| { | |
| "best_metric": 0.9845542806707855, | |
| "best_model_checkpoint": "audio/train/checkpoint/wav2vec2-base/superb_ks_42/checkpoint-14373", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 15970, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.6261740763932373, | |
| "grad_norm": 12.87492847442627, | |
| "learning_rate": 3.1308703819661866e-05, | |
| "loss": 1.25, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9582230067666961, | |
| "eval_loss": 0.17545583844184875, | |
| "eval_runtime": 7.033, | |
| "eval_samples_per_second": 966.585, | |
| "eval_steps_per_second": 3.839, | |
| "step": 1597 | |
| }, | |
| { | |
| "epoch": 1.2523481527864746, | |
| "grad_norm": 2.619173765182495, | |
| "learning_rate": 4.859806581785292e-05, | |
| "loss": 0.3308, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.878522229179712, | |
| "grad_norm": 6.4030070304870605, | |
| "learning_rate": 4.51193209490016e-05, | |
| "loss": 0.2502, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9713150926743159, | |
| "eval_loss": 0.140165776014328, | |
| "eval_runtime": 6.9118, | |
| "eval_samples_per_second": 983.532, | |
| "eval_steps_per_second": 3.906, | |
| "step": 3194 | |
| }, | |
| { | |
| "epoch": 2.504696305572949, | |
| "grad_norm": 0.2295457422733307, | |
| "learning_rate": 4.1640576080150286e-05, | |
| "loss": 0.2218, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9802883200941453, | |
| "eval_loss": 0.09560667723417282, | |
| "eval_runtime": 7.0467, | |
| "eval_samples_per_second": 964.712, | |
| "eval_steps_per_second": 3.832, | |
| "step": 4791 | |
| }, | |
| { | |
| "epoch": 3.1308703819661865, | |
| "grad_norm": 0.9765733480453491, | |
| "learning_rate": 3.8161831211298965e-05, | |
| "loss": 0.194, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.757044458359424, | |
| "grad_norm": 0.9296017289161682, | |
| "learning_rate": 3.468308634244764e-05, | |
| "loss": 0.1746, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9796999117387467, | |
| "eval_loss": 0.09171486645936966, | |
| "eval_runtime": 6.5552, | |
| "eval_samples_per_second": 1037.04, | |
| "eval_steps_per_second": 4.119, | |
| "step": 6388 | |
| }, | |
| { | |
| "epoch": 4.383218534752661, | |
| "grad_norm": 7.867722511291504, | |
| "learning_rate": 3.120434147359633e-05, | |
| "loss": 0.17, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9807296263606943, | |
| "eval_loss": 0.08932828158140182, | |
| "eval_runtime": 6.9043, | |
| "eval_samples_per_second": 984.61, | |
| "eval_steps_per_second": 3.911, | |
| "step": 7985 | |
| }, | |
| { | |
| "epoch": 5.009392611145898, | |
| "grad_norm": 1.299570083618164, | |
| "learning_rate": 2.7725596604745007e-05, | |
| "loss": 0.1554, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 5.635566687539136, | |
| "grad_norm": 1.2470015287399292, | |
| "learning_rate": 2.4246851735893692e-05, | |
| "loss": 0.1431, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9810238305383936, | |
| "eval_loss": 0.09325208514928818, | |
| "eval_runtime": 6.9991, | |
| "eval_samples_per_second": 971.269, | |
| "eval_steps_per_second": 3.858, | |
| "step": 9582 | |
| }, | |
| { | |
| "epoch": 6.261740763932373, | |
| "grad_norm": 1.0575257539749146, | |
| "learning_rate": 2.076810686704237e-05, | |
| "loss": 0.1346, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 6.88791484032561, | |
| "grad_norm": 3.034492015838623, | |
| "learning_rate": 1.7289361998191053e-05, | |
| "loss": 0.1238, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9830832597822889, | |
| "eval_loss": 0.09578000754117966, | |
| "eval_runtime": 6.8981, | |
| "eval_samples_per_second": 985.49, | |
| "eval_steps_per_second": 3.914, | |
| "step": 11179 | |
| }, | |
| { | |
| "epoch": 7.514088916718848, | |
| "grad_norm": 1.000091314315796, | |
| "learning_rate": 1.3810617129339735e-05, | |
| "loss": 0.116, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9833774639599883, | |
| "eval_loss": 0.09704861044883728, | |
| "eval_runtime": 6.9946, | |
| "eval_samples_per_second": 971.892, | |
| "eval_steps_per_second": 3.86, | |
| "step": 12776 | |
| }, | |
| { | |
| "epoch": 8.140262993112085, | |
| "grad_norm": 2.212705373764038, | |
| "learning_rate": 1.0331872260488417e-05, | |
| "loss": 0.1113, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 8.766437069505322, | |
| "grad_norm": 3.2522332668304443, | |
| "learning_rate": 6.853127391637097e-06, | |
| "loss": 0.0995, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9845542806707855, | |
| "eval_loss": 0.08534605801105499, | |
| "eval_runtime": 6.7022, | |
| "eval_samples_per_second": 1014.3, | |
| "eval_steps_per_second": 4.029, | |
| "step": 14373 | |
| }, | |
| { | |
| "epoch": 9.39261114589856, | |
| "grad_norm": 6.198681354522705, | |
| "learning_rate": 3.3743825227857787e-06, | |
| "loss": 0.0985, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9838187702265372, | |
| "eval_loss": 0.08294928818941116, | |
| "eval_runtime": 6.7918, | |
| "eval_samples_per_second": 1000.911, | |
| "eval_steps_per_second": 3.975, | |
| "step": 15970 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 15970, | |
| "total_flos": 4.6387626201984e+18, | |
| "train_loss": 0.22964708045190324, | |
| "train_runtime": 1401.6538, | |
| "train_samples_per_second": 364.527, | |
| "train_steps_per_second": 11.394 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 15970, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 4.6387626201984e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |