| { | |
| "best_metric": 0.9128113879003559, | |
| "best_model_checkpoint": "./SocalDisner-strong/run-1/checkpoint-4430", | |
| "epoch": 7.122186495176849, | |
| "global_step": 4430, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 2.3346599605377065e-05, | |
| "loss": 0.0902, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "eval_f1": 0.9015075376884422, | |
| "eval_loss": 0.04874825105071068, | |
| "eval_runtime": 13.1719, | |
| "eval_samples_per_second": 189.797, | |
| "eval_steps_per_second": 23.763, | |
| "step": 886 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 2.963685146642776e-05, | |
| "loss": 0.0494, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 2.5909883223667728e-05, | |
| "loss": 0.0284, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "eval_f1": 0.9087016748520338, | |
| "eval_loss": 0.049770649522542953, | |
| "eval_runtime": 13.2365, | |
| "eval_samples_per_second": 188.871, | |
| "eval_steps_per_second": 23.647, | |
| "step": 1772 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 2.21829149809077e-05, | |
| "loss": 0.0141, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 1.845594673814767e-05, | |
| "loss": 0.0075, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "eval_f1": 0.9092276830491474, | |
| "eval_loss": 0.0691143274307251, | |
| "eval_runtime": 13.1876, | |
| "eval_samples_per_second": 189.573, | |
| "eval_steps_per_second": 23.734, | |
| "step": 2658 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "learning_rate": 1.472897849538764e-05, | |
| "loss": 0.0029, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.63, | |
| "learning_rate": 1.1002010252627608e-05, | |
| "loss": 0.0024, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "eval_f1": 0.9093625498007968, | |
| "eval_loss": 0.06731252372264862, | |
| "eval_runtime": 13.216, | |
| "eval_samples_per_second": 189.165, | |
| "eval_steps_per_second": 23.683, | |
| "step": 3544 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "learning_rate": 7.275042009867579e-06, | |
| "loss": 0.0017, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.12, | |
| "eval_f1": 0.9128113879003559, | |
| "eval_loss": 0.07908429950475693, | |
| "eval_runtime": 13.1476, | |
| "eval_samples_per_second": 190.149, | |
| "eval_steps_per_second": 23.807, | |
| "step": 4430 | |
| } | |
| ], | |
| "max_steps": 4976, | |
| "num_train_epochs": 8, | |
| "total_flos": 1557691318801926.0, | |
| "trial_name": null, | |
| "trial_params": { | |
| "adam_epsilon": 1.0695999090544144e-10, | |
| "learning_rate": 3.198484145936658e-05, | |
| "num_train_epochs": 8, | |
| "per_device_eval_batch_size": 8, | |
| "per_device_train_batch_size": 32, | |
| "seed": 326, | |
| "warmup_steps": 685, | |
| "weight_decay": 7.355540557094741e-05 | |
| } | |
| } | |