INDEX

Explanations

quantifiable measures and technical terms

The neuron detects numeric quantities (digits and numbers) appearing 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

😥

-1.12

drm

-1.12

 tats

-1.08

 Pire

-1.05

戾

-1.02

 madame

-1.02

😓

-1.00

🤦

-0.99

おお

-0.98

via

-0.98

POSITIVE LOGITS

for

1.38

ഥ

1.14

 assist

1.05

 solve

1.04

dańsk

1.02

 spô

0.99

authier

0.99

我只

0.98

now

0.97

GetAxis

0.97

Activations Density 0.156%