| { | |
| "best_metric": 0.22972367703914642, | |
| "best_model_checkpoint": "bert-base-uncased-qnli/checkpoint-1638", | |
| "epoch": 9.0, | |
| "global_step": 7371, | |
| "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": 0.3436, | |
| "step": 819 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9035328574043566, | |
| "eval_loss": 0.24892905354499817, | |
| "eval_runtime": 6.5235, | |
| "eval_samples_per_second": 837.436, | |
| "eval_steps_per_second": 6.592, | |
| "step": 819 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.1962, | |
| "step": 1638 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9104887424492037, | |
| "eval_loss": 0.22972367703914642, | |
| "eval_runtime": 6.4904, | |
| "eval_samples_per_second": 841.708, | |
| "eval_steps_per_second": 6.625, | |
| "step": 1638 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.1049, | |
| "step": 2457 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9121361889071938, | |
| "eval_loss": 0.26201677322387695, | |
| "eval_runtime": 6.5072, | |
| "eval_samples_per_second": 839.533, | |
| "eval_steps_per_second": 6.608, | |
| "step": 2457 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.0662, | |
| "step": 3276 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9088412959912137, | |
| "eval_loss": 0.3533869981765747, | |
| "eval_runtime": 6.5309, | |
| "eval_samples_per_second": 836.486, | |
| "eval_steps_per_second": 6.584, | |
| "step": 3276 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.0487, | |
| "step": 4095 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9046311550430166, | |
| "eval_loss": 0.3687661290168762, | |
| "eval_runtime": 6.542, | |
| "eval_samples_per_second": 835.06, | |
| "eval_steps_per_second": 6.573, | |
| "step": 4095 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.0368, | |
| "step": 4914 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9073768991396669, | |
| "eval_loss": 0.3942682147026062, | |
| "eval_runtime": 6.6485, | |
| "eval_samples_per_second": 821.69, | |
| "eval_steps_per_second": 6.468, | |
| "step": 4914 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.0329, | |
| "step": 5733 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9092073952041003, | |
| "eval_loss": 0.42497432231903076, | |
| "eval_runtime": 6.5176, | |
| "eval_samples_per_second": 838.197, | |
| "eval_steps_per_second": 6.598, | |
| "step": 5733 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.0272, | |
| "step": 6552 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.90536335346879, | |
| "eval_loss": 0.4012433886528015, | |
| "eval_runtime": 6.5823, | |
| "eval_samples_per_second": 829.956, | |
| "eval_steps_per_second": 6.533, | |
| "step": 6552 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.0243, | |
| "step": 7371 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9040820062236866, | |
| "eval_loss": 0.4496908485889435, | |
| "eval_runtime": 6.5299, | |
| "eval_samples_per_second": 836.618, | |
| "eval_steps_per_second": 6.585, | |
| "step": 7371 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "step": 7371, | |
| "total_flos": 1.2401568924460646e+17, | |
| "train_loss": 0.09787335575006982, | |
| "train_runtime": 3170.4691, | |
| "train_samples_per_second": 1651.853, | |
| "train_steps_per_second": 12.916 | |
| } | |
| ], | |
| "max_steps": 40950, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.2401568924460646e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |