| { | |
| "best_metric": 0.03953862935304642, | |
| "best_model_checkpoint": "/workspace/training_output/checkpoint-6704", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 8380, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 1.8806682577565632e-05, | |
| "loss": 0.7239, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 1.7613365155131266e-05, | |
| "loss": 0.2996, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.6420047732696897e-05, | |
| "loss": 0.2045, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9733727810650887, | |
| "eval_loss": 0.11561588197946548, | |
| "eval_runtime": 250.554, | |
| "eval_samples_per_second": 9.443, | |
| "eval_steps_per_second": 1.181, | |
| "step": 1676 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 1.5226730310262532e-05, | |
| "loss": 0.1485, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 1.4033412887828164e-05, | |
| "loss": 0.141, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 1.2840095465393797e-05, | |
| "loss": 0.1241, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9809805579036348, | |
| "eval_loss": 0.07752275466918945, | |
| "eval_runtime": 191.0466, | |
| "eval_samples_per_second": 12.384, | |
| "eval_steps_per_second": 1.549, | |
| "step": 3352 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 1.1646778042959427e-05, | |
| "loss": 0.1247, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 1.045346062052506e-05, | |
| "loss": 0.0963, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 9.260143198090693e-06, | |
| "loss": 0.1046, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 8.066825775656326e-06, | |
| "loss": 0.1048, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9873203719357565, | |
| "eval_loss": 0.055058836936950684, | |
| "eval_runtime": 165.8573, | |
| "eval_samples_per_second": 14.265, | |
| "eval_steps_per_second": 1.785, | |
| "step": 5028 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 6.873508353221957e-06, | |
| "loss": 0.0834, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 5.68019093078759e-06, | |
| "loss": 0.0982, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 4.486873508353222e-06, | |
| "loss": 0.0675, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9915469146238377, | |
| "eval_loss": 0.03953862935304642, | |
| "eval_runtime": 267.8931, | |
| "eval_samples_per_second": 8.832, | |
| "eval_steps_per_second": 1.105, | |
| "step": 6704 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "learning_rate": 3.293556085918855e-06, | |
| "loss": 0.0724, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 2.100238663484487e-06, | |
| "loss": 0.0662, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "learning_rate": 9.069212410501194e-07, | |
| "loss": 0.0609, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9911242603550295, | |
| "eval_loss": 0.039820414036512375, | |
| "eval_runtime": 179.8609, | |
| "eval_samples_per_second": 13.155, | |
| "eval_steps_per_second": 1.646, | |
| "step": 8380 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 8380, | |
| "total_flos": 5.19422646293591e+18, | |
| "train_loss": 0.1531206039938551, | |
| "train_runtime": 7251.6972, | |
| "train_samples_per_second": 9.243, | |
| "train_steps_per_second": 1.156 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 8380, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 5.19422646293591e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |