| { | |
| "best_metric": 0.6773178026129387, | |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-2/checkpoint-14292", | |
| "epoch": 9.0, | |
| "eval_steps": 500, | |
| "global_step": 14292, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.5482791066169739, | |
| "learning_rate": 9.642076934385773e-06, | |
| "loss": 0.5031, | |
| "step": 1588 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6666929009916575, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.49551922082901, | |
| "eval_mcc": 0.0, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 23.2895, | |
| "eval_samples_per_second": 545.567, | |
| "eval_steps_per_second": 17.089, | |
| "step": 1588 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.0544217824935913, | |
| "learning_rate": 8.436817317587551e-06, | |
| "loss": 0.4941, | |
| "step": 3176 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6666929009916575, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.48784464597702026, | |
| "eval_mcc": 0.0, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 23.5003, | |
| "eval_samples_per_second": 540.673, | |
| "eval_steps_per_second": 16.936, | |
| "step": 3176 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.3678545951843262, | |
| "learning_rate": 7.23155770078933e-06, | |
| "loss": 0.4903, | |
| "step": 4764 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6667716039666299, | |
| "eval_f1": 0.003295668549905838, | |
| "eval_loss": 0.48430460691452026, | |
| "eval_mcc": 0.013927887759262332, | |
| "eval_precision": 0.5384615384615384, | |
| "eval_recall": 0.001652892561983471, | |
| "eval_runtime": 23.4354, | |
| "eval_samples_per_second": 542.171, | |
| "eval_steps_per_second": 16.983, | |
| "step": 4764 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.989233136177063, | |
| "learning_rate": 6.026298083991109e-06, | |
| "loss": 0.4879, | |
| "step": 6352 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6692900991657484, | |
| "eval_f1": 0.05742485419470615, | |
| "eval_loss": 0.4830881953239441, | |
| "eval_mcc": 0.0682424873004379, | |
| "eval_precision": 0.5739910313901345, | |
| "eval_recall": 0.030224321133412044, | |
| "eval_runtime": 23.418, | |
| "eval_samples_per_second": 542.574, | |
| "eval_steps_per_second": 16.995, | |
| "step": 6352 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.5889469385147095, | |
| "learning_rate": 4.821038467192886e-06, | |
| "loss": 0.4867, | |
| "step": 7940 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.669211396190776, | |
| "eval_f1": 0.049310110834652796, | |
| "eval_loss": 0.48194119334220886, | |
| "eval_mcc": 0.0653430061639141, | |
| "eval_precision": 0.5860215053763441, | |
| "eval_recall": 0.025737898465171192, | |
| "eval_runtime": 23.4842, | |
| "eval_samples_per_second": 541.044, | |
| "eval_steps_per_second": 16.948, | |
| "step": 7940 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.164549708366394, | |
| "learning_rate": 3.615778850394665e-06, | |
| "loss": 0.4851, | |
| "step": 9528 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6764520698882418, | |
| "eval_f1": 0.1451445206903722, | |
| "eval_loss": 0.48082929849624634, | |
| "eval_mcc": 0.12675760316839801, | |
| "eval_precision": 0.60801393728223, | |
| "eval_recall": 0.08240850059031878, | |
| "eval_runtime": 23.4779, | |
| "eval_samples_per_second": 541.19, | |
| "eval_steps_per_second": 16.952, | |
| "step": 9528 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.5342378616333008, | |
| "learning_rate": 2.410519233596443e-06, | |
| "loss": 0.4833, | |
| "step": 11116 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6767668817881316, | |
| "eval_f1": 0.1523219814241486, | |
| "eval_loss": 0.4799216091632843, | |
| "eval_mcc": 0.12939159899915856, | |
| "eval_precision": 0.6049180327868853, | |
| "eval_recall": 0.0871310507674144, | |
| "eval_runtime": 23.2444, | |
| "eval_samples_per_second": 546.625, | |
| "eval_steps_per_second": 17.122, | |
| "step": 11116 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.1848862171173096, | |
| "learning_rate": 1.2052596167982216e-06, | |
| "loss": 0.4837, | |
| "step": 12704 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6771603966629939, | |
| "eval_f1": 0.15527182866556835, | |
| "eval_loss": 0.47950050234794617, | |
| "eval_mcc": 0.13165465344587687, | |
| "eval_precision": 0.607085346215781, | |
| "eval_recall": 0.08902007083825265, | |
| "eval_runtime": 23.1542, | |
| "eval_samples_per_second": 548.757, | |
| "eval_steps_per_second": 17.189, | |
| "step": 12704 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 1.0717182159423828, | |
| "learning_rate": 0.0, | |
| "loss": 0.4838, | |
| "step": 14292 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.6773178026129387, | |
| "eval_f1": 0.15707236842105263, | |
| "eval_loss": 0.4793970286846161, | |
| "eval_mcc": 0.13265414298730402, | |
| "eval_precision": 0.6073131955484896, | |
| "eval_recall": 0.09020070838252657, | |
| "eval_runtime": 23.3667, | |
| "eval_samples_per_second": 543.764, | |
| "eval_steps_per_second": 17.033, | |
| "step": 14292 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 14292, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 500, | |
| "total_flos": 35180046686700.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.7681424055105933, | |
| "learning_rate": 1.0847336551183995e-05, | |
| "num_train_epochs": 9, | |
| "temperature": 16 | |
| } | |
| } | |