INDEX

Explanations

question indicator Q

The neuron responds to numeric literals (especially floating-point numbers) in source 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

馓

-2.00

 และ

-1.84

 gebruiken

-1.71

眛

-1.71

︲

-1.70

particularly

-1.70

嫆

-1.68

 cauſe

-1.67

 historische

-1.66

睃

-1.64

POSITIVE LOGITS

 with

1.56

</b>

1.44

 getting

1.41

){

1.40

1.37

 although

1.33

 even

1.32

 just

1.29

but

1.25

1.24

Activations Density 0.000%