INDEX

Explanations

code generation and explanation

The neuron fires on floating‐point numbers (decimal numerals) in the text.

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

्ड

0.85

َ

0.84

 செறிவு

0.79

 miguel

0.78

 heps

0.76

ﻴ

0.74

 appendages

0.73

 cardia

0.71

 Specimens

0.71

iril

0.70

POSITIVE LOGITS

gaan

0.91

to

0.89

ar

0.87

ain

0.84

transform

0.80

mapbox

0.80

ars

0.79

show

0.78

shader

0.78

ECUTIVE

0.78

Activations Density 0.000%