| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.3224, |
| "eval_steps": 500, |
| "global_step": 4500, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.02, |
| "learning_rate": 1.9600000000000002e-05, |
| "loss": 1.5719, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.04, |
| "learning_rate": 3.960000000000001e-05, |
| "loss": 0.3718, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.06, |
| "learning_rate": 5.96e-05, |
| "loss": 0.3261, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.08, |
| "learning_rate": 7.960000000000001e-05, |
| "loss": 0.3219, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.1, |
| "learning_rate": 9.960000000000001e-05, |
| "loss": 0.3252, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.1, |
| "eval_combined_wer": 0.19175330812854444, |
| "eval_f1_score": 0.6148314606741573, |
| "eval_label_f1": 0.7577528089887641, |
| "eval_loss": 0.3395695686340332, |
| "eval_runtime": 98.1651, |
| "eval_samples_per_second": 10.187, |
| "eval_steps_per_second": 0.163, |
| "eval_wer": 0.11928060336491975, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.12, |
| "learning_rate": 9.98830238119205e-05, |
| "loss": 0.321, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.14, |
| "learning_rate": 9.952307128483256e-05, |
| "loss": 0.3095, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.16, |
| "learning_rate": 9.892184733248666e-05, |
| "loss": 0.2932, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.18, |
| "learning_rate": 9.808228105754376e-05, |
| "loss": 0.2845, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.2, |
| "learning_rate": 9.700846274250251e-05, |
| "loss": 0.2729, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.2, |
| "eval_combined_wer": 0.17300724637681159, |
| "eval_f1_score": 0.6449152542372881, |
| "eval_label_f1": 0.7906779661016949, |
| "eval_loss": 0.31582510471343994, |
| "eval_runtime": 99.4294, |
| "eval_samples_per_second": 10.057, |
| "eval_steps_per_second": 0.161, |
| "eval_wer": 0.10578224714755366, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.22, |
| "learning_rate": 9.570562392225396e-05, |
| "loss": 0.2719, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.24, |
| "learning_rate": 9.418011189656941e-05, |
| "loss": 0.2593, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.26, |
| "learning_rate": 9.24393588066941e-05, |
| "loss": 0.2545, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.28, |
| "learning_rate": 9.049184542670199e-05, |
| "loss": 0.2467, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.3, |
| "learning_rate": 8.834705984601708e-05, |
| "loss": 0.2369, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.3, |
| "eval_combined_wer": 0.1735586011342155, |
| "eval_f1_score": 0.6916524701873935, |
| "eval_label_f1": 0.8083475298126065, |
| "eval_loss": 0.29707884788513184, |
| "eval_runtime": 99.4491, |
| "eval_samples_per_second": 10.055, |
| "eval_steps_per_second": 0.161, |
| "eval_wer": 0.10667182363179269, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.32, |
| "learning_rate": 8.601545124439535e-05, |
| "loss": 0.2262, |
| "step": 1600 |
| }, |
| { |
| "epoch": 0.34, |
| "learning_rate": 8.350837898457143e-05, |
| "loss": 0.2196, |
| "step": 1700 |
| }, |
| { |
| "epoch": 0.36, |
| "learning_rate": 8.083805727058513e-05, |
| "loss": 0.2168, |
| "step": 1800 |
| }, |
| { |
| "epoch": 0.38, |
| "learning_rate": 7.801749564140724e-05, |
| "loss": 0.2063, |
| "step": 1900 |
| }, |
| { |
| "epoch": 0.4, |
| "learning_rate": 7.506043558977321e-05, |
| "loss": 0.1967, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.4, |
| "eval_combined_wer": 0.1633585381222432, |
| "eval_f1_score": 0.6915401301518438, |
| "eval_label_f1": 0.8095444685466376, |
| "eval_loss": 0.282340943813324, |
| "eval_runtime": 99.8374, |
| "eval_samples_per_second": 10.016, |
| "eval_steps_per_second": 0.16, |
| "eval_wer": 0.09990330690388706, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.42, |
| "learning_rate": 7.1981283615012e-05, |
| "loss": 0.1919, |
| "step": 2100 |
| }, |
| { |
| "epoch": 0.44, |
| "learning_rate": 6.879504103602935e-05, |
| "loss": 0.1825, |
| "step": 2200 |
| }, |
| { |
| "epoch": 0.46, |
| "learning_rate": 6.551723090639007e-05, |
| "loss": 0.1746, |
| "step": 2300 |
| }, |
| { |
| "epoch": 0.48, |
| "learning_rate": 6.216382238756146e-05, |
| "loss": 0.169, |
| "step": 2400 |
| }, |
| { |
| "epoch": 0.5, |
| "learning_rate": 5.875115294876381e-05, |
| "loss": 0.1623, |
| "step": 2500 |
| }, |
| { |
| "epoch": 0.5, |
| "eval_combined_wer": 0.16934467548834278, |
| "eval_f1_score": 0.708774904620602, |
| "eval_label_f1": 0.8249258160237388, |
| "eval_loss": 0.280351847410202, |
| "eval_runtime": 102.5306, |
| "eval_samples_per_second": 9.753, |
| "eval_steps_per_second": 0.156, |
| "eval_wer": 0.10520208857087604, |
| "step": 2500 |
| }, |
| { |
| "epoch": 0.52, |
| "learning_rate": 5.52958487724626e-05, |
| "loss": 0.1576, |
| "step": 2600 |
| }, |
| { |
| "epoch": 0.54, |
| "learning_rate": 5.1814743753278795e-05, |
| "loss": 0.1509, |
| "step": 2700 |
| }, |
| { |
| "epoch": 0.56, |
| "learning_rate": 4.832479748494643e-05, |
| "loss": 0.1492, |
| "step": 2800 |
| }, |
| { |
| "epoch": 1.0, |
| "learning_rate": 4.484301263487665e-05, |
| "loss": 0.1398, |
| "step": 2900 |
| }, |
| { |
| "epoch": 1.02, |
| "learning_rate": 4.1386352108871174e-05, |
| "loss": 0.1146, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.02, |
| "eval_combined_wer": 0.15930214240705734, |
| "eval_f1_score": 0.7011543394613081, |
| "eval_label_f1": 0.8106028217186831, |
| "eval_loss": 0.28203457593917847, |
| "eval_runtime": 98.6107, |
| "eval_samples_per_second": 10.141, |
| "eval_steps_per_second": 0.162, |
| "eval_wer": 0.09510732933668536, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.04, |
| "learning_rate": 3.797165640955041e-05, |
| "loss": 0.1121, |
| "step": 3100 |
| }, |
| { |
| "epoch": 1.06, |
| "learning_rate": 3.461556159111748e-05, |
| "loss": 0.1121, |
| "step": 3200 |
| }, |
| { |
| "epoch": 1.08, |
| "learning_rate": 3.1334418210174263e-05, |
| "loss": 0.1055, |
| "step": 3300 |
| }, |
| { |
| "epoch": 1.1, |
| "learning_rate": 2.8144211667453368e-05, |
| "loss": 0.0997, |
| "step": 3400 |
| }, |
| { |
| "epoch": 1.12, |
| "learning_rate": 2.506048432855247e-05, |
| "loss": 0.0938, |
| "step": 3500 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_combined_wer": 0.14996849401386264, |
| "eval_f1_score": 0.7204861111111112, |
| "eval_label_f1": 0.8237847222222224, |
| "eval_loss": 0.2791779041290283, |
| "eval_runtime": 97.2167, |
| "eval_samples_per_second": 10.286, |
| "eval_steps_per_second": 0.165, |
| "eval_wer": 0.08752659060143106, |
| "step": 3500 |
| }, |
| { |
| "epoch": 1.14, |
| "learning_rate": 2.209825980309151e-05, |
| "loss": 0.0905, |
| "step": 3600 |
| }, |
| { |
| "epoch": 1.16, |
| "learning_rate": 1.9271969751196776e-05, |
| "loss": 0.0883, |
| "step": 3700 |
| }, |
| { |
| "epoch": 1.18, |
| "learning_rate": 1.659538357390341e-05, |
| "loss": 0.0849, |
| "step": 3800 |
| }, |
| { |
| "epoch": 1.2, |
| "learning_rate": 1.4081541330017705e-05, |
| "loss": 0.0942, |
| "step": 3900 |
| }, |
| { |
| "epoch": 1.22, |
| "learning_rate": 1.1742690206261292e-05, |
| "loss": 0.1001, |
| "step": 4000 |
| }, |
| { |
| "epoch": 1.22, |
| "eval_combined_wer": 0.1549306868304978, |
| "eval_f1_score": 0.7071583514099783, |
| "eval_label_f1": 0.8060737527114967, |
| "eval_loss": 0.2750394344329834, |
| "eval_runtime": 101.2486, |
| "eval_samples_per_second": 9.877, |
| "eval_steps_per_second": 0.158, |
| "eval_wer": 0.09282537226842004, |
| "step": 4000 |
| }, |
| { |
| "epoch": 1.24, |
| "learning_rate": 9.590224850208646e-06, |
| "loss": 0.0936, |
| "step": 4100 |
| }, |
| { |
| "epoch": 1.26, |
| "learning_rate": 7.63463185670939e-06, |
| "loss": 0.0974, |
| "step": 4200 |
| }, |
| { |
| "epoch": 1.28, |
| "learning_rate": 5.885438678252342e-06, |
| "loss": 0.0877, |
| "step": 4300 |
| }, |
| { |
| "epoch": 1.3, |
| "learning_rate": 4.351167208174639e-06, |
| "loss": 0.081, |
| "step": 4400 |
| }, |
| { |
| "epoch": 1.32, |
| "learning_rate": 3.039292262854088e-06, |
| "loss": 0.0848, |
| "step": 4500 |
| }, |
| { |
| "epoch": 1.32, |
| "eval_combined_wer": 0.1471329552614997, |
| "eval_f1_score": 0.7243010752688172, |
| "eval_label_f1": 0.8318279569892474, |
| "eval_loss": 0.2740628123283386, |
| "eval_runtime": 102.3764, |
| "eval_samples_per_second": 9.768, |
| "eval_steps_per_second": 0.156, |
| "eval_wer": 0.08601817830206923, |
| "step": 4500 |
| } |
| ], |
| "logging_steps": 100, |
| "max_steps": 5000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 9223372036854775807, |
| "save_steps": 500, |
| "total_flos": 1.6621999604711424e+20, |
| "train_batch_size": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
