wehub-resource-sync
38 lines
1.1 KiB
Bash
38 lines
1.1 KiB
Bash
# pip install math_verify # reward function
|
|
# pip install -U trl
|
|
# GPU memory: 80GiB
|
|
# You can set `--reward_model` to use a reward model to provide rewards.
|
|
# TransformersEngine to rollout
|
|
|
|
CUDA_VISIBLE_DEVICES=0 \
|
|
swift rlhf \
|
|
--rlhf_type grpo \
|
|
--model Qwen/Qwen2.5-7B \
|
|
--reward_funcs accuracy format \
|
|
--tuner_type lora \
|
|
--lora_rank 8 \
|
|
--lora_alpha 32 \
|
|
--target_modules all-linear \
|
|
--torch_dtype bfloat16 \
|
|
--dataset 'AI-MO/NuminaMath-TIR#1000' \
|
|
--load_from_cache_file true \
|
|
--max_completion_length 1024 \
|
|
--num_train_epochs 1 \
|
|
--per_device_train_batch_size 4 \
|
|
--per_device_eval_batch_size 4 \
|
|
--learning_rate 1e-5 \
|
|
--gradient_accumulation_steps 1 \
|
|
--eval_steps 100 \
|
|
--save_steps 100 \
|
|
--save_total_limit 2 \
|
|
--logging_steps 5 \
|
|
--max_length 2048 \
|
|
--output_dir output \
|
|
--warmup_ratio 0.05 \
|
|
--dataloader_num_workers 4 \
|
|
--dataset_num_proc 4 \
|
|
--num_generations 4 \
|
|
--temperature 0.9 \
|
|
--system 'examples/train/grpo/prompt.txt' \
|
|
--log_completions true
|