AnswerDotAI/fsdp_qlora
A training framework enabling fine-tuning of large language models using quantized LoRA and fully-sharded data parallelism.
Not currently ranked — collecting fresh signals.
star history
This repository provides scripts and integration for training large language models with QLoRA (Quantized Low-Rank Adaptation) combined with FSDP (Fully Sharded Data Parallel). It enables efficient fine-tuning of models like Llama-2 70B on limited GPU memory by combining 4-bit quantization with gradient checkpointing and CPU offloading. The project integrates with Axolotl training framework and uses libraries like bitsandbytes and Hugging Face transformers.
Frequently asked
- What is AnswerDotAI/fsdp_qlora?
- A training framework enabling fine-tuning of large language models using quantized LoRA and fully-sharded data parallelism.
- Is fsdp_qlora open source?
- Yes — AnswerDotAI/fsdp_qlora is open source, released under the Apache-2.0 license.
- What language is fsdp_qlora written in?
- AnswerDotAI/fsdp_qlora is primarily written in Jupyter Notebook.
- How popular is fsdp_qlora?
- AnswerDotAI/fsdp_qlora has 1.5k stars on GitHub.
- Where can I find fsdp_qlora?
- AnswerDotAI/fsdp_qlora is on GitHub at https://github.com/AnswerDotAI/fsdp_qlora.