The better, faster and cheaper way to use and fine tune #LLMs on your own data (with Ludwig and @predibase
!). Don't miss it: https://
my.demio.com/ref/CGdgLeAiL3
7QjX2g?utm_source=LinkedIn
…
OPEN SOURCE
-
Better, Faster, Cheaper LLM Fine-tuning with Ludwig and Predibase
By
–
-
Stability AI Releases 7B Parameter Fine-Tuned Chat Model
By
–
Try out the 7 billion parameter fine-tuned chat model (for research purposes) → https://
huggingface.co/spaces/stabili
tyai/stablelm-tuned-alpha-chat
… More information on accessing and using our models from our GitHub repository → -
Stability AI Launches StableLM Large Language Models Suite
By
–
Announcing StableLM We’re releasing the first of our large language models, starting with 3B and 7B param models, with 15-65B to follow. Our LLMs are released under CC BY-SA license. We’re also releasing RLHF-tuned models for research use. Read more→ https://
stability.ai/blog/stability
-ai-launches-the-first-of-its-stablelm-suite-of-language-models
… -
NYU and Hugging Face collaborate on open source AI standards
By
–
Yes indeed @NYUDataScience @huggingface collaboration with open source models (that may also be customised) is the way forwards. We should be working together on best case principle based approaches for data and overall standards for AI.
-
Huggingface.js Library Receives Major Improvements and Community Contributions
By
–
Check out Huggingface.js — really nice recent improvements from the team AND awesome external contributors https://
t.co/fRQGzi7Yfx -
UL2 Pretraining Code Found in Megatron-LM Repository
By
–
day 20+ of exploring the wilderness: i found ul2 pretraining code in megatron LM. https://
github.com/NVIDIA/Megatro
n-LM/pull/268
… a nice surprise. -
Google AI Releases Improved MC4 Corpus and uMT5 Models
By
–
Sharing a piece of work I contributed to while at @GoogleAI
: * a new improved Mc4 corpus (29T char tokens and 107 languages) that gets language sampling right with UniMax sampling. * open source pretrained uMT5 models trained on 1T tokens. * Unimax sampling solves some -
MLflow 2.3 Releases New LLM Management and Deployment Features
By
–
Introducing #MLflow 2.3! The latest release makes it easier than ever to manage & deploy #LLMs (including Dolly 2.0!) and is packed with innovation including 3 new model flavors: HuggingFace Transformers, OpenAI functions & LangChain. See what’s new! https://
bit.ly/40eO4fP -
Microsoft SPTAG: Distributed Vector Search Library for Large Scale
By
–
Bing Vector Search GitHub – microsoft/SPTAG: A distributed approximate nearest neighborhood search (ANN) library which provides a high quality vector index build, search and distributed online serving toolkits for large scale vector search scenario. https://
bit.ly/3LbJef8 -
Should I Open-Source JailbreakChat.com Code?
By
–
should I open-source the http://
jailbreakchat.com code? don't have a lot of time atm to work on improving it but feel like there are a lot of features that could be added to the site…