INDEX

Explanations

string delimiters

The neuron fires on quoted string literals in code, i.e. sequences of characters enclosed in quotes.

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

 новых

-1.51

それを

-1.47

 nieuwe

-1.45

autres

-1.45

AN

-1.43

 amelyek

-1.37

etsy

-1.34

be

-1.33

of

-1.32

 различ

-1.29

POSITIVE LOGITS

 because

1.49

reurs

1.48

 cerdas

1.48

<<<<

1.45

塒

1.41

 malang

1.41

aterdag

1.38

店長

1.38

erà

1.35

 fixa

1.35

Activations Density 0.035%