INDEX

Explanations

casting or type declaration

The neuron fires on numeric literals in code snippets.

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

ensitive

-1.40

oluble

-1.38

signable

-1.34

そして

-1.26

۸

-1.24

และ

-1.24

Sabes

-1.23

蟥

-1.21

 rembourse

-1.21

ḩ

-1.20

POSITIVE LOGITS

of

1.62

 that

1.48

ようです

1.35

an

1.28

 secoli

1.27

any

1.27

 your

1.26

 made

1.24

 lavoratori

1.20

為に

1.15

Activations Density 0.001%