| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.023408239700374533, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00046816479400749064, | |
| "eval_loss": 2.619614601135254, | |
| "eval_runtime": 19.9062, | |
| "eval_samples_per_second": 22.606, | |
| "eval_steps_per_second": 11.303, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0046816479400749065, | |
| "grad_norm": 2.752786874771118, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 4.7103, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.006086142322097378, | |
| "eval_loss": 1.8191447257995605, | |
| "eval_runtime": 18.9341, | |
| "eval_samples_per_second": 23.767, | |
| "eval_steps_per_second": 11.883, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.009363295880149813, | |
| "grad_norm": 4.815334320068359, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 3.4808, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.012172284644194757, | |
| "eval_loss": 1.5765701532363892, | |
| "eval_runtime": 18.9948, | |
| "eval_samples_per_second": 23.691, | |
| "eval_steps_per_second": 11.845, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.014044943820224719, | |
| "grad_norm": 4.303775310516357, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 3.4651, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.018258426966292134, | |
| "eval_loss": 1.4553884267807007, | |
| "eval_runtime": 19.023, | |
| "eval_samples_per_second": 23.656, | |
| "eval_steps_per_second": 11.828, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.018726591760299626, | |
| "grad_norm": 3.952723741531372, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 2.9012, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.023408239700374533, | |
| "grad_norm": 4.169103145599365, | |
| "learning_rate": 0.0, | |
| "loss": 3.01, | |
| "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": 8680794331545600.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |