| { | |
| "best_metric": 0.700058582308143, | |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-1/checkpoint-2140", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 2140, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.750048875808716, | |
| "learning_rate": 0.0008904595556519428, | |
| "loss": 0.3982, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6936145284124194, | |
| "eval_f1": 0.2496413199426112, | |
| "eval_loss": 0.3941108286380768, | |
| "eval_mcc": 0.2091898426756361, | |
| "eval_precision": 0.6796875, | |
| "eval_recall": 0.15289982425307558, | |
| "eval_runtime": 3.1574, | |
| "eval_samples_per_second": 540.639, | |
| "eval_steps_per_second": 17.103, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.170525074005127, | |
| "learning_rate": 0.0008014136000867486, | |
| "loss": 0.3881, | |
| "step": 428 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.3913939595222473, | |
| "eval_mcc": 0.0, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 3.2957, | |
| "eval_samples_per_second": 517.943, | |
| "eval_steps_per_second": 16.385, | |
| "step": 428 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.8615076541900635, | |
| "learning_rate": 0.0007123676445215543, | |
| "loss": 0.3832, | |
| "step": 642 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.69302870533099, | |
| "eval_f1": 0.2681564245810056, | |
| "eval_loss": 0.3964165151119232, | |
| "eval_mcc": 0.20820116163188562, | |
| "eval_precision": 0.6530612244897959, | |
| "eval_recall": 0.1687170474516696, | |
| "eval_runtime": 3.1559, | |
| "eval_samples_per_second": 540.891, | |
| "eval_steps_per_second": 17.111, | |
| "step": 642 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.2408286333084106, | |
| "learning_rate": 0.00062332168895636, | |
| "loss": 0.3935, | |
| "step": 856 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6924428822495606, | |
| "eval_f1": 0.25742574257425743, | |
| "eval_loss": 0.39223405718803406, | |
| "eval_mcc": 0.20514822121510018, | |
| "eval_precision": 0.6594202898550725, | |
| "eval_recall": 0.15992970123022848, | |
| "eval_runtime": 3.8392, | |
| "eval_samples_per_second": 444.629, | |
| "eval_steps_per_second": 14.066, | |
| "step": 856 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.090942144393921, | |
| "learning_rate": 0.0005342757333911657, | |
| "loss": 0.3871, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.389874130487442, | |
| "eval_mcc": 0.0, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 3.1327, | |
| "eval_samples_per_second": 544.899, | |
| "eval_steps_per_second": 17.238, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.2668867111206055, | |
| "learning_rate": 0.0004452297778259714, | |
| "loss": 0.3881, | |
| "step": 1284 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6977152899824253, | |
| "eval_f1": 0.26074498567335247, | |
| "eval_loss": 0.3898678421974182, | |
| "eval_mcc": 0.22568315247838705, | |
| "eval_precision": 0.7054263565891473, | |
| "eval_recall": 0.15992970123022848, | |
| "eval_runtime": 3.2109, | |
| "eval_samples_per_second": 531.633, | |
| "eval_steps_per_second": 16.818, | |
| "step": 1284 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.9589310884475708, | |
| "learning_rate": 0.00035618382226077714, | |
| "loss": 0.3856, | |
| "step": 1498 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.690099589923843, | |
| "eval_f1": 0.18238021638330756, | |
| "eval_loss": 0.3910105228424072, | |
| "eval_mcc": 0.19638682696493384, | |
| "eval_precision": 0.7564102564102564, | |
| "eval_recall": 0.10369068541300527, | |
| "eval_runtime": 3.168, | |
| "eval_samples_per_second": 538.82, | |
| "eval_steps_per_second": 17.045, | |
| "step": 1498 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.0469610691070557, | |
| "learning_rate": 0.00026713786669558284, | |
| "loss": 0.3795, | |
| "step": 1712 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6965436438195665, | |
| "eval_f1": 0.22686567164179106, | |
| "eval_loss": 0.3840750753879547, | |
| "eval_mcc": 0.22297453441408152, | |
| "eval_precision": 0.7524752475247525, | |
| "eval_recall": 0.1335676625659051, | |
| "eval_runtime": 3.305, | |
| "eval_samples_per_second": 516.497, | |
| "eval_steps_per_second": 16.339, | |
| "step": 1712 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.8197473883628845, | |
| "learning_rate": 0.00017809191113038857, | |
| "loss": 0.3762, | |
| "step": 1926 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.6936145284124194, | |
| "eval_f1": 0.19908116385911181, | |
| "eval_loss": 0.3833659291267395, | |
| "eval_mcc": 0.21257347787570094, | |
| "eval_precision": 0.7738095238095238, | |
| "eval_recall": 0.11423550087873462, | |
| "eval_runtime": 3.1266, | |
| "eval_samples_per_second": 545.952, | |
| "eval_steps_per_second": 17.271, | |
| "step": 1926 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 2.2055039405822754, | |
| "learning_rate": 8.904595556519429e-05, | |
| "loss": 0.3738, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.700058582308143, | |
| "eval_f1": 0.30997304582210244, | |
| "eval_loss": 0.3850247263908386, | |
| "eval_mcc": 0.23609006824017303, | |
| "eval_precision": 0.6647398843930635, | |
| "eval_recall": 0.20210896309314588, | |
| "eval_runtime": 3.1854, | |
| "eval_samples_per_second": 535.877, | |
| "eval_steps_per_second": 16.952, | |
| "step": 2140 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2354, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 11, | |
| "save_steps": 500, | |
| "total_flos": 5247756644400.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.6246941768140464, | |
| "learning_rate": 0.000979505511217137, | |
| "num_train_epochs": 11, | |
| "temperature": 32 | |
| } | |
| } | |