HPC & Quantum
27 subscribers
11.4K photos
668 videos
3 files
30.6K links
Download Telegram
HPC Guru (Twitter)

๐Ÿ‘‡๐Ÿงต by @CerebrasSystems on #SlimPajama โ€“ the largest deduplicated, multi-corpora, open-source, dataset for training LLMs

Cute mascot!

Cerebras built #SlimPajama to train a #LLaMA style model for partner @Opentensor

https://www.cerebras.net/blog/slimpajama-a-627b-token-cleaned-and-deduplicated-version-of-redpajama

#AI #GenerativeAI
-----------
@CerebrasSystems:
๐Ÿ“ฃ New dataset drop!
Introducing SlimPajama-627B: the largest extensively deduplicated, multi-corpora, open-source dataset for training large language models. ๐Ÿงตhttps://t.co/bwsSz4d9hs https://t.co/3Ow1OAEcS9