INDEX

Explanations

copyright, licensed, export

This neuron detects programming code syntax, especially JavaScript keywords and identifiers (e.g. import, from, let, new, module paths, and other code tokens).

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

 Colchester

-0.86

家

-0.85

verhältnisse

-0.80

_('

-0.77

未

-0.74

iment

-0.74

【

-0.73

 Enumerable

-0.71

思

-0.71

未

-0.71

POSITIVE LOGITS

loadData

0.81

かったです

0.73

游戏介绍

0.71

fabri

0.68

 alfabeto

0.68

ostructures

0.66

 Bibliote

0.65

Survival

0.65

 Шев

0.64

幸運

0.63

Activations Density 0.031%