INDEX

Explanations

open-source with a license

The neuron activates on numeric tokens and figures—i.e. it detects numbers or numerical expressions in the text.

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

 Betracht

0.84

 Мини

0.78

 Curso

0.73

 Пла

0.73

 Diskussion

0.73

 バッグ

0.71

 仕事

0.71

 envis

0.71

 Probleme

0.70

 ใจ

0.70

POSITIVE LOGITS

 undisclosed

0.86

 securities

0.82

 receivables

0.77

 contract

0.76

 lien

0.75

 liens

0.74

 unspecified

0.73

 forfeit

0.73

 escrow

0.73

 forfeiture

0.72

Activations Density 0.078%