| { | |
| "best_metric": 0.90164, | |
| "best_model_checkpoint": "../../checkpoint/imdb/roberta-base/checkpoint-14858", | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 15640, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.88836, | |
| "eval_loss": 0.2657841444015503, | |
| "eval_runtime": 31.5408, | |
| "eval_samples_per_second": 792.625, | |
| "eval_steps_per_second": 3.107, | |
| "step": 782 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 4.680306905370844e-05, | |
| "loss": 0.3125, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.89232, | |
| "eval_loss": 0.25732555985450745, | |
| "eval_runtime": 31.3388, | |
| "eval_samples_per_second": 797.734, | |
| "eval_steps_per_second": 3.127, | |
| "step": 1564 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 4.360613810741688e-05, | |
| "loss": 0.189, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8944, | |
| "eval_loss": 0.339549720287323, | |
| "eval_runtime": 31.3008, | |
| "eval_samples_per_second": 798.703, | |
| "eval_steps_per_second": 3.131, | |
| "step": 2346 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 4.040920716112532e-05, | |
| "loss": 0.1284, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.89492, | |
| "eval_loss": 0.4028278887271881, | |
| "eval_runtime": 31.3783, | |
| "eval_samples_per_second": 796.729, | |
| "eval_steps_per_second": 3.123, | |
| "step": 3128 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.88992, | |
| "eval_loss": 0.4815801978111267, | |
| "eval_runtime": 31.3726, | |
| "eval_samples_per_second": 796.873, | |
| "eval_steps_per_second": 3.124, | |
| "step": 3910 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "learning_rate": 3.721227621483376e-05, | |
| "loss": 0.09, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.89704, | |
| "eval_loss": 0.40239471197128296, | |
| "eval_runtime": 31.3511, | |
| "eval_samples_per_second": 797.42, | |
| "eval_steps_per_second": 3.126, | |
| "step": 4692 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "learning_rate": 3.40153452685422e-05, | |
| "loss": 0.0632, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.89572, | |
| "eval_loss": 0.4902121424674988, | |
| "eval_runtime": 31.3539, | |
| "eval_samples_per_second": 797.35, | |
| "eval_steps_per_second": 3.126, | |
| "step": 5474 | |
| }, | |
| { | |
| "epoch": 7.67, | |
| "learning_rate": 3.081841432225064e-05, | |
| "loss": 0.049, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.89104, | |
| "eval_loss": 0.5230616927146912, | |
| "eval_runtime": 31.2816, | |
| "eval_samples_per_second": 799.192, | |
| "eval_steps_per_second": 3.133, | |
| "step": 6256 | |
| }, | |
| { | |
| "epoch": 8.95, | |
| "learning_rate": 2.7621483375959077e-05, | |
| "loss": 0.0426, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.89764, | |
| "eval_loss": 0.5609263777732849, | |
| "eval_runtime": 31.2704, | |
| "eval_samples_per_second": 799.477, | |
| "eval_steps_per_second": 3.134, | |
| "step": 7038 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.89744, | |
| "eval_loss": 0.590167760848999, | |
| "eval_runtime": 31.3811, | |
| "eval_samples_per_second": 796.658, | |
| "eval_steps_per_second": 3.123, | |
| "step": 7820 | |
| }, | |
| { | |
| "epoch": 10.23, | |
| "learning_rate": 2.442455242966752e-05, | |
| "loss": 0.0343, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.89904, | |
| "eval_loss": 0.5456650853157043, | |
| "eval_runtime": 31.3005, | |
| "eval_samples_per_second": 798.71, | |
| "eval_steps_per_second": 3.131, | |
| "step": 8602 | |
| }, | |
| { | |
| "epoch": 11.51, | |
| "learning_rate": 2.122762148337596e-05, | |
| "loss": 0.0282, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8972, | |
| "eval_loss": 0.5991714596748352, | |
| "eval_runtime": 31.3289, | |
| "eval_samples_per_second": 797.985, | |
| "eval_steps_per_second": 3.128, | |
| "step": 9384 | |
| }, | |
| { | |
| "epoch": 12.79, | |
| "learning_rate": 1.80306905370844e-05, | |
| "loss": 0.0237, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8974, | |
| "eval_loss": 0.5605877041816711, | |
| "eval_runtime": 31.2841, | |
| "eval_samples_per_second": 799.128, | |
| "eval_steps_per_second": 3.133, | |
| "step": 10166 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.89972, | |
| "eval_loss": 0.5708814263343811, | |
| "eval_runtime": 31.4954, | |
| "eval_samples_per_second": 793.767, | |
| "eval_steps_per_second": 3.112, | |
| "step": 10948 | |
| }, | |
| { | |
| "epoch": 14.07, | |
| "learning_rate": 1.483375959079284e-05, | |
| "loss": 0.021, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8992, | |
| "eval_loss": 0.616736888885498, | |
| "eval_runtime": 31.4351, | |
| "eval_samples_per_second": 795.29, | |
| "eval_steps_per_second": 3.118, | |
| "step": 11730 | |
| }, | |
| { | |
| "epoch": 15.35, | |
| "learning_rate": 1.163682864450128e-05, | |
| "loss": 0.019, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9014, | |
| "eval_loss": 0.5849418640136719, | |
| "eval_runtime": 31.1725, | |
| "eval_samples_per_second": 801.99, | |
| "eval_steps_per_second": 3.144, | |
| "step": 12512 | |
| }, | |
| { | |
| "epoch": 16.62, | |
| "learning_rate": 8.439897698209718e-06, | |
| "loss": 0.0147, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.90072, | |
| "eval_loss": 0.60484778881073, | |
| "eval_runtime": 31.2733, | |
| "eval_samples_per_second": 799.404, | |
| "eval_steps_per_second": 3.134, | |
| "step": 13294 | |
| }, | |
| { | |
| "epoch": 17.9, | |
| "learning_rate": 5.242966751918159e-06, | |
| "loss": 0.014, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.89868, | |
| "eval_loss": 0.6006963849067688, | |
| "eval_runtime": 31.2309, | |
| "eval_samples_per_second": 800.488, | |
| "eval_steps_per_second": 3.138, | |
| "step": 14076 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.90164, | |
| "eval_loss": 0.5982226729393005, | |
| "eval_runtime": 31.229, | |
| "eval_samples_per_second": 800.537, | |
| "eval_steps_per_second": 3.138, | |
| "step": 14858 | |
| }, | |
| { | |
| "epoch": 19.18, | |
| "learning_rate": 2.0460358056265987e-06, | |
| "loss": 0.0129, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9014, | |
| "eval_loss": 0.599052906036377, | |
| "eval_runtime": 31.2461, | |
| "eval_samples_per_second": 800.099, | |
| "eval_steps_per_second": 3.136, | |
| "step": 15640 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 15640, | |
| "total_flos": 3.288888192e+16, | |
| "train_loss": 0.0671754312637212, | |
| "train_runtime": 2479.9785, | |
| "train_samples_per_second": 201.615, | |
| "train_steps_per_second": 6.307 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 15640, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 3.288888192e+16, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |