| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 2826, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5307855626326964, | |
| "grad_norm": 5.489389896392822, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 2.354, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0615711252653928, | |
| "grad_norm": 5.129901885986328, | |
| "learning_rate": 8.203054806828391e-06, | |
| "loss": 1.0543, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.5923566878980893, | |
| "grad_norm": 9.780948638916016, | |
| "learning_rate": 5.9568733153638815e-06, | |
| "loss": 0.8205, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.1231422505307855, | |
| "grad_norm": 8.44005012512207, | |
| "learning_rate": 3.710691823899371e-06, | |
| "loss": 0.7534, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.653927813163482, | |
| "grad_norm": 4.482905387878418, | |
| "learning_rate": 1.464510332434861e-06, | |
| "loss": 0.6727, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 2826, | |
| "total_flos": 2.377533515518771e+16, | |
| "train_loss": 1.0759811563350086, | |
| "train_runtime": 46811.0436, | |
| "train_samples_per_second": 1.93, | |
| "train_steps_per_second": 0.06 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2826, | |
| "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": 2.377533515518771e+16, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |