Bookmark Saya

Semantic Scholar Open Access 2023 19293 sitasi

LLaMA: Open and Efficient Foundation Language Models

Hugo Touvron Thibaut Lavril Gautier Izacard X. Martinet M. Lachaux +9 lainnya
Lihat Sumber

Abstrak

We introduce LLaMA, a collection of foundation language models ranging from 7B to 65B parameters. We train our models on trillions of tokens, and show that it is possible to train state-of-the-art models using publicly available datasets exclusively, without resorting to proprietary and inaccessible datasets. In particular, LLaMA-13B outperforms GPT-3 (175B) on most benchmarks, and LLaMA-65B is competitive with the best models, Chinchilla-70B and PaLM-540B. We release all our models to the research community.

Topik & Kata Kunci

Computer Science

Penulis (14)

H

Hugo Touvron

T

Thibaut Lavril

G

Gautier Izacard

X

X. Martinet

M

M. Lachaux

T

Timothée Lacroix

B

Baptiste Rozière

N

Naman Goyal

E

Eric Hambro

F

Faisal Azhar

A

Aur'elien Rodriguez

A

Armand Joulin

E

Edouard Grave

G

Guillaume Lample

Format Sitasi

APA MLA BibTeX
Touvron, H., Lavril, T., Izacard, G., Martinet, X., Lachaux, M., Lacroix, T. et al. (2023). LLaMA: Open and Efficient Foundation Language Models. https://www.semanticscholar.org/paper/57e849d0de13ed5f91d086936296721d4ff75a75

Akses Cepat

PDF tidak tersedia langsung

Cek di sumber asli →
Lihat di Sumber
Informasi Jurnal
Tahun Terbit
2023
Bahasa
en
Total Sitasi
19293×
Sumber Database
Semantic Scholar
Akses
Open Access ✓