INDEX

Explanations

object property/method access

The neuron detects numeric literals (numbers) 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

on

-1.55

 after

-1.48

 with

-1.43

 which

-1.42

 while

-1.41

and

-1.35

 before

-1.32

 that

-1.32

any

-1.30

 just

-1.28

POSITIVE LOGITS

particularly

1.45

 perioden

1.39

primarily

1.30

mediated

1.30

 ASSOCIATED

1.30

歺

1.30

quite

1.30

櫺

1.27

çou

1.27

mainly

1.27

Activations Density 0.036%