| { | |
| "best_metric": 0.6521328506217535, | |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-4/checkpoint-1588", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 7940, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.15466690063476562, | |
| "learning_rate": 0.00028711293555461854, | |
| "loss": 0.047, | |
| "step": 1588 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6521328506217535, | |
| "eval_f1": 0.006741573033707866, | |
| "eval_loss": 0.043082475662231445, | |
| "eval_mcc": -0.07334683954324206, | |
| "eval_precision": 0.06976744186046512, | |
| "eval_recall": 0.0035419126328217238, | |
| "eval_runtime": 23.3677, | |
| "eval_samples_per_second": 543.742, | |
| "eval_steps_per_second": 17.032, | |
| "step": 1588 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.2909219563007355, | |
| "learning_rate": 0.0002153347016659639, | |
| "loss": 0.0437, | |
| "step": 3176 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6499291673225248, | |
| "eval_f1": 0.010676156583629894, | |
| "eval_loss": 0.04202444478869438, | |
| "eval_mcc": -0.07414646393220078, | |
| "eval_precision": 0.09195402298850575, | |
| "eval_recall": 0.005667060212514758, | |
| "eval_runtime": 23.3008, | |
| "eval_samples_per_second": 545.304, | |
| "eval_steps_per_second": 17.081, | |
| "step": 3176 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.4530297815799713, | |
| "learning_rate": 0.00014355646777730927, | |
| "loss": 0.0429, | |
| "step": 4764 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6510310089721392, | |
| "eval_f1": 0.007165248544558891, | |
| "eval_loss": 0.04164046049118042, | |
| "eval_mcc": -0.07622110191236336, | |
| "eval_precision": 0.06926406926406926, | |
| "eval_recall": 0.0037780401416765055, | |
| "eval_runtime": 23.5928, | |
| "eval_samples_per_second": 538.553, | |
| "eval_steps_per_second": 16.87, | |
| "step": 4764 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.12990067899227142, | |
| "learning_rate": 7.177823388865464e-05, | |
| "loss": 0.0424, | |
| "step": 6352 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.651188414922084, | |
| "eval_f1": 0.006723442402510086, | |
| "eval_loss": 0.041306350380182266, | |
| "eval_mcc": -0.0764573922901941, | |
| "eval_precision": 0.06607929515418502, | |
| "eval_recall": 0.0035419126328217238, | |
| "eval_runtime": 23.6051, | |
| "eval_samples_per_second": 538.273, | |
| "eval_steps_per_second": 16.861, | |
| "step": 6352 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.14243847131729126, | |
| "learning_rate": 0.0, | |
| "loss": 0.042, | |
| "step": 7940 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6503226821973871, | |
| "eval_f1": 0.010247271107150813, | |
| "eval_loss": 0.04123334959149361, | |
| "eval_mcc": -0.0735498736578602, | |
| "eval_precision": 0.09055118110236221, | |
| "eval_recall": 0.005430932703659976, | |
| "eval_runtime": 23.3153, | |
| "eval_samples_per_second": 544.964, | |
| "eval_steps_per_second": 17.07, | |
| "step": 7940 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 7940, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 19544470381500.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.05479484259344969, | |
| "learning_rate": 0.0003588911694432732, | |
| "num_train_epochs": 5, | |
| "temperature": 13 | |
| } | |
| } | |