@BigScienceLLM profile picture
BigScience Large Model Training @BigScienceLLM

Follow the training of "BLOOM 🌸", the @BigScienceW multilingual 176B parameter open-science open-access language model, a research tool for the AI community.

Joined March 2022
  • Tweets 129
  • Following 1
  • Followers 8,347
5 Photos and videos
The BLOOM model is now officially released! Read more here: bigscience.huggingface.co/bl… Find the model here: huggingface.co/bigscience/bl…

BigScience Research Workshop@BigscienceW
12 Jul 2022
BLOOM is here. The largest open-access multilingual language model ever. Read more about it or get it at bigscience.huggingface.co/bl… hf.co/bigscience/bloom
6
59
252
BigScience Large Model Training retweeted
clem πŸ€—
@ClementDelangue
14 Nov 2022
The Bloom paper is out. Looks like it's doing worse than current GPT3 API in zero-shot generation tasks in English but better than other open-source LLMs & better than all in zs multi-lingual (which was the main goal). Proud of the work from the community! arxiv.org/abs/2211.05100
12
104
590
BigScience Large Model Training retweeted
Niklas Muennighoff@Muennighoff
4 Nov 2022
Crosslingual Generalization through Multitask Finetuning 🌸 Demo: huggingface.co/bigscience/bl… πŸ“œ arxiv.org/abs/2211.01786 πŸ’»github.com/bigscience-worksh… We present BLOOMZ & mT0, a family of models w/ up to 176B params that follow human instructions in >100 languages zero-shot. 1/7
10
75
279
The super-fast inference solutions are finally here for all to use:
Stas Bekman
@StasBekman
15 Sep 2022
Learn how you can get under 1msec per token generation time with BLOOM 176B model! Not one, but multiple super-fast solutions including Deepspeed-Inference, Accelerate and Deepspeed-ZeRO! huggingface.co/blog/bloom-in…
1
2
28
BigScience Large Model Training retweeted
clem πŸ€—
@ClementDelangue
31 Aug 2022
What do @StabilityAI @EMostaque #stablediffusion & @BigscienceW Bloom - aka the coolest new models ;) - have in common? They both use a new gen of ML licenses aimed at making ML more open & inclusive while keeping it harder to do harm with them. So cool! huggingface.co/blog/open_rai…
8
35
176
BigScience Large Model Training retweeted
Hugging Face
@huggingface
14 Jul 2022
The Technology Behind BLOOM Training🌸 Discover how @BigscienceW used @MSFTResearch DeepSpeed @nvidia Megatron-LM technologies to train the World's Largest Open Multilingual Language Model (BLOOM): huggingface.co/blog/bloom-me…

The Technology Behind BLOOM Training

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

huggingface.co
8
147
601
BigScience Large Model Training retweeted
BigScience Research Workshop@BigscienceW
12 Jul 2022
BLOOM is here. The largest open-access multilingual language model ever. Read more about it or get it at bigscience.huggingface.co/bl… hf.co/bigscience/bloom
29
757
2,665
BigScience Large Model Training retweeted
Saulnier Lucile@LucileSaulnier
28 Jun 2022
🌸@BigscienceW BLOOM's intermediate checkpoints have already shown some very cool capabilities! What's great about BLOOM is that you can ask it to generate the rest of a text - and this even if it is not yet fully trained yet! πŸ‘Ά 🧡 A thread with some examples
Saulnier Lucile@LucileSaulnier
27 Jun 2022
A milestone soon to be reached πŸš€πŸ’« Can't wait to see the capabilities and performance of this long-awaited checkpoint! What about you? Have you already prepared some prompts that you want to test? ✏️
5
25
144
β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“ 102%
15
6
175
For 111 days, we've enjoyed world-class hardware stability and throughput thanks to the hard work of our friends at @Genci_fr, @INS2I_CNRS, Megatron & DeepSpeed. Having reached our objective earlier than expected, we'll keep training for a few more days. Stay tuned, more soon ;)
3
25
303
β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“ 101%
68
78
968
β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“ 100%
40
306
2,184
β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“ 99%
16
75
725
β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“ 98%
4
12
228
β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“ 97%
2
8
165
β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–‘ 96%
1
2
78
β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–‘ 95%
1
8
172
β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–‘ 94%
6
1
76
β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–‘ 92%
2
8
167
β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–“β–‘ 91%
1
59
Load more