INDEX

Explanations

programming constructs

The neuron strongly responds to programming annotation tokens (e.g. Lombok or Java annotations like @NonNull, @Delegate).

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

及时

0.59

uma

0.58

 fists

0.56

ौरा

0.55

 timely

0.54

 offending

0.54

 tiểu

0.54

otra

0.54

hyth

0.53

digo

0.53

POSITIVE LOGITS

 superfici

0.64

lombok

0.61

 ToString

0.60

တ

0.60

 lombok

0.58

 فول

0.58

ゲ

0.57

extern

0.57

実験

0.56

 Builders

0.56

Activations Density 0.008%