configs/llm_finetuning/llama3-8b-dpo-qlora.yml

task: llm-dpo
base_model: meta-llama/Meta-Llama-3-8B-Instruct
project_name: autotrain-llama3-8b-dpo-qlora
log: tensorboard
backend: local

data:
  path: mlabonne/orpo-dpo-mix-40k
  train_split: train
  valid_split: null
  chat_template: chatml
  column_mapping:
    text_column: chosen
    rejected_text_column: rejected
    prompt_text_column: prompt

params:
  block_size: 1024
  model_max_length: 2048
  max_prompt_length: 512
  epochs: 3
  batch_size: 2
  lr: 3e-5
  peft: true
  quantization: int4
  target_modules: all-linear
  padding: right
  optimizer: adamw_torch
  scheduler: linear
  gradient_accumulation: 4
  mixed_precision: fp16

hub:
  username: ${HF_USERNAME}
  token: ${HF_TOKEN}
  push_to_hub: false