INDEX

Explanations

code syntax elements

The neuron is specialized for spotting numeric literals (constants) in code snippets.

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

if

-2.02

}$

-1.98

ie

-1.84

});

-1.74

-1.73

-1.70

las

-1.66

».

-1.66

ak

-1.65

POSITIVE LOGITS

its

1.88

 that

1.81

 aliviar

1.77

 Präsidenten

1.71

 sobrevivir

1.66

تاح

1.60

cesta

1.58

 allowed

1.56

 denominado

1.55

 llorar

1.52

Activations Density 0.004%