This model was created as part of the sft_Llama3.2-3B_R1Distill-sft experiment using the SkillFactory experiment management system.

{"model_name_or_path": "meta-llama/Llama-3.2-3B-Instruct", "trust_remote_code": true, "stage": "sft", "do_train": true, "finetuning_type": "full", "deepspeed": "/scratch/10416/zaynesprague/skill_factory_dir/skill-factory/thirdparty/LLaMA-Factory/examples/deepspeed/ds_z2_config.json", "dataset": "TAUR_dev__D_SFT_C_sft_Llama3_2_3B_R1Distill_sft_sft_data__sft_train", "template": "llama3", "cutoff_len": 16384, "max_samples": 1000000, "overwrite_cache": true, "preprocessing_num_workers": 1, "dataloader_num_workers": 0, "disable_tqdm": false, "output_dir": "/scratch/10416/zaynesprague/skill_inject_outputs/sf_experiments/sft_llama323b_r1distill_sft/llamafactory/checkpoints", "logging_steps": 10, "save_steps": 100000, "plot_loss": true, "overwrite_output_dir": true, "per_device_train_batch_size": 1, "gradient_accumulation_steps": 1, "learning_rate": 1e-06, "num_train_epochs": 2, "lr_scheduler_type": "cosine", "warmup_ratio": 0.05, "weight_decay": 0.0001, "adam_beta1": 0.9, "adam_beta2": 0.95, "bf16": true, "ddp_timeout": 180000000, "gradient_checkpointing": true, "save_only_model": true, "enable_masked_ranges": false, "save_strategy": "steps", "save_total_limit": 5, "sf_tracker_dataset_id": "TAUR-dev/D-ExpTracker__sft_Llama3.2-3B_R1Distill-sft__v1", "sf_eval_before_training": false, "sf_wandb_project": "sft_Llama3.2-3B_R1Distill-sft_sft", "sf_eval_steps": null, "run_name": "sft_Llama3.2-3B_R1Distill-sft_sft"}

🔗 View complete experiment details: Experiment Tracker Dataset