| { | |
| "best_metric": 0.7537959814071655, | |
| "best_model_checkpoint": "../../saves/LLaMA2-7B/lora/sft/checkpoint-100", | |
| "epoch": 0.14814814814814814, | |
| "eval_steps": 100, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.014814814814814815, | |
| "grad_norm": 0.7613758444786072, | |
| "learning_rate": 2.25e-05, | |
| "loss": 1.5912, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02962962962962963, | |
| "grad_norm": 1.294195294380188, | |
| "learning_rate": 4.75e-05, | |
| "loss": 1.6519, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.044444444444444446, | |
| "grad_norm": 1.4575605392456055, | |
| "learning_rate": 4.999751424211201e-05, | |
| "loss": 1.4853, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.05925925925925926, | |
| "grad_norm": 1.807989239692688, | |
| "learning_rate": 4.998892213342501e-05, | |
| "loss": 1.5093, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.07407407407407407, | |
| "grad_norm": 1.788489818572998, | |
| "learning_rate": 4.997419509448194e-05, | |
| "loss": 1.1533, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.08888888888888889, | |
| "grad_norm": 1.494881510734558, | |
| "learning_rate": 4.995333674085917e-05, | |
| "loss": 0.9519, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.1037037037037037, | |
| "grad_norm": 11.77552604675293, | |
| "learning_rate": 4.992635219340763e-05, | |
| "loss": 0.8711, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.11851851851851852, | |
| "grad_norm": 1.7387481927871704, | |
| "learning_rate": 4.989324807699557e-05, | |
| "loss": 0.7735, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.13333333333333333, | |
| "grad_norm": 1.1280796527862549, | |
| "learning_rate": 4.985403251888209e-05, | |
| "loss": 0.8349, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.14814814814814814, | |
| "grad_norm": 4.847592353820801, | |
| "learning_rate": 4.980871514672192e-05, | |
| "loss": 0.8172, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.14814814814814814, | |
| "eval_loss": 0.7537959814071655, | |
| "eval_runtime": 21.5094, | |
| "eval_samples_per_second": 27.895, | |
| "eval_steps_per_second": 27.895, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2025, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 1.0818125662715904e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |