INDEX

Explanations

dictionary etymology

This neuron fires on subword fragments (common prefixes, stems, and suffixes) rather than whole words, detecting morpheme pieces like “sne‐,” “aked,” “im-,” “effic-,” etc.

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

 this

-1.63

がこの

-1.62

同じく

-1.50

 nguyen

-1.32

 armazen

-1.32

 ocorrer

-1.30

 solcher

-1.29

are

-1.27

 these

-1.27

 then

-1.27

POSITIVE LOGITS

Ausstattung

1.42

湿

1.30

 चीज़

1.27

//{

1.23

КТ

1.23

%;

1.23

Ṭ

1.21

Ḥ

1.21

ccccccc

1.16

 растения

1.16

Activations Density 0.062%