| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.888888888888889, | |
| "eval_steps": 500, | |
| "global_step": 21, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25396825396825395, | |
| "grad_norm": 22.455032348632812, | |
| "learning_rate": 0.00018095238095238095, | |
| "loss": 71.5526, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.5079365079365079, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018095238095238095, | |
| "loss": 68.4573, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.7619047619047619, | |
| "grad_norm": 27.567764282226562, | |
| "learning_rate": 0.00016190476190476192, | |
| "loss": 66.9582, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.126984126984127, | |
| "grad_norm": 61.912845611572266, | |
| "learning_rate": 0.00014285714285714287, | |
| "loss": 78.8634, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.380952380952381, | |
| "grad_norm": 38.26054000854492, | |
| "learning_rate": 0.0001238095238095238, | |
| "loss": 53.5637, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.6349206349206349, | |
| "grad_norm": 19.41400146484375, | |
| "learning_rate": 0.00010476190476190477, | |
| "loss": 47.919, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.8888888888888888, | |
| "grad_norm": 21.822195053100586, | |
| "learning_rate": 8.571428571428571e-05, | |
| "loss": 47.9407, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.253968253968254, | |
| "grad_norm": 18.68866729736328, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 63.3071, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.507936507936508, | |
| "grad_norm": 28.52600860595703, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 43.1297, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.761904761904762, | |
| "grad_norm": 12.20601749420166, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 43.0936, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.888888888888889, | |
| "step": 21, | |
| "total_flos": 331427311589016.0, | |
| "train_loss": 57.708736782982236, | |
| "train_runtime": 785.4183, | |
| "train_samples_per_second": 0.481, | |
| "train_steps_per_second": 0.027 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 21, | |
| "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": 331427311589016.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |