| { | |
| "best_metric": 0.7562975981253661, | |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-23/checkpoint-1498", | |
| "epoch": 7.0, | |
| "eval_steps": 500, | |
| "global_step": 1498, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 8.174213409423828, | |
| "learning_rate": 0.00021193737357349756, | |
| "loss": 0.6053, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7041593438781488, | |
| "eval_f1": 0.4107351225204201, | |
| "eval_loss": 0.5531450510025024, | |
| "eval_mcc": 0.2654659366009494, | |
| "eval_precision": 0.6111111111111112, | |
| "eval_recall": 0.3093145869947276, | |
| "eval_runtime": 3.1397, | |
| "eval_samples_per_second": 543.682, | |
| "eval_steps_per_second": 17.199, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.549020290374756, | |
| "learning_rate": 0.00018544520187681036, | |
| "loss": 0.5455, | |
| "step": 428 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7229056824838899, | |
| "eval_f1": 0.4569460390355913, | |
| "eval_loss": 0.5337331295013428, | |
| "eval_mcc": 0.32023248430720136, | |
| "eval_precision": 0.6589403973509934, | |
| "eval_recall": 0.34973637961335674, | |
| "eval_runtime": 3.1777, | |
| "eval_samples_per_second": 537.185, | |
| "eval_steps_per_second": 16.994, | |
| "step": 428 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.2418973445892334, | |
| "learning_rate": 0.00015895303018012317, | |
| "loss": 0.514, | |
| "step": 642 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7311072056239016, | |
| "eval_f1": 0.43122676579925656, | |
| "eval_loss": 0.5188717246055603, | |
| "eval_mcc": 0.3396283739174578, | |
| "eval_precision": 0.7310924369747899, | |
| "eval_recall": 0.30579964850615116, | |
| "eval_runtime": 3.1525, | |
| "eval_samples_per_second": 541.47, | |
| "eval_steps_per_second": 17.129, | |
| "step": 642 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 4.082540988922119, | |
| "learning_rate": 0.000132460858483436, | |
| "loss": 0.5028, | |
| "step": 856 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7281780902167545, | |
| "eval_f1": 0.3796791443850267, | |
| "eval_loss": 0.521298348903656, | |
| "eval_mcc": 0.33395946974565643, | |
| "eval_precision": 0.7932960893854749, | |
| "eval_recall": 0.24956063268892795, | |
| "eval_runtime": 3.1071, | |
| "eval_samples_per_second": 549.395, | |
| "eval_steps_per_second": 17.38, | |
| "step": 856 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 5.98746919631958, | |
| "learning_rate": 0.00010596868678674878, | |
| "loss": 0.4947, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7492677211482133, | |
| "eval_f1": 0.5668016194331983, | |
| "eval_loss": 0.49519282579421997, | |
| "eval_mcc": 0.4052305275875567, | |
| "eval_precision": 0.6682577565632458, | |
| "eval_recall": 0.492091388400703, | |
| "eval_runtime": 3.1175, | |
| "eval_samples_per_second": 547.552, | |
| "eval_steps_per_second": 17.321, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 4.624367713928223, | |
| "learning_rate": 7.947651509006158e-05, | |
| "loss": 0.4792, | |
| "step": 1284 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7545401288810779, | |
| "eval_f1": 0.6080449017773621, | |
| "eval_loss": 0.49225014448165894, | |
| "eval_mcc": 0.4323544910894334, | |
| "eval_precision": 0.65, | |
| "eval_recall": 0.5711775043936731, | |
| "eval_runtime": 3.1259, | |
| "eval_samples_per_second": 546.074, | |
| "eval_steps_per_second": 17.275, | |
| "step": 1284 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 3.617922067642212, | |
| "learning_rate": 5.298434339337439e-05, | |
| "loss": 0.4667, | |
| "step": 1498 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7562975981253661, | |
| "eval_f1": 0.5315315315315315, | |
| "eval_loss": 0.4943116009235382, | |
| "eval_mcc": 0.4133753055226225, | |
| "eval_precision": 0.7398119122257053, | |
| "eval_recall": 0.4147627416520211, | |
| "eval_runtime": 3.1423, | |
| "eval_samples_per_second": 543.239, | |
| "eval_steps_per_second": 17.185, | |
| "step": 1498 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1926, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 500, | |
| "total_flos": 3673429651080.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.9926400370862213, | |
| "learning_rate": 0.00023842954527018476, | |
| "num_train_epochs": 9, | |
| "temperature": 23 | |
| } | |
| } | |