INDEX

Explanations

variable assignments

This neuron detects programming identifiers written in mixed-case (camelCase or PascalCase).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

has

-1.41

狲

-1.41

$*$

-1.36

astă

-1.34

 since

-1.30

 currently

-1.29

 manifiesta

-1.29

 cualidades

-1.28

 entiende

-1.27

(~

-1.27

POSITIVE LOGITS

↵

1.58

 tylko

1.48

 olsa

1.37

</h2>

1.35

！");

1.28

 이미

1.20

我們的

1.17

新竹

1.16

criptors

1.14

Familien

1.13

Activations Density 0.049%