| { | |
| "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": 5.8796, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.4610925470362255, | |
| "eval_loss": 3.108325958251953, | |
| "eval_runtime": 4.3048, | |
| "eval_samples_per_second": 1170.326, | |
| "eval_steps_per_second": 2.323, | |
| "step": 826 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 0.000125, | |
| "loss": 3.1165, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.0001875, | |
| "loss": 2.802, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.49653186960204804, | |
| "eval_loss": 2.7454917430877686, | |
| "eval_runtime": 4.0976, | |
| "eval_samples_per_second": 1229.489, | |
| "eval_steps_per_second": 2.44, | |
| "step": 1652 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 0.00025, | |
| "loss": 2.6268, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5115797107338558, | |
| "eval_loss": 2.573380947113037, | |
| "eval_runtime": 4.4435, | |
| "eval_samples_per_second": 1133.781, | |
| "eval_steps_per_second": 2.25, | |
| "step": 2478 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 0.0003125, | |
| "loss": 2.5157, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "learning_rate": 0.000375, | |
| "loss": 2.4165, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5211369966209564, | |
| "eval_loss": 2.4666714668273926, | |
| "eval_runtime": 4.525, | |
| "eval_samples_per_second": 1113.377, | |
| "eval_steps_per_second": 2.21, | |
| "step": 3304 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 0.00043750000000000006, | |
| "loss": 2.3502, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 0.0005, | |
| "loss": 2.2892, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5287937658050783, | |
| "eval_loss": 2.394850969314575, | |
| "eval_runtime": 4.6178, | |
| "eval_samples_per_second": 1090.991, | |
| "eval_steps_per_second": 2.166, | |
| "step": 4130 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 0.0005625000000000001, | |
| "loss": 2.2315, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5337701187510354, | |
| "eval_loss": 2.344557523727417, | |
| "eval_runtime": 4.6005, | |
| "eval_samples_per_second": 1095.088, | |
| "eval_steps_per_second": 2.174, | |
| "step": 4956 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 0.000625, | |
| "loss": 2.2096, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.66, | |
| "learning_rate": 0.0006875, | |
| "loss": 2.1587, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5373570314429236, | |
| "eval_loss": 2.3208389282226562, | |
| "eval_runtime": 4.4883, | |
| "eval_samples_per_second": 1122.477, | |
| "eval_steps_per_second": 2.228, | |
| "step": 5782 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "learning_rate": 0.00075, | |
| "loss": 2.139, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.87, | |
| "learning_rate": 0.0008125, | |
| "loss": 2.1253, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5394279069622053, | |
| "eval_loss": 2.3043758869171143, | |
| "eval_runtime": 4.5793, | |
| "eval_samples_per_second": 1100.17, | |
| "eval_steps_per_second": 2.184, | |
| "step": 6608 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "learning_rate": 0.0008750000000000001, | |
| "loss": 2.0858, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.5403687877641734, | |
| "eval_loss": 2.2939975261688232, | |
| "eval_runtime": 4.5389, | |
| "eval_samples_per_second": 1109.953, | |
| "eval_steps_per_second": 2.203, | |
| "step": 7434 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "learning_rate": 0.0009375, | |
| "loss": 2.0892, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 9.69, | |
| "learning_rate": 0.001, | |
| "loss": 2.0556, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5416644525230297, | |
| "eval_loss": 2.2877631187438965, | |
| "eval_runtime": 4.6002, | |
| "eval_samples_per_second": 1095.162, | |
| "eval_steps_per_second": 2.174, | |
| "step": 8260 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 8260, | |
| "total_flos": 2562796651806720.0, | |
| "train_loss": 2.5528629182903297, | |
| "train_runtime": 747.7642, | |
| "train_samples_per_second": 706.265, | |
| "train_steps_per_second": 11.046 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 8260, | |
| "num_train_epochs": 10, | |
| "save_steps": 2000, | |
| "total_flos": 2562796651806720.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |