| { |
| "best_metric": 15.23105066412153, |
| "best_model_checkpoint": "./whisper-small-dv-syn/checkpoint-4000", |
| "epoch": 2.7416038382453736, |
| "global_step": 4000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.03, |
| "learning_rate": 9.4e-06, |
| "loss": 4.2925, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.07, |
| "learning_rate": 1e-05, |
| "loss": 1.5599, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.1, |
| "learning_rate": 1e-05, |
| "loss": 0.7325, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.14, |
| "learning_rate": 1e-05, |
| "loss": 0.1344, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.17, |
| "learning_rate": 1e-05, |
| "loss": 0.0581, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.21, |
| "learning_rate": 1e-05, |
| "loss": 0.0471, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.24, |
| "learning_rate": 1e-05, |
| "loss": 0.0406, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.27, |
| "learning_rate": 1e-05, |
| "loss": 0.0377, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.31, |
| "learning_rate": 1e-05, |
| "loss": 0.034, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.34, |
| "learning_rate": 1e-05, |
| "loss": 0.0327, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.38, |
| "learning_rate": 1e-05, |
| "loss": 0.0304, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.41, |
| "learning_rate": 1e-05, |
| "loss": 0.0294, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.45, |
| "learning_rate": 1e-05, |
| "loss": 0.0275, |
| "step": 650 |
| }, |
| { |
| "epoch": 0.48, |
| "learning_rate": 1e-05, |
| "loss": 0.0268, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.51, |
| "learning_rate": 1e-05, |
| "loss": 0.0261, |
| "step": 750 |
| }, |
| { |
| "epoch": 0.55, |
| "learning_rate": 1e-05, |
| "loss": 0.0246, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.58, |
| "learning_rate": 1e-05, |
| "loss": 0.0242, |
| "step": 850 |
| }, |
| { |
| "epoch": 0.62, |
| "learning_rate": 1e-05, |
| "loss": 0.0243, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.65, |
| "learning_rate": 1e-05, |
| "loss": 0.0228, |
| "step": 950 |
| }, |
| { |
| "epoch": 0.69, |
| "learning_rate": 1e-05, |
| "loss": 0.0233, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.69, |
| "eval_loss": 0.23616750538349152, |
| "eval_runtime": 227.0219, |
| "eval_samples_per_second": 2.202, |
| "eval_steps_per_second": 0.07, |
| "eval_wer": 18.86582359023304, |
| "eval_wer_ortho": 75.35455404979514, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.72, |
| "learning_rate": 1e-05, |
| "loss": 0.0219, |
| "step": 1050 |
| }, |
| { |
| "epoch": 0.75, |
| "learning_rate": 1e-05, |
| "loss": 0.0218, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.79, |
| "learning_rate": 1e-05, |
| "loss": 0.0213, |
| "step": 1150 |
| }, |
| { |
| "epoch": 0.82, |
| "learning_rate": 1e-05, |
| "loss": 0.0207, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.86, |
| "learning_rate": 1e-05, |
| "loss": 0.021, |
| "step": 1250 |
| }, |
| { |
| "epoch": 0.89, |
| "learning_rate": 1e-05, |
| "loss": 0.0201, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.93, |
| "learning_rate": 1e-05, |
| "loss": 0.0206, |
| "step": 1350 |
| }, |
| { |
| "epoch": 0.96, |
| "learning_rate": 1e-05, |
| "loss": 0.0194, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.99, |
| "learning_rate": 1e-05, |
| "loss": 0.0193, |
| "step": 1450 |
| }, |
| { |
| "epoch": 1.03, |
| "learning_rate": 1e-05, |
| "loss": 0.0169, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.06, |
| "learning_rate": 1e-05, |
| "loss": 0.0157, |
| "step": 1550 |
| }, |
| { |
| "epoch": 1.1, |
| "learning_rate": 1e-05, |
| "loss": 0.0154, |
| "step": 1600 |
| }, |
| { |
| "epoch": 1.13, |
| "learning_rate": 1e-05, |
| "loss": 0.0159, |
| "step": 1650 |
| }, |
| { |
| "epoch": 1.17, |
| "learning_rate": 1e-05, |
| "loss": 0.0153, |
| "step": 1700 |
| }, |
| { |
| "epoch": 1.2, |
| "learning_rate": 1e-05, |
| "loss": 0.0156, |
| "step": 1750 |
| }, |
| { |
| "epoch": 1.23, |
| "learning_rate": 1e-05, |
| "loss": 0.0151, |
| "step": 1800 |
| }, |
| { |
| "epoch": 1.27, |
| "learning_rate": 1e-05, |
| "loss": 0.0151, |
| "step": 1850 |
| }, |
| { |
| "epoch": 1.3, |
| "learning_rate": 1e-05, |
| "loss": 0.0152, |
| "step": 1900 |
| }, |
| { |
| "epoch": 1.34, |
| "learning_rate": 1e-05, |
| "loss": 0.015, |
| "step": 1950 |
| }, |
| { |
| "epoch": 1.37, |
| "learning_rate": 1e-05, |
| "loss": 0.0147, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.37, |
| "eval_loss": 0.21357361972332, |
| "eval_runtime": 225.8202, |
| "eval_samples_per_second": 2.214, |
| "eval_steps_per_second": 0.071, |
| "eval_wer": 16.734271852382086, |
| "eval_wer_ortho": 70.94232587456666, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.41, |
| "learning_rate": 1e-05, |
| "loss": 0.0149, |
| "step": 2050 |
| }, |
| { |
| "epoch": 1.44, |
| "learning_rate": 1e-05, |
| "loss": 0.0148, |
| "step": 2100 |
| }, |
| { |
| "epoch": 1.47, |
| "learning_rate": 1e-05, |
| "loss": 0.0152, |
| "step": 2150 |
| }, |
| { |
| "epoch": 1.51, |
| "learning_rate": 1e-05, |
| "loss": 0.0146, |
| "step": 2200 |
| }, |
| { |
| "epoch": 1.54, |
| "learning_rate": 1e-05, |
| "loss": 0.0146, |
| "step": 2250 |
| }, |
| { |
| "epoch": 1.58, |
| "learning_rate": 1e-05, |
| "loss": 0.0141, |
| "step": 2300 |
| }, |
| { |
| "epoch": 1.61, |
| "learning_rate": 1e-05, |
| "loss": 0.0143, |
| "step": 2350 |
| }, |
| { |
| "epoch": 1.64, |
| "learning_rate": 1e-05, |
| "loss": 0.0144, |
| "step": 2400 |
| }, |
| { |
| "epoch": 1.68, |
| "learning_rate": 1e-05, |
| "loss": 0.0142, |
| "step": 2450 |
| }, |
| { |
| "epoch": 1.71, |
| "learning_rate": 1e-05, |
| "loss": 0.0145, |
| "step": 2500 |
| }, |
| { |
| "epoch": 1.75, |
| "learning_rate": 1e-05, |
| "loss": 0.014, |
| "step": 2550 |
| }, |
| { |
| "epoch": 1.78, |
| "learning_rate": 1e-05, |
| "loss": 0.0141, |
| "step": 2600 |
| }, |
| { |
| "epoch": 1.82, |
| "learning_rate": 1e-05, |
| "loss": 0.0142, |
| "step": 2650 |
| }, |
| { |
| "epoch": 1.85, |
| "learning_rate": 1e-05, |
| "loss": 0.0143, |
| "step": 2700 |
| }, |
| { |
| "epoch": 1.88, |
| "learning_rate": 1e-05, |
| "loss": 0.0137, |
| "step": 2750 |
| }, |
| { |
| "epoch": 1.92, |
| "learning_rate": 1e-05, |
| "loss": 0.0137, |
| "step": 2800 |
| }, |
| { |
| "epoch": 1.95, |
| "learning_rate": 1e-05, |
| "loss": 0.0144, |
| "step": 2850 |
| }, |
| { |
| "epoch": 1.99, |
| "learning_rate": 1e-05, |
| "loss": 0.0138, |
| "step": 2900 |
| }, |
| { |
| "epoch": 2.02, |
| "learning_rate": 1e-05, |
| "loss": 0.0113, |
| "step": 2950 |
| }, |
| { |
| "epoch": 2.06, |
| "learning_rate": 1e-05, |
| "loss": 0.0102, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.06, |
| "eval_loss": 0.20163850486278534, |
| "eval_runtime": 226.4401, |
| "eval_samples_per_second": 2.208, |
| "eval_steps_per_second": 0.071, |
| "eval_wer": 15.57305336832896, |
| "eval_wer_ortho": 65.64765206429247, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.09, |
| "learning_rate": 1e-05, |
| "loss": 0.0099, |
| "step": 3050 |
| }, |
| { |
| "epoch": 2.12, |
| "learning_rate": 1e-05, |
| "loss": 0.01, |
| "step": 3100 |
| }, |
| { |
| "epoch": 2.16, |
| "learning_rate": 1e-05, |
| "loss": 0.0099, |
| "step": 3150 |
| }, |
| { |
| "epoch": 2.19, |
| "learning_rate": 1e-05, |
| "loss": 0.0098, |
| "step": 3200 |
| }, |
| { |
| "epoch": 2.23, |
| "learning_rate": 1e-05, |
| "loss": 0.0107, |
| "step": 3250 |
| }, |
| { |
| "epoch": 2.26, |
| "learning_rate": 1e-05, |
| "loss": 0.0098, |
| "step": 3300 |
| }, |
| { |
| "epoch": 2.3, |
| "learning_rate": 1e-05, |
| "loss": 0.01, |
| "step": 3350 |
| }, |
| { |
| "epoch": 2.33, |
| "learning_rate": 1e-05, |
| "loss": 0.0103, |
| "step": 3400 |
| }, |
| { |
| "epoch": 2.36, |
| "learning_rate": 1e-05, |
| "loss": 0.0106, |
| "step": 3450 |
| }, |
| { |
| "epoch": 2.4, |
| "learning_rate": 1e-05, |
| "loss": 0.0102, |
| "step": 3500 |
| }, |
| { |
| "epoch": 2.43, |
| "learning_rate": 1e-05, |
| "loss": 0.0103, |
| "step": 3550 |
| }, |
| { |
| "epoch": 2.47, |
| "learning_rate": 1e-05, |
| "loss": 0.0099, |
| "step": 3600 |
| }, |
| { |
| "epoch": 2.5, |
| "learning_rate": 1e-05, |
| "loss": 0.0103, |
| "step": 3650 |
| }, |
| { |
| "epoch": 2.54, |
| "learning_rate": 1e-05, |
| "loss": 0.0102, |
| "step": 3700 |
| }, |
| { |
| "epoch": 2.57, |
| "learning_rate": 1e-05, |
| "loss": 0.0105, |
| "step": 3750 |
| }, |
| { |
| "epoch": 2.6, |
| "learning_rate": 1e-05, |
| "loss": 0.0104, |
| "step": 3800 |
| }, |
| { |
| "epoch": 2.64, |
| "learning_rate": 1e-05, |
| "loss": 0.0101, |
| "step": 3850 |
| }, |
| { |
| "epoch": 2.67, |
| "learning_rate": 1e-05, |
| "loss": 0.01, |
| "step": 3900 |
| }, |
| { |
| "epoch": 2.71, |
| "learning_rate": 1e-05, |
| "loss": 0.0101, |
| "step": 3950 |
| }, |
| { |
| "epoch": 2.74, |
| "learning_rate": 1e-05, |
| "loss": 0.01, |
| "step": 4000 |
| }, |
| { |
| "epoch": 2.74, |
| "eval_loss": 0.20110772550106049, |
| "eval_runtime": 225.7624, |
| "eval_samples_per_second": 2.215, |
| "eval_steps_per_second": 0.071, |
| "eval_wer": 15.23105066412153, |
| "eval_wer_ortho": 66.34100220611408, |
| "step": 4000 |
| } |
| ], |
| "max_steps": 4000, |
| "num_train_epochs": 3, |
| "total_flos": 1.4770435676258304e+20, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
