In Unsloth? I don't know, I'm not part of the dev team. You can still fine-tune with TRL or Axolotl.
OPEN SOURCE
-
FineLlama-3.1-8B Model Deployment Issues on Hugging Face
By
–
I've also made a Hugging Face Space to chat with the FineLlama-3.1-8B model we build in the article.
— Maxime Labonne @ ICLR (@maximelabonne) 29 juillet 2024
The model looks good but the space tends to crash after one answer. I'm interested if you know how to fix it!
🤗 Space: https://t.co/6bDNLHShqh pic.twitter.com/PLjzajao0xI've also made a Hugging Face Space to chat with the FineLlama-3.1-8B model we build in the article. The model looks good but the space tends to crash after one answer. I'm interested if you know how to fix it! Space: https://
huggingface.co/spaces/mlabonn
e/FineLlama-3.1-8B
… -
FineTome-100k: High-Quality Filtered Dataset Release
By
–
I'm also releasing FineTome-100k, a refiltered subset of arcee-ai/The-Tome using fineweb-edu-classifier. It's super high-quality data, perfect when you need extra samples to concatenate with a dataset. FineTome: https://
huggingface.co/datasets/mlabo
nne/FineTome-100k
… -
Fine-tune Llama 3.1 Ultra-Efficiently with Unsloth
By
–
Fine-tune Llama 3.1 Ultra-Efficiently with @UnslothAI New comprehensive guide about supervised fine-tuning on @huggingface
. Over the last year, I've done a lot of fine-tuning and blogging. This guide brings it all together. Article: https://
huggingface.co/blog/mlabonne/
sft-llama3
… -
PEFT v0.12.0 Release: OLoRA, X-LoRA, FourierFT Methods
By
–
in case you missed it last week: peft v0.12.0 just dropped With some cool new param-efficient methods like OLoRA, X-LoRA, FourierFT, and more
-
Microsoft Generative AI for Beginners: 18 Lessons Learning Guide
By
–
GitHub – microsoft/generative-ai-for-beginners: 18 Lessons, Get Started Building with Generative AI https://
bit.ly/461joDf https://
bit.ly/4fbeyb5
#AI #MachineLearning #DeepLearning #LLMs #DataScience -
MINT-1T Open-Sources 1 Trillion Token Multimodal Dataset
By
–
8/ MINT-1T – open-sources a large-scale multimodal interleaved dataset consisting of 1 trillion tokens which has 3.4 billion images; it also includes new sources such as PDFs and ArXiv papers.
-
OpenDevin Platform Enables Generalist Software Agents
By
–
4/ OpenDevin – presents a platform to develop generalist agents that interact with the world through software; features include 1) an interaction mechanism for interaction between agents, interfaces, and environments, 2) an environment including a sandboxed operating system and
-
Flowise Sequential Agents Documentation Released
By
–
Sunday reading Docs for Sequential Agents is now ready! https://
docs.flowiseai.com/using-flowise/
agentflows/sequential-agents
… Huge thanks to @toi500 for helping with this And @akirakudo911 for the intro guide in Japanese! https://
docs.flowiseai.com/getting-starte
d#community-guide
… -
Model Quantization Commands and Optimization Techniques
By
–
Thanks @bartowski1182
! Is it possible for you to provide the commands you used to quantize the models? I can't keep up with this pace