| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 50.0, | |
| "global_step": 40500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 4.9800000000000004e-05, | |
| "loss": 4.6222, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_loss": 5.947904586791992, | |
| "eval_runtime": 35.3873, | |
| "eval_samples_per_second": 21.505, | |
| "eval_steps_per_second": 1.356, | |
| "eval_wer": 0.5473815461346634, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 4.8084615384615386e-05, | |
| "loss": 1.1362, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "eval_loss": 7.979872226715088, | |
| "eval_runtime": 34.3508, | |
| "eval_samples_per_second": 22.154, | |
| "eval_steps_per_second": 1.397, | |
| "eval_wer": 0.509440684004275, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "learning_rate": 4.616153846153846e-05, | |
| "loss": 0.7814, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "eval_loss": 5.032960891723633, | |
| "eval_runtime": 33.9113, | |
| "eval_samples_per_second": 22.441, | |
| "eval_steps_per_second": 1.415, | |
| "eval_wer": 0.47239045244032773, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 7.41, | |
| "learning_rate": 4.423974358974359e-05, | |
| "loss": 0.6281, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.41, | |
| "eval_loss": 2.3483684062957764, | |
| "eval_runtime": 35.8392, | |
| "eval_samples_per_second": 21.234, | |
| "eval_steps_per_second": 1.339, | |
| "eval_wer": 0.5019593872461703, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 9.26, | |
| "learning_rate": 4.2316666666666674e-05, | |
| "loss": 0.5472, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 9.26, | |
| "eval_loss": 2.249516487121582, | |
| "eval_runtime": 35.6948, | |
| "eval_samples_per_second": 21.32, | |
| "eval_steps_per_second": 1.345, | |
| "eval_wer": 0.47933737085856787, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 11.11, | |
| "learning_rate": 4.039358974358974e-05, | |
| "loss": 0.4827, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 11.11, | |
| "eval_loss": 1.1529797315597534, | |
| "eval_runtime": 35.6048, | |
| "eval_samples_per_second": 21.373, | |
| "eval_steps_per_second": 1.348, | |
| "eval_wer": 0.47684360527253294, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 12.96, | |
| "learning_rate": 3.847051282051282e-05, | |
| "loss": 0.4327, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 12.96, | |
| "eval_loss": 1.6159653663635254, | |
| "eval_runtime": 34.1129, | |
| "eval_samples_per_second": 22.308, | |
| "eval_steps_per_second": 1.407, | |
| "eval_wer": 0.4645529034556466, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 14.81, | |
| "learning_rate": 3.6548717948717956e-05, | |
| "loss": 0.3989, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 14.81, | |
| "eval_loss": 3.263315439224243, | |
| "eval_runtime": 34.7016, | |
| "eval_samples_per_second": 21.93, | |
| "eval_steps_per_second": 1.383, | |
| "eval_wer": 0.47025293908086924, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "learning_rate": 3.4625641025641024e-05, | |
| "loss": 0.3522, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "eval_loss": 2.2337419986724854, | |
| "eval_runtime": 35.6982, | |
| "eval_samples_per_second": 21.318, | |
| "eval_steps_per_second": 1.345, | |
| "eval_wer": 0.4707873174207339, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 18.52, | |
| "learning_rate": 3.2702564102564105e-05, | |
| "loss": 0.3201, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 18.52, | |
| "eval_loss": 3.6878626346588135, | |
| "eval_runtime": 36.4839, | |
| "eval_samples_per_second": 20.859, | |
| "eval_steps_per_second": 1.316, | |
| "eval_wer": 0.45653722835767724, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 20.37, | |
| "learning_rate": 3.078076923076923e-05, | |
| "loss": 0.2899, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 20.37, | |
| "eval_loss": 5.438948631286621, | |
| "eval_runtime": 34.4996, | |
| "eval_samples_per_second": 22.058, | |
| "eval_steps_per_second": 1.391, | |
| "eval_wer": 0.45992162451015317, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 22.22, | |
| "learning_rate": 2.885897435897436e-05, | |
| "loss": 0.2776, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 22.22, | |
| "eval_loss": 3.528372049331665, | |
| "eval_runtime": 35.9097, | |
| "eval_samples_per_second": 21.192, | |
| "eval_steps_per_second": 1.337, | |
| "eval_wer": 0.4536872105450659, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 24.07, | |
| "learning_rate": 2.6935897435897438e-05, | |
| "loss": 0.2574, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 24.07, | |
| "eval_loss": 2.1759419441223145, | |
| "eval_runtime": 34.313, | |
| "eval_samples_per_second": 22.178, | |
| "eval_steps_per_second": 1.399, | |
| "eval_wer": 0.464909155682223, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 25.93, | |
| "learning_rate": 2.5012820512820513e-05, | |
| "loss": 0.2378, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 25.93, | |
| "eval_loss": 3.390052080154419, | |
| "eval_runtime": 34.4471, | |
| "eval_samples_per_second": 22.092, | |
| "eval_steps_per_second": 1.393, | |
| "eval_wer": 0.4447809048806555, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 27.78, | |
| "learning_rate": 2.3092307692307694e-05, | |
| "loss": 0.217, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 27.78, | |
| "eval_loss": 1.163241982460022, | |
| "eval_runtime": 36.0254, | |
| "eval_samples_per_second": 21.124, | |
| "eval_steps_per_second": 1.332, | |
| "eval_wer": 0.45653722835767724, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 29.63, | |
| "learning_rate": 2.1169230769230768e-05, | |
| "loss": 0.2115, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 29.63, | |
| "eval_loss": 1.7441022396087646, | |
| "eval_runtime": 35.1297, | |
| "eval_samples_per_second": 21.663, | |
| "eval_steps_per_second": 1.366, | |
| "eval_wer": 0.42322764517278233, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 31.48, | |
| "learning_rate": 1.9246153846153846e-05, | |
| "loss": 0.1959, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 31.48, | |
| "eval_loss": 3.4991888999938965, | |
| "eval_runtime": 36.7374, | |
| "eval_samples_per_second": 20.715, | |
| "eval_steps_per_second": 1.307, | |
| "eval_wer": 0.4303526897043107, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "learning_rate": 1.7323076923076924e-05, | |
| "loss": 0.187, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "eval_loss": 3.6162784099578857, | |
| "eval_runtime": 34.8093, | |
| "eval_samples_per_second": 21.862, | |
| "eval_steps_per_second": 1.379, | |
| "eval_wer": 0.43694335589597433, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 35.19, | |
| "learning_rate": 1.540128205128205e-05, | |
| "loss": 0.1748, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 35.19, | |
| "eval_loss": 3.603774309158325, | |
| "eval_runtime": 35.9258, | |
| "eval_samples_per_second": 21.183, | |
| "eval_steps_per_second": 1.336, | |
| "eval_wer": 0.4467402921268258, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 37.04, | |
| "learning_rate": 1.347820512820513e-05, | |
| "loss": 0.17, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 37.04, | |
| "eval_loss": 2.970829486846924, | |
| "eval_runtime": 35.2981, | |
| "eval_samples_per_second": 21.559, | |
| "eval_steps_per_second": 1.36, | |
| "eval_wer": 0.43623085144282153, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 38.89, | |
| "learning_rate": 1.1557692307692308e-05, | |
| "loss": 0.159, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 38.89, | |
| "eval_loss": 3.2044625282287598, | |
| "eval_runtime": 34.6143, | |
| "eval_samples_per_second": 21.985, | |
| "eval_steps_per_second": 1.387, | |
| "eval_wer": 0.42785892411827575, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 40.74, | |
| "learning_rate": 9.635897435897436e-06, | |
| "loss": 0.153, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 40.74, | |
| "eval_loss": 3.2426888942718506, | |
| "eval_runtime": 35.08, | |
| "eval_samples_per_second": 21.693, | |
| "eval_steps_per_second": 1.368, | |
| "eval_wer": 0.42874955468471676, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 42.59, | |
| "learning_rate": 7.712820512820514e-06, | |
| "loss": 0.1463, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 42.59, | |
| "eval_loss": 3.5439305305480957, | |
| "eval_runtime": 36.6846, | |
| "eval_samples_per_second": 20.744, | |
| "eval_steps_per_second": 1.308, | |
| "eval_wer": 0.4269682935518347, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 44.44, | |
| "learning_rate": 5.78974358974359e-06, | |
| "loss": 0.139, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 44.44, | |
| "eval_loss": 3.938081741333008, | |
| "eval_runtime": 34.7219, | |
| "eval_samples_per_second": 21.917, | |
| "eval_steps_per_second": 1.382, | |
| "eval_wer": 0.41503384396152476, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 46.3, | |
| "learning_rate": 3.867948717948718e-06, | |
| "loss": 0.1352, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 46.3, | |
| "eval_loss": 4.174356937408447, | |
| "eval_runtime": 36.3496, | |
| "eval_samples_per_second": 20.936, | |
| "eval_steps_per_second": 1.321, | |
| "eval_wer": 0.4091556822230139, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 48.15, | |
| "learning_rate": 1.9461538461538464e-06, | |
| "loss": 0.1369, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 48.15, | |
| "eval_loss": 4.227924823760986, | |
| "eval_runtime": 34.5902, | |
| "eval_samples_per_second": 22.0, | |
| "eval_steps_per_second": 1.388, | |
| "eval_wer": 0.4153900961881012, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 2.3076923076923076e-08, | |
| "loss": 0.1273, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_loss": 4.169058322906494, | |
| "eval_runtime": 34.2804, | |
| "eval_samples_per_second": 22.199, | |
| "eval_steps_per_second": 1.4, | |
| "eval_wer": 0.4132525828286427, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 40500, | |
| "total_flos": 5.1003805267852526e+20, | |
| "train_loss": 0.0, | |
| "train_runtime": 69.3888, | |
| "train_samples_per_second": 3734.751, | |
| "train_steps_per_second": 116.733 | |
| } | |
| ], | |
| "max_steps": 8100, | |
| "num_train_epochs": 10, | |
| "total_flos": 5.1003805267852526e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |