注意上图画圈的部分，如果选择LLM ORPO，那么chat-template要选择chatml
在huggingface运行
通过ngrok在colab运行UI界面
ngrok token
https://dashboard.ngrok.com/get-started/your-authtoken
本地微调命令
conda create -n autotrain python=3.10
conda activate autotrain
pip install autotrain-advanced
conda install pytorch torchvision torchaudio pytorch-cuda=12.1 -c pytorch -c nvidia
conda install -c "nvidia/label/cuda-12.1.0" cuda-nvccconda install xformers -c xformers
python -m nltk.downloader punkt
pip install flash-attn --no-build-isolation # if you want to use flash-attn
pip install deepspeed # if you want to use deepspeed
#运行微调，注意，请先设置配置文件的内容
autotrain --config 这里填自己文件名
本地微调的配置文件内容:
task: llm-orpo
base_model: unsloth/llama-3-8b-Instruct
project_name: autotrain-llama3-8b-orpo
log: tensorboard
backend: localdata:
path: argilla/distilabel-capybara-dpo-7k-binarized
train_split: train
valid_split: null
chat_template: chatml
column_mapping:
text_column: chosen
rejected_text_column: rejected
prompt_text_column: promptparams:
block_size: 1024
model_max_length: 8192
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: fp16hub:
username: leo009
token: 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
push_to_hub: true
