Training Large Language Models with Interpreter Feedback using WebAssembly
A fast, local, and secure approach to training LLMs for code with WebAssembly and interpreter-based rewards
Tuned, by Axolotl
Open Source AI Development
Subscribe:.rss.atom.json.md.m3u.pls
Dormant Last read · last published · next check
Read 5 days ago and current, but nothing has been published for 17 months.
Latest episodes
Enabling Long Context Training with Sequence Parallelism in Axolotl (v0.8.0)
Training large language models (LLMs) with long contexts has become an important capability as models continue to expand in both size and context length.
Accelerating LoRA Fine-tuning with Custom CUDA Kernels and Autograd Functions
We're excited to announce a performance improvement to Axolotl's LoRA and QLoRA fine-tuning capabilities.
Fine-tuning LLMs as a Cost Strategy
Tips from experts since GPUs and inference can be pricey
Supported LLM Models with Axolotl
Quick breakdown of compatibility with Axolotl v0.8.0
Axolotl Update v0.7.0 - with GRPO
New fixes and updates to parameters since December 2024
Process Reward Models
Training Process Reward Models in axolotl
Fine Tuning Llama 3.1 405B with Axolotl on a Lambda 1-Click Cluster
Personalizing SOTA Open Source AI
Welcome to Tuned, by Axolotl
Bringing you the latest news and techniques in AI Infrastructure