| { | |
| "best_metric": 0.657294511795044, | |
| "best_model_checkpoint": "hBERTv2_new_pretrain_w_init__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": 3.9200000000000004e-05, | |
| "loss": 0.6809, | |
| "step": 819 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6020501555921655, | |
| "eval_loss": 0.6640335321426392, | |
| "eval_runtime": 9.3726, | |
| "eval_samples_per_second": 582.869, | |
| "eval_steps_per_second": 4.588, | |
| "step": 819 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.8400000000000005e-05, | |
| "loss": 0.6539, | |
| "step": 1638 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5958264689730917, | |
| "eval_loss": 0.657294511795044, | |
| "eval_runtime": 9.3593, | |
| "eval_samples_per_second": 583.694, | |
| "eval_steps_per_second": 4.594, | |
| "step": 1638 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.76e-05, | |
| "loss": 0.6358, | |
| "step": 2457 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5941790225151016, | |
| "eval_loss": 0.6694244146347046, | |
| "eval_runtime": 9.3585, | |
| "eval_samples_per_second": 583.748, | |
| "eval_steps_per_second": 4.595, | |
| "step": 2457 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 3.680000000000001e-05, | |
| "loss": 0.6313, | |
| "step": 3276 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5789859051803039, | |
| "eval_loss": 0.6897775530815125, | |
| "eval_runtime": 9.3577, | |
| "eval_samples_per_second": 583.797, | |
| "eval_steps_per_second": 4.595, | |
| "step": 3276 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.6293, | |
| "step": 4095 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5879553358960278, | |
| "eval_loss": 0.6801490783691406, | |
| "eval_runtime": 9.3527, | |
| "eval_samples_per_second": 584.106, | |
| "eval_steps_per_second": 4.598, | |
| "step": 4095 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3.52e-05, | |
| "loss": 0.6253, | |
| "step": 4914 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5925315760571115, | |
| "eval_loss": 0.6828794479370117, | |
| "eval_runtime": 9.366, | |
| "eval_samples_per_second": 583.282, | |
| "eval_steps_per_second": 4.591, | |
| "step": 4914 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 3.44e-05, | |
| "loss": 0.6166, | |
| "step": 5733 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5916163280248947, | |
| "eval_loss": 0.6865497827529907, | |
| "eval_runtime": 9.3522, | |
| "eval_samples_per_second": 584.138, | |
| "eval_steps_per_second": 4.598, | |
| "step": 5733 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "step": 5733, | |
| "total_flos": 1.0729602688011469e+17, | |
| "train_loss": 0.639016756103984, | |
| "train_runtime": 3839.7768, | |
| "train_samples_per_second": 1363.92, | |
| "train_steps_per_second": 10.665 | |
| } | |
| ], | |
| "max_steps": 40950, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.0729602688011469e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |