INDEX

Explanations

specific words and what follows

The neuron detects decimal numeric values (floating-point numbers) 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

 again

-1.32

 with

-1.27

ränken

-1.22

denim

-1.20

colata

-1.20

pergillus

-1.19

 опять

-1.18

 кроссовки

-1.17

Seine

-1.16

гию

-1.16

POSITIVE LOGITS

 úgy

1.26

 their

1.24

 then

1.21

 zwarte

1.20

®:

1.19

 various

1.16

 ihres

1.16

 eivät

1.14

™

1.07

ścic

1.06

Activations Density 0.337%