| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 11115, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1349527665317139, | |
| "grad_norm": 7.277841567993164, | |
| "learning_rate": 4.775078722447144e-05, | |
| "loss": 3.0654, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2699055330634278, | |
| "grad_norm": 7.536296844482422, | |
| "learning_rate": 4.550157444894287e-05, | |
| "loss": 2.4088, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.4048582995951417, | |
| "grad_norm": 6.857296466827393, | |
| "learning_rate": 4.3252361673414306e-05, | |
| "loss": 2.1502, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.5398110661268556, | |
| "grad_norm": 20.20909309387207, | |
| "learning_rate": 4.100314889788574e-05, | |
| "loss": 1.9977, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.6747638326585695, | |
| "grad_norm": 6.654603004455566, | |
| "learning_rate": 3.8753936122357176e-05, | |
| "loss": 1.8983, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.8097165991902834, | |
| "grad_norm": 7.12253999710083, | |
| "learning_rate": 3.650472334682861e-05, | |
| "loss": 1.8103, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.9446693657219973, | |
| "grad_norm": 6.6242170333862305, | |
| "learning_rate": 3.4255510571300045e-05, | |
| "loss": 1.7543, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.0796221322537112, | |
| "grad_norm": 5.963022232055664, | |
| "learning_rate": 3.2006297795771486e-05, | |
| "loss": 1.6822, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.214574898785425, | |
| "grad_norm": 6.769837856292725, | |
| "learning_rate": 2.9757085020242914e-05, | |
| "loss": 1.6501, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.349527665317139, | |
| "grad_norm": 6.467240333557129, | |
| "learning_rate": 2.750787224471435e-05, | |
| "loss": 1.612, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.484480431848853, | |
| "grad_norm": 6.431339740753174, | |
| "learning_rate": 2.5258659469185787e-05, | |
| "loss": 1.5606, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.6194331983805668, | |
| "grad_norm": 7.2527289390563965, | |
| "learning_rate": 2.300944669365722e-05, | |
| "loss": 1.5413, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.7543859649122808, | |
| "grad_norm": 6.609572887420654, | |
| "learning_rate": 2.0760233918128656e-05, | |
| "loss": 1.5367, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.8893387314439947, | |
| "grad_norm": 6.267154216766357, | |
| "learning_rate": 1.851102114260009e-05, | |
| "loss": 1.5087, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.0242914979757085, | |
| "grad_norm": 6.8475117683410645, | |
| "learning_rate": 1.6261808367071525e-05, | |
| "loss": 1.4727, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.1592442645074224, | |
| "grad_norm": 6.945182800292969, | |
| "learning_rate": 1.401259559154296e-05, | |
| "loss": 1.4564, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.294197031039136, | |
| "grad_norm": 6.858353137969971, | |
| "learning_rate": 1.1763382816014395e-05, | |
| "loss": 1.4325, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.42914979757085, | |
| "grad_norm": 6.653971195220947, | |
| "learning_rate": 9.51417004048583e-06, | |
| "loss": 1.4041, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.564102564102564, | |
| "grad_norm": 5.984017848968506, | |
| "learning_rate": 7.264957264957266e-06, | |
| "loss": 1.4085, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.699055330634278, | |
| "grad_norm": 6.64699649810791, | |
| "learning_rate": 5.0157444894287e-06, | |
| "loss": 1.365, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.834008097165992, | |
| "grad_norm": 6.052412509918213, | |
| "learning_rate": 2.766531713900135e-06, | |
| "loss": 1.3799, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.968960863697706, | |
| "grad_norm": 6.4499711990356445, | |
| "learning_rate": 5.1731893837157e-07, | |
| "loss": 1.3994, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 11115, | |
| "total_flos": 1.1730479287894016e+16, | |
| "train_loss": 1.7010253453222524, | |
| "train_runtime": 4017.837, | |
| "train_samples_per_second": 11.064, | |
| "train_steps_per_second": 2.766 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 11115, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.1730479287894016e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |