Gemma: Open Models Based on Gemini Research and Technology
Gemma: Open Models Based on Gemini Research and Technology
This work introduces Gemma, a family of lightweight, state-of-the art open models built from the research and technology used to create Gemini models. Gemma models demonstrate strong performance across academic benchmarks for language understanding, reasoning, and safety. We release two sizes of models (2 billion and 7 billion parameters), and provide both pretrained and fine-tuned checkpoints. Gemma outperforms similarly sized open models on 11 out of 18 text-based tasks, and we present comprehensive evaluations of safety and responsibility aspects of the models, alongside a detailed description of model development. We believe the responsible release of LLMs is critical for improving the safety of frontier models, and for enabling the next wave of LLM innovations.
Demis Hassabis、Tom Hennigan、James Keeling、Evan Senter、Justin Chiu、Cl¨|ment Crepy、Ivan Grishchenko、Shree Pandya、Jacob Austin、Ross McIlroy、Rohan Anil、Zhitao Gong、Daniel Cer、Bobak Shahriari、Nithum Thain、Lars Lowe Sjoesund、Noah Fiedel、Katie Millican、Ramona Comanescu、Oscar Chang、Juliette Love、Justin Mao-Jones、Elena Buchatskaya、Alex Botev、L¨|onard Hussenot、Alek Andreev、Reena Jana、Zafarali Ahmed、Koray Kavukcuoglu、Lucas Dixon、Anna Bulanova、Am¨|lie H¨|liou、Daphne Ippolito、Kathy Yu、Charline Le Lan、Kathleen Kenealy、Rahma Chaabouni、Henryk Michalewski、Joelle Barral、Lisa Lee、Christopher A. Choquette-Choo、Ted Klimenko、Johan Ferret、Eric Ni、Tris Warkentin、George Tucker、Sertan Girgin、Ambrose Slone、Antonia Paterson、Geng Yan、Pouya Tafti、Siamak Shakeri、George-Christian Muraru、Armand Joulin、Vlad Feinberg、Jeremy Chen、Jeff Stanway、Adam Roberts、Ludovic Peran、Eli Collins、Andrea Tacchetti、Eric Noland、Cassidy Hardin、Katherine Lee、Zoubin Ghahramani、Cl¨|ment Farabet、Paige Bailey、Ryan Mullins、Sebastian Borgeaud、Alex Castro-Ros、Gemma Team、Jane Labanowski、Wojciech Stokowiec、Mihir Sanjay Kale、Robert Dadashi、Maciej Miku?a、Beth Tsai、Fernando Pereira、Jean-Baptiste Lespiau、Aakanksha Chowdhery、Nikolai Chinaev、Michael Sharman、Pier Giuseppe Sessa、Laurent Sifre、Jenny Brennan、Petko Yotov、Morgane Rivi¨¨re、Oscar Wahltinez、Sholto Douglas、Oriol Vinyals、Douglas Eck、Jeff Dean、David Reid、Shreya Pathak、Aditya Barua、Machel Reid、Minh Giang、Ruibo Liu、Mateo Wirth、Thomas Mesnard、Olivier Bachem、Soham De、Yu-hui Chen、Paul Michel、Samuel L Smith、Surya Bhupatiraju、Grigory Rozhdestvenskiy、Ian Tenney
计算技术、计算机技术
Demis Hassabis,Tom Hennigan,James Keeling,Evan Senter,Justin Chiu,Cl¨|ment Crepy,Ivan Grishchenko,Shree Pandya,Jacob Austin,Ross McIlroy,Rohan Anil,Zhitao Gong,Daniel Cer,Bobak Shahriari,Nithum Thain,Lars Lowe Sjoesund,Noah Fiedel,Katie Millican,Ramona Comanescu,Oscar Chang,Juliette Love,Justin Mao-Jones,Elena Buchatskaya,Alex Botev,L¨|onard Hussenot,Alek Andreev,Reena Jana,Zafarali Ahmed,Koray Kavukcuoglu,Lucas Dixon,Anna Bulanova,Am¨|lie H¨|liou,Daphne Ippolito,Kathy Yu,Charline Le Lan,Kathleen Kenealy,Rahma Chaabouni,Henryk Michalewski,Joelle Barral,Lisa Lee,Christopher A. Choquette-Choo,Ted Klimenko,Johan Ferret,Eric Ni,Tris Warkentin,George Tucker,Sertan Girgin,Ambrose Slone,Antonia Paterson,Geng Yan,Pouya Tafti,Siamak Shakeri,George-Christian Muraru,Armand Joulin,Vlad Feinberg,Jeremy Chen,Jeff Stanway,Adam Roberts,Ludovic Peran,Eli Collins,Andrea Tacchetti,Eric Noland,Cassidy Hardin,Katherine Lee,Zoubin Ghahramani,Cl¨|ment Farabet,Paige Bailey,Ryan Mullins,Sebastian Borgeaud,Alex Castro-Ros,Gemma Team,Jane Labanowski,Wojciech Stokowiec,Mihir Sanjay Kale,Robert Dadashi,Maciej Miku?a,Beth Tsai,Fernando Pereira,Jean-Baptiste Lespiau,Aakanksha Chowdhery,Nikolai Chinaev,Michael Sharman,Pier Giuseppe Sessa,Laurent Sifre,Jenny Brennan,Petko Yotov,Morgane Rivi¨¨re,Oscar Wahltinez,Sholto Douglas,Oriol Vinyals,Douglas Eck,Jeff Dean,David Reid,Shreya Pathak,Aditya Barua,Machel Reid,Minh Giang,Ruibo Liu,Mateo Wirth,Thomas Mesnard,Olivier Bachem,Soham De,Yu-hui Chen,Paul Michel,Samuel L Smith,Surya Bhupatiraju,Grigory Rozhdestvenskiy,Ian Tenney.Gemma: Open Models Based on Gemini Research and Technology[EB/OL].(2024-03-13)[2025-08-16].https://arxiv.org/abs/2403.08295.点此复制
评论