INDEX

Explanations

specific proper nouns

This neuron fires on the primary subject or heading term of a new section—i.e. the key entity or topic token that begins each document segment.

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

They

-1.75

who

-1.63

 should

-1.46

 Ін

-1.34

Они

-1.34

誰が

-1.34

 debía

-1.34

öpfe

-1.32

それにしても

-1.30

 Gebäude

-1.30

POSITIVE LOGITS

its

1.48

 других

1.41

1.37

 other

1.36

他の

1.34

 aufgrund

1.32

is

1.30

其他

1.24

 ấy

1.23

Especific

1.22

Activations Density 0.179%