| { | |
| "best_metric": 0.15105250983113577, | |
| "best_model_checkpoint": "xlsr_Paiwan/checkpoint-19740", | |
| "epoch": 30.0, | |
| "global_step": 19740, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.0002924620060790273, | |
| "loss": 2.0437, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.43691062927246094, | |
| "eval_runtime": 54.5716, | |
| "eval_samples_per_second": 54.79, | |
| "eval_steps_per_second": 6.853, | |
| "eval_wer": 0.57050328806054, | |
| "step": 658 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 0.00028486322188449846, | |
| "loss": 0.4605, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.3218730688095093, | |
| "eval_runtime": 53.5979, | |
| "eval_samples_per_second": 55.786, | |
| "eval_steps_per_second": 6.978, | |
| "eval_wer": 0.424407653415287, | |
| "step": 1316 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 0.0002772644376899696, | |
| "loss": 0.3486, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.2743721604347229, | |
| "eval_runtime": 52.9623, | |
| "eval_samples_per_second": 56.455, | |
| "eval_steps_per_second": 7.062, | |
| "eval_wer": 0.3111265325005783, | |
| "step": 1974 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 0.0002696656534954407, | |
| "loss": 0.3005, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 0.00026206686930091183, | |
| "loss": 0.243, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.2703934609889984, | |
| "eval_runtime": 52.6582, | |
| "eval_samples_per_second": 56.781, | |
| "eval_steps_per_second": 7.102, | |
| "eval_wer": 0.3132745117477942, | |
| "step": 2632 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "learning_rate": 0.000254468085106383, | |
| "loss": 0.2074, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.2540125250816345, | |
| "eval_runtime": 51.8815, | |
| "eval_samples_per_second": 57.631, | |
| "eval_steps_per_second": 7.209, | |
| "eval_wer": 0.29308350682396483, | |
| "step": 3290 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "learning_rate": 0.0002468693009118541, | |
| "loss": 0.1827, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.24963043630123138, | |
| "eval_runtime": 52.0611, | |
| "eval_samples_per_second": 57.433, | |
| "eval_steps_per_second": 7.184, | |
| "eval_wer": 0.24047453818446185, | |
| "step": 3948 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 0.0002392705167173252, | |
| "loss": 0.1649, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.84, | |
| "learning_rate": 0.00023167173252279633, | |
| "loss": 0.1371, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.24666966497898102, | |
| "eval_runtime": 52.0071, | |
| "eval_samples_per_second": 57.492, | |
| "eval_steps_per_second": 7.191, | |
| "eval_wer": 0.23730213806549685, | |
| "step": 4606 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "learning_rate": 0.00022407294832826744, | |
| "loss": 0.1258, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.25583532452583313, | |
| "eval_runtime": 52.0669, | |
| "eval_samples_per_second": 57.426, | |
| "eval_steps_per_second": 7.183, | |
| "eval_wer": 0.22309242919929942, | |
| "step": 5264 | |
| }, | |
| { | |
| "epoch": 8.36, | |
| "learning_rate": 0.00021647416413373858, | |
| "loss": 0.1078, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.2600269019603729, | |
| "eval_runtime": 52.0714, | |
| "eval_samples_per_second": 57.421, | |
| "eval_steps_per_second": 7.182, | |
| "eval_wer": 0.213310862165824, | |
| "step": 5922 | |
| }, | |
| { | |
| "epoch": 9.12, | |
| "learning_rate": 0.00020887537993920973, | |
| "loss": 0.1038, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 9.88, | |
| "learning_rate": 0.00020127659574468082, | |
| "loss": 0.0906, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.25083455443382263, | |
| "eval_runtime": 52.0294, | |
| "eval_samples_per_second": 57.467, | |
| "eval_steps_per_second": 7.188, | |
| "eval_wer": 0.20997323287399625, | |
| "step": 6580 | |
| }, | |
| { | |
| "epoch": 10.64, | |
| "learning_rate": 0.00019367781155015196, | |
| "loss": 0.0843, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.2840941548347473, | |
| "eval_runtime": 52.1825, | |
| "eval_samples_per_second": 57.299, | |
| "eval_steps_per_second": 7.167, | |
| "eval_wer": 0.19976206999107762, | |
| "step": 7238 | |
| }, | |
| { | |
| "epoch": 11.4, | |
| "learning_rate": 0.0001860790273556231, | |
| "loss": 0.0765, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.26433253288269043, | |
| "eval_runtime": 52.2172, | |
| "eval_samples_per_second": 57.261, | |
| "eval_steps_per_second": 7.162, | |
| "eval_wer": 0.19328508641485742, | |
| "step": 7896 | |
| }, | |
| { | |
| "epoch": 12.16, | |
| "learning_rate": 0.00017848024316109422, | |
| "loss": 0.0699, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 12.92, | |
| "learning_rate": 0.00017088145896656533, | |
| "loss": 0.0671, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.2808719873428345, | |
| "eval_runtime": 52.2538, | |
| "eval_samples_per_second": 57.221, | |
| "eval_steps_per_second": 7.157, | |
| "eval_wer": 0.18816298205611182, | |
| "step": 8554 | |
| }, | |
| { | |
| "epoch": 13.68, | |
| "learning_rate": 0.00016329787234042553, | |
| "loss": 0.0589, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.2795076072216034, | |
| "eval_runtime": 52.106, | |
| "eval_samples_per_second": 57.383, | |
| "eval_steps_per_second": 7.178, | |
| "eval_wer": 0.18647764449291168, | |
| "step": 9212 | |
| }, | |
| { | |
| "epoch": 14.44, | |
| "learning_rate": 0.00015569908814589664, | |
| "loss": 0.0566, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.27890855073928833, | |
| "eval_runtime": 52.1531, | |
| "eval_samples_per_second": 57.331, | |
| "eval_steps_per_second": 7.171, | |
| "eval_wer": 0.18158686097617396, | |
| "step": 9870 | |
| }, | |
| { | |
| "epoch": 15.2, | |
| "learning_rate": 0.00014810030395136776, | |
| "loss": 0.0514, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 15.96, | |
| "learning_rate": 0.0001405015197568389, | |
| "loss": 0.0507, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.28275853395462036, | |
| "eval_runtime": 52.1154, | |
| "eval_samples_per_second": 57.373, | |
| "eval_steps_per_second": 7.176, | |
| "eval_wer": 0.17676216912858134, | |
| "step": 10528 | |
| }, | |
| { | |
| "epoch": 16.72, | |
| "learning_rate": 0.00013291793313069907, | |
| "loss": 0.0452, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 0.2956693470478058, | |
| "eval_runtime": 56.2891, | |
| "eval_samples_per_second": 53.119, | |
| "eval_steps_per_second": 6.644, | |
| "eval_wer": 0.1839661610653977, | |
| "step": 11186 | |
| }, | |
| { | |
| "epoch": 17.48, | |
| "learning_rate": 0.0001253191489361702, | |
| "loss": 0.0414, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.272775262594223, | |
| "eval_runtime": 52.2587, | |
| "eval_samples_per_second": 57.215, | |
| "eval_steps_per_second": 7.157, | |
| "eval_wer": 0.17431677737021248, | |
| "step": 11844 | |
| }, | |
| { | |
| "epoch": 18.24, | |
| "learning_rate": 0.00011772036474164133, | |
| "loss": 0.0373, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 0.00011012158054711246, | |
| "loss": 0.0362, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 0.2982702851295471, | |
| "eval_runtime": 52.2158, | |
| "eval_samples_per_second": 57.262, | |
| "eval_steps_per_second": 7.163, | |
| "eval_wer": 0.17220184395756916, | |
| "step": 12502 | |
| }, | |
| { | |
| "epoch": 19.76, | |
| "learning_rate": 0.00010252279635258359, | |
| "loss": 0.0348, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.2988567054271698, | |
| "eval_runtime": 52.245, | |
| "eval_samples_per_second": 57.23, | |
| "eval_steps_per_second": 7.159, | |
| "eval_wer": 0.1638082019761409, | |
| "step": 13160 | |
| }, | |
| { | |
| "epoch": 20.52, | |
| "learning_rate": 9.493920972644377e-05, | |
| "loss": 0.0324, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_loss": 0.3093124032020569, | |
| "eval_runtime": 52.8274, | |
| "eval_samples_per_second": 56.599, | |
| "eval_steps_per_second": 7.08, | |
| "eval_wer": 0.16509698952447044, | |
| "step": 13818 | |
| }, | |
| { | |
| "epoch": 21.28, | |
| "learning_rate": 8.734042553191488e-05, | |
| "loss": 0.0303, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 0.3057407736778259, | |
| "eval_runtime": 52.5941, | |
| "eval_samples_per_second": 56.851, | |
| "eval_steps_per_second": 7.111, | |
| "eval_wer": 0.16489871451703514, | |
| "step": 14476 | |
| }, | |
| { | |
| "epoch": 22.04, | |
| "learning_rate": 7.974164133738601e-05, | |
| "loss": 0.0293, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 22.8, | |
| "learning_rate": 7.214285714285714e-05, | |
| "loss": 0.0267, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_loss": 0.3179123103618622, | |
| "eval_runtime": 52.2627, | |
| "eval_samples_per_second": 57.211, | |
| "eval_steps_per_second": 7.156, | |
| "eval_wer": 0.1582895476025247, | |
| "step": 15134 | |
| }, | |
| { | |
| "epoch": 23.56, | |
| "learning_rate": 6.455927051671732e-05, | |
| "loss": 0.026, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 0.3296581506729126, | |
| "eval_runtime": 52.0664, | |
| "eval_samples_per_second": 57.427, | |
| "eval_steps_per_second": 7.183, | |
| "eval_wer": 0.15799213509137172, | |
| "step": 15792 | |
| }, | |
| { | |
| "epoch": 24.32, | |
| "learning_rate": 5.696048632218844e-05, | |
| "loss": 0.0246, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 0.3182116746902466, | |
| "eval_runtime": 51.9316, | |
| "eval_samples_per_second": 57.576, | |
| "eval_steps_per_second": 7.202, | |
| "eval_wer": 0.15541455999471265, | |
| "step": 16450 | |
| }, | |
| { | |
| "epoch": 25.08, | |
| "learning_rate": 4.937689969604863e-05, | |
| "loss": 0.0235, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 25.84, | |
| "learning_rate": 4.179331306990881e-05, | |
| "loss": 0.0215, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 0.3294830620288849, | |
| "eval_runtime": 52.2884, | |
| "eval_samples_per_second": 57.183, | |
| "eval_steps_per_second": 7.153, | |
| "eval_wer": 0.15541455999471265, | |
| "step": 17108 | |
| }, | |
| { | |
| "epoch": 26.6, | |
| "learning_rate": 3.419452887537994e-05, | |
| "loss": 0.0199, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_loss": 0.33707642555236816, | |
| "eval_runtime": 51.7426, | |
| "eval_samples_per_second": 57.786, | |
| "eval_steps_per_second": 7.228, | |
| "eval_wer": 0.1535309474240772, | |
| "step": 17766 | |
| }, | |
| { | |
| "epoch": 27.36, | |
| "learning_rate": 2.6595744680851064e-05, | |
| "loss": 0.019, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 0.34009498357772827, | |
| "eval_runtime": 51.768, | |
| "eval_samples_per_second": 57.758, | |
| "eval_steps_per_second": 7.225, | |
| "eval_wer": 0.15174647235715938, | |
| "step": 18424 | |
| }, | |
| { | |
| "epoch": 28.12, | |
| "learning_rate": 1.8996960486322186e-05, | |
| "loss": 0.0185, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 28.88, | |
| "learning_rate": 1.139817629179331e-05, | |
| "loss": 0.0172, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_loss": 0.34455737471580505, | |
| "eval_runtime": 51.8182, | |
| "eval_samples_per_second": 57.702, | |
| "eval_steps_per_second": 7.218, | |
| "eval_wer": 0.15108555566570833, | |
| "step": 19082 | |
| }, | |
| { | |
| "epoch": 29.64, | |
| "learning_rate": 3.7993920972644374e-06, | |
| "loss": 0.0166, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 0.3466561436653137, | |
| "eval_runtime": 51.8971, | |
| "eval_samples_per_second": 57.614, | |
| "eval_steps_per_second": 7.207, | |
| "eval_wer": 0.15105250983113577, | |
| "step": 19740 | |
| } | |
| ], | |
| "max_steps": 19740, | |
| "num_train_epochs": 30, | |
| "total_flos": 7.422835365656386e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |