INDEX

Explanations

Ahead of the Bell

The main thing this neuron does is detect terminology and context related to motorsport racing events and venues.

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

 convict

-0.83

hio

-0.80

🎾

-0.80

 wymaga

-0.77

ھا

-0.74

🏀

-0.73

almente

-0.72

 гипер

-0.72

imus

-0.72

Squ

-0.71

POSITIVE LOGITS

 Speedway

1.09

 shale

1.03

 meetings

1.02

 réunion

0.90

 tapes

0.83

 mecánico

0.81

 Jawa

0.80

 Dirt

0.80

พ์

0.79

 dirt

0.78

Activations Density 0.013%