| { | |
| "best_metric": 0.39150252536899116, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-1/checkpoint-4276", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 6414, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 1.3464482832056169e-06, | |
| "loss": 0.6152, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 1.2326126150279351e-06, | |
| "loss": 0.5757, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1.1187769468502538e-06, | |
| "loss": 0.5878, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 1.0049412786725722e-06, | |
| "loss": 0.5571, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.5580433011054993, | |
| "eval_matthews_correlation": 0.20334509429779915, | |
| "eval_runtime": 0.5344, | |
| "eval_samples_per_second": 1951.85, | |
| "eval_steps_per_second": 123.511, | |
| "step": 2138 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 8.911056104948908e-07, | |
| "loss": 0.5247, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 7.772699423172092e-07, | |
| "loss": 0.516, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 6.634342741395277e-07, | |
| "loss": 0.518, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 5.495986059618463e-07, | |
| "loss": 0.4983, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.5595452189445496, | |
| "eval_matthews_correlation": 0.39150252536899116, | |
| "eval_runtime": 0.5424, | |
| "eval_samples_per_second": 1922.955, | |
| "eval_steps_per_second": 121.683, | |
| "step": 4276 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 4.357629377841647e-07, | |
| "loss": 0.4985, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 3.219272696064833e-07, | |
| "loss": 0.4795, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 2.080916014288018e-07, | |
| "loss": 0.4954, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 9.425593325112028e-08, | |
| "loss": 0.4859, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.5893857479095459, | |
| "eval_matthews_correlation": 0.3883727168442722, | |
| "eval_runtime": 0.5651, | |
| "eval_samples_per_second": 1845.532, | |
| "eval_steps_per_second": 116.783, | |
| "step": 6414 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 6414, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 97693154156472.0, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 1.4602839513832982e-06, | |
| "num_train_epochs": 3, | |
| "per_device_train_batch_size": 4, | |
| "seed": 39 | |
| } | |
| } | |