| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 119.99875156054931, | |
| "global_step": 24000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 7.960000000000001e-05, | |
| "loss": 7.6701, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.900460958480835, | |
| "eval_runtime": 252.4053, | |
| "eval_samples_per_second": 20.653, | |
| "eval_steps_per_second": 2.583, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 9.873191489361703e-05, | |
| "loss": 1.0784, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.4933229088783264, | |
| "eval_runtime": 254.2522, | |
| "eval_samples_per_second": 20.503, | |
| "eval_steps_per_second": 2.564, | |
| "eval_wer": 0.46976841118086854, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 9.702978723404256e-05, | |
| "loss": 0.3128, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.3993886709213257, | |
| "eval_runtime": 252.398, | |
| "eval_samples_per_second": 20.654, | |
| "eval_steps_per_second": 2.583, | |
| "eval_wer": 0.37059421158204603, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 9.532765957446809e-05, | |
| "loss": 0.2158, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.3813767731189728, | |
| "eval_runtime": 260.377, | |
| "eval_samples_per_second": 20.021, | |
| "eval_steps_per_second": 2.504, | |
| "eval_wer": 0.3755438038919426, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 9.362553191489361e-05, | |
| "loss": 0.1716, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.3713369071483612, | |
| "eval_runtime": 256.1576, | |
| "eval_samples_per_second": 20.351, | |
| "eval_steps_per_second": 2.545, | |
| "eval_wer": 0.33745799359158046, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 9.192340425531915e-05, | |
| "loss": 0.1419, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.3995266556739807, | |
| "eval_runtime": 252.1996, | |
| "eval_samples_per_second": 20.67, | |
| "eval_steps_per_second": 2.585, | |
| "eval_wer": 0.35183786177612214, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 9.022127659574468e-05, | |
| "loss": 0.1236, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.41960370540618896, | |
| "eval_runtime": 256.7174, | |
| "eval_samples_per_second": 20.306, | |
| "eval_steps_per_second": 2.54, | |
| "eval_wer": 0.34042774897751843, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 8.851914893617021e-05, | |
| "loss": 0.1136, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.3961251676082611, | |
| "eval_runtime": 257.0042, | |
| "eval_samples_per_second": 20.284, | |
| "eval_steps_per_second": 2.537, | |
| "eval_wer": 0.32029072342199183, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 8.682127659574468e-05, | |
| "loss": 0.1018, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.4284366965293884, | |
| "eval_runtime": 253.1831, | |
| "eval_samples_per_second": 20.59, | |
| "eval_steps_per_second": 2.575, | |
| "eval_wer": 0.3373798421340558, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 8.511914893617021e-05, | |
| "loss": 0.0948, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.4246082305908203, | |
| "eval_runtime": 250.1867, | |
| "eval_samples_per_second": 20.836, | |
| "eval_steps_per_second": 2.606, | |
| "eval_wer": 0.32953864589574594, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "learning_rate": 8.341702127659575e-05, | |
| "loss": 0.0837, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 0.41016092896461487, | |
| "eval_runtime": 255.5501, | |
| "eval_samples_per_second": 20.399, | |
| "eval_steps_per_second": 2.551, | |
| "eval_wer": 0.31726886706437074, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 8.171489361702128e-05, | |
| "loss": 0.0823, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 0.42129427194595337, | |
| "eval_runtime": 254.0415, | |
| "eval_samples_per_second": 20.52, | |
| "eval_steps_per_second": 2.567, | |
| "eval_wer": 0.3245109021283247, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "learning_rate": 8.001276595744681e-05, | |
| "loss": 0.079, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 0.4289919435977936, | |
| "eval_runtime": 256.8225, | |
| "eval_samples_per_second": 20.298, | |
| "eval_steps_per_second": 2.539, | |
| "eval_wer": 0.3159923932581343, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "learning_rate": 7.831063829787234e-05, | |
| "loss": 0.0736, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 0.45376530289649963, | |
| "eval_runtime": 250.495, | |
| "eval_samples_per_second": 20.811, | |
| "eval_steps_per_second": 2.603, | |
| "eval_wer": 0.3086982572224972, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 7.66127659574468e-05, | |
| "loss": 0.0702, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 0.4636918306350708, | |
| "eval_runtime": 251.4184, | |
| "eval_samples_per_second": 20.734, | |
| "eval_steps_per_second": 2.593, | |
| "eval_wer": 0.31151170969338576, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "learning_rate": 7.491063829787235e-05, | |
| "loss": 0.0668, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_loss": 0.4827902615070343, | |
| "eval_runtime": 248.7074, | |
| "eval_samples_per_second": 20.96, | |
| "eval_steps_per_second": 2.622, | |
| "eval_wer": 0.3185192903847657, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "learning_rate": 7.320851063829788e-05, | |
| "loss": 0.0646, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_loss": 0.4427362084388733, | |
| "eval_runtime": 252.8917, | |
| "eval_samples_per_second": 20.614, | |
| "eval_steps_per_second": 2.578, | |
| "eval_wer": 0.311381457264178, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "learning_rate": 7.15063829787234e-05, | |
| "loss": 0.06, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_loss": 0.4571715295314789, | |
| "eval_runtime": 252.2048, | |
| "eval_samples_per_second": 20.67, | |
| "eval_steps_per_second": 2.585, | |
| "eval_wer": 0.311615911636752, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "learning_rate": 6.980425531914893e-05, | |
| "loss": 0.0584, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_loss": 0.4944392740726471, | |
| "eval_runtime": 251.4845, | |
| "eval_samples_per_second": 20.729, | |
| "eval_steps_per_second": 2.593, | |
| "eval_wer": 0.30804699507645816, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 6.810212765957446e-05, | |
| "loss": 0.0545, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 0.4599160850048065, | |
| "eval_runtime": 253.7962, | |
| "eval_samples_per_second": 20.54, | |
| "eval_steps_per_second": 2.569, | |
| "eval_wer": 0.3099747310287337, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "learning_rate": 6.64e-05, | |
| "loss": 0.0536, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_loss": 0.4586937427520752, | |
| "eval_runtime": 256.4427, | |
| "eval_samples_per_second": 20.328, | |
| "eval_steps_per_second": 2.542, | |
| "eval_wer": 0.30338395811081875, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "learning_rate": 6.469787234042553e-05, | |
| "loss": 0.0506, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_loss": 0.46839669346809387, | |
| "eval_runtime": 253.5984, | |
| "eval_samples_per_second": 20.556, | |
| "eval_steps_per_second": 2.571, | |
| "eval_wer": 0.3020814338187407, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "learning_rate": 6.299574468085107e-05, | |
| "loss": 0.0504, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_loss": 0.4694036841392517, | |
| "eval_runtime": 252.7537, | |
| "eval_samples_per_second": 20.625, | |
| "eval_steps_per_second": 2.58, | |
| "eval_wer": 0.30710917758616196, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "learning_rate": 6.12936170212766e-05, | |
| "loss": 0.048, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_loss": 0.4677378535270691, | |
| "eval_runtime": 252.9941, | |
| "eval_samples_per_second": 20.605, | |
| "eval_steps_per_second": 2.577, | |
| "eval_wer": 0.2971578919946857, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 5.9591489361702134e-05, | |
| "loss": 0.045, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_loss": 0.46043330430984497, | |
| "eval_runtime": 249.2435, | |
| "eval_samples_per_second": 20.915, | |
| "eval_steps_per_second": 2.616, | |
| "eval_wer": 0.29778310365488314, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "learning_rate": 5.7893617021276604e-05, | |
| "loss": 0.0436, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_loss": 0.4766680598258972, | |
| "eval_runtime": 251.6774, | |
| "eval_samples_per_second": 20.713, | |
| "eval_steps_per_second": 2.591, | |
| "eval_wer": 0.29731419490973504, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "learning_rate": 5.619148936170213e-05, | |
| "loss": 0.0422, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_loss": 0.4668702185153961, | |
| "eval_runtime": 252.6671, | |
| "eval_samples_per_second": 20.632, | |
| "eval_steps_per_second": 2.58, | |
| "eval_wer": 0.2975486492823091, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "learning_rate": 5.448936170212766e-05, | |
| "loss": 0.0404, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_loss": 0.4487351179122925, | |
| "eval_runtime": 250.621, | |
| "eval_samples_per_second": 20.8, | |
| "eval_steps_per_second": 2.602, | |
| "eval_wer": 0.2965066298486467, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "learning_rate": 5.278723404255319e-05, | |
| "loss": 0.04, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_loss": 0.49005138874053955, | |
| "eval_runtime": 254.1559, | |
| "eval_samples_per_second": 20.511, | |
| "eval_steps_per_second": 2.565, | |
| "eval_wer": 0.2988511735743872, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "learning_rate": 5.108510638297873e-05, | |
| "loss": 0.0375, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_loss": 0.46530571579933167, | |
| "eval_runtime": 255.1439, | |
| "eval_samples_per_second": 20.432, | |
| "eval_steps_per_second": 2.555, | |
| "eval_wer": 0.2963763774194389, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "learning_rate": 4.938297872340426e-05, | |
| "loss": 0.0353, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_loss": 0.5175049901008606, | |
| "eval_runtime": 253.9576, | |
| "eval_samples_per_second": 20.527, | |
| "eval_steps_per_second": 2.567, | |
| "eval_wer": 0.2963242764477558, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "learning_rate": 4.768085106382979e-05, | |
| "loss": 0.0346, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_loss": 0.5116600394248962, | |
| "eval_runtime": 256.8702, | |
| "eval_samples_per_second": 20.294, | |
| "eval_steps_per_second": 2.538, | |
| "eval_wer": 0.2914528355953839, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "learning_rate": 4.597872340425532e-05, | |
| "loss": 0.0341, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_loss": 0.4671209454536438, | |
| "eval_runtime": 255.8006, | |
| "eval_samples_per_second": 20.379, | |
| "eval_steps_per_second": 2.549, | |
| "eval_wer": 0.2945528434105296, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "learning_rate": 4.4280851063829785e-05, | |
| "loss": 0.0332, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_loss": 0.48402974009513855, | |
| "eval_runtime": 251.9224, | |
| "eval_samples_per_second": 20.693, | |
| "eval_steps_per_second": 2.588, | |
| "eval_wer": 0.2882746763227134, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "learning_rate": 4.257872340425532e-05, | |
| "loss": 0.0307, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_loss": 0.4735446870326996, | |
| "eval_runtime": 251.6191, | |
| "eval_samples_per_second": 20.718, | |
| "eval_steps_per_second": 2.591, | |
| "eval_wer": 0.2853570219084586, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "learning_rate": 4.0876595744680854e-05, | |
| "loss": 0.0296, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_loss": 0.49666687846183777, | |
| "eval_runtime": 255.1334, | |
| "eval_samples_per_second": 20.432, | |
| "eval_steps_per_second": 2.556, | |
| "eval_wer": 0.2870242530023185, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "learning_rate": 3.917446808510639e-05, | |
| "loss": 0.0288, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_loss": 0.4831916093826294, | |
| "eval_runtime": 250.196, | |
| "eval_samples_per_second": 20.836, | |
| "eval_steps_per_second": 2.606, | |
| "eval_wer": 0.28111079271628414, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "learning_rate": 3.747234042553192e-05, | |
| "loss": 0.0278, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_loss": 0.5073911547660828, | |
| "eval_runtime": 251.6593, | |
| "eval_samples_per_second": 20.715, | |
| "eval_steps_per_second": 2.591, | |
| "eval_wer": 0.2897595540156824, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "learning_rate": 3.577021276595745e-05, | |
| "loss": 0.0264, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_loss": 0.49896690249443054, | |
| "eval_runtime": 253.2919, | |
| "eval_samples_per_second": 20.581, | |
| "eval_steps_per_second": 2.574, | |
| "eval_wer": 0.28066793445697763, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "learning_rate": 3.406808510638298e-05, | |
| "loss": 0.0259, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_loss": 0.4933984577655792, | |
| "eval_runtime": 250.843, | |
| "eval_samples_per_second": 20.782, | |
| "eval_steps_per_second": 2.599, | |
| "eval_wer": 0.28478391121994423, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "learning_rate": 3.2365957446808515e-05, | |
| "loss": 0.0256, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_loss": 0.5108290314674377, | |
| "eval_runtime": 254.0809, | |
| "eval_samples_per_second": 20.517, | |
| "eval_steps_per_second": 2.566, | |
| "eval_wer": 0.2834032354703415, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "learning_rate": 3.066382978723404e-05, | |
| "loss": 0.0241, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_loss": 0.5010423064231873, | |
| "eval_runtime": 253.2137, | |
| "eval_samples_per_second": 20.587, | |
| "eval_steps_per_second": 2.575, | |
| "eval_wer": 0.28137129757469975, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "learning_rate": 2.8961702127659574e-05, | |
| "loss": 0.0225, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "eval_loss": 0.5098214745521545, | |
| "eval_runtime": 251.7862, | |
| "eval_samples_per_second": 20.704, | |
| "eval_steps_per_second": 2.589, | |
| "eval_wer": 0.2809544898012348, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "learning_rate": 2.725957446808511e-05, | |
| "loss": 0.0214, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_loss": 0.5001631379127502, | |
| "eval_runtime": 251.7107, | |
| "eval_samples_per_second": 20.71, | |
| "eval_steps_per_second": 2.59, | |
| "eval_wer": 0.27485867611430953, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "learning_rate": 2.5565957446808516e-05, | |
| "loss": 0.0212, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_loss": 0.5039480328559875, | |
| "eval_runtime": 256.5398, | |
| "eval_samples_per_second": 20.32, | |
| "eval_steps_per_second": 2.542, | |
| "eval_wer": 0.27517128194440826, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "learning_rate": 2.3863829787234044e-05, | |
| "loss": 0.0198, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_loss": 0.5044199824333191, | |
| "eval_runtime": 250.6127, | |
| "eval_samples_per_second": 20.801, | |
| "eval_steps_per_second": 2.602, | |
| "eval_wer": 0.2775939771276734, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "learning_rate": 2.2161702127659575e-05, | |
| "loss": 0.0194, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "eval_loss": 0.5091202855110168, | |
| "eval_runtime": 252.5846, | |
| "eval_samples_per_second": 20.639, | |
| "eval_steps_per_second": 2.581, | |
| "eval_wer": 0.2784796936462865, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "learning_rate": 2.0459574468085106e-05, | |
| "loss": 0.0193, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_loss": 0.5078316926956177, | |
| "eval_runtime": 254.2064, | |
| "eval_samples_per_second": 20.507, | |
| "eval_steps_per_second": 2.565, | |
| "eval_wer": 0.2763435538072785, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "learning_rate": 1.875744680851064e-05, | |
| "loss": 0.0189, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "eval_loss": 0.50649094581604, | |
| "eval_runtime": 250.3726, | |
| "eval_samples_per_second": 20.821, | |
| "eval_steps_per_second": 2.604, | |
| "eval_wer": 0.2715502644124313, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 1.7055319148936173e-05, | |
| "loss": 0.0169, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_loss": 0.5111807584762573, | |
| "eval_runtime": 251.2786, | |
| "eval_samples_per_second": 20.746, | |
| "eval_steps_per_second": 2.595, | |
| "eval_wer": 0.2712897595540157, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 102.0, | |
| "learning_rate": 1.5353191489361704e-05, | |
| "loss": 0.0162, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 102.0, | |
| "eval_loss": 0.5116817951202393, | |
| "eval_runtime": 249.9762, | |
| "eval_samples_per_second": 20.854, | |
| "eval_steps_per_second": 2.608, | |
| "eval_wer": 0.27011748769114546, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 104.0, | |
| "learning_rate": 1.3651063829787234e-05, | |
| "loss": 0.0155, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 104.0, | |
| "eval_loss": 0.5181553363800049, | |
| "eval_runtime": 252.5167, | |
| "eval_samples_per_second": 20.644, | |
| "eval_steps_per_second": 2.582, | |
| "eval_wer": 0.26858050902649333, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 106.0, | |
| "learning_rate": 1.1948936170212767e-05, | |
| "loss": 0.0162, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 106.0, | |
| "eval_loss": 0.5102177262306213, | |
| "eval_runtime": 253.88, | |
| "eval_samples_per_second": 20.533, | |
| "eval_steps_per_second": 2.568, | |
| "eval_wer": 0.26886706437075053, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 108.0, | |
| "learning_rate": 1.0246808510638298e-05, | |
| "loss": 0.015, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 108.0, | |
| "eval_loss": 0.5183060169219971, | |
| "eval_runtime": 256.9895, | |
| "eval_samples_per_second": 20.285, | |
| "eval_steps_per_second": 2.537, | |
| "eval_wer": 0.27050824497876885, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 110.0, | |
| "learning_rate": 8.54468085106383e-06, | |
| "loss": 0.0148, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 110.0, | |
| "eval_loss": 0.524269700050354, | |
| "eval_runtime": 251.4198, | |
| "eval_samples_per_second": 20.734, | |
| "eval_steps_per_second": 2.593, | |
| "eval_wer": 0.269075468257483, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 112.0, | |
| "learning_rate": 6.846808510638299e-06, | |
| "loss": 0.0142, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 112.0, | |
| "eval_loss": 0.5115846991539001, | |
| "eval_runtime": 252.9673, | |
| "eval_samples_per_second": 20.607, | |
| "eval_steps_per_second": 2.577, | |
| "eval_wer": 0.269075468257483, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 114.0, | |
| "learning_rate": 5.14468085106383e-06, | |
| "loss": 0.0133, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 114.0, | |
| "eval_loss": 0.5274414420127869, | |
| "eval_runtime": 253.5713, | |
| "eval_samples_per_second": 20.558, | |
| "eval_steps_per_second": 2.571, | |
| "eval_wer": 0.2676166410503556, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 116.0, | |
| "learning_rate": 3.4425531914893614e-06, | |
| "loss": 0.0138, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 116.0, | |
| "eval_loss": 0.5216977596282959, | |
| "eval_runtime": 253.8491, | |
| "eval_samples_per_second": 20.536, | |
| "eval_steps_per_second": 2.568, | |
| "eval_wer": 0.26722588376273215, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 118.0, | |
| "learning_rate": 1.7404255319148937e-06, | |
| "loss": 0.0135, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 118.0, | |
| "eval_loss": 0.5230008363723755, | |
| "eval_runtime": 257.6327, | |
| "eval_samples_per_second": 20.234, | |
| "eval_steps_per_second": 2.531, | |
| "eval_wer": 0.2673300857060984, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 120.0, | |
| "learning_rate": 3.8297872340425535e-08, | |
| "loss": 0.0134, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 120.0, | |
| "eval_loss": 0.5197370648384094, | |
| "eval_runtime": 253.9198, | |
| "eval_samples_per_second": 20.53, | |
| "eval_steps_per_second": 2.568, | |
| "eval_wer": 0.2666527730742178, | |
| "step": 24000 | |
| } | |
| ], | |
| "max_steps": 24000, | |
| "num_train_epochs": 120, | |
| "total_flos": 1.951590305315055e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |