| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 28.79581151832461, | |
| "global_step": 22000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 4.94e-05, | |
| "loss": 6.5844, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_loss": 2.985063076019287, | |
| "eval_runtime": 66.7207, | |
| "eval_samples_per_second": 15.348, | |
| "eval_steps_per_second": 1.918, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 9.94e-05, | |
| "loss": 2.1717, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_loss": 1.0769672393798828, | |
| "eval_runtime": 66.0484, | |
| "eval_samples_per_second": 15.504, | |
| "eval_steps_per_second": 1.938, | |
| "eval_wer": 0.8259404828747895, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 9.774635036496352e-05, | |
| "loss": 1.0058, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "eval_loss": 0.7441346049308777, | |
| "eval_runtime": 66.5848, | |
| "eval_samples_per_second": 15.379, | |
| "eval_steps_per_second": 1.922, | |
| "eval_wer": 0.6882650196518809, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 9.546532846715329e-05, | |
| "loss": 0.7316, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "eval_loss": 0.6406537294387817, | |
| "eval_runtime": 65.9141, | |
| "eval_samples_per_second": 15.535, | |
| "eval_steps_per_second": 1.942, | |
| "eval_wer": 0.617630544637844, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 9.318430656934307e-05, | |
| "loss": 0.6249, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "eval_loss": 0.5814340114593506, | |
| "eval_runtime": 66.4954, | |
| "eval_samples_per_second": 15.4, | |
| "eval_steps_per_second": 1.925, | |
| "eval_wer": 0.5783267827063447, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 9.090328467153285e-05, | |
| "loss": 0.5541, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "eval_loss": 0.5387497544288635, | |
| "eval_runtime": 66.9564, | |
| "eval_samples_per_second": 15.294, | |
| "eval_steps_per_second": 1.912, | |
| "eval_wer": 0.5449747332959012, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 8.862226277372263e-05, | |
| "loss": 0.4727, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "eval_loss": 0.5657380223274231, | |
| "eval_runtime": 66.1646, | |
| "eval_samples_per_second": 15.477, | |
| "eval_steps_per_second": 1.935, | |
| "eval_wer": 0.5357664233576642, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 8.634124087591242e-05, | |
| "loss": 0.4444, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "eval_loss": 0.5406317710876465, | |
| "eval_runtime": 65.9956, | |
| "eval_samples_per_second": 15.516, | |
| "eval_steps_per_second": 1.94, | |
| "eval_wer": 0.5164514317798989, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.89, | |
| "learning_rate": 8.406021897810219e-05, | |
| "loss": 0.4057, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.89, | |
| "eval_loss": 0.5271017551422119, | |
| "eval_runtime": 66.6343, | |
| "eval_samples_per_second": 15.367, | |
| "eval_steps_per_second": 1.921, | |
| "eval_wer": 0.5116226838854576, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 6.54, | |
| "learning_rate": 8.177919708029196e-05, | |
| "loss": 0.3663, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.54, | |
| "eval_loss": 0.5105618238449097, | |
| "eval_runtime": 66.576, | |
| "eval_samples_per_second": 15.381, | |
| "eval_steps_per_second": 1.923, | |
| "eval_wer": 0.4793935991016283, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 7.949817518248176e-05, | |
| "loss": 0.3406, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "eval_loss": 0.5662369132041931, | |
| "eval_runtime": 67.3476, | |
| "eval_samples_per_second": 15.205, | |
| "eval_steps_per_second": 1.901, | |
| "eval_wer": 0.48568220101066817, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 7.85, | |
| "learning_rate": 7.721715328467154e-05, | |
| "loss": 0.3142, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.85, | |
| "eval_loss": 0.5632680654525757, | |
| "eval_runtime": 66.2027, | |
| "eval_samples_per_second": 15.468, | |
| "eval_steps_per_second": 1.933, | |
| "eval_wer": 0.4879281302638967, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 8.51, | |
| "learning_rate": 7.493613138686131e-05, | |
| "loss": 0.3002, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 8.51, | |
| "eval_loss": 0.5218114852905273, | |
| "eval_runtime": 67.5136, | |
| "eval_samples_per_second": 15.167, | |
| "eval_steps_per_second": 1.896, | |
| "eval_wer": 0.45659741718135877, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 9.16, | |
| "learning_rate": 7.265967153284671e-05, | |
| "loss": 0.2845, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 9.16, | |
| "eval_loss": 0.5436348915100098, | |
| "eval_runtime": 66.2104, | |
| "eval_samples_per_second": 15.466, | |
| "eval_steps_per_second": 1.933, | |
| "eval_wer": 0.45457608085345313, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 9.82, | |
| "learning_rate": 7.038321167883212e-05, | |
| "loss": 0.2603, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 9.82, | |
| "eval_loss": 0.5183758735656738, | |
| "eval_runtime": 67.776, | |
| "eval_samples_per_second": 15.109, | |
| "eval_steps_per_second": 1.889, | |
| "eval_wer": 0.4415496911847277, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 10.47, | |
| "learning_rate": 6.810218978102189e-05, | |
| "loss": 0.2438, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 10.47, | |
| "eval_loss": 0.5246397256851196, | |
| "eval_runtime": 66.5435, | |
| "eval_samples_per_second": 15.388, | |
| "eval_steps_per_second": 1.924, | |
| "eval_wer": 0.43638405390230206, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 11.13, | |
| "learning_rate": 6.582116788321169e-05, | |
| "loss": 0.2396, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 11.13, | |
| "eval_loss": 0.5663778781890869, | |
| "eval_runtime": 66.6323, | |
| "eval_samples_per_second": 15.368, | |
| "eval_steps_per_second": 1.921, | |
| "eval_wer": 0.4428972487366648, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 11.78, | |
| "learning_rate": 6.354014598540147e-05, | |
| "loss": 0.2229, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 11.78, | |
| "eval_loss": 0.5550942420959473, | |
| "eval_runtime": 65.2937, | |
| "eval_samples_per_second": 15.683, | |
| "eval_steps_per_second": 1.96, | |
| "eval_wer": 0.4415496911847277, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 12.43, | |
| "learning_rate": 6.125912408759124e-05, | |
| "loss": 0.2088, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 12.43, | |
| "eval_loss": 0.5503721833229065, | |
| "eval_runtime": 66.2676, | |
| "eval_samples_per_second": 15.453, | |
| "eval_steps_per_second": 1.932, | |
| "eval_wer": 0.4257158899494666, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 13.09, | |
| "learning_rate": 5.897810218978103e-05, | |
| "loss": 0.2042, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 13.09, | |
| "eval_loss": 0.5646582245826721, | |
| "eval_runtime": 66.4465, | |
| "eval_samples_per_second": 15.411, | |
| "eval_steps_per_second": 1.926, | |
| "eval_wer": 0.4398652442448063, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 13.74, | |
| "learning_rate": 5.669708029197081e-05, | |
| "loss": 0.1975, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 13.74, | |
| "eval_loss": 0.5424100160598755, | |
| "eval_runtime": 66.9743, | |
| "eval_samples_per_second": 15.289, | |
| "eval_steps_per_second": 1.911, | |
| "eval_wer": 0.4186412128017967, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "learning_rate": 5.441605839416058e-05, | |
| "loss": 0.1856, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "eval_loss": 0.5583313703536987, | |
| "eval_runtime": 66.5902, | |
| "eval_samples_per_second": 15.378, | |
| "eval_steps_per_second": 1.922, | |
| "eval_wer": 0.42515440763615947, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 15.05, | |
| "learning_rate": 5.213503649635036e-05, | |
| "loss": 0.1872, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 15.05, | |
| "eval_loss": 0.6118334531784058, | |
| "eval_runtime": 66.3517, | |
| "eval_samples_per_second": 15.433, | |
| "eval_steps_per_second": 1.929, | |
| "eval_wer": 0.433464345873105, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 15.71, | |
| "learning_rate": 4.985401459854015e-05, | |
| "loss": 0.1676, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 15.71, | |
| "eval_loss": 0.6066869497299194, | |
| "eval_runtime": 67.2768, | |
| "eval_samples_per_second": 15.221, | |
| "eval_steps_per_second": 1.903, | |
| "eval_wer": 0.43627175743964064, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 16.36, | |
| "learning_rate": 4.757299270072993e-05, | |
| "loss": 0.1536, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 16.36, | |
| "eval_loss": 0.5462669730186462, | |
| "eval_runtime": 66.3968, | |
| "eval_samples_per_second": 15.422, | |
| "eval_steps_per_second": 1.928, | |
| "eval_wer": 0.412914093206064, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 17.02, | |
| "learning_rate": 4.5291970802919706e-05, | |
| "loss": 0.1582, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 17.02, | |
| "eval_loss": 0.5799742937088013, | |
| "eval_runtime": 65.9177, | |
| "eval_samples_per_second": 15.535, | |
| "eval_steps_per_second": 1.942, | |
| "eval_wer": 0.41021897810218977, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 17.67, | |
| "learning_rate": 4.3015510948905114e-05, | |
| "loss": 0.1429, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 17.67, | |
| "eval_loss": 0.5899107456207275, | |
| "eval_runtime": 66.1113, | |
| "eval_samples_per_second": 15.489, | |
| "eval_steps_per_second": 1.936, | |
| "eval_wer": 0.4150477259966311, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 18.32, | |
| "learning_rate": 4.0734489051094895e-05, | |
| "loss": 0.1451, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 18.32, | |
| "eval_loss": 0.6129721999168396, | |
| "eval_runtime": 67.4516, | |
| "eval_samples_per_second": 15.181, | |
| "eval_steps_per_second": 1.898, | |
| "eval_wer": 0.4138124649073554, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 18.98, | |
| "learning_rate": 3.845346715328467e-05, | |
| "loss": 0.1356, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 18.98, | |
| "eval_loss": 0.6040089130401611, | |
| "eval_runtime": 66.75, | |
| "eval_samples_per_second": 15.341, | |
| "eval_steps_per_second": 1.918, | |
| "eval_wer": 0.4123526108927569, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 19.63, | |
| "learning_rate": 3.617244525547445e-05, | |
| "loss": 0.134, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 19.63, | |
| "eval_loss": 0.5997006893157959, | |
| "eval_runtime": 67.809, | |
| "eval_samples_per_second": 15.101, | |
| "eval_steps_per_second": 1.888, | |
| "eval_wer": 0.40819764177428414, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 20.29, | |
| "learning_rate": 3.389598540145986e-05, | |
| "loss": 0.1305, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 20.29, | |
| "eval_loss": 0.5769977569580078, | |
| "eval_runtime": 66.9982, | |
| "eval_samples_per_second": 15.284, | |
| "eval_steps_per_second": 1.91, | |
| "eval_wer": 0.40235822571588997, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 20.94, | |
| "learning_rate": 3.161496350364963e-05, | |
| "loss": 0.1222, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 20.94, | |
| "eval_loss": 0.5980133414268494, | |
| "eval_runtime": 66.1854, | |
| "eval_samples_per_second": 15.472, | |
| "eval_steps_per_second": 1.934, | |
| "eval_wer": 0.391577765300393, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 21.6, | |
| "learning_rate": 2.9333941605839417e-05, | |
| "loss": 0.1214, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 21.6, | |
| "eval_loss": 0.5917083024978638, | |
| "eval_runtime": 67.5557, | |
| "eval_samples_per_second": 15.158, | |
| "eval_steps_per_second": 1.895, | |
| "eval_wer": 0.4005614823133071, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 22.25, | |
| "learning_rate": 2.70529197080292e-05, | |
| "loss": 0.1184, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 22.25, | |
| "eval_loss": 0.578889787197113, | |
| "eval_runtime": 66.5407, | |
| "eval_samples_per_second": 15.389, | |
| "eval_steps_per_second": 1.924, | |
| "eval_wer": 0.39460976979225154, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 22.91, | |
| "learning_rate": 2.477189781021898e-05, | |
| "loss": 0.1132, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 22.91, | |
| "eval_loss": 0.5992549657821655, | |
| "eval_runtime": 66.4759, | |
| "eval_samples_per_second": 15.404, | |
| "eval_steps_per_second": 1.926, | |
| "eval_wer": 0.3892195395845031, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 23.56, | |
| "learning_rate": 2.249087591240876e-05, | |
| "loss": 0.1052, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 23.56, | |
| "eval_loss": 0.5946719646453857, | |
| "eval_runtime": 66.8036, | |
| "eval_samples_per_second": 15.329, | |
| "eval_steps_per_second": 1.916, | |
| "eval_wer": 0.3889949466591802, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 24.21, | |
| "learning_rate": 2.0209854014598544e-05, | |
| "loss": 0.113, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 24.21, | |
| "eval_loss": 0.6009370684623718, | |
| "eval_runtime": 65.8523, | |
| "eval_samples_per_second": 15.55, | |
| "eval_steps_per_second": 1.944, | |
| "eval_wer": 0.3841661987647389, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 24.87, | |
| "learning_rate": 1.792883211678832e-05, | |
| "loss": 0.1021, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 24.87, | |
| "eval_loss": 0.6255775690078735, | |
| "eval_runtime": 67.7689, | |
| "eval_samples_per_second": 15.11, | |
| "eval_steps_per_second": 1.889, | |
| "eval_wer": 0.3841661987647389, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 25.52, | |
| "learning_rate": 1.5652372262773723e-05, | |
| "loss": 0.0966, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 25.52, | |
| "eval_loss": 0.6213188171386719, | |
| "eval_runtime": 68.5057, | |
| "eval_samples_per_second": 14.948, | |
| "eval_steps_per_second": 1.868, | |
| "eval_wer": 0.3869736103312746, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 26.18, | |
| "learning_rate": 1.3371350364963504e-05, | |
| "loss": 0.0981, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 26.18, | |
| "eval_loss": 0.6331676244735718, | |
| "eval_runtime": 67.2276, | |
| "eval_samples_per_second": 15.232, | |
| "eval_steps_per_second": 1.904, | |
| "eval_wer": 0.37866367209432905, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 26.83, | |
| "learning_rate": 1.1090328467153285e-05, | |
| "loss": 0.0983, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 26.83, | |
| "eval_loss": 0.6225576996803284, | |
| "eval_runtime": 66.5907, | |
| "eval_samples_per_second": 15.378, | |
| "eval_steps_per_second": 1.922, | |
| "eval_wer": 0.3806850084222347, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 27.49, | |
| "learning_rate": 8.809306569343067e-06, | |
| "loss": 0.0915, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 27.49, | |
| "eval_loss": 0.6176589727401733, | |
| "eval_runtime": 66.1318, | |
| "eval_samples_per_second": 15.484, | |
| "eval_steps_per_second": 1.936, | |
| "eval_wer": 0.3785513756316676, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 28.14, | |
| "learning_rate": 6.532846715328468e-06, | |
| "loss": 0.0898, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 28.14, | |
| "eval_loss": 0.6252412796020508, | |
| "eval_runtime": 66.2396, | |
| "eval_samples_per_second": 15.459, | |
| "eval_steps_per_second": 1.932, | |
| "eval_wer": 0.3828186412128018, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 28.8, | |
| "learning_rate": 4.251824817518248e-06, | |
| "loss": 0.0916, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 28.8, | |
| "eval_loss": 0.6163517832756042, | |
| "eval_runtime": 66.1166, | |
| "eval_samples_per_second": 15.488, | |
| "eval_steps_per_second": 1.936, | |
| "eval_wer": 0.3760808534531162, | |
| "step": 22000 | |
| } | |
| ], | |
| "max_steps": 22920, | |
| "num_train_epochs": 30, | |
| "total_flos": 6.988254857647251e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |