| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.005016051364365971, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00010032102728731942, | |
| "eval_loss": 1.6005431413650513, | |
| "eval_runtime": 48.9287, | |
| "eval_samples_per_second": 85.798, | |
| "eval_steps_per_second": 42.899, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0010032102728731941, | |
| "grad_norm": 18.93649673461914, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 6.5127, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0013041733547351525, | |
| "eval_loss": 1.5364631414413452, | |
| "eval_runtime": 48.476, | |
| "eval_samples_per_second": 86.599, | |
| "eval_steps_per_second": 43.3, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0020064205457463883, | |
| "grad_norm": 25.3447265625, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 6.1434, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.002608346709470305, | |
| "eval_loss": 1.4999792575836182, | |
| "eval_runtime": 48.7504, | |
| "eval_samples_per_second": 86.112, | |
| "eval_steps_per_second": 43.056, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.003009630818619583, | |
| "grad_norm": 19.626148223876953, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 6.0478, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.003912520064205458, | |
| "eval_loss": 1.4834470748901367, | |
| "eval_runtime": 48.6482, | |
| "eval_samples_per_second": 86.293, | |
| "eval_steps_per_second": 43.146, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0040128410914927765, | |
| "grad_norm": 22.51584815979004, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 5.7603, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.005016051364365971, | |
| "grad_norm": 20.262388229370117, | |
| "learning_rate": 0.0, | |
| "loss": 5.777, | |
| "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": 877285657804800.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |