INDEX

Explanations

code snippets and technical descriptions

This neuron fires on identifiers or literals used as branch labels in switch (or equivalent) statements—i.e. the constant that follows `case` (or the right‐hand side of an `==`/`===` check).

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

‘

-1.56

of

-1.52

-1.43

-1.38

-1.36

-1.31

-1.30

-1.27

An

-1.27

POSITIVE LOGITS

หรือ

1.63

菇

1.51

 или

1.33

蠵

1.30

 یا

1.28

aien

1.27

﹖

1.26

 火鍋

1.22

饜

1.20

 saisi

1.16

Activations Density 0.002%