| { | |
| "best_metric": 1.523019790649414, | |
| "best_model_checkpoint": "./mbartLarge_mid_en-ko1/checkpoint-1500", | |
| "epoch": 5.609573672400898, | |
| "eval_steps": 1500, | |
| "global_step": 7500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 5e-05, | |
| "loss": 1.7061, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.952812382030955e-05, | |
| "loss": 1.619, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 4.9056247640619104e-05, | |
| "loss": 1.4343, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_bleu": 11.2393, | |
| "eval_gen_len": 16.3612, | |
| "eval_loss": 1.523019790649414, | |
| "eval_runtime": 172.8656, | |
| "eval_samples_per_second": 15.457, | |
| "eval_steps_per_second": 0.966, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 4.8584371460928654e-05, | |
| "loss": 1.1601, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 4.8112495281238205e-05, | |
| "loss": 1.1956, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 4.7640619101547756e-05, | |
| "loss": 0.9153, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "eval_bleu": 11.9955, | |
| "eval_gen_len": 16.2665, | |
| "eval_loss": 1.600303053855896, | |
| "eval_runtime": 166.8515, | |
| "eval_samples_per_second": 16.014, | |
| "eval_steps_per_second": 1.001, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 4.7168742921857306e-05, | |
| "loss": 0.79, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 4.669686674216686e-05, | |
| "loss": 0.8315, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 4.622499056247641e-05, | |
| "loss": 0.4823, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "eval_bleu": 11.5074, | |
| "eval_gen_len": 16.1329, | |
| "eval_loss": 1.7483373880386353, | |
| "eval_runtime": 163.6405, | |
| "eval_samples_per_second": 16.328, | |
| "eval_steps_per_second": 1.021, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 4.575311438278596e-05, | |
| "loss": 0.5193, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 4.528123820309551e-05, | |
| "loss": 0.4565, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 4.480936202340506e-05, | |
| "loss": 0.3021, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "eval_bleu": 11.452, | |
| "eval_gen_len": 16.2631, | |
| "eval_loss": 1.888548731803894, | |
| "eval_runtime": 159.6603, | |
| "eval_samples_per_second": 16.736, | |
| "eval_steps_per_second": 1.046, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "learning_rate": 4.433748584371461e-05, | |
| "loss": 0.3266, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 4.386560966402416e-05, | |
| "loss": 0.2338, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 4.339373348433372e-05, | |
| "loss": 0.1974, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "eval_bleu": 12.065, | |
| "eval_gen_len": 16.0183, | |
| "eval_loss": 2.0004332065582275, | |
| "eval_runtime": 159.6772, | |
| "eval_samples_per_second": 16.734, | |
| "eval_steps_per_second": 1.046, | |
| "step": 7500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 53480, | |
| "num_train_epochs": 40, | |
| "save_steps": 1500, | |
| "total_flos": 2.6005557215232e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |