INDEX

Explanations

python class definitions

The neuron activates on dotted attribute accesses in code (e.g. module or class references like `tk.Frame`, `models.Model`, `torch.nn.Module`).

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

to

-1.30

be

-1.10

 significativa

-1.09

 inny

-1.07

-1.05

局面

-1.05

by

-1.02

it

-1.02

……。

-1.02

POSITIVE LOGITS

我們

1.23

Gambar

1.18

 herencia

1.16

yyr

1.15

بله

1.15

 понравился

1.13

گاهی

1.10

 yapan

1.08

گوشی

1.07

ఁ

1.07

Activations Density 0.005%