Demonstration of transformer-based ALBERT model on a 14nm analog AI inference chip

Abstract A Lite Bidirectional Encoder Representations from Transformers model is demonstrated on an analog inference chip fabricated at 14nm node with phase change memory. The 7.1 million unique analog weights shared across 12 layers are mapped to a single chip, accurately programmed into the conduc...

Celý popis

Podrobná bibliografie

Hlavní autoři:	An Chen, Stefano Ambrogio, Pritish Narayanan, Atsuya Okazaki, Charles Mackin, Andrea Fasoli, Malte J. Rasch, Alexander Friz, Jose Luquin, Takeo Yasuda, Masatoshi Ishii, Takuto Kanamori, Kohji Hosokawa, Timothy Philicelli, Seiji Munetoh, Vijay Narayanan, Hsinyu Tsai, Geoffrey W. Burr
Médium:	Artigo
Jazyk:	Inglês
Vydáno:	Nature Portfolio 2025-09-01
Edice:	Nature Communications
On-line přístup:	https://doi.org/10.1038/s41467-025-63794-4
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!