INDEX

Explanations

. / exists region Comple

The neuron detects numeric tokens—especially floating-point numbers (digits with decimal points)—wherever they occur.

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

ที่มา

-1.28

 rozdí

-1.27

intelligible

-1.25

鱂

-1.23

፨

-1.22

 cór

-1.20

 arquite

-1.18

tranh

-1.17

 ซึ่ง

-1.16

maining

-1.15

POSITIVE LOGITS

就可以了

1.34

就行了

1.22

 europea

1.17

 rosario

1.12

is

1.05

 romano

1.04

 cristi

1.02

才行

1.01

 emporio

1.01

0.99

Activations Density 0.000%