02

to

-1.63

 ihrem

-1.59

 whose

-1.58

It

-1.47

 this

-1.44

 Their

-1.41

POSITIVE LOGITS

 arbres

1.30

 inoltre

1.30

 écl

1.24

 profondeur

1.24

 forêts

1.21

 morado

1.21

 сьогодні

1.20

 dnes

1.17

 promesse

1.16

𝕋

1.16

Activations Density 0.218%