INDEX

Explanations

sets of items or structures

The neuron fires on references to numbered or quantified groupings of items—e.g. “six-pack,” “two-book set,” “5-tuple,” “five-disk set,” 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

 Erdoğan

-1.47

-1.43

槩

-1.42

 estadounid

-1.41

October

-1.37

ورا

-1.33

ι

-1.32

łady

-1.27

score

-1.26

 pihak

-1.26

POSITIVE LOGITS

of

2.77

 когда

1.46

 Miß

1.41

 Mereka

1.35

operazione

1.34

وردار

1.34

 gagal

1.31

下次

1.31

not

1.30

 otras

1.30

Activations Density 0.064%