INDEX

Explanations

code syntax snippets

The neuron is specialized to detect numeric literals (numbers or number-like tokens) in code.

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

gründe

-1.42

 первым

-1.27

Friedrich

-1.24

 использовании

-1.23

 передает

-1.18

 отвечает

-1.16

ulex

-1.15

 hoppas

-1.14

-1.13

POSITIVE LOGITS

 instead

1.42

幸いです

1.35

 couldn

1.32

 part

1.30

');

1.28

 tanpa

1.27

 ícones

1.26

глежда

1.25

 beyond

1.25

 among

1.23

Activations Density 0.081%