INDEX

Explanations

questions and specific phrases

This neuron activates on numeric literals (sequences of digits, including signed integers) in the text.

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

<bos>

-1.66

谟

-1.41

Zubereitung

-1.40

 prodej

-1.38

tembro

-1.37

utuhan

-1.36

ってみました

-1.35

felizmente

-1.35

dây

-1.34

ればいい

-1.31

POSITIVE LOGITS

1.57

1.40

the

1.30

1.23

 erläut

1.22

“

1.20

1.19

<b>

1.18

/***

1.17

Activations Density 0.000%