|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 9.312782393237137, |
|
"global_step": 102910, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 0.0001188, |
|
"loss": 6.3637, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"eval_loss": 2.970613718032837, |
|
"eval_runtime": 1204.8934, |
|
"eval_samples_per_second": 9.49, |
|
"eval_steps_per_second": 1.187, |
|
"eval_wer": 1.0, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"learning_rate": 0.0002388, |
|
"loss": 2.9537, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"eval_loss": 2.926468849182129, |
|
"eval_runtime": 1196.9703, |
|
"eval_samples_per_second": 9.553, |
|
"eval_steps_per_second": 1.195, |
|
"eval_wer": 1.0, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"learning_rate": 0.00029971291866028705, |
|
"loss": 2.322, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"eval_loss": 0.9299482107162476, |
|
"eval_runtime": 1200.9258, |
|
"eval_samples_per_second": 9.522, |
|
"eval_steps_per_second": 1.191, |
|
"eval_wer": 0.7969234236772412, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.08, |
|
"learning_rate": 0.0002991270383751586, |
|
"loss": 1.1527, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 0.08, |
|
"eval_loss": 0.6302356719970703, |
|
"eval_runtime": 1200.2547, |
|
"eval_samples_per_second": 9.527, |
|
"eval_steps_per_second": 1.191, |
|
"eval_wer": 0.6488195187919085, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"learning_rate": 0.00029854115809003026, |
|
"loss": 0.9892, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"eval_loss": 0.5595320463180542, |
|
"eval_runtime": 1199.9051, |
|
"eval_samples_per_second": 9.53, |
|
"eval_steps_per_second": 1.192, |
|
"eval_wer": 0.6062207697075562, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"learning_rate": 0.00029795527780490183, |
|
"loss": 0.8919, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"eval_loss": 0.5005972981452942, |
|
"eval_runtime": 1200.1109, |
|
"eval_samples_per_second": 9.528, |
|
"eval_steps_per_second": 1.192, |
|
"eval_wer": 0.5602411675212712, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 0.0002973693975197734, |
|
"loss": 0.8609, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"eval_loss": 0.4540606439113617, |
|
"eval_runtime": 1277.6037, |
|
"eval_samples_per_second": 8.95, |
|
"eval_steps_per_second": 1.119, |
|
"eval_wer": 0.5377472248830788, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"learning_rate": 0.00029678351723464504, |
|
"loss": 0.8229, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"eval_loss": 0.42465054988861084, |
|
"eval_runtime": 1258.3012, |
|
"eval_samples_per_second": 9.088, |
|
"eval_steps_per_second": 1.136, |
|
"eval_wer": 0.5064968727108807, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"learning_rate": 0.0002961976369495166, |
|
"loss": 0.7813, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"eval_loss": 0.4217381477355957, |
|
"eval_runtime": 1266.2898, |
|
"eval_samples_per_second": 9.03, |
|
"eval_steps_per_second": 1.129, |
|
"eval_wer": 0.5103735842677636, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 0.08, |
|
"learning_rate": 0.0002956117566643882, |
|
"loss": 0.7586, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.08, |
|
"eval_loss": 0.3938300311565399, |
|
"eval_runtime": 1264.268, |
|
"eval_samples_per_second": 9.045, |
|
"eval_steps_per_second": 1.131, |
|
"eval_wer": 0.48733870513326194, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"learning_rate": 0.0002950258763792598, |
|
"loss": 0.7416, |
|
"step": 2200 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"eval_loss": 0.3816222548484802, |
|
"eval_runtime": 1263.3113, |
|
"eval_samples_per_second": 9.052, |
|
"eval_steps_per_second": 1.132, |
|
"eval_wer": 0.46917225446554345, |
|
"step": 2200 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"learning_rate": 0.0002944399960941314, |
|
"loss": 0.7077, |
|
"step": 2400 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"eval_loss": 0.4034684896469116, |
|
"eval_runtime": 1262.8373, |
|
"eval_samples_per_second": 9.055, |
|
"eval_steps_per_second": 1.132, |
|
"eval_wer": 0.4645404857158956, |
|
"step": 2400 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"learning_rate": 0.000293854115809003, |
|
"loss": 0.7359, |
|
"step": 2600 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"eval_loss": 0.3767629861831665, |
|
"eval_runtime": 1266.8299, |
|
"eval_samples_per_second": 9.026, |
|
"eval_steps_per_second": 1.129, |
|
"eval_wer": 0.46494618808812754, |
|
"step": 2600 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"learning_rate": 0.0002932682355238746, |
|
"loss": 0.7335, |
|
"step": 2800 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"eval_loss": 0.37543314695358276, |
|
"eval_runtime": 1278.5608, |
|
"eval_samples_per_second": 8.944, |
|
"eval_steps_per_second": 1.118, |
|
"eval_wer": 0.4669296219079281, |
|
"step": 2800 |
|
}, |
|
{ |
|
"epoch": 0.17, |
|
"learning_rate": 0.0002926823552387462, |
|
"loss": 0.7204, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 0.17, |
|
"eval_loss": 0.3659396171569824, |
|
"eval_runtime": 1288.786, |
|
"eval_samples_per_second": 8.873, |
|
"eval_steps_per_second": 1.11, |
|
"eval_wer": 0.4529103510452471, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 0.19, |
|
"learning_rate": 0.00029209647495361777, |
|
"loss": 0.7003, |
|
"step": 3200 |
|
}, |
|
{ |
|
"epoch": 0.19, |
|
"eval_loss": 0.3635198473930359, |
|
"eval_runtime": 1309.43, |
|
"eval_samples_per_second": 8.733, |
|
"eval_steps_per_second": 1.092, |
|
"eval_wer": 0.43374091395728853, |
|
"step": 3200 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"learning_rate": 0.0002915105946684894, |
|
"loss": 0.7048, |
|
"step": 3400 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"eval_loss": 0.33965152502059937, |
|
"eval_runtime": 1308.7374, |
|
"eval_samples_per_second": 8.737, |
|
"eval_steps_per_second": 1.093, |
|
"eval_wer": 0.4208824026596044, |
|
"step": 3400 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"learning_rate": 0.000290924714383361, |
|
"loss": 0.6782, |
|
"step": 3600 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"eval_loss": 0.34997662901878357, |
|
"eval_runtime": 1307.483, |
|
"eval_samples_per_second": 8.746, |
|
"eval_steps_per_second": 1.094, |
|
"eval_wer": 0.43742604383839523, |
|
"step": 3600 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"learning_rate": 0.00029033883409823255, |
|
"loss": 0.6748, |
|
"step": 3800 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"eval_loss": 0.328862726688385, |
|
"eval_runtime": 1308.3222, |
|
"eval_samples_per_second": 8.74, |
|
"eval_steps_per_second": 1.093, |
|
"eval_wer": 0.4091283033752183, |
|
"step": 3800 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"learning_rate": 0.0002897558832145298, |
|
"loss": 0.6673, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"eval_loss": 0.328796923160553, |
|
"eval_runtime": 1280.3163, |
|
"eval_samples_per_second": 8.931, |
|
"eval_steps_per_second": 1.117, |
|
"eval_wer": 0.42204316222460136, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"learning_rate": 0.0002891700029294014, |
|
"loss": 0.682, |
|
"step": 4200 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"eval_loss": 0.32723408937454224, |
|
"eval_runtime": 1263.2249, |
|
"eval_samples_per_second": 9.052, |
|
"eval_steps_per_second": 1.132, |
|
"eval_wer": 0.40732518172085425, |
|
"step": 4200 |
|
}, |
|
{ |
|
"epoch": 0.31, |
|
"learning_rate": 0.000288584122644273, |
|
"loss": 0.6484, |
|
"step": 4400 |
|
}, |
|
{ |
|
"epoch": 0.31, |
|
"eval_loss": 0.3117896616458893, |
|
"eval_runtime": 1263.1269, |
|
"eval_samples_per_second": 9.053, |
|
"eval_steps_per_second": 1.132, |
|
"eval_wer": 0.39746436017355047, |
|
"step": 4400 |
|
}, |
|
{ |
|
"epoch": 0.33, |
|
"learning_rate": 0.0002879982423591446, |
|
"loss": 0.6614, |
|
"step": 4600 |
|
}, |
|
{ |
|
"epoch": 0.33, |
|
"eval_loss": 0.3179396390914917, |
|
"eval_runtime": 1264.4151, |
|
"eval_samples_per_second": 9.044, |
|
"eval_steps_per_second": 1.131, |
|
"eval_wer": 0.3975094382149096, |
|
"step": 4600 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"learning_rate": 0.0002874123620740162, |
|
"loss": 0.6557, |
|
"step": 4800 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"eval_loss": 0.31483808159828186, |
|
"eval_runtime": 1264.2237, |
|
"eval_samples_per_second": 9.045, |
|
"eval_steps_per_second": 1.131, |
|
"eval_wer": 0.39069138445934526, |
|
"step": 4800 |
|
}, |
|
{ |
|
"epoch": 0.37, |
|
"learning_rate": 0.00028682648178888776, |
|
"loss": 0.6501, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 0.37, |
|
"eval_loss": 0.32537421584129333, |
|
"eval_runtime": 1269.162, |
|
"eval_samples_per_second": 9.01, |
|
"eval_steps_per_second": 1.127, |
|
"eval_wer": 0.3900828309009974, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 0.39, |
|
"learning_rate": 0.00028624060150375934, |
|
"loss": 0.6487, |
|
"step": 5200 |
|
}, |
|
{ |
|
"epoch": 0.39, |
|
"eval_loss": 0.29997891187667847, |
|
"eval_runtime": 1266.7543, |
|
"eval_samples_per_second": 9.027, |
|
"eval_steps_per_second": 1.129, |
|
"eval_wer": 0.3837042880486843, |
|
"step": 5200 |
|
}, |
|
{ |
|
"epoch": 0.41, |
|
"learning_rate": 0.00028565472121863097, |
|
"loss": 0.6493, |
|
"step": 5400 |
|
}, |
|
{ |
|
"epoch": 0.41, |
|
"eval_loss": 0.2948152720928192, |
|
"eval_runtime": 1259.7136, |
|
"eval_samples_per_second": 9.077, |
|
"eval_steps_per_second": 1.135, |
|
"eval_wer": 0.3774271707894292, |
|
"step": 5400 |
|
}, |
|
{ |
|
"epoch": 0.43, |
|
"learning_rate": 0.00028506884093350255, |
|
"loss": 0.6302, |
|
"step": 5600 |
|
}, |
|
{ |
|
"epoch": 0.43, |
|
"eval_loss": 0.29879939556121826, |
|
"eval_runtime": 1259.867, |
|
"eval_samples_per_second": 9.076, |
|
"eval_steps_per_second": 1.135, |
|
"eval_wer": 0.3797824984504423, |
|
"step": 5600 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"learning_rate": 0.0002844829606483741, |
|
"loss": 0.6508, |
|
"step": 5800 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"eval_loss": 0.29643869400024414, |
|
"eval_runtime": 1261.7676, |
|
"eval_samples_per_second": 9.063, |
|
"eval_steps_per_second": 1.133, |
|
"eval_wer": 0.3807742153603426, |
|
"step": 5800 |
|
}, |
|
{ |
|
"epoch": 0.47, |
|
"learning_rate": 0.00028389708036324576, |
|
"loss": 0.6188, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 0.47, |
|
"eval_loss": 0.29014137387275696, |
|
"eval_runtime": 1270.9486, |
|
"eval_samples_per_second": 8.997, |
|
"eval_steps_per_second": 1.125, |
|
"eval_wer": 0.3699554854341579, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 0.49, |
|
"learning_rate": 0.0002833112000781174, |
|
"loss": 0.6215, |
|
"step": 6200 |
|
}, |
|
{ |
|
"epoch": 0.49, |
|
"eval_loss": 0.29307612776756287, |
|
"eval_runtime": 1276.8284, |
|
"eval_samples_per_second": 8.956, |
|
"eval_steps_per_second": 1.12, |
|
"eval_wer": 0.3614244661069477, |
|
"step": 6200 |
|
}, |
|
{ |
|
"epoch": 0.51, |
|
"learning_rate": 0.0002827253197929889, |
|
"loss": 0.6035, |
|
"step": 6400 |
|
}, |
|
{ |
|
"epoch": 0.51, |
|
"eval_loss": 0.2853919267654419, |
|
"eval_runtime": 1270.8617, |
|
"eval_samples_per_second": 8.998, |
|
"eval_steps_per_second": 1.125, |
|
"eval_wer": 0.3639150278920381, |
|
"step": 6400 |
|
}, |
|
{ |
|
"epoch": 0.52, |
|
"learning_rate": 0.00028213943950786054, |
|
"loss": 0.613, |
|
"step": 6600 |
|
}, |
|
{ |
|
"epoch": 0.52, |
|
"eval_loss": 0.2793109714984894, |
|
"eval_runtime": 1273.0596, |
|
"eval_samples_per_second": 8.982, |
|
"eval_steps_per_second": 1.123, |
|
"eval_wer": 0.3593170676734096, |
|
"step": 6600 |
|
}, |
|
{ |
|
"epoch": 0.54, |
|
"learning_rate": 0.0002815535592227321, |
|
"loss": 0.5967, |
|
"step": 6800 |
|
}, |
|
{ |
|
"epoch": 0.54, |
|
"eval_loss": 0.2713634669780731, |
|
"eval_runtime": 1279.2102, |
|
"eval_samples_per_second": 8.939, |
|
"eval_steps_per_second": 1.118, |
|
"eval_wer": 0.35412182340677295, |
|
"step": 6800 |
|
}, |
|
{ |
|
"epoch": 0.56, |
|
"learning_rate": 0.00028096767893760375, |
|
"loss": 0.6118, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 0.56, |
|
"eval_loss": 0.27592843770980835, |
|
"eval_runtime": 1269.9249, |
|
"eval_samples_per_second": 9.004, |
|
"eval_steps_per_second": 1.126, |
|
"eval_wer": 0.34716853552713134, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 0.58, |
|
"learning_rate": 0.00028038179865247533, |
|
"loss": 0.6016, |
|
"step": 7200 |
|
}, |
|
{ |
|
"epoch": 0.58, |
|
"eval_loss": 0.27817872166633606, |
|
"eval_runtime": 1278.4857, |
|
"eval_samples_per_second": 8.944, |
|
"eval_steps_per_second": 1.119, |
|
"eval_wer": 0.35648842057812585, |
|
"step": 7200 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"learning_rate": 0.0002797959183673469, |
|
"loss": 0.5965, |
|
"step": 7400 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"eval_loss": 0.2640519440174103, |
|
"eval_runtime": 1280.4206, |
|
"eval_samples_per_second": 8.931, |
|
"eval_steps_per_second": 1.117, |
|
"eval_wer": 0.3394714599650645, |
|
"step": 7400 |
|
}, |
|
{ |
|
"epoch": 0.62, |
|
"learning_rate": 0.00027921003808221854, |
|
"loss": 0.596, |
|
"step": 7600 |
|
}, |
|
{ |
|
"epoch": 0.62, |
|
"eval_loss": 0.2649860978126526, |
|
"eval_runtime": 1278.933, |
|
"eval_samples_per_second": 8.941, |
|
"eval_steps_per_second": 1.118, |
|
"eval_wer": 0.3421197948949118, |
|
"step": 7600 |
|
}, |
|
{ |
|
"epoch": 0.64, |
|
"learning_rate": 0.0002786241577970901, |
|
"loss": 0.6075, |
|
"step": 7800 |
|
}, |
|
{ |
|
"epoch": 0.64, |
|
"eval_loss": 0.26620569825172424, |
|
"eval_runtime": 1277.4585, |
|
"eval_samples_per_second": 8.951, |
|
"eval_steps_per_second": 1.119, |
|
"eval_wer": 0.3415225108469037, |
|
"step": 7800 |
|
}, |
|
{ |
|
"epoch": 0.66, |
|
"learning_rate": 0.0002780382775119617, |
|
"loss": 0.5988, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 0.66, |
|
"eval_loss": 0.2684703469276428, |
|
"eval_runtime": 1278.0005, |
|
"eval_samples_per_second": 8.948, |
|
"eval_steps_per_second": 1.119, |
|
"eval_wer": 0.336383614131966, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 0.68, |
|
"learning_rate": 0.0002774523972268333, |
|
"loss": 0.5973, |
|
"step": 8200 |
|
}, |
|
{ |
|
"epoch": 0.68, |
|
"eval_loss": 0.25252845883369446, |
|
"eval_runtime": 1274.619, |
|
"eval_samples_per_second": 8.971, |
|
"eval_steps_per_second": 1.122, |
|
"eval_wer": 0.33392686087789486, |
|
"step": 8200 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"learning_rate": 0.0002768665169417049, |
|
"loss": 0.5809, |
|
"step": 8400 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"eval_loss": 0.2525635361671448, |
|
"eval_runtime": 1273.2438, |
|
"eval_samples_per_second": 8.981, |
|
"eval_steps_per_second": 1.123, |
|
"eval_wer": 0.322713698089818, |
|
"step": 8400 |
|
}, |
|
{ |
|
"epoch": 0.72, |
|
"learning_rate": 0.0002762806366565765, |
|
"loss": 0.574, |
|
"step": 8600 |
|
}, |
|
{ |
|
"epoch": 0.72, |
|
"eval_loss": 0.25077855587005615, |
|
"eval_runtime": 1281.2675, |
|
"eval_samples_per_second": 8.925, |
|
"eval_steps_per_second": 1.116, |
|
"eval_wer": 0.32358145038598074, |
|
"step": 8600 |
|
}, |
|
{ |
|
"epoch": 0.74, |
|
"learning_rate": 0.0002756947563714481, |
|
"loss": 0.5893, |
|
"step": 8800 |
|
}, |
|
{ |
|
"epoch": 0.74, |
|
"eval_loss": 0.25881966948509216, |
|
"eval_runtime": 1276.5376, |
|
"eval_samples_per_second": 8.958, |
|
"eval_steps_per_second": 1.12, |
|
"eval_wer": 0.3342649461880881, |
|
"step": 8800 |
|
}, |
|
{ |
|
"epoch": 0.76, |
|
"learning_rate": 0.0002751088760863197, |
|
"loss": 0.5876, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 0.76, |
|
"eval_loss": 0.25579118728637695, |
|
"eval_runtime": 1273.8923, |
|
"eval_samples_per_second": 8.976, |
|
"eval_steps_per_second": 1.123, |
|
"eval_wer": 0.3283709922803854, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 0.78, |
|
"learning_rate": 0.00027452299580119126, |
|
"loss": 0.5686, |
|
"step": 9200 |
|
}, |
|
{ |
|
"epoch": 0.78, |
|
"eval_loss": 0.24226322770118713, |
|
"eval_runtime": 1284.5776, |
|
"eval_samples_per_second": 8.902, |
|
"eval_steps_per_second": 1.113, |
|
"eval_wer": 0.311004676846791, |
|
"step": 9200 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"learning_rate": 0.00027393711551606284, |
|
"loss": 0.5798, |
|
"step": 9400 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"eval_loss": 0.24731135368347168, |
|
"eval_runtime": 1276.3349, |
|
"eval_samples_per_second": 8.959, |
|
"eval_steps_per_second": 1.12, |
|
"eval_wer": 0.3168648222234744, |
|
"step": 9400 |
|
}, |
|
{ |
|
"epoch": 0.82, |
|
"learning_rate": 0.00027335123523093447, |
|
"loss": 0.5758, |
|
"step": 9600 |
|
}, |
|
{ |
|
"epoch": 0.82, |
|
"eval_loss": 0.2408532351255417, |
|
"eval_runtime": 1298.8974, |
|
"eval_samples_per_second": 8.804, |
|
"eval_steps_per_second": 1.101, |
|
"eval_wer": 0.3150053530174114, |
|
"step": 9600 |
|
}, |
|
{ |
|
"epoch": 0.84, |
|
"learning_rate": 0.00027276535494580605, |
|
"loss": 0.5717, |
|
"step": 9800 |
|
}, |
|
{ |
|
"epoch": 0.84, |
|
"eval_loss": 0.24253633618354797, |
|
"eval_runtime": 1290.9724, |
|
"eval_samples_per_second": 8.858, |
|
"eval_steps_per_second": 1.108, |
|
"eval_wer": 0.30879585282019495, |
|
"step": 9800 |
|
}, |
|
{ |
|
"epoch": 0.86, |
|
"learning_rate": 0.0002721794746606776, |
|
"loss": 0.5515, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 0.86, |
|
"eval_loss": 0.24737118184566498, |
|
"eval_runtime": 1307.8378, |
|
"eval_samples_per_second": 8.743, |
|
"eval_steps_per_second": 1.093, |
|
"eval_wer": 0.31223305347382657, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 0.87, |
|
"learning_rate": 0.00027159359437554925, |
|
"loss": 0.5719, |
|
"step": 10200 |
|
}, |
|
{ |
|
"epoch": 0.87, |
|
"eval_loss": 0.23542268574237823, |
|
"eval_runtime": 1311.0297, |
|
"eval_samples_per_second": 8.722, |
|
"eval_steps_per_second": 1.091, |
|
"eval_wer": 0.30239477094720235, |
|
"step": 10200 |
|
}, |
|
{ |
|
"epoch": 0.89, |
|
"learning_rate": 0.00027100771409042083, |
|
"loss": 0.5614, |
|
"step": 10400 |
|
}, |
|
{ |
|
"epoch": 0.89, |
|
"eval_loss": 0.247614324092865, |
|
"eval_runtime": 1303.8784, |
|
"eval_samples_per_second": 8.77, |
|
"eval_steps_per_second": 1.097, |
|
"eval_wer": 0.3179579647264326, |
|
"step": 10400 |
|
}, |
|
{ |
|
"epoch": 0.91, |
|
"learning_rate": 0.00027042476320671805, |
|
"loss": 0.5593, |
|
"step": 10600 |
|
}, |
|
{ |
|
"epoch": 0.91, |
|
"eval_loss": 0.23083819448947906, |
|
"eval_runtime": 1300.6119, |
|
"eval_samples_per_second": 8.792, |
|
"eval_steps_per_second": 1.099, |
|
"eval_wer": 0.2990251873556094, |
|
"step": 10600 |
|
}, |
|
{ |
|
"epoch": 0.93, |
|
"learning_rate": 0.0002698388829215897, |
|
"loss": 0.5617, |
|
"step": 10800 |
|
}, |
|
{ |
|
"epoch": 0.93, |
|
"eval_loss": 0.235725536942482, |
|
"eval_runtime": 1286.8437, |
|
"eval_samples_per_second": 8.886, |
|
"eval_steps_per_second": 1.111, |
|
"eval_wer": 0.29992674818279147, |
|
"step": 10800 |
|
}, |
|
{ |
|
"epoch": 0.95, |
|
"learning_rate": 0.00026925300263646126, |
|
"loss": 0.5596, |
|
"step": 11000 |
|
}, |
|
{ |
|
"epoch": 0.95, |
|
"eval_loss": 0.24363787472248077, |
|
"eval_runtime": 1282.1875, |
|
"eval_samples_per_second": 8.918, |
|
"eval_steps_per_second": 1.115, |
|
"eval_wer": 0.30588831915253284, |
|
"step": 11000 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"learning_rate": 0.00026866712235133284, |
|
"loss": 0.5505, |
|
"step": 11200 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"eval_loss": 0.2367905229330063, |
|
"eval_runtime": 1272.677, |
|
"eval_samples_per_second": 8.985, |
|
"eval_steps_per_second": 1.124, |
|
"eval_wer": 0.30030991153434383, |
|
"step": 11200 |
|
}, |
|
{ |
|
"epoch": 1.04, |
|
"learning_rate": 0.00026808124206620447, |
|
"loss": 0.5445, |
|
"step": 11400 |
|
}, |
|
{ |
|
"epoch": 1.04, |
|
"eval_loss": 0.2314738780260086, |
|
"eval_runtime": 1260.4441, |
|
"eval_samples_per_second": 9.072, |
|
"eval_steps_per_second": 1.135, |
|
"eval_wer": 0.2962866963430439, |
|
"step": 11400 |
|
}, |
|
{ |
|
"epoch": 1.06, |
|
"learning_rate": 0.00026749536178107604, |
|
"loss": 0.5259, |
|
"step": 11600 |
|
}, |
|
{ |
|
"epoch": 1.06, |
|
"eval_loss": 0.2309747040271759, |
|
"eval_runtime": 1241.2773, |
|
"eval_samples_per_second": 9.212, |
|
"eval_steps_per_second": 1.152, |
|
"eval_wer": 0.2933002761030033, |
|
"step": 11600 |
|
}, |
|
{ |
|
"epoch": 1.08, |
|
"learning_rate": 0.0002669094814959476, |
|
"loss": 0.5268, |
|
"step": 11800 |
|
}, |
|
{ |
|
"epoch": 1.08, |
|
"eval_loss": 0.22442255914211273, |
|
"eval_runtime": 1246.8773, |
|
"eval_samples_per_second": 9.171, |
|
"eval_steps_per_second": 1.147, |
|
"eval_wer": 0.2917788922071336, |
|
"step": 11800 |
|
}, |
|
{ |
|
"epoch": 1.1, |
|
"learning_rate": 0.00026632360121081925, |
|
"loss": 0.5479, |
|
"step": 12000 |
|
}, |
|
{ |
|
"epoch": 1.1, |
|
"eval_loss": 0.22898849844932556, |
|
"eval_runtime": 1248.9274, |
|
"eval_samples_per_second": 9.156, |
|
"eval_steps_per_second": 1.145, |
|
"eval_wer": 0.2952048233504254, |
|
"step": 12000 |
|
}, |
|
{ |
|
"epoch": 1.12, |
|
"learning_rate": 0.00026573772092569083, |
|
"loss": 0.5308, |
|
"step": 12200 |
|
}, |
|
{ |
|
"epoch": 1.12, |
|
"eval_loss": 0.2276710867881775, |
|
"eval_runtime": 1248.0355, |
|
"eval_samples_per_second": 9.162, |
|
"eval_steps_per_second": 1.146, |
|
"eval_wer": 0.28650476136811853, |
|
"step": 12200 |
|
}, |
|
{ |
|
"epoch": 1.14, |
|
"learning_rate": 0.0002651518406405624, |
|
"loss": 0.5345, |
|
"step": 12400 |
|
}, |
|
{ |
|
"epoch": 1.14, |
|
"eval_loss": 0.22329020500183105, |
|
"eval_runtime": 1258.1958, |
|
"eval_samples_per_second": 9.088, |
|
"eval_steps_per_second": 1.137, |
|
"eval_wer": 0.29397644672338985, |
|
"step": 12400 |
|
}, |
|
{ |
|
"epoch": 1.16, |
|
"learning_rate": 0.00026456596035543404, |
|
"loss": 0.5324, |
|
"step": 12600 |
|
}, |
|
{ |
|
"epoch": 1.16, |
|
"eval_loss": 0.23222756385803223, |
|
"eval_runtime": 1257.7533, |
|
"eval_samples_per_second": 9.092, |
|
"eval_steps_per_second": 1.137, |
|
"eval_wer": 0.2929734603031498, |
|
"step": 12600 |
|
}, |
|
{ |
|
"epoch": 1.17, |
|
"learning_rate": 0.00026398300947173126, |
|
"loss": 0.5374, |
|
"step": 12800 |
|
}, |
|
{ |
|
"epoch": 1.17, |
|
"eval_loss": 0.21313965320587158, |
|
"eval_runtime": 1251.7467, |
|
"eval_samples_per_second": 9.135, |
|
"eval_steps_per_second": 1.142, |
|
"eval_wer": 0.27523525102834284, |
|
"step": 12800 |
|
}, |
|
{ |
|
"epoch": 1.19, |
|
"learning_rate": 0.00026339712918660283, |
|
"loss": 0.5177, |
|
"step": 13000 |
|
}, |
|
{ |
|
"epoch": 1.19, |
|
"eval_loss": 0.2149176001548767, |
|
"eval_runtime": 1254.8601, |
|
"eval_samples_per_second": 9.113, |
|
"eval_steps_per_second": 1.14, |
|
"eval_wer": 0.2795063954471178, |
|
"step": 13000 |
|
}, |
|
{ |
|
"epoch": 1.21, |
|
"learning_rate": 0.00026281124890147447, |
|
"loss": 0.5234, |
|
"step": 13200 |
|
}, |
|
{ |
|
"epoch": 1.21, |
|
"eval_loss": 0.21338462829589844, |
|
"eval_runtime": 1267.9076, |
|
"eval_samples_per_second": 9.019, |
|
"eval_steps_per_second": 1.128, |
|
"eval_wer": 0.2745478108976165, |
|
"step": 13200 |
|
}, |
|
{ |
|
"epoch": 1.23, |
|
"learning_rate": 0.00026222536861634604, |
|
"loss": 0.5208, |
|
"step": 13400 |
|
}, |
|
{ |
|
"epoch": 1.23, |
|
"eval_loss": 0.21556991338729858, |
|
"eval_runtime": 1278.0567, |
|
"eval_samples_per_second": 8.947, |
|
"eval_steps_per_second": 1.119, |
|
"eval_wer": 0.27523525102834284, |
|
"step": 13400 |
|
}, |
|
{ |
|
"epoch": 1.25, |
|
"learning_rate": 0.0002616394883312176, |
|
"loss": 0.5139, |
|
"step": 13600 |
|
}, |
|
{ |
|
"epoch": 1.25, |
|
"eval_loss": 0.21141602098941803, |
|
"eval_runtime": 1279.0428, |
|
"eval_samples_per_second": 8.94, |
|
"eval_steps_per_second": 1.118, |
|
"eval_wer": 0.2733307037809207, |
|
"step": 13600 |
|
}, |
|
{ |
|
"epoch": 1.27, |
|
"learning_rate": 0.00026105360804608925, |
|
"loss": 0.5034, |
|
"step": 13800 |
|
}, |
|
{ |
|
"epoch": 1.27, |
|
"eval_loss": 0.21080410480499268, |
|
"eval_runtime": 1283.2325, |
|
"eval_samples_per_second": 8.911, |
|
"eval_steps_per_second": 1.114, |
|
"eval_wer": 0.2685524313968558, |
|
"step": 13800 |
|
}, |
|
{ |
|
"epoch": 1.29, |
|
"learning_rate": 0.00026046772776096083, |
|
"loss": 0.5283, |
|
"step": 14000 |
|
}, |
|
{ |
|
"epoch": 1.29, |
|
"eval_loss": 0.21575190126895905, |
|
"eval_runtime": 1258.3659, |
|
"eval_samples_per_second": 9.087, |
|
"eval_steps_per_second": 1.136, |
|
"eval_wer": 0.2740406829323266, |
|
"step": 14000 |
|
}, |
|
{ |
|
"epoch": 1.31, |
|
"learning_rate": 0.0002598818474758324, |
|
"loss": 0.5166, |
|
"step": 14200 |
|
}, |
|
{ |
|
"epoch": 1.31, |
|
"eval_loss": 0.20629876852035522, |
|
"eval_runtime": 1254.3394, |
|
"eval_samples_per_second": 9.116, |
|
"eval_steps_per_second": 1.14, |
|
"eval_wer": 0.2717980503747112, |
|
"step": 14200 |
|
}, |
|
{ |
|
"epoch": 1.33, |
|
"learning_rate": 0.00025929596719070404, |
|
"loss": 0.5214, |
|
"step": 14400 |
|
}, |
|
{ |
|
"epoch": 1.33, |
|
"eval_loss": 0.20648950338363647, |
|
"eval_runtime": 1257.7224, |
|
"eval_samples_per_second": 9.092, |
|
"eval_steps_per_second": 1.137, |
|
"eval_wer": 0.2600552206006649, |
|
"step": 14400 |
|
}, |
|
{ |
|
"epoch": 1.35, |
|
"learning_rate": 0.0002587100869055756, |
|
"loss": 0.511, |
|
"step": 14600 |
|
}, |
|
{ |
|
"epoch": 1.35, |
|
"eval_loss": 0.21166543662548065, |
|
"eval_runtime": 1257.0914, |
|
"eval_samples_per_second": 9.096, |
|
"eval_steps_per_second": 1.138, |
|
"eval_wer": 0.2707837944441314, |
|
"step": 14600 |
|
}, |
|
{ |
|
"epoch": 1.37, |
|
"learning_rate": 0.0002581242066204472, |
|
"loss": 0.5086, |
|
"step": 14800 |
|
}, |
|
{ |
|
"epoch": 1.37, |
|
"eval_loss": 0.20775602757930756, |
|
"eval_runtime": 1250.3755, |
|
"eval_samples_per_second": 9.145, |
|
"eval_steps_per_second": 1.144, |
|
"eval_wer": 0.26485603200540936, |
|
"step": 14800 |
|
}, |
|
{ |
|
"epoch": 1.39, |
|
"learning_rate": 0.00025753832633531877, |
|
"loss": 0.5134, |
|
"step": 15000 |
|
}, |
|
{ |
|
"epoch": 1.39, |
|
"eval_loss": 0.20706060528755188, |
|
"eval_runtime": 1256.0437, |
|
"eval_samples_per_second": 9.104, |
|
"eval_steps_per_second": 1.138, |
|
"eval_wer": 0.2688341691553502, |
|
"step": 15000 |
|
}, |
|
{ |
|
"epoch": 1.41, |
|
"learning_rate": 0.00025695537545161604, |
|
"loss": 0.5021, |
|
"step": 15200 |
|
}, |
|
{ |
|
"epoch": 1.41, |
|
"eval_loss": 0.2014995664358139, |
|
"eval_runtime": 1266.4014, |
|
"eval_samples_per_second": 9.03, |
|
"eval_steps_per_second": 1.129, |
|
"eval_wer": 0.26383050656448975, |
|
"step": 15200 |
|
}, |
|
{ |
|
"epoch": 1.43, |
|
"learning_rate": 0.0002563694951664876, |
|
"loss": 0.506, |
|
"step": 15400 |
|
}, |
|
{ |
|
"epoch": 1.43, |
|
"eval_loss": 0.20584553480148315, |
|
"eval_runtime": 1282.7956, |
|
"eval_samples_per_second": 8.914, |
|
"eval_steps_per_second": 1.115, |
|
"eval_wer": 0.2588268439736294, |
|
"step": 15400 |
|
}, |
|
{ |
|
"epoch": 1.45, |
|
"learning_rate": 0.00025578361488135925, |
|
"loss": 0.5139, |
|
"step": 15600 |
|
}, |
|
{ |
|
"epoch": 1.45, |
|
"eval_loss": 0.20228460431098938, |
|
"eval_runtime": 1297.6236, |
|
"eval_samples_per_second": 8.812, |
|
"eval_steps_per_second": 1.102, |
|
"eval_wer": 0.25989744745590804, |
|
"step": 15600 |
|
}, |
|
{ |
|
"epoch": 1.47, |
|
"learning_rate": 0.0002551977345962308, |
|
"loss": 0.4951, |
|
"step": 15800 |
|
}, |
|
{ |
|
"epoch": 1.47, |
|
"eval_loss": 0.20315276086330414, |
|
"eval_runtime": 1298.6234, |
|
"eval_samples_per_second": 8.805, |
|
"eval_steps_per_second": 1.101, |
|
"eval_wer": 0.26025807178678084, |
|
"step": 15800 |
|
}, |
|
{ |
|
"epoch": 1.49, |
|
"learning_rate": 0.0002546118543111024, |
|
"loss": 0.492, |
|
"step": 16000 |
|
}, |
|
{ |
|
"epoch": 1.49, |
|
"eval_loss": 0.19616416096687317, |
|
"eval_runtime": 1288.7568, |
|
"eval_samples_per_second": 8.873, |
|
"eval_steps_per_second": 1.11, |
|
"eval_wer": 0.25178340001126953, |
|
"step": 16000 |
|
}, |
|
{ |
|
"epoch": 1.51, |
|
"learning_rate": 0.000254025974025974, |
|
"loss": 0.4886, |
|
"step": 16200 |
|
}, |
|
{ |
|
"epoch": 1.51, |
|
"eval_loss": 0.1980104148387909, |
|
"eval_runtime": 1289.7166, |
|
"eval_samples_per_second": 8.866, |
|
"eval_steps_per_second": 1.109, |
|
"eval_wer": 0.24772637628895025, |
|
"step": 16200 |
|
}, |
|
{ |
|
"epoch": 1.52, |
|
"learning_rate": 0.0002534400937408456, |
|
"loss": 0.4847, |
|
"step": 16400 |
|
}, |
|
{ |
|
"epoch": 1.52, |
|
"eval_loss": 0.19249635934829712, |
|
"eval_runtime": 1251.9702, |
|
"eval_samples_per_second": 9.134, |
|
"eval_steps_per_second": 1.142, |
|
"eval_wer": 0.2432974587254184, |
|
"step": 16400 |
|
}, |
|
{ |
|
"epoch": 1.54, |
|
"learning_rate": 0.0002528542134557172, |
|
"loss": 0.4818, |
|
"step": 16600 |
|
}, |
|
{ |
|
"epoch": 1.54, |
|
"eval_loss": 0.1978258192539215, |
|
"eval_runtime": 1264.8774, |
|
"eval_samples_per_second": 9.04, |
|
"eval_steps_per_second": 1.131, |
|
"eval_wer": 0.25403730207922465, |
|
"step": 16600 |
|
}, |
|
{ |
|
"epoch": 1.56, |
|
"learning_rate": 0.00025226833317058876, |
|
"loss": 0.4823, |
|
"step": 16800 |
|
}, |
|
{ |
|
"epoch": 1.56, |
|
"eval_loss": 0.1943204402923584, |
|
"eval_runtime": 1276.7796, |
|
"eval_samples_per_second": 8.956, |
|
"eval_steps_per_second": 1.12, |
|
"eval_wer": 0.24585563757254747, |
|
"step": 16800 |
|
}, |
|
{ |
|
"epoch": 1.58, |
|
"learning_rate": 0.0002516824528854604, |
|
"loss": 0.4834, |
|
"step": 17000 |
|
}, |
|
{ |
|
"epoch": 1.58, |
|
"eval_loss": 0.19206224381923676, |
|
"eval_runtime": 1273.2063, |
|
"eval_samples_per_second": 8.981, |
|
"eval_steps_per_second": 1.123, |
|
"eval_wer": 0.24493153772468587, |
|
"step": 17000 |
|
}, |
|
{ |
|
"epoch": 1.6, |
|
"learning_rate": 0.000251096572600332, |
|
"loss": 0.4814, |
|
"step": 17200 |
|
}, |
|
{ |
|
"epoch": 1.6, |
|
"eval_loss": 0.18929176032543182, |
|
"eval_runtime": 1279.8788, |
|
"eval_samples_per_second": 8.934, |
|
"eval_steps_per_second": 1.117, |
|
"eval_wer": 0.2424973234912943, |
|
"step": 17200 |
|
}, |
|
{ |
|
"epoch": 1.62, |
|
"learning_rate": 0.00025051069231520355, |
|
"loss": 0.4961, |
|
"step": 17400 |
|
}, |
|
{ |
|
"epoch": 1.62, |
|
"eval_loss": 0.19354866445064545, |
|
"eval_runtime": 1283.269, |
|
"eval_samples_per_second": 8.911, |
|
"eval_steps_per_second": 1.114, |
|
"eval_wer": 0.2469037020341466, |
|
"step": 17400 |
|
}, |
|
{ |
|
"epoch": 1.64, |
|
"learning_rate": 0.0002499277414315008, |
|
"loss": 0.4899, |
|
"step": 17600 |
|
}, |
|
{ |
|
"epoch": 1.64, |
|
"eval_loss": 0.19293488562107086, |
|
"eval_runtime": 1281.2009, |
|
"eval_samples_per_second": 8.925, |
|
"eval_steps_per_second": 1.116, |
|
"eval_wer": 0.2515467402941342, |
|
"step": 17600 |
|
}, |
|
{ |
|
"epoch": 1.66, |
|
"learning_rate": 0.0002493418611463724, |
|
"loss": 0.4859, |
|
"step": 17800 |
|
}, |
|
{ |
|
"epoch": 1.66, |
|
"eval_loss": 0.1929776519536972, |
|
"eval_runtime": 1274.0955, |
|
"eval_samples_per_second": 8.975, |
|
"eval_steps_per_second": 1.122, |
|
"eval_wer": 0.2457429424691497, |
|
"step": 17800 |
|
}, |
|
{ |
|
"epoch": 1.68, |
|
"learning_rate": 0.000248755980861244, |
|
"loss": 0.4889, |
|
"step": 18000 |
|
}, |
|
{ |
|
"epoch": 1.68, |
|
"eval_loss": 0.18940649926662445, |
|
"eval_runtime": 1293.2691, |
|
"eval_samples_per_second": 8.842, |
|
"eval_steps_per_second": 1.106, |
|
"eval_wer": 0.24051388967149379, |
|
"step": 18000 |
|
}, |
|
{ |
|
"epoch": 1.7, |
|
"learning_rate": 0.0002481701005761156, |
|
"loss": 0.4718, |
|
"step": 18200 |
|
}, |
|
{ |
|
"epoch": 1.7, |
|
"eval_loss": 0.18651717901229858, |
|
"eval_runtime": 1283.6643, |
|
"eval_samples_per_second": 8.908, |
|
"eval_steps_per_second": 1.114, |
|
"eval_wer": 0.23402265171578296, |
|
"step": 18200 |
|
}, |
|
{ |
|
"epoch": 1.72, |
|
"learning_rate": 0.0002475842202909872, |
|
"loss": 0.4784, |
|
"step": 18400 |
|
}, |
|
{ |
|
"epoch": 1.72, |
|
"eval_loss": 0.18498806655406952, |
|
"eval_runtime": 1277.8059, |
|
"eval_samples_per_second": 8.949, |
|
"eval_steps_per_second": 1.119, |
|
"eval_wer": 0.23561165267369133, |
|
"step": 18400 |
|
}, |
|
{ |
|
"epoch": 1.74, |
|
"learning_rate": 0.00024699834000585876, |
|
"loss": 0.4843, |
|
"step": 18600 |
|
}, |
|
{ |
|
"epoch": 1.74, |
|
"eval_loss": 0.18745319545269012, |
|
"eval_runtime": 1301.5017, |
|
"eval_samples_per_second": 8.786, |
|
"eval_steps_per_second": 1.099, |
|
"eval_wer": 0.2299092804417648, |
|
"step": 18600 |
|
}, |
|
{ |
|
"epoch": 1.76, |
|
"learning_rate": 0.00024641538912215604, |
|
"loss": 0.4768, |
|
"step": 18800 |
|
}, |
|
{ |
|
"epoch": 1.76, |
|
"eval_loss": 0.18831437826156616, |
|
"eval_runtime": 1283.9645, |
|
"eval_samples_per_second": 8.906, |
|
"eval_steps_per_second": 1.114, |
|
"eval_wer": 0.23689637685242576, |
|
"step": 18800 |
|
}, |
|
{ |
|
"epoch": 1.78, |
|
"learning_rate": 0.0002458295088370276, |
|
"loss": 0.4599, |
|
"step": 19000 |
|
}, |
|
{ |
|
"epoch": 1.78, |
|
"eval_loss": 0.18521185219287872, |
|
"eval_runtime": 1294.064, |
|
"eval_samples_per_second": 8.837, |
|
"eval_steps_per_second": 1.105, |
|
"eval_wer": 0.2295373866005522, |
|
"step": 19000 |
|
}, |
|
{ |
|
"epoch": 1.8, |
|
"learning_rate": 0.0002452436285518992, |
|
"loss": 0.4944, |
|
"step": 19200 |
|
}, |
|
{ |
|
"epoch": 1.8, |
|
"eval_loss": 0.1821286529302597, |
|
"eval_runtime": 1279.5843, |
|
"eval_samples_per_second": 8.936, |
|
"eval_steps_per_second": 1.118, |
|
"eval_wer": 0.23201667887530286, |
|
"step": 19200 |
|
}, |
|
{ |
|
"epoch": 1.82, |
|
"learning_rate": 0.0002446577482667708, |
|
"loss": 0.4753, |
|
"step": 19400 |
|
}, |
|
{ |
|
"epoch": 1.82, |
|
"eval_loss": 0.18405307829380035, |
|
"eval_runtime": 1292.3701, |
|
"eval_samples_per_second": 8.848, |
|
"eval_steps_per_second": 1.106, |
|
"eval_wer": 0.23243365075787456, |
|
"step": 19400 |
|
}, |
|
{ |
|
"epoch": 1.84, |
|
"learning_rate": 0.0002440718679816424, |
|
"loss": 0.4675, |
|
"step": 19600 |
|
}, |
|
{ |
|
"epoch": 1.84, |
|
"eval_loss": 0.18435348570346832, |
|
"eval_runtime": 1281.9809, |
|
"eval_samples_per_second": 8.92, |
|
"eval_steps_per_second": 1.115, |
|
"eval_wer": 0.23523975883247872, |
|
"step": 19600 |
|
}, |
|
{ |
|
"epoch": 1.86, |
|
"learning_rate": 0.00024348598769651398, |
|
"loss": 0.4615, |
|
"step": 19800 |
|
}, |
|
{ |
|
"epoch": 1.86, |
|
"eval_loss": 0.17844724655151367, |
|
"eval_runtime": 1282.3895, |
|
"eval_samples_per_second": 8.917, |
|
"eval_steps_per_second": 1.115, |
|
"eval_wer": 0.2273511015946357, |
|
"step": 19800 |
|
}, |
|
{ |
|
"epoch": 1.87, |
|
"learning_rate": 0.00024290010741138558, |
|
"loss": 0.4723, |
|
"step": 20000 |
|
}, |
|
{ |
|
"epoch": 1.87, |
|
"eval_loss": 0.1748301386833191, |
|
"eval_runtime": 1290.0733, |
|
"eval_samples_per_second": 8.864, |
|
"eval_steps_per_second": 1.108, |
|
"eval_wer": 0.2285231306699724, |
|
"step": 20000 |
|
}, |
|
{ |
|
"epoch": 1.89, |
|
"learning_rate": 0.00024231422712625716, |
|
"loss": 0.4644, |
|
"step": 20200 |
|
}, |
|
{ |
|
"epoch": 1.89, |
|
"eval_loss": 0.17795561254024506, |
|
"eval_runtime": 1276.663, |
|
"eval_samples_per_second": 8.957, |
|
"eval_steps_per_second": 1.12, |
|
"eval_wer": 0.22774553445652787, |
|
"step": 20200 |
|
}, |
|
{ |
|
"epoch": 1.91, |
|
"learning_rate": 0.00024172834684112876, |
|
"loss": 0.4516, |
|
"step": 20400 |
|
}, |
|
{ |
|
"epoch": 1.91, |
|
"eval_loss": 0.18295042216777802, |
|
"eval_runtime": 1278.8615, |
|
"eval_samples_per_second": 8.942, |
|
"eval_steps_per_second": 1.118, |
|
"eval_wer": 0.23694145489378488, |
|
"step": 20400 |
|
}, |
|
{ |
|
"epoch": 1.93, |
|
"learning_rate": 0.0002411424665560004, |
|
"loss": 0.4656, |
|
"step": 20600 |
|
}, |
|
{ |
|
"epoch": 1.93, |
|
"eval_loss": 0.17588302493095398, |
|
"eval_runtime": 1272.4536, |
|
"eval_samples_per_second": 8.987, |
|
"eval_steps_per_second": 1.124, |
|
"eval_wer": 0.22738491012565504, |
|
"step": 20600 |
|
}, |
|
{ |
|
"epoch": 1.95, |
|
"learning_rate": 0.00024055658627087194, |
|
"loss": 0.4649, |
|
"step": 20800 |
|
}, |
|
{ |
|
"epoch": 1.95, |
|
"eval_loss": 0.1794005036354065, |
|
"eval_runtime": 1274.9253, |
|
"eval_samples_per_second": 8.969, |
|
"eval_steps_per_second": 1.122, |
|
"eval_wer": 0.22744125767735393, |
|
"step": 20800 |
|
}, |
|
{ |
|
"epoch": 2.02, |
|
"learning_rate": 0.0002399736353871692, |
|
"loss": 0.458, |
|
"step": 21000 |
|
}, |
|
{ |
|
"epoch": 2.02, |
|
"eval_loss": 0.18070010840892792, |
|
"eval_runtime": 1348.5976, |
|
"eval_samples_per_second": 8.479, |
|
"eval_steps_per_second": 1.06, |
|
"eval_wer": 0.23123908266185833, |
|
"step": 21000 |
|
}, |
|
{ |
|
"epoch": 2.04, |
|
"learning_rate": 0.0002393877551020408, |
|
"loss": 0.4605, |
|
"step": 21200 |
|
}, |
|
{ |
|
"epoch": 2.04, |
|
"eval_loss": 0.1788109391927719, |
|
"eval_runtime": 1335.4279, |
|
"eval_samples_per_second": 8.563, |
|
"eval_steps_per_second": 1.071, |
|
"eval_wer": 0.22942469149715444, |
|
"step": 21200 |
|
}, |
|
{ |
|
"epoch": 2.06, |
|
"learning_rate": 0.00023880187481691237, |
|
"loss": 0.446, |
|
"step": 21400 |
|
}, |
|
{ |
|
"epoch": 2.06, |
|
"eval_loss": 0.17524349689483643, |
|
"eval_runtime": 1337.1415, |
|
"eval_samples_per_second": 8.552, |
|
"eval_steps_per_second": 1.069, |
|
"eval_wer": 0.22274187186566743, |
|
"step": 21400 |
|
}, |
|
{ |
|
"epoch": 2.08, |
|
"learning_rate": 0.00023821599453178398, |
|
"loss": 0.4442, |
|
"step": 21600 |
|
}, |
|
{ |
|
"epoch": 2.08, |
|
"eval_loss": 0.16901686787605286, |
|
"eval_runtime": 1340.3966, |
|
"eval_samples_per_second": 8.531, |
|
"eval_steps_per_second": 1.067, |
|
"eval_wer": 0.21865103961232885, |
|
"step": 21600 |
|
}, |
|
{ |
|
"epoch": 2.1, |
|
"learning_rate": 0.00023763011424665558, |
|
"loss": 0.4557, |
|
"step": 21800 |
|
}, |
|
{ |
|
"epoch": 2.1, |
|
"eval_loss": 0.1691495031118393, |
|
"eval_runtime": 1342.8042, |
|
"eval_samples_per_second": 8.516, |
|
"eval_steps_per_second": 1.065, |
|
"eval_wer": 0.21790725192990365, |
|
"step": 21800 |
|
}, |
|
{ |
|
"epoch": 2.12, |
|
"learning_rate": 0.00023704423396152716, |
|
"loss": 0.4435, |
|
"step": 22000 |
|
}, |
|
{ |
|
"epoch": 2.12, |
|
"eval_loss": 0.17447154223918915, |
|
"eval_runtime": 1342.6406, |
|
"eval_samples_per_second": 8.517, |
|
"eval_steps_per_second": 1.065, |
|
"eval_wer": 0.21763678368174902, |
|
"step": 22000 |
|
}, |
|
{ |
|
"epoch": 2.14, |
|
"learning_rate": 0.00023645835367639876, |
|
"loss": 0.4427, |
|
"step": 22200 |
|
}, |
|
{ |
|
"epoch": 2.14, |
|
"eval_loss": 0.16724492609500885, |
|
"eval_runtime": 1348.4035, |
|
"eval_samples_per_second": 8.48, |
|
"eval_steps_per_second": 1.061, |
|
"eval_wer": 0.2169944215923818, |
|
"step": 22200 |
|
}, |
|
{ |
|
"epoch": 2.16, |
|
"learning_rate": 0.00023587247339127036, |
|
"loss": 0.4439, |
|
"step": 22400 |
|
}, |
|
{ |
|
"epoch": 2.16, |
|
"eval_loss": 0.16558879613876343, |
|
"eval_runtime": 1349.3597, |
|
"eval_samples_per_second": 8.474, |
|
"eval_steps_per_second": 1.06, |
|
"eval_wer": 0.2161153997858793, |
|
"step": 22400 |
|
}, |
|
{ |
|
"epoch": 2.17, |
|
"learning_rate": 0.00023528659310614194, |
|
"loss": 0.4416, |
|
"step": 22600 |
|
}, |
|
{ |
|
"epoch": 2.17, |
|
"eval_loss": 0.1659516543149948, |
|
"eval_runtime": 1345.2434, |
|
"eval_samples_per_second": 8.5, |
|
"eval_steps_per_second": 1.063, |
|
"eval_wer": 0.21102158111230068, |
|
"step": 22600 |
|
}, |
|
{ |
|
"epoch": 2.19, |
|
"learning_rate": 0.00023470071282101355, |
|
"loss": 0.434, |
|
"step": 22800 |
|
}, |
|
{ |
|
"epoch": 2.19, |
|
"eval_loss": 0.16529063880443573, |
|
"eval_runtime": 1343.6004, |
|
"eval_samples_per_second": 8.511, |
|
"eval_steps_per_second": 1.064, |
|
"eval_wer": 0.21015382881613795, |
|
"step": 22800 |
|
}, |
|
{ |
|
"epoch": 2.21, |
|
"learning_rate": 0.00023411483253588512, |
|
"loss": 0.4431, |
|
"step": 23000 |
|
}, |
|
{ |
|
"epoch": 2.21, |
|
"eval_loss": 0.16421248018741608, |
|
"eval_runtime": 1343.5764, |
|
"eval_samples_per_second": 8.511, |
|
"eval_steps_per_second": 1.064, |
|
"eval_wer": 0.2060742660731391, |
|
"step": 23000 |
|
}, |
|
{ |
|
"epoch": 2.23, |
|
"learning_rate": 0.00023352895225075675, |
|
"loss": 0.4328, |
|
"step": 23200 |
|
}, |
|
{ |
|
"epoch": 2.23, |
|
"eval_loss": 0.16530516743659973, |
|
"eval_runtime": 1340.835, |
|
"eval_samples_per_second": 8.528, |
|
"eval_steps_per_second": 1.066, |
|
"eval_wer": 0.2064123513833324, |
|
"step": 23200 |
|
}, |
|
{ |
|
"epoch": 2.25, |
|
"learning_rate": 0.00023294307196562836, |
|
"loss": 0.4291, |
|
"step": 23400 |
|
}, |
|
{ |
|
"epoch": 2.25, |
|
"eval_loss": 0.16716133058071136, |
|
"eval_runtime": 1349.4559, |
|
"eval_samples_per_second": 8.474, |
|
"eval_steps_per_second": 1.06, |
|
"eval_wer": 0.20953400574745026, |
|
"step": 23400 |
|
}, |
|
{ |
|
"epoch": 2.27, |
|
"learning_rate": 0.00023235719168049993, |
|
"loss": 0.4249, |
|
"step": 23600 |
|
}, |
|
{ |
|
"epoch": 2.27, |
|
"eval_loss": 0.1660962998867035, |
|
"eval_runtime": 1342.5077, |
|
"eval_samples_per_second": 8.518, |
|
"eval_steps_per_second": 1.065, |
|
"eval_wer": 0.20847467177551135, |
|
"step": 23600 |
|
}, |
|
{ |
|
"epoch": 2.29, |
|
"learning_rate": 0.00023177131139537154, |
|
"loss": 0.4476, |
|
"step": 23800 |
|
}, |
|
{ |
|
"epoch": 2.29, |
|
"eval_loss": 0.167561873793602, |
|
"eval_runtime": 1343.4696, |
|
"eval_samples_per_second": 8.512, |
|
"eval_steps_per_second": 1.064, |
|
"eval_wer": 0.21436862568321408, |
|
"step": 23800 |
|
}, |
|
{ |
|
"epoch": 2.31, |
|
"learning_rate": 0.00023118543111024314, |
|
"loss": 0.4246, |
|
"step": 24000 |
|
}, |
|
{ |
|
"epoch": 2.31, |
|
"eval_loss": 0.1651725172996521, |
|
"eval_runtime": 1345.4703, |
|
"eval_samples_per_second": 8.499, |
|
"eval_steps_per_second": 1.063, |
|
"eval_wer": 0.21284724178734435, |
|
"step": 24000 |
|
}, |
|
{ |
|
"epoch": 2.33, |
|
"learning_rate": 0.00023060248022654034, |
|
"loss": 0.445, |
|
"step": 24200 |
|
}, |
|
{ |
|
"epoch": 2.33, |
|
"eval_loss": 0.16498707234859467, |
|
"eval_runtime": 1347.0845, |
|
"eval_samples_per_second": 8.489, |
|
"eval_steps_per_second": 1.062, |
|
"eval_wer": 0.2074716853552713, |
|
"step": 24200 |
|
}, |
|
{ |
|
"epoch": 2.35, |
|
"learning_rate": 0.00023001659994141194, |
|
"loss": 0.4303, |
|
"step": 24400 |
|
}, |
|
{ |
|
"epoch": 2.35, |
|
"eval_loss": 0.15905718505382538, |
|
"eval_runtime": 1346.9157, |
|
"eval_samples_per_second": 8.49, |
|
"eval_steps_per_second": 1.062, |
|
"eval_wer": 0.20397813714994084, |
|
"step": 24400 |
|
}, |
|
{ |
|
"epoch": 2.37, |
|
"learning_rate": 0.00022943071965628354, |
|
"loss": 0.4194, |
|
"step": 24600 |
|
}, |
|
{ |
|
"epoch": 2.37, |
|
"eval_loss": 0.1634632647037506, |
|
"eval_runtime": 1345.7282, |
|
"eval_samples_per_second": 8.497, |
|
"eval_steps_per_second": 1.063, |
|
"eval_wer": 0.20863244492026822, |
|
"step": 24600 |
|
}, |
|
{ |
|
"epoch": 2.39, |
|
"learning_rate": 0.00022884483937115512, |
|
"loss": 0.4211, |
|
"step": 24800 |
|
}, |
|
{ |
|
"epoch": 2.39, |
|
"eval_loss": 0.16573943197727203, |
|
"eval_runtime": 1346.4203, |
|
"eval_samples_per_second": 8.493, |
|
"eval_steps_per_second": 1.062, |
|
"eval_wer": 0.21404180988336058, |
|
"step": 24800 |
|
}, |
|
{ |
|
"epoch": 2.41, |
|
"learning_rate": 0.00022825895908602673, |
|
"loss": 0.4238, |
|
"step": 25000 |
|
}, |
|
{ |
|
"epoch": 2.41, |
|
"eval_loss": 0.16177457571029663, |
|
"eval_runtime": 1353.3476, |
|
"eval_samples_per_second": 8.449, |
|
"eval_steps_per_second": 1.057, |
|
"eval_wer": 0.20668281963148702, |
|
"step": 25000 |
|
}, |
|
{ |
|
"epoch": 2.43, |
|
"learning_rate": 0.00022767307880089836, |
|
"loss": 0.4229, |
|
"step": 25200 |
|
}, |
|
{ |
|
"epoch": 2.43, |
|
"eval_loss": 0.1606961190700531, |
|
"eval_runtime": 1347.2789, |
|
"eval_samples_per_second": 8.487, |
|
"eval_steps_per_second": 1.061, |
|
"eval_wer": 0.20163407899926747, |
|
"step": 25200 |
|
}, |
|
{ |
|
"epoch": 2.45, |
|
"learning_rate": 0.0002270871985157699, |
|
"loss": 0.434, |
|
"step": 25400 |
|
}, |
|
{ |
|
"epoch": 2.45, |
|
"eval_loss": 0.1586761772632599, |
|
"eval_runtime": 1345.6532, |
|
"eval_samples_per_second": 8.498, |
|
"eval_steps_per_second": 1.063, |
|
"eval_wer": 0.19963937566912718, |
|
"step": 25400 |
|
}, |
|
{ |
|
"epoch": 2.47, |
|
"learning_rate": 0.00022650131823064154, |
|
"loss": 0.4232, |
|
"step": 25600 |
|
}, |
|
{ |
|
"epoch": 2.47, |
|
"eval_loss": 0.15713459253311157, |
|
"eval_runtime": 1346.3424, |
|
"eval_samples_per_second": 8.493, |
|
"eval_steps_per_second": 1.062, |
|
"eval_wer": 0.20052966698596947, |
|
"step": 25600 |
|
}, |
|
{ |
|
"epoch": 2.49, |
|
"learning_rate": 0.0002259154379455131, |
|
"loss": 0.4132, |
|
"step": 25800 |
|
}, |
|
{ |
|
"epoch": 2.49, |
|
"eval_loss": 0.15517787635326385, |
|
"eval_runtime": 1347.0293, |
|
"eval_samples_per_second": 8.489, |
|
"eval_steps_per_second": 1.062, |
|
"eval_wer": 0.2024680227644109, |
|
"step": 25800 |
|
}, |
|
{ |
|
"epoch": 2.51, |
|
"learning_rate": 0.00022532955766038472, |
|
"loss": 0.4239, |
|
"step": 26000 |
|
}, |
|
{ |
|
"epoch": 2.51, |
|
"eval_loss": 0.15804125368595123, |
|
"eval_runtime": 1349.3993, |
|
"eval_samples_per_second": 8.474, |
|
"eval_steps_per_second": 1.06, |
|
"eval_wer": 0.20016904265509664, |
|
"step": 26000 |
|
}, |
|
{ |
|
"epoch": 2.52, |
|
"learning_rate": 0.00022474367737525632, |
|
"loss": 0.4097, |
|
"step": 26200 |
|
}, |
|
{ |
|
"epoch": 2.52, |
|
"eval_loss": 0.1611405611038208, |
|
"eval_runtime": 1344.7866, |
|
"eval_samples_per_second": 8.503, |
|
"eval_steps_per_second": 1.063, |
|
"eval_wer": 0.20251310080576998, |
|
"step": 26200 |
|
}, |
|
{ |
|
"epoch": 2.54, |
|
"learning_rate": 0.00022416072649155354, |
|
"loss": 0.3975, |
|
"step": 26400 |
|
}, |
|
{ |
|
"epoch": 2.54, |
|
"eval_loss": 0.15802615880966187, |
|
"eval_runtime": 1349.3436, |
|
"eval_samples_per_second": 8.474, |
|
"eval_steps_per_second": 1.06, |
|
"eval_wer": 0.20389925057756242, |
|
"step": 26400 |
|
}, |
|
{ |
|
"epoch": 2.56, |
|
"learning_rate": 0.00022357484620642512, |
|
"loss": 0.418, |
|
"step": 26600 |
|
}, |
|
{ |
|
"epoch": 2.56, |
|
"eval_loss": 0.1567668914794922, |
|
"eval_runtime": 1346.9229, |
|
"eval_samples_per_second": 8.49, |
|
"eval_steps_per_second": 1.062, |
|
"eval_wer": 0.20108187299261848, |
|
"step": 26600 |
|
}, |
|
{ |
|
"epoch": 2.58, |
|
"learning_rate": 0.00022298896592129672, |
|
"loss": 0.4126, |
|
"step": 26800 |
|
}, |
|
{ |
|
"epoch": 2.58, |
|
"eval_loss": 0.15674176812171936, |
|
"eval_runtime": 1347.6959, |
|
"eval_samples_per_second": 8.485, |
|
"eval_steps_per_second": 1.061, |
|
"eval_wer": 0.19748689919423001, |
|
"step": 26800 |
|
}, |
|
{ |
|
"epoch": 2.6, |
|
"learning_rate": 0.0002224030856361683, |
|
"loss": 0.4227, |
|
"step": 27000 |
|
}, |
|
{ |
|
"epoch": 2.6, |
|
"eval_loss": 0.15419313311576843, |
|
"eval_runtime": 1347.3912, |
|
"eval_samples_per_second": 8.487, |
|
"eval_steps_per_second": 1.061, |
|
"eval_wer": 0.19683326759452302, |
|
"step": 27000 |
|
}, |
|
{ |
|
"epoch": 2.62, |
|
"learning_rate": 0.0002218172053510399, |
|
"loss": 0.4176, |
|
"step": 27200 |
|
}, |
|
{ |
|
"epoch": 2.62, |
|
"eval_loss": 0.1559024453163147, |
|
"eval_runtime": 1346.8455, |
|
"eval_samples_per_second": 8.49, |
|
"eval_steps_per_second": 1.062, |
|
"eval_wer": 0.19721643094607538, |
|
"step": 27200 |
|
}, |
|
{ |
|
"epoch": 2.64, |
|
"learning_rate": 0.00022123132506591154, |
|
"loss": 0.4158, |
|
"step": 27400 |
|
}, |
|
{ |
|
"epoch": 2.64, |
|
"eval_loss": 0.15258029103279114, |
|
"eval_runtime": 1345.1608, |
|
"eval_samples_per_second": 8.501, |
|
"eval_steps_per_second": 1.063, |
|
"eval_wer": 0.1984673465937905, |
|
"step": 27400 |
|
}, |
|
{ |
|
"epoch": 2.66, |
|
"learning_rate": 0.00022064544478078309, |
|
"loss": 0.4064, |
|
"step": 27600 |
|
}, |
|
{ |
|
"epoch": 2.66, |
|
"eval_loss": 0.1543867439031601, |
|
"eval_runtime": 1350.12, |
|
"eval_samples_per_second": 8.47, |
|
"eval_steps_per_second": 1.059, |
|
"eval_wer": 0.19683326759452302, |
|
"step": 27600 |
|
}, |
|
{ |
|
"epoch": 2.68, |
|
"learning_rate": 0.00022005956449565472, |
|
"loss": 0.4054, |
|
"step": 27800 |
|
}, |
|
{ |
|
"epoch": 2.68, |
|
"eval_loss": 0.1523013412952423, |
|
"eval_runtime": 1347.6703, |
|
"eval_samples_per_second": 8.485, |
|
"eval_steps_per_second": 1.061, |
|
"eval_wer": 0.19182960500366258, |
|
"step": 27800 |
|
}, |
|
{ |
|
"epoch": 2.7, |
|
"learning_rate": 0.00021947368421052632, |
|
"loss": 0.4046, |
|
"step": 28000 |
|
}, |
|
{ |
|
"epoch": 2.7, |
|
"eval_loss": 0.15553198754787445, |
|
"eval_runtime": 1346.2686, |
|
"eval_samples_per_second": 8.494, |
|
"eval_steps_per_second": 1.062, |
|
"eval_wer": 0.19630360060855356, |
|
"step": 28000 |
|
}, |
|
{ |
|
"epoch": 2.72, |
|
"learning_rate": 0.0002188878039253979, |
|
"loss": 0.4052, |
|
"step": 28200 |
|
}, |
|
{ |
|
"epoch": 2.72, |
|
"eval_loss": 0.15202964842319489, |
|
"eval_runtime": 1350.4067, |
|
"eval_samples_per_second": 8.468, |
|
"eval_steps_per_second": 1.059, |
|
"eval_wer": 0.19546965684341017, |
|
"step": 28200 |
|
}, |
|
{ |
|
"epoch": 2.74, |
|
"learning_rate": 0.0002183019236402695, |
|
"loss": 0.4139, |
|
"step": 28400 |
|
}, |
|
{ |
|
"epoch": 2.74, |
|
"eval_loss": 0.1537465751171112, |
|
"eval_runtime": 1348.6125, |
|
"eval_samples_per_second": 8.479, |
|
"eval_steps_per_second": 1.06, |
|
"eval_wer": 0.19421874119569504, |
|
"step": 28400 |
|
}, |
|
{ |
|
"epoch": 2.76, |
|
"learning_rate": 0.00021771604335514108, |
|
"loss": 0.4043, |
|
"step": 28600 |
|
}, |
|
{ |
|
"epoch": 2.76, |
|
"eval_loss": 0.15482832491397858, |
|
"eval_runtime": 1353.2391, |
|
"eval_samples_per_second": 8.45, |
|
"eval_steps_per_second": 1.057, |
|
"eval_wer": 0.1914013636107511, |
|
"step": 28600 |
|
}, |
|
{ |
|
"epoch": 2.78, |
|
"learning_rate": 0.00021713016307001268, |
|
"loss": 0.4028, |
|
"step": 28800 |
|
}, |
|
{ |
|
"epoch": 2.78, |
|
"eval_loss": 0.15318334102630615, |
|
"eval_runtime": 1349.2472, |
|
"eval_samples_per_second": 8.475, |
|
"eval_steps_per_second": 1.06, |
|
"eval_wer": 0.19257339268608778, |
|
"step": 28800 |
|
}, |
|
{ |
|
"epoch": 2.8, |
|
"learning_rate": 0.0002165442827848843, |
|
"loss": 0.4195, |
|
"step": 29000 |
|
}, |
|
{ |
|
"epoch": 2.8, |
|
"eval_loss": 0.15186063945293427, |
|
"eval_runtime": 1355.3879, |
|
"eval_samples_per_second": 8.437, |
|
"eval_steps_per_second": 1.055, |
|
"eval_wer": 0.19338479743055165, |
|
"step": 29000 |
|
}, |
|
{ |
|
"epoch": 2.82, |
|
"learning_rate": 0.00021595840249975586, |
|
"loss": 0.4086, |
|
"step": 29200 |
|
}, |
|
{ |
|
"epoch": 2.82, |
|
"eval_loss": 0.1497160643339157, |
|
"eval_runtime": 1351.97, |
|
"eval_samples_per_second": 8.458, |
|
"eval_steps_per_second": 1.058, |
|
"eval_wer": 0.19105200879021805, |
|
"step": 29200 |
|
}, |
|
{ |
|
"epoch": 2.84, |
|
"learning_rate": 0.00021537252221462747, |
|
"loss": 0.39, |
|
"step": 29400 |
|
}, |
|
{ |
|
"epoch": 2.84, |
|
"eval_loss": 0.1531924307346344, |
|
"eval_runtime": 1349.4993, |
|
"eval_samples_per_second": 8.474, |
|
"eval_steps_per_second": 1.06, |
|
"eval_wer": 0.1928889389756015, |
|
"step": 29400 |
|
}, |
|
{ |
|
"epoch": 2.86, |
|
"learning_rate": 0.00021478957133092472, |
|
"loss": 0.3936, |
|
"step": 29600 |
|
}, |
|
{ |
|
"epoch": 2.86, |
|
"eval_loss": 0.14919565618038177, |
|
"eval_runtime": 1346.555, |
|
"eval_samples_per_second": 8.492, |
|
"eval_steps_per_second": 1.062, |
|
"eval_wer": 0.18807685806051727, |
|
"step": 29600 |
|
}, |
|
{ |
|
"epoch": 2.87, |
|
"learning_rate": 0.00021420369104579627, |
|
"loss": 0.4065, |
|
"step": 29800 |
|
}, |
|
{ |
|
"epoch": 2.87, |
|
"eval_loss": 0.14806747436523438, |
|
"eval_runtime": 1359.5314, |
|
"eval_samples_per_second": 8.411, |
|
"eval_steps_per_second": 1.052, |
|
"eval_wer": 0.187738772750324, |
|
"step": 29800 |
|
}, |
|
{ |
|
"epoch": 2.89, |
|
"learning_rate": 0.0002136178107606679, |
|
"loss": 0.4047, |
|
"step": 30000 |
|
}, |
|
{ |
|
"epoch": 2.89, |
|
"eval_loss": 0.1506689190864563, |
|
"eval_runtime": 1347.9533, |
|
"eval_samples_per_second": 8.483, |
|
"eval_steps_per_second": 1.061, |
|
"eval_wer": 0.192764974361864, |
|
"step": 30000 |
|
}, |
|
{ |
|
"epoch": 2.02, |
|
"learning_rate": 0.0002130319304755395, |
|
"loss": 0.3972, |
|
"step": 30200 |
|
}, |
|
{ |
|
"epoch": 2.02, |
|
"eval_loss": 0.14971239864826202, |
|
"eval_runtime": 1295.4974, |
|
"eval_samples_per_second": 8.827, |
|
"eval_steps_per_second": 1.104, |
|
"eval_wer": 0.18893334084634023, |
|
"step": 30200 |
|
}, |
|
{ |
|
"epoch": 2.04, |
|
"learning_rate": 0.00021244605019041108, |
|
"loss": 0.3955, |
|
"step": 30400 |
|
}, |
|
{ |
|
"epoch": 2.04, |
|
"eval_loss": 0.15231847763061523, |
|
"eval_runtime": 1286.7728, |
|
"eval_samples_per_second": 8.887, |
|
"eval_steps_per_second": 1.111, |
|
"eval_wer": 0.1934974925339494, |
|
"step": 30400 |
|
}, |
|
{ |
|
"epoch": 2.06, |
|
"learning_rate": 0.00021186016990528268, |
|
"loss": 0.3864, |
|
"step": 30600 |
|
}, |
|
{ |
|
"epoch": 2.06, |
|
"eval_loss": 0.14847034215927124, |
|
"eval_runtime": 1289.1638, |
|
"eval_samples_per_second": 8.87, |
|
"eval_steps_per_second": 1.109, |
|
"eval_wer": 0.1906688454386657, |
|
"step": 30600 |
|
}, |
|
{ |
|
"epoch": 2.08, |
|
"learning_rate": 0.00021127428962015426, |
|
"loss": 0.3847, |
|
"step": 30800 |
|
}, |
|
{ |
|
"epoch": 2.08, |
|
"eval_loss": 0.14542284607887268, |
|
"eval_runtime": 1293.2483, |
|
"eval_samples_per_second": 8.842, |
|
"eval_steps_per_second": 1.106, |
|
"eval_wer": 0.18785146785372175, |
|
"step": 30800 |
|
}, |
|
{ |
|
"epoch": 2.1, |
|
"learning_rate": 0.00021068840933502586, |
|
"loss": 0.3971, |
|
"step": 31000 |
|
}, |
|
{ |
|
"epoch": 2.1, |
|
"eval_loss": 0.14737120270729065, |
|
"eval_runtime": 1285.34, |
|
"eval_samples_per_second": 8.896, |
|
"eval_steps_per_second": 1.113, |
|
"eval_wer": 0.18689355947484082, |
|
"step": 31000 |
|
}, |
|
{ |
|
"epoch": 2.12, |
|
"learning_rate": 0.00021010252904989747, |
|
"loss": 0.3945, |
|
"step": 31200 |
|
}, |
|
{ |
|
"epoch": 2.12, |
|
"eval_loss": 0.14705486595630646, |
|
"eval_runtime": 1285.8175, |
|
"eval_samples_per_second": 8.893, |
|
"eval_steps_per_second": 1.112, |
|
"eval_wer": 0.18310700400067617, |
|
"step": 31200 |
|
}, |
|
{ |
|
"epoch": 2.14, |
|
"learning_rate": 0.00020951664876476904, |
|
"loss": 0.3921, |
|
"step": 31400 |
|
}, |
|
{ |
|
"epoch": 2.14, |
|
"eval_loss": 0.14592072367668152, |
|
"eval_runtime": 1297.0191, |
|
"eval_samples_per_second": 8.816, |
|
"eval_steps_per_second": 1.103, |
|
"eval_wer": 0.18582295599256213, |
|
"step": 31400 |
|
}, |
|
{ |
|
"epoch": 2.16, |
|
"learning_rate": 0.00020893076847964065, |
|
"loss": 0.3884, |
|
"step": 31600 |
|
}, |
|
{ |
|
"epoch": 2.16, |
|
"eval_loss": 0.14477235078811646, |
|
"eval_runtime": 1294.3727, |
|
"eval_samples_per_second": 8.834, |
|
"eval_steps_per_second": 1.105, |
|
"eval_wer": 0.18609342424071673, |
|
"step": 31600 |
|
}, |
|
{ |
|
"epoch": 2.17, |
|
"learning_rate": 0.00020834488819451225, |
|
"loss": 0.3834, |
|
"step": 31800 |
|
}, |
|
{ |
|
"epoch": 2.17, |
|
"eval_loss": 0.14469768106937408, |
|
"eval_runtime": 1297.5153, |
|
"eval_samples_per_second": 8.813, |
|
"eval_steps_per_second": 1.102, |
|
"eval_wer": 0.18483123908266186, |
|
"step": 31800 |
|
}, |
|
{ |
|
"epoch": 2.19, |
|
"learning_rate": 0.00020776193731080944, |
|
"loss": 0.382, |
|
"step": 32000 |
|
}, |
|
{ |
|
"epoch": 2.19, |
|
"eval_loss": 0.14314468204975128, |
|
"eval_runtime": 1293.9163, |
|
"eval_samples_per_second": 8.838, |
|
"eval_steps_per_second": 1.105, |
|
"eval_wer": 0.17969234236772413, |
|
"step": 32000 |
|
}, |
|
{ |
|
"epoch": 2.21, |
|
"learning_rate": 0.00020717605702568108, |
|
"loss": 0.3847, |
|
"step": 32200 |
|
}, |
|
{ |
|
"epoch": 2.21, |
|
"eval_loss": 0.14419673383235931, |
|
"eval_runtime": 1295.5312, |
|
"eval_samples_per_second": 8.826, |
|
"eval_steps_per_second": 1.104, |
|
"eval_wer": 0.18103341409815743, |
|
"step": 32200 |
|
}, |
|
{ |
|
"epoch": 2.23, |
|
"learning_rate": 0.0002065931061419783, |
|
"loss": 0.3824, |
|
"step": 32400 |
|
}, |
|
{ |
|
"epoch": 2.23, |
|
"eval_loss": 0.14198584854602814, |
|
"eval_runtime": 1291.5798, |
|
"eval_samples_per_second": 8.853, |
|
"eval_steps_per_second": 1.107, |
|
"eval_wer": 0.1794218741195695, |
|
"step": 32400 |
|
}, |
|
{ |
|
"epoch": 2.25, |
|
"learning_rate": 0.0002060072258568499, |
|
"loss": 0.375, |
|
"step": 32600 |
|
}, |
|
{ |
|
"epoch": 2.25, |
|
"eval_loss": 0.1454899162054062, |
|
"eval_runtime": 1296.0759, |
|
"eval_samples_per_second": 8.823, |
|
"eval_steps_per_second": 1.103, |
|
"eval_wer": 0.1824984504423283, |
|
"step": 32600 |
|
}, |
|
{ |
|
"epoch": 2.27, |
|
"learning_rate": 0.00020542134557172148, |
|
"loss": 0.3649, |
|
"step": 32800 |
|
}, |
|
{ |
|
"epoch": 2.27, |
|
"eval_loss": 0.14184387028217316, |
|
"eval_runtime": 1293.9484, |
|
"eval_samples_per_second": 8.837, |
|
"eval_steps_per_second": 1.105, |
|
"eval_wer": 0.17908378880937623, |
|
"step": 32800 |
|
}, |
|
{ |
|
"epoch": 2.29, |
|
"learning_rate": 0.00020483546528659308, |
|
"loss": 0.3835, |
|
"step": 33000 |
|
}, |
|
{ |
|
"epoch": 2.29, |
|
"eval_loss": 0.14400140941143036, |
|
"eval_runtime": 1291.8948, |
|
"eval_samples_per_second": 8.851, |
|
"eval_steps_per_second": 1.107, |
|
"eval_wer": 0.18164196765650534, |
|
"step": 33000 |
|
}, |
|
{ |
|
"epoch": 2.31, |
|
"learning_rate": 0.00020424958500146466, |
|
"loss": 0.3708, |
|
"step": 33200 |
|
}, |
|
{ |
|
"epoch": 2.31, |
|
"eval_loss": 0.14257806539535522, |
|
"eval_runtime": 1287.163, |
|
"eval_samples_per_second": 8.884, |
|
"eval_steps_per_second": 1.111, |
|
"eval_wer": 0.181055953118837, |
|
"step": 33200 |
|
}, |
|
{ |
|
"epoch": 2.33, |
|
"learning_rate": 0.00020366370471633626, |
|
"loss": 0.3903, |
|
"step": 33400 |
|
}, |
|
{ |
|
"epoch": 2.33, |
|
"eval_loss": 0.14411504566669464, |
|
"eval_runtime": 1294.8904, |
|
"eval_samples_per_second": 8.831, |
|
"eval_steps_per_second": 1.104, |
|
"eval_wer": 0.18088691046374036, |
|
"step": 33400 |
|
}, |
|
{ |
|
"epoch": 2.35, |
|
"learning_rate": 0.0002030778244312079, |
|
"loss": 0.3901, |
|
"step": 33600 |
|
}, |
|
{ |
|
"epoch": 2.35, |
|
"eval_loss": 0.14472506940364838, |
|
"eval_runtime": 1287.7937, |
|
"eval_samples_per_second": 8.88, |
|
"eval_steps_per_second": 1.11, |
|
"eval_wer": 0.18023327886403337, |
|
"step": 33600 |
|
}, |
|
{ |
|
"epoch": 2.37, |
|
"learning_rate": 0.00020249194414607944, |
|
"loss": 0.372, |
|
"step": 33800 |
|
}, |
|
{ |
|
"epoch": 2.37, |
|
"eval_loss": 0.14397132396697998, |
|
"eval_runtime": 1288.3131, |
|
"eval_samples_per_second": 8.876, |
|
"eval_steps_per_second": 1.11, |
|
"eval_wer": 0.17937679607821042, |
|
"step": 33800 |
|
}, |
|
{ |
|
"epoch": 2.39, |
|
"learning_rate": 0.00020190606386095107, |
|
"loss": 0.373, |
|
"step": 34000 |
|
}, |
|
{ |
|
"epoch": 2.39, |
|
"eval_loss": 0.14335143566131592, |
|
"eval_runtime": 1294.3533, |
|
"eval_samples_per_second": 8.835, |
|
"eval_steps_per_second": 1.105, |
|
"eval_wer": 0.17952329971262748, |
|
"step": 34000 |
|
}, |
|
{ |
|
"epoch": 2.41, |
|
"learning_rate": 0.00020132018357582268, |
|
"loss": 0.382, |
|
"step": 34200 |
|
}, |
|
{ |
|
"epoch": 2.41, |
|
"eval_loss": 0.14260776340961456, |
|
"eval_runtime": 1294.0006, |
|
"eval_samples_per_second": 8.837, |
|
"eval_steps_per_second": 1.105, |
|
"eval_wer": 0.17981630698146164, |
|
"step": 34200 |
|
}, |
|
{ |
|
"epoch": 2.43, |
|
"learning_rate": 0.00020073430329069426, |
|
"loss": 0.3735, |
|
"step": 34400 |
|
}, |
|
{ |
|
"epoch": 2.43, |
|
"eval_loss": 0.14174862205982208, |
|
"eval_runtime": 1291.0684, |
|
"eval_samples_per_second": 8.857, |
|
"eval_steps_per_second": 1.108, |
|
"eval_wer": 0.17719051107229392, |
|
"step": 34400 |
|
}, |
|
{ |
|
"epoch": 2.45, |
|
"learning_rate": 0.00020014842300556586, |
|
"loss": 0.3919, |
|
"step": 34600 |
|
}, |
|
{ |
|
"epoch": 2.45, |
|
"eval_loss": 0.14172929525375366, |
|
"eval_runtime": 1295.6993, |
|
"eval_samples_per_second": 8.825, |
|
"eval_steps_per_second": 1.104, |
|
"eval_wer": 0.18066152025694485, |
|
"step": 34600 |
|
}, |
|
{ |
|
"epoch": 2.47, |
|
"learning_rate": 0.00019956254272043744, |
|
"loss": 0.3694, |
|
"step": 34800 |
|
}, |
|
{ |
|
"epoch": 2.47, |
|
"eval_loss": 0.1413334310054779, |
|
"eval_runtime": 1291.4925, |
|
"eval_samples_per_second": 8.854, |
|
"eval_steps_per_second": 1.107, |
|
"eval_wer": 0.1777990646306418, |
|
"step": 34800 |
|
}, |
|
{ |
|
"epoch": 2.49, |
|
"learning_rate": 0.00019897666243530904, |
|
"loss": 0.3747, |
|
"step": 35000 |
|
}, |
|
{ |
|
"epoch": 2.49, |
|
"eval_loss": 0.14046485722064972, |
|
"eval_runtime": 1291.2845, |
|
"eval_samples_per_second": 8.856, |
|
"eval_steps_per_second": 1.107, |
|
"eval_wer": 0.1763565673071505, |
|
"step": 35000 |
|
}, |
|
{ |
|
"epoch": 2.51, |
|
"learning_rate": 0.00019839078215018064, |
|
"loss": 0.3728, |
|
"step": 35200 |
|
}, |
|
{ |
|
"epoch": 2.51, |
|
"eval_loss": 0.14019745588302612, |
|
"eval_runtime": 1277.9327, |
|
"eval_samples_per_second": 8.948, |
|
"eval_steps_per_second": 1.119, |
|
"eval_wer": 0.17606356003831633, |
|
"step": 35200 |
|
}, |
|
{ |
|
"epoch": 2.52, |
|
"learning_rate": 0.00019780490186505222, |
|
"loss": 0.3604, |
|
"step": 35400 |
|
}, |
|
{ |
|
"epoch": 2.52, |
|
"eval_loss": 0.14010049402713776, |
|
"eval_runtime": 1269.2145, |
|
"eval_samples_per_second": 9.01, |
|
"eval_steps_per_second": 1.127, |
|
"eval_wer": 0.1762551417140925, |
|
"step": 35400 |
|
}, |
|
{ |
|
"epoch": 2.54, |
|
"learning_rate": 0.00019721902157992383, |
|
"loss": 0.354, |
|
"step": 35600 |
|
}, |
|
{ |
|
"epoch": 2.54, |
|
"eval_loss": 0.13699594140052795, |
|
"eval_runtime": 1266.9571, |
|
"eval_samples_per_second": 9.026, |
|
"eval_steps_per_second": 1.129, |
|
"eval_wer": 0.17405758719783626, |
|
"step": 35600 |
|
}, |
|
{ |
|
"epoch": 2.56, |
|
"learning_rate": 0.00019663314129479543, |
|
"loss": 0.3731, |
|
"step": 35800 |
|
}, |
|
{ |
|
"epoch": 2.56, |
|
"eval_loss": 0.137326180934906, |
|
"eval_runtime": 1281.6045, |
|
"eval_samples_per_second": 8.922, |
|
"eval_steps_per_second": 1.116, |
|
"eval_wer": 0.1728292105708007, |
|
"step": 35800 |
|
}, |
|
{ |
|
"epoch": 2.58, |
|
"learning_rate": 0.000196047261009667, |
|
"loss": 0.3704, |
|
"step": 36000 |
|
}, |
|
{ |
|
"epoch": 2.58, |
|
"eval_loss": 0.14074836671352386, |
|
"eval_runtime": 1278.0727, |
|
"eval_samples_per_second": 8.947, |
|
"eval_steps_per_second": 1.119, |
|
"eval_wer": 0.17788922071336, |
|
"step": 36000 |
|
}, |
|
{ |
|
"epoch": 2.6, |
|
"learning_rate": 0.00019546431012596425, |
|
"loss": 0.3737, |
|
"step": 36200 |
|
}, |
|
{ |
|
"epoch": 2.6, |
|
"eval_loss": 0.1350381225347519, |
|
"eval_runtime": 1289.4945, |
|
"eval_samples_per_second": 8.868, |
|
"eval_steps_per_second": 1.109, |
|
"eval_wer": 0.1738547360117203, |
|
"step": 36200 |
|
}, |
|
{ |
|
"epoch": 2.62, |
|
"learning_rate": 0.00019487842984083586, |
|
"loss": 0.3638, |
|
"step": 36400 |
|
}, |
|
{ |
|
"epoch": 2.62, |
|
"eval_loss": 0.13623464107513428, |
|
"eval_runtime": 1306.2961, |
|
"eval_samples_per_second": 8.754, |
|
"eval_steps_per_second": 1.095, |
|
"eval_wer": 0.1758494393418606, |
|
"step": 36400 |
|
}, |
|
{ |
|
"epoch": 2.64, |
|
"learning_rate": 0.00019429254955570743, |
|
"loss": 0.3726, |
|
"step": 36600 |
|
}, |
|
{ |
|
"epoch": 2.64, |
|
"eval_loss": 0.1357557475566864, |
|
"eval_runtime": 1277.9297, |
|
"eval_samples_per_second": 8.948, |
|
"eval_steps_per_second": 1.119, |
|
"eval_wer": 0.17244604721924833, |
|
"step": 36600 |
|
}, |
|
{ |
|
"epoch": 2.66, |
|
"learning_rate": 0.00019370666927057904, |
|
"loss": 0.362, |
|
"step": 36800 |
|
}, |
|
{ |
|
"epoch": 2.66, |
|
"eval_loss": 0.136088564991951, |
|
"eval_runtime": 1273.2522, |
|
"eval_samples_per_second": 8.981, |
|
"eval_steps_per_second": 1.123, |
|
"eval_wer": 0.17147686933002762, |
|
"step": 36800 |
|
}, |
|
{ |
|
"epoch": 2.68, |
|
"learning_rate": 0.00019312078898545064, |
|
"loss": 0.3707, |
|
"step": 37000 |
|
}, |
|
{ |
|
"epoch": 2.68, |
|
"eval_loss": 0.13885775208473206, |
|
"eval_runtime": 1286.0056, |
|
"eval_samples_per_second": 8.892, |
|
"eval_steps_per_second": 1.112, |
|
"eval_wer": 0.17355045923254633, |
|
"step": 37000 |
|
}, |
|
{ |
|
"epoch": 2.7, |
|
"learning_rate": 0.00019253490870032222, |
|
"loss": 0.3682, |
|
"step": 37200 |
|
}, |
|
{ |
|
"epoch": 2.7, |
|
"eval_loss": 0.13624690473079681, |
|
"eval_runtime": 1286.554, |
|
"eval_samples_per_second": 8.888, |
|
"eval_steps_per_second": 1.111, |
|
"eval_wer": 0.17115005353017412, |
|
"step": 37200 |
|
}, |
|
{ |
|
"epoch": 2.72, |
|
"learning_rate": 0.00019194902841519382, |
|
"loss": 0.3553, |
|
"step": 37400 |
|
}, |
|
{ |
|
"epoch": 2.72, |
|
"eval_loss": 0.13790255784988403, |
|
"eval_runtime": 1291.8798, |
|
"eval_samples_per_second": 8.851, |
|
"eval_steps_per_second": 1.107, |
|
"eval_wer": 0.17359553727390545, |
|
"step": 37400 |
|
}, |
|
{ |
|
"epoch": 2.74, |
|
"learning_rate": 0.0001913631481300654, |
|
"loss": 0.3721, |
|
"step": 37600 |
|
}, |
|
{ |
|
"epoch": 2.74, |
|
"eval_loss": 0.1351141631603241, |
|
"eval_runtime": 1288.5954, |
|
"eval_samples_per_second": 8.874, |
|
"eval_steps_per_second": 1.11, |
|
"eval_wer": 0.16746492364906745, |
|
"step": 37600 |
|
}, |
|
{ |
|
"epoch": 2.76, |
|
"learning_rate": 0.000190777267844937, |
|
"loss": 0.3615, |
|
"step": 37800 |
|
}, |
|
{ |
|
"epoch": 2.76, |
|
"eval_loss": 0.13291703164577484, |
|
"eval_runtime": 1285.1295, |
|
"eval_samples_per_second": 8.898, |
|
"eval_steps_per_second": 1.113, |
|
"eval_wer": 0.1660449653462557, |
|
"step": 37800 |
|
}, |
|
{ |
|
"epoch": 2.78, |
|
"learning_rate": 0.0001901913875598086, |
|
"loss": 0.3484, |
|
"step": 38000 |
|
}, |
|
{ |
|
"epoch": 2.78, |
|
"eval_loss": 0.1341363489627838, |
|
"eval_runtime": 1287.2816, |
|
"eval_samples_per_second": 8.883, |
|
"eval_steps_per_second": 1.111, |
|
"eval_wer": 0.16912154166901447, |
|
"step": 38000 |
|
}, |
|
{ |
|
"epoch": 2.8, |
|
"learning_rate": 0.00018960550727468019, |
|
"loss": 0.3738, |
|
"step": 38200 |
|
}, |
|
{ |
|
"epoch": 2.8, |
|
"eval_loss": 0.13387584686279297, |
|
"eval_runtime": 1267.1408, |
|
"eval_samples_per_second": 9.024, |
|
"eval_steps_per_second": 1.129, |
|
"eval_wer": 0.1686031441933848, |
|
"step": 38200 |
|
}, |
|
{ |
|
"epoch": 2.82, |
|
"learning_rate": 0.0001890196269895518, |
|
"loss": 0.3599, |
|
"step": 38400 |
|
}, |
|
{ |
|
"epoch": 2.82, |
|
"eval_loss": 0.13417528569698334, |
|
"eval_runtime": 1259.8892, |
|
"eval_samples_per_second": 9.076, |
|
"eval_steps_per_second": 1.135, |
|
"eval_wer": 0.16907646362765538, |
|
"step": 38400 |
|
}, |
|
{ |
|
"epoch": 2.84, |
|
"learning_rate": 0.0001884337467044234, |
|
"loss": 0.3534, |
|
"step": 38600 |
|
}, |
|
{ |
|
"epoch": 2.84, |
|
"eval_loss": 0.13447332382202148, |
|
"eval_runtime": 1271.9609, |
|
"eval_samples_per_second": 8.99, |
|
"eval_steps_per_second": 1.124, |
|
"eval_wer": 0.1683664844762495, |
|
"step": 38600 |
|
}, |
|
{ |
|
"epoch": 2.86, |
|
"learning_rate": 0.00018784786641929497, |
|
"loss": 0.3502, |
|
"step": 38800 |
|
}, |
|
{ |
|
"epoch": 2.86, |
|
"eval_loss": 0.13398751616477966, |
|
"eval_runtime": 1260.4382, |
|
"eval_samples_per_second": 9.072, |
|
"eval_steps_per_second": 1.135, |
|
"eval_wer": 0.16712683833887418, |
|
"step": 38800 |
|
}, |
|
{ |
|
"epoch": 2.87, |
|
"learning_rate": 0.00018726198613416658, |
|
"loss": 0.3597, |
|
"step": 39000 |
|
}, |
|
{ |
|
"epoch": 2.87, |
|
"eval_loss": 0.1342509537935257, |
|
"eval_runtime": 1268.7112, |
|
"eval_samples_per_second": 9.013, |
|
"eval_steps_per_second": 1.127, |
|
"eval_wer": 0.17053023046148644, |
|
"step": 39000 |
|
}, |
|
{ |
|
"epoch": 2.89, |
|
"learning_rate": 0.00018667610584903815, |
|
"loss": 0.3558, |
|
"step": 39200 |
|
}, |
|
{ |
|
"epoch": 2.89, |
|
"eval_loss": 0.13048462569713593, |
|
"eval_runtime": 1294.9536, |
|
"eval_samples_per_second": 8.83, |
|
"eval_steps_per_second": 1.104, |
|
"eval_wer": 0.1658646531808193, |
|
"step": 39200 |
|
}, |
|
{ |
|
"epoch": 2.91, |
|
"learning_rate": 0.00018609022556390976, |
|
"loss": 0.3685, |
|
"step": 39400 |
|
}, |
|
{ |
|
"epoch": 2.91, |
|
"eval_loss": 0.13240975141525269, |
|
"eval_runtime": 1294.9408, |
|
"eval_samples_per_second": 8.831, |
|
"eval_steps_per_second": 1.104, |
|
"eval_wer": 0.16629289457373078, |
|
"step": 39400 |
|
}, |
|
{ |
|
"epoch": 2.93, |
|
"learning_rate": 0.00018550434527878136, |
|
"loss": 0.3709, |
|
"step": 39600 |
|
}, |
|
{ |
|
"epoch": 2.93, |
|
"eval_loss": 0.13161210715770721, |
|
"eval_runtime": 1282.8283, |
|
"eval_samples_per_second": 8.914, |
|
"eval_steps_per_second": 1.115, |
|
"eval_wer": 0.1704626133994478, |
|
"step": 39600 |
|
}, |
|
{ |
|
"epoch": 3.02, |
|
"learning_rate": 0.00018491846499365294, |
|
"loss": 0.3561, |
|
"step": 39800 |
|
}, |
|
{ |
|
"epoch": 3.02, |
|
"eval_loss": 0.13415904343128204, |
|
"eval_runtime": 1283.3514, |
|
"eval_samples_per_second": 8.91, |
|
"eval_steps_per_second": 1.114, |
|
"eval_wer": 0.16870456978644277, |
|
"step": 39800 |
|
}, |
|
{ |
|
"epoch": 3.04, |
|
"learning_rate": 0.00018433258470852454, |
|
"loss": 0.3465, |
|
"step": 40000 |
|
}, |
|
{ |
|
"epoch": 3.04, |
|
"eval_loss": 0.12848955392837524, |
|
"eval_runtime": 1271.6159, |
|
"eval_samples_per_second": 8.992, |
|
"eval_steps_per_second": 1.125, |
|
"eval_wer": 0.1640051839747563, |
|
"step": 40000 |
|
}, |
|
{ |
|
"epoch": 3.06, |
|
"learning_rate": 0.00018374670442339615, |
|
"loss": 0.3381, |
|
"step": 40200 |
|
}, |
|
{ |
|
"epoch": 3.06, |
|
"eval_loss": 0.13284260034561157, |
|
"eval_runtime": 1266.2009, |
|
"eval_samples_per_second": 9.031, |
|
"eval_steps_per_second": 1.129, |
|
"eval_wer": 0.1692116977517327, |
|
"step": 40200 |
|
}, |
|
{ |
|
"epoch": 3.08, |
|
"learning_rate": 0.00018316082413826772, |
|
"loss": 0.3391, |
|
"step": 40400 |
|
}, |
|
{ |
|
"epoch": 3.08, |
|
"eval_loss": 0.1303485929965973, |
|
"eval_runtime": 1268.7911, |
|
"eval_samples_per_second": 9.013, |
|
"eval_steps_per_second": 1.127, |
|
"eval_wer": 0.16649574575984674, |
|
"step": 40400 |
|
}, |
|
{ |
|
"epoch": 3.1, |
|
"learning_rate": 0.00018257494385313933, |
|
"loss": 0.3657, |
|
"step": 40600 |
|
}, |
|
{ |
|
"epoch": 3.1, |
|
"eval_loss": 0.13046273589134216, |
|
"eval_runtime": 1268.0254, |
|
"eval_samples_per_second": 9.018, |
|
"eval_steps_per_second": 1.128, |
|
"eval_wer": 0.1679945906350369, |
|
"step": 40600 |
|
}, |
|
{ |
|
"epoch": 3.12, |
|
"learning_rate": 0.0001819890635680109, |
|
"loss": 0.3401, |
|
"step": 40800 |
|
}, |
|
{ |
|
"epoch": 3.12, |
|
"eval_loss": 0.12838347256183624, |
|
"eval_runtime": 1274.5116, |
|
"eval_samples_per_second": 8.972, |
|
"eval_steps_per_second": 1.122, |
|
"eval_wer": 0.16308108412689468, |
|
"step": 40800 |
|
}, |
|
{ |
|
"epoch": 3.14, |
|
"learning_rate": 0.0001814031832828825, |
|
"loss": 0.3534, |
|
"step": 41000 |
|
}, |
|
{ |
|
"epoch": 3.14, |
|
"eval_loss": 0.1312979757785797, |
|
"eval_runtime": 1280.3407, |
|
"eval_samples_per_second": 8.931, |
|
"eval_steps_per_second": 1.117, |
|
"eval_wer": 0.16876091733814166, |
|
"step": 41000 |
|
}, |
|
{ |
|
"epoch": 3.16, |
|
"learning_rate": 0.0001808173029977541, |
|
"loss": 0.3451, |
|
"step": 41200 |
|
}, |
|
{ |
|
"epoch": 3.16, |
|
"eval_loss": 0.1285577118396759, |
|
"eval_runtime": 1281.0478, |
|
"eval_samples_per_second": 8.926, |
|
"eval_steps_per_second": 1.116, |
|
"eval_wer": 0.16237110497548882, |
|
"step": 41200 |
|
}, |
|
{ |
|
"epoch": 3.17, |
|
"learning_rate": 0.0001802314227126257, |
|
"loss": 0.3436, |
|
"step": 41400 |
|
}, |
|
{ |
|
"epoch": 3.17, |
|
"eval_loss": 0.1303623467683792, |
|
"eval_runtime": 1282.6637, |
|
"eval_samples_per_second": 8.915, |
|
"eval_steps_per_second": 1.115, |
|
"eval_wer": 0.1643319997746098, |
|
"step": 41400 |
|
}, |
|
{ |
|
"epoch": 3.19, |
|
"learning_rate": 0.0001796455424274973, |
|
"loss": 0.3406, |
|
"step": 41600 |
|
}, |
|
{ |
|
"epoch": 3.19, |
|
"eval_loss": 0.12939453125, |
|
"eval_runtime": 1277.6832, |
|
"eval_samples_per_second": 8.95, |
|
"eval_steps_per_second": 1.119, |
|
"eval_wer": 0.1627993463684003, |
|
"step": 41600 |
|
}, |
|
{ |
|
"epoch": 3.21, |
|
"learning_rate": 0.0001790596621423689, |
|
"loss": 0.3449, |
|
"step": 41800 |
|
}, |
|
{ |
|
"epoch": 3.21, |
|
"eval_loss": 0.12918172776699066, |
|
"eval_runtime": 1279.0857, |
|
"eval_samples_per_second": 8.94, |
|
"eval_steps_per_second": 1.118, |
|
"eval_wer": 0.165627993463684, |
|
"step": 41800 |
|
}, |
|
{ |
|
"epoch": 3.23, |
|
"learning_rate": 0.00017847671125866612, |
|
"loss": 0.3374, |
|
"step": 42000 |
|
}, |
|
{ |
|
"epoch": 3.23, |
|
"eval_loss": 0.1299501359462738, |
|
"eval_runtime": 1278.9732, |
|
"eval_samples_per_second": 8.941, |
|
"eval_steps_per_second": 1.118, |
|
"eval_wer": 0.1630247365751958, |
|
"step": 42000 |
|
}, |
|
{ |
|
"epoch": 3.25, |
|
"learning_rate": 0.00017789083097353772, |
|
"loss": 0.3392, |
|
"step": 42200 |
|
}, |
|
{ |
|
"epoch": 3.25, |
|
"eval_loss": 0.12759089469909668, |
|
"eval_runtime": 1268.8237, |
|
"eval_samples_per_second": 9.012, |
|
"eval_steps_per_second": 1.127, |
|
"eval_wer": 0.16211190623767396, |
|
"step": 42200 |
|
}, |
|
{ |
|
"epoch": 3.27, |
|
"learning_rate": 0.00017730495068840932, |
|
"loss": 0.3234, |
|
"step": 42400 |
|
}, |
|
{ |
|
"epoch": 3.27, |
|
"eval_loss": 0.12891671061515808, |
|
"eval_runtime": 1261.6618, |
|
"eval_samples_per_second": 9.063, |
|
"eval_steps_per_second": 1.133, |
|
"eval_wer": 0.16194286358257734, |
|
"step": 42400 |
|
}, |
|
{ |
|
"epoch": 3.29, |
|
"learning_rate": 0.0001767190704032809, |
|
"loss": 0.3506, |
|
"step": 42600 |
|
}, |
|
{ |
|
"epoch": 3.29, |
|
"eval_loss": 0.1317521631717682, |
|
"eval_runtime": 1268.2708, |
|
"eval_samples_per_second": 9.016, |
|
"eval_steps_per_second": 1.128, |
|
"eval_wer": 0.1603764016453485, |
|
"step": 42600 |
|
}, |
|
{ |
|
"epoch": 3.31, |
|
"learning_rate": 0.0001761331901181525, |
|
"loss": 0.3317, |
|
"step": 42800 |
|
}, |
|
{ |
|
"epoch": 3.31, |
|
"eval_loss": 0.12691594660282135, |
|
"eval_runtime": 1266.6129, |
|
"eval_samples_per_second": 9.028, |
|
"eval_steps_per_second": 1.129, |
|
"eval_wer": 0.16011720290753367, |
|
"step": 42800 |
|
}, |
|
{ |
|
"epoch": 3.33, |
|
"learning_rate": 0.0001755473098330241, |
|
"loss": 0.3412, |
|
"step": 43000 |
|
}, |
|
{ |
|
"epoch": 3.33, |
|
"eval_loss": 0.12849685549736023, |
|
"eval_runtime": 1268.3183, |
|
"eval_samples_per_second": 9.016, |
|
"eval_steps_per_second": 1.127, |
|
"eval_wer": 0.15849439341860597, |
|
"step": 43000 |
|
}, |
|
{ |
|
"epoch": 3.35, |
|
"learning_rate": 0.0001749614295478957, |
|
"loss": 0.3432, |
|
"step": 43200 |
|
}, |
|
{ |
|
"epoch": 3.35, |
|
"eval_loss": 0.12799765169620514, |
|
"eval_runtime": 1270.7987, |
|
"eval_samples_per_second": 8.998, |
|
"eval_steps_per_second": 1.125, |
|
"eval_wer": 0.1610300332450555, |
|
"step": 43200 |
|
}, |
|
{ |
|
"epoch": 3.37, |
|
"learning_rate": 0.0001743755492627673, |
|
"loss": 0.3274, |
|
"step": 43400 |
|
}, |
|
{ |
|
"epoch": 3.37, |
|
"eval_loss": 0.1282772719860077, |
|
"eval_runtime": 1270.8967, |
|
"eval_samples_per_second": 8.998, |
|
"eval_steps_per_second": 1.125, |
|
"eval_wer": 0.15816757761875247, |
|
"step": 43400 |
|
}, |
|
{ |
|
"epoch": 3.39, |
|
"learning_rate": 0.00017378966897763887, |
|
"loss": 0.3378, |
|
"step": 43600 |
|
}, |
|
{ |
|
"epoch": 3.39, |
|
"eval_loss": 0.12611497938632965, |
|
"eval_runtime": 1268.024, |
|
"eval_samples_per_second": 9.018, |
|
"eval_steps_per_second": 1.128, |
|
"eval_wer": 0.15873105313574126, |
|
"step": 43600 |
|
}, |
|
{ |
|
"epoch": 3.41, |
|
"learning_rate": 0.00017320378869251047, |
|
"loss": 0.3335, |
|
"step": 43800 |
|
}, |
|
{ |
|
"epoch": 3.41, |
|
"eval_loss": 0.12488209456205368, |
|
"eval_runtime": 1262.74, |
|
"eval_samples_per_second": 9.056, |
|
"eval_steps_per_second": 1.132, |
|
"eval_wer": 0.157637910632783, |
|
"step": 43800 |
|
}, |
|
{ |
|
"epoch": 3.43, |
|
"learning_rate": 0.00017261790840738208, |
|
"loss": 0.323, |
|
"step": 44000 |
|
}, |
|
{ |
|
"epoch": 3.43, |
|
"eval_loss": 0.1285601556301117, |
|
"eval_runtime": 1264.5781, |
|
"eval_samples_per_second": 9.043, |
|
"eval_steps_per_second": 1.131, |
|
"eval_wer": 0.15807742153603427, |
|
"step": 44000 |
|
}, |
|
{ |
|
"epoch": 3.45, |
|
"learning_rate": 0.00017203202812225365, |
|
"loss": 0.3427, |
|
"step": 44200 |
|
}, |
|
{ |
|
"epoch": 3.45, |
|
"eval_loss": 0.12231270223855972, |
|
"eval_runtime": 1271.7465, |
|
"eval_samples_per_second": 8.992, |
|
"eval_steps_per_second": 1.124, |
|
"eval_wer": 0.15785203132923875, |
|
"step": 44200 |
|
}, |
|
{ |
|
"epoch": 3.47, |
|
"learning_rate": 0.00017144614783712526, |
|
"loss": 0.3276, |
|
"step": 44400 |
|
}, |
|
{ |
|
"epoch": 3.47, |
|
"eval_loss": 0.12326563894748688, |
|
"eval_runtime": 1274.0885, |
|
"eval_samples_per_second": 8.975, |
|
"eval_steps_per_second": 1.122, |
|
"eval_wer": 0.1624725305685468, |
|
"step": 44400 |
|
}, |
|
{ |
|
"epoch": 3.49, |
|
"learning_rate": 0.0001708602675519969, |
|
"loss": 0.3291, |
|
"step": 44600 |
|
}, |
|
{ |
|
"epoch": 3.49, |
|
"eval_loss": 0.12240497022867203, |
|
"eval_runtime": 1272.556, |
|
"eval_samples_per_second": 8.986, |
|
"eval_steps_per_second": 1.124, |
|
"eval_wer": 0.15554178170958471, |
|
"step": 44600 |
|
}, |
|
{ |
|
"epoch": 3.51, |
|
"learning_rate": 0.00017027438726686844, |
|
"loss": 0.3374, |
|
"step": 44800 |
|
}, |
|
{ |
|
"epoch": 3.51, |
|
"eval_loss": 0.12430132925510406, |
|
"eval_runtime": 1271.1476, |
|
"eval_samples_per_second": 8.996, |
|
"eval_steps_per_second": 1.125, |
|
"eval_wer": 0.15789710937059784, |
|
"step": 44800 |
|
}, |
|
{ |
|
"epoch": 3.52, |
|
"learning_rate": 0.00016968850698174007, |
|
"loss": 0.3383, |
|
"step": 45000 |
|
}, |
|
{ |
|
"epoch": 3.52, |
|
"eval_loss": 0.12537476420402527, |
|
"eval_runtime": 1276.5687, |
|
"eval_samples_per_second": 8.958, |
|
"eval_steps_per_second": 1.12, |
|
"eval_wer": 0.15711951315715333, |
|
"step": 45000 |
|
}, |
|
{ |
|
"epoch": 3.54, |
|
"learning_rate": 0.00016910262669661162, |
|
"loss": 0.3231, |
|
"step": 45200 |
|
}, |
|
{ |
|
"epoch": 3.54, |
|
"eval_loss": 0.1274578869342804, |
|
"eval_runtime": 1279.812, |
|
"eval_samples_per_second": 8.935, |
|
"eval_steps_per_second": 1.117, |
|
"eval_wer": 0.1589000957908379, |
|
"step": 45200 |
|
}, |
|
{ |
|
"epoch": 3.56, |
|
"learning_rate": 0.00016851674641148325, |
|
"loss": 0.3265, |
|
"step": 45400 |
|
}, |
|
{ |
|
"epoch": 3.56, |
|
"eval_loss": 0.12590159475803375, |
|
"eval_runtime": 1276.0756, |
|
"eval_samples_per_second": 8.961, |
|
"eval_steps_per_second": 1.121, |
|
"eval_wer": 0.15724347777089084, |
|
"step": 45400 |
|
}, |
|
{ |
|
"epoch": 3.58, |
|
"learning_rate": 0.00016793086612635485, |
|
"loss": 0.3334, |
|
"step": 45600 |
|
}, |
|
{ |
|
"epoch": 3.58, |
|
"eval_loss": 0.12398205697536469, |
|
"eval_runtime": 1283.216, |
|
"eval_samples_per_second": 8.911, |
|
"eval_steps_per_second": 1.114, |
|
"eval_wer": 0.1549895757029357, |
|
"step": 45600 |
|
}, |
|
{ |
|
"epoch": 3.6, |
|
"learning_rate": 0.00016734791524265207, |
|
"loss": 0.3314, |
|
"step": 45800 |
|
}, |
|
{ |
|
"epoch": 3.6, |
|
"eval_loss": 0.12473876029253006, |
|
"eval_runtime": 1280.783, |
|
"eval_samples_per_second": 8.928, |
|
"eval_steps_per_second": 1.117, |
|
"eval_wer": 0.15710824364681356, |
|
"step": 45800 |
|
}, |
|
{ |
|
"epoch": 3.62, |
|
"learning_rate": 0.00016676203495752365, |
|
"loss": 0.3372, |
|
"step": 46000 |
|
}, |
|
{ |
|
"epoch": 3.62, |
|
"eval_loss": 0.1239311620593071, |
|
"eval_runtime": 1282.1393, |
|
"eval_samples_per_second": 8.919, |
|
"eval_steps_per_second": 1.115, |
|
"eval_wer": 0.15447117822730602, |
|
"step": 46000 |
|
}, |
|
{ |
|
"epoch": 3.64, |
|
"learning_rate": 0.00016617615467239526, |
|
"loss": 0.3362, |
|
"step": 46200 |
|
}, |
|
{ |
|
"epoch": 3.64, |
|
"eval_loss": 0.1215914860367775, |
|
"eval_runtime": 1278.6758, |
|
"eval_samples_per_second": 8.943, |
|
"eval_steps_per_second": 1.118, |
|
"eval_wer": 0.15436975263424804, |
|
"step": 46200 |
|
}, |
|
{ |
|
"epoch": 3.66, |
|
"learning_rate": 0.00016559027438726683, |
|
"loss": 0.3209, |
|
"step": 46400 |
|
}, |
|
{ |
|
"epoch": 3.66, |
|
"eval_loss": 0.1207798570394516, |
|
"eval_runtime": 1279.7208, |
|
"eval_samples_per_second": 8.936, |
|
"eval_steps_per_second": 1.117, |
|
"eval_wer": 0.15486561108919816, |
|
"step": 46400 |
|
}, |
|
{ |
|
"epoch": 3.68, |
|
"learning_rate": 0.00016500732350356408, |
|
"loss": 0.3202, |
|
"step": 46600 |
|
}, |
|
{ |
|
"epoch": 3.68, |
|
"eval_loss": 0.11999519914388657, |
|
"eval_runtime": 1274.3631, |
|
"eval_samples_per_second": 8.973, |
|
"eval_steps_per_second": 1.122, |
|
"eval_wer": 0.15501211472361526, |
|
"step": 46600 |
|
}, |
|
{ |
|
"epoch": 3.7, |
|
"learning_rate": 0.00016442144321843568, |
|
"loss": 0.3135, |
|
"step": 46800 |
|
}, |
|
{ |
|
"epoch": 3.7, |
|
"eval_loss": 0.12257663160562515, |
|
"eval_runtime": 1258.8007, |
|
"eval_samples_per_second": 9.084, |
|
"eval_steps_per_second": 1.136, |
|
"eval_wer": 0.15324280160027046, |
|
"step": 46800 |
|
}, |
|
{ |
|
"epoch": 3.72, |
|
"learning_rate": 0.0001638355629333073, |
|
"loss": 0.3148, |
|
"step": 47000 |
|
}, |
|
{ |
|
"epoch": 3.72, |
|
"eval_loss": 0.11898791790008545, |
|
"eval_runtime": 1289.0474, |
|
"eval_samples_per_second": 8.871, |
|
"eval_steps_per_second": 1.109, |
|
"eval_wer": 0.15274694314532034, |
|
"step": 47000 |
|
}, |
|
{ |
|
"epoch": 3.74, |
|
"learning_rate": 0.00016324968264817887, |
|
"loss": 0.3339, |
|
"step": 47200 |
|
}, |
|
{ |
|
"epoch": 3.74, |
|
"eval_loss": 0.12242983281612396, |
|
"eval_runtime": 1285.1481, |
|
"eval_samples_per_second": 8.898, |
|
"eval_steps_per_second": 1.113, |
|
"eval_wer": 0.15426832704119006, |
|
"step": 47200 |
|
}, |
|
{ |
|
"epoch": 3.76, |
|
"learning_rate": 0.00016266380236305047, |
|
"loss": 0.3238, |
|
"step": 47400 |
|
}, |
|
{ |
|
"epoch": 3.76, |
|
"eval_loss": 0.12152837961912155, |
|
"eval_runtime": 1287.6381, |
|
"eval_samples_per_second": 8.881, |
|
"eval_steps_per_second": 1.111, |
|
"eval_wer": 0.15251028342818504, |
|
"step": 47400 |
|
}, |
|
{ |
|
"epoch": 3.78, |
|
"learning_rate": 0.00016207792207792205, |
|
"loss": 0.3141, |
|
"step": 47600 |
|
}, |
|
{ |
|
"epoch": 3.78, |
|
"eval_loss": 0.11933837085962296, |
|
"eval_runtime": 1282.8603, |
|
"eval_samples_per_second": 8.914, |
|
"eval_steps_per_second": 1.115, |
|
"eval_wer": 0.1536034259311433, |
|
"step": 47600 |
|
}, |
|
{ |
|
"epoch": 3.8, |
|
"learning_rate": 0.00016149204179279365, |
|
"loss": 0.3354, |
|
"step": 47800 |
|
}, |
|
{ |
|
"epoch": 3.8, |
|
"eval_loss": 0.12152481079101562, |
|
"eval_runtime": 1280.7316, |
|
"eval_samples_per_second": 8.928, |
|
"eval_steps_per_second": 1.117, |
|
"eval_wer": 0.15500084521327548, |
|
"step": 47800 |
|
}, |
|
{ |
|
"epoch": 3.82, |
|
"learning_rate": 0.00016090616150766525, |
|
"loss": 0.3217, |
|
"step": 48000 |
|
}, |
|
{ |
|
"epoch": 3.82, |
|
"eval_loss": 0.12007062137126923, |
|
"eval_runtime": 1281.3849, |
|
"eval_samples_per_second": 8.924, |
|
"eval_steps_per_second": 1.116, |
|
"eval_wer": 0.15225108469037021, |
|
"step": 48000 |
|
}, |
|
{ |
|
"epoch": 3.84, |
|
"learning_rate": 0.00016032028122253683, |
|
"loss": 0.3248, |
|
"step": 48200 |
|
}, |
|
{ |
|
"epoch": 3.84, |
|
"eval_loss": 0.11909425258636475, |
|
"eval_runtime": 1287.6101, |
|
"eval_samples_per_second": 8.881, |
|
"eval_steps_per_second": 1.111, |
|
"eval_wer": 0.15036907646362765, |
|
"step": 48200 |
|
}, |
|
{ |
|
"epoch": 3.86, |
|
"learning_rate": 0.00015973440093740844, |
|
"loss": 0.3169, |
|
"step": 48400 |
|
}, |
|
{ |
|
"epoch": 3.86, |
|
"eval_loss": 0.12102551758289337, |
|
"eval_runtime": 1285.0012, |
|
"eval_samples_per_second": 8.899, |
|
"eval_steps_per_second": 1.113, |
|
"eval_wer": 0.15103397757367443, |
|
"step": 48400 |
|
}, |
|
{ |
|
"epoch": 3.87, |
|
"learning_rate": 0.00015914852065228, |
|
"loss": 0.3235, |
|
"step": 48600 |
|
}, |
|
{ |
|
"epoch": 3.87, |
|
"eval_loss": 0.12175353616476059, |
|
"eval_runtime": 1286.2751, |
|
"eval_samples_per_second": 8.89, |
|
"eval_steps_per_second": 1.112, |
|
"eval_wer": 0.15386262466895814, |
|
"step": 48600 |
|
}, |
|
{ |
|
"epoch": 3.89, |
|
"learning_rate": 0.00015856264036715162, |
|
"loss": 0.3262, |
|
"step": 48800 |
|
}, |
|
{ |
|
"epoch": 3.89, |
|
"eval_loss": 0.11936228722333908, |
|
"eval_runtime": 1280.7325, |
|
"eval_samples_per_second": 8.928, |
|
"eval_steps_per_second": 1.117, |
|
"eval_wer": 0.15041415450498677, |
|
"step": 48800 |
|
}, |
|
{ |
|
"epoch": 3.91, |
|
"learning_rate": 0.00015797676008202325, |
|
"loss": 0.3183, |
|
"step": 49000 |
|
}, |
|
{ |
|
"epoch": 3.91, |
|
"eval_loss": 0.116153784096241, |
|
"eval_runtime": 1283.8223, |
|
"eval_samples_per_second": 8.907, |
|
"eval_steps_per_second": 1.114, |
|
"eval_wer": 0.14855468529892377, |
|
"step": 49000 |
|
}, |
|
{ |
|
"epoch": 3.93, |
|
"learning_rate": 0.0001573908797968948, |
|
"loss": 0.3229, |
|
"step": 49200 |
|
}, |
|
{ |
|
"epoch": 3.93, |
|
"eval_loss": 0.11911948770284653, |
|
"eval_runtime": 1284.0466, |
|
"eval_samples_per_second": 8.905, |
|
"eval_steps_per_second": 1.114, |
|
"eval_wer": 0.14968163633290132, |
|
"step": 49200 |
|
}, |
|
{ |
|
"epoch": 4.02, |
|
"learning_rate": 0.00015680499951176643, |
|
"loss": 0.322, |
|
"step": 49400 |
|
}, |
|
{ |
|
"epoch": 4.02, |
|
"eval_loss": 0.11941692978143692, |
|
"eval_runtime": 629.3496, |
|
"eval_samples_per_second": 18.17, |
|
"eval_steps_per_second": 2.272, |
|
"eval_wer": 0.15029018989124923, |
|
"step": 49400 |
|
}, |
|
{ |
|
"epoch": 4.04, |
|
"learning_rate": 0.00015621911922663803, |
|
"loss": 0.309, |
|
"step": 49600 |
|
}, |
|
{ |
|
"epoch": 4.04, |
|
"eval_loss": 0.11948557198047638, |
|
"eval_runtime": 625.3187, |
|
"eval_samples_per_second": 18.287, |
|
"eval_steps_per_second": 2.287, |
|
"eval_wer": 0.14937735955372738, |
|
"step": 49600 |
|
}, |
|
{ |
|
"epoch": 4.06, |
|
"learning_rate": 0.0001556332389415096, |
|
"loss": 0.3013, |
|
"step": 49800 |
|
}, |
|
{ |
|
"epoch": 4.06, |
|
"eval_loss": 0.11908197402954102, |
|
"eval_runtime": 625.6323, |
|
"eval_samples_per_second": 18.278, |
|
"eval_steps_per_second": 2.286, |
|
"eval_wer": 0.1490392742435341, |
|
"step": 49800 |
|
}, |
|
{ |
|
"epoch": 4.08, |
|
"learning_rate": 0.00015504735865638121, |
|
"loss": 0.3107, |
|
"step": 50000 |
|
}, |
|
{ |
|
"epoch": 4.08, |
|
"eval_loss": 0.11991952359676361, |
|
"eval_runtime": 618.7466, |
|
"eval_samples_per_second": 18.481, |
|
"eval_steps_per_second": 2.311, |
|
"eval_wer": 0.14963655829154224, |
|
"step": 50000 |
|
}, |
|
{ |
|
"epoch": 4.1, |
|
"learning_rate": 0.00015446440777267843, |
|
"loss": 0.3205, |
|
"step": 50200 |
|
}, |
|
{ |
|
"epoch": 4.1, |
|
"eval_loss": 0.11845077574253082, |
|
"eval_runtime": 622.4777, |
|
"eval_samples_per_second": 18.37, |
|
"eval_steps_per_second": 2.297, |
|
"eval_wer": 0.1494224375950865, |
|
"step": 50200 |
|
}, |
|
{ |
|
"epoch": 4.12, |
|
"learning_rate": 0.00015387852748755, |
|
"loss": 0.3053, |
|
"step": 50400 |
|
}, |
|
{ |
|
"epoch": 4.12, |
|
"eval_loss": 0.11776554584503174, |
|
"eval_runtime": 626.5582, |
|
"eval_samples_per_second": 18.25, |
|
"eval_steps_per_second": 2.282, |
|
"eval_wer": 0.14600777596213443, |
|
"step": 50400 |
|
}, |
|
{ |
|
"epoch": 4.14, |
|
"learning_rate": 0.00015329264720242161, |
|
"loss": 0.3039, |
|
"step": 50600 |
|
}, |
|
{ |
|
"epoch": 4.14, |
|
"eval_loss": 0.11720620095729828, |
|
"eval_runtime": 626.1062, |
|
"eval_samples_per_second": 18.264, |
|
"eval_steps_per_second": 2.284, |
|
"eval_wer": 0.14637966980334705, |
|
"step": 50600 |
|
}, |
|
{ |
|
"epoch": 4.16, |
|
"learning_rate": 0.00015270676691729325, |
|
"loss": 0.3122, |
|
"step": 50800 |
|
}, |
|
{ |
|
"epoch": 4.16, |
|
"eval_loss": 0.11798885464668274, |
|
"eval_runtime": 617.4335, |
|
"eval_samples_per_second": 18.52, |
|
"eval_steps_per_second": 2.316, |
|
"eval_wer": 0.147889784188877, |
|
"step": 50800 |
|
}, |
|
{ |
|
"epoch": 4.17, |
|
"learning_rate": 0.0001521208866321648, |
|
"loss": 0.3144, |
|
"step": 51000 |
|
}, |
|
{ |
|
"epoch": 4.17, |
|
"eval_loss": 0.1195751279592514, |
|
"eval_runtime": 623.8518, |
|
"eval_samples_per_second": 18.33, |
|
"eval_steps_per_second": 2.292, |
|
"eval_wer": 0.15110159463571307, |
|
"step": 51000 |
|
}, |
|
{ |
|
"epoch": 4.19, |
|
"learning_rate": 0.00015153500634703643, |
|
"loss": 0.3075, |
|
"step": 51200 |
|
}, |
|
{ |
|
"epoch": 4.19, |
|
"eval_loss": 0.11787259578704834, |
|
"eval_runtime": 625.9862, |
|
"eval_samples_per_second": 18.267, |
|
"eval_steps_per_second": 2.284, |
|
"eval_wer": 0.14795740125091564, |
|
"step": 51200 |
|
}, |
|
{ |
|
"epoch": 4.21, |
|
"learning_rate": 0.00015094912606190798, |
|
"loss": 0.311, |
|
"step": 51400 |
|
}, |
|
{ |
|
"epoch": 4.21, |
|
"eval_loss": 0.12053581327199936, |
|
"eval_runtime": 618.1157, |
|
"eval_samples_per_second": 18.5, |
|
"eval_steps_per_second": 2.313, |
|
"eval_wer": 0.14891530962979657, |
|
"step": 51400 |
|
}, |
|
{ |
|
"epoch": 4.23, |
|
"learning_rate": 0.0001503632457767796, |
|
"loss": 0.3015, |
|
"step": 51600 |
|
}, |
|
{ |
|
"epoch": 4.23, |
|
"eval_loss": 0.11665117740631104, |
|
"eval_runtime": 622.4687, |
|
"eval_samples_per_second": 18.37, |
|
"eval_steps_per_second": 2.297, |
|
"eval_wer": 0.14651490392742436, |
|
"step": 51600 |
|
}, |
|
{ |
|
"epoch": 4.25, |
|
"learning_rate": 0.00014977736549165119, |
|
"loss": 0.3106, |
|
"step": 51800 |
|
}, |
|
{ |
|
"epoch": 4.25, |
|
"eval_loss": 0.11705104261636734, |
|
"eval_runtime": 625.4889, |
|
"eval_samples_per_second": 18.282, |
|
"eval_steps_per_second": 2.286, |
|
"eval_wer": 0.15017749478785147, |
|
"step": 51800 |
|
}, |
|
{ |
|
"epoch": 4.27, |
|
"learning_rate": 0.0001491914852065228, |
|
"loss": 0.2971, |
|
"step": 52000 |
|
}, |
|
{ |
|
"epoch": 4.27, |
|
"eval_loss": 0.1164257824420929, |
|
"eval_runtime": 623.9511, |
|
"eval_samples_per_second": 18.327, |
|
"eval_steps_per_second": 2.292, |
|
"eval_wer": 0.14739392573392687, |
|
"step": 52000 |
|
}, |
|
{ |
|
"epoch": 4.29, |
|
"learning_rate": 0.0001486056049213944, |
|
"loss": 0.3113, |
|
"step": 52200 |
|
}, |
|
{ |
|
"epoch": 4.29, |
|
"eval_loss": 0.11716081202030182, |
|
"eval_runtime": 620.6594, |
|
"eval_samples_per_second": 18.424, |
|
"eval_steps_per_second": 2.304, |
|
"eval_wer": 0.1467741026652392, |
|
"step": 52200 |
|
}, |
|
{ |
|
"epoch": 4.31, |
|
"learning_rate": 0.00014801972463626597, |
|
"loss": 0.2966, |
|
"step": 52400 |
|
}, |
|
{ |
|
"epoch": 4.31, |
|
"eval_loss": 0.11439959704875946, |
|
"eval_runtime": 625.7976, |
|
"eval_samples_per_second": 18.273, |
|
"eval_steps_per_second": 2.285, |
|
"eval_wer": 0.145928889389756, |
|
"step": 52400 |
|
}, |
|
{ |
|
"epoch": 4.33, |
|
"learning_rate": 0.00014743384435113757, |
|
"loss": 0.3167, |
|
"step": 52600 |
|
}, |
|
{ |
|
"epoch": 4.33, |
|
"eval_loss": 0.11525892466306686, |
|
"eval_runtime": 624.008, |
|
"eval_samples_per_second": 18.325, |
|
"eval_steps_per_second": 2.292, |
|
"eval_wer": 0.1450047895418944, |
|
"step": 52600 |
|
}, |
|
{ |
|
"epoch": 4.35, |
|
"learning_rate": 0.00014684796406600918, |
|
"loss": 0.3101, |
|
"step": 52800 |
|
}, |
|
{ |
|
"epoch": 4.35, |
|
"eval_loss": 0.11492661386728287, |
|
"eval_runtime": 624.0638, |
|
"eval_samples_per_second": 18.323, |
|
"eval_steps_per_second": 2.291, |
|
"eval_wer": 0.1450160590522342, |
|
"step": 52800 |
|
}, |
|
{ |
|
"epoch": 4.37, |
|
"learning_rate": 0.00014626208378088076, |
|
"loss": 0.2972, |
|
"step": 53000 |
|
}, |
|
{ |
|
"epoch": 4.37, |
|
"eval_loss": 0.11436022818088531, |
|
"eval_runtime": 621.413, |
|
"eval_samples_per_second": 18.402, |
|
"eval_steps_per_second": 2.301, |
|
"eval_wer": 0.14395672508029525, |
|
"step": 53000 |
|
}, |
|
{ |
|
"epoch": 4.39, |
|
"learning_rate": 0.00014567620349575236, |
|
"loss": 0.2949, |
|
"step": 53200 |
|
}, |
|
{ |
|
"epoch": 4.39, |
|
"eval_loss": 0.11512508243322372, |
|
"eval_runtime": 622.0613, |
|
"eval_samples_per_second": 18.382, |
|
"eval_steps_per_second": 2.299, |
|
"eval_wer": 0.14511748464529217, |
|
"step": 53200 |
|
}, |
|
{ |
|
"epoch": 4.41, |
|
"learning_rate": 0.00014509032321062394, |
|
"loss": 0.2998, |
|
"step": 53400 |
|
}, |
|
{ |
|
"epoch": 4.41, |
|
"eval_loss": 0.11325063556432724, |
|
"eval_runtime": 625.3597, |
|
"eval_samples_per_second": 18.285, |
|
"eval_steps_per_second": 2.287, |
|
"eval_wer": 0.14369752634248042, |
|
"step": 53400 |
|
}, |
|
{ |
|
"epoch": 4.43, |
|
"learning_rate": 0.00014450444292549554, |
|
"loss": 0.3033, |
|
"step": 53600 |
|
}, |
|
{ |
|
"epoch": 4.43, |
|
"eval_loss": 0.11504179239273071, |
|
"eval_runtime": 631.902, |
|
"eval_samples_per_second": 18.096, |
|
"eval_steps_per_second": 2.263, |
|
"eval_wer": 0.14047444638530457, |
|
"step": 53600 |
|
}, |
|
{ |
|
"epoch": 4.45, |
|
"learning_rate": 0.00014391856264036714, |
|
"loss": 0.3064, |
|
"step": 53800 |
|
}, |
|
{ |
|
"epoch": 4.45, |
|
"eval_loss": 0.1125454306602478, |
|
"eval_runtime": 621.7758, |
|
"eval_samples_per_second": 18.391, |
|
"eval_steps_per_second": 2.3, |
|
"eval_wer": 0.1426381923705415, |
|
"step": 53800 |
|
}, |
|
{ |
|
"epoch": 4.47, |
|
"learning_rate": 0.00014333268235523872, |
|
"loss": 0.2942, |
|
"step": 54000 |
|
}, |
|
{ |
|
"epoch": 4.47, |
|
"eval_loss": 0.11342581361532211, |
|
"eval_runtime": 632.3216, |
|
"eval_samples_per_second": 18.084, |
|
"eval_steps_per_second": 2.262, |
|
"eval_wer": 0.14604158449315377, |
|
"step": 54000 |
|
}, |
|
{ |
|
"epoch": 4.49, |
|
"learning_rate": 0.00014274680207011033, |
|
"loss": 0.2907, |
|
"step": 54200 |
|
}, |
|
{ |
|
"epoch": 4.49, |
|
"eval_loss": 0.11338940262794495, |
|
"eval_runtime": 631.9817, |
|
"eval_samples_per_second": 18.094, |
|
"eval_steps_per_second": 2.263, |
|
"eval_wer": 0.1434721361356849, |
|
"step": 54200 |
|
}, |
|
{ |
|
"epoch": 4.51, |
|
"learning_rate": 0.00014216385118640757, |
|
"loss": 0.3061, |
|
"step": 54400 |
|
}, |
|
{ |
|
"epoch": 4.51, |
|
"eval_loss": 0.11131487786769867, |
|
"eval_runtime": 619.3147, |
|
"eval_samples_per_second": 18.464, |
|
"eval_steps_per_second": 2.309, |
|
"eval_wer": 0.14222122048796978, |
|
"step": 54400 |
|
}, |
|
{ |
|
"epoch": 4.52, |
|
"learning_rate": 0.00014157797090127915, |
|
"loss": 0.2872, |
|
"step": 54600 |
|
}, |
|
{ |
|
"epoch": 4.52, |
|
"eval_loss": 0.11378352344036102, |
|
"eval_runtime": 627.6236, |
|
"eval_samples_per_second": 18.22, |
|
"eval_steps_per_second": 2.278, |
|
"eval_wer": 0.14586127232771737, |
|
"step": 54600 |
|
}, |
|
{ |
|
"epoch": 4.54, |
|
"learning_rate": 0.00014099209061615075, |
|
"loss": 0.291, |
|
"step": 54800 |
|
}, |
|
{ |
|
"epoch": 4.54, |
|
"eval_loss": 0.10948885232210159, |
|
"eval_runtime": 631.6042, |
|
"eval_samples_per_second": 18.105, |
|
"eval_steps_per_second": 2.264, |
|
"eval_wer": 0.14297627768073476, |
|
"step": 54800 |
|
}, |
|
{ |
|
"epoch": 4.56, |
|
"learning_rate": 0.00014040621033102236, |
|
"loss": 0.2916, |
|
"step": 55000 |
|
}, |
|
{ |
|
"epoch": 4.56, |
|
"eval_loss": 0.11261311173439026, |
|
"eval_runtime": 629.602, |
|
"eval_samples_per_second": 18.162, |
|
"eval_steps_per_second": 2.271, |
|
"eval_wer": 0.13918972220657014, |
|
"step": 55000 |
|
}, |
|
{ |
|
"epoch": 4.58, |
|
"learning_rate": 0.00013982033004589393, |
|
"loss": 0.295, |
|
"step": 55200 |
|
}, |
|
{ |
|
"epoch": 4.58, |
|
"eval_loss": 0.11116189509630203, |
|
"eval_runtime": 631.1044, |
|
"eval_samples_per_second": 18.119, |
|
"eval_steps_per_second": 2.266, |
|
"eval_wer": 0.13987716233729644, |
|
"step": 55200 |
|
}, |
|
{ |
|
"epoch": 4.6, |
|
"learning_rate": 0.00013923444976076554, |
|
"loss": 0.2953, |
|
"step": 55400 |
|
}, |
|
{ |
|
"epoch": 4.6, |
|
"eval_loss": 0.11264733970165253, |
|
"eval_runtime": 630.3645, |
|
"eval_samples_per_second": 18.14, |
|
"eval_steps_per_second": 2.269, |
|
"eval_wer": 0.13953907702710316, |
|
"step": 55400 |
|
}, |
|
{ |
|
"epoch": 4.62, |
|
"learning_rate": 0.00013864856947563714, |
|
"loss": 0.2934, |
|
"step": 55600 |
|
}, |
|
{ |
|
"epoch": 4.62, |
|
"eval_loss": 0.11191383749246597, |
|
"eval_runtime": 638.5461, |
|
"eval_samples_per_second": 17.908, |
|
"eval_steps_per_second": 2.239, |
|
"eval_wer": 0.1429199301290359, |
|
"step": 55600 |
|
}, |
|
{ |
|
"epoch": 4.64, |
|
"learning_rate": 0.00013806268919050872, |
|
"loss": 0.2995, |
|
"step": 55800 |
|
}, |
|
{ |
|
"epoch": 4.64, |
|
"eval_loss": 0.11275320500135422, |
|
"eval_runtime": 636.369, |
|
"eval_samples_per_second": 17.969, |
|
"eval_steps_per_second": 2.247, |
|
"eval_wer": 0.1417929790950583, |
|
"step": 55800 |
|
}, |
|
{ |
|
"epoch": 4.66, |
|
"learning_rate": 0.00013747680890538032, |
|
"loss": 0.2908, |
|
"step": 56000 |
|
}, |
|
{ |
|
"epoch": 4.66, |
|
"eval_loss": 0.11625451594591141, |
|
"eval_runtime": 629.3289, |
|
"eval_samples_per_second": 18.17, |
|
"eval_steps_per_second": 2.272, |
|
"eval_wer": 0.14449766157660449, |
|
"step": 56000 |
|
}, |
|
{ |
|
"epoch": 4.68, |
|
"learning_rate": 0.0001368909286202519, |
|
"loss": 0.2982, |
|
"step": 56200 |
|
}, |
|
{ |
|
"epoch": 4.68, |
|
"eval_loss": 0.1119445189833641, |
|
"eval_runtime": 627.9078, |
|
"eval_samples_per_second": 18.211, |
|
"eval_steps_per_second": 2.277, |
|
"eval_wer": 0.14262692286020173, |
|
"step": 56200 |
|
}, |
|
{ |
|
"epoch": 4.7, |
|
"learning_rate": 0.0001363050483351235, |
|
"loss": 0.2907, |
|
"step": 56400 |
|
}, |
|
{ |
|
"epoch": 4.7, |
|
"eval_loss": 0.11290817707777023, |
|
"eval_runtime": 625.336, |
|
"eval_samples_per_second": 18.286, |
|
"eval_steps_per_second": 2.287, |
|
"eval_wer": 0.1417028230123401, |
|
"step": 56400 |
|
}, |
|
{ |
|
"epoch": 4.72, |
|
"learning_rate": 0.0001357191680499951, |
|
"loss": 0.2884, |
|
"step": 56600 |
|
}, |
|
{ |
|
"epoch": 4.72, |
|
"eval_loss": 0.11267752945423126, |
|
"eval_runtime": 623.6566, |
|
"eval_samples_per_second": 18.335, |
|
"eval_steps_per_second": 2.293, |
|
"eval_wer": 0.14219868146729026, |
|
"step": 56600 |
|
}, |
|
{ |
|
"epoch": 4.74, |
|
"learning_rate": 0.00013513328776486669, |
|
"loss": 0.2967, |
|
"step": 56800 |
|
}, |
|
{ |
|
"epoch": 4.74, |
|
"eval_loss": 0.11243223398923874, |
|
"eval_runtime": 624.4114, |
|
"eval_samples_per_second": 18.313, |
|
"eval_steps_per_second": 2.29, |
|
"eval_wer": 0.14279596551529836, |
|
"step": 56800 |
|
}, |
|
{ |
|
"epoch": 4.76, |
|
"learning_rate": 0.0001345474074797383, |
|
"loss": 0.2938, |
|
"step": 57000 |
|
}, |
|
{ |
|
"epoch": 4.76, |
|
"eval_loss": 0.11092329025268555, |
|
"eval_runtime": 621.6714, |
|
"eval_samples_per_second": 18.394, |
|
"eval_steps_per_second": 2.3, |
|
"eval_wer": 0.13847974305516425, |
|
"step": 57000 |
|
}, |
|
{ |
|
"epoch": 4.78, |
|
"learning_rate": 0.00013396445659603554, |
|
"loss": 0.2814, |
|
"step": 57200 |
|
}, |
|
{ |
|
"epoch": 4.78, |
|
"eval_loss": 0.11275891214609146, |
|
"eval_runtime": 621.8116, |
|
"eval_samples_per_second": 18.39, |
|
"eval_steps_per_second": 2.3, |
|
"eval_wer": 0.13922353073758945, |
|
"step": 57200 |
|
}, |
|
{ |
|
"epoch": 4.8, |
|
"learning_rate": 0.00013337857631090711, |
|
"loss": 0.2972, |
|
"step": 57400 |
|
}, |
|
{ |
|
"epoch": 4.8, |
|
"eval_loss": 0.1118135005235672, |
|
"eval_runtime": 626.7556, |
|
"eval_samples_per_second": 18.245, |
|
"eval_steps_per_second": 2.282, |
|
"eval_wer": 0.14214233391559136, |
|
"step": 57400 |
|
}, |
|
{ |
|
"epoch": 4.82, |
|
"learning_rate": 0.00013279269602577872, |
|
"loss": 0.2824, |
|
"step": 57600 |
|
}, |
|
{ |
|
"epoch": 4.82, |
|
"eval_loss": 0.11107096076011658, |
|
"eval_runtime": 623.6389, |
|
"eval_samples_per_second": 18.336, |
|
"eval_steps_per_second": 2.293, |
|
"eval_wer": 0.14098157435059447, |
|
"step": 57600 |
|
}, |
|
{ |
|
"epoch": 4.84, |
|
"learning_rate": 0.00013220681574065032, |
|
"loss": 0.2876, |
|
"step": 57800 |
|
}, |
|
{ |
|
"epoch": 4.84, |
|
"eval_loss": 0.11149298399686813, |
|
"eval_runtime": 623.7048, |
|
"eval_samples_per_second": 18.334, |
|
"eval_steps_per_second": 2.293, |
|
"eval_wer": 0.13867132473094043, |
|
"step": 57800 |
|
}, |
|
{ |
|
"epoch": 4.86, |
|
"learning_rate": 0.0001316209354555219, |
|
"loss": 0.2821, |
|
"step": 58000 |
|
}, |
|
{ |
|
"epoch": 4.86, |
|
"eval_loss": 0.10840985924005508, |
|
"eval_runtime": 625.8552, |
|
"eval_samples_per_second": 18.271, |
|
"eval_steps_per_second": 2.285, |
|
"eval_wer": 0.1385473601172029, |
|
"step": 58000 |
|
}, |
|
{ |
|
"epoch": 4.87, |
|
"learning_rate": 0.0001310350551703935, |
|
"loss": 0.2816, |
|
"step": 58200 |
|
}, |
|
{ |
|
"epoch": 4.87, |
|
"eval_loss": 0.11007586866617203, |
|
"eval_runtime": 625.5014, |
|
"eval_samples_per_second": 18.281, |
|
"eval_steps_per_second": 2.286, |
|
"eval_wer": 0.13653011776638305, |
|
"step": 58200 |
|
}, |
|
{ |
|
"epoch": 4.89, |
|
"learning_rate": 0.00013045210428669075, |
|
"loss": 0.2992, |
|
"step": 58400 |
|
}, |
|
{ |
|
"epoch": 4.89, |
|
"eval_loss": 0.10946887731552124, |
|
"eval_runtime": 620.2255, |
|
"eval_samples_per_second": 18.437, |
|
"eval_steps_per_second": 2.306, |
|
"eval_wer": 0.14015890009579085, |
|
"step": 58400 |
|
}, |
|
{ |
|
"epoch": 4.91, |
|
"learning_rate": 0.00012986622400156233, |
|
"loss": 0.2912, |
|
"step": 58600 |
|
}, |
|
{ |
|
"epoch": 4.91, |
|
"eval_loss": 0.10773813724517822, |
|
"eval_runtime": 625.4145, |
|
"eval_samples_per_second": 18.284, |
|
"eval_steps_per_second": 2.286, |
|
"eval_wer": 0.1377584943934186, |
|
"step": 58600 |
|
}, |
|
{ |
|
"epoch": 4.93, |
|
"learning_rate": 0.00012928034371643393, |
|
"loss": 0.2956, |
|
"step": 58800 |
|
}, |
|
{ |
|
"epoch": 4.93, |
|
"eval_loss": 0.10763510316610336, |
|
"eval_runtime": 627.2025, |
|
"eval_samples_per_second": 18.232, |
|
"eval_steps_per_second": 2.28, |
|
"eval_wer": 0.13772468586239928, |
|
"step": 58800 |
|
}, |
|
{ |
|
"epoch": 4.95, |
|
"learning_rate": 0.00012869446343130554, |
|
"loss": 0.3034, |
|
"step": 59000 |
|
}, |
|
{ |
|
"epoch": 4.95, |
|
"eval_loss": 0.10751395672559738, |
|
"eval_runtime": 627.4457, |
|
"eval_samples_per_second": 18.225, |
|
"eval_steps_per_second": 2.279, |
|
"eval_wer": 0.13743167859356512, |
|
"step": 59000 |
|
}, |
|
{ |
|
"epoch": 4.97, |
|
"learning_rate": 0.0001281085831461771, |
|
"loss": 0.3127, |
|
"step": 59200 |
|
}, |
|
{ |
|
"epoch": 4.97, |
|
"eval_loss": 0.10920178145170212, |
|
"eval_runtime": 626.5767, |
|
"eval_samples_per_second": 18.25, |
|
"eval_steps_per_second": 2.282, |
|
"eval_wer": 0.1391559136755508, |
|
"step": 59200 |
|
}, |
|
{ |
|
"epoch": 4.99, |
|
"learning_rate": 0.00012752270286104872, |
|
"loss": 0.3078, |
|
"step": 59400 |
|
}, |
|
{ |
|
"epoch": 4.99, |
|
"eval_loss": 0.10621806979179382, |
|
"eval_runtime": 627.2538, |
|
"eval_samples_per_second": 18.23, |
|
"eval_steps_per_second": 2.28, |
|
"eval_wer": 0.1387389417929791, |
|
"step": 59400 |
|
}, |
|
{ |
|
"epoch": 5.01, |
|
"learning_rate": 0.00012693682257592032, |
|
"loss": 0.2782, |
|
"step": 59600 |
|
}, |
|
{ |
|
"epoch": 5.01, |
|
"eval_loss": 0.1064351499080658, |
|
"eval_runtime": 627.6844, |
|
"eval_samples_per_second": 18.218, |
|
"eval_steps_per_second": 2.278, |
|
"eval_wer": 0.13708232377303206, |
|
"step": 59600 |
|
}, |
|
{ |
|
"epoch": 5.03, |
|
"learning_rate": 0.0001263509422907919, |
|
"loss": 0.2751, |
|
"step": 59800 |
|
}, |
|
{ |
|
"epoch": 5.03, |
|
"eval_loss": 0.10696742683649063, |
|
"eval_runtime": 622.6157, |
|
"eval_samples_per_second": 18.366, |
|
"eval_steps_per_second": 2.297, |
|
"eval_wer": 0.13444525835352453, |
|
"step": 59800 |
|
}, |
|
{ |
|
"epoch": 5.05, |
|
"learning_rate": 0.0001257650620056635, |
|
"loss": 0.2582, |
|
"step": 60000 |
|
}, |
|
{ |
|
"epoch": 5.05, |
|
"eval_loss": 0.10697264224290848, |
|
"eval_runtime": 630.057, |
|
"eval_samples_per_second": 18.149, |
|
"eval_steps_per_second": 2.27, |
|
"eval_wer": 0.13461430100862118, |
|
"step": 60000 |
|
}, |
|
{ |
|
"epoch": 5.07, |
|
"learning_rate": 0.00012517918172053508, |
|
"loss": 0.2641, |
|
"step": 60200 |
|
}, |
|
{ |
|
"epoch": 5.07, |
|
"eval_loss": 0.10726083815097809, |
|
"eval_runtime": 627.7279, |
|
"eval_samples_per_second": 18.216, |
|
"eval_steps_per_second": 2.278, |
|
"eval_wer": 0.13570744351157943, |
|
"step": 60200 |
|
}, |
|
{ |
|
"epoch": 5.09, |
|
"learning_rate": 0.00012459330143540668, |
|
"loss": 0.2709, |
|
"step": 60400 |
|
}, |
|
{ |
|
"epoch": 5.09, |
|
"eval_loss": 0.1082502007484436, |
|
"eval_runtime": 631.7872, |
|
"eval_samples_per_second": 18.099, |
|
"eval_steps_per_second": 2.263, |
|
"eval_wer": 0.13455795345692229, |
|
"step": 60400 |
|
}, |
|
{ |
|
"epoch": 5.11, |
|
"learning_rate": 0.0001240074211502783, |
|
"loss": 0.2834, |
|
"step": 60600 |
|
}, |
|
{ |
|
"epoch": 5.11, |
|
"eval_loss": 0.10695182532072067, |
|
"eval_runtime": 628.9494, |
|
"eval_samples_per_second": 18.181, |
|
"eval_steps_per_second": 2.274, |
|
"eval_wer": 0.13725136642812868, |
|
"step": 60600 |
|
}, |
|
{ |
|
"epoch": 5.13, |
|
"learning_rate": 0.00012342154086514986, |
|
"loss": 0.2723, |
|
"step": 60800 |
|
}, |
|
{ |
|
"epoch": 5.13, |
|
"eval_loss": 0.10714374482631683, |
|
"eval_runtime": 627.1877, |
|
"eval_samples_per_second": 18.232, |
|
"eval_steps_per_second": 2.28, |
|
"eval_wer": 0.1403166732405477, |
|
"step": 60800 |
|
}, |
|
{ |
|
"epoch": 5.02, |
|
"learning_rate": 0.00012283566058002147, |
|
"loss": 0.2895, |
|
"step": 61000 |
|
}, |
|
{ |
|
"epoch": 5.02, |
|
"eval_loss": 0.10630907863378525, |
|
"eval_runtime": 724.1222, |
|
"eval_samples_per_second": 15.792, |
|
"eval_steps_per_second": 1.975, |
|
"eval_wer": 0.1340395559812926, |
|
"step": 61000 |
|
}, |
|
{ |
|
"epoch": 5.04, |
|
"learning_rate": 0.00012224978029489307, |
|
"loss": 0.2791, |
|
"step": 61200 |
|
}, |
|
{ |
|
"epoch": 5.04, |
|
"eval_loss": 0.10776282101869583, |
|
"eval_runtime": 714.1344, |
|
"eval_samples_per_second": 16.012, |
|
"eval_steps_per_second": 2.002, |
|
"eval_wer": 0.13479461317405758, |
|
"step": 61200 |
|
}, |
|
{ |
|
"epoch": 5.06, |
|
"learning_rate": 0.00012166390000976466, |
|
"loss": 0.2689, |
|
"step": 61400 |
|
}, |
|
{ |
|
"epoch": 5.06, |
|
"eval_loss": 0.10609008371829987, |
|
"eval_runtime": 717.7674, |
|
"eval_samples_per_second": 15.931, |
|
"eval_steps_per_second": 1.992, |
|
"eval_wer": 0.13410717304333128, |
|
"step": 61400 |
|
}, |
|
{ |
|
"epoch": 5.08, |
|
"learning_rate": 0.00012107801972463625, |
|
"loss": 0.2639, |
|
"step": 61600 |
|
}, |
|
{ |
|
"epoch": 5.08, |
|
"eval_loss": 0.10747672617435455, |
|
"eval_runtime": 708.6123, |
|
"eval_samples_per_second": 16.137, |
|
"eval_steps_per_second": 2.018, |
|
"eval_wer": 0.13658646531808194, |
|
"step": 61600 |
|
}, |
|
{ |
|
"epoch": 5.1, |
|
"learning_rate": 0.00012049213943950784, |
|
"loss": 0.2811, |
|
"step": 61800 |
|
}, |
|
{ |
|
"epoch": 5.1, |
|
"eval_loss": 0.10639283061027527, |
|
"eval_runtime": 699.5143, |
|
"eval_samples_per_second": 16.347, |
|
"eval_steps_per_second": 2.044, |
|
"eval_wer": 0.13498619484983376, |
|
"step": 61800 |
|
}, |
|
{ |
|
"epoch": 5.12, |
|
"learning_rate": 0.00011990625915437945, |
|
"loss": 0.2708, |
|
"step": 62000 |
|
}, |
|
{ |
|
"epoch": 5.12, |
|
"eval_loss": 0.1074819564819336, |
|
"eval_runtime": 707.2617, |
|
"eval_samples_per_second": 16.168, |
|
"eval_steps_per_second": 2.022, |
|
"eval_wer": 0.13471572660167916, |
|
"step": 62000 |
|
}, |
|
{ |
|
"epoch": 5.14, |
|
"learning_rate": 0.00011932037886925104, |
|
"loss": 0.2711, |
|
"step": 62200 |
|
}, |
|
{ |
|
"epoch": 5.14, |
|
"eval_loss": 0.10513726621866226, |
|
"eval_runtime": 711.5872, |
|
"eval_samples_per_second": 16.07, |
|
"eval_steps_per_second": 2.01, |
|
"eval_wer": 0.13240547698202512, |
|
"step": 62200 |
|
}, |
|
{ |
|
"epoch": 5.16, |
|
"learning_rate": 0.00011873742798554829, |
|
"loss": 0.2739, |
|
"step": 62400 |
|
}, |
|
{ |
|
"epoch": 5.16, |
|
"eval_loss": 0.10640209913253784, |
|
"eval_runtime": 710.3202, |
|
"eval_samples_per_second": 16.098, |
|
"eval_steps_per_second": 2.013, |
|
"eval_wer": 0.1323265904096467, |
|
"step": 62400 |
|
}, |
|
{ |
|
"epoch": 5.17, |
|
"learning_rate": 0.00011815154770041988, |
|
"loss": 0.2737, |
|
"step": 62600 |
|
}, |
|
{ |
|
"epoch": 5.17, |
|
"eval_loss": 0.10621224343776703, |
|
"eval_runtime": 709.0146, |
|
"eval_samples_per_second": 16.128, |
|
"eval_steps_per_second": 2.017, |
|
"eval_wer": 0.13518904603594975, |
|
"step": 62600 |
|
}, |
|
{ |
|
"epoch": 5.19, |
|
"learning_rate": 0.00011756566741529147, |
|
"loss": 0.266, |
|
"step": 62800 |
|
}, |
|
{ |
|
"epoch": 5.19, |
|
"eval_loss": 0.10414595156908035, |
|
"eval_runtime": 703.9804, |
|
"eval_samples_per_second": 16.243, |
|
"eval_steps_per_second": 2.031, |
|
"eval_wer": 0.13109821378261113, |
|
"step": 62800 |
|
}, |
|
{ |
|
"epoch": 5.21, |
|
"learning_rate": 0.00011697978713016306, |
|
"loss": 0.272, |
|
"step": 63000 |
|
}, |
|
{ |
|
"epoch": 5.21, |
|
"eval_loss": 0.10723841935396194, |
|
"eval_runtime": 704.9142, |
|
"eval_samples_per_second": 16.222, |
|
"eval_steps_per_second": 2.029, |
|
"eval_wer": 0.13118836986532936, |
|
"step": 63000 |
|
}, |
|
{ |
|
"epoch": 5.23, |
|
"learning_rate": 0.00011639390684503466, |
|
"loss": 0.2631, |
|
"step": 63200 |
|
}, |
|
{ |
|
"epoch": 5.23, |
|
"eval_loss": 0.10395067185163498, |
|
"eval_runtime": 709.9351, |
|
"eval_samples_per_second": 16.107, |
|
"eval_steps_per_second": 2.014, |
|
"eval_wer": 0.13021919197610865, |
|
"step": 63200 |
|
}, |
|
{ |
|
"epoch": 5.25, |
|
"learning_rate": 0.00011580802655990625, |
|
"loss": 0.2677, |
|
"step": 63400 |
|
}, |
|
{ |
|
"epoch": 5.25, |
|
"eval_loss": 0.10302858054637909, |
|
"eval_runtime": 716.0431, |
|
"eval_samples_per_second": 15.97, |
|
"eval_steps_per_second": 1.997, |
|
"eval_wer": 0.13177438440299769, |
|
"step": 63400 |
|
}, |
|
{ |
|
"epoch": 5.27, |
|
"learning_rate": 0.00011522214627477784, |
|
"loss": 0.2537, |
|
"step": 63600 |
|
}, |
|
{ |
|
"epoch": 5.27, |
|
"eval_loss": 0.105635866522789, |
|
"eval_runtime": 717.4391, |
|
"eval_samples_per_second": 15.939, |
|
"eval_steps_per_second": 1.993, |
|
"eval_wer": 0.13281117935425707, |
|
"step": 63600 |
|
}, |
|
{ |
|
"epoch": 5.29, |
|
"learning_rate": 0.00011463626598964943, |
|
"loss": 0.2817, |
|
"step": 63800 |
|
}, |
|
{ |
|
"epoch": 5.29, |
|
"eval_loss": 0.10486125200986862, |
|
"eval_runtime": 713.1716, |
|
"eval_samples_per_second": 16.034, |
|
"eval_steps_per_second": 2.005, |
|
"eval_wer": 0.1323265904096467, |
|
"step": 63800 |
|
}, |
|
{ |
|
"epoch": 5.31, |
|
"learning_rate": 0.00011405038570452104, |
|
"loss": 0.261, |
|
"step": 64000 |
|
}, |
|
{ |
|
"epoch": 5.31, |
|
"eval_loss": 0.10378312319517136, |
|
"eval_runtime": 707.893, |
|
"eval_samples_per_second": 16.154, |
|
"eval_steps_per_second": 2.02, |
|
"eval_wer": 0.1305572772863019, |
|
"step": 64000 |
|
}, |
|
{ |
|
"epoch": 5.33, |
|
"learning_rate": 0.00011346450541939263, |
|
"loss": 0.2822, |
|
"step": 64200 |
|
}, |
|
{ |
|
"epoch": 5.33, |
|
"eval_loss": 0.10581466555595398, |
|
"eval_runtime": 716.437, |
|
"eval_samples_per_second": 15.961, |
|
"eval_steps_per_second": 1.996, |
|
"eval_wer": 0.12975714205217784, |
|
"step": 64200 |
|
}, |
|
{ |
|
"epoch": 5.35, |
|
"learning_rate": 0.00011287862513426422, |
|
"loss": 0.2725, |
|
"step": 64400 |
|
}, |
|
{ |
|
"epoch": 5.35, |
|
"eval_loss": 0.10494490712881088, |
|
"eval_runtime": 715.1416, |
|
"eval_samples_per_second": 15.99, |
|
"eval_steps_per_second": 2.0, |
|
"eval_wer": 0.13041077365188483, |
|
"step": 64400 |
|
}, |
|
{ |
|
"epoch": 5.37, |
|
"learning_rate": 0.00011229274484913581, |
|
"loss": 0.2601, |
|
"step": 64600 |
|
}, |
|
{ |
|
"epoch": 5.37, |
|
"eval_loss": 0.1038796454668045, |
|
"eval_runtime": 716.2768, |
|
"eval_samples_per_second": 15.964, |
|
"eval_steps_per_second": 1.996, |
|
"eval_wer": 0.12908097143179129, |
|
"step": 64600 |
|
}, |
|
{ |
|
"epoch": 5.39, |
|
"learning_rate": 0.0001117068645640074, |
|
"loss": 0.2637, |
|
"step": 64800 |
|
}, |
|
{ |
|
"epoch": 5.39, |
|
"eval_loss": 0.10526276379823685, |
|
"eval_runtime": 717.256, |
|
"eval_samples_per_second": 15.943, |
|
"eval_steps_per_second": 1.994, |
|
"eval_wer": 0.1309291711275145, |
|
"step": 64800 |
|
}, |
|
{ |
|
"epoch": 5.41, |
|
"learning_rate": 0.00011112098427887902, |
|
"loss": 0.2646, |
|
"step": 65000 |
|
}, |
|
{ |
|
"epoch": 5.41, |
|
"eval_loss": 0.10593494772911072, |
|
"eval_runtime": 711.2752, |
|
"eval_samples_per_second": 16.077, |
|
"eval_steps_per_second": 2.01, |
|
"eval_wer": 0.13076012847241789, |
|
"step": 65000 |
|
}, |
|
{ |
|
"epoch": 5.43, |
|
"learning_rate": 0.00011053510399375061, |
|
"loss": 0.2593, |
|
"step": 65200 |
|
}, |
|
{ |
|
"epoch": 5.43, |
|
"eval_loss": 0.10260586440563202, |
|
"eval_runtime": 710.5911, |
|
"eval_samples_per_second": 16.092, |
|
"eval_steps_per_second": 2.012, |
|
"eval_wer": 0.12839353130106496, |
|
"step": 65200 |
|
}, |
|
{ |
|
"epoch": 5.45, |
|
"learning_rate": 0.0001099492237086222, |
|
"loss": 0.2764, |
|
"step": 65400 |
|
}, |
|
{ |
|
"epoch": 5.45, |
|
"eval_loss": 0.1039929986000061, |
|
"eval_runtime": 711.9108, |
|
"eval_samples_per_second": 16.062, |
|
"eval_steps_per_second": 2.009, |
|
"eval_wer": 0.12882177269397643, |
|
"step": 65400 |
|
}, |
|
{ |
|
"epoch": 5.47, |
|
"learning_rate": 0.00010936334342349379, |
|
"loss": 0.2557, |
|
"step": 65600 |
|
}, |
|
{ |
|
"epoch": 5.47, |
|
"eval_loss": 0.10381390899419785, |
|
"eval_runtime": 718.0919, |
|
"eval_samples_per_second": 15.924, |
|
"eval_steps_per_second": 1.991, |
|
"eval_wer": 0.12913731898349018, |
|
"step": 65600 |
|
}, |
|
{ |
|
"epoch": 5.49, |
|
"learning_rate": 0.00010877746313836539, |
|
"loss": 0.2571, |
|
"step": 65800 |
|
}, |
|
{ |
|
"epoch": 5.49, |
|
"eval_loss": 0.10302343964576721, |
|
"eval_runtime": 711.5912, |
|
"eval_samples_per_second": 16.07, |
|
"eval_steps_per_second": 2.01, |
|
"eval_wer": 0.1304671212035837, |
|
"step": 65800 |
|
}, |
|
{ |
|
"epoch": 5.51, |
|
"learning_rate": 0.00010819158285323698, |
|
"loss": 0.2703, |
|
"step": 66000 |
|
}, |
|
{ |
|
"epoch": 5.51, |
|
"eval_loss": 0.1036553829908371, |
|
"eval_runtime": 714.6453, |
|
"eval_samples_per_second": 16.001, |
|
"eval_steps_per_second": 2.001, |
|
"eval_wer": 0.12699611201893277, |
|
"step": 66000 |
|
}, |
|
{ |
|
"epoch": 5.52, |
|
"learning_rate": 0.00010760570256810857, |
|
"loss": 0.2603, |
|
"step": 66200 |
|
}, |
|
{ |
|
"epoch": 5.52, |
|
"eval_loss": 0.102656789124012, |
|
"eval_runtime": 711.5823, |
|
"eval_samples_per_second": 16.07, |
|
"eval_steps_per_second": 2.01, |
|
"eval_wer": 0.12859638248718094, |
|
"step": 66200 |
|
}, |
|
{ |
|
"epoch": 5.54, |
|
"learning_rate": 0.00010701982228298016, |
|
"loss": 0.2501, |
|
"step": 66400 |
|
}, |
|
{ |
|
"epoch": 5.54, |
|
"eval_loss": 0.1020718440413475, |
|
"eval_runtime": 708.9234, |
|
"eval_samples_per_second": 16.13, |
|
"eval_steps_per_second": 2.017, |
|
"eval_wer": 0.12927255310756747, |
|
"step": 66400 |
|
}, |
|
{ |
|
"epoch": 5.56, |
|
"learning_rate": 0.00010643394199785177, |
|
"loss": 0.2615, |
|
"step": 66600 |
|
}, |
|
{ |
|
"epoch": 5.56, |
|
"eval_loss": 0.10426247119903564, |
|
"eval_runtime": 712.0372, |
|
"eval_samples_per_second": 16.06, |
|
"eval_steps_per_second": 2.008, |
|
"eval_wer": 0.1299036456865949, |
|
"step": 66600 |
|
}, |
|
{ |
|
"epoch": 5.58, |
|
"learning_rate": 0.00010584806171272336, |
|
"loss": 0.2669, |
|
"step": 66800 |
|
}, |
|
{ |
|
"epoch": 5.58, |
|
"eval_loss": 0.10160184651613235, |
|
"eval_runtime": 715.9766, |
|
"eval_samples_per_second": 15.971, |
|
"eval_steps_per_second": 1.997, |
|
"eval_wer": 0.12859638248718094, |
|
"step": 66800 |
|
}, |
|
{ |
|
"epoch": 5.6, |
|
"learning_rate": 0.00010526218142759495, |
|
"loss": 0.2737, |
|
"step": 67000 |
|
}, |
|
{ |
|
"epoch": 5.6, |
|
"eval_loss": 0.0998615026473999, |
|
"eval_runtime": 721.1944, |
|
"eval_samples_per_second": 15.856, |
|
"eval_steps_per_second": 1.983, |
|
"eval_wer": 0.1280554459908717, |
|
"step": 67000 |
|
}, |
|
{ |
|
"epoch": 5.62, |
|
"learning_rate": 0.0001046792305438922, |
|
"loss": 0.2686, |
|
"step": 67200 |
|
}, |
|
{ |
|
"epoch": 5.62, |
|
"eval_loss": 0.1005900576710701, |
|
"eval_runtime": 720.5097, |
|
"eval_samples_per_second": 15.871, |
|
"eval_steps_per_second": 1.985, |
|
"eval_wer": 0.12931763114892658, |
|
"step": 67200 |
|
}, |
|
{ |
|
"epoch": 5.64, |
|
"learning_rate": 0.00010409335025876379, |
|
"loss": 0.2638, |
|
"step": 67400 |
|
}, |
|
{ |
|
"epoch": 5.64, |
|
"eval_loss": 0.10066718608140945, |
|
"eval_runtime": 716.9838, |
|
"eval_samples_per_second": 15.949, |
|
"eval_steps_per_second": 1.994, |
|
"eval_wer": 0.12787513382543528, |
|
"step": 67400 |
|
}, |
|
{ |
|
"epoch": 5.66, |
|
"learning_rate": 0.00010351039937506102, |
|
"loss": 0.2657, |
|
"step": 67600 |
|
}, |
|
{ |
|
"epoch": 5.66, |
|
"eval_loss": 0.10130422562360764, |
|
"eval_runtime": 718.8149, |
|
"eval_samples_per_second": 15.908, |
|
"eval_steps_per_second": 1.989, |
|
"eval_wer": 0.12775116921169774, |
|
"step": 67600 |
|
}, |
|
{ |
|
"epoch": 5.68, |
|
"learning_rate": 0.00010292451908993261, |
|
"loss": 0.2648, |
|
"step": 67800 |
|
}, |
|
{ |
|
"epoch": 5.68, |
|
"eval_loss": 0.10149582475423813, |
|
"eval_runtime": 721.2563, |
|
"eval_samples_per_second": 15.854, |
|
"eval_steps_per_second": 1.983, |
|
"eval_wer": 0.12784132529441597, |
|
"step": 67800 |
|
}, |
|
{ |
|
"epoch": 5.7, |
|
"learning_rate": 0.0001023386388048042, |
|
"loss": 0.2524, |
|
"step": 68000 |
|
}, |
|
{ |
|
"epoch": 5.7, |
|
"eval_loss": 0.09992196410894394, |
|
"eval_runtime": 715.5922, |
|
"eval_samples_per_second": 15.98, |
|
"eval_steps_per_second": 1.998, |
|
"eval_wer": 0.12749197047388292, |
|
"step": 68000 |
|
}, |
|
{ |
|
"epoch": 5.72, |
|
"learning_rate": 0.0001017527585196758, |
|
"loss": 0.2546, |
|
"step": 68200 |
|
}, |
|
{ |
|
"epoch": 5.72, |
|
"eval_loss": 0.09948063641786575, |
|
"eval_runtime": 707.0025, |
|
"eval_samples_per_second": 16.174, |
|
"eval_steps_per_second": 2.023, |
|
"eval_wer": 0.12589170000563477, |
|
"step": 68200 |
|
}, |
|
{ |
|
"epoch": 5.74, |
|
"learning_rate": 0.0001011668782345474, |
|
"loss": 0.2622, |
|
"step": 68400 |
|
}, |
|
{ |
|
"epoch": 5.74, |
|
"eval_loss": 0.10015752166509628, |
|
"eval_runtime": 713.843, |
|
"eval_samples_per_second": 16.019, |
|
"eval_steps_per_second": 2.003, |
|
"eval_wer": 0.12627486335718713, |
|
"step": 68400 |
|
}, |
|
{ |
|
"epoch": 5.76, |
|
"learning_rate": 0.00010058099794941899, |
|
"loss": 0.2604, |
|
"step": 68600 |
|
}, |
|
{ |
|
"epoch": 5.76, |
|
"eval_loss": 0.1018703505396843, |
|
"eval_runtime": 721.4552, |
|
"eval_samples_per_second": 15.85, |
|
"eval_steps_per_second": 1.982, |
|
"eval_wer": 0.1254071110610244, |
|
"step": 68600 |
|
}, |
|
{ |
|
"epoch": 5.78, |
|
"learning_rate": 9.999511766429058e-05, |
|
"loss": 0.2458, |
|
"step": 68800 |
|
}, |
|
{ |
|
"epoch": 5.78, |
|
"eval_loss": 0.1008736789226532, |
|
"eval_runtime": 701.5996, |
|
"eval_samples_per_second": 16.298, |
|
"eval_steps_per_second": 2.038, |
|
"eval_wer": 0.12576773539189723, |
|
"step": 68800 |
|
}, |
|
{ |
|
"epoch": 5.8, |
|
"learning_rate": 9.940923737916218e-05, |
|
"loss": 0.2608, |
|
"step": 69000 |
|
}, |
|
{ |
|
"epoch": 5.8, |
|
"eval_loss": 0.10070067644119263, |
|
"eval_runtime": 712.046, |
|
"eval_samples_per_second": 16.059, |
|
"eval_steps_per_second": 2.008, |
|
"eval_wer": 0.12631994139854624, |
|
"step": 69000 |
|
}, |
|
{ |
|
"epoch": 5.82, |
|
"learning_rate": 9.882335709403377e-05, |
|
"loss": 0.2633, |
|
"step": 69200 |
|
}, |
|
{ |
|
"epoch": 5.82, |
|
"eval_loss": 0.09837665408849716, |
|
"eval_runtime": 708.3742, |
|
"eval_samples_per_second": 16.143, |
|
"eval_steps_per_second": 2.019, |
|
"eval_wer": 0.12583535245393587, |
|
"step": 69200 |
|
}, |
|
{ |
|
"epoch": 5.84, |
|
"learning_rate": 9.824040621033102e-05, |
|
"loss": 0.2526, |
|
"step": 69400 |
|
}, |
|
{ |
|
"epoch": 5.84, |
|
"eval_loss": 0.10069932788610458, |
|
"eval_runtime": 725.7524, |
|
"eval_samples_per_second": 15.756, |
|
"eval_steps_per_second": 1.97, |
|
"eval_wer": 0.12542965008170395, |
|
"step": 69400 |
|
}, |
|
{ |
|
"epoch": 5.86, |
|
"learning_rate": 9.765452592520261e-05, |
|
"loss": 0.2476, |
|
"step": 69600 |
|
}, |
|
{ |
|
"epoch": 5.86, |
|
"eval_loss": 0.10035388171672821, |
|
"eval_runtime": 710.9719, |
|
"eval_samples_per_second": 16.084, |
|
"eval_steps_per_second": 2.011, |
|
"eval_wer": 0.12496760015777314, |
|
"step": 69600 |
|
}, |
|
{ |
|
"epoch": 5.87, |
|
"learning_rate": 9.70686456400742e-05, |
|
"loss": 0.2607, |
|
"step": 69800 |
|
}, |
|
{ |
|
"epoch": 5.87, |
|
"eval_loss": 0.10133133828639984, |
|
"eval_runtime": 715.255, |
|
"eval_samples_per_second": 15.987, |
|
"eval_steps_per_second": 1.999, |
|
"eval_wer": 0.12650025356398265, |
|
"step": 69800 |
|
}, |
|
{ |
|
"epoch": 5.89, |
|
"learning_rate": 9.648276535494579e-05, |
|
"loss": 0.2606, |
|
"step": 70000 |
|
}, |
|
{ |
|
"epoch": 5.89, |
|
"eval_loss": 0.09976745396852493, |
|
"eval_runtime": 710.9411, |
|
"eval_samples_per_second": 16.084, |
|
"eval_steps_per_second": 2.011, |
|
"eval_wer": 0.1242914295373866, |
|
"step": 70000 |
|
}, |
|
{ |
|
"epoch": 5.91, |
|
"learning_rate": 9.58968850698174e-05, |
|
"loss": 0.258, |
|
"step": 70200 |
|
}, |
|
{ |
|
"epoch": 5.91, |
|
"eval_loss": 0.09873379021883011, |
|
"eval_runtime": 708.9039, |
|
"eval_samples_per_second": 16.131, |
|
"eval_steps_per_second": 2.017, |
|
"eval_wer": 0.12497886966811292, |
|
"step": 70200 |
|
}, |
|
{ |
|
"epoch": 5.93, |
|
"learning_rate": 9.531100478468898e-05, |
|
"loss": 0.262, |
|
"step": 70400 |
|
}, |
|
{ |
|
"epoch": 5.93, |
|
"eval_loss": 0.09941908717155457, |
|
"eval_runtime": 713.9341, |
|
"eval_samples_per_second": 16.017, |
|
"eval_steps_per_second": 2.003, |
|
"eval_wer": 0.12625232433650757, |
|
"step": 70400 |
|
}, |
|
{ |
|
"epoch": 5.95, |
|
"learning_rate": 9.472512449956058e-05, |
|
"loss": 0.2684, |
|
"step": 70600 |
|
}, |
|
{ |
|
"epoch": 5.95, |
|
"eval_loss": 0.09816844016313553, |
|
"eval_runtime": 710.6891, |
|
"eval_samples_per_second": 16.09, |
|
"eval_steps_per_second": 2.012, |
|
"eval_wer": 0.12258973347608046, |
|
"step": 70600 |
|
}, |
|
{ |
|
"epoch": 6.02, |
|
"learning_rate": 9.413924421443217e-05, |
|
"loss": 0.247, |
|
"step": 70800 |
|
}, |
|
{ |
|
"epoch": 6.02, |
|
"eval_loss": 0.10028348863124847, |
|
"eval_runtime": 662.0676, |
|
"eval_samples_per_second": 17.272, |
|
"eval_steps_per_second": 2.16, |
|
"eval_wer": 0.12441539415112414, |
|
"step": 70800 |
|
}, |
|
{ |
|
"epoch": 6.04, |
|
"learning_rate": 9.355336392930377e-05, |
|
"loss": 0.2482, |
|
"step": 71000 |
|
}, |
|
{ |
|
"epoch": 6.04, |
|
"eval_loss": 0.0990142971277237, |
|
"eval_runtime": 649.2239, |
|
"eval_samples_per_second": 17.613, |
|
"eval_steps_per_second": 2.203, |
|
"eval_wer": 0.1230405138896715, |
|
"step": 71000 |
|
}, |
|
{ |
|
"epoch": 6.06, |
|
"learning_rate": 9.296748364417536e-05, |
|
"loss": 0.2385, |
|
"step": 71200 |
|
}, |
|
{ |
|
"epoch": 6.06, |
|
"eval_loss": 0.09822484850883484, |
|
"eval_runtime": 660.7912, |
|
"eval_samples_per_second": 17.305, |
|
"eval_steps_per_second": 2.164, |
|
"eval_wer": 0.12372795402039781, |
|
"step": 71200 |
|
}, |
|
{ |
|
"epoch": 6.08, |
|
"learning_rate": 9.238160335904695e-05, |
|
"loss": 0.2406, |
|
"step": 71400 |
|
}, |
|
{ |
|
"epoch": 6.08, |
|
"eval_loss": 0.09724367409944534, |
|
"eval_runtime": 660.8872, |
|
"eval_samples_per_second": 17.302, |
|
"eval_steps_per_second": 2.164, |
|
"eval_wer": 0.12193610187637346, |
|
"step": 71400 |
|
}, |
|
{ |
|
"epoch": 6.1, |
|
"learning_rate": 9.179572307391854e-05, |
|
"loss": 0.2579, |
|
"step": 71600 |
|
}, |
|
{ |
|
"epoch": 6.1, |
|
"eval_loss": 0.0971040278673172, |
|
"eval_runtime": 674.9194, |
|
"eval_samples_per_second": 16.943, |
|
"eval_steps_per_second": 2.119, |
|
"eval_wer": 0.12212768355214966, |
|
"step": 71600 |
|
}, |
|
{ |
|
"epoch": 6.12, |
|
"learning_rate": 9.120984278879016e-05, |
|
"loss": 0.2512, |
|
"step": 71800 |
|
}, |
|
{ |
|
"epoch": 6.12, |
|
"eval_loss": 0.09761016815900803, |
|
"eval_runtime": 670.6191, |
|
"eval_samples_per_second": 17.051, |
|
"eval_steps_per_second": 2.132, |
|
"eval_wer": 0.12180086775229616, |
|
"step": 71800 |
|
}, |
|
{ |
|
"epoch": 6.14, |
|
"learning_rate": 9.062396250366175e-05, |
|
"loss": 0.2435, |
|
"step": 72000 |
|
}, |
|
{ |
|
"epoch": 6.14, |
|
"eval_loss": 0.09776688367128372, |
|
"eval_runtime": 668.7679, |
|
"eval_samples_per_second": 17.099, |
|
"eval_steps_per_second": 2.138, |
|
"eval_wer": 0.12095565447681299, |
|
"step": 72000 |
|
}, |
|
{ |
|
"epoch": 6.16, |
|
"learning_rate": 9.003808221853334e-05, |
|
"loss": 0.2526, |
|
"step": 72200 |
|
}, |
|
{ |
|
"epoch": 6.16, |
|
"eval_loss": 0.09796804934740067, |
|
"eval_runtime": 666.8275, |
|
"eval_samples_per_second": 17.148, |
|
"eval_steps_per_second": 2.144, |
|
"eval_wer": 0.1238744576548149, |
|
"step": 72200 |
|
}, |
|
{ |
|
"epoch": 6.17, |
|
"learning_rate": 8.945220193340493e-05, |
|
"loss": 0.2472, |
|
"step": 72400 |
|
}, |
|
{ |
|
"epoch": 6.17, |
|
"eval_loss": 0.09801238030195236, |
|
"eval_runtime": 673.7542, |
|
"eval_samples_per_second": 16.972, |
|
"eval_steps_per_second": 2.122, |
|
"eval_wer": 0.12542965008170395, |
|
"step": 72400 |
|
}, |
|
{ |
|
"epoch": 6.19, |
|
"learning_rate": 8.886632164827653e-05, |
|
"loss": 0.2438, |
|
"step": 72600 |
|
}, |
|
{ |
|
"epoch": 6.19, |
|
"eval_loss": 0.0977615937590599, |
|
"eval_runtime": 664.5529, |
|
"eval_samples_per_second": 17.207, |
|
"eval_steps_per_second": 2.152, |
|
"eval_wer": 0.12182340677297572, |
|
"step": 72600 |
|
}, |
|
{ |
|
"epoch": 6.21, |
|
"learning_rate": 8.828044136314813e-05, |
|
"loss": 0.2375, |
|
"step": 72800 |
|
}, |
|
{ |
|
"epoch": 6.21, |
|
"eval_loss": 0.09897564351558685, |
|
"eval_runtime": 659.823, |
|
"eval_samples_per_second": 17.33, |
|
"eval_steps_per_second": 2.167, |
|
"eval_wer": 0.12198117991773258, |
|
"step": 72800 |
|
}, |
|
{ |
|
"epoch": 6.23, |
|
"learning_rate": 8.769456107801972e-05, |
|
"loss": 0.2419, |
|
"step": 73000 |
|
}, |
|
{ |
|
"epoch": 6.23, |
|
"eval_loss": 0.09719140082597733, |
|
"eval_runtime": 670.2577, |
|
"eval_samples_per_second": 17.061, |
|
"eval_steps_per_second": 2.134, |
|
"eval_wer": 0.11912999380176931, |
|
"step": 73000 |
|
}, |
|
{ |
|
"epoch": 6.25, |
|
"learning_rate": 8.71086807928913e-05, |
|
"loss": 0.2441, |
|
"step": 73200 |
|
}, |
|
{ |
|
"epoch": 6.25, |
|
"eval_loss": 0.09728872776031494, |
|
"eval_runtime": 692.8136, |
|
"eval_samples_per_second": 16.505, |
|
"eval_steps_per_second": 2.064, |
|
"eval_wer": 0.12141770440074379, |
|
"step": 73200 |
|
}, |
|
{ |
|
"epoch": 6.27, |
|
"learning_rate": 8.652280050776291e-05, |
|
"loss": 0.2314, |
|
"step": 73400 |
|
}, |
|
{ |
|
"epoch": 6.27, |
|
"eval_loss": 0.09700626879930496, |
|
"eval_runtime": 686.6118, |
|
"eval_samples_per_second": 16.654, |
|
"eval_steps_per_second": 2.083, |
|
"eval_wer": 0.12061756916661971, |
|
"step": 73400 |
|
}, |
|
{ |
|
"epoch": 6.29, |
|
"learning_rate": 8.59369202226345e-05, |
|
"loss": 0.2533, |
|
"step": 73600 |
|
}, |
|
{ |
|
"epoch": 6.29, |
|
"eval_loss": 0.09654780477285385, |
|
"eval_runtime": 683.2657, |
|
"eval_samples_per_second": 16.736, |
|
"eval_steps_per_second": 2.093, |
|
"eval_wer": 0.1190398377190511, |
|
"step": 73600 |
|
}, |
|
{ |
|
"epoch": 6.31, |
|
"learning_rate": 8.535103993750609e-05, |
|
"loss": 0.2357, |
|
"step": 73800 |
|
}, |
|
{ |
|
"epoch": 6.31, |
|
"eval_loss": 0.09661402553319931, |
|
"eval_runtime": 682.2598, |
|
"eval_samples_per_second": 16.76, |
|
"eval_steps_per_second": 2.096, |
|
"eval_wer": 0.12017805826336846, |
|
"step": 73800 |
|
}, |
|
{ |
|
"epoch": 6.33, |
|
"learning_rate": 8.476515965237768e-05, |
|
"loss": 0.2466, |
|
"step": 74000 |
|
}, |
|
{ |
|
"epoch": 6.33, |
|
"eval_loss": 0.09645663946866989, |
|
"eval_runtime": 689.4454, |
|
"eval_samples_per_second": 16.586, |
|
"eval_steps_per_second": 2.074, |
|
"eval_wer": 0.11943427058094326, |
|
"step": 74000 |
|
}, |
|
{ |
|
"epoch": 6.35, |
|
"learning_rate": 8.417927936724929e-05, |
|
"loss": 0.246, |
|
"step": 74200 |
|
}, |
|
{ |
|
"epoch": 6.35, |
|
"eval_loss": 0.09545984119176865, |
|
"eval_runtime": 679.0158, |
|
"eval_samples_per_second": 16.841, |
|
"eval_steps_per_second": 2.106, |
|
"eval_wer": 0.11880317800191581, |
|
"step": 74200 |
|
}, |
|
{ |
|
"epoch": 6.37, |
|
"learning_rate": 8.359339908212088e-05, |
|
"loss": 0.2309, |
|
"step": 74400 |
|
}, |
|
{ |
|
"epoch": 6.37, |
|
"eval_loss": 0.09694214910268784, |
|
"eval_runtime": 677.6907, |
|
"eval_samples_per_second": 16.873, |
|
"eval_steps_per_second": 2.11, |
|
"eval_wer": 0.11869048289851807, |
|
"step": 74400 |
|
}, |
|
{ |
|
"epoch": 6.39, |
|
"learning_rate": 8.300751879699247e-05, |
|
"loss": 0.2334, |
|
"step": 74600 |
|
}, |
|
{ |
|
"epoch": 6.39, |
|
"eval_loss": 0.0962836816906929, |
|
"eval_runtime": 683.8832, |
|
"eval_samples_per_second": 16.721, |
|
"eval_steps_per_second": 2.091, |
|
"eval_wer": 0.12112469713190961, |
|
"step": 74600 |
|
}, |
|
{ |
|
"epoch": 6.41, |
|
"learning_rate": 8.242163851186406e-05, |
|
"loss": 0.2343, |
|
"step": 74800 |
|
}, |
|
{ |
|
"epoch": 6.41, |
|
"eval_loss": 0.09781304746866226, |
|
"eval_runtime": 685.7457, |
|
"eval_samples_per_second": 16.675, |
|
"eval_steps_per_second": 2.085, |
|
"eval_wer": 0.11866794387783851, |
|
"step": 74800 |
|
}, |
|
{ |
|
"epoch": 6.43, |
|
"learning_rate": 8.183575822673568e-05, |
|
"loss": 0.2387, |
|
"step": 75000 |
|
}, |
|
{ |
|
"epoch": 6.43, |
|
"eval_loss": 0.09602531045675278, |
|
"eval_runtime": 672.2565, |
|
"eval_samples_per_second": 17.01, |
|
"eval_steps_per_second": 2.127, |
|
"eval_wer": 0.11829605003662591, |
|
"step": 75000 |
|
}, |
|
{ |
|
"epoch": 6.45, |
|
"learning_rate": 8.124987794160727e-05, |
|
"loss": 0.25, |
|
"step": 75200 |
|
}, |
|
{ |
|
"epoch": 6.45, |
|
"eval_loss": 0.09615187346935272, |
|
"eval_runtime": 682.05, |
|
"eval_samples_per_second": 16.766, |
|
"eval_steps_per_second": 2.097, |
|
"eval_wer": 0.1192764974361864, |
|
"step": 75200 |
|
}, |
|
{ |
|
"epoch": 6.47, |
|
"learning_rate": 8.06669270579045e-05, |
|
"loss": 0.2317, |
|
"step": 75400 |
|
}, |
|
{ |
|
"epoch": 6.47, |
|
"eval_loss": 0.09546995908021927, |
|
"eval_runtime": 690.2183, |
|
"eval_samples_per_second": 16.567, |
|
"eval_steps_per_second": 2.072, |
|
"eval_wer": 0.11836366709866457, |
|
"step": 75400 |
|
}, |
|
{ |
|
"epoch": 6.49, |
|
"learning_rate": 8.008104677277609e-05, |
|
"loss": 0.2301, |
|
"step": 75600 |
|
}, |
|
{ |
|
"epoch": 6.49, |
|
"eval_loss": 0.09542080760002136, |
|
"eval_runtime": 681.8498, |
|
"eval_samples_per_second": 16.771, |
|
"eval_steps_per_second": 2.097, |
|
"eval_wer": 0.11834112807798501, |
|
"step": 75600 |
|
}, |
|
{ |
|
"epoch": 6.51, |
|
"learning_rate": 7.949516648764768e-05, |
|
"loss": 0.2405, |
|
"step": 75800 |
|
}, |
|
{ |
|
"epoch": 6.51, |
|
"eval_loss": 0.09432049840688705, |
|
"eval_runtime": 677.928, |
|
"eval_samples_per_second": 16.868, |
|
"eval_steps_per_second": 2.109, |
|
"eval_wer": 0.11672958809939708, |
|
"step": 75800 |
|
}, |
|
{ |
|
"epoch": 6.52, |
|
"learning_rate": 7.891221560394493e-05, |
|
"loss": 0.2345, |
|
"step": 76000 |
|
}, |
|
{ |
|
"epoch": 6.52, |
|
"eval_loss": 0.09489911794662476, |
|
"eval_runtime": 688.6398, |
|
"eval_samples_per_second": 16.605, |
|
"eval_steps_per_second": 2.077, |
|
"eval_wer": 0.11725925508536654, |
|
"step": 76000 |
|
}, |
|
{ |
|
"epoch": 6.54, |
|
"learning_rate": 7.832633531881652e-05, |
|
"loss": 0.2267, |
|
"step": 76200 |
|
}, |
|
{ |
|
"epoch": 6.54, |
|
"eval_loss": 0.09413082152605057, |
|
"eval_runtime": 680.2461, |
|
"eval_samples_per_second": 16.81, |
|
"eval_steps_per_second": 2.102, |
|
"eval_wer": 0.11842001465036345, |
|
"step": 76200 |
|
}, |
|
{ |
|
"epoch": 6.56, |
|
"learning_rate": 7.774045503368811e-05, |
|
"loss": 0.2293, |
|
"step": 76400 |
|
}, |
|
{ |
|
"epoch": 6.56, |
|
"eval_loss": 0.09459416568279266, |
|
"eval_runtime": 658.8476, |
|
"eval_samples_per_second": 17.356, |
|
"eval_steps_per_second": 2.17, |
|
"eval_wer": 0.11623372964444695, |
|
"step": 76400 |
|
}, |
|
{ |
|
"epoch": 6.58, |
|
"learning_rate": 7.715457474855971e-05, |
|
"loss": 0.2312, |
|
"step": 76600 |
|
}, |
|
{ |
|
"epoch": 6.58, |
|
"eval_loss": 0.09610540419816971, |
|
"eval_runtime": 661.5363, |
|
"eval_samples_per_second": 17.286, |
|
"eval_steps_per_second": 2.162, |
|
"eval_wer": 0.1151405871414887, |
|
"step": 76600 |
|
}, |
|
{ |
|
"epoch": 6.6, |
|
"learning_rate": 7.65686944634313e-05, |
|
"loss": 0.2343, |
|
"step": 76800 |
|
}, |
|
{ |
|
"epoch": 6.6, |
|
"eval_loss": 0.09503426402807236, |
|
"eval_runtime": 680.9869, |
|
"eval_samples_per_second": 16.792, |
|
"eval_steps_per_second": 2.1, |
|
"eval_wer": 0.11631261621682538, |
|
"step": 76800 |
|
}, |
|
{ |
|
"epoch": 6.62, |
|
"learning_rate": 7.59828141783029e-05, |
|
"loss": 0.2366, |
|
"step": 77000 |
|
}, |
|
{ |
|
"epoch": 6.62, |
|
"eval_loss": 0.09373170137405396, |
|
"eval_runtime": 677.558, |
|
"eval_samples_per_second": 16.877, |
|
"eval_steps_per_second": 2.111, |
|
"eval_wer": 0.11597453090663211, |
|
"step": 77000 |
|
}, |
|
{ |
|
"epoch": 6.64, |
|
"learning_rate": 7.539693389317448e-05, |
|
"loss": 0.2439, |
|
"step": 77200 |
|
}, |
|
{ |
|
"epoch": 6.64, |
|
"eval_loss": 0.09325527399778366, |
|
"eval_runtime": 664.5936, |
|
"eval_samples_per_second": 17.206, |
|
"eval_steps_per_second": 2.152, |
|
"eval_wer": 0.11638023327886403, |
|
"step": 77200 |
|
}, |
|
{ |
|
"epoch": 6.66, |
|
"learning_rate": 7.481105360804607e-05, |
|
"loss": 0.2307, |
|
"step": 77400 |
|
}, |
|
{ |
|
"epoch": 6.66, |
|
"eval_loss": 0.09506206959486008, |
|
"eval_runtime": 678.3069, |
|
"eval_samples_per_second": 16.858, |
|
"eval_steps_per_second": 2.108, |
|
"eval_wer": 0.1172817941060461, |
|
"step": 77400 |
|
}, |
|
{ |
|
"epoch": 6.68, |
|
"learning_rate": 7.422517332291768e-05, |
|
"loss": 0.2387, |
|
"step": 77600 |
|
}, |
|
{ |
|
"epoch": 6.68, |
|
"eval_loss": 0.0931260883808136, |
|
"eval_runtime": 685.8319, |
|
"eval_samples_per_second": 16.673, |
|
"eval_steps_per_second": 2.085, |
|
"eval_wer": 0.11590691384459345, |
|
"step": 77600 |
|
}, |
|
{ |
|
"epoch": 6.7, |
|
"learning_rate": 7.363929303778927e-05, |
|
"loss": 0.2268, |
|
"step": 77800 |
|
}, |
|
{ |
|
"epoch": 6.7, |
|
"eval_loss": 0.0957166850566864, |
|
"eval_runtime": 683.3247, |
|
"eval_samples_per_second": 16.734, |
|
"eval_steps_per_second": 2.093, |
|
"eval_wer": 0.11711275145094946, |
|
"step": 77800 |
|
}, |
|
{ |
|
"epoch": 6.72, |
|
"learning_rate": 7.305341275266087e-05, |
|
"loss": 0.2316, |
|
"step": 78000 |
|
}, |
|
{ |
|
"epoch": 6.72, |
|
"eval_loss": 0.09374788403511047, |
|
"eval_runtime": 684.0333, |
|
"eval_samples_per_second": 16.717, |
|
"eval_steps_per_second": 2.091, |
|
"eval_wer": 0.1139798275764918, |
|
"step": 78000 |
|
}, |
|
{ |
|
"epoch": 6.74, |
|
"learning_rate": 7.246753246753246e-05, |
|
"loss": 0.2319, |
|
"step": 78200 |
|
}, |
|
{ |
|
"epoch": 6.74, |
|
"eval_loss": 0.09412852674722672, |
|
"eval_runtime": 683.7255, |
|
"eval_samples_per_second": 16.725, |
|
"eval_steps_per_second": 2.091, |
|
"eval_wer": 0.1150504310587705, |
|
"step": 78200 |
|
}, |
|
{ |
|
"epoch": 6.76, |
|
"learning_rate": 7.188165218240405e-05, |
|
"loss": 0.2399, |
|
"step": 78400 |
|
}, |
|
{ |
|
"epoch": 6.76, |
|
"eval_loss": 0.09368865191936493, |
|
"eval_runtime": 695.8, |
|
"eval_samples_per_second": 16.434, |
|
"eval_steps_per_second": 2.055, |
|
"eval_wer": 0.11510677861046938, |
|
"step": 78400 |
|
}, |
|
{ |
|
"epoch": 6.78, |
|
"learning_rate": 7.129577189727565e-05, |
|
"loss": 0.2282, |
|
"step": 78600 |
|
}, |
|
{ |
|
"epoch": 6.78, |
|
"eval_loss": 0.09468115866184235, |
|
"eval_runtime": 692.8715, |
|
"eval_samples_per_second": 16.504, |
|
"eval_steps_per_second": 2.064, |
|
"eval_wer": 0.11432918239702484, |
|
"step": 78600 |
|
}, |
|
{ |
|
"epoch": 6.8, |
|
"learning_rate": 7.070989161214725e-05, |
|
"loss": 0.2406, |
|
"step": 78800 |
|
}, |
|
{ |
|
"epoch": 6.8, |
|
"eval_loss": 0.09301230311393738, |
|
"eval_runtime": 675.0001, |
|
"eval_samples_per_second": 16.941, |
|
"eval_steps_per_second": 2.119, |
|
"eval_wer": 0.11670704907871753, |
|
"step": 78800 |
|
}, |
|
{ |
|
"epoch": 6.82, |
|
"learning_rate": 7.012401132701884e-05, |
|
"loss": 0.2367, |
|
"step": 79000 |
|
}, |
|
{ |
|
"epoch": 6.82, |
|
"eval_loss": 0.09292880445718765, |
|
"eval_runtime": 678.9288, |
|
"eval_samples_per_second": 16.843, |
|
"eval_steps_per_second": 2.106, |
|
"eval_wer": 0.11399109708683158, |
|
"step": 79000 |
|
}, |
|
{ |
|
"epoch": 6.84, |
|
"learning_rate": 6.953813104189044e-05, |
|
"loss": 0.2225, |
|
"step": 79200 |
|
}, |
|
{ |
|
"epoch": 6.84, |
|
"eval_loss": 0.09368370473384857, |
|
"eval_runtime": 662.1759, |
|
"eval_samples_per_second": 17.269, |
|
"eval_steps_per_second": 2.16, |
|
"eval_wer": 0.11385586296275427, |
|
"step": 79200 |
|
}, |
|
{ |
|
"epoch": 6.86, |
|
"learning_rate": 6.895225075676203e-05, |
|
"loss": 0.2242, |
|
"step": 79400 |
|
}, |
|
{ |
|
"epoch": 6.86, |
|
"eval_loss": 0.09303466975688934, |
|
"eval_runtime": 669.81, |
|
"eval_samples_per_second": 17.072, |
|
"eval_steps_per_second": 2.135, |
|
"eval_wer": 0.11382205443173495, |
|
"step": 79400 |
|
}, |
|
{ |
|
"epoch": 6.87, |
|
"learning_rate": 6.836637047163362e-05, |
|
"loss": 0.2319, |
|
"step": 79600 |
|
}, |
|
{ |
|
"epoch": 6.87, |
|
"eval_loss": 0.09515677392482758, |
|
"eval_runtime": 675.7005, |
|
"eval_samples_per_second": 16.923, |
|
"eval_steps_per_second": 2.116, |
|
"eval_wer": 0.11462218966585902, |
|
"step": 79600 |
|
}, |
|
{ |
|
"epoch": 6.89, |
|
"learning_rate": 6.778049018650522e-05, |
|
"loss": 0.2383, |
|
"step": 79800 |
|
}, |
|
{ |
|
"epoch": 6.89, |
|
"eval_loss": 0.09353630244731903, |
|
"eval_runtime": 680.3127, |
|
"eval_samples_per_second": 16.808, |
|
"eval_steps_per_second": 2.102, |
|
"eval_wer": 0.11421648729362709, |
|
"step": 79800 |
|
}, |
|
{ |
|
"epoch": 6.91, |
|
"learning_rate": 6.719460990137682e-05, |
|
"loss": 0.228, |
|
"step": 80000 |
|
}, |
|
{ |
|
"epoch": 6.91, |
|
"eval_loss": 0.09211862087249756, |
|
"eval_runtime": 677.5612, |
|
"eval_samples_per_second": 16.877, |
|
"eval_steps_per_second": 2.111, |
|
"eval_wer": 0.11403617512819068, |
|
"step": 80000 |
|
}, |
|
{ |
|
"epoch": 6.93, |
|
"learning_rate": 6.660872961624841e-05, |
|
"loss": 0.232, |
|
"step": 80200 |
|
}, |
|
{ |
|
"epoch": 6.93, |
|
"eval_loss": 0.09219860285520554, |
|
"eval_runtime": 667.4301, |
|
"eval_samples_per_second": 17.133, |
|
"eval_steps_per_second": 2.143, |
|
"eval_wer": 0.11532089930692511, |
|
"step": 80200 |
|
}, |
|
{ |
|
"epoch": 6.95, |
|
"learning_rate": 6.602577873254564e-05, |
|
"loss": 0.2403, |
|
"step": 80400 |
|
}, |
|
{ |
|
"epoch": 6.95, |
|
"eval_loss": 0.09069675952196121, |
|
"eval_runtime": 670.4564, |
|
"eval_samples_per_second": 17.056, |
|
"eval_steps_per_second": 2.133, |
|
"eval_wer": 0.1124020961289232, |
|
"step": 80400 |
|
}, |
|
{ |
|
"epoch": 6.97, |
|
"learning_rate": 6.543989844741723e-05, |
|
"loss": 0.2574, |
|
"step": 80600 |
|
}, |
|
{ |
|
"epoch": 6.97, |
|
"eval_loss": 0.09040801972150803, |
|
"eval_runtime": 680.0348, |
|
"eval_samples_per_second": 16.815, |
|
"eval_steps_per_second": 2.103, |
|
"eval_wer": 0.11377697639037584, |
|
"step": 80600 |
|
}, |
|
{ |
|
"epoch": 6.99, |
|
"learning_rate": 6.485694756371448e-05, |
|
"loss": 0.2617, |
|
"step": 80800 |
|
}, |
|
{ |
|
"epoch": 6.99, |
|
"eval_loss": 0.08986030519008636, |
|
"eval_runtime": 664.2336, |
|
"eval_samples_per_second": 17.215, |
|
"eval_steps_per_second": 2.153, |
|
"eval_wer": 0.11516312616216826, |
|
"step": 80800 |
|
}, |
|
{ |
|
"epoch": 7.01, |
|
"learning_rate": 6.427106727858607e-05, |
|
"loss": 0.2235, |
|
"step": 81000 |
|
}, |
|
{ |
|
"epoch": 7.01, |
|
"eval_loss": 0.0911833718419075, |
|
"eval_runtime": 673.0343, |
|
"eval_samples_per_second": 16.99, |
|
"eval_steps_per_second": 2.125, |
|
"eval_wer": 0.11346143010086211, |
|
"step": 81000 |
|
}, |
|
{ |
|
"epoch": 7.03, |
|
"learning_rate": 6.368518699345766e-05, |
|
"loss": 0.2211, |
|
"step": 81200 |
|
}, |
|
{ |
|
"epoch": 7.03, |
|
"eval_loss": 0.09049921482801437, |
|
"eval_runtime": 664.6179, |
|
"eval_samples_per_second": 17.205, |
|
"eval_steps_per_second": 2.152, |
|
"eval_wer": 0.11221051445314702, |
|
"step": 81200 |
|
}, |
|
{ |
|
"epoch": 7.05, |
|
"learning_rate": 6.309930670832925e-05, |
|
"loss": 0.2141, |
|
"step": 81400 |
|
}, |
|
{ |
|
"epoch": 7.05, |
|
"eval_loss": 0.09099965542554855, |
|
"eval_runtime": 653.4138, |
|
"eval_samples_per_second": 17.5, |
|
"eval_steps_per_second": 2.189, |
|
"eval_wer": 0.11219924494280724, |
|
"step": 81400 |
|
}, |
|
{ |
|
"epoch": 7.07, |
|
"learning_rate": 6.251342642320086e-05, |
|
"loss": 0.2077, |
|
"step": 81600 |
|
}, |
|
{ |
|
"epoch": 7.07, |
|
"eval_loss": 0.0929802879691124, |
|
"eval_runtime": 654.1785, |
|
"eval_samples_per_second": 17.48, |
|
"eval_steps_per_second": 2.186, |
|
"eval_wer": 0.11382205443173495, |
|
"step": 81600 |
|
}, |
|
{ |
|
"epoch": 7.09, |
|
"learning_rate": 6.192754613807245e-05, |
|
"loss": 0.2094, |
|
"step": 81800 |
|
}, |
|
{ |
|
"epoch": 7.09, |
|
"eval_loss": 0.09322798252105713, |
|
"eval_runtime": 659.1034, |
|
"eval_samples_per_second": 17.349, |
|
"eval_steps_per_second": 2.17, |
|
"eval_wer": 0.1118386206119344, |
|
"step": 81800 |
|
}, |
|
{ |
|
"epoch": 7.11, |
|
"learning_rate": 6.134166585294405e-05, |
|
"loss": 0.221, |
|
"step": 82000 |
|
}, |
|
{ |
|
"epoch": 7.11, |
|
"eval_loss": 0.09150510281324387, |
|
"eval_runtime": 660.2794, |
|
"eval_samples_per_second": 17.318, |
|
"eval_steps_per_second": 2.166, |
|
"eval_wer": 0.110452470840142, |
|
"step": 82000 |
|
}, |
|
{ |
|
"epoch": 7.13, |
|
"learning_rate": 6.0755785567815635e-05, |
|
"loss": 0.2132, |
|
"step": 82200 |
|
}, |
|
{ |
|
"epoch": 7.13, |
|
"eval_loss": 0.092040054500103, |
|
"eval_runtime": 657.507, |
|
"eval_samples_per_second": 17.391, |
|
"eval_steps_per_second": 2.175, |
|
"eval_wer": 0.11068913055727729, |
|
"step": 82200 |
|
}, |
|
{ |
|
"epoch": 7.15, |
|
"learning_rate": 6.016990528268724e-05, |
|
"loss": 0.2157, |
|
"step": 82400 |
|
}, |
|
{ |
|
"epoch": 7.15, |
|
"eval_loss": 0.09085466712713242, |
|
"eval_runtime": 665.7916, |
|
"eval_samples_per_second": 17.175, |
|
"eval_steps_per_second": 2.148, |
|
"eval_wer": 0.1125373302530005, |
|
"step": 82400 |
|
}, |
|
{ |
|
"epoch": 7.17, |
|
"learning_rate": 5.958402499755883e-05, |
|
"loss": 0.2246, |
|
"step": 82600 |
|
}, |
|
{ |
|
"epoch": 7.17, |
|
"eval_loss": 0.09079894423484802, |
|
"eval_runtime": 660.1771, |
|
"eval_samples_per_second": 17.321, |
|
"eval_steps_per_second": 2.166, |
|
"eval_wer": 0.11159069138445935, |
|
"step": 82600 |
|
}, |
|
{ |
|
"epoch": 7.19, |
|
"learning_rate": 5.899814471243042e-05, |
|
"loss": 0.2181, |
|
"step": 82800 |
|
}, |
|
{ |
|
"epoch": 7.19, |
|
"eval_loss": 0.09172764420509338, |
|
"eval_runtime": 661.7619, |
|
"eval_samples_per_second": 17.28, |
|
"eval_steps_per_second": 2.161, |
|
"eval_wer": 0.11192877669465262, |
|
"step": 82800 |
|
}, |
|
{ |
|
"epoch": 7.2, |
|
"learning_rate": 5.841226442730202e-05, |
|
"loss": 0.2203, |
|
"step": 83000 |
|
}, |
|
{ |
|
"epoch": 7.2, |
|
"eval_loss": 0.09090258926153183, |
|
"eval_runtime": 662.2479, |
|
"eval_samples_per_second": 17.267, |
|
"eval_steps_per_second": 2.159, |
|
"eval_wer": 0.11027215867470558, |
|
"step": 83000 |
|
}, |
|
{ |
|
"epoch": 7.22, |
|
"learning_rate": 5.782638414217361e-05, |
|
"loss": 0.2191, |
|
"step": 83200 |
|
}, |
|
{ |
|
"epoch": 7.22, |
|
"eval_loss": 0.09051696956157684, |
|
"eval_runtime": 662.1226, |
|
"eval_samples_per_second": 17.27, |
|
"eval_steps_per_second": 2.16, |
|
"eval_wer": 0.109798839240435, |
|
"step": 83200 |
|
}, |
|
{ |
|
"epoch": 7.24, |
|
"learning_rate": 5.7240503857045205e-05, |
|
"loss": 0.2252, |
|
"step": 83400 |
|
}, |
|
{ |
|
"epoch": 7.24, |
|
"eval_loss": 0.08852972090244293, |
|
"eval_runtime": 663.1512, |
|
"eval_samples_per_second": 17.243, |
|
"eval_steps_per_second": 2.156, |
|
"eval_wer": 0.1117484645292162, |
|
"step": 83400 |
|
}, |
|
{ |
|
"epoch": 7.26, |
|
"learning_rate": 5.6654623571916796e-05, |
|
"loss": 0.2222, |
|
"step": 83600 |
|
}, |
|
{ |
|
"epoch": 7.26, |
|
"eval_loss": 0.0903807133436203, |
|
"eval_runtime": 667.5255, |
|
"eval_samples_per_second": 17.13, |
|
"eval_steps_per_second": 2.142, |
|
"eval_wer": 0.11038485377810334, |
|
"step": 83600 |
|
}, |
|
{ |
|
"epoch": 7.28, |
|
"learning_rate": 5.606874328678839e-05, |
|
"loss": 0.2127, |
|
"step": 83800 |
|
}, |
|
{ |
|
"epoch": 7.28, |
|
"eval_loss": 0.09016119688749313, |
|
"eval_runtime": 657.2492, |
|
"eval_samples_per_second": 17.398, |
|
"eval_steps_per_second": 2.176, |
|
"eval_wer": 0.11103848537781033, |
|
"step": 83800 |
|
}, |
|
{ |
|
"epoch": 7.3, |
|
"learning_rate": 5.5482863001659984e-05, |
|
"loss": 0.2145, |
|
"step": 84000 |
|
}, |
|
{ |
|
"epoch": 7.3, |
|
"eval_loss": 0.09145020693540573, |
|
"eval_runtime": 658.2658, |
|
"eval_samples_per_second": 17.371, |
|
"eval_steps_per_second": 2.172, |
|
"eval_wer": 0.11002422944723052, |
|
"step": 84000 |
|
}, |
|
{ |
|
"epoch": 7.32, |
|
"learning_rate": 5.489698271653159e-05, |
|
"loss": 0.2256, |
|
"step": 84200 |
|
}, |
|
{ |
|
"epoch": 7.32, |
|
"eval_loss": 0.09001456201076508, |
|
"eval_runtime": 660.2198, |
|
"eval_samples_per_second": 17.32, |
|
"eval_steps_per_second": 2.166, |
|
"eval_wer": 0.11011438552994872, |
|
"step": 84200 |
|
}, |
|
{ |
|
"epoch": 7.34, |
|
"learning_rate": 5.431110243140318e-05, |
|
"loss": 0.2334, |
|
"step": 84400 |
|
}, |
|
{ |
|
"epoch": 7.34, |
|
"eval_loss": 0.09009411185979843, |
|
"eval_runtime": 657.4443, |
|
"eval_samples_per_second": 17.393, |
|
"eval_steps_per_second": 2.175, |
|
"eval_wer": 0.11022708063334648, |
|
"step": 84400 |
|
}, |
|
{ |
|
"epoch": 7.36, |
|
"learning_rate": 5.3725222146274776e-05, |
|
"loss": 0.2171, |
|
"step": 84600 |
|
}, |
|
{ |
|
"epoch": 7.36, |
|
"eval_loss": 0.08927768468856812, |
|
"eval_runtime": 658.5101, |
|
"eval_samples_per_second": 17.365, |
|
"eval_steps_per_second": 2.172, |
|
"eval_wer": 0.11080182566067505, |
|
"step": 84600 |
|
}, |
|
{ |
|
"epoch": 7.38, |
|
"learning_rate": 5.3139341861146366e-05, |
|
"loss": 0.2247, |
|
"step": 84800 |
|
}, |
|
{ |
|
"epoch": 7.38, |
|
"eval_loss": 0.08871866762638092, |
|
"eval_runtime": 657.6113, |
|
"eval_samples_per_second": 17.389, |
|
"eval_steps_per_second": 2.175, |
|
"eval_wer": 0.10952837099228038, |
|
"step": 84800 |
|
}, |
|
{ |
|
"epoch": 7.4, |
|
"learning_rate": 5.2553461576017963e-05, |
|
"loss": 0.225, |
|
"step": 85000 |
|
}, |
|
{ |
|
"epoch": 7.4, |
|
"eval_loss": 0.08847151696681976, |
|
"eval_runtime": 657.4131, |
|
"eval_samples_per_second": 17.394, |
|
"eval_steps_per_second": 2.175, |
|
"eval_wer": 0.10948329295092128, |
|
"step": 85000 |
|
}, |
|
{ |
|
"epoch": 7.42, |
|
"learning_rate": 5.1967581290889554e-05, |
|
"loss": 0.2263, |
|
"step": 85200 |
|
}, |
|
{ |
|
"epoch": 7.42, |
|
"eval_loss": 0.08873692154884338, |
|
"eval_runtime": 658.6562, |
|
"eval_samples_per_second": 17.361, |
|
"eval_steps_per_second": 2.171, |
|
"eval_wer": 0.10950583197160084, |
|
"step": 85200 |
|
}, |
|
{ |
|
"epoch": 7.44, |
|
"learning_rate": 5.1384630407186794e-05, |
|
"loss": 0.205, |
|
"step": 85400 |
|
}, |
|
{ |
|
"epoch": 7.44, |
|
"eval_loss": 0.08996891975402832, |
|
"eval_runtime": 662.3112, |
|
"eval_samples_per_second": 17.265, |
|
"eval_steps_per_second": 2.159, |
|
"eval_wer": 0.10919028568208711, |
|
"step": 85400 |
|
}, |
|
{ |
|
"epoch": 7.46, |
|
"learning_rate": 5.0798750122058385e-05, |
|
"loss": 0.2153, |
|
"step": 85600 |
|
}, |
|
{ |
|
"epoch": 7.46, |
|
"eval_loss": 0.09001825749874115, |
|
"eval_runtime": 663.8308, |
|
"eval_samples_per_second": 17.226, |
|
"eval_steps_per_second": 2.154, |
|
"eval_wer": 0.10957344903363948, |
|
"step": 85600 |
|
}, |
|
{ |
|
"epoch": 7.48, |
|
"learning_rate": 5.021286983692998e-05, |
|
"loss": 0.2175, |
|
"step": 85800 |
|
}, |
|
{ |
|
"epoch": 7.48, |
|
"eval_loss": 0.08824647217988968, |
|
"eval_runtime": 665.8794, |
|
"eval_samples_per_second": 17.173, |
|
"eval_steps_per_second": 2.148, |
|
"eval_wer": 0.10828872485490505, |
|
"step": 85800 |
|
}, |
|
{ |
|
"epoch": 7.5, |
|
"learning_rate": 4.962698955180157e-05, |
|
"loss": 0.2169, |
|
"step": 86000 |
|
}, |
|
{ |
|
"epoch": 7.5, |
|
"eval_loss": 0.08869241178035736, |
|
"eval_runtime": 664.8058, |
|
"eval_samples_per_second": 17.201, |
|
"eval_steps_per_second": 2.151, |
|
"eval_wer": 0.10858173212373923, |
|
"step": 86000 |
|
}, |
|
{ |
|
"epoch": 7.52, |
|
"learning_rate": 4.904110926667317e-05, |
|
"loss": 0.2234, |
|
"step": 86200 |
|
}, |
|
{ |
|
"epoch": 7.52, |
|
"eval_loss": 0.08755665272474289, |
|
"eval_runtime": 662.3879, |
|
"eval_samples_per_second": 17.263, |
|
"eval_steps_per_second": 2.159, |
|
"eval_wer": 0.10818729926184707, |
|
"step": 86200 |
|
}, |
|
{ |
|
"epoch": 7.54, |
|
"learning_rate": 4.845522898154476e-05, |
|
"loss": 0.2112, |
|
"step": 86400 |
|
}, |
|
{ |
|
"epoch": 7.54, |
|
"eval_loss": 0.08796130865812302, |
|
"eval_runtime": 660.9811, |
|
"eval_samples_per_second": 17.3, |
|
"eval_steps_per_second": 2.163, |
|
"eval_wer": 0.10853665408238013, |
|
"step": 86400 |
|
}, |
|
{ |
|
"epoch": 7.55, |
|
"learning_rate": 4.7869348696416365e-05, |
|
"loss": 0.2122, |
|
"step": 86600 |
|
}, |
|
{ |
|
"epoch": 7.55, |
|
"eval_loss": 0.08818641304969788, |
|
"eval_runtime": 659.8525, |
|
"eval_samples_per_second": 17.33, |
|
"eval_steps_per_second": 2.167, |
|
"eval_wer": 0.10849157604102101, |
|
"step": 86600 |
|
}, |
|
{ |
|
"epoch": 7.57, |
|
"learning_rate": 4.728639781271359e-05, |
|
"loss": 0.2134, |
|
"step": 86800 |
|
}, |
|
{ |
|
"epoch": 7.57, |
|
"eval_loss": 0.08850608021020889, |
|
"eval_runtime": 663.3892, |
|
"eval_samples_per_second": 17.237, |
|
"eval_steps_per_second": 2.156, |
|
"eval_wer": 0.10823237730320617, |
|
"step": 86800 |
|
}, |
|
{ |
|
"epoch": 7.59, |
|
"learning_rate": 4.6700517527585196e-05, |
|
"loss": 0.2218, |
|
"step": 87000 |
|
}, |
|
{ |
|
"epoch": 7.59, |
|
"eval_loss": 0.08732796460390091, |
|
"eval_runtime": 663.6661, |
|
"eval_samples_per_second": 17.23, |
|
"eval_steps_per_second": 2.155, |
|
"eval_wer": 0.10806333464810954, |
|
"step": 87000 |
|
}, |
|
{ |
|
"epoch": 7.61, |
|
"learning_rate": 4.6114637242456786e-05, |
|
"loss": 0.2098, |
|
"step": 87200 |
|
}, |
|
{ |
|
"epoch": 7.61, |
|
"eval_loss": 0.08799753338098526, |
|
"eval_runtime": 660.746, |
|
"eval_samples_per_second": 17.306, |
|
"eval_steps_per_second": 2.164, |
|
"eval_wer": 0.1077590578689356, |
|
"step": 87200 |
|
}, |
|
{ |
|
"epoch": 7.63, |
|
"learning_rate": 4.5528756957328383e-05, |
|
"loss": 0.2185, |
|
"step": 87400 |
|
}, |
|
{ |
|
"epoch": 7.63, |
|
"eval_loss": 0.08814089745283127, |
|
"eval_runtime": 659.5596, |
|
"eval_samples_per_second": 17.337, |
|
"eval_steps_per_second": 2.168, |
|
"eval_wer": 0.10757874570349918, |
|
"step": 87400 |
|
}, |
|
{ |
|
"epoch": 7.65, |
|
"learning_rate": 4.4942876672199974e-05, |
|
"loss": 0.2036, |
|
"step": 87600 |
|
}, |
|
{ |
|
"epoch": 7.65, |
|
"eval_loss": 0.08803264051675797, |
|
"eval_runtime": 659.0171, |
|
"eval_samples_per_second": 17.352, |
|
"eval_steps_per_second": 2.17, |
|
"eval_wer": 0.10823237730320617, |
|
"step": 87600 |
|
}, |
|
{ |
|
"epoch": 7.67, |
|
"learning_rate": 4.435699638707157e-05, |
|
"loss": 0.2064, |
|
"step": 87800 |
|
}, |
|
{ |
|
"epoch": 7.67, |
|
"eval_loss": 0.08712273836135864, |
|
"eval_runtime": 663.9763, |
|
"eval_samples_per_second": 17.222, |
|
"eval_steps_per_second": 2.154, |
|
"eval_wer": 0.10770271031723672, |
|
"step": 87800 |
|
}, |
|
{ |
|
"epoch": 7.69, |
|
"learning_rate": 4.377111610194316e-05, |
|
"loss": 0.2141, |
|
"step": 88000 |
|
}, |
|
{ |
|
"epoch": 7.69, |
|
"eval_loss": 0.08767861127853394, |
|
"eval_runtime": 653.4457, |
|
"eval_samples_per_second": 17.5, |
|
"eval_steps_per_second": 2.188, |
|
"eval_wer": 0.10619259593170677, |
|
"step": 88000 |
|
}, |
|
{ |
|
"epoch": 7.71, |
|
"learning_rate": 4.318523581681476e-05, |
|
"loss": 0.2168, |
|
"step": 88200 |
|
}, |
|
{ |
|
"epoch": 7.71, |
|
"eval_loss": 0.0869930163025856, |
|
"eval_runtime": 653.5133, |
|
"eval_samples_per_second": 17.498, |
|
"eval_steps_per_second": 2.188, |
|
"eval_wer": 0.10565165943539753, |
|
"step": 88200 |
|
}, |
|
{ |
|
"epoch": 7.73, |
|
"learning_rate": 4.259935553168635e-05, |
|
"loss": 0.2157, |
|
"step": 88400 |
|
}, |
|
{ |
|
"epoch": 7.73, |
|
"eval_loss": 0.0869561955332756, |
|
"eval_runtime": 657.2321, |
|
"eval_samples_per_second": 17.399, |
|
"eval_steps_per_second": 2.176, |
|
"eval_wer": 0.10656448977291937, |
|
"step": 88400 |
|
}, |
|
{ |
|
"epoch": 7.75, |
|
"learning_rate": 4.2013475246557954e-05, |
|
"loss": 0.2151, |
|
"step": 88600 |
|
}, |
|
{ |
|
"epoch": 7.75, |
|
"eval_loss": 0.08736027032136917, |
|
"eval_runtime": 655.7458, |
|
"eval_samples_per_second": 17.438, |
|
"eval_steps_per_second": 2.181, |
|
"eval_wer": 0.10630529103510453, |
|
"step": 88600 |
|
}, |
|
{ |
|
"epoch": 7.77, |
|
"learning_rate": 4.1427594961429544e-05, |
|
"loss": 0.2115, |
|
"step": 88800 |
|
}, |
|
{ |
|
"epoch": 7.77, |
|
"eval_loss": 0.0879904106259346, |
|
"eval_runtime": 656.2337, |
|
"eval_samples_per_second": 17.425, |
|
"eval_steps_per_second": 2.179, |
|
"eval_wer": 0.10746605060010142, |
|
"step": 88800 |
|
}, |
|
{ |
|
"epoch": 7.79, |
|
"learning_rate": 4.084171467630114e-05, |
|
"loss": 0.2125, |
|
"step": 89000 |
|
}, |
|
{ |
|
"epoch": 7.79, |
|
"eval_loss": 0.08856356143951416, |
|
"eval_runtime": 658.7488, |
|
"eval_samples_per_second": 17.359, |
|
"eval_steps_per_second": 2.171, |
|
"eval_wer": 0.10648560320054094, |
|
"step": 89000 |
|
}, |
|
{ |
|
"epoch": 7.81, |
|
"learning_rate": 4.0258763792598375e-05, |
|
"loss": 0.2036, |
|
"step": 89200 |
|
}, |
|
{ |
|
"epoch": 7.81, |
|
"eval_loss": 0.08776041120290756, |
|
"eval_runtime": 656.4464, |
|
"eval_samples_per_second": 17.42, |
|
"eval_steps_per_second": 2.178, |
|
"eval_wer": 0.10588831915253283, |
|
"step": 89200 |
|
}, |
|
{ |
|
"epoch": 7.83, |
|
"learning_rate": 3.967288350746997e-05, |
|
"loss": 0.2199, |
|
"step": 89400 |
|
}, |
|
{ |
|
"epoch": 7.83, |
|
"eval_loss": 0.08602162450551987, |
|
"eval_runtime": 658.2289, |
|
"eval_samples_per_second": 17.372, |
|
"eval_steps_per_second": 2.172, |
|
"eval_wer": 0.10615878740068745, |
|
"step": 89400 |
|
}, |
|
{ |
|
"epoch": 8.02, |
|
"learning_rate": 3.908700322234156e-05, |
|
"loss": 0.2055, |
|
"step": 89600 |
|
}, |
|
{ |
|
"epoch": 8.02, |
|
"eval_loss": 0.08735963702201843, |
|
"eval_runtime": 649.7905, |
|
"eval_samples_per_second": 17.598, |
|
"eval_steps_per_second": 2.201, |
|
"eval_wer": 0.10638417760748295, |
|
"step": 89600 |
|
}, |
|
{ |
|
"epoch": 8.04, |
|
"learning_rate": 3.850112293721316e-05, |
|
"loss": 0.2011, |
|
"step": 89800 |
|
}, |
|
{ |
|
"epoch": 8.04, |
|
"eval_loss": 0.08764158189296722, |
|
"eval_runtime": 640.7104, |
|
"eval_samples_per_second": 17.847, |
|
"eval_steps_per_second": 2.232, |
|
"eval_wer": 0.10567419845607709, |
|
"step": 89800 |
|
}, |
|
{ |
|
"epoch": 8.06, |
|
"learning_rate": 3.791524265208475e-05, |
|
"loss": 0.2018, |
|
"step": 90000 |
|
}, |
|
{ |
|
"epoch": 8.06, |
|
"eval_loss": 0.08688807487487793, |
|
"eval_runtime": 642.3443, |
|
"eval_samples_per_second": 17.802, |
|
"eval_steps_per_second": 2.226, |
|
"eval_wer": 0.10522341804248606, |
|
"step": 90000 |
|
}, |
|
{ |
|
"epoch": 8.08, |
|
"learning_rate": 3.732936236695635e-05, |
|
"loss": 0.2052, |
|
"step": 90200 |
|
}, |
|
{ |
|
"epoch": 8.08, |
|
"eval_loss": 0.08650859445333481, |
|
"eval_runtime": 640.5273, |
|
"eval_samples_per_second": 17.852, |
|
"eval_steps_per_second": 2.233, |
|
"eval_wer": 0.10556150335267933, |
|
"step": 90200 |
|
}, |
|
{ |
|
"epoch": 8.1, |
|
"learning_rate": 3.6743482081827945e-05, |
|
"loss": 0.2128, |
|
"step": 90400 |
|
}, |
|
{ |
|
"epoch": 8.1, |
|
"eval_loss": 0.08721867948770523, |
|
"eval_runtime": 644.1835, |
|
"eval_samples_per_second": 17.751, |
|
"eval_steps_per_second": 2.22, |
|
"eval_wer": 0.10627148250408519, |
|
"step": 90400 |
|
}, |
|
{ |
|
"epoch": 8.12, |
|
"learning_rate": 3.616053119812518e-05, |
|
"loss": 0.2054, |
|
"step": 90600 |
|
}, |
|
{ |
|
"epoch": 8.12, |
|
"eval_loss": 0.08693686872720718, |
|
"eval_runtime": 640.4586, |
|
"eval_samples_per_second": 17.854, |
|
"eval_steps_per_second": 2.233, |
|
"eval_wer": 0.10453597791175974, |
|
"step": 90600 |
|
}, |
|
{ |
|
"epoch": 8.14, |
|
"learning_rate": 3.5574650912996776e-05, |
|
"loss": 0.2064, |
|
"step": 90800 |
|
}, |
|
{ |
|
"epoch": 8.14, |
|
"eval_loss": 0.08682414889335632, |
|
"eval_runtime": 645.5348, |
|
"eval_samples_per_second": 17.714, |
|
"eval_steps_per_second": 2.215, |
|
"eval_wer": 0.10422043162224601, |
|
"step": 90800 |
|
}, |
|
{ |
|
"epoch": 8.16, |
|
"learning_rate": 3.4988770627868374e-05, |
|
"loss": 0.2003, |
|
"step": 91000 |
|
}, |
|
{ |
|
"epoch": 8.16, |
|
"eval_loss": 0.08698801696300507, |
|
"eval_runtime": 642.0155, |
|
"eval_samples_per_second": 17.811, |
|
"eval_steps_per_second": 2.227, |
|
"eval_wer": 0.10504310587704964, |
|
"step": 91000 |
|
}, |
|
{ |
|
"epoch": 8.17, |
|
"learning_rate": 3.4402890342739964e-05, |
|
"loss": 0.2051, |
|
"step": 91200 |
|
}, |
|
{ |
|
"epoch": 8.17, |
|
"eval_loss": 0.08681213855743408, |
|
"eval_runtime": 639.0151, |
|
"eval_samples_per_second": 17.895, |
|
"eval_steps_per_second": 2.238, |
|
"eval_wer": 0.10517834000112695, |
|
"step": 91200 |
|
}, |
|
{ |
|
"epoch": 8.19, |
|
"learning_rate": 3.381701005761156e-05, |
|
"loss": 0.205, |
|
"step": 91400 |
|
}, |
|
{ |
|
"epoch": 8.19, |
|
"eval_loss": 0.08681774884462357, |
|
"eval_runtime": 647.2922, |
|
"eval_samples_per_second": 17.666, |
|
"eval_steps_per_second": 2.209, |
|
"eval_wer": 0.10460359497379838, |
|
"step": 91400 |
|
}, |
|
{ |
|
"epoch": 8.21, |
|
"learning_rate": 3.323112977248315e-05, |
|
"loss": 0.2045, |
|
"step": 91600 |
|
}, |
|
{ |
|
"epoch": 8.21, |
|
"eval_loss": 0.08683190494775772, |
|
"eval_runtime": 640.6094, |
|
"eval_samples_per_second": 17.85, |
|
"eval_steps_per_second": 2.232, |
|
"eval_wer": 0.10384853778103341, |
|
"step": 91600 |
|
}, |
|
{ |
|
"epoch": 8.23, |
|
"learning_rate": 3.264524948735475e-05, |
|
"loss": 0.2073, |
|
"step": 91800 |
|
}, |
|
{ |
|
"epoch": 8.23, |
|
"eval_loss": 0.08680137246847153, |
|
"eval_runtime": 644.4426, |
|
"eval_samples_per_second": 17.744, |
|
"eval_steps_per_second": 2.219, |
|
"eval_wer": 0.10420916211190624, |
|
"step": 91800 |
|
}, |
|
{ |
|
"epoch": 8.25, |
|
"learning_rate": 3.205936920222635e-05, |
|
"loss": 0.1989, |
|
"step": 92000 |
|
}, |
|
{ |
|
"epoch": 8.25, |
|
"eval_loss": 0.08712614327669144, |
|
"eval_runtime": 642.565, |
|
"eval_samples_per_second": 17.796, |
|
"eval_steps_per_second": 2.225, |
|
"eval_wer": 0.10425424015326534, |
|
"step": 92000 |
|
}, |
|
{ |
|
"epoch": 8.27, |
|
"learning_rate": 3.147348891709794e-05, |
|
"loss": 0.1929, |
|
"step": 92200 |
|
}, |
|
{ |
|
"epoch": 8.27, |
|
"eval_loss": 0.08767101913690567, |
|
"eval_runtime": 645.2302, |
|
"eval_samples_per_second": 17.722, |
|
"eval_steps_per_second": 2.216, |
|
"eval_wer": 0.10409646700850848, |
|
"step": 92200 |
|
}, |
|
{ |
|
"epoch": 8.29, |
|
"learning_rate": 3.0887608631969534e-05, |
|
"loss": 0.2116, |
|
"step": 92400 |
|
}, |
|
{ |
|
"epoch": 8.29, |
|
"eval_loss": 0.08523967117071152, |
|
"eval_runtime": 645.8547, |
|
"eval_samples_per_second": 17.705, |
|
"eval_steps_per_second": 2.214, |
|
"eval_wer": 0.10381472925001409, |
|
"step": 92400 |
|
}, |
|
{ |
|
"epoch": 8.31, |
|
"learning_rate": 3.030172834684113e-05, |
|
"loss": 0.1991, |
|
"step": 92600 |
|
}, |
|
{ |
|
"epoch": 8.31, |
|
"eval_loss": 0.08694258332252502, |
|
"eval_runtime": 638.7288, |
|
"eval_samples_per_second": 17.903, |
|
"eval_steps_per_second": 2.239, |
|
"eval_wer": 0.10360060855355835, |
|
"step": 92600 |
|
}, |
|
{ |
|
"epoch": 8.33, |
|
"learning_rate": 2.971584806171272e-05, |
|
"loss": 0.2045, |
|
"step": 92800 |
|
}, |
|
{ |
|
"epoch": 8.33, |
|
"eval_loss": 0.08679590374231339, |
|
"eval_runtime": 641.8956, |
|
"eval_samples_per_second": 17.814, |
|
"eval_steps_per_second": 2.228, |
|
"eval_wer": 0.10344283540880149, |
|
"step": 92800 |
|
}, |
|
{ |
|
"epoch": 8.35, |
|
"learning_rate": 2.9129967776584313e-05, |
|
"loss": 0.2081, |
|
"step": 93000 |
|
}, |
|
{ |
|
"epoch": 8.35, |
|
"eval_loss": 0.0871407687664032, |
|
"eval_runtime": 641.4271, |
|
"eval_samples_per_second": 17.827, |
|
"eval_steps_per_second": 2.229, |
|
"eval_wer": 0.10428804868428466, |
|
"step": 93000 |
|
}, |
|
{ |
|
"epoch": 8.37, |
|
"learning_rate": 2.8544087491455907e-05, |
|
"loss": 0.1915, |
|
"step": 93200 |
|
}, |
|
{ |
|
"epoch": 8.37, |
|
"eval_loss": 0.0871988832950592, |
|
"eval_runtime": 645.7411, |
|
"eval_samples_per_second": 17.708, |
|
"eval_steps_per_second": 2.215, |
|
"eval_wer": 0.10367949512593677, |
|
"step": 93200 |
|
}, |
|
{ |
|
"epoch": 8.39, |
|
"learning_rate": 2.7958207206327504e-05, |
|
"loss": 0.1893, |
|
"step": 93400 |
|
}, |
|
{ |
|
"epoch": 8.39, |
|
"eval_loss": 0.08661910146474838, |
|
"eval_runtime": 641.6227, |
|
"eval_samples_per_second": 17.822, |
|
"eval_steps_per_second": 2.229, |
|
"eval_wer": 0.1032512537330253, |
|
"step": 93400 |
|
}, |
|
{ |
|
"epoch": 8.41, |
|
"learning_rate": 2.7372326921199098e-05, |
|
"loss": 0.1927, |
|
"step": 93600 |
|
}, |
|
{ |
|
"epoch": 8.41, |
|
"eval_loss": 0.08692566305398941, |
|
"eval_runtime": 645.917, |
|
"eval_samples_per_second": 17.704, |
|
"eval_steps_per_second": 2.214, |
|
"eval_wer": 0.1028342818504536, |
|
"step": 93600 |
|
}, |
|
{ |
|
"epoch": 8.43, |
|
"learning_rate": 2.6786446636070692e-05, |
|
"loss": 0.1987, |
|
"step": 93800 |
|
}, |
|
{ |
|
"epoch": 8.43, |
|
"eval_loss": 0.08724019676446915, |
|
"eval_runtime": 641.5, |
|
"eval_samples_per_second": 17.825, |
|
"eval_steps_per_second": 2.229, |
|
"eval_wer": 0.10263143066433764, |
|
"step": 93800 |
|
}, |
|
{ |
|
"epoch": 8.45, |
|
"learning_rate": 2.6200566350942286e-05, |
|
"loss": 0.2088, |
|
"step": 94000 |
|
}, |
|
{ |
|
"epoch": 8.45, |
|
"eval_loss": 0.08689233660697937, |
|
"eval_runtime": 638.7681, |
|
"eval_samples_per_second": 17.902, |
|
"eval_steps_per_second": 2.239, |
|
"eval_wer": 0.10304840254690933, |
|
"step": 94000 |
|
}, |
|
{ |
|
"epoch": 8.47, |
|
"learning_rate": 2.5614686065813883e-05, |
|
"loss": 0.1929, |
|
"step": 94200 |
|
}, |
|
{ |
|
"epoch": 8.47, |
|
"eval_loss": 0.08683472126722336, |
|
"eval_runtime": 643.1191, |
|
"eval_samples_per_second": 17.781, |
|
"eval_steps_per_second": 2.224, |
|
"eval_wer": 0.10273285625739562, |
|
"step": 94200 |
|
}, |
|
{ |
|
"epoch": 8.49, |
|
"learning_rate": 2.5028805780685477e-05, |
|
"loss": 0.1931, |
|
"step": 94400 |
|
}, |
|
{ |
|
"epoch": 8.49, |
|
"eval_loss": 0.08680200576782227, |
|
"eval_runtime": 643.6223, |
|
"eval_samples_per_second": 17.767, |
|
"eval_steps_per_second": 2.222, |
|
"eval_wer": 0.1028342818504536, |
|
"step": 94400 |
|
}, |
|
{ |
|
"epoch": 8.51, |
|
"learning_rate": 2.4445854896982714e-05, |
|
"loss": 0.2008, |
|
"step": 94600 |
|
}, |
|
{ |
|
"epoch": 8.51, |
|
"eval_loss": 0.08561909943819046, |
|
"eval_runtime": 640.5789, |
|
"eval_samples_per_second": 17.851, |
|
"eval_steps_per_second": 2.232, |
|
"eval_wer": 0.10308221107792867, |
|
"step": 94600 |
|
}, |
|
{ |
|
"epoch": 8.52, |
|
"learning_rate": 2.3859974611854308e-05, |
|
"loss": 0.1971, |
|
"step": 94800 |
|
}, |
|
{ |
|
"epoch": 8.52, |
|
"eval_loss": 0.0856679379940033, |
|
"eval_runtime": 643.6696, |
|
"eval_samples_per_second": 17.765, |
|
"eval_steps_per_second": 2.222, |
|
"eval_wer": 0.10286809038147293, |
|
"step": 94800 |
|
}, |
|
{ |
|
"epoch": 8.54, |
|
"learning_rate": 2.3274094326725902e-05, |
|
"loss": 0.1915, |
|
"step": 95000 |
|
}, |
|
{ |
|
"epoch": 8.54, |
|
"eval_loss": 0.08586091548204422, |
|
"eval_runtime": 644.3313, |
|
"eval_samples_per_second": 17.747, |
|
"eval_steps_per_second": 2.219, |
|
"eval_wer": 0.10251873556093988, |
|
"step": 95000 |
|
}, |
|
{ |
|
"epoch": 8.56, |
|
"learning_rate": 2.2691143443023142e-05, |
|
"loss": 0.2011, |
|
"step": 95200 |
|
}, |
|
{ |
|
"epoch": 8.56, |
|
"eval_loss": 0.08547823131084442, |
|
"eval_runtime": 642.5453, |
|
"eval_samples_per_second": 17.796, |
|
"eval_steps_per_second": 2.226, |
|
"eval_wer": 0.10172986983715558, |
|
"step": 95200 |
|
}, |
|
{ |
|
"epoch": 8.58, |
|
"learning_rate": 2.2105263157894733e-05, |
|
"loss": 0.1931, |
|
"step": 95400 |
|
}, |
|
{ |
|
"epoch": 8.58, |
|
"eval_loss": 0.0860673114657402, |
|
"eval_runtime": 644.3949, |
|
"eval_samples_per_second": 17.745, |
|
"eval_steps_per_second": 2.219, |
|
"eval_wer": 0.10129035893390433, |
|
"step": 95400 |
|
}, |
|
{ |
|
"epoch": 8.6, |
|
"learning_rate": 2.1519382872766327e-05, |
|
"loss": 0.2008, |
|
"step": 95600 |
|
}, |
|
{ |
|
"epoch": 8.6, |
|
"eval_loss": 0.08597339689731598, |
|
"eval_runtime": 643.5265, |
|
"eval_samples_per_second": 17.769, |
|
"eval_steps_per_second": 2.222, |
|
"eval_wer": 0.10132416746492365, |
|
"step": 95600 |
|
}, |
|
{ |
|
"epoch": 8.62, |
|
"learning_rate": 2.093350258763792e-05, |
|
"loss": 0.2045, |
|
"step": 95800 |
|
}, |
|
{ |
|
"epoch": 8.62, |
|
"eval_loss": 0.08469708263874054, |
|
"eval_runtime": 649.2866, |
|
"eval_samples_per_second": 17.612, |
|
"eval_steps_per_second": 2.202, |
|
"eval_wer": 0.10094100411337127, |
|
"step": 95800 |
|
}, |
|
{ |
|
"epoch": 8.64, |
|
"learning_rate": 2.0347622302509518e-05, |
|
"loss": 0.2027, |
|
"step": 96000 |
|
}, |
|
{ |
|
"epoch": 8.64, |
|
"eval_loss": 0.0852479636669159, |
|
"eval_runtime": 641.0893, |
|
"eval_samples_per_second": 17.837, |
|
"eval_steps_per_second": 2.231, |
|
"eval_wer": 0.10149321012002029, |
|
"step": 96000 |
|
}, |
|
{ |
|
"epoch": 8.66, |
|
"learning_rate": 1.9761742017381112e-05, |
|
"loss": 0.1984, |
|
"step": 96200 |
|
}, |
|
{ |
|
"epoch": 8.66, |
|
"eval_loss": 0.08589636534452438, |
|
"eval_runtime": 636.8856, |
|
"eval_samples_per_second": 17.955, |
|
"eval_steps_per_second": 2.245, |
|
"eval_wer": 0.10201160759564996, |
|
"step": 96200 |
|
}, |
|
{ |
|
"epoch": 8.68, |
|
"learning_rate": 1.9175861732252706e-05, |
|
"loss": 0.1946, |
|
"step": 96400 |
|
}, |
|
{ |
|
"epoch": 8.68, |
|
"eval_loss": 0.08547361940145493, |
|
"eval_runtime": 640.9334, |
|
"eval_samples_per_second": 17.841, |
|
"eval_steps_per_second": 2.231, |
|
"eval_wer": 0.1017636783681749, |
|
"step": 96400 |
|
}, |
|
{ |
|
"epoch": 8.7, |
|
"learning_rate": 1.8589981447124303e-05, |
|
"loss": 0.1991, |
|
"step": 96600 |
|
}, |
|
{ |
|
"epoch": 8.7, |
|
"eval_loss": 0.08481213450431824, |
|
"eval_runtime": 641.2548, |
|
"eval_samples_per_second": 17.832, |
|
"eval_steps_per_second": 2.23, |
|
"eval_wer": 0.101808756409534, |
|
"step": 96600 |
|
}, |
|
{ |
|
"epoch": 8.72, |
|
"learning_rate": 1.8004101161995897e-05, |
|
"loss": 0.1936, |
|
"step": 96800 |
|
}, |
|
{ |
|
"epoch": 8.72, |
|
"eval_loss": 0.08544305711984634, |
|
"eval_runtime": 645.1596, |
|
"eval_samples_per_second": 17.724, |
|
"eval_steps_per_second": 2.217, |
|
"eval_wer": 0.10132416746492365, |
|
"step": 96800 |
|
}, |
|
{ |
|
"epoch": 8.74, |
|
"learning_rate": 1.7418220876867494e-05, |
|
"loss": 0.1965, |
|
"step": 97000 |
|
}, |
|
{ |
|
"epoch": 8.74, |
|
"eval_loss": 0.0854332447052002, |
|
"eval_runtime": 639.5304, |
|
"eval_samples_per_second": 17.88, |
|
"eval_steps_per_second": 2.236, |
|
"eval_wer": 0.10099735166507015, |
|
"step": 97000 |
|
}, |
|
{ |
|
"epoch": 8.76, |
|
"learning_rate": 1.6832340591739088e-05, |
|
"loss": 0.1906, |
|
"step": 97200 |
|
}, |
|
{ |
|
"epoch": 8.76, |
|
"eval_loss": 0.08577127009630203, |
|
"eval_runtime": 639.6566, |
|
"eval_samples_per_second": 17.877, |
|
"eval_steps_per_second": 2.236, |
|
"eval_wer": 0.10094100411337127, |
|
"step": 97200 |
|
}, |
|
{ |
|
"epoch": 8.78, |
|
"learning_rate": 1.6246460306610682e-05, |
|
"loss": 0.1938, |
|
"step": 97400 |
|
}, |
|
{ |
|
"epoch": 8.78, |
|
"eval_loss": 0.08535514771938324, |
|
"eval_runtime": 643.0482, |
|
"eval_samples_per_second": 17.782, |
|
"eval_steps_per_second": 2.224, |
|
"eval_wer": 0.10100862117540993, |
|
"step": 97400 |
|
}, |
|
{ |
|
"epoch": 8.8, |
|
"learning_rate": 1.5660580021482276e-05, |
|
"loss": 0.2027, |
|
"step": 97600 |
|
}, |
|
{ |
|
"epoch": 8.8, |
|
"eval_loss": 0.08470670878887177, |
|
"eval_runtime": 638.835, |
|
"eval_samples_per_second": 17.9, |
|
"eval_steps_per_second": 2.238, |
|
"eval_wer": 0.10085084803065307, |
|
"step": 97600 |
|
}, |
|
{ |
|
"epoch": 8.82, |
|
"learning_rate": 1.507469973635387e-05, |
|
"loss": 0.1932, |
|
"step": 97800 |
|
}, |
|
{ |
|
"epoch": 8.82, |
|
"eval_loss": 0.0843362957239151, |
|
"eval_runtime": 643.5712, |
|
"eval_samples_per_second": 17.768, |
|
"eval_steps_per_second": 2.222, |
|
"eval_wer": 0.10097481264439061, |
|
"step": 97800 |
|
}, |
|
{ |
|
"epoch": 8.84, |
|
"learning_rate": 1.4488819451225464e-05, |
|
"loss": 0.1857, |
|
"step": 98000 |
|
}, |
|
{ |
|
"epoch": 8.84, |
|
"eval_loss": 0.08466717600822449, |
|
"eval_runtime": 644.2589, |
|
"eval_samples_per_second": 17.749, |
|
"eval_steps_per_second": 2.22, |
|
"eval_wer": 0.1001408688792472, |
|
"step": 98000 |
|
}, |
|
{ |
|
"epoch": 8.86, |
|
"learning_rate": 1.390293916609706e-05, |
|
"loss": 0.1871, |
|
"step": 98200 |
|
}, |
|
{ |
|
"epoch": 8.86, |
|
"eval_loss": 0.08466171473264694, |
|
"eval_runtime": 645.4241, |
|
"eval_samples_per_second": 17.717, |
|
"eval_steps_per_second": 2.216, |
|
"eval_wer": 0.10034372006536316, |
|
"step": 98200 |
|
}, |
|
{ |
|
"epoch": 8.87, |
|
"learning_rate": 1.3317058880968653e-05, |
|
"loss": 0.1917, |
|
"step": 98400 |
|
}, |
|
{ |
|
"epoch": 8.87, |
|
"eval_loss": 0.08486426621675491, |
|
"eval_runtime": 640.712, |
|
"eval_samples_per_second": 17.847, |
|
"eval_steps_per_second": 2.232, |
|
"eval_wer": 0.10045641516876092, |
|
"step": 98400 |
|
}, |
|
{ |
|
"epoch": 8.89, |
|
"learning_rate": 1.2731178595840249e-05, |
|
"loss": 0.2035, |
|
"step": 98600 |
|
}, |
|
{ |
|
"epoch": 8.89, |
|
"eval_loss": 0.08421877771615982, |
|
"eval_runtime": 643.294, |
|
"eval_samples_per_second": 17.776, |
|
"eval_steps_per_second": 2.223, |
|
"eval_wer": 0.100050712796529, |
|
"step": 98600 |
|
}, |
|
{ |
|
"epoch": 8.91, |
|
"learning_rate": 1.2145298310711843e-05, |
|
"loss": 0.1946, |
|
"step": 98800 |
|
}, |
|
{ |
|
"epoch": 8.91, |
|
"eval_loss": 0.08399970084428787, |
|
"eval_runtime": 640.6136, |
|
"eval_samples_per_second": 17.85, |
|
"eval_steps_per_second": 2.232, |
|
"eval_wer": 0.1000957908378881, |
|
"step": 98800 |
|
}, |
|
{ |
|
"epoch": 8.93, |
|
"learning_rate": 1.1559418025583438e-05, |
|
"loss": 0.2038, |
|
"step": 99000 |
|
}, |
|
{ |
|
"epoch": 8.93, |
|
"eval_loss": 0.08454040437936783, |
|
"eval_runtime": 643.6723, |
|
"eval_samples_per_second": 17.765, |
|
"eval_steps_per_second": 2.222, |
|
"eval_wer": 0.10106496872710881, |
|
"step": 99000 |
|
}, |
|
{ |
|
"epoch": 8.95, |
|
"learning_rate": 1.0973537740455032e-05, |
|
"loss": 0.2043, |
|
"step": 99200 |
|
}, |
|
{ |
|
"epoch": 8.95, |
|
"eval_loss": 0.08372523635625839, |
|
"eval_runtime": 644.4701, |
|
"eval_samples_per_second": 17.743, |
|
"eval_steps_per_second": 2.219, |
|
"eval_wer": 0.10090719558235195, |
|
"step": 99200 |
|
}, |
|
{ |
|
"epoch": 8.97, |
|
"learning_rate": 1.0387657455326628e-05, |
|
"loss": 0.2244, |
|
"step": 99400 |
|
}, |
|
{ |
|
"epoch": 8.97, |
|
"eval_loss": 0.08242122083902359, |
|
"eval_runtime": 643.4311, |
|
"eval_samples_per_second": 17.772, |
|
"eval_steps_per_second": 2.222, |
|
"eval_wer": 0.10090719558235195, |
|
"step": 99400 |
|
}, |
|
{ |
|
"epoch": 8.99, |
|
"learning_rate": 9.801777170198222e-06, |
|
"loss": 0.2166, |
|
"step": 99600 |
|
}, |
|
{ |
|
"epoch": 8.99, |
|
"eval_loss": 0.08180003613233566, |
|
"eval_runtime": 642.4939, |
|
"eval_samples_per_second": 17.798, |
|
"eval_steps_per_second": 2.226, |
|
"eval_wer": 0.09989293965177214, |
|
"step": 99600 |
|
}, |
|
{ |
|
"epoch": 9.01, |
|
"learning_rate": 9.215896885069816e-06, |
|
"loss": 0.1912, |
|
"step": 99800 |
|
}, |
|
{ |
|
"epoch": 9.01, |
|
"eval_loss": 0.0824543908238411, |
|
"eval_runtime": 639.0239, |
|
"eval_samples_per_second": 17.894, |
|
"eval_steps_per_second": 2.238, |
|
"eval_wer": 0.0998591311207528, |
|
"step": 99800 |
|
}, |
|
{ |
|
"epoch": 9.03, |
|
"learning_rate": 8.630016599941411e-06, |
|
"loss": 0.1859, |
|
"step": 100000 |
|
}, |
|
{ |
|
"epoch": 9.03, |
|
"eval_loss": 0.08340790867805481, |
|
"eval_runtime": 641.2003, |
|
"eval_samples_per_second": 17.834, |
|
"eval_steps_per_second": 2.23, |
|
"eval_wer": 0.09992674818279146, |
|
"step": 100000 |
|
}, |
|
{ |
|
"epoch": 9.05, |
|
"learning_rate": 8.044136314813005e-06, |
|
"loss": 0.1773, |
|
"step": 100200 |
|
}, |
|
{ |
|
"epoch": 9.05, |
|
"eval_loss": 0.08390292525291443, |
|
"eval_runtime": 643.4716, |
|
"eval_samples_per_second": 17.771, |
|
"eval_steps_per_second": 2.222, |
|
"eval_wer": 0.0995773933622584, |
|
"step": 100200 |
|
}, |
|
{ |
|
"epoch": 9.07, |
|
"learning_rate": 7.458256029684601e-06, |
|
"loss": 0.1782, |
|
"step": 100400 |
|
}, |
|
{ |
|
"epoch": 9.07, |
|
"eval_loss": 0.08477972447872162, |
|
"eval_runtime": 656.0288, |
|
"eval_samples_per_second": 17.431, |
|
"eval_steps_per_second": 2.18, |
|
"eval_wer": 0.09971262748633572, |
|
"step": 100400 |
|
}, |
|
{ |
|
"epoch": 9.09, |
|
"learning_rate": 6.875305145981837e-06, |
|
"loss": 0.1756, |
|
"step": 100600 |
|
}, |
|
{ |
|
"epoch": 9.09, |
|
"eval_loss": 0.08468695729970932, |
|
"eval_runtime": 639.7418, |
|
"eval_samples_per_second": 17.874, |
|
"eval_steps_per_second": 2.235, |
|
"eval_wer": 0.09964501042429706, |
|
"step": 100600 |
|
}, |
|
{ |
|
"epoch": 9.11, |
|
"learning_rate": 6.289424860853432e-06, |
|
"loss": 0.1922, |
|
"step": 100800 |
|
}, |
|
{ |
|
"epoch": 9.11, |
|
"eval_loss": 0.08449111133813858, |
|
"eval_runtime": 644.0375, |
|
"eval_samples_per_second": 17.755, |
|
"eval_steps_per_second": 2.22, |
|
"eval_wer": 0.09962247140361752, |
|
"step": 100800 |
|
}, |
|
{ |
|
"epoch": 9.13, |
|
"learning_rate": 5.7035445757250265e-06, |
|
"loss": 0.1859, |
|
"step": 101000 |
|
}, |
|
{ |
|
"epoch": 9.13, |
|
"eval_loss": 0.08459131419658661, |
|
"eval_runtime": 642.2544, |
|
"eval_samples_per_second": 17.804, |
|
"eval_steps_per_second": 2.227, |
|
"eval_wer": 0.09975770552769482, |
|
"step": 101000 |
|
}, |
|
{ |
|
"epoch": 9.15, |
|
"learning_rate": 5.117664290596621e-06, |
|
"loss": 0.1845, |
|
"step": 101200 |
|
}, |
|
{ |
|
"epoch": 9.15, |
|
"eval_loss": 0.08437124639749527, |
|
"eval_runtime": 642.6621, |
|
"eval_samples_per_second": 17.793, |
|
"eval_steps_per_second": 2.225, |
|
"eval_wer": 0.09970135797599594, |
|
"step": 101200 |
|
}, |
|
{ |
|
"epoch": 9.17, |
|
"learning_rate": 4.531784005468216e-06, |
|
"loss": 0.1956, |
|
"step": 101400 |
|
}, |
|
{ |
|
"epoch": 9.17, |
|
"eval_loss": 0.08400815725326538, |
|
"eval_runtime": 639.7693, |
|
"eval_samples_per_second": 17.874, |
|
"eval_steps_per_second": 2.235, |
|
"eval_wer": 0.09974643601735504, |
|
"step": 101400 |
|
}, |
|
{ |
|
"epoch": 9.19, |
|
"learning_rate": 3.945903720339811e-06, |
|
"loss": 0.1915, |
|
"step": 101600 |
|
}, |
|
{ |
|
"epoch": 9.19, |
|
"eval_loss": 0.0842614620923996, |
|
"eval_runtime": 639.7904, |
|
"eval_samples_per_second": 17.873, |
|
"eval_steps_per_second": 2.235, |
|
"eval_wer": 0.09966754944497662, |
|
"step": 101600 |
|
}, |
|
{ |
|
"epoch": 9.2, |
|
"learning_rate": 3.360023435211405e-06, |
|
"loss": 0.1876, |
|
"step": 101800 |
|
}, |
|
{ |
|
"epoch": 9.2, |
|
"eval_loss": 0.08413951843976974, |
|
"eval_runtime": 638.0015, |
|
"eval_samples_per_second": 17.923, |
|
"eval_steps_per_second": 2.241, |
|
"eval_wer": 0.09988167014143236, |
|
"step": 101800 |
|
}, |
|
{ |
|
"epoch": 9.22, |
|
"learning_rate": 2.7741431500829998e-06, |
|
"loss": 0.1898, |
|
"step": 102000 |
|
}, |
|
{ |
|
"epoch": 9.22, |
|
"eval_loss": 0.08397097885608673, |
|
"eval_runtime": 635.7725, |
|
"eval_samples_per_second": 17.986, |
|
"eval_steps_per_second": 2.249, |
|
"eval_wer": 0.09959993238293796, |
|
"step": 102000 |
|
}, |
|
{ |
|
"epoch": 9.24, |
|
"learning_rate": 2.188262864954594e-06, |
|
"loss": 0.2012, |
|
"step": 102200 |
|
}, |
|
{ |
|
"epoch": 9.24, |
|
"eval_loss": 0.08381262421607971, |
|
"eval_runtime": 637.045, |
|
"eval_samples_per_second": 17.95, |
|
"eval_steps_per_second": 2.245, |
|
"eval_wer": 0.09993801769313124, |
|
"step": 102200 |
|
}, |
|
{ |
|
"epoch": 9.26, |
|
"learning_rate": 1.6023825798261886e-06, |
|
"loss": 0.1891, |
|
"step": 102400 |
|
}, |
|
{ |
|
"epoch": 9.26, |
|
"eval_loss": 0.0839412659406662, |
|
"eval_runtime": 636.5331, |
|
"eval_samples_per_second": 17.965, |
|
"eval_steps_per_second": 2.247, |
|
"eval_wer": 0.0996788189553164, |
|
"step": 102400 |
|
}, |
|
{ |
|
"epoch": 9.28, |
|
"learning_rate": 1.0165022946977834e-06, |
|
"loss": 0.1746, |
|
"step": 102600 |
|
}, |
|
{ |
|
"epoch": 9.28, |
|
"eval_loss": 0.08403033018112183, |
|
"eval_runtime": 633.7847, |
|
"eval_samples_per_second": 18.042, |
|
"eval_steps_per_second": 2.256, |
|
"eval_wer": 0.09998309573449034, |
|
"step": 102600 |
|
}, |
|
{ |
|
"epoch": 9.3, |
|
"learning_rate": 4.3355141099501997e-07, |
|
"loss": 0.1859, |
|
"step": 102800 |
|
}, |
|
{ |
|
"epoch": 9.3, |
|
"eval_loss": 0.08405481278896332, |
|
"eval_runtime": 633.0898, |
|
"eval_samples_per_second": 18.062, |
|
"eval_steps_per_second": 2.259, |
|
"eval_wer": 0.09982532258973348, |
|
"step": 102800 |
|
}, |
|
{ |
|
"epoch": 9.31, |
|
"step": 102910, |
|
"total_flos": 1.7369851149894887e+20, |
|
"train_loss": 0.025842480747058687, |
|
"train_runtime": 56787.2686, |
|
"train_samples_per_second": 18.122, |
|
"train_steps_per_second": 1.812 |
|
} |
|
], |
|
"max_steps": 102910, |
|
"num_train_epochs": 10, |
|
"total_flos": 1.7369851149894887e+20, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|