Automate fine-tuning of Large Language Models with Axolotl using LoRA/QLoRA and DPO/KTO/ORPO/GRPO.
npx skills add https://github.com/Ced3-han/Harness-Settings --skill axolotl-ced3-han
Please help me install this Agent Skill. Skill: axolotl Source: https://github.com/Ced3-han/Harness-Settings/tree/main/skills/axolotl Command: npx skills add https://github.com/Ced3-han/Harness-Settings --skill axolotl-ced3-han
This Skill provides expert guidance for fine-tuning Large Language Models (LLMs) with Axolotl, offering a comprehensive set of tools and resources for efficient model customization and enhancement.
references/ for API usage, dataset formats, and other relevant information.Use the axolotl skill to generate a custom fine-tuned model for text generation based on your requirements.
High-intent search queries and answers about installing and using this skill.
Fine-tuning large language models with LoRA and QLoRA requires configuring YAML files for your specific model. This Skill automates the training process using axolotl, torch, and peft to apply these efficient parameter tuning techniques.
Yes, you can use DPO or KTO for LLM alignment training. This Skill supports multiple preference alignment techniques including DPO, KTO, ORPO, and GRPO, allowing you to customize large language models for specific text or code generation tasks.
Yes, axolotl supports multimodal large language model training. This Skill provides comprehensive configuration management and documentation to streamline the fine-tuning process for multimodal models alongside standard text-based models.
You need DeepSpeed and accelerate for distributed LLM fine-tuning and memory optimization. These dependencies work with torch and transformers to manage training scale and accelerate the fine-tuning process across different hardware configurations.
The best way to manage configurations for 100+ models is using YAML configuration files. This Skill handles configuration management for various models, ensuring optimal settings for different use cases like question answering or code generation.
NITISH-gitbitFine-tune large language models with Axolotl for LoRA, DPO, GRPO, and multimodal training.
Ced3-hanGuide fine-tuning of large language models with LLaMA-Factory WebUI.
HarriesConnects images and text in a shared embedding space using CLIP.
jonnabioGuide fine-tuning of large language models with Axolotl configurations and advanced techniques.