Some tips & tricks for reducing how much memory you use when loading larger LLMs & other models in PyTorch: https://
bit.ly/489XajJ v/
@rasbt
OPEN SOURCE
-
Memory optimization tips for loading large LLMs PyTorch
By
–
-
AutoTrain Advanced: Easy Local and Hub Model Fine-tuning
By
–
The easiest way to fine-tune models Locally or on Hugging Face Hub, now via Python "pip install autotrain-advanced"
-
Mistral Releases New 3B and 8B Parameter Models
By
–
We just released two small models, with 3B and 8B parameters. Ministral 3B is exceptionally strong, outperforming Llama 3 8B and our previous Mistral 7B on instruction following benchmarks. https://
mistral.ai/news/ministrau
x/
… -
Open-source model merging technique simplifies parameter tuning
By
–
Hope you'll like this technique. This makes model merging 1/ simpler by removing parameter tuning and 2/ more computationally efficient than alternatives. You can play with it today. The code is open-sourced and available on GitHub
-
Run Any GGUF Model from Hugging Face Hub with Ollama
By
–
Now you can run any GGUF model from Hugging Face Hub with Ollama
-
Differentiable Adaptive Merging: Faster Automated Model Fusion
By
–
Differentiable Adaptive Merging I participated in a paper with @arcee_ai about a new automated merging technique (no manual parameter tuning!) Compared to other algorithms, like Evolutionary Merging, it is significantly faster to run and achieves the same level of performance.
-
OSS Team Working on Fix Implementation Soon
By
–
yes – oss team is at work on this right now – expect a fix soon
-
H2O AI Partners with AI Verify Foundation for Responsible AI Testing
By
–
We’re partnering with the AI Verify Foundation to advance responsible AI adoption in Singapore! We’re contributing benchmarks and code to AI Verify’s open-source Project #Moonshot toolkit for LLM application testing – https://
h2o.ai/blog/2024/h2o-
ai-collaboration-with-ai-verify-foundation/
… -
New Model Beats Llama-405B and All Open Source Competitors
By
–
Thank you. Please note we beat every open source model, including Llama-405B.