| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 8260, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 6.25e-05, | |
| "loss": 5.7733, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.4626414056321562, | |
| "eval_loss": 3.1065430641174316, | |
| "eval_runtime": 4.0915, | |
| "eval_samples_per_second": 1231.319, | |
| "eval_steps_per_second": 2.444, | |
| "step": 826 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 0.000125, | |
| "loss": 3.1078, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.0001875, | |
| "loss": 2.7977, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.49604892580170235, | |
| "eval_loss": 2.750513792037964, | |
| "eval_runtime": 4.2814, | |
| "eval_samples_per_second": 1176.724, | |
| "eval_steps_per_second": 2.336, | |
| "step": 1652 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 0.00025, | |
| "loss": 2.6214, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5098120426490952, | |
| "eval_loss": 2.5736496448516846, | |
| "eval_runtime": 4.397, | |
| "eval_samples_per_second": 1145.772, | |
| "eval_steps_per_second": 2.274, | |
| "step": 2478 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 0.0003125, | |
| "loss": 2.513, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "learning_rate": 0.000375, | |
| "loss": 2.4167, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5208181599372329, | |
| "eval_loss": 2.4625895023345947, | |
| "eval_runtime": 4.4219, | |
| "eval_samples_per_second": 1139.337, | |
| "eval_steps_per_second": 2.261, | |
| "step": 3304 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 0.00043750000000000006, | |
| "loss": 2.3476, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 0.0005, | |
| "loss": 2.291, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5284624257219932, | |
| "eval_loss": 2.394136667251587, | |
| "eval_runtime": 4.4258, | |
| "eval_samples_per_second": 1138.337, | |
| "eval_steps_per_second": 2.26, | |
| "step": 4130 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 0.0005625000000000001, | |
| "loss": 2.2345, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5323009693260355, | |
| "eval_loss": 2.3510043621063232, | |
| "eval_runtime": 4.5875, | |
| "eval_samples_per_second": 1098.192, | |
| "eval_steps_per_second": 2.18, | |
| "step": 4956 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 0.000625, | |
| "loss": 2.2073, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.66, | |
| "learning_rate": 0.0006875, | |
| "loss": 2.1641, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5357347153757428, | |
| "eval_loss": 2.32427716255188, | |
| "eval_runtime": 4.4956, | |
| "eval_samples_per_second": 1120.641, | |
| "eval_steps_per_second": 2.224, | |
| "step": 5782 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "learning_rate": 0.00075, | |
| "loss": 2.14, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.87, | |
| "learning_rate": 0.0008125, | |
| "loss": 2.1255, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5391278253775245, | |
| "eval_loss": 2.3039562702178955, | |
| "eval_runtime": 4.4937, | |
| "eval_samples_per_second": 1121.12, | |
| "eval_steps_per_second": 2.225, | |
| "step": 6608 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "learning_rate": 0.0008750000000000001, | |
| "loss": 2.0883, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.5396185838024713, | |
| "eval_loss": 2.294677257537842, | |
| "eval_runtime": 4.5308, | |
| "eval_samples_per_second": 1111.95, | |
| "eval_steps_per_second": 2.207, | |
| "step": 7434 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "learning_rate": 0.0009375, | |
| "loss": 2.096, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 9.69, | |
| "learning_rate": 0.001, | |
| "loss": 2.0612, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5411096141763542, | |
| "eval_loss": 2.2898333072662354, | |
| "eval_runtime": 4.54, | |
| "eval_samples_per_second": 1109.693, | |
| "eval_steps_per_second": 2.203, | |
| "step": 8260 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 8260, | |
| "total_flos": 2562796651806720.0, | |
| "train_loss": 2.546341962098498, | |
| "train_runtime": 742.1102, | |
| "train_samples_per_second": 711.646, | |
| "train_steps_per_second": 11.13 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 8260, | |
| "num_train_epochs": 10, | |
| "save_steps": 2000, | |
| "total_flos": 2562796651806720.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |