| { | |
| "best_metric": 0.6232164195970928, | |
| "best_model_checkpoint": "./fine-tune/roberta-base/cola/checkpoint-5350", | |
| "epoch": 10.0, | |
| "global_step": 5350, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 1.9288128852654602e-05, | |
| "loss": 0.5497, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.5504242181777954, | |
| "eval_matthews_correlation": 0.4613462408273676, | |
| "eval_runtime": 1.7936, | |
| "eval_samples_per_second": 581.514, | |
| "eval_steps_per_second": 73.038, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 1.729966196062836e-05, | |
| "loss": 0.3786, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.48504745960235596, | |
| "eval_matthews_correlation": 0.5470285194988502, | |
| "eval_runtime": 1.8903, | |
| "eval_samples_per_second": 551.77, | |
| "eval_steps_per_second": 69.302, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 1.531119506860211e-05, | |
| "loss": 0.2733, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.5036450624465942, | |
| "eval_matthews_correlation": 0.5791579814417055, | |
| "eval_runtime": 2.2071, | |
| "eval_samples_per_second": 472.565, | |
| "eval_steps_per_second": 59.354, | |
| "step": 1605 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 1.3322728176575861e-05, | |
| "loss": 0.2204, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.5532266497612, | |
| "eval_matthews_correlation": 0.6138670730002377, | |
| "eval_runtime": 1.8422, | |
| "eval_samples_per_second": 566.18, | |
| "eval_steps_per_second": 71.112, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 1.1334261284549613e-05, | |
| "loss": 0.164, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.9516334533691406, | |
| "eval_matthews_correlation": 0.5934048733741295, | |
| "eval_runtime": 1.8822, | |
| "eval_samples_per_second": 554.136, | |
| "eval_steps_per_second": 69.599, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 9.345794392523365e-06, | |
| "loss": 0.1351, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.9050658345222473, | |
| "eval_matthews_correlation": 0.5754342635449895, | |
| "eval_runtime": 1.846, | |
| "eval_samples_per_second": 565.019, | |
| "eval_steps_per_second": 70.966, | |
| "step": 3210 | |
| }, | |
| { | |
| "epoch": 6.54, | |
| "learning_rate": 7.357327500497118e-06, | |
| "loss": 0.1065, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.9006233215332031, | |
| "eval_matthews_correlation": 0.6161067903576496, | |
| "eval_runtime": 1.877, | |
| "eval_samples_per_second": 555.681, | |
| "eval_steps_per_second": 69.793, | |
| "step": 3745 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 5.36886060847087e-06, | |
| "loss": 0.0874, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.9457036852836609, | |
| "eval_matthews_correlation": 0.6157170770507245, | |
| "eval_runtime": 1.8772, | |
| "eval_samples_per_second": 555.609, | |
| "eval_steps_per_second": 69.784, | |
| "step": 4280 | |
| }, | |
| { | |
| "epoch": 8.41, | |
| "learning_rate": 3.3803937164446215e-06, | |
| "loss": 0.0579, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 1.0371503829956055, | |
| "eval_matthews_correlation": 0.6006530974238766, | |
| "eval_runtime": 1.8847, | |
| "eval_samples_per_second": 553.403, | |
| "eval_steps_per_second": 69.507, | |
| "step": 4815 | |
| }, | |
| { | |
| "epoch": 9.35, | |
| "learning_rate": 1.3919268244183738e-06, | |
| "loss": 0.0451, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.0570732355117798, | |
| "eval_matthews_correlation": 0.6232164195970928, | |
| "eval_runtime": 1.8418, | |
| "eval_samples_per_second": 566.299, | |
| "eval_steps_per_second": 71.127, | |
| "step": 5350 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 5350, | |
| "total_flos": 5624656585958400.0, | |
| "train_loss": 0.191466952528909, | |
| "train_runtime": 617.5978, | |
| "train_samples_per_second": 138.456, | |
| "train_steps_per_second": 8.663 | |
| } | |
| ], | |
| "max_steps": 5350, | |
| "num_train_epochs": 10, | |
| "total_flos": 5624656585958400.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |