| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 12.67427122940431, | |
| "global_step": 10000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.63, | |
| "eval_loss": 2.2638113498687744, | |
| "eval_runtime": 149.9992, | |
| "eval_samples_per_second": 31.487, | |
| "eval_steps_per_second": 2.627, | |
| "eval_wer": 0.9359337678636492, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 9.970000000000001e-05, | |
| "loss": 2.6089, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "eval_loss": 0.7277476787567139, | |
| "eval_runtime": 148.226, | |
| "eval_samples_per_second": 31.864, | |
| "eval_steps_per_second": 2.658, | |
| "eval_wer": 0.24067851503987692, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_loss": 0.5800275802612305, | |
| "eval_runtime": 148.0487, | |
| "eval_samples_per_second": 31.902, | |
| "eval_steps_per_second": 2.661, | |
| "eval_wer": 0.1745475891664305, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 8.892222222222223e-05, | |
| "loss": 0.6019, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "eval_loss": 0.48867326974868774, | |
| "eval_runtime": 148.4526, | |
| "eval_samples_per_second": 31.815, | |
| "eval_steps_per_second": 2.654, | |
| "eval_wer": 0.15135014776729688, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "eval_loss": 0.466572642326355, | |
| "eval_runtime": 148.0238, | |
| "eval_samples_per_second": 31.907, | |
| "eval_steps_per_second": 2.662, | |
| "eval_wer": 0.14213999433221328, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 7.78111111111111e-05, | |
| "loss": 0.4722, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "eval_loss": 0.44257038831710815, | |
| "eval_runtime": 149.7003, | |
| "eval_samples_per_second": 31.55, | |
| "eval_steps_per_second": 2.632, | |
| "eval_wer": 0.14505485607870128, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "eval_loss": 0.41759932041168213, | |
| "eval_runtime": 149.0588, | |
| "eval_samples_per_second": 31.685, | |
| "eval_steps_per_second": 2.643, | |
| "eval_wer": 0.12481276061697907, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 6.671111111111111e-05, | |
| "loss": 0.4278, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "eval_loss": 0.4364745318889618, | |
| "eval_runtime": 148.5505, | |
| "eval_samples_per_second": 31.794, | |
| "eval_steps_per_second": 2.652, | |
| "eval_wer": 0.12388162422573985, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "eval_loss": 0.3815610408782959, | |
| "eval_runtime": 149.3382, | |
| "eval_samples_per_second": 31.626, | |
| "eval_steps_per_second": 2.638, | |
| "eval_wer": 0.1177280272053763, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "learning_rate": 5.560000000000001e-05, | |
| "loss": 0.369, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "eval_loss": 0.4113306403160095, | |
| "eval_runtime": 159.7075, | |
| "eval_samples_per_second": 29.573, | |
| "eval_steps_per_second": 2.467, | |
| "eval_wer": 0.11716124853244808, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "eval_loss": 0.3862614035606384, | |
| "eval_runtime": 152.4609, | |
| "eval_samples_per_second": 30.978, | |
| "eval_steps_per_second": 2.584, | |
| "eval_wer": 0.1230112141208858, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "learning_rate": 4.448888888888889e-05, | |
| "loss": 0.341, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "eval_loss": 0.384976863861084, | |
| "eval_runtime": 159.7493, | |
| "eval_samples_per_second": 29.565, | |
| "eval_steps_per_second": 2.466, | |
| "eval_wer": 0.1116149143759362, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 8.24, | |
| "eval_loss": 0.401400089263916, | |
| "eval_runtime": 155.917, | |
| "eval_samples_per_second": 30.292, | |
| "eval_steps_per_second": 2.527, | |
| "eval_wer": 0.11406420792680458, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "learning_rate": 3.337777777777778e-05, | |
| "loss": 0.3119, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "eval_loss": 0.39530250430107117, | |
| "eval_runtime": 165.7417, | |
| "eval_samples_per_second": 28.496, | |
| "eval_steps_per_second": 2.377, | |
| "eval_wer": 0.10782964252459415, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 9.51, | |
| "eval_loss": 0.4018384516239166, | |
| "eval_runtime": 163.4263, | |
| "eval_samples_per_second": 28.9, | |
| "eval_steps_per_second": 2.411, | |
| "eval_wer": 0.10801182138374965, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 10.14, | |
| "learning_rate": 2.2277777777777778e-05, | |
| "loss": 0.3008, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 10.14, | |
| "eval_loss": 0.3963571786880493, | |
| "eval_runtime": 172.5171, | |
| "eval_samples_per_second": 27.377, | |
| "eval_steps_per_second": 2.284, | |
| "eval_wer": 0.10744504271082142, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 10.77, | |
| "eval_loss": 0.39167362451553345, | |
| "eval_runtime": 165.6389, | |
| "eval_samples_per_second": 28.514, | |
| "eval_steps_per_second": 2.379, | |
| "eval_wer": 0.10780940042913242, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 11.41, | |
| "learning_rate": 1.1166666666666668e-05, | |
| "loss": 0.2741, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 11.41, | |
| "eval_loss": 0.3961273431777954, | |
| "eval_runtime": 164.9191, | |
| "eval_samples_per_second": 28.638, | |
| "eval_steps_per_second": 2.389, | |
| "eval_wer": 0.10568398040565159, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "eval_loss": 0.39744970202445984, | |
| "eval_runtime": 164.8733, | |
| "eval_samples_per_second": 28.646, | |
| "eval_steps_per_second": 2.39, | |
| "eval_wer": 0.10529938059187888, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 12.67, | |
| "learning_rate": 5.555555555555556e-08, | |
| "loss": 0.2531, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 12.67, | |
| "eval_loss": 0.40417608618736267, | |
| "eval_runtime": 164.6908, | |
| "eval_samples_per_second": 28.678, | |
| "eval_steps_per_second": 2.392, | |
| "eval_wer": 0.10485405449172099, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 12.67, | |
| "step": 10000, | |
| "total_flos": 2.7136568655380324e+19, | |
| "train_loss": 0.5960800704956055, | |
| "train_runtime": 13155.4001, | |
| "train_samples_per_second": 18.243, | |
| "train_steps_per_second": 0.76 | |
| } | |
| ], | |
| "max_steps": 10000, | |
| "num_train_epochs": 13, | |
| "total_flos": 2.7136568655380324e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |