| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.896551724137931, | |
| "eval_steps": 500, | |
| "global_step": 21, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13793103448275862, | |
| "grad_norm": 14.019169682273084, | |
| "learning_rate": 6.666666666666666e-07, | |
| "loss": 1.507, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.27586206896551724, | |
| "grad_norm": 14.71228068046855, | |
| "learning_rate": 1.3333333333333332e-06, | |
| "loss": 1.4883, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.41379310344827586, | |
| "grad_norm": 14.342453028916033, | |
| "learning_rate": 2e-06, | |
| "loss": 1.4746, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.5517241379310345, | |
| "grad_norm": 11.865842851033273, | |
| "learning_rate": 1.984807753012208e-06, | |
| "loss": 1.4302, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.6896551724137931, | |
| "grad_norm": 9.644001731592224, | |
| "learning_rate": 1.9396926207859082e-06, | |
| "loss": 1.2784, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.8275862068965517, | |
| "grad_norm": 9.22710365972269, | |
| "learning_rate": 1.8660254037844386e-06, | |
| "loss": 1.2302, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.9655172413793104, | |
| "grad_norm": 8.622558679008812, | |
| "learning_rate": 1.766044443118978e-06, | |
| "loss": 1.1959, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.103448275862069, | |
| "grad_norm": 9.106960231342002, | |
| "learning_rate": 1.6427876096865393e-06, | |
| "loss": 1.0703, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.2413793103448276, | |
| "grad_norm": 8.29490409484742, | |
| "learning_rate": 1.5e-06, | |
| "loss": 1.042, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.3793103448275863, | |
| "grad_norm": 6.446031536810181, | |
| "learning_rate": 1.3420201433256689e-06, | |
| "loss": 0.9649, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.5172413793103448, | |
| "grad_norm": 5.764192061204997, | |
| "learning_rate": 1.1736481776669305e-06, | |
| "loss": 0.869, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.6551724137931034, | |
| "grad_norm": 5.6349278706228345, | |
| "learning_rate": 1e-06, | |
| "loss": 0.9707, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.793103448275862, | |
| "grad_norm": 4.868404389323644, | |
| "learning_rate": 8.263518223330696e-07, | |
| "loss": 0.9837, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.9310344827586206, | |
| "grad_norm": 4.44321806590041, | |
| "learning_rate": 6.579798566743313e-07, | |
| "loss": 0.9148, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.0689655172413794, | |
| "grad_norm": 4.338533070335014, | |
| "learning_rate": 5.000000000000002e-07, | |
| "loss": 0.7761, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.206896551724138, | |
| "grad_norm": 4.45792051740593, | |
| "learning_rate": 3.5721239031346063e-07, | |
| "loss": 0.8401, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.344827586206897, | |
| "grad_norm": 4.512087188956715, | |
| "learning_rate": 2.339555568810221e-07, | |
| "loss": 0.7472, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.4827586206896552, | |
| "grad_norm": 4.223326488483186, | |
| "learning_rate": 1.3397459621556128e-07, | |
| "loss": 0.8066, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.6206896551724137, | |
| "grad_norm": 4.252642416848704, | |
| "learning_rate": 6.030737921409168e-08, | |
| "loss": 0.8122, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 2.7586206896551726, | |
| "grad_norm": 4.1503783715330105, | |
| "learning_rate": 1.519224698779198e-08, | |
| "loss": 0.8134, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.896551724137931, | |
| "grad_norm": 3.741071114000594, | |
| "learning_rate": 0.0, | |
| "loss": 0.7617, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.896551724137931, | |
| "step": 21, | |
| "total_flos": 1395378216960.0, | |
| "train_loss": 1.0465347681726729, | |
| "train_runtime": 5265.2341, | |
| "train_samples_per_second": 0.132, | |
| "train_steps_per_second": 0.004 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 21, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1395378216960.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |