INDEX

Explanations

python imports

The neuron fires on import statements that reference well-known machine-learning libraries (e.g. “sklearn,” “keras,” “torch,” “torchvision”).

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

 from

-1.76

 this

-1.50

providedIn

-1.41

感じで

-1.34

なのかな

-1.33

isissez

-1.33

 casó

-1.32

 different

-1.31

はコチラ

-1.31

 creciendo

-1.30

POSITIVE LOGITS

1.45

1.38

in

1.35

讲述

1.34

そこは

1.33

もあって

1.31

それから

1.28

1.27

そこで

1.27

is

1.26

Activations Density 0.005%