{ "best_metric": 0.7738199234008789, "best_model_checkpoint": "../artifacts/LlaMa3-QLoRA-PatentMatch-v0.1/checkpoint-20", "epoch": 0.18867924528301888, "eval_steps": 20, "global_step": 20, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.009433962264150943, "grad_norm": 20.337095260620117, "learning_rate": 2e-05, "loss": 0.5913, "step": 1 }, { "epoch": 0.018867924528301886, "grad_norm": 11.506393432617188, "learning_rate": 2e-05, "loss": 0.6775, "step": 2 }, { "epoch": 0.02830188679245283, "grad_norm": 9.703904151916504, "learning_rate": 2e-05, "loss": 0.576, "step": 3 }, { "epoch": 0.03773584905660377, "grad_norm": 11.118324279785156, "learning_rate": 2e-05, "loss": 0.5084, "step": 4 }, { "epoch": 0.04716981132075472, "grad_norm": 13.329315185546875, "learning_rate": 2e-05, "loss": 0.5712, "step": 5 }, { "epoch": 0.05660377358490566, "grad_norm": 29.63173484802246, "learning_rate": 2e-05, "loss": 0.7234, "step": 6 }, { "epoch": 0.0660377358490566, "grad_norm": 17.787134170532227, "learning_rate": 2e-05, "loss": 0.6053, "step": 7 }, { "epoch": 0.07547169811320754, "grad_norm": 30.639591217041016, "learning_rate": 2e-05, "loss": 0.6873, "step": 8 }, { "epoch": 0.08490566037735849, "grad_norm": 9.612072944641113, "learning_rate": 2e-05, "loss": 0.5541, "step": 9 }, { "epoch": 0.09433962264150944, "grad_norm": 8.989519119262695, "learning_rate": 2e-05, "loss": 0.7858, "step": 10 }, { "epoch": 0.10377358490566038, "grad_norm": 17.486469268798828, "learning_rate": 2e-05, "loss": 0.9176, "step": 11 }, { "epoch": 0.11320754716981132, "grad_norm": 35.29791259765625, "learning_rate": 2e-05, "loss": 0.6558, "step": 12 }, { "epoch": 0.12264150943396226, "grad_norm": 19.468692779541016, "learning_rate": 2e-05, "loss": 0.6085, "step": 13 }, { "epoch": 0.1320754716981132, "grad_norm": 9.410886764526367, "learning_rate": 2e-05, "loss": 0.6229, "step": 14 }, { "epoch": 0.14150943396226415, "grad_norm": 11.87700080871582, "learning_rate": 2e-05, "loss": 0.5764, "step": 15 }, { "epoch": 0.1509433962264151, "grad_norm": 20.188251495361328, "learning_rate": 2e-05, "loss": 0.8275, "step": 16 }, { "epoch": 0.16037735849056603, "grad_norm": 28.298933029174805, "learning_rate": 2e-05, "loss": 0.5896, "step": 17 }, { "epoch": 0.16981132075471697, "grad_norm": 47.8366813659668, "learning_rate": 2e-05, "loss": 0.8496, "step": 18 }, { "epoch": 0.1792452830188679, "grad_norm": 36.19501495361328, "learning_rate": 2e-05, "loss": 0.6756, "step": 19 }, { "epoch": 0.18867924528301888, "grad_norm": 22.574682235717773, "learning_rate": 2e-05, "loss": 0.6366, "step": 20 }, { "epoch": 0.18867924528301888, "eval_loss": 0.7738199234008789, "eval_runtime": 18.5622, "eval_samples_per_second": 15.893, "eval_steps_per_second": 3.179, "step": 20 } ], "logging_steps": 1, "max_steps": 106, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 20, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 7622883106406400.0, "train_batch_size": 5, "trial_name": null, "trial_params": null }