| { | |
| "best_metric": 0.5127451929115474, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-5/checkpoint-8552", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 8552, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 1.6879780082730344e-05, | |
| "loss": 0.5883, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 1.583160695290357e-05, | |
| "loss": 0.5765, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1.47834338230768e-05, | |
| "loss": 0.5618, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 1.3735260693250027e-05, | |
| "loss": 0.5987, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.5104408264160156, | |
| "eval_matthews_correlation": 0.447681158536829, | |
| "eval_runtime": 0.6946, | |
| "eval_samples_per_second": 1501.589, | |
| "eval_steps_per_second": 95.019, | |
| "step": 2138 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 1.2687087563423256e-05, | |
| "loss": 0.4436, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 1.1638914433596482e-05, | |
| "loss": 0.4754, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 1.0590741303769709e-05, | |
| "loss": 0.4799, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 9.542568173942937e-06, | |
| "loss": 0.5073, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.7683575749397278, | |
| "eval_matthews_correlation": 0.4766173298807751, | |
| "eval_runtime": 0.5113, | |
| "eval_samples_per_second": 2039.725, | |
| "eval_steps_per_second": 129.072, | |
| "step": 4276 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 8.494395044116164e-06, | |
| "loss": 0.411, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 7.446221914289393e-06, | |
| "loss": 0.3135, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 6.398048784462619e-06, | |
| "loss": 0.319, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 5.349875654635847e-06, | |
| "loss": 0.3068, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.9311069250106812, | |
| "eval_matthews_correlation": 0.4971890871020268, | |
| "eval_runtime": 0.6011, | |
| "eval_samples_per_second": 1735.215, | |
| "eval_steps_per_second": 109.803, | |
| "step": 6414 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 4.301702524809075e-06, | |
| "loss": 0.3312, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 3.253529394982302e-06, | |
| "loss": 0.2016, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 2.2053562651555294e-06, | |
| "loss": 0.1942, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 1.157183135328757e-06, | |
| "loss": 0.2323, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 1.0901000550198435e-07, | |
| "loss": 0.1824, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.1264679431915283, | |
| "eval_matthews_correlation": 0.5127451929115474, | |
| "eval_runtime": 0.5395, | |
| "eval_samples_per_second": 1933.135, | |
| "eval_steps_per_second": 122.327, | |
| "step": 8552 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 8552, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "total_flos": 138903192682500.0, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 1.7927953212557118e-05, | |
| "num_train_epochs": 4, | |
| "per_device_train_batch_size": 4, | |
| "seed": 35 | |
| } | |
| } | |