INDEX

Explanations

great-grand family relationships

This neuron fires on tokens involved in extended family relationship terms (e.g. “great,” “grand,” “uncle,” “son,” etc.), i.e. mentions of genealogical relations.

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

our

-1.54

or

-1.48

 creates

-1.44

 that

-1.38

 helps

-1.37

 enables

-1.37

 begins

-1.34

 which

-1.34

if

-1.34

and

-1.33

POSITIVE LOGITS

several

1.40

 camiset

1.36

 cuatro

1.31

 pimenta

1.30

tifications

1.27



1.27

 seksi

1.26

 และ

1.23

 новое

1.23

觚

1.22

Activations Density 0.001%