| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 28.846153846153847, | |
| "global_step": 10500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 4.99e-05, | |
| "loss": 3.5823, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "eval_loss": 3.32504940032959, | |
| "eval_runtime": 45.0187, | |
| "eval_samples_per_second": 14.394, | |
| "eval_steps_per_second": 0.466, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 9.970000000000001e-05, | |
| "loss": 1.6148, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "eval_loss": 0.9707893133163452, | |
| "eval_runtime": 41.6379, | |
| "eval_samples_per_second": 15.563, | |
| "eval_steps_per_second": 0.504, | |
| "eval_wer": 0.5453111956219274, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 9.50100806451613e-05, | |
| "loss": 0.7901, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "eval_loss": 0.7987369298934937, | |
| "eval_runtime": 42.0128, | |
| "eval_samples_per_second": 15.424, | |
| "eval_steps_per_second": 0.5, | |
| "eval_wer": 0.4292737222892125, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "learning_rate": 8.996975806451613e-05, | |
| "loss": 0.5529, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "eval_loss": 0.803723156452179, | |
| "eval_runtime": 41.9975, | |
| "eval_samples_per_second": 15.43, | |
| "eval_steps_per_second": 0.5, | |
| "eval_wer": 0.4209257026249884, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.87, | |
| "learning_rate": 8.492943548387097e-05, | |
| "loss": 0.4404, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.87, | |
| "eval_loss": 0.7943486571311951, | |
| "eval_runtime": 42.0152, | |
| "eval_samples_per_second": 15.423, | |
| "eval_steps_per_second": 0.5, | |
| "eval_wer": 0.45524533902235415, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.24, | |
| "learning_rate": 7.988911290322581e-05, | |
| "loss": 0.3528, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.24, | |
| "eval_loss": 0.8733369708061218, | |
| "eval_runtime": 42.4523, | |
| "eval_samples_per_second": 15.264, | |
| "eval_steps_per_second": 0.495, | |
| "eval_wer": 0.37658844263055374, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 9.62, | |
| "learning_rate": 7.484879032258065e-05, | |
| "loss": 0.3008, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.62, | |
| "eval_loss": 0.914967954158783, | |
| "eval_runtime": 42.0647, | |
| "eval_samples_per_second": 15.405, | |
| "eval_steps_per_second": 0.499, | |
| "eval_wer": 0.3725999443465356, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 10.99, | |
| "learning_rate": 6.98084677419355e-05, | |
| "loss": 0.2625, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.99, | |
| "eval_loss": 0.9753163456916809, | |
| "eval_runtime": 41.804, | |
| "eval_samples_per_second": 15.501, | |
| "eval_steps_per_second": 0.502, | |
| "eval_wer": 0.3690752249327521, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 12.36, | |
| "learning_rate": 6.476814516129032e-05, | |
| "loss": 0.2153, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 12.36, | |
| "eval_loss": 0.9292902946472168, | |
| "eval_runtime": 53.6443, | |
| "eval_samples_per_second": 12.08, | |
| "eval_steps_per_second": 0.391, | |
| "eval_wer": 0.35284296447453856, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 13.74, | |
| "learning_rate": 5.9727822580645166e-05, | |
| "loss": 0.1926, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 13.74, | |
| "eval_loss": 0.9629846215248108, | |
| "eval_runtime": 43.3856, | |
| "eval_samples_per_second": 14.936, | |
| "eval_steps_per_second": 0.484, | |
| "eval_wer": 0.3697245153510806, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 15.11, | |
| "learning_rate": 5.46875e-05, | |
| "loss": 0.1634, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 15.11, | |
| "eval_loss": 0.9719225764274597, | |
| "eval_runtime": 43.1527, | |
| "eval_samples_per_second": 15.016, | |
| "eval_steps_per_second": 0.487, | |
| "eval_wer": 0.37473332714961505, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 16.48, | |
| "learning_rate": 4.964717741935484e-05, | |
| "loss": 0.1422, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 16.48, | |
| "eval_loss": 1.0191898345947266, | |
| "eval_runtime": 43.4487, | |
| "eval_samples_per_second": 14.914, | |
| "eval_steps_per_second": 0.483, | |
| "eval_wer": 0.38085520823671276, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 17.86, | |
| "learning_rate": 4.460685483870968e-05, | |
| "loss": 0.1253, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 17.86, | |
| "eval_loss": 1.0450010299682617, | |
| "eval_runtime": 42.8089, | |
| "eval_samples_per_second": 15.137, | |
| "eval_steps_per_second": 0.491, | |
| "eval_wer": 0.37686670995269456, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 19.23, | |
| "learning_rate": 3.956653225806452e-05, | |
| "loss": 0.1068, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 19.23, | |
| "eval_loss": 1.0906848907470703, | |
| "eval_runtime": 43.3307, | |
| "eval_samples_per_second": 14.955, | |
| "eval_steps_per_second": 0.485, | |
| "eval_wer": 0.359243112883777, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 20.6, | |
| "learning_rate": 3.4526209677419356e-05, | |
| "loss": 0.095, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 20.6, | |
| "eval_loss": 1.0443437099456787, | |
| "eval_runtime": 53.9592, | |
| "eval_samples_per_second": 12.009, | |
| "eval_steps_per_second": 0.389, | |
| "eval_wer": 0.35794453204711996, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 21.98, | |
| "learning_rate": 2.9495967741935488e-05, | |
| "loss": 0.0814, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 21.98, | |
| "eval_loss": 1.140635371208191, | |
| "eval_runtime": 43.1083, | |
| "eval_samples_per_second": 15.032, | |
| "eval_steps_per_second": 0.487, | |
| "eval_wer": 0.36258232074946667, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 23.35, | |
| "learning_rate": 2.4455645161290326e-05, | |
| "loss": 0.069, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 23.35, | |
| "eval_loss": 1.1894210577011108, | |
| "eval_runtime": 42.7802, | |
| "eval_samples_per_second": 15.147, | |
| "eval_steps_per_second": 0.491, | |
| "eval_wer": 0.3690752249327521, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 24.73, | |
| "learning_rate": 1.942540322580645e-05, | |
| "loss": 0.0623, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 24.73, | |
| "eval_loss": 1.2040750980377197, | |
| "eval_runtime": 43.2859, | |
| "eval_samples_per_second": 14.97, | |
| "eval_steps_per_second": 0.485, | |
| "eval_wer": 0.3687042018365643, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 26.1, | |
| "learning_rate": 1.4385080645161292e-05, | |
| "loss": 0.0554, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 26.1, | |
| "eval_loss": 1.2136509418487549, | |
| "eval_runtime": 42.5198, | |
| "eval_samples_per_second": 15.24, | |
| "eval_steps_per_second": 0.494, | |
| "eval_wer": 0.35720248585474446, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 27.47, | |
| "learning_rate": 9.344758064516129e-06, | |
| "loss": 0.0475, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 27.47, | |
| "eval_loss": 1.2355363368988037, | |
| "eval_runtime": 43.577, | |
| "eval_samples_per_second": 14.87, | |
| "eval_steps_per_second": 0.482, | |
| "eval_wer": 0.3565531954364159, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 28.85, | |
| "learning_rate": 4.304435483870968e-06, | |
| "loss": 0.042, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 28.85, | |
| "eval_loss": 1.2368338108062744, | |
| "eval_runtime": 43.2519, | |
| "eval_samples_per_second": 14.982, | |
| "eval_steps_per_second": 0.486, | |
| "eval_wer": 0.35553288192189964, | |
| "step": 10500 | |
| } | |
| ], | |
| "max_steps": 10920, | |
| "num_train_epochs": 30, | |
| "total_flos": 1.5691152295536976e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |