INDEX

Explanations

subjects and categories

The neuron detects high-level topical section headings or category labels (e.g. “Education,” “Research,” “Music,” “Film,” 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

 their

-1.80

 your

-1.70

 both

-1.63

for

-1.52

 både

-1.50

 mengatakan

-1.49

to

-1.48

儛

-1.39

[]{"

-1.38

 Your

-1.33

POSITIVE LOGITS

1.33

Gareth

1.21

感じです

1.15

Ofic

1.09

が付いて

1.07

–

1.07

 laajenta

1.07

也非常

1.06

 desempen

1.05

 пър

1.04

Activations Density 0.137%