| { | |
| "best_metric": 0.6971034407615662, | |
| "best_model_checkpoint": "hBERTv2_new_pretrain_w_init_48_rte/checkpoint-40", | |
| "epoch": 7.0, | |
| "global_step": 140, | |
| "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": 0.755, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 0.7425682544708252, | |
| "eval_runtime": 0.4916, | |
| "eval_samples_per_second": 563.487, | |
| "eval_steps_per_second": 6.103, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.8400000000000005e-05, | |
| "loss": 0.7127, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.48736462093862815, | |
| "eval_loss": 0.6971034407615662, | |
| "eval_runtime": 0.501, | |
| "eval_samples_per_second": 552.935, | |
| "eval_steps_per_second": 5.988, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.76e-05, | |
| "loss": 0.7149, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5306859205776173, | |
| "eval_loss": 0.7048454284667969, | |
| "eval_runtime": 0.4961, | |
| "eval_samples_per_second": 558.334, | |
| "eval_steps_per_second": 6.047, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 3.680000000000001e-05, | |
| "loss": 0.6699, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.49458483754512633, | |
| "eval_loss": 0.7651455402374268, | |
| "eval_runtime": 0.4953, | |
| "eval_samples_per_second": 559.23, | |
| "eval_steps_per_second": 6.057, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.6432, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.48014440433212996, | |
| "eval_loss": 0.702499270439148, | |
| "eval_runtime": 0.4961, | |
| "eval_samples_per_second": 558.37, | |
| "eval_steps_per_second": 6.047, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3.52e-05, | |
| "loss": 0.6008, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5379061371841155, | |
| "eval_loss": 0.7388982176780701, | |
| "eval_runtime": 0.5048, | |
| "eval_samples_per_second": 548.748, | |
| "eval_steps_per_second": 5.943, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 3.44e-05, | |
| "loss": 0.5142, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5342960288808665, | |
| "eval_loss": 0.9628210067749023, | |
| "eval_runtime": 0.4987, | |
| "eval_samples_per_second": 555.482, | |
| "eval_steps_per_second": 6.016, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "step": 140, | |
| "total_flos": 2550691654533120.0, | |
| "train_loss": 0.6586811883108956, | |
| "train_runtime": 121.339, | |
| "train_samples_per_second": 1026.051, | |
| "train_steps_per_second": 8.241 | |
| } | |
| ], | |
| "max_steps": 1000, | |
| "num_train_epochs": 50, | |
| "total_flos": 2550691654533120.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |