INDEX

Explanations

0 followed by a delimiter

The neuron activates on numeric literal tokens (numbers) 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

 میں

-1.40

-1.37

<i>

-1.34

an

-1.31

 wana

-1.30

家乡

-1.30

 שלהם

-1.27

ſchen

-1.27

私は

-1.26

他们在

-1.25

POSITIVE LOGITS

1.80

 that

1.52

☫

1.48

 inmueble

1.45

 minimise

1.43

 Tatsache

1.37

kenal

1.36

this

1.36

 question

1.35

 method

1.34

Activations Density 0.013%