INDEX

Explanations

code elements and definitions

The neuron fires strongly on tokens that begin a new line (i.e. line-initial tokens).

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



-1.58

his

-1.50

 them

-1.46

it

-1.44

部分を

-1.43

 celles

-1.40

nitř

-1.33

它

-1.31

 inspirasi

-1.31

ţ

-1.28

POSITIVE LOGITS

の

1.49

1.48

」

1.37

 как

1.37

now

1.34

的

1.31

 wobei

1.30

\}=\

1.29

}$.

1.24

 fotografico

1.24

Activations Density 0.286%