| { | |
| "best_metric": 0.8502932169961132, | |
| "best_model_checkpoint": "test-klue/ynat-hps/run-0/checkpoint-4568", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 4568, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 1.5575177105745416e-05, | |
| "loss": 0.7213, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 1.4080438419973303e-05, | |
| "loss": 0.5522, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_f1": 0.8362527412907268, | |
| "eval_loss": 0.6280386447906494, | |
| "eval_runtime": 12.2023, | |
| "eval_samples_per_second": 746.334, | |
| "eval_steps_per_second": 2.95, | |
| "step": 1142 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 1.258569973420119e-05, | |
| "loss": 0.4071, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 1.1090961048429078e-05, | |
| "loss": 0.4276, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_f1": 0.8415413879849883, | |
| "eval_loss": 0.7244753241539001, | |
| "eval_runtime": 12.1177, | |
| "eval_samples_per_second": 751.545, | |
| "eval_steps_per_second": 2.971, | |
| "step": 2284 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 9.596222362656965e-06, | |
| "loss": 0.337, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 8.101483676884852e-06, | |
| "loss": 0.2512, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_f1": 0.8449137319112024, | |
| "eval_loss": 0.8435599207878113, | |
| "eval_runtime": 12.1493, | |
| "eval_samples_per_second": 749.593, | |
| "eval_steps_per_second": 2.963, | |
| "step": 3426 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 6.606744991112739e-06, | |
| "loss": 0.184, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 5.112006305340626e-06, | |
| "loss": 0.1224, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 3.6172676195685133e-06, | |
| "loss": 0.114, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_f1": 0.8502932169961132, | |
| "eval_loss": 0.9528796672821045, | |
| "eval_runtime": 12.1904, | |
| "eval_samples_per_second": 747.06, | |
| "eval_steps_per_second": 2.953, | |
| "step": 4568 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 5710, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 171806898397200.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 1.706991579151753e-05, | |
| "num_train_epochs": 5, | |
| "per_device_train_batch_size": 4, | |
| "seed": 2 | |
| } | |
| } | |