INDEX

Explanations

materials and their applications

The neuron detects technical material or composite names (e.g. specific nanoparticles, hydrogels, layered composites) typically found in scientific paper titles.

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

すっ

-1.19

-1.18

事も

-1.12

ことも

-1.11

 either

-1.08

のだろう

-1.05

ようになります

-1.02

 beweisen

-1.00

ゎ

-0.99

ほんと

-0.98

POSITIVE LOGITS

for

2.05

 with

1.80

สำหรับ

1.45

PointerException

1.27

ksessa

1.23

 محفوظة

1.20

riteria

1.16

idable

1.16

in

1.16

 juventud

1.16

Activations Density 0.236%