| { | |
| "best_metric": 0.6487034559249878, | |
| "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_qnli/checkpoint-1638", | |
| "epoch": 7.0, | |
| "global_step": 5733, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.6754, | |
| "step": 819 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6177924217462932, | |
| "eval_loss": 0.6491163372993469, | |
| "eval_runtime": 8.8255, | |
| "eval_samples_per_second": 619.001, | |
| "eval_steps_per_second": 4.872, | |
| "step": 819 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6369, | |
| "step": 1638 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6093721398498994, | |
| "eval_loss": 0.6487034559249878, | |
| "eval_runtime": 8.8229, | |
| "eval_samples_per_second": 619.186, | |
| "eval_steps_per_second": 4.874, | |
| "step": 1638 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7001221001221e-05, | |
| "loss": 0.6125, | |
| "step": 2457 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6088229910305692, | |
| "eval_loss": 0.6555132269859314, | |
| "eval_runtime": 8.8047, | |
| "eval_samples_per_second": 620.464, | |
| "eval_steps_per_second": 4.884, | |
| "step": 2457 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.6001221001221e-05, | |
| "loss": 0.5942, | |
| "step": 3276 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6027823540179389, | |
| "eval_loss": 0.6647323369979858, | |
| "eval_runtime": 8.7921, | |
| "eval_samples_per_second": 621.356, | |
| "eval_steps_per_second": 4.891, | |
| "step": 3276 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5001221001221004e-05, | |
| "loss": 0.5805, | |
| "step": 4095 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5934468240893283, | |
| "eval_loss": 0.6735221147537231, | |
| "eval_runtime": 8.6724, | |
| "eval_samples_per_second": 629.928, | |
| "eval_steps_per_second": 4.958, | |
| "step": 4095 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.400244200244201e-05, | |
| "loss": 0.5689, | |
| "step": 4914 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5978400146439685, | |
| "eval_loss": 0.6893225312232971, | |
| "eval_runtime": 8.8418, | |
| "eval_samples_per_second": 617.861, | |
| "eval_steps_per_second": 4.863, | |
| "step": 4914 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.3002442002442004e-05, | |
| "loss": 0.5587, | |
| "step": 5733 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.589602782354018, | |
| "eval_loss": 0.7054734230041504, | |
| "eval_runtime": 8.8696, | |
| "eval_samples_per_second": 615.924, | |
| "eval_steps_per_second": 4.848, | |
| "step": 5733 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "step": 5733, | |
| "total_flos": 2.2988974586855424e+16, | |
| "train_loss": 0.603873611681752, | |
| "train_runtime": 4016.5926, | |
| "train_samples_per_second": 1303.879, | |
| "train_steps_per_second": 10.195 | |
| } | |
| ], | |
| "max_steps": 40950, | |
| "num_train_epochs": 50, | |
| "total_flos": 2.2988974586855424e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |