Collama: a quick implementation of the idea of Wikipedia-style crowd-sourced fine-tuning for LLMs.
OPEN SOURCE
-
Open Science Exchange Accelerates AI Progress and Innovation
By
–
In fact, the very reason we've seen such a fast progress in in AI in the last decade is *precisely* because of the quick and open exchange of papers, data, and code. Making the process less open would inevitably lead to a slow down of progress, which would give an opportunity
-
LoRA Exchange LoRAX Serve Hundreds Fine-Tuned LLMs Cost
By
–
Link to the blog: https://
predibase.com/blog/lora-exch
ange-lorax-serve-100s-of-fine-tuned-llms-for-the-cost-of-one
… -
PyTorch Satellite Cloud Generator Tool Released
By
–
strath-ai/SatelliteCloudGenerator: A PyTorch-based tool to generate clouds for satellite images. https://
bit.ly/3PKsXAi
#AI #MachineLearning #DeepLearning #LLMs #DataScience -
China’s AI Development Strategy Without Open Source Dependence
By
–
Repeat after me: AI is not a weapon. With or without open source AI, China is not far behind. Like many other countries, they want control over AI technology and must develop their own homegrown stack.
-
Float8 Training Implementation with Stateful Tensor Subclass
By
–
my bad, didn't realize it was not public yet. We'll work on publishing it asap. We basically have a stateful Tensor subclass and a Float8Linear layer that work together to get to working Float8 training.
-
Getting Started with RAG: Chroma and Mistral AI Guide
By
–
Retrieval augmented generation is all the rage, but getting started can be overwhelming. To help you get started and navigate all of the moving pieces, we wrote a blog post about how to do RAG with @trychroma and @MistralAI
's 7B-instruct. -
Jonathan McKinney to Present LangChain and Open Source LLMs at H2O Conference
By
–
Jonathan McKinney(
@pseudotensor
) http://
H2O.ai's Director of Research, will delve deep into Mastering #LangChain and Agents with Open Source #LLMs using h2oGPT at the upcoming #H2OGenAIWorld Conference Register for the event here: https://
h2o.ai/events/generat
ive-ai-summit/
… -
LLaMA-13B Model Classifies Text-to-Image Prompts for Safety
By
–
To go alongside sdxl-emoji I also trained a llama-13b fine-tune to classify text to image prompts as safe (0) or toxic (10). That model is also approaching 1M runs: https://
replicate.com/fofr/prompt-cl
assifier
… Trained on all the toxic prompts people tried after hitting #1 on Hacker News. -
Fine-Tuning MusicGen: Create Custom Music AI Models
By
–
How to fine tune MusicGen, any style you want, any duration you want. It’s lots of fun.