| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.04681647940074907, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0009363295880149813, | |
| "eval_loss": 2.619614601135254, | |
| "eval_runtime": 19.951, | |
| "eval_samples_per_second": 22.555, | |
| "eval_steps_per_second": 11.278, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.009363295880149813, | |
| "grad_norm": 3.152433395385742, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 9.6871, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.012172284644194757, | |
| "eval_loss": 1.9787213802337646, | |
| "eval_runtime": 18.9672, | |
| "eval_samples_per_second": 23.725, | |
| "eval_steps_per_second": 11.863, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.018726591760299626, | |
| "grad_norm": 4.919699192047119, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 7.9577, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.024344569288389514, | |
| "eval_loss": 1.6615052223205566, | |
| "eval_runtime": 18.9799, | |
| "eval_samples_per_second": 23.709, | |
| "eval_steps_per_second": 11.855, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.028089887640449437, | |
| "grad_norm": 5.763250350952148, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 6.6716, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03651685393258427, | |
| "eval_loss": 1.5344139337539673, | |
| "eval_runtime": 18.9954, | |
| "eval_samples_per_second": 23.69, | |
| "eval_steps_per_second": 11.845, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.03745318352059925, | |
| "grad_norm": 7.3046369552612305, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 6.4393, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.04681647940074907, | |
| "grad_norm": 6.708037853240967, | |
| "learning_rate": 0.0, | |
| "loss": 6.3186, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.741215447908352e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |