INDEX

Explanations

defining technical terms

The neuron activates on occurrences of domain-specific concept names or technical terms (often multi-word) that serve as the defined topic of a section.

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

of

-1.88

and

-1.55

 ecuador

-1.45

 marseille

-1.40

 gabri

-1.34

可能な

-1.33

 uruguay

-1.30

 anderson

-1.29

 santiago

-1.27

 augustus

-1.26

POSITIVE LOGITS

 those

1.88

 when

1.70

for

1.63

Karakteristik

1.63

 people

1.54

you

1.53

1.52

 sonre

1.39

but

1.38

那些

1.37

Activations Density 0.445%