INDEX

Explanations

the [word]

The neuron fires on numeric tokens (integers or decimals), i.e. explicit numbers in the text.

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

-1.43

na

-1.34

aaa

-1.31

ÑO

-1.28

</h5>

-1.28

on

-1.27

ka

-1.26

我和

-1.24

の前

-1.23

 bezw

-1.23

POSITIVE LOGITS

 Surprisingly

1.55

𖣘

1.54

有名な

1.53

 utaf

1.46

 other

1.43

six

1.41

их

1.41

 Перейти

1.39

 three

1.38

 noteworthy

1.38

Activations Density 0.048%