| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.888888888888889, | |
| "eval_steps": 500, | |
| "global_step": 28, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25396825396825395, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.642857142857143e-05, | |
| "loss": 72.1443, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.5079365079365079, | |
| "grad_norm": 19.469039916992188, | |
| "learning_rate": 8.92857142857143e-05, | |
| "loss": 72.0489, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.7619047619047619, | |
| "grad_norm": 20.524240493774414, | |
| "learning_rate": 8.571428571428571e-05, | |
| "loss": 69.7797, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.126984126984127, | |
| "grad_norm": 49.45151138305664, | |
| "learning_rate": 7.857142857142858e-05, | |
| "loss": 89.8185, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.380952380952381, | |
| "grad_norm": 32.59318161010742, | |
| "learning_rate": 7.142857142857143e-05, | |
| "loss": 62.2748, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.6349206349206349, | |
| "grad_norm": 32.76274871826172, | |
| "learning_rate": 6.428571428571429e-05, | |
| "loss": 57.2675, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.8888888888888888, | |
| "grad_norm": 36.317630767822266, | |
| "learning_rate": 5.714285714285714e-05, | |
| "loss": 56.7585, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.253968253968254, | |
| "grad_norm": 35.975807189941406, | |
| "learning_rate": 5e-05, | |
| "loss": 74.1119, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.507936507936508, | |
| "grad_norm": 20.22718620300293, | |
| "learning_rate": 4.2857142857142856e-05, | |
| "loss": 49.704, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.761904761904762, | |
| "grad_norm": 18.7318172454834, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 49.5238, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.126984126984127, | |
| "grad_norm": 34.112125396728516, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 66.7037, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 3.380952380952381, | |
| "grad_norm": 19.139488220214844, | |
| "learning_rate": 2.1428571428571428e-05, | |
| "loss": 46.9201, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 3.634920634920635, | |
| "grad_norm": 18.72429656982422, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 47.8016, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 3.888888888888889, | |
| "grad_norm": 17.140914916992188, | |
| "learning_rate": 7.142857142857143e-06, | |
| "loss": 46.4582, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 3.888888888888889, | |
| "step": 28, | |
| "total_flos": 445653973809192.0, | |
| "train_loss": 61.52253450666155, | |
| "train_runtime": 1057.0707, | |
| "train_samples_per_second": 0.477, | |
| "train_steps_per_second": 0.026 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 28, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "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": 445653973809192.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |