INDEX

Explanations

layout proofs proofs layouts

The main thing this neuron does is detect numeric sequences and number‐based identifiers (e.g. page numbers, section numbers, file or figure numbers).

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

⠁

-0.75

想像

-0.72

霖

-0.72

дан

-0.70

And

-0.70

 peraturan

-0.69

čnosti

-0.69

 suspens

-0.69

ässä

-0.69

داشت

-0.68

POSITIVE LOGITS

 layout

1.41

Layout

1.26

 Layout

1.09

 layouts

0.96

 proofs

0.95

layout

0.91

 LAYOUT

0.90

Final

0.89

 izman

0.88

饬

0.88

Activations Density 0.025%