| ### model | |
| model_name_or_path: Qwen/Qwen2.5-Coder-7B | |
| #model_name_or_path: meta-llama/Meta-Llama-3-8B-Instruct | |
| trust_remote_code: true | |
| ### method | |
| stage: sft | |
| do_train: true | |
| finetuning_type: lora | |
| lora_rank: 128 | |
| lora_target: all | |
| ### dataset | |
| dataset: std_premises | |
| template: qwen | |
| cutoff_len: 2048 | |
| max_samples: 50000 | |
| overwrite_cache: true | |
| preprocessing_num_workers: 16 | |
| dataloader_num_workers: 4 | |
| ### output | |
| output_dir: /projects/bdkg/jhu11/saves/qwen2.5-coder-7b-3epoch-r128-purged-json-formatted/lora/sft | |
| logging_steps: 10 | |
| save_steps: 500 | |
| plot_loss: true | |
| overwrite_output_dir: true | |
| save_only_model: false | |
| ### train | |
| per_device_train_batch_size: 2 | |
| gradient_accumulation_steps: 8 | |
| learning_rate: 1.0e-4 | |
| num_train_epochs: 3.0 | |
| lr_scheduler_type: cosine | |
| warmup_ratio: 0.1 | |
| bf16: true | |
| ddp_timeout: 180000000 | |
| resume_from_checkpoint: null | |
| ### eval | |
| eval_dataset: std_premises_evaluation | |
| ### val_size: 0.1 | |
| per_device_eval_batch_size: 2 | |
| eval_strategy: steps | |
| eval_steps: 500 | |