| { | |
| "best_metric": 0.9015151515151515, | |
| "best_model_checkpoint": "dit-base-rvlcdip-finetuned-grp-actual/checkpoint-112", | |
| "epoch": 6.72, | |
| "eval_steps": 500, | |
| "global_step": 126, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.8692, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_accuracy": 0.8560606060606061, | |
| "eval_loss": 0.6972441673278809, | |
| "eval_runtime": 15.3616, | |
| "eval_samples_per_second": 17.186, | |
| "eval_steps_per_second": 0.586, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 4.690265486725664e-05, | |
| "loss": 0.7995, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 4.247787610619469e-05, | |
| "loss": 0.7348, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_accuracy": 0.8598484848484849, | |
| "eval_loss": 0.6350475549697876, | |
| "eval_runtime": 16.1781, | |
| "eval_samples_per_second": 16.318, | |
| "eval_steps_per_second": 0.556, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 3.8053097345132744e-05, | |
| "loss": 0.6855, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 3.3628318584070804e-05, | |
| "loss": 0.6655, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.8712121212121212, | |
| "eval_loss": 0.533867359161377, | |
| "eval_runtime": 15.3125, | |
| "eval_samples_per_second": 17.241, | |
| "eval_steps_per_second": 0.588, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 2.9203539823008852e-05, | |
| "loss": 0.7083, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 2.4778761061946905e-05, | |
| "loss": 0.7167, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8901515151515151, | |
| "eval_loss": 0.5045967102050781, | |
| "eval_runtime": 15.2445, | |
| "eval_samples_per_second": 17.318, | |
| "eval_steps_per_second": 0.59, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 2.0353982300884957e-05, | |
| "loss": 0.6926, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 1.592920353982301e-05, | |
| "loss": 0.694, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.5026077628135681, | |
| "eval_runtime": 15.2119, | |
| "eval_samples_per_second": 17.355, | |
| "eval_steps_per_second": 0.592, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 1.1504424778761062e-05, | |
| "loss": 0.6366, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.87, | |
| "learning_rate": 7.079646017699115e-06, | |
| "loss": 0.6638, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "eval_accuracy": 0.9015151515151515, | |
| "eval_loss": 0.46005359292030334, | |
| "eval_runtime": 15.7633, | |
| "eval_samples_per_second": 16.748, | |
| "eval_steps_per_second": 0.571, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 2.6548672566371683e-06, | |
| "loss": 0.6618, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "eval_accuracy": 0.8977272727272727, | |
| "eval_loss": 0.45817553997039795, | |
| "eval_runtime": 17.722, | |
| "eval_samples_per_second": 14.897, | |
| "eval_steps_per_second": 0.508, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 126, | |
| "total_flos": 1.2388649195611423e+18, | |
| "train_loss": 0.7077309669010223, | |
| "train_runtime": 1389.7418, | |
| "train_samples_per_second": 11.968, | |
| "train_steps_per_second": 0.091 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 126, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "total_flos": 1.2388649195611423e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |