| { | |
| "best_metric": 9.556300454211874, | |
| "best_model_checkpoint": "./whisper-small-lv/checkpoint-134000", | |
| "epoch": 30.861354214647626, | |
| "eval_steps": 1000, | |
| "global_step": 134000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 10.87213134765625, | |
| "learning_rate": 4.92e-06, | |
| "loss": 1.6097, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 5.319793701171875, | |
| "learning_rate": 9.920000000000002e-06, | |
| "loss": 0.4717, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 5.155162811279297, | |
| "learning_rate": 9.994327876412267e-06, | |
| "loss": 0.3114, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 4.760919570922852, | |
| "learning_rate": 9.9885635231727e-06, | |
| "loss": 0.2588, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "eval_loss": 0.29197508096694946, | |
| "eval_runtime": 858.8311, | |
| "eval_samples_per_second": 7.862, | |
| "eval_steps_per_second": 0.246, | |
| "eval_wer": 31.485411666699726, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 4.729026794433594, | |
| "learning_rate": 9.982799169933133e-06, | |
| "loss": 0.2303, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 4.435351848602295, | |
| "learning_rate": 9.977034816693568e-06, | |
| "loss": 0.2142, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 4.544332504272461, | |
| "learning_rate": 9.971270463454001e-06, | |
| "loss": 0.1945, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 3.7481513023376465, | |
| "learning_rate": 9.965506110214434e-06, | |
| "loss": 0.1786, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eval_loss": 0.2217991054058075, | |
| "eval_runtime": 859.3081, | |
| "eval_samples_per_second": 7.857, | |
| "eval_steps_per_second": 0.246, | |
| "eval_wer": 25.128428902949402, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 3.2256762981414795, | |
| "learning_rate": 9.959741756974867e-06, | |
| "loss": 0.1682, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 3.6426024436950684, | |
| "learning_rate": 9.953977403735302e-06, | |
| "loss": 0.1599, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 4.202388286590576, | |
| "learning_rate": 9.948213050495735e-06, | |
| "loss": 0.1499, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 4.096649169921875, | |
| "learning_rate": 9.942448697256168e-06, | |
| "loss": 0.1433, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "eval_loss": 0.18866823613643646, | |
| "eval_runtime": 871.5329, | |
| "eval_samples_per_second": 7.747, | |
| "eval_steps_per_second": 0.242, | |
| "eval_wer": 21.99853224110915, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 3.3941469192504883, | |
| "learning_rate": 9.936684344016601e-06, | |
| "loss": 0.1369, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 3.8051373958587646, | |
| "learning_rate": 9.930919990777036e-06, | |
| "loss": 0.1302, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 4.107229232788086, | |
| "learning_rate": 9.925155637537469e-06, | |
| "loss": 0.1232, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 3.520193338394165, | |
| "learning_rate": 9.919391284297902e-06, | |
| "loss": 0.1222, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_loss": 0.16477710008621216, | |
| "eval_runtime": 864.1221, | |
| "eval_samples_per_second": 7.814, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 19.57871352916675, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 4.162296295166016, | |
| "learning_rate": 9.913626931058335e-06, | |
| "loss": 0.1125, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 2.3214519023895264, | |
| "learning_rate": 9.90786257781877e-06, | |
| "loss": 0.0832, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "grad_norm": 3.2043275833129883, | |
| "learning_rate": 9.902098224579203e-06, | |
| "loss": 0.0666, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 2.7628753185272217, | |
| "learning_rate": 9.896333871339636e-06, | |
| "loss": 0.0645, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_loss": 0.15404021739959717, | |
| "eval_runtime": 868.0851, | |
| "eval_samples_per_second": 7.778, | |
| "eval_steps_per_second": 0.243, | |
| "eval_wer": 18.128805759961917, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "grad_norm": 2.9250879287719727, | |
| "learning_rate": 9.89056951810007e-06, | |
| "loss": 0.0624, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "grad_norm": 2.2111053466796875, | |
| "learning_rate": 9.884805164860504e-06, | |
| "loss": 0.0653, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 3.372081995010376, | |
| "learning_rate": 9.879040811620937e-06, | |
| "loss": 0.0625, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "grad_norm": 1.3998991250991821, | |
| "learning_rate": 9.87327645838137e-06, | |
| "loss": 0.06, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 0.14527586102485657, | |
| "eval_runtime": 867.9425, | |
| "eval_samples_per_second": 7.779, | |
| "eval_steps_per_second": 0.243, | |
| "eval_wer": 17.113275284130353, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 2.6196086406707764, | |
| "learning_rate": 9.867512105141804e-06, | |
| "loss": 0.0585, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 2.8574118614196777, | |
| "learning_rate": 9.861747751902238e-06, | |
| "loss": 0.0604, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 2.2665085792541504, | |
| "learning_rate": 9.85598339866267e-06, | |
| "loss": 0.0581, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "grad_norm": 2.5183279514312744, | |
| "learning_rate": 9.850219045423104e-06, | |
| "loss": 0.0561, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "eval_loss": 0.13737037777900696, | |
| "eval_runtime": 874.2241, | |
| "eval_samples_per_second": 7.723, | |
| "eval_steps_per_second": 0.241, | |
| "eval_wer": 16.181050042644348, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "grad_norm": 2.2866740226745605, | |
| "learning_rate": 9.844454692183538e-06, | |
| "loss": 0.0555, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 3.0550711154937744, | |
| "learning_rate": 9.838690338943971e-06, | |
| "loss": 0.0551, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "grad_norm": 2.4893910884857178, | |
| "learning_rate": 9.832925985704405e-06, | |
| "loss": 0.0535, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 3.216590642929077, | |
| "learning_rate": 9.827161632464838e-06, | |
| "loss": 0.0538, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "eval_loss": 0.12932516634464264, | |
| "eval_runtime": 868.0214, | |
| "eval_samples_per_second": 7.779, | |
| "eval_steps_per_second": 0.243, | |
| "eval_wer": 15.659400599004304, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 2.0171263217926025, | |
| "learning_rate": 9.821397279225272e-06, | |
| "loss": 0.0512, | |
| "step": 8250 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 2.5316624641418457, | |
| "learning_rate": 9.815655983398664e-06, | |
| "loss": 0.0505, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 1.9595179557800293, | |
| "learning_rate": 9.809891630159097e-06, | |
| "loss": 0.0442, | |
| "step": 8750 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "grad_norm": 2.2406492233276367, | |
| "learning_rate": 9.80412727691953e-06, | |
| "loss": 0.021, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "eval_loss": 0.1289907842874527, | |
| "eval_runtime": 855.2407, | |
| "eval_samples_per_second": 7.895, | |
| "eval_steps_per_second": 0.247, | |
| "eval_wer": 14.508994981851359, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "grad_norm": 1.1805692911148071, | |
| "learning_rate": 9.798362923679965e-06, | |
| "loss": 0.0219, | |
| "step": 9250 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "grad_norm": 1.599218487739563, | |
| "learning_rate": 9.792598570440398e-06, | |
| "loss": 0.0224, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 2.063124418258667, | |
| "learning_rate": 9.78683421720083e-06, | |
| "loss": 0.0234, | |
| "step": 9750 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 2.7844250202178955, | |
| "learning_rate": 9.781092921374222e-06, | |
| "loss": 0.0231, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "eval_loss": 0.12939909100532532, | |
| "eval_runtime": 857.9085, | |
| "eval_samples_per_second": 7.87, | |
| "eval_steps_per_second": 0.246, | |
| "eval_wer": 14.405855167899716, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "grad_norm": 1.8661818504333496, | |
| "learning_rate": 9.775328568134657e-06, | |
| "loss": 0.0226, | |
| "step": 10250 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "grad_norm": 1.9496829509735107, | |
| "learning_rate": 9.76956421489509e-06, | |
| "loss": 0.0221, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 2.1408982276916504, | |
| "learning_rate": 9.763799861655523e-06, | |
| "loss": 0.0229, | |
| "step": 10750 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "grad_norm": 2.1212332248687744, | |
| "learning_rate": 9.758035508415956e-06, | |
| "loss": 0.0223, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "eval_loss": 0.12141337990760803, | |
| "eval_runtime": 863.1344, | |
| "eval_samples_per_second": 7.823, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 13.654124600829084, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "grad_norm": 2.9175450801849365, | |
| "learning_rate": 9.75227115517639e-06, | |
| "loss": 0.0216, | |
| "step": 11250 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "grad_norm": 2.1823952198028564, | |
| "learning_rate": 9.746506801936824e-06, | |
| "loss": 0.022, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "grad_norm": 2.2630937099456787, | |
| "learning_rate": 9.740742448697257e-06, | |
| "loss": 0.0216, | |
| "step": 11750 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "grad_norm": 1.8593631982803345, | |
| "learning_rate": 9.73497809545769e-06, | |
| "loss": 0.0217, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "eval_loss": 0.12216547876596451, | |
| "eval_runtime": 860.7599, | |
| "eval_samples_per_second": 7.844, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 13.650157684907867, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "grad_norm": 1.9656251668930054, | |
| "learning_rate": 9.729213742218125e-06, | |
| "loss": 0.022, | |
| "step": 12250 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 2.4857194423675537, | |
| "learning_rate": 9.723449388978558e-06, | |
| "loss": 0.022, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "grad_norm": 2.0210490226745605, | |
| "learning_rate": 9.71768503573899e-06, | |
| "loss": 0.0219, | |
| "step": 12750 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "grad_norm": 1.4191805124282837, | |
| "learning_rate": 9.711920682499424e-06, | |
| "loss": 0.0223, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_loss": 0.1153900995850563, | |
| "eval_runtime": 867.7389, | |
| "eval_samples_per_second": 7.781, | |
| "eval_steps_per_second": 0.243, | |
| "eval_wer": 13.041236091001052, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "grad_norm": 0.8872393369674683, | |
| "learning_rate": 9.706156329259858e-06, | |
| "loss": 0.0113, | |
| "step": 13250 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "grad_norm": 1.113614559173584, | |
| "learning_rate": 9.700391976020292e-06, | |
| "loss": 0.0099, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "grad_norm": 2.4680979251861572, | |
| "learning_rate": 9.694627622780725e-06, | |
| "loss": 0.0105, | |
| "step": 13750 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "grad_norm": 0.9727824926376343, | |
| "learning_rate": 9.688886326954116e-06, | |
| "loss": 0.0107, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "eval_loss": 0.12547273933887482, | |
| "eval_runtime": 855.5477, | |
| "eval_samples_per_second": 7.892, | |
| "eval_steps_per_second": 0.247, | |
| "eval_wer": 13.029335343237399, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "grad_norm": 0.43644458055496216, | |
| "learning_rate": 9.68312197371455e-06, | |
| "loss": 0.0103, | |
| "step": 14250 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "grad_norm": 0.9514774680137634, | |
| "learning_rate": 9.677357620474984e-06, | |
| "loss": 0.0105, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "grad_norm": 1.8206623792648315, | |
| "learning_rate": 9.671593267235417e-06, | |
| "loss": 0.0111, | |
| "step": 14750 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "grad_norm": 1.4537785053253174, | |
| "learning_rate": 9.66582891399585e-06, | |
| "loss": 0.0106, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "eval_loss": 0.12262353301048279, | |
| "eval_runtime": 865.9365, | |
| "eval_samples_per_second": 7.797, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 12.821072257373507, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "grad_norm": 1.7742477655410767, | |
| "learning_rate": 9.660064560756283e-06, | |
| "loss": 0.011, | |
| "step": 15250 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "grad_norm": 1.4441837072372437, | |
| "learning_rate": 9.654300207516718e-06, | |
| "loss": 0.0117, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "grad_norm": 0.9210599064826965, | |
| "learning_rate": 9.64853585427715e-06, | |
| "loss": 0.0121, | |
| "step": 15750 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "grad_norm": 3.534360647201538, | |
| "learning_rate": 9.642771501037584e-06, | |
| "loss": 0.0116, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "eval_loss": 0.12254315614700317, | |
| "eval_runtime": 864.7376, | |
| "eval_samples_per_second": 7.808, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 12.650494872761172, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "grad_norm": 1.7790424823760986, | |
| "learning_rate": 9.637007147798017e-06, | |
| "loss": 0.0115, | |
| "step": 16250 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "grad_norm": 1.789859414100647, | |
| "learning_rate": 9.631242794558452e-06, | |
| "loss": 0.0116, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "grad_norm": 1.898427963256836, | |
| "learning_rate": 9.625478441318885e-06, | |
| "loss": 0.0115, | |
| "step": 16750 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "grad_norm": 2.6319220066070557, | |
| "learning_rate": 9.619714088079318e-06, | |
| "loss": 0.0118, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "eval_loss": 0.1181536465883255, | |
| "eval_runtime": 866.1146, | |
| "eval_samples_per_second": 7.796, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 12.787353472043161, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "grad_norm": 2.027708053588867, | |
| "learning_rate": 9.61394973483975e-06, | |
| "loss": 0.0113, | |
| "step": 17250 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "grad_norm": 0.5843227505683899, | |
| "learning_rate": 9.608185381600186e-06, | |
| "loss": 0.0092, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "grad_norm": 1.0360699892044067, | |
| "learning_rate": 9.602421028360619e-06, | |
| "loss": 0.0059, | |
| "step": 17750 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "grad_norm": 2.0503389835357666, | |
| "learning_rate": 9.596656675121052e-06, | |
| "loss": 0.0069, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "eval_loss": 0.12232159078121185, | |
| "eval_runtime": 874.5807, | |
| "eval_samples_per_second": 7.72, | |
| "eval_steps_per_second": 0.241, | |
| "eval_wer": 12.555288890651964, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "grad_norm": 1.467950701713562, | |
| "learning_rate": 9.590892321881485e-06, | |
| "loss": 0.0062, | |
| "step": 18250 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "grad_norm": 1.5525797605514526, | |
| "learning_rate": 9.58512796864192e-06, | |
| "loss": 0.0066, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 1.1637401580810547, | |
| "learning_rate": 9.57938667281531e-06, | |
| "loss": 0.0074, | |
| "step": 18750 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "grad_norm": 0.30910471081733704, | |
| "learning_rate": 9.573622319575744e-06, | |
| "loss": 0.0069, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "eval_loss": 0.12405180186033249, | |
| "eval_runtime": 867.8174, | |
| "eval_samples_per_second": 7.78, | |
| "eval_steps_per_second": 0.243, | |
| "eval_wer": 12.285538608009203, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "grad_norm": 2.687534809112549, | |
| "learning_rate": 9.567857966336179e-06, | |
| "loss": 0.0077, | |
| "step": 19250 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "grad_norm": 1.3895576000213623, | |
| "learning_rate": 9.562093613096612e-06, | |
| "loss": 0.0074, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "grad_norm": 1.3012845516204834, | |
| "learning_rate": 9.556329259857045e-06, | |
| "loss": 0.0077, | |
| "step": 19750 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "grad_norm": 1.299644112586975, | |
| "learning_rate": 9.55056490661748e-06, | |
| "loss": 0.0075, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "eval_loss": 0.12338556349277496, | |
| "eval_runtime": 867.2958, | |
| "eval_samples_per_second": 7.785, | |
| "eval_steps_per_second": 0.243, | |
| "eval_wer": 12.166531130372691, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 4.66, | |
| "grad_norm": 1.9311870336532593, | |
| "learning_rate": 9.544800553377913e-06, | |
| "loss": 0.0074, | |
| "step": 20250 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "grad_norm": 0.8604072332382202, | |
| "learning_rate": 9.539036200138346e-06, | |
| "loss": 0.0072, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "grad_norm": 1.3878774642944336, | |
| "learning_rate": 9.533271846898779e-06, | |
| "loss": 0.0075, | |
| "step": 20750 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "grad_norm": 0.29127371311187744, | |
| "learning_rate": 9.52753055107217e-06, | |
| "loss": 0.0075, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "eval_loss": 0.12316736578941345, | |
| "eval_runtime": 874.2652, | |
| "eval_samples_per_second": 7.723, | |
| "eval_steps_per_second": 0.241, | |
| "eval_wer": 12.237935616954598, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "grad_norm": 0.9361172914505005, | |
| "learning_rate": 9.521789255245563e-06, | |
| "loss": 0.0077, | |
| "step": 21250 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "grad_norm": 0.9276347160339355, | |
| "learning_rate": 9.516024902005996e-06, | |
| "loss": 0.0083, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "grad_norm": 0.9278080463409424, | |
| "learning_rate": 9.510260548766429e-06, | |
| "loss": 0.0073, | |
| "step": 21750 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "grad_norm": 1.3185243606567383, | |
| "learning_rate": 9.504496195526862e-06, | |
| "loss": 0.0048, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "eval_loss": 0.12173668295145035, | |
| "eval_runtime": 864.9202, | |
| "eval_samples_per_second": 7.807, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 11.658765892456909, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "grad_norm": 0.7605379223823547, | |
| "learning_rate": 9.498731842287297e-06, | |
| "loss": 0.0051, | |
| "step": 22250 | |
| }, | |
| { | |
| "epoch": 5.18, | |
| "grad_norm": 0.9030295014381409, | |
| "learning_rate": 9.49296748904773e-06, | |
| "loss": 0.0049, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "grad_norm": 0.5184769630432129, | |
| "learning_rate": 9.487203135808163e-06, | |
| "loss": 0.0051, | |
| "step": 22750 | |
| }, | |
| { | |
| "epoch": 5.3, | |
| "grad_norm": 0.11390486359596252, | |
| "learning_rate": 9.481438782568596e-06, | |
| "loss": 0.0052, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 5.3, | |
| "eval_loss": 0.1290554255247116, | |
| "eval_runtime": 859.7689, | |
| "eval_samples_per_second": 7.853, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 12.015788325366444, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "grad_norm": 0.6451162695884705, | |
| "learning_rate": 9.475674429329031e-06, | |
| "loss": 0.0054, | |
| "step": 23250 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "grad_norm": 1.7149757146835327, | |
| "learning_rate": 9.469910076089464e-06, | |
| "loss": 0.0058, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 5.47, | |
| "grad_norm": 0.8278890252113342, | |
| "learning_rate": 9.464145722849897e-06, | |
| "loss": 0.0055, | |
| "step": 23750 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "grad_norm": 1.2663037776947021, | |
| "learning_rate": 9.458381369610332e-06, | |
| "loss": 0.0053, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "eval_loss": 0.12593936920166016, | |
| "eval_runtime": 862.6306, | |
| "eval_samples_per_second": 7.827, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 11.84719439871472, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 5.58, | |
| "grad_norm": 0.23468834161758423, | |
| "learning_rate": 9.452617016370765e-06, | |
| "loss": 0.006, | |
| "step": 24250 | |
| }, | |
| { | |
| "epoch": 5.64, | |
| "grad_norm": 2.092036247253418, | |
| "learning_rate": 9.446852663131198e-06, | |
| "loss": 0.0057, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "grad_norm": 1.3907628059387207, | |
| "learning_rate": 9.441088309891631e-06, | |
| "loss": 0.0058, | |
| "step": 24750 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.8375486135482788, | |
| "learning_rate": 9.435347014065022e-06, | |
| "loss": 0.0064, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "eval_loss": 0.12751518189907074, | |
| "eval_runtime": 855.0746, | |
| "eval_samples_per_second": 7.896, | |
| "eval_steps_per_second": 0.247, | |
| "eval_wer": 12.069341690302874, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 5.82, | |
| "grad_norm": 1.525996208190918, | |
| "learning_rate": 9.429582660825455e-06, | |
| "loss": 0.0064, | |
| "step": 25250 | |
| }, | |
| { | |
| "epoch": 5.87, | |
| "grad_norm": 1.7118737697601318, | |
| "learning_rate": 9.423818307585888e-06, | |
| "loss": 0.0061, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "grad_norm": 1.0815613269805908, | |
| "learning_rate": 9.418053954346323e-06, | |
| "loss": 0.0063, | |
| "step": 25750 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "grad_norm": 1.5884686708450317, | |
| "learning_rate": 9.412289601106756e-06, | |
| "loss": 0.0056, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "eval_loss": 0.1244601458311081, | |
| "eval_runtime": 859.6409, | |
| "eval_samples_per_second": 7.854, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 11.96620187635123, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "grad_norm": 1.1554352045059204, | |
| "learning_rate": 9.40652524786719e-06, | |
| "loss": 0.0044, | |
| "step": 26250 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "grad_norm": 1.1166563034057617, | |
| "learning_rate": 9.400760894627624e-06, | |
| "loss": 0.0037, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 6.16, | |
| "grad_norm": 0.11678412556648254, | |
| "learning_rate": 9.394996541388057e-06, | |
| "loss": 0.0043, | |
| "step": 26750 | |
| }, | |
| { | |
| "epoch": 6.22, | |
| "grad_norm": 0.4431307315826416, | |
| "learning_rate": 9.38923218814849e-06, | |
| "loss": 0.0039, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 6.22, | |
| "eval_loss": 0.12606889009475708, | |
| "eval_runtime": 869.3384, | |
| "eval_samples_per_second": 7.767, | |
| "eval_steps_per_second": 0.243, | |
| "eval_wer": 12.134795803002955, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "grad_norm": 0.2918953597545624, | |
| "learning_rate": 9.383467834908923e-06, | |
| "loss": 0.0042, | |
| "step": 27250 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "grad_norm": 1.4891304969787598, | |
| "learning_rate": 9.377703481669358e-06, | |
| "loss": 0.0043, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "grad_norm": 1.534620761871338, | |
| "learning_rate": 9.371939128429791e-06, | |
| "loss": 0.0043, | |
| "step": 27750 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "grad_norm": 0.4249553680419922, | |
| "learning_rate": 9.366174775190224e-06, | |
| "loss": 0.0048, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "eval_loss": 0.12901394069194794, | |
| "eval_runtime": 878.5914, | |
| "eval_samples_per_second": 7.685, | |
| "eval_steps_per_second": 0.24, | |
| "eval_wer": 11.863062062399587, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 6.51, | |
| "grad_norm": 0.7702826261520386, | |
| "learning_rate": 9.360410421950657e-06, | |
| "loss": 0.005, | |
| "step": 28250 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "grad_norm": 1.9649180173873901, | |
| "learning_rate": 9.354646068711092e-06, | |
| "loss": 0.0054, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "grad_norm": 2.576911211013794, | |
| "learning_rate": 9.348881715471525e-06, | |
| "loss": 0.005, | |
| "step": 28750 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "grad_norm": 1.0807923078536987, | |
| "learning_rate": 9.343117362231958e-06, | |
| "loss": 0.0047, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "eval_loss": 0.12914614379405975, | |
| "eval_runtime": 865.8374, | |
| "eval_samples_per_second": 7.798, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 11.928516175099668, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 6.74, | |
| "grad_norm": 1.7776553630828857, | |
| "learning_rate": 9.337353008992391e-06, | |
| "loss": 0.0048, | |
| "step": 29250 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "grad_norm": 1.3761777877807617, | |
| "learning_rate": 9.331588655752826e-06, | |
| "loss": 0.0055, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "grad_norm": 2.5688774585723877, | |
| "learning_rate": 9.325824302513259e-06, | |
| "loss": 0.0049, | |
| "step": 29750 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "grad_norm": 0.4477705657482147, | |
| "learning_rate": 9.320059949273692e-06, | |
| "loss": 0.0051, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "eval_loss": 0.1264464259147644, | |
| "eval_runtime": 860.2965, | |
| "eval_samples_per_second": 7.848, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 11.884880099966281, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "grad_norm": 2.3739101886749268, | |
| "learning_rate": 9.314318653447085e-06, | |
| "loss": 0.0061, | |
| "step": 30250 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "grad_norm": 0.3228144645690918, | |
| "learning_rate": 9.308554300207518e-06, | |
| "loss": 0.0041, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "grad_norm": 1.3196961879730225, | |
| "learning_rate": 9.302789946967951e-06, | |
| "loss": 0.003, | |
| "step": 30750 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "grad_norm": 1.6070516109466553, | |
| "learning_rate": 9.297025593728384e-06, | |
| "loss": 0.0036, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "eval_loss": 0.1240488588809967, | |
| "eval_runtime": 867.7656, | |
| "eval_samples_per_second": 7.781, | |
| "eval_steps_per_second": 0.243, | |
| "eval_wer": 11.387032151853543, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "grad_norm": 0.3357069492340088, | |
| "learning_rate": 9.291284297901775e-06, | |
| "loss": 0.0032, | |
| "step": 31250 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "grad_norm": 0.24253146350383759, | |
| "learning_rate": 9.28551994466221e-06, | |
| "loss": 0.0039, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "grad_norm": 2.30023455619812, | |
| "learning_rate": 9.279755591422643e-06, | |
| "loss": 0.0042, | |
| "step": 31750 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "grad_norm": 0.30979445576667786, | |
| "learning_rate": 9.273991238183076e-06, | |
| "loss": 0.0041, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "eval_loss": 0.12816853821277618, | |
| "eval_runtime": 929.7626, | |
| "eval_samples_per_second": 7.262, | |
| "eval_steps_per_second": 0.227, | |
| "eval_wer": 11.621080191205348, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 7.43, | |
| "grad_norm": 1.5548877716064453, | |
| "learning_rate": 9.26822688494351e-06, | |
| "loss": 0.0043, | |
| "step": 32250 | |
| }, | |
| { | |
| "epoch": 7.49, | |
| "grad_norm": 0.23581494390964508, | |
| "learning_rate": 9.262462531703944e-06, | |
| "loss": 0.004, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 7.54, | |
| "grad_norm": 0.1430416703224182, | |
| "learning_rate": 9.256698178464377e-06, | |
| "loss": 0.0042, | |
| "step": 32750 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "grad_norm": 1.1397004127502441, | |
| "learning_rate": 9.25093382522481e-06, | |
| "loss": 0.0042, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "eval_loss": 0.13014425337314606, | |
| "eval_runtime": 972.7499, | |
| "eval_samples_per_second": 6.941, | |
| "eval_steps_per_second": 0.217, | |
| "eval_wer": 11.75198841660551, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 7.66, | |
| "grad_norm": 2.46602463722229, | |
| "learning_rate": 9.245169471985243e-06, | |
| "loss": 0.0047, | |
| "step": 33250 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "grad_norm": 0.6565764546394348, | |
| "learning_rate": 9.239405118745678e-06, | |
| "loss": 0.0039, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 7.77, | |
| "grad_norm": 0.5344523787498474, | |
| "learning_rate": 9.233640765506111e-06, | |
| "loss": 0.0048, | |
| "step": 33750 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "grad_norm": 1.3736058473587036, | |
| "learning_rate": 9.227876412266544e-06, | |
| "loss": 0.0042, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "eval_loss": 0.12513211369514465, | |
| "eval_runtime": 868.4948, | |
| "eval_samples_per_second": 7.774, | |
| "eval_steps_per_second": 0.243, | |
| "eval_wer": 11.38901560981415, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 7.89, | |
| "grad_norm": 0.4617059826850891, | |
| "learning_rate": 9.222112059026977e-06, | |
| "loss": 0.0041, | |
| "step": 34250 | |
| }, | |
| { | |
| "epoch": 7.95, | |
| "grad_norm": 0.7892825603485107, | |
| "learning_rate": 9.216347705787412e-06, | |
| "loss": 0.004, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.2047586441040039, | |
| "learning_rate": 9.210583352547845e-06, | |
| "loss": 0.0043, | |
| "step": 34750 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "grad_norm": 0.3856526017189026, | |
| "learning_rate": 9.204818999308278e-06, | |
| "loss": 0.0029, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "eval_loss": 0.12612038850784302, | |
| "eval_runtime": 863.5386, | |
| "eval_samples_per_second": 7.819, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 11.222405141123033, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "grad_norm": 1.465882420539856, | |
| "learning_rate": 9.199054646068711e-06, | |
| "loss": 0.0028, | |
| "step": 35250 | |
| }, | |
| { | |
| "epoch": 8.18, | |
| "grad_norm": 0.45589736104011536, | |
| "learning_rate": 9.193290292829146e-06, | |
| "loss": 0.0028, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 8.23, | |
| "grad_norm": 1.3802299499511719, | |
| "learning_rate": 9.187525939589579e-06, | |
| "loss": 0.0032, | |
| "step": 35750 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "grad_norm": 1.1035882234573364, | |
| "learning_rate": 9.181761586350012e-06, | |
| "loss": 0.0034, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "eval_loss": 0.1280359923839569, | |
| "eval_runtime": 861.7939, | |
| "eval_samples_per_second": 7.835, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 11.510006545411269, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 8.35, | |
| "grad_norm": 1.5274240970611572, | |
| "learning_rate": 9.176020290523405e-06, | |
| "loss": 0.0039, | |
| "step": 36250 | |
| }, | |
| { | |
| "epoch": 8.41, | |
| "grad_norm": 2.005580186843872, | |
| "learning_rate": 9.170255937283838e-06, | |
| "loss": 0.0033, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 8.46, | |
| "grad_norm": 0.9045758843421936, | |
| "learning_rate": 9.164491584044271e-06, | |
| "loss": 0.0046, | |
| "step": 36750 | |
| }, | |
| { | |
| "epoch": 8.52, | |
| "grad_norm": 1.8008005619049072, | |
| "learning_rate": 9.158727230804704e-06, | |
| "loss": 0.0035, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 8.52, | |
| "eval_loss": 0.13088497519493103, | |
| "eval_runtime": 856.6628, | |
| "eval_samples_per_second": 7.882, | |
| "eval_steps_per_second": 0.246, | |
| "eval_wer": 11.527857667056747, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 8.58, | |
| "grad_norm": 1.5927373170852661, | |
| "learning_rate": 9.152962877565137e-06, | |
| "loss": 0.0037, | |
| "step": 37250 | |
| }, | |
| { | |
| "epoch": 8.64, | |
| "grad_norm": 1.2567116022109985, | |
| "learning_rate": 9.14722158173853e-06, | |
| "loss": 0.0035, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 8.69, | |
| "grad_norm": 1.1662616729736328, | |
| "learning_rate": 9.141457228498963e-06, | |
| "loss": 0.0039, | |
| "step": 37750 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "grad_norm": 1.5941380262374878, | |
| "learning_rate": 9.135692875259396e-06, | |
| "loss": 0.0043, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "eval_loss": 0.1274692267179489, | |
| "eval_runtime": 865.0862, | |
| "eval_samples_per_second": 7.805, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 11.513973461332487, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 8.81, | |
| "grad_norm": 1.7074130773544312, | |
| "learning_rate": 9.12992852201983e-06, | |
| "loss": 0.004, | |
| "step": 38250 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "grad_norm": 1.6893212795257568, | |
| "learning_rate": 9.124164168780264e-06, | |
| "loss": 0.004, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 8.92, | |
| "grad_norm": 1.044587254524231, | |
| "learning_rate": 9.118399815540697e-06, | |
| "loss": 0.0035, | |
| "step": 38750 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "grad_norm": 0.12974177300930023, | |
| "learning_rate": 9.11263546230113e-06, | |
| "loss": 0.004, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "eval_loss": 0.12444661557674408, | |
| "eval_runtime": 872.943, | |
| "eval_samples_per_second": 7.735, | |
| "eval_steps_per_second": 0.242, | |
| "eval_wer": 11.139099906777476, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "grad_norm": 1.5708075761795044, | |
| "learning_rate": 9.106871109061563e-06, | |
| "loss": 0.0031, | |
| "step": 39250 | |
| }, | |
| { | |
| "epoch": 9.1, | |
| "grad_norm": 1.6801316738128662, | |
| "learning_rate": 9.101106755821998e-06, | |
| "loss": 0.0023, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 9.15, | |
| "grad_norm": 1.937129020690918, | |
| "learning_rate": 9.095342402582431e-06, | |
| "loss": 0.0025, | |
| "step": 39750 | |
| }, | |
| { | |
| "epoch": 9.21, | |
| "grad_norm": 0.7914566993713379, | |
| "learning_rate": 9.089578049342864e-06, | |
| "loss": 0.0032, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 9.21, | |
| "eval_loss": 0.12887698411941528, | |
| "eval_runtime": 862.542, | |
| "eval_samples_per_second": 7.828, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 11.561576452387092, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 9.27, | |
| "grad_norm": 0.04653190076351166, | |
| "learning_rate": 9.083813696103297e-06, | |
| "loss": 0.0027, | |
| "step": 40250 | |
| }, | |
| { | |
| "epoch": 9.33, | |
| "grad_norm": 0.34394124150276184, | |
| "learning_rate": 9.07807240027669e-06, | |
| "loss": 0.0033, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 9.39, | |
| "grad_norm": 0.496722936630249, | |
| "learning_rate": 9.072308047037123e-06, | |
| "loss": 0.003, | |
| "step": 40750 | |
| }, | |
| { | |
| "epoch": 9.44, | |
| "grad_norm": 0.2492765635251999, | |
| "learning_rate": 9.066543693797556e-06, | |
| "loss": 0.0029, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 9.44, | |
| "eval_loss": 0.12612169981002808, | |
| "eval_runtime": 862.7524, | |
| "eval_samples_per_second": 7.826, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 11.252157010532162, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "grad_norm": 2.209688425064087, | |
| "learning_rate": 9.06077934055799e-06, | |
| "loss": 0.0031, | |
| "step": 41250 | |
| }, | |
| { | |
| "epoch": 9.56, | |
| "grad_norm": 0.5090010166168213, | |
| "learning_rate": 9.055014987318424e-06, | |
| "loss": 0.0032, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 9.62, | |
| "grad_norm": 0.561119794845581, | |
| "learning_rate": 9.049250634078857e-06, | |
| "loss": 0.0039, | |
| "step": 41750 | |
| }, | |
| { | |
| "epoch": 9.67, | |
| "grad_norm": 0.20059019327163696, | |
| "learning_rate": 9.04348628083929e-06, | |
| "loss": 0.0031, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 9.67, | |
| "eval_loss": 0.1292741298675537, | |
| "eval_runtime": 864.5766, | |
| "eval_samples_per_second": 7.81, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 11.369181030208065, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 9.73, | |
| "grad_norm": 0.5912229418754578, | |
| "learning_rate": 9.037721927599723e-06, | |
| "loss": 0.0032, | |
| "step": 42250 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "grad_norm": 0.274827241897583, | |
| "learning_rate": 9.031957574360158e-06, | |
| "loss": 0.0034, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 9.85, | |
| "grad_norm": 1.7666079998016357, | |
| "learning_rate": 9.026193221120591e-06, | |
| "loss": 0.0037, | |
| "step": 42750 | |
| }, | |
| { | |
| "epoch": 9.9, | |
| "grad_norm": 0.4969967305660248, | |
| "learning_rate": 9.020428867881024e-06, | |
| "loss": 0.0031, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 9.9, | |
| "eval_loss": 0.1289782077074051, | |
| "eval_runtime": 871.6635, | |
| "eval_samples_per_second": 7.746, | |
| "eval_steps_per_second": 0.242, | |
| "eval_wer": 11.055794672431919, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 9.96, | |
| "grad_norm": 2.4347357749938965, | |
| "learning_rate": 9.014687572054416e-06, | |
| "loss": 0.0033, | |
| "step": 43250 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "grad_norm": 0.36096087098121643, | |
| "learning_rate": 9.008946276227809e-06, | |
| "loss": 0.004, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 10.08, | |
| "grad_norm": 0.30262959003448486, | |
| "learning_rate": 9.003181922988242e-06, | |
| "loss": 0.0022, | |
| "step": 43750 | |
| }, | |
| { | |
| "epoch": 10.13, | |
| "grad_norm": 0.9803399443626404, | |
| "learning_rate": 8.997417569748675e-06, | |
| "loss": 0.0021, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 10.13, | |
| "eval_loss": 0.12753725051879883, | |
| "eval_runtime": 848.9932, | |
| "eval_samples_per_second": 7.953, | |
| "eval_steps_per_second": 0.249, | |
| "eval_wer": 11.154967570462345, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 10.19, | |
| "grad_norm": 0.08501572906970978, | |
| "learning_rate": 8.991653216509108e-06, | |
| "loss": 0.0023, | |
| "step": 44250 | |
| }, | |
| { | |
| "epoch": 10.25, | |
| "grad_norm": 0.17812615633010864, | |
| "learning_rate": 8.985888863269543e-06, | |
| "loss": 0.0022, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 10.31, | |
| "grad_norm": 0.192487895488739, | |
| "learning_rate": 8.980124510029976e-06, | |
| "loss": 0.0023, | |
| "step": 44750 | |
| }, | |
| { | |
| "epoch": 10.36, | |
| "grad_norm": 1.9704803228378296, | |
| "learning_rate": 8.974360156790409e-06, | |
| "loss": 0.0026, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 10.36, | |
| "eval_loss": 0.1309252232313156, | |
| "eval_runtime": 858.8815, | |
| "eval_samples_per_second": 7.861, | |
| "eval_steps_per_second": 0.246, | |
| "eval_wer": 11.109348037368347, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 10.42, | |
| "grad_norm": 0.22806860506534576, | |
| "learning_rate": 8.968595803550842e-06, | |
| "loss": 0.0027, | |
| "step": 45250 | |
| }, | |
| { | |
| "epoch": 10.48, | |
| "grad_norm": 0.2697317600250244, | |
| "learning_rate": 8.962831450311276e-06, | |
| "loss": 0.0031, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 10.54, | |
| "grad_norm": 0.503118097782135, | |
| "learning_rate": 8.95706709707171e-06, | |
| "loss": 0.0034, | |
| "step": 45750 | |
| }, | |
| { | |
| "epoch": 10.59, | |
| "grad_norm": 1.2143537998199463, | |
| "learning_rate": 8.951302743832143e-06, | |
| "loss": 0.0031, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 10.59, | |
| "eval_loss": 0.13377217948436737, | |
| "eval_runtime": 857.5578, | |
| "eval_samples_per_second": 7.874, | |
| "eval_steps_per_second": 0.246, | |
| "eval_wer": 11.539758414820398, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 10.65, | |
| "grad_norm": 0.5072382688522339, | |
| "learning_rate": 8.945538390592576e-06, | |
| "loss": 0.0034, | |
| "step": 46250 | |
| }, | |
| { | |
| "epoch": 10.71, | |
| "grad_norm": 0.45895808935165405, | |
| "learning_rate": 8.939774037353009e-06, | |
| "loss": 0.0029, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 10.77, | |
| "grad_norm": 1.3219285011291504, | |
| "learning_rate": 8.934009684113443e-06, | |
| "loss": 0.0031, | |
| "step": 46750 | |
| }, | |
| { | |
| "epoch": 10.82, | |
| "grad_norm": 0.9804620742797852, | |
| "learning_rate": 8.928245330873877e-06, | |
| "loss": 0.0035, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 10.82, | |
| "eval_loss": 0.12685059010982513, | |
| "eval_runtime": 862.1359, | |
| "eval_samples_per_second": 7.832, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 11.055794672431919, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 10.88, | |
| "grad_norm": 0.4526994228363037, | |
| "learning_rate": 8.92248097763431e-06, | |
| "loss": 0.0034, | |
| "step": 47250 | |
| }, | |
| { | |
| "epoch": 10.94, | |
| "grad_norm": 0.8010832667350769, | |
| "learning_rate": 8.916739681807703e-06, | |
| "loss": 0.0064, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.5813642144203186, | |
| "learning_rate": 8.910975328568136e-06, | |
| "loss": 0.0031, | |
| "step": 47750 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "grad_norm": 0.029226483777165413, | |
| "learning_rate": 8.905210975328569e-06, | |
| "loss": 0.0021, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "eval_loss": 0.1262122392654419, | |
| "eval_runtime": 848.9631, | |
| "eval_samples_per_second": 7.953, | |
| "eval_steps_per_second": 0.249, | |
| "eval_wer": 11.028026260983397, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 11.11, | |
| "grad_norm": 1.0615992546081543, | |
| "learning_rate": 8.899446622089002e-06, | |
| "loss": 0.0021, | |
| "step": 48250 | |
| }, | |
| { | |
| "epoch": 11.17, | |
| "grad_norm": 1.0726251602172852, | |
| "learning_rate": 8.893682268849437e-06, | |
| "loss": 0.0019, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 11.23, | |
| "grad_norm": 0.3940158486366272, | |
| "learning_rate": 8.88791791560987e-06, | |
| "loss": 0.0026, | |
| "step": 48750 | |
| }, | |
| { | |
| "epoch": 11.29, | |
| "grad_norm": 0.5974787473678589, | |
| "learning_rate": 8.882153562370303e-06, | |
| "loss": 0.0025, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 11.29, | |
| "eval_loss": 0.12749157845973969, | |
| "eval_runtime": 861.4829, | |
| "eval_samples_per_second": 7.838, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 11.279925421980682, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 11.34, | |
| "grad_norm": 1.4202208518981934, | |
| "learning_rate": 8.876389209130736e-06, | |
| "loss": 0.0023, | |
| "step": 49250 | |
| }, | |
| { | |
| "epoch": 11.4, | |
| "grad_norm": 1.4934862852096558, | |
| "learning_rate": 8.87062485589117e-06, | |
| "loss": 0.0026, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 11.46, | |
| "grad_norm": 1.8105831146240234, | |
| "learning_rate": 8.864860502651604e-06, | |
| "loss": 0.0026, | |
| "step": 49750 | |
| }, | |
| { | |
| "epoch": 11.52, | |
| "grad_norm": 0.6586973071098328, | |
| "learning_rate": 8.859096149412037e-06, | |
| "loss": 0.0024, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 11.52, | |
| "eval_loss": 0.13289304077625275, | |
| "eval_runtime": 861.8006, | |
| "eval_samples_per_second": 7.835, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 11.06769542019557, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 11.57, | |
| "grad_norm": 0.08556759357452393, | |
| "learning_rate": 8.85333179617247e-06, | |
| "loss": 0.0027, | |
| "step": 50250 | |
| }, | |
| { | |
| "epoch": 11.63, | |
| "grad_norm": 0.31020841002464294, | |
| "learning_rate": 8.847567442932904e-06, | |
| "loss": 0.0028, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 11.69, | |
| "grad_norm": 0.2600391209125519, | |
| "learning_rate": 8.841803089693337e-06, | |
| "loss": 0.003, | |
| "step": 50750 | |
| }, | |
| { | |
| "epoch": 11.75, | |
| "grad_norm": 1.2706468105316162, | |
| "learning_rate": 8.83603873645377e-06, | |
| "loss": 0.0031, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 11.75, | |
| "eval_loss": 0.13028421998023987, | |
| "eval_runtime": 861.6838, | |
| "eval_samples_per_second": 7.836, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 10.93083682091358, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 11.8, | |
| "grad_norm": 0.8451604247093201, | |
| "learning_rate": 8.830274383214205e-06, | |
| "loss": 0.003, | |
| "step": 51250 | |
| }, | |
| { | |
| "epoch": 11.86, | |
| "grad_norm": 1.8342962265014648, | |
| "learning_rate": 8.824510029974638e-06, | |
| "loss": 0.0033, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 11.92, | |
| "grad_norm": 0.15427502989768982, | |
| "learning_rate": 8.818745676735071e-06, | |
| "loss": 0.0028, | |
| "step": 51750 | |
| }, | |
| { | |
| "epoch": 11.98, | |
| "grad_norm": 1.4967540502548218, | |
| "learning_rate": 8.812981323495504e-06, | |
| "loss": 0.0028, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 11.98, | |
| "eval_loss": 0.13051362335681915, | |
| "eval_runtime": 865.4409, | |
| "eval_samples_per_second": 7.802, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 11.228355515004859, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 12.03, | |
| "grad_norm": 1.1010907888412476, | |
| "learning_rate": 8.807240027668896e-06, | |
| "loss": 0.0022, | |
| "step": 52250 | |
| }, | |
| { | |
| "epoch": 12.09, | |
| "grad_norm": 0.7162982821464539, | |
| "learning_rate": 8.801475674429329e-06, | |
| "loss": 0.0018, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 12.15, | |
| "grad_norm": 0.30756324529647827, | |
| "learning_rate": 8.795711321189762e-06, | |
| "loss": 0.0023, | |
| "step": 52750 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "grad_norm": 1.2408708333969116, | |
| "learning_rate": 8.789946967950197e-06, | |
| "loss": 0.002, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "eval_loss": 0.13401812314987183, | |
| "eval_runtime": 859.7266, | |
| "eval_samples_per_second": 7.854, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 11.105381121447131, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 12.26, | |
| "grad_norm": 1.0242916345596313, | |
| "learning_rate": 8.78418261471063e-06, | |
| "loss": 0.0022, | |
| "step": 53250 | |
| }, | |
| { | |
| "epoch": 12.32, | |
| "grad_norm": 0.08462110161781311, | |
| "learning_rate": 8.778418261471063e-06, | |
| "loss": 0.0023, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 12.38, | |
| "grad_norm": 0.19610953330993652, | |
| "learning_rate": 8.772653908231497e-06, | |
| "loss": 0.0021, | |
| "step": 53750 | |
| }, | |
| { | |
| "epoch": 12.44, | |
| "grad_norm": 2.1351606845855713, | |
| "learning_rate": 8.76688955499193e-06, | |
| "loss": 0.0025, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 12.44, | |
| "eval_loss": 0.12935225665569305, | |
| "eval_runtime": 862.1306, | |
| "eval_samples_per_second": 7.832, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 10.899101493543844, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 12.49, | |
| "grad_norm": 1.4103162288665771, | |
| "learning_rate": 8.761125201752364e-06, | |
| "loss": 0.0021, | |
| "step": 54250 | |
| }, | |
| { | |
| "epoch": 12.55, | |
| "grad_norm": 0.27089348435401917, | |
| "learning_rate": 8.755360848512797e-06, | |
| "loss": 0.0023, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 12.61, | |
| "grad_norm": 2.1209752559661865, | |
| "learning_rate": 8.749596495273231e-06, | |
| "loss": 0.0026, | |
| "step": 54750 | |
| }, | |
| { | |
| "epoch": 12.67, | |
| "grad_norm": 0.4095695912837982, | |
| "learning_rate": 8.743855199446623e-06, | |
| "loss": 0.0025, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 12.67, | |
| "eval_loss": 0.12996189296245575, | |
| "eval_runtime": 859.3218, | |
| "eval_samples_per_second": 7.857, | |
| "eval_steps_per_second": 0.246, | |
| "eval_wer": 10.893151119662019, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 12.72, | |
| "grad_norm": 0.7925490736961365, | |
| "learning_rate": 8.738090846207057e-06, | |
| "loss": 0.0027, | |
| "step": 55250 | |
| }, | |
| { | |
| "epoch": 12.78, | |
| "grad_norm": 0.7041000723838806, | |
| "learning_rate": 8.73232649296749e-06, | |
| "loss": 0.0021, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 12.84, | |
| "grad_norm": 1.976866364479065, | |
| "learning_rate": 8.726562139727924e-06, | |
| "loss": 0.0028, | |
| "step": 55750 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "grad_norm": 0.39583584666252136, | |
| "learning_rate": 8.720820843901315e-06, | |
| "loss": 0.0025, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "eval_loss": 0.13089019060134888, | |
| "eval_runtime": 856.9052, | |
| "eval_samples_per_second": 7.88, | |
| "eval_steps_per_second": 0.246, | |
| "eval_wer": 10.791994763670983, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 12.95, | |
| "grad_norm": 0.4872288405895233, | |
| "learning_rate": 8.715056490661748e-06, | |
| "loss": 0.0027, | |
| "step": 56250 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "grad_norm": 2.2456576824188232, | |
| "learning_rate": 8.709292137422181e-06, | |
| "loss": 0.0025, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 13.07, | |
| "grad_norm": 0.20264917612075806, | |
| "learning_rate": 8.703527784182614e-06, | |
| "loss": 0.0018, | |
| "step": 56750 | |
| }, | |
| { | |
| "epoch": 13.13, | |
| "grad_norm": 0.3795398771762848, | |
| "learning_rate": 8.697763430943049e-06, | |
| "loss": 0.0016, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 13.13, | |
| "eval_loss": 0.12614800035953522, | |
| "eval_runtime": 860.8165, | |
| "eval_samples_per_second": 7.844, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 10.69083840767995, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 13.19, | |
| "grad_norm": 2.3237907886505127, | |
| "learning_rate": 8.691999077703482e-06, | |
| "loss": 0.0017, | |
| "step": 57250 | |
| }, | |
| { | |
| "epoch": 13.24, | |
| "grad_norm": 0.7427815794944763, | |
| "learning_rate": 8.686234724463915e-06, | |
| "loss": 0.002, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 13.3, | |
| "grad_norm": 0.1203819140791893, | |
| "learning_rate": 8.68047037122435e-06, | |
| "loss": 0.002, | |
| "step": 57750 | |
| }, | |
| { | |
| "epoch": 13.36, | |
| "grad_norm": 0.11056603491306305, | |
| "learning_rate": 8.674706017984783e-06, | |
| "loss": 0.0024, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 13.36, | |
| "eval_loss": 0.13082969188690186, | |
| "eval_runtime": 861.4091, | |
| "eval_samples_per_second": 7.838, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 10.873316540055933, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 13.42, | |
| "grad_norm": 0.3026280701160431, | |
| "learning_rate": 8.668941664745216e-06, | |
| "loss": 0.002, | |
| "step": 58250 | |
| }, | |
| { | |
| "epoch": 13.47, | |
| "grad_norm": 1.0132057666778564, | |
| "learning_rate": 8.663177311505649e-06, | |
| "loss": 0.0023, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 13.53, | |
| "grad_norm": 0.26512446999549866, | |
| "learning_rate": 8.657412958266084e-06, | |
| "loss": 0.0025, | |
| "step": 58750 | |
| }, | |
| { | |
| "epoch": 13.59, | |
| "grad_norm": 1.13190758228302, | |
| "learning_rate": 8.651648605026517e-06, | |
| "loss": 0.0022, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 13.59, | |
| "eval_loss": 0.13342134654521942, | |
| "eval_runtime": 863.3109, | |
| "eval_samples_per_second": 7.821, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 10.825713549001328, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 13.65, | |
| "grad_norm": 0.338905930519104, | |
| "learning_rate": 8.64588425178695e-06, | |
| "loss": 0.0022, | |
| "step": 59250 | |
| }, | |
| { | |
| "epoch": 13.7, | |
| "grad_norm": 0.8769376873970032, | |
| "learning_rate": 8.640119898547383e-06, | |
| "loss": 0.0021, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 13.76, | |
| "grad_norm": 0.6874887943267822, | |
| "learning_rate": 8.634355545307818e-06, | |
| "loss": 0.0023, | |
| "step": 59750 | |
| }, | |
| { | |
| "epoch": 13.82, | |
| "grad_norm": 0.38465040922164917, | |
| "learning_rate": 8.62859119206825e-06, | |
| "loss": 0.0027, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 13.82, | |
| "eval_loss": 0.1326436996459961, | |
| "eval_runtime": 862.9972, | |
| "eval_samples_per_second": 7.824, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 10.978439811968185, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 13.88, | |
| "grad_norm": 0.63032466173172, | |
| "learning_rate": 8.622826838828684e-06, | |
| "loss": 0.0025, | |
| "step": 60250 | |
| }, | |
| { | |
| "epoch": 13.93, | |
| "grad_norm": 1.5229909420013428, | |
| "learning_rate": 8.617085543002077e-06, | |
| "loss": 0.0028, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 13.99, | |
| "grad_norm": 1.10037100315094, | |
| "learning_rate": 8.61132118976251e-06, | |
| "loss": 0.0023, | |
| "step": 60750 | |
| }, | |
| { | |
| "epoch": 14.05, | |
| "grad_norm": 0.09242303669452667, | |
| "learning_rate": 8.605556836522943e-06, | |
| "loss": 0.0014, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 14.05, | |
| "eval_loss": 0.13251730799674988, | |
| "eval_runtime": 858.8112, | |
| "eval_samples_per_second": 7.862, | |
| "eval_steps_per_second": 0.246, | |
| "eval_wer": 10.694805323601166, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 14.11, | |
| "grad_norm": 2.284032106399536, | |
| "learning_rate": 8.599792483283378e-06, | |
| "loss": 0.002, | |
| "step": 61250 | |
| }, | |
| { | |
| "epoch": 14.16, | |
| "grad_norm": 0.04196714237332344, | |
| "learning_rate": 8.59402813004381e-06, | |
| "loss": 0.0016, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 14.22, | |
| "grad_norm": 1.1200813055038452, | |
| "learning_rate": 8.588263776804244e-06, | |
| "loss": 0.0017, | |
| "step": 61750 | |
| }, | |
| { | |
| "epoch": 14.28, | |
| "grad_norm": 0.09567234665155411, | |
| "learning_rate": 8.582499423564677e-06, | |
| "loss": 0.002, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 14.28, | |
| "eval_loss": 0.13003449141979218, | |
| "eval_runtime": 856.542, | |
| "eval_samples_per_second": 7.883, | |
| "eval_steps_per_second": 0.246, | |
| "eval_wer": 10.75629252038003, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 14.34, | |
| "grad_norm": 0.9637365937232971, | |
| "learning_rate": 8.576781185151026e-06, | |
| "loss": 0.0019, | |
| "step": 62250 | |
| }, | |
| { | |
| "epoch": 14.39, | |
| "grad_norm": 0.9736013412475586, | |
| "learning_rate": 8.571016831911461e-06, | |
| "loss": 0.0019, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 14.45, | |
| "grad_norm": 0.8420233130455017, | |
| "learning_rate": 8.565252478671894e-06, | |
| "loss": 0.0022, | |
| "step": 62750 | |
| }, | |
| { | |
| "epoch": 14.51, | |
| "grad_norm": 0.3317677974700928, | |
| "learning_rate": 8.559488125432327e-06, | |
| "loss": 0.0022, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 14.51, | |
| "eval_loss": 0.13171714544296265, | |
| "eval_runtime": 847.1312, | |
| "eval_samples_per_second": 7.97, | |
| "eval_steps_per_second": 0.249, | |
| "eval_wer": 10.712656445246644, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 14.57, | |
| "grad_norm": 1.777006983757019, | |
| "learning_rate": 8.55372377219276e-06, | |
| "loss": 0.0024, | |
| "step": 63250 | |
| }, | |
| { | |
| "epoch": 14.62, | |
| "grad_norm": 0.9992290735244751, | |
| "learning_rate": 8.547959418953195e-06, | |
| "loss": 0.0023, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 14.68, | |
| "grad_norm": 1.5052534341812134, | |
| "learning_rate": 8.542195065713628e-06, | |
| "loss": 0.0024, | |
| "step": 63750 | |
| }, | |
| { | |
| "epoch": 14.74, | |
| "grad_norm": 0.3194350600242615, | |
| "learning_rate": 8.536430712474061e-06, | |
| "loss": 0.0025, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 14.74, | |
| "eval_loss": 0.12849073112010956, | |
| "eval_runtime": 852.5941, | |
| "eval_samples_per_second": 7.919, | |
| "eval_steps_per_second": 0.247, | |
| "eval_wer": 10.587698593728307, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 14.8, | |
| "grad_norm": 0.10011152178049088, | |
| "learning_rate": 8.530666359234494e-06, | |
| "loss": 0.0021, | |
| "step": 64250 | |
| }, | |
| { | |
| "epoch": 14.85, | |
| "grad_norm": 2.3586697578430176, | |
| "learning_rate": 8.524902005994929e-06, | |
| "loss": 0.0024, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 14.91, | |
| "grad_norm": 0.3267650008201599, | |
| "learning_rate": 8.519137652755362e-06, | |
| "loss": 0.0023, | |
| "step": 64750 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "grad_norm": 0.058121953159570694, | |
| "learning_rate": 8.513373299515795e-06, | |
| "loss": 0.0025, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "eval_loss": 0.12760809063911438, | |
| "eval_runtime": 865.2509, | |
| "eval_samples_per_second": 7.804, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 10.496459527540313, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 15.03, | |
| "grad_norm": 0.024574730545282364, | |
| "learning_rate": 8.50760894627623e-06, | |
| "loss": 0.0016, | |
| "step": 65250 | |
| }, | |
| { | |
| "epoch": 15.09, | |
| "grad_norm": 0.04577196016907692, | |
| "learning_rate": 8.501844593036663e-06, | |
| "loss": 0.0016, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 15.14, | |
| "grad_norm": 1.1313352584838867, | |
| "learning_rate": 8.496080239797096e-06, | |
| "loss": 0.0018, | |
| "step": 65750 | |
| }, | |
| { | |
| "epoch": 15.2, | |
| "grad_norm": 0.08705839514732361, | |
| "learning_rate": 8.490315886557529e-06, | |
| "loss": 0.0013, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 15.2, | |
| "eval_loss": 0.13220834732055664, | |
| "eval_runtime": 857.8748, | |
| "eval_samples_per_second": 7.871, | |
| "eval_steps_per_second": 0.246, | |
| "eval_wer": 10.619433921098043, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 15.26, | |
| "grad_norm": 1.0441516637802124, | |
| "learning_rate": 8.484551533317962e-06, | |
| "loss": 0.0014, | |
| "step": 66250 | |
| }, | |
| { | |
| "epoch": 15.32, | |
| "grad_norm": 1.5698387622833252, | |
| "learning_rate": 8.478787180078397e-06, | |
| "loss": 0.0016, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 15.37, | |
| "grad_norm": 1.2932049036026, | |
| "learning_rate": 8.47302282683883e-06, | |
| "loss": 0.0021, | |
| "step": 66750 | |
| }, | |
| { | |
| "epoch": 15.43, | |
| "grad_norm": 0.2987584173679352, | |
| "learning_rate": 8.467258473599263e-06, | |
| "loss": 0.0022, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 15.43, | |
| "eval_loss": 0.13362213969230652, | |
| "eval_runtime": 860.791, | |
| "eval_samples_per_second": 7.844, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 10.728524108931511, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 15.49, | |
| "grad_norm": 0.15311214327812195, | |
| "learning_rate": 8.461494120359696e-06, | |
| "loss": 0.0019, | |
| "step": 67250 | |
| }, | |
| { | |
| "epoch": 15.55, | |
| "grad_norm": 0.13635623455047607, | |
| "learning_rate": 8.455752824533087e-06, | |
| "loss": 0.0021, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 15.6, | |
| "grad_norm": 0.4941750168800354, | |
| "learning_rate": 8.449988471293522e-06, | |
| "loss": 0.002, | |
| "step": 67750 | |
| }, | |
| { | |
| "epoch": 15.66, | |
| "grad_norm": 0.04697488993406296, | |
| "learning_rate": 8.444224118053955e-06, | |
| "loss": 0.0025, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 15.66, | |
| "eval_loss": 0.13006770610809326, | |
| "eval_runtime": 863.9415, | |
| "eval_samples_per_second": 7.815, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 10.506376817343357, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 15.72, | |
| "grad_norm": 1.116538166999817, | |
| "learning_rate": 8.438459764814388e-06, | |
| "loss": 0.0021, | |
| "step": 68250 | |
| }, | |
| { | |
| "epoch": 15.78, | |
| "grad_norm": 0.79182368516922, | |
| "learning_rate": 8.432695411574821e-06, | |
| "loss": 0.0019, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 15.83, | |
| "grad_norm": 0.18393570184707642, | |
| "learning_rate": 8.426931058335256e-06, | |
| "loss": 0.0018, | |
| "step": 68750 | |
| }, | |
| { | |
| "epoch": 15.89, | |
| "grad_norm": 0.31326913833618164, | |
| "learning_rate": 8.421166705095689e-06, | |
| "loss": 0.0022, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 15.89, | |
| "eval_loss": 0.1315659135580063, | |
| "eval_runtime": 863.1675, | |
| "eval_samples_per_second": 7.822, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 10.50042644346153, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 15.95, | |
| "grad_norm": 2.326347827911377, | |
| "learning_rate": 8.415402351856122e-06, | |
| "loss": 0.0018, | |
| "step": 69250 | |
| }, | |
| { | |
| "epoch": 16.01, | |
| "grad_norm": 0.02481095865368843, | |
| "learning_rate": 8.409637998616555e-06, | |
| "loss": 0.0019, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 16.06, | |
| "grad_norm": 0.3131246566772461, | |
| "learning_rate": 8.40387364537699e-06, | |
| "loss": 0.0013, | |
| "step": 69750 | |
| }, | |
| { | |
| "epoch": 16.12, | |
| "grad_norm": 0.11796487122774124, | |
| "learning_rate": 8.398109292137423e-06, | |
| "loss": 0.0019, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 16.12, | |
| "eval_loss": 0.13220052421092987, | |
| "eval_runtime": 862.684, | |
| "eval_samples_per_second": 7.827, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 10.432988872800841, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 16.18, | |
| "grad_norm": 0.130937397480011, | |
| "learning_rate": 8.392344938897856e-06, | |
| "loss": 0.0017, | |
| "step": 70250 | |
| }, | |
| { | |
| "epoch": 16.24, | |
| "grad_norm": 0.2919136881828308, | |
| "learning_rate": 8.386580585658289e-06, | |
| "loss": 0.0016, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 16.29, | |
| "grad_norm": 0.09921443462371826, | |
| "learning_rate": 8.380816232418724e-06, | |
| "loss": 0.0014, | |
| "step": 70750 | |
| }, | |
| { | |
| "epoch": 16.35, | |
| "grad_norm": 0.490876168012619, | |
| "learning_rate": 8.375051879179157e-06, | |
| "loss": 0.0018, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 16.35, | |
| "eval_loss": 0.1313299685716629, | |
| "eval_runtime": 868.192, | |
| "eval_samples_per_second": 7.777, | |
| "eval_steps_per_second": 0.243, | |
| "eval_wer": 10.671003828073863, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 16.41, | |
| "grad_norm": 0.18082498013973236, | |
| "learning_rate": 8.36928752593959e-06, | |
| "loss": 0.0018, | |
| "step": 71250 | |
| }, | |
| { | |
| "epoch": 16.47, | |
| "grad_norm": 0.24912157654762268, | |
| "learning_rate": 8.363523172700023e-06, | |
| "loss": 0.0017, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 16.52, | |
| "grad_norm": 0.12575741112232208, | |
| "learning_rate": 8.357758819460458e-06, | |
| "loss": 0.0017, | |
| "step": 71750 | |
| }, | |
| { | |
| "epoch": 16.58, | |
| "grad_norm": 1.444352626800537, | |
| "learning_rate": 8.35199446622089e-06, | |
| "loss": 0.0021, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 16.58, | |
| "eval_loss": 0.13000106811523438, | |
| "eval_runtime": 860.5334, | |
| "eval_samples_per_second": 7.846, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 10.692821865640557, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 16.64, | |
| "grad_norm": 0.48708584904670715, | |
| "learning_rate": 8.346230112981324e-06, | |
| "loss": 0.0023, | |
| "step": 72250 | |
| }, | |
| { | |
| "epoch": 16.7, | |
| "grad_norm": 1.296075463294983, | |
| "learning_rate": 8.340488817154715e-06, | |
| "loss": 0.002, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 16.75, | |
| "grad_norm": 0.3294328451156616, | |
| "learning_rate": 8.334747521328108e-06, | |
| "loss": 0.0021, | |
| "step": 72750 | |
| }, | |
| { | |
| "epoch": 16.81, | |
| "grad_norm": 0.04183769226074219, | |
| "learning_rate": 8.328983168088541e-06, | |
| "loss": 0.0023, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 16.81, | |
| "eval_loss": 0.12990330159664154, | |
| "eval_runtime": 882.105, | |
| "eval_samples_per_second": 7.654, | |
| "eval_steps_per_second": 0.239, | |
| "eval_wer": 10.319931769046155, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 16.87, | |
| "grad_norm": 0.42695850133895874, | |
| "learning_rate": 8.323241872261933e-06, | |
| "loss": 0.002, | |
| "step": 73250 | |
| }, | |
| { | |
| "epoch": 16.93, | |
| "grad_norm": 0.2880449593067169, | |
| "learning_rate": 8.317477519022367e-06, | |
| "loss": 0.0021, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "grad_norm": 1.023850917816162, | |
| "learning_rate": 8.3117131657828e-06, | |
| "loss": 0.0017, | |
| "step": 73750 | |
| }, | |
| { | |
| "epoch": 17.04, | |
| "grad_norm": 1.6181273460388184, | |
| "learning_rate": 8.305948812543234e-06, | |
| "loss": 0.0015, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 17.04, | |
| "eval_loss": 0.13121721148490906, | |
| "eval_runtime": 859.3639, | |
| "eval_samples_per_second": 7.857, | |
| "eval_steps_per_second": 0.246, | |
| "eval_wer": 10.403237003391713, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 17.1, | |
| "grad_norm": 0.26772114634513855, | |
| "learning_rate": 8.300184459303667e-06, | |
| "loss": 0.0014, | |
| "step": 74250 | |
| }, | |
| { | |
| "epoch": 17.16, | |
| "grad_norm": 3.353755235671997, | |
| "learning_rate": 8.294420106064101e-06, | |
| "loss": 0.0009, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 17.22, | |
| "grad_norm": 1.0177912712097168, | |
| "learning_rate": 8.288655752824534e-06, | |
| "loss": 0.0013, | |
| "step": 74750 | |
| }, | |
| { | |
| "epoch": 17.27, | |
| "grad_norm": 1.6711407899856567, | |
| "learning_rate": 8.282891399584967e-06, | |
| "loss": 0.0016, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 17.27, | |
| "eval_loss": 0.13091835379600525, | |
| "eval_runtime": 876.3454, | |
| "eval_samples_per_second": 7.705, | |
| "eval_steps_per_second": 0.241, | |
| "eval_wer": 10.928853362952973, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 17.33, | |
| "grad_norm": 2.060011148452759, | |
| "learning_rate": 8.2771270463454e-06, | |
| "loss": 0.0015, | |
| "step": 75250 | |
| }, | |
| { | |
| "epoch": 17.39, | |
| "grad_norm": 0.3818737864494324, | |
| "learning_rate": 8.271362693105834e-06, | |
| "loss": 0.0015, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 17.45, | |
| "grad_norm": 0.08700312674045563, | |
| "learning_rate": 8.265598339866268e-06, | |
| "loss": 0.0014, | |
| "step": 75750 | |
| }, | |
| { | |
| "epoch": 17.5, | |
| "grad_norm": 0.17121294140815735, | |
| "learning_rate": 8.259833986626701e-06, | |
| "loss": 0.0019, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 17.5, | |
| "eval_loss": 0.13205237686634064, | |
| "eval_runtime": 856.9406, | |
| "eval_samples_per_second": 7.879, | |
| "eval_steps_per_second": 0.246, | |
| "eval_wer": 10.40918737727354, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 17.56, | |
| "grad_norm": 0.05054645985364914, | |
| "learning_rate": 8.254069633387134e-06, | |
| "loss": 0.0017, | |
| "step": 76250 | |
| }, | |
| { | |
| "epoch": 17.62, | |
| "grad_norm": 0.03382011130452156, | |
| "learning_rate": 8.248305280147568e-06, | |
| "loss": 0.002, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 17.68, | |
| "grad_norm": 0.9019294381141663, | |
| "learning_rate": 8.242540926908002e-06, | |
| "loss": 0.0016, | |
| "step": 76750 | |
| }, | |
| { | |
| "epoch": 17.73, | |
| "grad_norm": 0.12265095859766006, | |
| "learning_rate": 8.236776573668435e-06, | |
| "loss": 0.0016, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 17.73, | |
| "eval_loss": 0.12913843989372253, | |
| "eval_runtime": 853.8374, | |
| "eval_samples_per_second": 7.908, | |
| "eval_steps_per_second": 0.247, | |
| "eval_wer": 10.337782890691633, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 17.79, | |
| "grad_norm": 1.6407935619354248, | |
| "learning_rate": 8.231012220428868e-06, | |
| "loss": 0.0019, | |
| "step": 77250 | |
| }, | |
| { | |
| "epoch": 17.85, | |
| "grad_norm": 0.5586839914321899, | |
| "learning_rate": 8.22527092460226e-06, | |
| "loss": 0.0016, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 17.91, | |
| "grad_norm": 2.7545528411865234, | |
| "learning_rate": 8.219506571362694e-06, | |
| "loss": 0.0019, | |
| "step": 77750 | |
| }, | |
| { | |
| "epoch": 17.96, | |
| "grad_norm": 0.10660357773303986, | |
| "learning_rate": 8.213742218123128e-06, | |
| "loss": 0.0019, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 17.96, | |
| "eval_loss": 0.12859106063842773, | |
| "eval_runtime": 860.0021, | |
| "eval_samples_per_second": 7.851, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 10.310014479243112, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 18.02, | |
| "grad_norm": 0.057170018553733826, | |
| "learning_rate": 8.20797786488356e-06, | |
| "loss": 0.0019, | |
| "step": 78250 | |
| }, | |
| { | |
| "epoch": 18.08, | |
| "grad_norm": 0.8357897996902466, | |
| "learning_rate": 8.202213511643994e-06, | |
| "loss": 0.0013, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 18.14, | |
| "grad_norm": 0.2379799783229828, | |
| "learning_rate": 8.196449158404428e-06, | |
| "loss": 0.0015, | |
| "step": 78750 | |
| }, | |
| { | |
| "epoch": 18.19, | |
| "grad_norm": 0.020079199224710464, | |
| "learning_rate": 8.190684805164861e-06, | |
| "loss": 0.0015, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 18.19, | |
| "eval_loss": 0.13000167906284332, | |
| "eval_runtime": 861.6458, | |
| "eval_samples_per_second": 7.836, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 10.403237003391713, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 18.25, | |
| "grad_norm": 1.2580838203430176, | |
| "learning_rate": 8.184920451925294e-06, | |
| "loss": 0.0012, | |
| "step": 79250 | |
| }, | |
| { | |
| "epoch": 18.31, | |
| "grad_norm": 0.2000938206911087, | |
| "learning_rate": 8.179156098685728e-06, | |
| "loss": 0.0012, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 18.37, | |
| "grad_norm": 0.07198411971330643, | |
| "learning_rate": 8.173391745446162e-06, | |
| "loss": 0.0019, | |
| "step": 79750 | |
| }, | |
| { | |
| "epoch": 18.42, | |
| "grad_norm": 0.19537071883678436, | |
| "learning_rate": 8.167627392206595e-06, | |
| "loss": 0.0016, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 18.42, | |
| "eval_loss": 0.13053175806999207, | |
| "eval_runtime": 860.9809, | |
| "eval_samples_per_second": 7.842, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 10.30009718944007, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 18.48, | |
| "grad_norm": 0.3851453363895416, | |
| "learning_rate": 8.161863038967028e-06, | |
| "loss": 0.002, | |
| "step": 80250 | |
| }, | |
| { | |
| "epoch": 18.54, | |
| "grad_norm": 1.5883591175079346, | |
| "learning_rate": 8.156098685727461e-06, | |
| "loss": 0.0016, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 18.6, | |
| "grad_norm": 0.21738067269325256, | |
| "learning_rate": 8.150334332487896e-06, | |
| "loss": 0.0018, | |
| "step": 80750 | |
| }, | |
| { | |
| "epoch": 18.65, | |
| "grad_norm": 0.2855512201786041, | |
| "learning_rate": 8.14456997924833e-06, | |
| "loss": 0.0016, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 18.65, | |
| "eval_loss": 0.13212721049785614, | |
| "eval_runtime": 864.7213, | |
| "eval_samples_per_second": 7.808, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 10.468691116091794, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 18.71, | |
| "grad_norm": 0.42549794912338257, | |
| "learning_rate": 8.138805626008762e-06, | |
| "loss": 0.0018, | |
| "step": 81250 | |
| }, | |
| { | |
| "epoch": 18.77, | |
| "grad_norm": 0.1744888573884964, | |
| "learning_rate": 8.133064330182154e-06, | |
| "loss": 0.0017, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 18.83, | |
| "grad_norm": 0.4819861650466919, | |
| "learning_rate": 8.127299976942587e-06, | |
| "loss": 0.0018, | |
| "step": 81750 | |
| }, | |
| { | |
| "epoch": 18.89, | |
| "grad_norm": 0.4029196798801422, | |
| "learning_rate": 8.121535623703021e-06, | |
| "loss": 0.0017, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 18.89, | |
| "eval_loss": 0.1323300451040268, | |
| "eval_runtime": 863.1777, | |
| "eval_samples_per_second": 7.822, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 10.282246067794592, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 18.94, | |
| "grad_norm": 1.2428728342056274, | |
| "learning_rate": 8.115771270463455e-06, | |
| "loss": 0.0018, | |
| "step": 82250 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.11099331825971603, | |
| "learning_rate": 8.110006917223888e-06, | |
| "loss": 0.0015, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 19.06, | |
| "grad_norm": 0.02784816548228264, | |
| "learning_rate": 8.10424256398432e-06, | |
| "loss": 0.0011, | |
| "step": 82750 | |
| }, | |
| { | |
| "epoch": 19.12, | |
| "grad_norm": 0.06138940528035164, | |
| "learning_rate": 8.098478210744755e-06, | |
| "loss": 0.0012, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 19.12, | |
| "eval_loss": 0.13197582960128784, | |
| "eval_runtime": 855.8423, | |
| "eval_samples_per_second": 7.889, | |
| "eval_steps_per_second": 0.247, | |
| "eval_wer": 10.191007001606602, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 19.17, | |
| "grad_norm": 1.163352608680725, | |
| "learning_rate": 8.092713857505188e-06, | |
| "loss": 0.0011, | |
| "step": 83250 | |
| }, | |
| { | |
| "epoch": 19.23, | |
| "grad_norm": 0.19741587340831757, | |
| "learning_rate": 8.086949504265622e-06, | |
| "loss": 0.0011, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 19.29, | |
| "grad_norm": 0.019343126565217972, | |
| "learning_rate": 8.081185151026055e-06, | |
| "loss": 0.001, | |
| "step": 83750 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "grad_norm": 1.0601211786270142, | |
| "learning_rate": 8.075420797786488e-06, | |
| "loss": 0.0012, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "eval_loss": 0.13453201949596405, | |
| "eval_runtime": 855.5505, | |
| "eval_samples_per_second": 7.892, | |
| "eval_steps_per_second": 0.247, | |
| "eval_wer": 10.371501676021976, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 19.4, | |
| "grad_norm": 0.49607208371162415, | |
| "learning_rate": 8.069656444546922e-06, | |
| "loss": 0.0012, | |
| "step": 84250 | |
| }, | |
| { | |
| "epoch": 19.46, | |
| "grad_norm": 0.5492927432060242, | |
| "learning_rate": 8.063892091307355e-06, | |
| "loss": 0.0014, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 19.52, | |
| "grad_norm": 0.07558491080999374, | |
| "learning_rate": 8.058127738067788e-06, | |
| "loss": 0.0017, | |
| "step": 84750 | |
| }, | |
| { | |
| "epoch": 19.58, | |
| "grad_norm": 0.2147219181060791, | |
| "learning_rate": 8.052363384828223e-06, | |
| "loss": 0.0021, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 19.58, | |
| "eval_loss": 0.13526426255702972, | |
| "eval_runtime": 862.7391, | |
| "eval_samples_per_second": 7.826, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 10.405220461352322, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 19.63, | |
| "grad_norm": 0.6740741729736328, | |
| "learning_rate": 8.046599031588656e-06, | |
| "loss": 0.0013, | |
| "step": 85250 | |
| }, | |
| { | |
| "epoch": 19.69, | |
| "grad_norm": 0.10711955279111862, | |
| "learning_rate": 8.04083467834909e-06, | |
| "loss": 0.0018, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 19.75, | |
| "grad_norm": 0.09672527760267258, | |
| "learning_rate": 8.035070325109522e-06, | |
| "loss": 0.0013, | |
| "step": 85750 | |
| }, | |
| { | |
| "epoch": 19.81, | |
| "grad_norm": 2.067840814590454, | |
| "learning_rate": 8.029305971869957e-06, | |
| "loss": 0.0018, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 19.81, | |
| "eval_loss": 0.13280877470970154, | |
| "eval_runtime": 867.5864, | |
| "eval_samples_per_second": 7.783, | |
| "eval_steps_per_second": 0.243, | |
| "eval_wer": 10.284229525755203, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 19.86, | |
| "grad_norm": 1.6060312986373901, | |
| "learning_rate": 8.023587733456307e-06, | |
| "loss": 0.0019, | |
| "step": 86250 | |
| }, | |
| { | |
| "epoch": 19.92, | |
| "grad_norm": 1.6165310144424438, | |
| "learning_rate": 8.01782338021674e-06, | |
| "loss": 0.0017, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 19.98, | |
| "grad_norm": 0.09126902371644974, | |
| "learning_rate": 8.012059026977173e-06, | |
| "loss": 0.0017, | |
| "step": 86750 | |
| }, | |
| { | |
| "epoch": 20.04, | |
| "grad_norm": 0.06976249814033508, | |
| "learning_rate": 8.006294673737606e-06, | |
| "loss": 0.0016, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 20.04, | |
| "eval_loss": 0.13026085495948792, | |
| "eval_runtime": 862.5128, | |
| "eval_samples_per_second": 7.828, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 10.250510740424856, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 20.09, | |
| "grad_norm": 0.02732229419052601, | |
| "learning_rate": 8.00053032049804e-06, | |
| "loss": 0.0011, | |
| "step": 87250 | |
| }, | |
| { | |
| "epoch": 20.15, | |
| "grad_norm": 0.67994225025177, | |
| "learning_rate": 7.994765967258474e-06, | |
| "loss": 0.0014, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 20.21, | |
| "grad_norm": 0.10088305920362473, | |
| "learning_rate": 7.989001614018907e-06, | |
| "loss": 0.001, | |
| "step": 87750 | |
| }, | |
| { | |
| "epoch": 20.27, | |
| "grad_norm": 0.11000318080186844, | |
| "learning_rate": 7.98323726077934e-06, | |
| "loss": 0.001, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 20.27, | |
| "eval_loss": 0.12965452671051025, | |
| "eval_runtime": 856.312, | |
| "eval_samples_per_second": 7.885, | |
| "eval_steps_per_second": 0.246, | |
| "eval_wer": 10.175139337921733, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 20.32, | |
| "grad_norm": 0.09820275008678436, | |
| "learning_rate": 7.977472907539775e-06, | |
| "loss": 0.0014, | |
| "step": 88250 | |
| }, | |
| { | |
| "epoch": 20.38, | |
| "grad_norm": 0.023317014798521996, | |
| "learning_rate": 7.971708554300208e-06, | |
| "loss": 0.0014, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 20.44, | |
| "grad_norm": 0.012326917611062527, | |
| "learning_rate": 7.96594420106064e-06, | |
| "loss": 0.0013, | |
| "step": 88750 | |
| }, | |
| { | |
| "epoch": 20.5, | |
| "grad_norm": 0.0349048413336277, | |
| "learning_rate": 7.960179847821076e-06, | |
| "loss": 0.0011, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 20.5, | |
| "eval_loss": 0.13150520622730255, | |
| "eval_runtime": 865.4062, | |
| "eval_samples_per_second": 7.802, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 10.230676160818772, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 20.56, | |
| "grad_norm": 0.019593272358179092, | |
| "learning_rate": 7.954415494581509e-06, | |
| "loss": 0.0014, | |
| "step": 89250 | |
| }, | |
| { | |
| "epoch": 20.61, | |
| "grad_norm": 1.248373031616211, | |
| "learning_rate": 7.948651141341942e-06, | |
| "loss": 0.0016, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 20.67, | |
| "grad_norm": 0.050250094383955, | |
| "learning_rate": 7.942886788102375e-06, | |
| "loss": 0.0014, | |
| "step": 89750 | |
| }, | |
| { | |
| "epoch": 20.73, | |
| "grad_norm": 0.13364633917808533, | |
| "learning_rate": 7.93712243486281e-06, | |
| "loss": 0.0017, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 20.73, | |
| "eval_loss": 0.13066557049751282, | |
| "eval_runtime": 845.7111, | |
| "eval_samples_per_second": 7.984, | |
| "eval_steps_per_second": 0.249, | |
| "eval_wer": 10.333815974770415, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 20.79, | |
| "grad_norm": 1.2950758934020996, | |
| "learning_rate": 7.931358081623242e-06, | |
| "loss": 0.0014, | |
| "step": 90250 | |
| }, | |
| { | |
| "epoch": 20.84, | |
| "grad_norm": 0.03415424004197121, | |
| "learning_rate": 7.925593728383676e-06, | |
| "loss": 0.0014, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 20.9, | |
| "grad_norm": 0.10316485166549683, | |
| "learning_rate": 7.919852432557069e-06, | |
| "loss": 0.0016, | |
| "step": 90750 | |
| }, | |
| { | |
| "epoch": 20.96, | |
| "grad_norm": 0.19864577054977417, | |
| "learning_rate": 7.914088079317502e-06, | |
| "loss": 0.0013, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 20.96, | |
| "eval_loss": 0.12971824407577515, | |
| "eval_runtime": 865.4465, | |
| "eval_samples_per_second": 7.802, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 10.139437094630779, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 21.02, | |
| "grad_norm": 1.215699553489685, | |
| "learning_rate": 7.908323726077935e-06, | |
| "loss": 0.0012, | |
| "step": 91250 | |
| }, | |
| { | |
| "epoch": 21.07, | |
| "grad_norm": 0.38624367117881775, | |
| "learning_rate": 7.90255937283837e-06, | |
| "loss": 0.0013, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 21.13, | |
| "grad_norm": 0.0665733814239502, | |
| "learning_rate": 7.896795019598802e-06, | |
| "loss": 0.0013, | |
| "step": 91750 | |
| }, | |
| { | |
| "epoch": 21.19, | |
| "grad_norm": 1.9759693145751953, | |
| "learning_rate": 7.891030666359236e-06, | |
| "loss": 0.0013, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 21.19, | |
| "eval_loss": 0.1314748227596283, | |
| "eval_runtime": 846.5805, | |
| "eval_samples_per_second": 7.976, | |
| "eval_steps_per_second": 0.249, | |
| "eval_wer": 10.242576908582421, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 21.25, | |
| "grad_norm": 1.912919521331787, | |
| "learning_rate": 7.885266313119669e-06, | |
| "loss": 0.0014, | |
| "step": 92250 | |
| }, | |
| { | |
| "epoch": 21.3, | |
| "grad_norm": 2.836052417755127, | |
| "learning_rate": 7.879501959880103e-06, | |
| "loss": 0.0015, | |
| "step": 92500 | |
| }, | |
| { | |
| "epoch": 21.36, | |
| "grad_norm": 0.8251251578330994, | |
| "learning_rate": 7.873737606640536e-06, | |
| "loss": 0.0013, | |
| "step": 92750 | |
| }, | |
| { | |
| "epoch": 21.42, | |
| "grad_norm": 0.017601924017071724, | |
| "learning_rate": 7.86797325340097e-06, | |
| "loss": 0.0013, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 21.42, | |
| "eval_loss": 0.1311779022216797, | |
| "eval_runtime": 857.0196, | |
| "eval_samples_per_second": 7.878, | |
| "eval_steps_per_second": 0.246, | |
| "eval_wer": 10.206874665291469, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 21.48, | |
| "grad_norm": 0.2548398971557617, | |
| "learning_rate": 7.86223195757436e-06, | |
| "loss": 0.0013, | |
| "step": 93250 | |
| }, | |
| { | |
| "epoch": 21.53, | |
| "grad_norm": 0.050698913633823395, | |
| "learning_rate": 7.856467604334794e-06, | |
| "loss": 0.0012, | |
| "step": 93500 | |
| }, | |
| { | |
| "epoch": 21.59, | |
| "grad_norm": 0.15648888051509857, | |
| "learning_rate": 7.850703251095227e-06, | |
| "loss": 0.0009, | |
| "step": 93750 | |
| }, | |
| { | |
| "epoch": 21.65, | |
| "grad_norm": 0.6790530681610107, | |
| "learning_rate": 7.844938897855662e-06, | |
| "loss": 0.0016, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 21.65, | |
| "eval_loss": 0.13513444364070892, | |
| "eval_runtime": 855.2346, | |
| "eval_samples_per_second": 7.895, | |
| "eval_steps_per_second": 0.247, | |
| "eval_wer": 10.276295693912768, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 21.71, | |
| "grad_norm": 0.8243337869644165, | |
| "learning_rate": 7.839197602029053e-06, | |
| "loss": 0.0013, | |
| "step": 94250 | |
| }, | |
| { | |
| "epoch": 21.76, | |
| "grad_norm": 0.05715777724981308, | |
| "learning_rate": 7.833433248789486e-06, | |
| "loss": 0.0015, | |
| "step": 94500 | |
| }, | |
| { | |
| "epoch": 21.82, | |
| "grad_norm": 0.10455400496721268, | |
| "learning_rate": 7.827668895549921e-06, | |
| "loss": 0.0018, | |
| "step": 94750 | |
| }, | |
| { | |
| "epoch": 21.88, | |
| "grad_norm": 0.06954806298017502, | |
| "learning_rate": 7.821904542310354e-06, | |
| "loss": 0.0012, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 21.88, | |
| "eval_loss": 0.13230295479297638, | |
| "eval_runtime": 864.2691, | |
| "eval_samples_per_second": 7.812, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 9.964892794097228, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "grad_norm": 0.35232919454574585, | |
| "learning_rate": 7.816140189070787e-06, | |
| "loss": 0.0016, | |
| "step": 95250 | |
| }, | |
| { | |
| "epoch": 21.99, | |
| "grad_norm": 0.27226054668426514, | |
| "learning_rate": 7.81037583583122e-06, | |
| "loss": 0.0015, | |
| "step": 95500 | |
| }, | |
| { | |
| "epoch": 22.05, | |
| "grad_norm": 0.03192569315433502, | |
| "learning_rate": 7.804611482591655e-06, | |
| "loss": 0.0009, | |
| "step": 95750 | |
| }, | |
| { | |
| "epoch": 22.11, | |
| "grad_norm": 0.772456705570221, | |
| "learning_rate": 7.798847129352088e-06, | |
| "loss": 0.0009, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 22.11, | |
| "eval_loss": 0.13247309625148773, | |
| "eval_runtime": 846.1981, | |
| "eval_samples_per_second": 7.979, | |
| "eval_steps_per_second": 0.249, | |
| "eval_wer": 10.012495785151835, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 22.17, | |
| "grad_norm": 0.05217467620968819, | |
| "learning_rate": 7.793082776112521e-06, | |
| "loss": 0.0009, | |
| "step": 96250 | |
| }, | |
| { | |
| "epoch": 22.22, | |
| "grad_norm": 0.03777674213051796, | |
| "learning_rate": 7.787318422872956e-06, | |
| "loss": 0.0012, | |
| "step": 96500 | |
| }, | |
| { | |
| "epoch": 22.28, | |
| "grad_norm": 0.20706115663051605, | |
| "learning_rate": 7.781554069633389e-06, | |
| "loss": 0.0011, | |
| "step": 96750 | |
| }, | |
| { | |
| "epoch": 22.34, | |
| "grad_norm": 0.07679043710231781, | |
| "learning_rate": 7.775789716393822e-06, | |
| "loss": 0.0008, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 22.34, | |
| "eval_loss": 0.13062846660614014, | |
| "eval_runtime": 859.6538, | |
| "eval_samples_per_second": 7.854, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 9.780431203760637, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 22.4, | |
| "grad_norm": 0.2341454029083252, | |
| "learning_rate": 7.770025363154255e-06, | |
| "loss": 0.001, | |
| "step": 97250 | |
| }, | |
| { | |
| "epoch": 22.46, | |
| "grad_norm": 0.5662194490432739, | |
| "learning_rate": 7.764284067327646e-06, | |
| "loss": 0.0013, | |
| "step": 97500 | |
| }, | |
| { | |
| "epoch": 22.51, | |
| "grad_norm": 0.08940891921520233, | |
| "learning_rate": 7.75851971408808e-06, | |
| "loss": 0.0012, | |
| "step": 97750 | |
| }, | |
| { | |
| "epoch": 22.57, | |
| "grad_norm": 0.28177976608276367, | |
| "learning_rate": 7.752755360848512e-06, | |
| "loss": 0.0015, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 22.57, | |
| "eval_loss": 0.13175997138023376, | |
| "eval_runtime": 865.6294, | |
| "eval_samples_per_second": 7.8, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 10.066049150088265, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 22.63, | |
| "grad_norm": 0.4019626975059509, | |
| "learning_rate": 7.746991007608947e-06, | |
| "loss": 0.0012, | |
| "step": 98250 | |
| }, | |
| { | |
| "epoch": 22.69, | |
| "grad_norm": 0.06278575956821442, | |
| "learning_rate": 7.74122665436938e-06, | |
| "loss": 0.0019, | |
| "step": 98500 | |
| }, | |
| { | |
| "epoch": 22.74, | |
| "grad_norm": 2.755483388900757, | |
| "learning_rate": 7.735462301129813e-06, | |
| "loss": 0.0013, | |
| "step": 98750 | |
| }, | |
| { | |
| "epoch": 22.8, | |
| "grad_norm": 0.478965163230896, | |
| "learning_rate": 7.729697947890248e-06, | |
| "loss": 0.0012, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 22.8, | |
| "eval_loss": 0.13538804650306702, | |
| "eval_runtime": 848.5461, | |
| "eval_samples_per_second": 7.957, | |
| "eval_steps_per_second": 0.249, | |
| "eval_wer": 9.996628121466966, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 22.86, | |
| "grad_norm": 0.18010005354881287, | |
| "learning_rate": 7.723933594650681e-06, | |
| "loss": 0.0014, | |
| "step": 99250 | |
| }, | |
| { | |
| "epoch": 22.92, | |
| "grad_norm": 0.32389816641807556, | |
| "learning_rate": 7.718169241411114e-06, | |
| "loss": 0.0011, | |
| "step": 99500 | |
| }, | |
| { | |
| "epoch": 22.97, | |
| "grad_norm": 0.031109152361750603, | |
| "learning_rate": 7.712427945584507e-06, | |
| "loss": 0.0013, | |
| "step": 99750 | |
| }, | |
| { | |
| "epoch": 23.03, | |
| "grad_norm": 0.40925395488739014, | |
| "learning_rate": 7.70666359234494e-06, | |
| "loss": 0.0013, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 23.03, | |
| "eval_loss": 0.13106635212898254, | |
| "eval_runtime": 855.4169, | |
| "eval_samples_per_second": 7.893, | |
| "eval_steps_per_second": 0.247, | |
| "eval_wer": 9.673324473887776, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 23.09, | |
| "grad_norm": 2.2084646224975586, | |
| "learning_rate": 7.700922296518331e-06, | |
| "loss": 0.0015, | |
| "step": 100250 | |
| }, | |
| { | |
| "epoch": 23.15, | |
| "grad_norm": 0.03353571146726608, | |
| "learning_rate": 7.695157943278765e-06, | |
| "loss": 0.0007, | |
| "step": 100500 | |
| }, | |
| { | |
| "epoch": 23.2, | |
| "grad_norm": 0.27213382720947266, | |
| "learning_rate": 7.689393590039198e-06, | |
| "loss": 0.0008, | |
| "step": 100750 | |
| }, | |
| { | |
| "epoch": 23.26, | |
| "grad_norm": 1.3013813495635986, | |
| "learning_rate": 7.68362923679963e-06, | |
| "loss": 0.001, | |
| "step": 101000 | |
| }, | |
| { | |
| "epoch": 23.26, | |
| "eval_loss": 0.13327930867671967, | |
| "eval_runtime": 860.1438, | |
| "eval_samples_per_second": 7.85, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 10.161255132197473, | |
| "step": 101000 | |
| }, | |
| { | |
| "epoch": 23.32, | |
| "grad_norm": 1.0605800151824951, | |
| "learning_rate": 7.677864883560065e-06, | |
| "loss": 0.0016, | |
| "step": 101250 | |
| }, | |
| { | |
| "epoch": 23.38, | |
| "grad_norm": 1.6323225498199463, | |
| "learning_rate": 7.672100530320498e-06, | |
| "loss": 0.0016, | |
| "step": 101500 | |
| }, | |
| { | |
| "epoch": 23.43, | |
| "grad_norm": 0.3052765130996704, | |
| "learning_rate": 7.666336177080931e-06, | |
| "loss": 0.001, | |
| "step": 101750 | |
| }, | |
| { | |
| "epoch": 23.49, | |
| "grad_norm": 1.3335562944412231, | |
| "learning_rate": 7.660571823841365e-06, | |
| "loss": 0.0013, | |
| "step": 102000 | |
| }, | |
| { | |
| "epoch": 23.49, | |
| "eval_loss": 0.1332116574048996, | |
| "eval_runtime": 862.515, | |
| "eval_samples_per_second": 7.828, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 10.036297280679136, | |
| "step": 102000 | |
| }, | |
| { | |
| "epoch": 23.55, | |
| "grad_norm": 1.166803240776062, | |
| "learning_rate": 7.6548074706018e-06, | |
| "loss": 0.0014, | |
| "step": 102250 | |
| }, | |
| { | |
| "epoch": 23.61, | |
| "grad_norm": 0.06531043350696564, | |
| "learning_rate": 7.649043117362232e-06, | |
| "loss": 0.0013, | |
| "step": 102500 | |
| }, | |
| { | |
| "epoch": 23.66, | |
| "grad_norm": 0.2673274874687195, | |
| "learning_rate": 7.643278764122665e-06, | |
| "loss": 0.0013, | |
| "step": 102750 | |
| }, | |
| { | |
| "epoch": 23.72, | |
| "grad_norm": 0.007462978828698397, | |
| "learning_rate": 7.6375144108831e-06, | |
| "loss": 0.0013, | |
| "step": 103000 | |
| }, | |
| { | |
| "epoch": 23.72, | |
| "eval_loss": 0.13317851722240448, | |
| "eval_runtime": 854.6131, | |
| "eval_samples_per_second": 7.901, | |
| "eval_steps_per_second": 0.247, | |
| "eval_wer": 10.012495785151835, | |
| "step": 103000 | |
| }, | |
| { | |
| "epoch": 23.78, | |
| "grad_norm": 1.090072512626648, | |
| "learning_rate": 7.631750057643533e-06, | |
| "loss": 0.0016, | |
| "step": 103250 | |
| }, | |
| { | |
| "epoch": 23.84, | |
| "grad_norm": 0.10874152928590775, | |
| "learning_rate": 7.625985704403966e-06, | |
| "loss": 0.0014, | |
| "step": 103500 | |
| }, | |
| { | |
| "epoch": 23.89, | |
| "grad_norm": 0.3460334837436676, | |
| "learning_rate": 7.620221351164399e-06, | |
| "loss": 0.0012, | |
| "step": 103750 | |
| }, | |
| { | |
| "epoch": 23.95, | |
| "grad_norm": 0.0227680541574955, | |
| "learning_rate": 7.614456997924834e-06, | |
| "loss": 0.0012, | |
| "step": 104000 | |
| }, | |
| { | |
| "epoch": 23.95, | |
| "eval_loss": 0.13361206650733948, | |
| "eval_runtime": 864.3568, | |
| "eval_samples_per_second": 7.812, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 10.058115318245829, | |
| "step": 104000 | |
| }, | |
| { | |
| "epoch": 24.01, | |
| "grad_norm": 1.1514167785644531, | |
| "learning_rate": 7.608692644685267e-06, | |
| "loss": 0.0015, | |
| "step": 104250 | |
| }, | |
| { | |
| "epoch": 24.07, | |
| "grad_norm": 0.04206667095422745, | |
| "learning_rate": 7.6029282914457e-06, | |
| "loss": 0.0009, | |
| "step": 104500 | |
| }, | |
| { | |
| "epoch": 24.12, | |
| "grad_norm": 0.03346676006913185, | |
| "learning_rate": 7.597163938206133e-06, | |
| "loss": 0.0007, | |
| "step": 104750 | |
| }, | |
| { | |
| "epoch": 24.18, | |
| "grad_norm": 0.005905906204134226, | |
| "learning_rate": 7.591399584966568e-06, | |
| "loss": 0.0009, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 24.18, | |
| "eval_loss": 0.13135230541229248, | |
| "eval_runtime": 859.844, | |
| "eval_samples_per_second": 7.853, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 9.89745522343654, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 24.24, | |
| "grad_norm": 0.48378223180770874, | |
| "learning_rate": 7.585635231727001e-06, | |
| "loss": 0.001, | |
| "step": 105250 | |
| }, | |
| { | |
| "epoch": 24.3, | |
| "grad_norm": 0.013876562006771564, | |
| "learning_rate": 7.579870878487434e-06, | |
| "loss": 0.0009, | |
| "step": 105500 | |
| }, | |
| { | |
| "epoch": 24.36, | |
| "grad_norm": 0.7280595898628235, | |
| "learning_rate": 7.574106525247867e-06, | |
| "loss": 0.0013, | |
| "step": 105750 | |
| }, | |
| { | |
| "epoch": 24.41, | |
| "grad_norm": 1.0038032531738281, | |
| "learning_rate": 7.568342172008302e-06, | |
| "loss": 0.0012, | |
| "step": 106000 | |
| }, | |
| { | |
| "epoch": 24.41, | |
| "eval_loss": 0.13419808447360992, | |
| "eval_runtime": 868.3234, | |
| "eval_samples_per_second": 7.776, | |
| "eval_steps_per_second": 0.243, | |
| "eval_wer": 10.222742328976338, | |
| "step": 106000 | |
| }, | |
| { | |
| "epoch": 24.47, | |
| "grad_norm": 1.810659646987915, | |
| "learning_rate": 7.562600876181692e-06, | |
| "loss": 0.0015, | |
| "step": 106250 | |
| }, | |
| { | |
| "epoch": 24.53, | |
| "grad_norm": 0.027089903131127357, | |
| "learning_rate": 7.556836522942127e-06, | |
| "loss": 0.0014, | |
| "step": 106500 | |
| }, | |
| { | |
| "epoch": 24.59, | |
| "grad_norm": 0.1833794265985489, | |
| "learning_rate": 7.55107216970256e-06, | |
| "loss": 0.001, | |
| "step": 106750 | |
| }, | |
| { | |
| "epoch": 24.64, | |
| "grad_norm": 0.4848494529724121, | |
| "learning_rate": 7.5453308738759516e-06, | |
| "loss": 0.0011, | |
| "step": 107000 | |
| }, | |
| { | |
| "epoch": 24.64, | |
| "eval_loss": 0.134404718875885, | |
| "eval_runtime": 871.0128, | |
| "eval_samples_per_second": 7.752, | |
| "eval_steps_per_second": 0.242, | |
| "eval_wer": 9.812166531130373, | |
| "step": 107000 | |
| }, | |
| { | |
| "epoch": 24.7, | |
| "grad_norm": 0.03687746450304985, | |
| "learning_rate": 7.5395665206363855e-06, | |
| "loss": 0.0011, | |
| "step": 107250 | |
| }, | |
| { | |
| "epoch": 24.76, | |
| "grad_norm": 1.3746637105941772, | |
| "learning_rate": 7.5338021673968185e-06, | |
| "loss": 0.0015, | |
| "step": 107500 | |
| }, | |
| { | |
| "epoch": 24.82, | |
| "grad_norm": 0.5606046319007874, | |
| "learning_rate": 7.528037814157252e-06, | |
| "loss": 0.0016, | |
| "step": 107750 | |
| }, | |
| { | |
| "epoch": 24.87, | |
| "grad_norm": 0.23363818228244781, | |
| "learning_rate": 7.522273460917686e-06, | |
| "loss": 0.0011, | |
| "step": 108000 | |
| }, | |
| { | |
| "epoch": 24.87, | |
| "eval_loss": 0.131892129778862, | |
| "eval_runtime": 866.9375, | |
| "eval_samples_per_second": 7.788, | |
| "eval_steps_per_second": 0.243, | |
| "eval_wer": 9.820100362972807, | |
| "step": 108000 | |
| }, | |
| { | |
| "epoch": 24.93, | |
| "grad_norm": 0.14785051345825195, | |
| "learning_rate": 7.516509107678119e-06, | |
| "loss": 0.0012, | |
| "step": 108250 | |
| }, | |
| { | |
| "epoch": 24.99, | |
| "grad_norm": 0.17186763882637024, | |
| "learning_rate": 7.5107447544385524e-06, | |
| "loss": 0.001, | |
| "step": 108500 | |
| }, | |
| { | |
| "epoch": 25.05, | |
| "grad_norm": 0.7404451370239258, | |
| "learning_rate": 7.5049804011989855e-06, | |
| "loss": 0.0012, | |
| "step": 108750 | |
| }, | |
| { | |
| "epoch": 25.1, | |
| "grad_norm": 0.11903788894414902, | |
| "learning_rate": 7.49921604795942e-06, | |
| "loss": 0.0008, | |
| "step": 109000 | |
| }, | |
| { | |
| "epoch": 25.1, | |
| "eval_loss": 0.1291409283876419, | |
| "eval_runtime": 862.8202, | |
| "eval_samples_per_second": 7.826, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 9.744728960469683, | |
| "step": 109000 | |
| }, | |
| { | |
| "epoch": 25.16, | |
| "grad_norm": 0.028790809214115143, | |
| "learning_rate": 7.493451694719853e-06, | |
| "loss": 0.0008, | |
| "step": 109250 | |
| }, | |
| { | |
| "epoch": 25.22, | |
| "grad_norm": 0.08509263396263123, | |
| "learning_rate": 7.487687341480286e-06, | |
| "loss": 0.0009, | |
| "step": 109500 | |
| }, | |
| { | |
| "epoch": 25.28, | |
| "grad_norm": 3.7736191749572754, | |
| "learning_rate": 7.481922988240719e-06, | |
| "loss": 0.001, | |
| "step": 109750 | |
| }, | |
| { | |
| "epoch": 25.33, | |
| "grad_norm": 1.2739046812057495, | |
| "learning_rate": 7.476158635001154e-06, | |
| "loss": 0.0011, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 25.33, | |
| "eval_loss": 0.12932626903057098, | |
| "eval_runtime": 872.4168, | |
| "eval_samples_per_second": 7.739, | |
| "eval_steps_per_second": 0.242, | |
| "eval_wer": 9.643572604478647, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 25.39, | |
| "grad_norm": 1.4399291276931763, | |
| "learning_rate": 7.470394281761587e-06, | |
| "loss": 0.0009, | |
| "step": 110250 | |
| }, | |
| { | |
| "epoch": 25.45, | |
| "grad_norm": 0.06352319568395615, | |
| "learning_rate": 7.46462992852202e-06, | |
| "loss": 0.0011, | |
| "step": 110500 | |
| }, | |
| { | |
| "epoch": 25.51, | |
| "grad_norm": 0.0745418593287468, | |
| "learning_rate": 7.458865575282453e-06, | |
| "loss": 0.0008, | |
| "step": 110750 | |
| }, | |
| { | |
| "epoch": 25.56, | |
| "grad_norm": 2.0110867023468018, | |
| "learning_rate": 7.453101222042888e-06, | |
| "loss": 0.0015, | |
| "step": 111000 | |
| }, | |
| { | |
| "epoch": 25.56, | |
| "eval_loss": 0.1350392997264862, | |
| "eval_runtime": 859.6078, | |
| "eval_samples_per_second": 7.855, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 9.728861296784814, | |
| "step": 111000 | |
| }, | |
| { | |
| "epoch": 25.62, | |
| "grad_norm": 1.9989393949508667, | |
| "learning_rate": 7.447359926216279e-06, | |
| "loss": 0.0012, | |
| "step": 111250 | |
| }, | |
| { | |
| "epoch": 25.68, | |
| "grad_norm": 0.7481454610824585, | |
| "learning_rate": 7.441595572976713e-06, | |
| "loss": 0.0012, | |
| "step": 111500 | |
| }, | |
| { | |
| "epoch": 25.74, | |
| "grad_norm": 0.42684781551361084, | |
| "learning_rate": 7.435854277150104e-06, | |
| "loss": 0.0013, | |
| "step": 111750 | |
| }, | |
| { | |
| "epoch": 25.79, | |
| "grad_norm": 0.21273395419120789, | |
| "learning_rate": 7.430089923910537e-06, | |
| "loss": 0.0009, | |
| "step": 112000 | |
| }, | |
| { | |
| "epoch": 25.79, | |
| "eval_loss": 0.13176259398460388, | |
| "eval_runtime": 853.5624, | |
| "eval_samples_per_second": 7.91, | |
| "eval_steps_per_second": 0.247, | |
| "eval_wer": 9.653489894281691, | |
| "step": 112000 | |
| }, | |
| { | |
| "epoch": 25.85, | |
| "grad_norm": 0.798125147819519, | |
| "learning_rate": 7.424325570670972e-06, | |
| "loss": 0.0011, | |
| "step": 112250 | |
| }, | |
| { | |
| "epoch": 25.91, | |
| "grad_norm": 0.3447425663471222, | |
| "learning_rate": 7.418561217431405e-06, | |
| "loss": 0.0012, | |
| "step": 112500 | |
| }, | |
| { | |
| "epoch": 25.97, | |
| "grad_norm": 1.2573052644729614, | |
| "learning_rate": 7.412819921604797e-06, | |
| "loss": 0.0009, | |
| "step": 112750 | |
| }, | |
| { | |
| "epoch": 26.02, | |
| "grad_norm": 0.2099132090806961, | |
| "learning_rate": 7.40705556836523e-06, | |
| "loss": 0.001, | |
| "step": 113000 | |
| }, | |
| { | |
| "epoch": 26.02, | |
| "eval_loss": 0.1325974017381668, | |
| "eval_runtime": 859.4858, | |
| "eval_samples_per_second": 7.856, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 9.685225221651427, | |
| "step": 113000 | |
| }, | |
| { | |
| "epoch": 26.08, | |
| "grad_norm": 0.06933692842721939, | |
| "learning_rate": 7.401291215125663e-06, | |
| "loss": 0.0007, | |
| "step": 113250 | |
| }, | |
| { | |
| "epoch": 26.14, | |
| "grad_norm": 0.12258395552635193, | |
| "learning_rate": 7.395526861886096e-06, | |
| "loss": 0.0008, | |
| "step": 113500 | |
| }, | |
| { | |
| "epoch": 26.2, | |
| "grad_norm": 0.2960332930088043, | |
| "learning_rate": 7.389762508646531e-06, | |
| "loss": 0.0008, | |
| "step": 113750 | |
| }, | |
| { | |
| "epoch": 26.26, | |
| "grad_norm": 0.03250851482152939, | |
| "learning_rate": 7.383998155406964e-06, | |
| "loss": 0.0006, | |
| "step": 114000 | |
| }, | |
| { | |
| "epoch": 26.26, | |
| "eval_loss": 0.1351398229598999, | |
| "eval_runtime": 869.8853, | |
| "eval_samples_per_second": 7.762, | |
| "eval_steps_per_second": 0.243, | |
| "eval_wer": 9.845885316460718, | |
| "step": 114000 | |
| }, | |
| { | |
| "epoch": 26.31, | |
| "grad_norm": 0.2106870859861374, | |
| "learning_rate": 7.378233802167397e-06, | |
| "loss": 0.0013, | |
| "step": 114250 | |
| }, | |
| { | |
| "epoch": 26.37, | |
| "grad_norm": 0.6854831576347351, | |
| "learning_rate": 7.372469448927832e-06, | |
| "loss": 0.0012, | |
| "step": 114500 | |
| }, | |
| { | |
| "epoch": 26.43, | |
| "grad_norm": 1.116827368736267, | |
| "learning_rate": 7.366705095688265e-06, | |
| "loss": 0.0009, | |
| "step": 114750 | |
| }, | |
| { | |
| "epoch": 26.49, | |
| "grad_norm": 0.060012660920619965, | |
| "learning_rate": 7.360940742448698e-06, | |
| "loss": 0.0011, | |
| "step": 115000 | |
| }, | |
| { | |
| "epoch": 26.49, | |
| "eval_loss": 0.13296382129192352, | |
| "eval_runtime": 858.1234, | |
| "eval_samples_per_second": 7.868, | |
| "eval_steps_per_second": 0.246, | |
| "eval_wer": 9.891504849554714, | |
| "step": 115000 | |
| }, | |
| { | |
| "epoch": 26.54, | |
| "grad_norm": 0.2378009855747223, | |
| "learning_rate": 7.355176389209131e-06, | |
| "loss": 0.0011, | |
| "step": 115250 | |
| }, | |
| { | |
| "epoch": 26.6, | |
| "grad_norm": 0.07352372258901596, | |
| "learning_rate": 7.349412035969566e-06, | |
| "loss": 0.0015, | |
| "step": 115500 | |
| }, | |
| { | |
| "epoch": 26.66, | |
| "grad_norm": 0.9573454856872559, | |
| "learning_rate": 7.343647682729999e-06, | |
| "loss": 0.0009, | |
| "step": 115750 | |
| }, | |
| { | |
| "epoch": 26.72, | |
| "grad_norm": 2.2814531326293945, | |
| "learning_rate": 7.337883329490432e-06, | |
| "loss": 0.0013, | |
| "step": 116000 | |
| }, | |
| { | |
| "epoch": 26.72, | |
| "eval_loss": 0.13392788171768188, | |
| "eval_runtime": 863.6577, | |
| "eval_samples_per_second": 7.818, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 10.036297280679136, | |
| "step": 116000 | |
| }, | |
| { | |
| "epoch": 26.77, | |
| "grad_norm": 0.2348572313785553, | |
| "learning_rate": 7.332118976250865e-06, | |
| "loss": 0.001, | |
| "step": 116250 | |
| }, | |
| { | |
| "epoch": 26.83, | |
| "grad_norm": 0.3893941640853882, | |
| "learning_rate": 7.3263546230112995e-06, | |
| "loss": 0.0011, | |
| "step": 116500 | |
| }, | |
| { | |
| "epoch": 26.89, | |
| "grad_norm": 1.920695424079895, | |
| "learning_rate": 7.320590269771733e-06, | |
| "loss": 0.0012, | |
| "step": 116750 | |
| }, | |
| { | |
| "epoch": 26.95, | |
| "grad_norm": 0.17352449893951416, | |
| "learning_rate": 7.314825916532166e-06, | |
| "loss": 0.0011, | |
| "step": 117000 | |
| }, | |
| { | |
| "epoch": 26.95, | |
| "eval_loss": 0.13272777199745178, | |
| "eval_runtime": 857.9116, | |
| "eval_samples_per_second": 7.87, | |
| "eval_steps_per_second": 0.246, | |
| "eval_wer": 10.083900271733741, | |
| "step": 117000 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "grad_norm": 0.03900158777832985, | |
| "learning_rate": 7.309061563292599e-06, | |
| "loss": 0.0012, | |
| "step": 117250 | |
| }, | |
| { | |
| "epoch": 27.06, | |
| "grad_norm": 0.03566018119454384, | |
| "learning_rate": 7.3032972100530334e-06, | |
| "loss": 0.0006, | |
| "step": 117500 | |
| }, | |
| { | |
| "epoch": 27.12, | |
| "grad_norm": 0.057484470307826996, | |
| "learning_rate": 7.2975328568134665e-06, | |
| "loss": 0.0006, | |
| "step": 117750 | |
| }, | |
| { | |
| "epoch": 27.18, | |
| "grad_norm": 0.03549089655280113, | |
| "learning_rate": 7.2917685035738995e-06, | |
| "loss": 0.0008, | |
| "step": 118000 | |
| }, | |
| { | |
| "epoch": 27.18, | |
| "eval_loss": 0.13313508033752441, | |
| "eval_runtime": 854.7837, | |
| "eval_samples_per_second": 7.899, | |
| "eval_steps_per_second": 0.247, | |
| "eval_wer": 9.935140924688103, | |
| "step": 118000 | |
| }, | |
| { | |
| "epoch": 27.23, | |
| "grad_norm": 0.009218300692737103, | |
| "learning_rate": 7.286004150334333e-06, | |
| "loss": 0.0008, | |
| "step": 118250 | |
| }, | |
| { | |
| "epoch": 27.29, | |
| "grad_norm": 0.0295584537088871, | |
| "learning_rate": 7.2802397970947665e-06, | |
| "loss": 0.0008, | |
| "step": 118500 | |
| }, | |
| { | |
| "epoch": 27.35, | |
| "grad_norm": 0.41446352005004883, | |
| "learning_rate": 7.2744754438552e-06, | |
| "loss": 0.0009, | |
| "step": 118750 | |
| }, | |
| { | |
| "epoch": 27.41, | |
| "grad_norm": 0.4680302143096924, | |
| "learning_rate": 7.2687110906156335e-06, | |
| "loss": 0.0009, | |
| "step": 119000 | |
| }, | |
| { | |
| "epoch": 27.41, | |
| "eval_loss": 0.1315101832151413, | |
| "eval_runtime": 865.9401, | |
| "eval_samples_per_second": 7.797, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 9.919273261003234, | |
| "step": 119000 | |
| }, | |
| { | |
| "epoch": 27.46, | |
| "grad_norm": 4.383372783660889, | |
| "learning_rate": 7.2629467373760665e-06, | |
| "loss": 0.0009, | |
| "step": 119250 | |
| }, | |
| { | |
| "epoch": 27.52, | |
| "grad_norm": 0.050101153552532196, | |
| "learning_rate": 7.2571823841365e-06, | |
| "loss": 0.0013, | |
| "step": 119500 | |
| }, | |
| { | |
| "epoch": 27.58, | |
| "grad_norm": 0.22907772660255432, | |
| "learning_rate": 7.251418030896934e-06, | |
| "loss": 0.0011, | |
| "step": 119750 | |
| }, | |
| { | |
| "epoch": 27.64, | |
| "grad_norm": 0.2788342237472534, | |
| "learning_rate": 7.245653677657367e-06, | |
| "loss": 0.0014, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 27.64, | |
| "eval_loss": 0.1343310922384262, | |
| "eval_runtime": 855.2599, | |
| "eval_samples_per_second": 7.895, | |
| "eval_steps_per_second": 0.247, | |
| "eval_wer": 9.832001110736458, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 27.69, | |
| "grad_norm": 0.027169860899448395, | |
| "learning_rate": 7.239889324417801e-06, | |
| "loss": 0.0011, | |
| "step": 120250 | |
| }, | |
| { | |
| "epoch": 27.75, | |
| "grad_norm": 0.3283644914627075, | |
| "learning_rate": 7.234124971178234e-06, | |
| "loss": 0.001, | |
| "step": 120500 | |
| }, | |
| { | |
| "epoch": 27.81, | |
| "grad_norm": 0.9259850382804871, | |
| "learning_rate": 7.228383675351626e-06, | |
| "loss": 0.0008, | |
| "step": 120750 | |
| }, | |
| { | |
| "epoch": 27.87, | |
| "grad_norm": 0.20162805914878845, | |
| "learning_rate": 7.22261932211206e-06, | |
| "loss": 0.0013, | |
| "step": 121000 | |
| }, | |
| { | |
| "epoch": 27.87, | |
| "eval_loss": 0.1318797767162323, | |
| "eval_runtime": 854.2094, | |
| "eval_samples_per_second": 7.904, | |
| "eval_steps_per_second": 0.247, | |
| "eval_wer": 9.94902513041236, | |
| "step": 121000 | |
| }, | |
| { | |
| "epoch": 27.92, | |
| "grad_norm": 0.029574675485491753, | |
| "learning_rate": 7.2168549688724935e-06, | |
| "loss": 0.0008, | |
| "step": 121250 | |
| }, | |
| { | |
| "epoch": 27.98, | |
| "grad_norm": 0.36377763748168945, | |
| "learning_rate": 7.2110906156329266e-06, | |
| "loss": 0.0007, | |
| "step": 121500 | |
| }, | |
| { | |
| "epoch": 28.04, | |
| "grad_norm": 0.15056151151657104, | |
| "learning_rate": 7.20532626239336e-06, | |
| "loss": 0.0008, | |
| "step": 121750 | |
| }, | |
| { | |
| "epoch": 28.1, | |
| "grad_norm": 0.8695113658905029, | |
| "learning_rate": 7.1995619091537935e-06, | |
| "loss": 0.0006, | |
| "step": 122000 | |
| }, | |
| { | |
| "epoch": 28.1, | |
| "eval_loss": 0.135031595826149, | |
| "eval_runtime": 850.3861, | |
| "eval_samples_per_second": 7.94, | |
| "eval_steps_per_second": 0.248, | |
| "eval_wer": 9.851835690342543, | |
| "step": 122000 | |
| }, | |
| { | |
| "epoch": 28.16, | |
| "grad_norm": 0.009558724239468575, | |
| "learning_rate": 7.193797555914227e-06, | |
| "loss": 0.0008, | |
| "step": 122250 | |
| }, | |
| { | |
| "epoch": 28.21, | |
| "grad_norm": 0.4410775303840637, | |
| "learning_rate": 7.1880332026746605e-06, | |
| "loss": 0.0008, | |
| "step": 122500 | |
| }, | |
| { | |
| "epoch": 28.27, | |
| "grad_norm": 0.3716282546520233, | |
| "learning_rate": 7.182268849435094e-06, | |
| "loss": 0.0007, | |
| "step": 122750 | |
| }, | |
| { | |
| "epoch": 28.33, | |
| "grad_norm": 0.018515612930059433, | |
| "learning_rate": 7.176527553608486e-06, | |
| "loss": 0.001, | |
| "step": 123000 | |
| }, | |
| { | |
| "epoch": 28.33, | |
| "eval_loss": 0.13689374923706055, | |
| "eval_runtime": 864.0622, | |
| "eval_samples_per_second": 7.814, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 9.968859710018446, | |
| "step": 123000 | |
| }, | |
| { | |
| "epoch": 28.39, | |
| "grad_norm": 0.07072445005178452, | |
| "learning_rate": 7.170763200368919e-06, | |
| "loss": 0.0011, | |
| "step": 123250 | |
| }, | |
| { | |
| "epoch": 28.44, | |
| "grad_norm": 0.7424390316009521, | |
| "learning_rate": 7.164998847129353e-06, | |
| "loss": 0.0013, | |
| "step": 123500 | |
| }, | |
| { | |
| "epoch": 28.5, | |
| "grad_norm": 0.031032560393214226, | |
| "learning_rate": 7.159234493889787e-06, | |
| "loss": 0.001, | |
| "step": 123750 | |
| }, | |
| { | |
| "epoch": 28.56, | |
| "grad_norm": 1.3974592685699463, | |
| "learning_rate": 7.15347014065022e-06, | |
| "loss": 0.001, | |
| "step": 124000 | |
| }, | |
| { | |
| "epoch": 28.56, | |
| "eval_loss": 0.13558751344680786, | |
| "eval_runtime": 859.3374, | |
| "eval_samples_per_second": 7.857, | |
| "eval_steps_per_second": 0.246, | |
| "eval_wer": 9.889521391594105, | |
| "step": 124000 | |
| }, | |
| { | |
| "epoch": 28.62, | |
| "grad_norm": 2.0855116844177246, | |
| "learning_rate": 7.147705787410653e-06, | |
| "loss": 0.0008, | |
| "step": 124250 | |
| }, | |
| { | |
| "epoch": 28.67, | |
| "grad_norm": 0.008047823794186115, | |
| "learning_rate": 7.141941434171087e-06, | |
| "loss": 0.0009, | |
| "step": 124500 | |
| }, | |
| { | |
| "epoch": 28.73, | |
| "grad_norm": 0.1283062994480133, | |
| "learning_rate": 7.13617708093152e-06, | |
| "loss": 0.0009, | |
| "step": 124750 | |
| }, | |
| { | |
| "epoch": 28.79, | |
| "grad_norm": 0.0888696163892746, | |
| "learning_rate": 7.1304127276919536e-06, | |
| "loss": 0.0009, | |
| "step": 125000 | |
| }, | |
| { | |
| "epoch": 28.79, | |
| "eval_loss": 0.13283681869506836, | |
| "eval_runtime": 852.2523, | |
| "eval_samples_per_second": 7.923, | |
| "eval_steps_per_second": 0.248, | |
| "eval_wer": 9.724894380863597, | |
| "step": 125000 | |
| }, | |
| { | |
| "epoch": 28.85, | |
| "grad_norm": 0.011558549478650093, | |
| "learning_rate": 7.1246483744523875e-06, | |
| "loss": 0.0009, | |
| "step": 125250 | |
| }, | |
| { | |
| "epoch": 28.9, | |
| "grad_norm": 0.0639125183224678, | |
| "learning_rate": 7.1188840212128205e-06, | |
| "loss": 0.0009, | |
| "step": 125500 | |
| }, | |
| { | |
| "epoch": 28.96, | |
| "grad_norm": 0.038483619689941406, | |
| "learning_rate": 7.1131196679732536e-06, | |
| "loss": 0.0009, | |
| "step": 125750 | |
| }, | |
| { | |
| "epoch": 29.02, | |
| "grad_norm": 0.053386442363262177, | |
| "learning_rate": 7.1073553147336875e-06, | |
| "loss": 0.0008, | |
| "step": 126000 | |
| }, | |
| { | |
| "epoch": 29.02, | |
| "eval_loss": 0.13507717847824097, | |
| "eval_runtime": 865.2057, | |
| "eval_samples_per_second": 7.804, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 9.788365035603071, | |
| "step": 126000 | |
| }, | |
| { | |
| "epoch": 29.08, | |
| "grad_norm": 0.5218244791030884, | |
| "learning_rate": 7.101590961494121e-06, | |
| "loss": 0.0005, | |
| "step": 126250 | |
| }, | |
| { | |
| "epoch": 29.13, | |
| "grad_norm": 0.046457044780254364, | |
| "learning_rate": 7.0958266082545544e-06, | |
| "loss": 0.0005, | |
| "step": 126500 | |
| }, | |
| { | |
| "epoch": 29.19, | |
| "grad_norm": 0.017029544338583946, | |
| "learning_rate": 7.0900622550149875e-06, | |
| "loss": 0.0007, | |
| "step": 126750 | |
| }, | |
| { | |
| "epoch": 29.25, | |
| "grad_norm": 0.09011877328157425, | |
| "learning_rate": 7.08432095918838e-06, | |
| "loss": 0.0007, | |
| "step": 127000 | |
| }, | |
| { | |
| "epoch": 29.25, | |
| "eval_loss": 0.13347487151622772, | |
| "eval_runtime": 859.4608, | |
| "eval_samples_per_second": 7.856, | |
| "eval_steps_per_second": 0.246, | |
| "eval_wer": 9.659440268163516, | |
| "step": 127000 | |
| }, | |
| { | |
| "epoch": 29.31, | |
| "grad_norm": 0.025057533755898476, | |
| "learning_rate": 7.078556605948813e-06, | |
| "loss": 0.0006, | |
| "step": 127250 | |
| }, | |
| { | |
| "epoch": 29.36, | |
| "grad_norm": 0.673778235912323, | |
| "learning_rate": 7.072815310122205e-06, | |
| "loss": 0.0011, | |
| "step": 127500 | |
| }, | |
| { | |
| "epoch": 29.42, | |
| "grad_norm": 0.3776152729988098, | |
| "learning_rate": 7.067050956882638e-06, | |
| "loss": 0.0008, | |
| "step": 127750 | |
| }, | |
| { | |
| "epoch": 29.48, | |
| "grad_norm": 0.062430597841739655, | |
| "learning_rate": 7.061286603643072e-06, | |
| "loss": 0.0007, | |
| "step": 128000 | |
| }, | |
| { | |
| "epoch": 29.48, | |
| "eval_loss": 0.1375785619020462, | |
| "eval_runtime": 869.3107, | |
| "eval_samples_per_second": 7.767, | |
| "eval_steps_per_second": 0.243, | |
| "eval_wer": 9.968859710018446, | |
| "step": 128000 | |
| }, | |
| { | |
| "epoch": 29.54, | |
| "grad_norm": 0.054513975977897644, | |
| "learning_rate": 7.055545307816464e-06, | |
| "loss": 0.001, | |
| "step": 128250 | |
| }, | |
| { | |
| "epoch": 29.59, | |
| "grad_norm": 0.19355349242687225, | |
| "learning_rate": 7.049780954576897e-06, | |
| "loss": 0.001, | |
| "step": 128500 | |
| }, | |
| { | |
| "epoch": 29.65, | |
| "grad_norm": 0.01914064772427082, | |
| "learning_rate": 7.04401660133733e-06, | |
| "loss": 0.0009, | |
| "step": 128750 | |
| }, | |
| { | |
| "epoch": 29.71, | |
| "grad_norm": 0.566556453704834, | |
| "learning_rate": 7.038252248097764e-06, | |
| "loss": 0.0006, | |
| "step": 129000 | |
| }, | |
| { | |
| "epoch": 29.71, | |
| "eval_loss": 0.13309036195278168, | |
| "eval_runtime": 871.0297, | |
| "eval_samples_per_second": 7.752, | |
| "eval_steps_per_second": 0.242, | |
| "eval_wer": 9.810183073169764, | |
| "step": 129000 | |
| }, | |
| { | |
| "epoch": 29.77, | |
| "grad_norm": 0.0412914976477623, | |
| "learning_rate": 7.032487894858197e-06, | |
| "loss": 0.001, | |
| "step": 129250 | |
| }, | |
| { | |
| "epoch": 29.82, | |
| "grad_norm": 0.22140279412269592, | |
| "learning_rate": 7.026723541618631e-06, | |
| "loss": 0.0009, | |
| "step": 129500 | |
| }, | |
| { | |
| "epoch": 29.88, | |
| "grad_norm": 1.2251185178756714, | |
| "learning_rate": 7.020959188379064e-06, | |
| "loss": 0.0009, | |
| "step": 129750 | |
| }, | |
| { | |
| "epoch": 29.94, | |
| "grad_norm": 0.04262693226337433, | |
| "learning_rate": 7.015194835139498e-06, | |
| "loss": 0.0012, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 29.94, | |
| "eval_loss": 0.13242684304714203, | |
| "eval_runtime": 871.8865, | |
| "eval_samples_per_second": 7.744, | |
| "eval_steps_per_second": 0.242, | |
| "eval_wer": 9.822083820933415, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 0.08901894092559814, | |
| "learning_rate": 7.009430481899931e-06, | |
| "loss": 0.0009, | |
| "step": 130250 | |
| }, | |
| { | |
| "epoch": 30.06, | |
| "grad_norm": 0.27203115820884705, | |
| "learning_rate": 7.003666128660365e-06, | |
| "loss": 0.0006, | |
| "step": 130500 | |
| }, | |
| { | |
| "epoch": 30.11, | |
| "grad_norm": 0.07738136500120163, | |
| "learning_rate": 6.997924832833756e-06, | |
| "loss": 0.0006, | |
| "step": 130750 | |
| }, | |
| { | |
| "epoch": 30.17, | |
| "grad_norm": 0.07566077262163162, | |
| "learning_rate": 6.99216047959419e-06, | |
| "loss": 0.0006, | |
| "step": 131000 | |
| }, | |
| { | |
| "epoch": 30.17, | |
| "eval_loss": 0.13523517549037933, | |
| "eval_runtime": 849.9774, | |
| "eval_samples_per_second": 7.944, | |
| "eval_steps_per_second": 0.248, | |
| "eval_wer": 9.86770335402741, | |
| "step": 131000 | |
| }, | |
| { | |
| "epoch": 30.23, | |
| "grad_norm": 0.04363285005092621, | |
| "learning_rate": 6.986396126354623e-06, | |
| "loss": 0.0007, | |
| "step": 131250 | |
| }, | |
| { | |
| "epoch": 30.29, | |
| "grad_norm": 0.8319553136825562, | |
| "learning_rate": 6.980631773115057e-06, | |
| "loss": 0.0006, | |
| "step": 131500 | |
| }, | |
| { | |
| "epoch": 30.34, | |
| "grad_norm": 0.05418200418353081, | |
| "learning_rate": 6.97486741987549e-06, | |
| "loss": 0.0009, | |
| "step": 131750 | |
| }, | |
| { | |
| "epoch": 30.4, | |
| "grad_norm": 0.023919006809592247, | |
| "learning_rate": 6.969103066635924e-06, | |
| "loss": 0.001, | |
| "step": 132000 | |
| }, | |
| { | |
| "epoch": 30.4, | |
| "eval_loss": 0.13232696056365967, | |
| "eval_runtime": 869.6064, | |
| "eval_samples_per_second": 7.764, | |
| "eval_steps_per_second": 0.243, | |
| "eval_wer": 9.730844754745423, | |
| "step": 132000 | |
| }, | |
| { | |
| "epoch": 30.46, | |
| "grad_norm": 0.04671400040388107, | |
| "learning_rate": 6.963338713396357e-06, | |
| "loss": 0.001, | |
| "step": 132250 | |
| }, | |
| { | |
| "epoch": 30.52, | |
| "grad_norm": 0.01245614979416132, | |
| "learning_rate": 6.957574360156791e-06, | |
| "loss": 0.0009, | |
| "step": 132500 | |
| }, | |
| { | |
| "epoch": 30.57, | |
| "grad_norm": 1.1713380813598633, | |
| "learning_rate": 6.951810006917224e-06, | |
| "loss": 0.001, | |
| "step": 132750 | |
| }, | |
| { | |
| "epoch": 30.63, | |
| "grad_norm": 0.01822400651872158, | |
| "learning_rate": 6.946045653677658e-06, | |
| "loss": 0.0008, | |
| "step": 133000 | |
| }, | |
| { | |
| "epoch": 30.63, | |
| "eval_loss": 0.13655835390090942, | |
| "eval_runtime": 855.6562, | |
| "eval_samples_per_second": 7.891, | |
| "eval_steps_per_second": 0.247, | |
| "eval_wer": 9.90935597120019, | |
| "step": 133000 | |
| }, | |
| { | |
| "epoch": 30.69, | |
| "grad_norm": 0.11439549922943115, | |
| "learning_rate": 6.940281300438091e-06, | |
| "loss": 0.0009, | |
| "step": 133250 | |
| }, | |
| { | |
| "epoch": 30.75, | |
| "grad_norm": 0.2022576779127121, | |
| "learning_rate": 6.934516947198525e-06, | |
| "loss": 0.0011, | |
| "step": 133500 | |
| }, | |
| { | |
| "epoch": 30.8, | |
| "grad_norm": 0.0333305299282074, | |
| "learning_rate": 6.928752593958958e-06, | |
| "loss": 0.001, | |
| "step": 133750 | |
| }, | |
| { | |
| "epoch": 30.86, | |
| "grad_norm": 0.27165189385414124, | |
| "learning_rate": 6.922988240719391e-06, | |
| "loss": 0.0008, | |
| "step": 134000 | |
| }, | |
| { | |
| "epoch": 30.86, | |
| "eval_loss": 0.13061845302581787, | |
| "eval_runtime": 861.0795, | |
| "eval_samples_per_second": 7.841, | |
| "eval_steps_per_second": 0.245, | |
| "eval_wer": 9.556300454211874, | |
| "step": 134000 | |
| } | |
| ], | |
| "logging_steps": 250, | |
| "max_steps": 434200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 1000, | |
| "total_flos": 1.2371858121498627e+21, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |