| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.188679245283019, | |
| "eval_steps": 50, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1509433962264151, | |
| "grad_norm": 4.232557773590088, | |
| "learning_rate": 5e-06, | |
| "loss": 1.8485, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.1509433962264151, | |
| "eval_loss": 1.865347146987915, | |
| "eval_runtime": 3.8366, | |
| "eval_samples_per_second": 5.995, | |
| "eval_steps_per_second": 0.782, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.3018867924528302, | |
| "grad_norm": 4.266393184661865, | |
| "learning_rate": 1e-05, | |
| "loss": 1.7945, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.4528301886792453, | |
| "grad_norm": 4.149282932281494, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.8377, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.6037735849056604, | |
| "grad_norm": 4.348968505859375, | |
| "learning_rate": 2e-05, | |
| "loss": 1.772, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.7547169811320755, | |
| "grad_norm": 4.1208930015563965, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.7184, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.9056603773584906, | |
| "grad_norm": 4.385045051574707, | |
| "learning_rate": 3e-05, | |
| "loss": 1.7101, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.1132075471698113, | |
| "grad_norm": 7.531089782714844, | |
| "learning_rate": 3.5e-05, | |
| "loss": 2.8336, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.2641509433962264, | |
| "grad_norm": 3.102412223815918, | |
| "learning_rate": 4e-05, | |
| "loss": 1.357, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.4150943396226414, | |
| "grad_norm": 2.1879281997680664, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.1962, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.5660377358490565, | |
| "grad_norm": 1.4032061100006104, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0743, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.7169811320754715, | |
| "grad_norm": 0.9539682865142822, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 0.958, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.8679245283018868, | |
| "grad_norm": 1.0121053457260132, | |
| "learning_rate": 6e-05, | |
| "loss": 0.8695, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.0754716981132075, | |
| "grad_norm": 2.0145821571350098, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 1.4344, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.2264150943396226, | |
| "grad_norm": 1.0492981672286987, | |
| "learning_rate": 7e-05, | |
| "loss": 0.7102, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.3773584905660377, | |
| "grad_norm": 1.0005221366882324, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.6064, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.5283018867924527, | |
| "grad_norm": 1.096711277961731, | |
| "learning_rate": 8e-05, | |
| "loss": 0.6273, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.6792452830188678, | |
| "grad_norm": 0.9175000786781311, | |
| "learning_rate": 8.5e-05, | |
| "loss": 0.5188, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.830188679245283, | |
| "grad_norm": 0.8022292852401733, | |
| "learning_rate": 9e-05, | |
| "loss": 0.4369, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.0377358490566038, | |
| "grad_norm": 1.2239990234375, | |
| "learning_rate": 9.5e-05, | |
| "loss": 0.6315, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 3.188679245283019, | |
| "grad_norm": 0.5746318697929382, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3028, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.1184619093229568e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |