INDEX

Explanations

struct definitions and usage

The neuron fires on programming‐language code tokens—identifiers, keywords, data types, and symbols typically found 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

 betekent

-1.58

 sarung

-1.57

 schrijven

-1.41

 they

-1.38

 relojes

-1.36

ḧ

-1.34

 ícones

-1.34

резать

-1.33

 renang

-1.32

 глю

-1.29

POSITIVE LOGITS

 other

1.90

 with

1.49

 anderen

1.42

しちゃ

1.30

อื่น

1.29

 andere

1.24

the

1.22

 otras

1.20

 while

1.19

has

1.12

Activations Density 0.111%