INDEX

Explanations

SVG/XML attributes

This neuron fires on numeric literal tokens (especially floating-point numbers) in code or markup.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 maniere

-1.58

 veľ

-1.55

 tehn

-1.52

%\

-1.48

 ľudí

-1.46

 però

-1.45

 poft

-1.45

៧

-1.44

 fournir

-1.39

 specif

-1.37

POSITIVE LOGITS

on

1.72

 Since

1.60

 These

1.59

we

1.52

In

1.44

 Because

1.43

1.42

or

1.41

by

1.36

 first

1.34

Activations Density 0.030%