| { | |
| "best_metric": 0.5966597659617285, | |
| "best_model_checkpoint": "test-klue/ynat/run-0/checkpoint-2764", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 2764, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 1.549617863520091e-06, | |
| "loss": 1.0402, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1.2073877699865019e-06, | |
| "loss": 0.8825, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_f1": 0.5922490852804443, | |
| "eval_loss": 0.8286302089691162, | |
| "eval_runtime": 9.5242, | |
| "eval_samples_per_second": 248.734, | |
| "eval_steps_per_second": 1.995, | |
| "step": 1382 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 8.65157676452913e-07, | |
| "loss": 0.8276, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 5.22927582919324e-07, | |
| "loss": 0.7615, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 1.8069748938573499e-07, | |
| "loss": 0.7606, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_f1": 0.5966597659617285, | |
| "eval_loss": 0.8385059237480164, | |
| "eval_runtime": 9.5681, | |
| "eval_samples_per_second": 247.594, | |
| "eval_steps_per_second": 1.986, | |
| "step": 2764 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2764, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "total_flos": 417767579078724.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 1.8850033551830082e-06, | |
| "num_train_epochs": 2, | |
| "per_device_train_batch_size": 4, | |
| "seed": 19 | |
| } | |
| } | |