INDEX

Explanations

type names and concepts

The main thing this neuron does is detect domain-specific technical jargon and terminology.

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

🎐

0.81

opuerto

0.73

॓

0.71

ᆨ

0.71

 registrer

0.69

bcc

0.69

несе

0.68

xls

0.68

靂

0.68

තුව

0.68

POSITIVE LOGITS

或是

0.70

 hỗn

0.63

 deven

0.63

 Takeuchi

0.62

 Travers

0.62

า

0.62

 ਜਾਂ

0.61

 Ча

0.61

 Shabb

0.61

 혹은

0.60

Activations Density 2.474%