#####   ######  #####    ###    #   #  ###  #   #  ######
##  ##  ##      ##  ##  ## ##   #   #   #   #   #  ##
#####   ####    #####   #   #   #   #   #   #   #  ####
##  #   ##      ##      ## ##    # #    #    # #   ##
##   #  ######  ##       ###      #    ###    #    ######

$ curl repovive.com/roadmaps/llm-fine-tuning/fine-tuning-tools-frameworks/axolotl

░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░█████████████████████████████████████████████████████████████████████████████████

#####   ######  #####    ###    #   #  ###  #   #  ######
##  ##  ##      ##  ##  ## ##   #   #   #   #   #  ##
#####   ####    #####   #   #   #   #   #   #   #  ####
##  #   ##      ##      ## ##    # #    #    # #   ##
##   #  ######  ##       ###      #    ###    #    ######

$ curl repovive.com/roadmaps/llm-fine-tuning/fine-tuning-tools-frameworks/axolotl

Repovive

Axolotl - Fine-Tuning Tools & Frameworks | LLM Fine-Tuning | Repovive

Repovive.

LLM Fine-Tuning10 sections · 313 units1h total

Open in Course

Axolotl

Axolotl is configuration-driven fine-tuning. Write a YAML file specifying model, data, method, and hyperparameters. No code required.

base_model: meta-llama/Llama-2-7b-hf
load_in_4bit: true
adapter: qlora
lora_r: 16
datasets:
  - path: data.jsonl
    type: alpaca

Your config fully documents the training run. Others can replicate exactly. Axolotl supports dozens of options. Start simple and add complexity as needed.