| { | |
| "best_metric": 0.7633274751025191, | |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-12/checkpoint-1712", | |
| "epoch": 8.0, | |
| "eval_steps": 500, | |
| "global_step": 1712, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 9.038531303405762, | |
| "learning_rate": 0.0001519444662250004, | |
| "loss": 0.5879, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7076742823667252, | |
| "eval_f1": 0.5434583714547119, | |
| "eval_loss": 0.5341159105300903, | |
| "eval_mcc": 0.329604437952711, | |
| "eval_precision": 0.566793893129771, | |
| "eval_recall": 0.5219683655536028, | |
| "eval_runtime": 3.1577, | |
| "eval_samples_per_second": 540.589, | |
| "eval_steps_per_second": 17.101, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.545409917831421, | |
| "learning_rate": 0.00013295140794687536, | |
| "loss": 0.5171, | |
| "step": 428 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7422378441710603, | |
| "eval_f1": 0.5111111111111112, | |
| "eval_loss": 0.5047069191932678, | |
| "eval_mcc": 0.3761458173289304, | |
| "eval_precision": 0.6948640483383686, | |
| "eval_recall": 0.40421792618629176, | |
| "eval_runtime": 3.8964, | |
| "eval_samples_per_second": 438.1, | |
| "eval_steps_per_second": 13.859, | |
| "step": 428 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.3753821849823, | |
| "learning_rate": 0.0001139583496687503, | |
| "loss": 0.4951, | |
| "step": 642 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7527826596367897, | |
| "eval_f1": 0.546236559139785, | |
| "eval_loss": 0.5026707649230957, | |
| "eval_mcc": 0.40677405413312046, | |
| "eval_precision": 0.703601108033241, | |
| "eval_recall": 0.44639718804920914, | |
| "eval_runtime": 3.1728, | |
| "eval_samples_per_second": 538.017, | |
| "eval_steps_per_second": 17.02, | |
| "step": 642 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 4.286716938018799, | |
| "learning_rate": 9.496529139062526e-05, | |
| "loss": 0.4835, | |
| "step": 856 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7521968365553603, | |
| "eval_f1": 0.5273743016759778, | |
| "eval_loss": 0.49232202768325806, | |
| "eval_mcc": 0.4025712218580689, | |
| "eval_precision": 0.7239263803680982, | |
| "eval_recall": 0.4147627416520211, | |
| "eval_runtime": 3.2254, | |
| "eval_samples_per_second": 529.238, | |
| "eval_steps_per_second": 16.742, | |
| "step": 856 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 6.082200527191162, | |
| "learning_rate": 7.59722331125002e-05, | |
| "loss": 0.4751, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7545401288810779, | |
| "eval_f1": 0.5855588526211671, | |
| "eval_loss": 0.4886399805545807, | |
| "eval_mcc": 0.42175813043842797, | |
| "eval_precision": 0.669683257918552, | |
| "eval_recall": 0.5202108963093146, | |
| "eval_runtime": 3.1296, | |
| "eval_samples_per_second": 545.437, | |
| "eval_steps_per_second": 17.255, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 3.644395589828491, | |
| "learning_rate": 5.697917483437515e-05, | |
| "loss": 0.4705, | |
| "step": 1284 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7568834212067955, | |
| "eval_f1": 0.5966958211856171, | |
| "eval_loss": 0.48512187600135803, | |
| "eval_mcc": 0.43040150867070137, | |
| "eval_precision": 0.6673913043478261, | |
| "eval_recall": 0.539543057996485, | |
| "eval_runtime": 3.1955, | |
| "eval_samples_per_second": 534.196, | |
| "eval_steps_per_second": 16.899, | |
| "step": 1284 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 6.025336265563965, | |
| "learning_rate": 3.79861165562501e-05, | |
| "loss": 0.4638, | |
| "step": 1498 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7598125366139425, | |
| "eval_f1": 0.5824847250509165, | |
| "eval_loss": 0.4841197729110718, | |
| "eval_mcc": 0.4304303132828965, | |
| "eval_precision": 0.6924939467312349, | |
| "eval_recall": 0.5026362038664324, | |
| "eval_runtime": 3.1724, | |
| "eval_samples_per_second": 538.074, | |
| "eval_steps_per_second": 17.022, | |
| "step": 1498 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 7.094229698181152, | |
| "learning_rate": 1.899305827812505e-05, | |
| "loss": 0.4594, | |
| "step": 1712 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7633274751025191, | |
| "eval_f1": 0.5935613682092556, | |
| "eval_loss": 0.48480337858200073, | |
| "eval_mcc": 0.44066073676066914, | |
| "eval_precision": 0.6941176470588235, | |
| "eval_recall": 0.5184534270650264, | |
| "eval_runtime": 3.347, | |
| "eval_samples_per_second": 510.014, | |
| "eval_steps_per_second": 16.134, | |
| "step": 1712 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1926, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 500, | |
| "total_flos": 4198205315520.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.9671338526849426, | |
| "learning_rate": 0.00017093752450312547, | |
| "num_train_epochs": 9, | |
| "temperature": 10 | |
| } | |
| } | |