INDEX

Explanations

say that

The neuron fires on occurrences of the verb “say” when used to preface or introduce someone’s statement or opinion.

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

 What

-1.38

を使う

-1.38

 latter

-1.34

 übrig

-1.34

is

-1.34

him

-1.32

 Here

-1.30

 naudoti

-1.28

もあり

-1.28

 said

-1.27

POSITIVE LOGITS

Ꮠ

1.41

_____________

1.38

Even

1.36

 compagnon

1.34

܆

1.30

television

1.30

 gouverneur

1.29

______________

1.27

 حتى

1.27

ݯ

1.26

Activations Density 0.018%