INDEX

Explanations

links to research papers

The neuron detects numeric strings—especially decimal‐formatted numbers and identifiers like DOIs—in reference and citation metadata.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 monographs

0.61

doi

0.56

selected

0.54

))

0.53

ONG

0.52

ISBN

0.52

matic

0.52

0.51

 Mitte

0.51

 isothermal

0.50

POSITIVE LOGITS

 मजदूरों

0.65

 cuánt

0.61

ኋ

0.61

 почи

0.60

 chụp

0.60

 evitare

0.59

ákat

0.59

佰

0.59

 chasse

0.58

 yelling

0.58

Activations Density 0.039%