INDEX

Explanations

second list item

This neuron detects ordinal list or section markers (e.g. “First,” “Second,” “Third,” “Fourth”) in structured academic or formal documents.

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

でしたが

-1.66

to

-1.60

 così

-1.58

-1.44

 quite

-1.43

%;

-1.42

_;

-1.42

his

-1.41

ejs

-1.38

IsContent

-1.38

POSITIVE LOGITS

 phénom

2.05

1.87

不得不说

1.81

 绿色

1.59

 壁纸

1.53

 levier

1.52

 manuell

1.51

 verhind

1.50

ウィキ

1.43

的那种

1.43

Activations Density 0.010%