| export MODEL_PATH="Qwen/Qwen2.5-7B-Instruct" | |
| CUDA_VISIBLE_DEVICES=5,6,7,8,9 accelerate launch \ | |
| --config_file ./accelerate_config_rm.yaml \ | |
| --main_process_port 29500 \ | |
| ./scripts/train_rm.py \ | |
| --model_name $MODEL_PATH \ | |
| --learning_rate 1e-5 \ | |
| --max_length 4096 \ | |
| --train_batch_size 1 \ | |
| --val_batch_size 1 \ | |
| --accumulation_steps 8 \ | |
| --num_epochs 30 \ | |
| --evaluation_steps 50 \ | |
| --reward_data_path ../data/sotopia_pi_bc_episodes_rm.json \ | |
| --template_path ../evals/qwen2.5-7b.jinja \ | |
| --checkpoint_dir ../rm_checkpoints_qwen2.5-7b | |