INDEX

Explanations

piece of code

This neuron fires on singular-count nouns that often occur after “piece of,” i.e. common English countable items (legislation, data, apparel, audio, code, hardware, 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

of

-1.41

zação

-1.40

 bedste

-1.34

 Catedral

-1.28

 作り方

-1.27

contrast

-1.27

Buyer

-1.25

zystać

-1.23

水上

-1.21

咈

-1.21

POSITIVE LOGITS

 that

2.56

the

1.69

 Когда

1.55

 όταν

1.54

 this

1.51

;-)

1.48

能夠

1.46

 which

1.45

 другими

1.43

もう一度

1.42

Activations Density 0.046%