| { | |
| "best_metric": 0.6896100044250488, | |
| "best_model_checkpoint": "hBERTv2_rte/checkpoint-70", | |
| "epoch": 12.0, | |
| "global_step": 120, | |
| "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": 0.7231, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.4548736462093863, | |
| "eval_loss": 0.7174825668334961, | |
| "eval_runtime": 0.3321, | |
| "eval_samples_per_second": 834.009, | |
| "eval_steps_per_second": 6.022, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.702, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.4729241877256318, | |
| "eval_loss": 0.7052958011627197, | |
| "eval_runtime": 0.3299, | |
| "eval_samples_per_second": 839.582, | |
| "eval_steps_per_second": 6.062, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6982, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.4584837545126354, | |
| "eval_loss": 0.6975917220115662, | |
| "eval_runtime": 0.3315, | |
| "eval_samples_per_second": 835.597, | |
| "eval_steps_per_second": 6.033, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.7008, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.4657039711191336, | |
| "eval_loss": 0.726104199886322, | |
| "eval_runtime": 0.3329, | |
| "eval_samples_per_second": 832.075, | |
| "eval_steps_per_second": 6.008, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.7022, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.49458483754512633, | |
| "eval_loss": 0.7142461538314819, | |
| "eval_runtime": 0.3322, | |
| "eval_samples_per_second": 833.768, | |
| "eval_steps_per_second": 6.02, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.6867, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.48014440433212996, | |
| "eval_loss": 0.694332480430603, | |
| "eval_runtime": 0.3388, | |
| "eval_samples_per_second": 817.49, | |
| "eval_steps_per_second": 5.902, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.6796, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5487364620938628, | |
| "eval_loss": 0.6896100044250488, | |
| "eval_runtime": 0.3322, | |
| "eval_samples_per_second": 833.85, | |
| "eval_steps_per_second": 6.021, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.6614, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.516245487364621, | |
| "eval_loss": 0.7151486277580261, | |
| "eval_runtime": 0.3309, | |
| "eval_samples_per_second": 837.215, | |
| "eval_steps_per_second": 6.045, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.6303, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 0.7244121432304382, | |
| "eval_runtime": 0.3318, | |
| "eval_samples_per_second": 834.768, | |
| "eval_steps_per_second": 6.027, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 4e-05, | |
| "loss": 0.602, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.4729241877256318, | |
| "eval_loss": 0.7570322751998901, | |
| "eval_runtime": 0.3338, | |
| "eval_samples_per_second": 829.795, | |
| "eval_steps_per_second": 5.991, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.5761, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.5379061371841155, | |
| "eval_loss": 0.7605395913124084, | |
| "eval_runtime": 0.334, | |
| "eval_samples_per_second": 829.304, | |
| "eval_steps_per_second": 5.988, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.5664, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.5234657039711191, | |
| "eval_loss": 0.8159566521644592, | |
| "eval_runtime": 0.3341, | |
| "eval_samples_per_second": 829.158, | |
| "eval_steps_per_second": 5.987, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "step": 120, | |
| "total_flos": 3721656539283456.0, | |
| "train_loss": 0.660737669467926, | |
| "train_runtime": 170.3259, | |
| "train_samples_per_second": 730.951, | |
| "train_steps_per_second": 2.936 | |
| } | |
| ], | |
| "max_steps": 500, | |
| "num_train_epochs": 50, | |
| "total_flos": 3721656539283456.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |