INDEX

Explanations

programming code namespaces

The neuron chiefly activates on numeric literal tokens (digits and related symbols) in code.

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

🏷

-0.90

oiden

-0.84

ênero

-0.82

гмента

-0.78

 zarządz

-0.77

社長

-0.76

BigDecimal

-0.75

 Компа

-0.75

ತ

-0.74

erir

-0.74

POSITIVE LOGITS

icycle

0.89

vertido

0.85

 nombrado

0.82

oughed

0.81

として

0.81

ahead

0.80

 Knie

0.79

 선택

0.79

ramenta

0.78

在你

0.78

Activations Density 0.009%