They use Llama-2 as a base model, so that should be okay. I am not a lawyer and can't give advice on the dataset aspect but here is a table from the paper that might be helpful
LLMS
-
Web2 Lessons for Generative AI: Avoiding Past Mistakes
By
–
Since the generative AI boom began, I’ve been asking experts what the evolution of Web2 can teach us for the LLM era. How can companies avoid past mistakes and mitigate unseen risks? My latest story begins to scratch the surface of that question.
-
Simon Willison’s Talk on LLM Developments and Challenges
By
–
Such a great talk "Catching up on the weird world of LLMs" by Simon Willison(
@simonw
). The talk is the "last few years of LLMs developments compressed in just 35 minutes". It covers a lot of things such as what LLMs are, use-cases, how they’re trained and challenges involved in -
Theory Behind Q, K, V Matrices in Language Models
By
–
Not sure whether there is any theory behind it vs just empirical observation that it works well. Maybe an intuition is that the Q, K, V matrices work well for language in general, and you don't want to screw them up. Whereas the other ones are more like the extraction params.
-
AI Assistants Surpassing Experts and Multi-Agent Collaboration Predicted in 2015
By
–
Nous interagirons avec des assistants intelligents meilleurs que tous les experts sur tous les sujets. Puis ces assistants intelligents discuteront les uns avec les autres. C'est exactement ce que j'expliquais dans cette conférence en 2015 https://
youtu.be/K5a1uthRHf8 -
Generative Models and Diffusion Models in Chemistry
By
–
Exploring the Promise of Generative Models in Chemistry: An Introduction to Diffusion Models https://
bit.ly/3Yt08es #AI #MachineLearning #DeepLearning #LLMs #DataScience -
Fallbacks Now Work for All Chains Implementation
By
–
Fallbacks also works for all chains! Gist showing that here: https://
gist.github.com/hwchase17/26ec
a9f20031c349bd72fdc669c05a1f
… -
LLM Development Methods: Prompting, Fine-tuning, Few-shot
By
–
In addition to prompting LLMs, many developers are now also experimenting with fine-tuning. I describe in The Batch how to choose from the growing menu of options for building applications with LLMs: Prompting, few-shot, fine-tuning, pre-training.
-
Predibase and LangChain Partnership for Fine-tuning Open-Source LLMs
By
–
Predibase + @langchain
: The easiest way to #finetune and productionize open-source #LLMs! Our partnership brings production-grade workflows and managed infrastructure to teams that want to build on state-of-the-art open-source #models. Learn more: https://
pbase.ai/3qxmAXd -
RAGAS Framework for RAG Pipeline Evaluation Webinar
By
–
LangChain "RAG Evaluation" Webinar RAGAS is an open-source evaluation framework for your Retrieval Augmented Generation (RAG) pipelines I'm VERY excited to be doing a webinar with them next week! RAGAS Repo: https://
github.com/explodinggradi
ents/ragas
… Webinar: