| { | |
| "best_metric": 0.9974986599964266, | |
| "best_model_checkpoint": "./DISTEMIST-huggingface-med/run-4/checkpoint-10632", | |
| "epoch": 9.53542600896861, | |
| "global_step": 10632, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 3.718581391614352e-05, | |
| "loss": 0.1783, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_f1": 0.7232514486248266, | |
| "eval_loss": 0.06977316737174988, | |
| "eval_runtime": 17.2132, | |
| "eval_samples_per_second": 517.974, | |
| "eval_steps_per_second": 64.776, | |
| "step": 886 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 3.544000105623068e-05, | |
| "loss": 0.1046, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 3.3694188196317836e-05, | |
| "loss": 0.0712, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "eval_f1": 0.8131887201735358, | |
| "eval_loss": 0.039787087589502335, | |
| "eval_runtime": 17.176, | |
| "eval_samples_per_second": 519.097, | |
| "eval_steps_per_second": 64.916, | |
| "step": 1772 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 3.1948375336404995e-05, | |
| "loss": 0.0586, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 3.0202562476492158e-05, | |
| "loss": 0.0456, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "eval_f1": 0.8951260803706673, | |
| "eval_loss": 0.020952457562088966, | |
| "eval_runtime": 17.1751, | |
| "eval_samples_per_second": 519.123, | |
| "eval_steps_per_second": 64.919, | |
| "step": 2658 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 2.8456749616579317e-05, | |
| "loss": 0.0318, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 2.6710936756666475e-05, | |
| "loss": 0.0277, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "eval_f1": 0.9308534306309513, | |
| "eval_loss": 0.011393814347684383, | |
| "eval_runtime": 17.3437, | |
| "eval_samples_per_second": 514.078, | |
| "eval_steps_per_second": 64.289, | |
| "step": 3544 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 2.496512389675363e-05, | |
| "loss": 0.0192, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "eval_f1": 0.9564366998577525, | |
| "eval_loss": 0.008351404219865799, | |
| "eval_runtime": 17.1902, | |
| "eval_samples_per_second": 518.667, | |
| "eval_steps_per_second": 64.862, | |
| "step": 4430 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 2.321931103684079e-05, | |
| "loss": 0.0158, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 2.147349817692795e-05, | |
| "loss": 0.0133, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "eval_f1": 0.9696861943283848, | |
| "eval_loss": 0.00492077087983489, | |
| "eval_runtime": 17.1702, | |
| "eval_samples_per_second": 519.271, | |
| "eval_steps_per_second": 64.938, | |
| "step": 5316 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 1.9727685317015108e-05, | |
| "loss": 0.0108, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 1.7981872457102267e-05, | |
| "loss": 0.0079, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "eval_f1": 0.9758359340169416, | |
| "eval_loss": 0.004112700931727886, | |
| "eval_runtime": 17.19, | |
| "eval_samples_per_second": 518.675, | |
| "eval_steps_per_second": 64.863, | |
| "step": 6202 | |
| }, | |
| { | |
| "epoch": 5.83, | |
| "learning_rate": 1.6236059597189426e-05, | |
| "loss": 0.0065, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "learning_rate": 1.4490246737276583e-05, | |
| "loss": 0.0052, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 6.36, | |
| "eval_f1": 0.9876058849754793, | |
| "eval_loss": 0.0018256960902363062, | |
| "eval_runtime": 17.1652, | |
| "eval_samples_per_second": 519.424, | |
| "eval_steps_per_second": 64.957, | |
| "step": 7088 | |
| }, | |
| { | |
| "epoch": 6.73, | |
| "learning_rate": 1.2744433877363742e-05, | |
| "loss": 0.0046, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 7.15, | |
| "eval_f1": 0.9915080003575579, | |
| "eval_loss": 0.001805759035050869, | |
| "eval_runtime": 17.2049, | |
| "eval_samples_per_second": 518.225, | |
| "eval_steps_per_second": 64.807, | |
| "step": 7974 | |
| }, | |
| { | |
| "epoch": 7.17, | |
| "learning_rate": 1.0998621017450902e-05, | |
| "loss": 0.0032, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 7.62, | |
| "learning_rate": 9.25280815753806e-06, | |
| "loss": 0.0034, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 7.95, | |
| "eval_f1": 0.9946399857066286, | |
| "eval_loss": 0.0007319775759242475, | |
| "eval_runtime": 17.3742, | |
| "eval_samples_per_second": 513.174, | |
| "eval_steps_per_second": 64.176, | |
| "step": 8860 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 7.5069952976252175e-06, | |
| "loss": 0.0026, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 8.52, | |
| "learning_rate": 5.761182437712377e-06, | |
| "loss": 0.0011, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 8.74, | |
| "eval_f1": 0.9957135202714771, | |
| "eval_loss": 0.0006867917836643755, | |
| "eval_runtime": 17.1265, | |
| "eval_samples_per_second": 520.596, | |
| "eval_steps_per_second": 65.104, | |
| "step": 9746 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "learning_rate": 4.015369577799535e-06, | |
| "loss": 0.002, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 9.42, | |
| "learning_rate": 2.269556717886694e-06, | |
| "loss": 0.0014, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 9.54, | |
| "eval_f1": 0.9974986599964266, | |
| "eval_loss": 0.0003464366600383073, | |
| "eval_runtime": 17.2334, | |
| "eval_samples_per_second": 517.369, | |
| "eval_steps_per_second": 64.7, | |
| "step": 10632 | |
| } | |
| ], | |
| "max_steps": 11150, | |
| "num_train_epochs": 10, | |
| "total_flos": 2805348170687904.0, | |
| "trial_name": null, | |
| "trial_params": { | |
| "adam_epsilon": 7.614260573309694e-10, | |
| "learning_rate": 3.890369377029776e-05, | |
| "num_train_epochs": 10, | |
| "per_device_eval_batch_size": 8, | |
| "per_device_train_batch_size": 8, | |
| "seed": 326, | |
| "warmup_steps": 8, | |
| "weight_decay": 2.2273844202742368e-07 | |
| } | |
| } | |