| { | |
| "best_metric": 21.483863040629096, | |
| "best_model_checkpoint": "./att1/checkpoint-11000", | |
| "epoch": 1.4468333333333334, | |
| "global_step": 11000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 9.9e-06, | |
| "loss": 0.6056, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 9.569565217391306e-06, | |
| "loss": 0.5195, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_loss": 0.4892498850822449, | |
| "eval_runtime": 5699.9901, | |
| "eval_samples_per_second": 0.585, | |
| "eval_steps_per_second": 0.037, | |
| "eval_wer": 26.316759501965926, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.134782608695652e-06, | |
| "loss": 0.494, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 8.700000000000001e-06, | |
| "loss": 0.4744, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_loss": 0.45853909850120544, | |
| "eval_runtime": 5781.8818, | |
| "eval_samples_per_second": 0.576, | |
| "eval_steps_per_second": 0.036, | |
| "eval_wer": 25.868283093053734, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 8.26608695652174e-06, | |
| "loss": 0.4652, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 7.831304347826087e-06, | |
| "loss": 0.4449, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.4347614645957947, | |
| "eval_runtime": 5767.4751, | |
| "eval_samples_per_second": 0.578, | |
| "eval_steps_per_second": 0.036, | |
| "eval_wer": 24.18086500655308, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 7.396521739130436e-06, | |
| "loss": 0.4431, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 6.961739130434784e-06, | |
| "loss": 0.4267, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_loss": 0.41802194714546204, | |
| "eval_runtime": 5718.7598, | |
| "eval_samples_per_second": 0.583, | |
| "eval_steps_per_second": 0.037, | |
| "eval_wer": 23.638188073394495, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 6.526956521739131e-06, | |
| "loss": 0.4231, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 6.092173913043479e-06, | |
| "loss": 0.4163, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_loss": 0.4041951596736908, | |
| "eval_runtime": 6102.971, | |
| "eval_samples_per_second": 0.546, | |
| "eval_steps_per_second": 0.034, | |
| "eval_wer": 23.937172346002622, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 5.6573913043478265e-06, | |
| "loss": 0.4094, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 5.222608695652175e-06, | |
| "loss": 0.3171, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "eval_loss": 0.4081740379333496, | |
| "eval_runtime": 5663.9498, | |
| "eval_samples_per_second": 0.588, | |
| "eval_steps_per_second": 0.037, | |
| "eval_wer": 23.087319790301443, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 4.787826086956522e-06, | |
| "loss": 0.273, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 4.35304347826087e-06, | |
| "loss": 0.2763, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "eval_loss": 0.4000445604324341, | |
| "eval_runtime": 5952.601, | |
| "eval_samples_per_second": 0.56, | |
| "eval_steps_per_second": 0.035, | |
| "eval_wer": 22.94601900393185, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 3.918260869565217e-06, | |
| "loss": 0.2727, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 3.4843478260869567e-06, | |
| "loss": 0.279, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_loss": 0.39390885829925537, | |
| "eval_runtime": 5726.1575, | |
| "eval_samples_per_second": 0.582, | |
| "eval_steps_per_second": 0.036, | |
| "eval_wer": 22.493446920052424, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 3.049565217391305e-06, | |
| "loss": 0.2759, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 2.6147826086956524e-06, | |
| "loss": 0.2841, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "eval_loss": 0.38072848320007324, | |
| "eval_runtime": 6058.2634, | |
| "eval_samples_per_second": 0.55, | |
| "eval_steps_per_second": 0.034, | |
| "eval_wer": 22.52211664482307, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 2.1800000000000003e-06, | |
| "loss": 0.2712, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 1.745217391304348e-06, | |
| "loss": 0.27, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "eval_loss": 0.37836796045303345, | |
| "eval_runtime": 5966.0536, | |
| "eval_samples_per_second": 0.558, | |
| "eval_steps_per_second": 0.035, | |
| "eval_wer": 21.907765399737876, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 1.311304347826087e-06, | |
| "loss": 0.2869, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 8.773913043478262e-07, | |
| "loss": 0.2646, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "eval_loss": 0.3694765865802765, | |
| "eval_runtime": 5884.5805, | |
| "eval_samples_per_second": 0.566, | |
| "eval_steps_per_second": 0.036, | |
| "eval_wer": 21.483863040629096, | |
| "step": 11000 | |
| } | |
| ], | |
| "max_steps": 12000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 3.736950850289664e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |