INDEX

Explanations

code snippets and configuration

The neuron detects numeric literal tokens (sequences of digits) in code or 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

Another

-1.59

 another

-1.42

 manquante

-1.32

nother

-1.30

฿

-1.29

-1.27

 condicionado

-1.23

Portale

-1.22



-1.21

%;

-1.20

POSITIVE LOGITS

そして

1.46

etc

1.32

それが

1.23

などが

1.16

<bos>

1.13

があり

1.11

時には

1.09

以及

1.08

 einigen

1.05

が含ま

1.03

Activations Density 0.028%