| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 1240, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8064516129032258, | |
| "grad_norm": 2.2094340324401855, | |
| "learning_rate": 2.8814516129032256e-05, | |
| "loss": 0.821, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.6129032258064515, | |
| "grad_norm": 4.53880500793457, | |
| "learning_rate": 2.760483870967742e-05, | |
| "loss": 0.3767, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.4193548387096775, | |
| "grad_norm": 2.487438917160034, | |
| "learning_rate": 2.6395161290322583e-05, | |
| "loss": 0.228, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.225806451612903, | |
| "grad_norm": 1.6553895473480225, | |
| "learning_rate": 2.5185483870967742e-05, | |
| "loss": 0.1713, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.032258064516129, | |
| "grad_norm": 0.6422418355941772, | |
| "learning_rate": 2.3975806451612904e-05, | |
| "loss": 0.1186, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.838709677419355, | |
| "grad_norm": 1.2269980907440186, | |
| "learning_rate": 2.2766129032258065e-05, | |
| "loss": 0.0827, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.645161290322581, | |
| "grad_norm": 1.165654182434082, | |
| "learning_rate": 2.1556451612903224e-05, | |
| "loss": 0.0598, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 6.451612903225806, | |
| "grad_norm": 1.4293478727340698, | |
| "learning_rate": 2.034677419354839e-05, | |
| "loss": 0.0373, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 7.258064516129032, | |
| "grad_norm": 1.201370120048523, | |
| "learning_rate": 1.913709677419355e-05, | |
| "loss": 0.0339, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 8.064516129032258, | |
| "grad_norm": 1.2729601860046387, | |
| "learning_rate": 1.792741935483871e-05, | |
| "loss": 0.0318, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 8.870967741935484, | |
| "grad_norm": 0.45044800639152527, | |
| "learning_rate": 1.671774193548387e-05, | |
| "loss": 0.0234, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 9.67741935483871, | |
| "grad_norm": 0.26697471737861633, | |
| "learning_rate": 1.5508064516129033e-05, | |
| "loss": 0.0219, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 10.483870967741936, | |
| "grad_norm": 0.787188708782196, | |
| "learning_rate": 1.4298387096774195e-05, | |
| "loss": 0.0132, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 11.290322580645162, | |
| "grad_norm": 2.1090033054351807, | |
| "learning_rate": 1.3088709677419355e-05, | |
| "loss": 0.0106, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 12.096774193548388, | |
| "grad_norm": 0.08617769926786423, | |
| "learning_rate": 1.1879032258064517e-05, | |
| "loss": 0.0097, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 12.903225806451612, | |
| "grad_norm": 0.5370274782180786, | |
| "learning_rate": 1.0669354838709679e-05, | |
| "loss": 0.0096, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 13.709677419354838, | |
| "grad_norm": 0.9249340891838074, | |
| "learning_rate": 9.459677419354839e-06, | |
| "loss": 0.009, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 14.516129032258064, | |
| "grad_norm": 0.43685030937194824, | |
| "learning_rate": 8.25e-06, | |
| "loss": 0.0068, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 15.32258064516129, | |
| "grad_norm": 0.44842422008514404, | |
| "learning_rate": 7.040322580645162e-06, | |
| "loss": 0.005, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 16.129032258064516, | |
| "grad_norm": 0.44923698902130127, | |
| "learning_rate": 5.8306451612903235e-06, | |
| "loss": 0.0054, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 16.93548387096774, | |
| "grad_norm": 0.49068453907966614, | |
| "learning_rate": 4.620967741935484e-06, | |
| "loss": 0.0043, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 17.741935483870968, | |
| "grad_norm": 0.147027388215065, | |
| "learning_rate": 3.4112903225806454e-06, | |
| "loss": 0.0028, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 18.548387096774192, | |
| "grad_norm": 1.196764349937439, | |
| "learning_rate": 2.2016129032258064e-06, | |
| "loss": 0.0035, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 19.35483870967742, | |
| "grad_norm": 0.022890331223607063, | |
| "learning_rate": 9.919354838709678e-07, | |
| "loss": 0.0027, | |
| "step": 1200 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 1240, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "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": 1192381370848800.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |