INDEX

Explanations

SQL table definitions

This neuron activates on programming-language syntax—especially function definitions and their type annotations (e.g. “def extract(index: Integer) = udf(…)”).

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

 without

-1.45

on

-1.37

as

-1.30

 although

-1.30

 when

-1.25

 where

-1.24

 seems

-1.23

 wants

-1.23

 Those

-1.22

 before

-1.21

POSITIVE LOGITS

 rapporter

1.39

และ

1.30

 และ

1.30

וּ

1.23

 лв

1.22

 Seeder

1.18

 бампер

1.16



1.16

难受

1.15

摸了

1.13

Activations Density 0.023%