| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.6178010471204187, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 6.25e-06, | |
| "loss": 7.3722, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_runtime": 4.9957, | |
| "eval_samples_per_second": 20.017, | |
| "eval_steps_per_second": 10.009, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 9.895833333333334e-06, | |
| "loss": 1.5052, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "eval_runtime": 5.5484, | |
| "eval_samples_per_second": 18.023, | |
| "eval_steps_per_second": 9.012, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 9.635416666666668e-06, | |
| "loss": 1.3479, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_runtime": 5.4316, | |
| "eval_samples_per_second": 18.411, | |
| "eval_steps_per_second": 9.205, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 9.375000000000001e-06, | |
| "loss": 1.3177, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_runtime": 5.4246, | |
| "eval_samples_per_second": 18.435, | |
| "eval_steps_per_second": 9.217, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 9.114583333333334e-06, | |
| "loss": 1.2962, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_runtime": 5.0933, | |
| "eval_samples_per_second": 19.634, | |
| "eval_steps_per_second": 9.817, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 8.854166666666667e-06, | |
| "loss": 1.2824, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "eval_runtime": 6.1894, | |
| "eval_samples_per_second": 16.157, | |
| "eval_steps_per_second": 8.078, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 8.59375e-06, | |
| "loss": 1.2784, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "eval_runtime": 4.7573, | |
| "eval_samples_per_second": 21.02, | |
| "eval_steps_per_second": 10.51, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 1.2655, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "eval_runtime": 5.5685, | |
| "eval_samples_per_second": 17.958, | |
| "eval_steps_per_second": 8.979, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 8.072916666666667e-06, | |
| "loss": 1.2553, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "eval_runtime": 6.732, | |
| "eval_samples_per_second": 14.854, | |
| "eval_steps_per_second": 7.427, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 7.8125e-06, | |
| "loss": 1.2627, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "eval_runtime": 4.5523, | |
| "eval_samples_per_second": 21.967, | |
| "eval_steps_per_second": 10.983, | |
| "step": 5000 | |
| } | |
| ], | |
| "max_steps": 20000, | |
| "num_train_epochs": 11, | |
| "total_flos": 4.184046875234304e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |