| { | |
| "best_metric": 0.7223198594024605, | |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-24/checkpoint-428", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 428, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 4.061065673828125, | |
| "learning_rate": 0.0005772102319780231, | |
| "loss": 0.5743, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7012302284710018, | |
| "eval_f1": 0.3342036553524804, | |
| "eval_loss": 0.548787534236908, | |
| "eval_mcc": 0.24244041680325173, | |
| "eval_precision": 0.649746192893401, | |
| "eval_recall": 0.22495606326889278, | |
| "eval_runtime": 3.1573, | |
| "eval_samples_per_second": 540.652, | |
| "eval_steps_per_second": 17.103, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.865691900253296, | |
| "learning_rate": 0.0005432566889204924, | |
| "loss": 0.5363, | |
| "step": 428 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7223198594024605, | |
| "eval_f1": 0.5380116959064327, | |
| "eval_loss": 0.5190436244010925, | |
| "eval_mcc": 0.3470929601359411, | |
| "eval_precision": 0.6039387308533917, | |
| "eval_recall": 0.4850615114235501, | |
| "eval_runtime": 3.2428, | |
| "eval_samples_per_second": 526.395, | |
| "eval_steps_per_second": 16.652, | |
| "step": 428 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3852, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 18, | |
| "save_steps": 500, | |
| "total_flos": 1049551328880.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.9335511903353643, | |
| "learning_rate": 0.0006111637750355539, | |
| "num_train_epochs": 18, | |
| "temperature": 13 | |
| } | |
| } | |