INDEX

Explanations

category or cat

The neuron fires on occurrences of code identifiers naming data models or entities—especially “Category” (and similar model names like “Author”).

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

Also

-1.69

 dialami

-1.62

-1.59

That

-1.59

They

-1.54

Everyone

-1.48

 That

-1.47

but

-1.46

También

-1.45

 what

-1.41

POSITIVE LOGITS

 eenen

1.58



1.49

 geschenk

1.45

 hanno

1.44

ются

1.44

 eenige

1.44

 gouden

1.43

 gangen

1.41

 famed

1.34

 affix

1.32

Activations Density 0.032%