| { | |
| "best_metric": 0.18234074115753174, | |
| "best_model_checkpoint": "/scratch/skscla001/results/mms-1b-all-bem-genbed-f-model/checkpoint-4400", | |
| "epoch": 6.438356164383562, | |
| "eval_steps": 100, | |
| "global_step": 4700, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.136986301369863, | |
| "grad_norm": 2.5438549518585205, | |
| "learning_rate": 0.00028799999999999995, | |
| "loss": 6.6556, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.136986301369863, | |
| "eval_loss": 0.5950818061828613, | |
| "eval_runtime": 66.923, | |
| "eval_samples_per_second": 14.494, | |
| "eval_steps_per_second": 1.823, | |
| "eval_wer": 0.6643033623375229, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.273972602739726, | |
| "grad_norm": 1.7334426641464233, | |
| "learning_rate": 0.0002986788990825688, | |
| "loss": 0.4415, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.273972602739726, | |
| "eval_loss": 0.27343931794166565, | |
| "eval_runtime": 66.3712, | |
| "eval_samples_per_second": 14.615, | |
| "eval_steps_per_second": 1.838, | |
| "eval_wer": 0.45654742722096897, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.410958904109589, | |
| "grad_norm": 2.5106163024902344, | |
| "learning_rate": 0.00029730275229357796, | |
| "loss": 0.3448, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.410958904109589, | |
| "eval_loss": 0.248216450214386, | |
| "eval_runtime": 65.6824, | |
| "eval_samples_per_second": 14.768, | |
| "eval_steps_per_second": 1.857, | |
| "eval_wer": 0.4289397357396068, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.547945205479452, | |
| "grad_norm": 1.4925711154937744, | |
| "learning_rate": 0.00029592660550458713, | |
| "loss": 0.3459, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.547945205479452, | |
| "eval_loss": 0.23923924565315247, | |
| "eval_runtime": 65.8079, | |
| "eval_samples_per_second": 14.74, | |
| "eval_steps_per_second": 1.854, | |
| "eval_wer": 0.4148673326887958, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.684931506849315, | |
| "grad_norm": 3.7458858489990234, | |
| "learning_rate": 0.0002945504587155963, | |
| "loss": 0.3184, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.684931506849315, | |
| "eval_loss": 0.23041340708732605, | |
| "eval_runtime": 66.6329, | |
| "eval_samples_per_second": 14.557, | |
| "eval_steps_per_second": 1.831, | |
| "eval_wer": 0.40852938016972823, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.821917808219178, | |
| "grad_norm": 2.308128595352173, | |
| "learning_rate": 0.00029317431192660547, | |
| "loss": 0.3058, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.821917808219178, | |
| "eval_loss": 0.2372046560049057, | |
| "eval_runtime": 66.0123, | |
| "eval_samples_per_second": 14.694, | |
| "eval_steps_per_second": 1.848, | |
| "eval_wer": 0.4107852615748201, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.958904109589041, | |
| "grad_norm": 2.1398093700408936, | |
| "learning_rate": 0.00029181192660550457, | |
| "loss": 0.3077, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.958904109589041, | |
| "eval_loss": 0.22706551849842072, | |
| "eval_runtime": 65.7716, | |
| "eval_samples_per_second": 14.748, | |
| "eval_steps_per_second": 1.855, | |
| "eval_wer": 0.4172306370179396, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.095890410958904, | |
| "grad_norm": 1.1179248094558716, | |
| "learning_rate": 0.00029043577981651374, | |
| "loss": 0.2812, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.095890410958904, | |
| "eval_loss": 0.22172598540782928, | |
| "eval_runtime": 65.6625, | |
| "eval_samples_per_second": 14.773, | |
| "eval_steps_per_second": 1.858, | |
| "eval_wer": 0.39832420238478894, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.2328767123287672, | |
| "grad_norm": 0.7546290159225464, | |
| "learning_rate": 0.0002890596330275229, | |
| "loss": 0.3297, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.2328767123287672, | |
| "eval_loss": 0.22088122367858887, | |
| "eval_runtime": 66.7304, | |
| "eval_samples_per_second": 14.536, | |
| "eval_steps_per_second": 1.828, | |
| "eval_wer": 0.3984316253088409, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.36986301369863, | |
| "grad_norm": 0.5848907232284546, | |
| "learning_rate": 0.0002876834862385321, | |
| "loss": 0.2817, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.36986301369863, | |
| "eval_loss": 0.21634483337402344, | |
| "eval_runtime": 66.3551, | |
| "eval_samples_per_second": 14.618, | |
| "eval_steps_per_second": 1.839, | |
| "eval_wer": 0.41239660543559997, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.5068493150684932, | |
| "grad_norm": 1.0077754259109497, | |
| "learning_rate": 0.00028630733944954125, | |
| "loss": 0.2927, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.5068493150684932, | |
| "eval_loss": 0.21464507281780243, | |
| "eval_runtime": 66.426, | |
| "eval_samples_per_second": 14.603, | |
| "eval_steps_per_second": 1.837, | |
| "eval_wer": 0.3862928348909657, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.643835616438356, | |
| "grad_norm": 0.9891207218170166, | |
| "learning_rate": 0.0002849311926605504, | |
| "loss": 0.2806, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.643835616438356, | |
| "eval_loss": 0.21057289838790894, | |
| "eval_runtime": 65.8444, | |
| "eval_samples_per_second": 14.732, | |
| "eval_steps_per_second": 1.853, | |
| "eval_wer": 0.38511118272639383, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.7808219178082192, | |
| "grad_norm": 0.7662916779518127, | |
| "learning_rate": 0.0002835550458715596, | |
| "loss": 0.2574, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.7808219178082192, | |
| "eval_loss": 0.20976857841014862, | |
| "eval_runtime": 66.4117, | |
| "eval_samples_per_second": 14.606, | |
| "eval_steps_per_second": 1.837, | |
| "eval_wer": 0.3866151036631217, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.9178082191780823, | |
| "grad_norm": 0.8207290768623352, | |
| "learning_rate": 0.00028217889908256877, | |
| "loss": 0.2829, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.9178082191780823, | |
| "eval_loss": 0.20674893260002136, | |
| "eval_runtime": 66.1157, | |
| "eval_samples_per_second": 14.671, | |
| "eval_steps_per_second": 1.845, | |
| "eval_wer": 0.37716188634654635, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.0547945205479454, | |
| "grad_norm": 1.1322788000106812, | |
| "learning_rate": 0.000280802752293578, | |
| "loss": 0.2764, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0547945205479454, | |
| "eval_loss": 0.20762823522090912, | |
| "eval_runtime": 66.5673, | |
| "eval_samples_per_second": 14.572, | |
| "eval_steps_per_second": 1.833, | |
| "eval_wer": 0.37888065313137825, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.191780821917808, | |
| "grad_norm": 1.9598041772842407, | |
| "learning_rate": 0.0002794266055045871, | |
| "loss": 0.2635, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.191780821917808, | |
| "eval_loss": 0.20759092271327972, | |
| "eval_runtime": 66.2145, | |
| "eval_samples_per_second": 14.649, | |
| "eval_steps_per_second": 1.842, | |
| "eval_wer": 0.37694704049844235, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.328767123287671, | |
| "grad_norm": 1.0646597146987915, | |
| "learning_rate": 0.0002780504587155963, | |
| "loss": 0.2761, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.328767123287671, | |
| "eval_loss": 0.20681439340114594, | |
| "eval_runtime": 66.1689, | |
| "eval_samples_per_second": 14.659, | |
| "eval_steps_per_second": 1.844, | |
| "eval_wer": 0.38006230529595014, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.4657534246575343, | |
| "grad_norm": 2.4462172985076904, | |
| "learning_rate": 0.0002766743119266055, | |
| "loss": 0.2854, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.4657534246575343, | |
| "eval_loss": 0.19937340915203094, | |
| "eval_runtime": 66.2725, | |
| "eval_samples_per_second": 14.637, | |
| "eval_steps_per_second": 1.841, | |
| "eval_wer": 0.3644859813084112, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.602739726027397, | |
| "grad_norm": 1.0019258260726929, | |
| "learning_rate": 0.0002752981651376147, | |
| "loss": 0.2557, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.602739726027397, | |
| "eval_loss": 0.20158326625823975, | |
| "eval_runtime": 66.179, | |
| "eval_samples_per_second": 14.657, | |
| "eval_steps_per_second": 1.843, | |
| "eval_wer": 0.3860779890428617, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.73972602739726, | |
| "grad_norm": 1.5527012348175049, | |
| "learning_rate": 0.0002739220183486238, | |
| "loss": 0.2717, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.73972602739726, | |
| "eval_loss": 0.2010820060968399, | |
| "eval_runtime": 66.4692, | |
| "eval_samples_per_second": 14.593, | |
| "eval_steps_per_second": 1.835, | |
| "eval_wer": 0.37340208400472663, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.8767123287671232, | |
| "grad_norm": 1.1809673309326172, | |
| "learning_rate": 0.000272545871559633, | |
| "loss": 0.2504, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.8767123287671232, | |
| "eval_loss": 0.19885794818401337, | |
| "eval_runtime": 66.3483, | |
| "eval_samples_per_second": 14.62, | |
| "eval_steps_per_second": 1.839, | |
| "eval_wer": 0.367386400257815, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.0136986301369864, | |
| "grad_norm": 1.5838390588760376, | |
| "learning_rate": 0.0002711697247706422, | |
| "loss": 0.2606, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.0136986301369864, | |
| "eval_loss": 0.19898578524589539, | |
| "eval_runtime": 66.1937, | |
| "eval_samples_per_second": 14.654, | |
| "eval_steps_per_second": 1.843, | |
| "eval_wer": 0.3834998388656139, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.1506849315068495, | |
| "grad_norm": 0.551179051399231, | |
| "learning_rate": 0.00026979357798165136, | |
| "loss": 0.2583, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.1506849315068495, | |
| "eval_loss": 0.20280733704566956, | |
| "eval_runtime": 66.4916, | |
| "eval_samples_per_second": 14.588, | |
| "eval_steps_per_second": 1.835, | |
| "eval_wer": 0.3666344397894511, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.287671232876712, | |
| "grad_norm": 0.7909404635429382, | |
| "learning_rate": 0.00026841743119266053, | |
| "loss": 0.2591, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.287671232876712, | |
| "eval_loss": 0.19522929191589355, | |
| "eval_runtime": 66.4208, | |
| "eval_samples_per_second": 14.604, | |
| "eval_steps_per_second": 1.837, | |
| "eval_wer": 0.35073584702975613, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.4246575342465753, | |
| "grad_norm": 0.699164867401123, | |
| "learning_rate": 0.0002670412844036697, | |
| "loss": 0.2408, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.4246575342465753, | |
| "eval_loss": 0.19884993135929108, | |
| "eval_runtime": 66.6309, | |
| "eval_samples_per_second": 14.558, | |
| "eval_steps_per_second": 1.831, | |
| "eval_wer": 0.36373402084004725, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.5616438356164384, | |
| "grad_norm": 0.6602646112442017, | |
| "learning_rate": 0.0002656651376146789, | |
| "loss": 0.2485, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.5616438356164384, | |
| "eval_loss": 0.19720342755317688, | |
| "eval_runtime": 66.8315, | |
| "eval_samples_per_second": 14.514, | |
| "eval_steps_per_second": 1.825, | |
| "eval_wer": 0.35932968095391554, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.6986301369863015, | |
| "grad_norm": 0.8821811676025391, | |
| "learning_rate": 0.00026428899082568804, | |
| "loss": 0.2474, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.6986301369863015, | |
| "eval_loss": 0.1949489265680313, | |
| "eval_runtime": 66.4835, | |
| "eval_samples_per_second": 14.59, | |
| "eval_steps_per_second": 1.835, | |
| "eval_wer": 0.353421420131056, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.8356164383561646, | |
| "grad_norm": 0.9004201292991638, | |
| "learning_rate": 0.0002629128440366972, | |
| "loss": 0.2398, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.8356164383561646, | |
| "eval_loss": 0.19590793550014496, | |
| "eval_runtime": 66.4984, | |
| "eval_samples_per_second": 14.587, | |
| "eval_steps_per_second": 1.835, | |
| "eval_wer": 0.3697497045869589, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.9726027397260273, | |
| "grad_norm": 0.9938905835151672, | |
| "learning_rate": 0.0002615366972477064, | |
| "loss": 0.2512, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 3.9726027397260273, | |
| "eval_loss": 0.19063439965248108, | |
| "eval_runtime": 66.702, | |
| "eval_samples_per_second": 14.542, | |
| "eval_steps_per_second": 1.829, | |
| "eval_wer": 0.3558921473842518, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 4.109589041095891, | |
| "grad_norm": 53.745086669921875, | |
| "learning_rate": 0.0002601743119266055, | |
| "loss": 0.2266, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.109589041095891, | |
| "eval_loss": 0.19052593410015106, | |
| "eval_runtime": 66.4298, | |
| "eval_samples_per_second": 14.602, | |
| "eval_steps_per_second": 1.837, | |
| "eval_wer": 0.3481576968525083, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.2465753424657535, | |
| "grad_norm": 0.7372691035270691, | |
| "learning_rate": 0.00025879816513761465, | |
| "loss": 0.2538, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 4.2465753424657535, | |
| "eval_loss": 0.19160659611225128, | |
| "eval_runtime": 66.6561, | |
| "eval_samples_per_second": 14.552, | |
| "eval_steps_per_second": 1.83, | |
| "eval_wer": 0.35213234504243207, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 4.383561643835616, | |
| "grad_norm": 0.9205912947654724, | |
| "learning_rate": 0.0002574220183486238, | |
| "loss": 0.2268, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 4.383561643835616, | |
| "eval_loss": 0.1914052814245224, | |
| "eval_runtime": 66.9123, | |
| "eval_samples_per_second": 14.497, | |
| "eval_steps_per_second": 1.823, | |
| "eval_wer": 0.38951552261252553, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 4.52054794520548, | |
| "grad_norm": 0.7206313610076904, | |
| "learning_rate": 0.000256045871559633, | |
| "loss": 0.2249, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 4.52054794520548, | |
| "eval_loss": 0.18965989351272583, | |
| "eval_runtime": 66.7204, | |
| "eval_samples_per_second": 14.538, | |
| "eval_steps_per_second": 1.829, | |
| "eval_wer": 0.3417123214093888, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 4.657534246575342, | |
| "grad_norm": 0.8181961178779602, | |
| "learning_rate": 0.00025466972477064217, | |
| "loss": 0.2416, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 4.657534246575342, | |
| "eval_loss": 0.18771447241306305, | |
| "eval_runtime": 66.4891, | |
| "eval_samples_per_second": 14.589, | |
| "eval_steps_per_second": 1.835, | |
| "eval_wer": 0.34579439252336447, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 4.794520547945205, | |
| "grad_norm": 0.9659631848335266, | |
| "learning_rate": 0.00025329357798165134, | |
| "loss": 0.2421, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.794520547945205, | |
| "eval_loss": 0.18715131282806396, | |
| "eval_runtime": 66.81, | |
| "eval_samples_per_second": 14.519, | |
| "eval_steps_per_second": 1.826, | |
| "eval_wer": 0.3411752067891288, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.931506849315069, | |
| "grad_norm": 0.8366265296936035, | |
| "learning_rate": 0.0002519174311926605, | |
| "loss": 0.244, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.931506849315069, | |
| "eval_loss": 0.1854562610387802, | |
| "eval_runtime": 66.011, | |
| "eval_samples_per_second": 14.695, | |
| "eval_steps_per_second": 1.848, | |
| "eval_wer": 0.352776882586744, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 5.068493150684931, | |
| "grad_norm": 1.7714653015136719, | |
| "learning_rate": 0.00025054128440366974, | |
| "loss": 0.2371, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 5.068493150684931, | |
| "eval_loss": 0.18706142902374268, | |
| "eval_runtime": 66.1616, | |
| "eval_samples_per_second": 14.661, | |
| "eval_steps_per_second": 1.844, | |
| "eval_wer": 0.34472016328284455, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 5.205479452054795, | |
| "grad_norm": 2.374345302581787, | |
| "learning_rate": 0.00024916513761467885, | |
| "loss": 0.2383, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 5.205479452054795, | |
| "eval_loss": 0.18328477442264557, | |
| "eval_runtime": 66.7399, | |
| "eval_samples_per_second": 14.534, | |
| "eval_steps_per_second": 1.828, | |
| "eval_wer": 0.35234719089053607, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 5.342465753424658, | |
| "grad_norm": 1.9919145107269287, | |
| "learning_rate": 0.000247788990825688, | |
| "loss": 0.2409, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 5.342465753424658, | |
| "eval_loss": 0.18861110508441925, | |
| "eval_runtime": 66.6306, | |
| "eval_samples_per_second": 14.558, | |
| "eval_steps_per_second": 1.831, | |
| "eval_wer": 0.34869481147276826, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 5.47945205479452, | |
| "grad_norm": 2.0718414783477783, | |
| "learning_rate": 0.00024641284403669725, | |
| "loss": 0.2312, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.47945205479452, | |
| "eval_loss": 0.18479366600513458, | |
| "eval_runtime": 66.8869, | |
| "eval_samples_per_second": 14.502, | |
| "eval_steps_per_second": 1.824, | |
| "eval_wer": 0.3437533569663766, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.616438356164384, | |
| "grad_norm": 2.795865774154663, | |
| "learning_rate": 0.0002450366972477064, | |
| "loss": 0.2261, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 5.616438356164384, | |
| "eval_loss": 0.18661826848983765, | |
| "eval_runtime": 67.285, | |
| "eval_samples_per_second": 14.416, | |
| "eval_steps_per_second": 1.813, | |
| "eval_wer": 0.3468686217638844, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 5.7534246575342465, | |
| "grad_norm": 2.0424089431762695, | |
| "learning_rate": 0.00024366055045871556, | |
| "loss": 0.2169, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 5.7534246575342465, | |
| "eval_loss": 0.18412768840789795, | |
| "eval_runtime": 66.3619, | |
| "eval_samples_per_second": 14.617, | |
| "eval_steps_per_second": 1.838, | |
| "eval_wer": 0.33763025029541305, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 5.890410958904109, | |
| "grad_norm": 1.5393688678741455, | |
| "learning_rate": 0.00024228440366972474, | |
| "loss": 0.2283, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 5.890410958904109, | |
| "eval_loss": 0.1864866465330124, | |
| "eval_runtime": 66.1121, | |
| "eval_samples_per_second": 14.672, | |
| "eval_steps_per_second": 1.845, | |
| "eval_wer": 0.3411752067891288, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 6.027397260273973, | |
| "grad_norm": 0.7328805327415466, | |
| "learning_rate": 0.00024090825688073393, | |
| "loss": 0.2182, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 6.027397260273973, | |
| "eval_loss": 0.18234074115753174, | |
| "eval_runtime": 66.2722, | |
| "eval_samples_per_second": 14.637, | |
| "eval_steps_per_second": 1.841, | |
| "eval_wer": 0.34310881942206467, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 6.164383561643835, | |
| "grad_norm": 0.7929720878601074, | |
| "learning_rate": 0.0002395321100917431, | |
| "loss": 0.2141, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 6.164383561643835, | |
| "eval_loss": 0.18578113615512848, | |
| "eval_runtime": 66.1558, | |
| "eval_samples_per_second": 14.662, | |
| "eval_steps_per_second": 1.844, | |
| "eval_wer": 0.34031582339671285, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 6.301369863013699, | |
| "grad_norm": 0.8493778705596924, | |
| "learning_rate": 0.00023815596330275225, | |
| "loss": 0.2127, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 6.301369863013699, | |
| "eval_loss": 0.18759718537330627, | |
| "eval_runtime": 66.2656, | |
| "eval_samples_per_second": 14.638, | |
| "eval_steps_per_second": 1.841, | |
| "eval_wer": 0.3355892147384252, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 6.438356164383562, | |
| "grad_norm": 0.8041571378707886, | |
| "learning_rate": 0.00023677981651376145, | |
| "loss": 0.229, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 6.438356164383562, | |
| "eval_loss": 0.18627603352069855, | |
| "eval_runtime": 66.3029, | |
| "eval_samples_per_second": 14.63, | |
| "eval_steps_per_second": 1.84, | |
| "eval_wer": 0.33612632935868514, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 6.438356164383562, | |
| "step": 4700, | |
| "total_flos": 1.108496718380256e+19, | |
| "train_loss": 0.39921932220458983, | |
| "train_runtime": 6096.3108, | |
| "train_samples_per_second": 14.355, | |
| "train_steps_per_second": 3.592 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 21900, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "save_steps": 400, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.108496718380256e+19, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |