| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.867046907267184, | |
| "global_step": 160000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 3.9921644212803626e-05, | |
| "loss": 4.8649, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 3.984297374373095e-05, | |
| "loss": 4.2624, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 3.9764303274658276e-05, | |
| "loss": 4.1126, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3.9685711476054684e-05, | |
| "loss": 3.9901, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 3.9607041006982005e-05, | |
| "loss": 3.8152, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 3.9528370537909334e-05, | |
| "loss": 3.4152, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 3.944970006883666e-05, | |
| "loss": 3.0102, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 3.937102959976399e-05, | |
| "loss": 2.6704, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 3.929235913069132e-05, | |
| "loss": 2.4043, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 3.921376733208772e-05, | |
| "loss": 2.1653, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 3.913509686301505e-05, | |
| "loss": 1.9766, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 3.905642639394238e-05, | |
| "loss": 1.7996, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 3.897775592486971e-05, | |
| "loss": 1.66, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 3.8899085455797035e-05, | |
| "loss": 1.5483, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 3.8820414986724364e-05, | |
| "loss": 1.4356, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 3.8741823188120765e-05, | |
| "loss": 1.335, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 3.8663231389517165e-05, | |
| "loss": 1.255, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 3.8584560920444494e-05, | |
| "loss": 1.1804, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 3.8505969121840894e-05, | |
| "loss": 1.1252, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 3.8427377323237294e-05, | |
| "loss": 1.0548, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_cer": 0.390676029242807, | |
| "eval_loss": 0.9201487302780151, | |
| "eval_runtime": 4761.7643, | |
| "eval_samples_per_second": 15.186, | |
| "eval_steps_per_second": 0.475, | |
| "eval_wer": 0.49385302835019235, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 3.8348785524633695e-05, | |
| "loss": 0.9557, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 3.8270115055561024e-05, | |
| "loss": 0.9954, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 3.8191444586488346e-05, | |
| "loss": 0.973, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 3.811277411741568e-05, | |
| "loss": 0.9175, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 3.803410364834301e-05, | |
| "loss": 0.8689, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 3.795543317927033e-05, | |
| "loss": 0.8241, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 3.787692005113581e-05, | |
| "loss": 0.802, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 3.779824958206314e-05, | |
| "loss": 0.7675, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 3.771957911299046e-05, | |
| "loss": 0.7455, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 3.764106598485594e-05, | |
| "loss": 0.7133, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 3.756239551578327e-05, | |
| "loss": 0.688, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 3.748372504671059e-05, | |
| "loss": 0.6583, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 3.740505457763792e-05, | |
| "loss": 0.6379, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 3.7326462779034327e-05, | |
| "loss": 0.6216, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 3.724787098043072e-05, | |
| "loss": 0.6076, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 3.716920051135805e-05, | |
| "loss": 0.5834, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 3.7090530042285384e-05, | |
| "loss": 0.5629, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 3.7011859573212706e-05, | |
| "loss": 0.552, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 3.6933189104140035e-05, | |
| "loss": 0.533, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 3.685459730553644e-05, | |
| "loss": 0.5201, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_cer": 0.25913786335133093, | |
| "eval_loss": 0.621722936630249, | |
| "eval_runtime": 4764.6248, | |
| "eval_samples_per_second": 15.177, | |
| "eval_steps_per_second": 0.474, | |
| "eval_wer": 0.32157834292047605, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 3.6775926836463764e-05, | |
| "loss": 0.4791, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 3.669725636739109e-05, | |
| "loss": 0.4229, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 3.661858589831842e-05, | |
| "loss": 0.4253, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 3.653999409971482e-05, | |
| "loss": 0.4222, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 3.646140230111122e-05, | |
| "loss": 0.3999, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 3.638273183203855e-05, | |
| "loss": 0.4048, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 3.630406136296588e-05, | |
| "loss": 0.3977, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 3.622539089389321e-05, | |
| "loss": 0.3918, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 3.614672042482054e-05, | |
| "loss": 0.387, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 3.606812862621694e-05, | |
| "loss": 0.3815, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 3.5989458157144266e-05, | |
| "loss": 0.3732, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 3.591086635854067e-05, | |
| "loss": 0.3737, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 3.5832195889467995e-05, | |
| "loss": 0.3589, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 3.5753525420395324e-05, | |
| "loss": 0.359, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 3.5674854951322646e-05, | |
| "loss": 0.3536, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 3.5596341823188125e-05, | |
| "loss": 0.346, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 3.5517750024584525e-05, | |
| "loss": 0.3398, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 3.543907955551185e-05, | |
| "loss": 0.3382, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 3.536040908643918e-05, | |
| "loss": 0.3305, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 3.528173861736651e-05, | |
| "loss": 0.3248, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "eval_cer": 0.19179082390807053, | |
| "eval_loss": 0.41522958874702454, | |
| "eval_runtime": 4766.524, | |
| "eval_samples_per_second": 15.171, | |
| "eval_steps_per_second": 0.474, | |
| "eval_wer": 0.24954789141369815, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.520306814829383e-05, | |
| "loss": 0.3219, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 3.512439767922116e-05, | |
| "loss": 0.2627, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 3.50457272101485e-05, | |
| "loss": 0.2599, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 3.496705674107582e-05, | |
| "loss": 0.2617, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 3.488838627200315e-05, | |
| "loss": 0.2612, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 3.480979447339955e-05, | |
| "loss": 0.263, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "learning_rate": 3.473112400432688e-05, | |
| "loss": 0.2573, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 3.465253220572328e-05, | |
| "loss": 0.2533, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 3.4573861736650606e-05, | |
| "loss": 0.256, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 3.4495191267577935e-05, | |
| "loss": 0.2498, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 3.4416520798505264e-05, | |
| "loss": 0.248, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 3.4337928999901664e-05, | |
| "loss": 0.2458, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 3.4259337201298065e-05, | |
| "loss": 0.2498, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 3.418066673222539e-05, | |
| "loss": 0.2423, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 3.4102074933621794e-05, | |
| "loss": 0.2374, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 3.402340446454912e-05, | |
| "loss": 0.2427, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 3.394473399547645e-05, | |
| "loss": 0.2374, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 3.386606352640378e-05, | |
| "loss": 0.24, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 3.378739305733111e-05, | |
| "loss": 0.2303, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 3.370872258825844e-05, | |
| "loss": 0.2337, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "eval_cer": 0.17718038202544326, | |
| "eval_loss": 0.568325936794281, | |
| "eval_runtime": 4765.8542, | |
| "eval_samples_per_second": 15.173, | |
| "eval_steps_per_second": 0.474, | |
| "eval_wer": 0.23509144980666905, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 3.3630052119185766e-05, | |
| "loss": 0.2298, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 3.355138165011309e-05, | |
| "loss": 0.1984, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 3.3472789851509495e-05, | |
| "loss": 0.1805, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 3.3394198052905895e-05, | |
| "loss": 0.1812, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "learning_rate": 3.331552758383322e-05, | |
| "loss": 0.1835, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "learning_rate": 3.323685711476055e-05, | |
| "loss": 0.1796, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 3.315818664568788e-05, | |
| "loss": 0.186, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "learning_rate": 3.3079516176615203e-05, | |
| "loss": 0.1844, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "learning_rate": 3.300084570754253e-05, | |
| "loss": 0.1815, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "learning_rate": 3.292225390893893e-05, | |
| "loss": 0.1796, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 3.284366211033533e-05, | |
| "loss": 0.1806, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 3.276499164126266e-05, | |
| "loss": 0.1789, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 3.268639984265907e-05, | |
| "loss": 0.1793, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 3.260772937358639e-05, | |
| "loss": 0.1771, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 3.252905890451372e-05, | |
| "loss": 0.1815, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 3.245038843544105e-05, | |
| "loss": 0.18, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "learning_rate": 3.237179663683745e-05, | |
| "loss": 0.1767, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "learning_rate": 3.229312616776478e-05, | |
| "loss": 0.1737, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 3.2214455698692106e-05, | |
| "loss": 0.1731, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "learning_rate": 3.2135863900088506e-05, | |
| "loss": 0.1754, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "eval_cer": 0.18740275862352454, | |
| "eval_loss": 0.3622461259365082, | |
| "eval_runtime": 4765.3375, | |
| "eval_samples_per_second": 15.175, | |
| "eval_steps_per_second": 0.474, | |
| "eval_wer": 0.2296571196445318, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 3.205727210148491e-05, | |
| "loss": 0.172, | |
| "step": 101000 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 3.1978601632412236e-05, | |
| "loss": 0.1624, | |
| "step": 102000 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 3.1899931163339564e-05, | |
| "loss": 0.1304, | |
| "step": 103000 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 3.182126069426689e-05, | |
| "loss": 0.1349, | |
| "step": 104000 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 3.174259022519422e-05, | |
| "loss": 0.1351, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "learning_rate": 3.166399842659062e-05, | |
| "loss": 0.1366, | |
| "step": 106000 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "learning_rate": 3.158532795751795e-05, | |
| "loss": 0.1363, | |
| "step": 107000 | |
| }, | |
| { | |
| "epoch": 5.31, | |
| "learning_rate": 3.150681482938342e-05, | |
| "loss": 0.1348, | |
| "step": 108000 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "learning_rate": 3.142814436031075e-05, | |
| "loss": 0.1348, | |
| "step": 109000 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "learning_rate": 3.134947389123808e-05, | |
| "loss": 0.1371, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "learning_rate": 3.127088209263448e-05, | |
| "loss": 0.1374, | |
| "step": 111000 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "learning_rate": 3.119221162356181e-05, | |
| "loss": 0.1333, | |
| "step": 112000 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "learning_rate": 3.111354115448914e-05, | |
| "loss": 0.1362, | |
| "step": 113000 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 3.103494935588554e-05, | |
| "loss": 0.1355, | |
| "step": 114000 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 3.095627888681287e-05, | |
| "loss": 0.1369, | |
| "step": 115000 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "learning_rate": 3.0877608417740196e-05, | |
| "loss": 0.1369, | |
| "step": 116000 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "learning_rate": 3.079893794866752e-05, | |
| "loss": 0.1368, | |
| "step": 117000 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "learning_rate": 3.0720424820533e-05, | |
| "loss": 0.1372, | |
| "step": 118000 | |
| }, | |
| { | |
| "epoch": 5.85, | |
| "learning_rate": 3.0641754351460326e-05, | |
| "loss": 0.1329, | |
| "step": 119000 | |
| }, | |
| { | |
| "epoch": 5.9, | |
| "learning_rate": 3.056316255285672e-05, | |
| "loss": 0.1339, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 5.9, | |
| "eval_cer": 0.16900559795182515, | |
| "eval_loss": 0.891129732131958, | |
| "eval_runtime": 4771.5866, | |
| "eval_samples_per_second": 15.155, | |
| "eval_steps_per_second": 0.474, | |
| "eval_wer": 0.2033450518809021, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 5.95, | |
| "learning_rate": 3.048449208378405e-05, | |
| "loss": 0.1319, | |
| "step": 121000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3.0405900285180455e-05, | |
| "loss": 0.1363, | |
| "step": 122000 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 3.032722981610778e-05, | |
| "loss": 0.1022, | |
| "step": 123000 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "learning_rate": 3.024855934703511e-05, | |
| "loss": 0.0999, | |
| "step": 124000 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 3.0169888877962438e-05, | |
| "loss": 0.1014, | |
| "step": 125000 | |
| }, | |
| { | |
| "epoch": 6.2, | |
| "learning_rate": 3.0091218408889763e-05, | |
| "loss": 0.1034, | |
| "step": 126000 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 3.0012547939817095e-05, | |
| "loss": 0.1034, | |
| "step": 127000 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 2.9933956141213496e-05, | |
| "loss": 0.101, | |
| "step": 128000 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "learning_rate": 2.985528567214082e-05, | |
| "loss": 0.1054, | |
| "step": 129000 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "learning_rate": 2.977661520306815e-05, | |
| "loss": 0.1026, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 6.44, | |
| "learning_rate": 2.969794473399548e-05, | |
| "loss": 0.1068, | |
| "step": 131000 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "learning_rate": 2.961943160586095e-05, | |
| "loss": 0.1081, | |
| "step": 132000 | |
| }, | |
| { | |
| "epoch": 6.54, | |
| "learning_rate": 2.954076113678828e-05, | |
| "loss": 0.106, | |
| "step": 133000 | |
| }, | |
| { | |
| "epoch": 6.59, | |
| "learning_rate": 2.946209066771561e-05, | |
| "loss": 0.105, | |
| "step": 134000 | |
| }, | |
| { | |
| "epoch": 6.64, | |
| "learning_rate": 2.9383498869112008e-05, | |
| "loss": 0.1072, | |
| "step": 135000 | |
| }, | |
| { | |
| "epoch": 6.69, | |
| "learning_rate": 2.9304828400039337e-05, | |
| "loss": 0.1052, | |
| "step": 136000 | |
| }, | |
| { | |
| "epoch": 6.74, | |
| "learning_rate": 2.922615793096667e-05, | |
| "loss": 0.1069, | |
| "step": 137000 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "learning_rate": 2.9147487461893994e-05, | |
| "loss": 0.1075, | |
| "step": 138000 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 2.9068974333759466e-05, | |
| "loss": 0.1057, | |
| "step": 139000 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "learning_rate": 2.899038253515587e-05, | |
| "loss": 0.1054, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "eval_cer": 0.152140834629885, | |
| "eval_loss": 0.7343257069587708, | |
| "eval_runtime": 4764.4162, | |
| "eval_samples_per_second": 15.178, | |
| "eval_steps_per_second": 0.474, | |
| "eval_wer": 0.19141319665595113, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "learning_rate": 2.8911712066083196e-05, | |
| "loss": 0.1085, | |
| "step": 141000 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "learning_rate": 2.8833041597010524e-05, | |
| "loss": 0.1064, | |
| "step": 142000 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 2.8754528468876e-05, | |
| "loss": 0.0993, | |
| "step": 143000 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "learning_rate": 2.8675857999803325e-05, | |
| "loss": 0.079, | |
| "step": 144000 | |
| }, | |
| { | |
| "epoch": 7.13, | |
| "learning_rate": 2.8597187530730654e-05, | |
| "loss": 0.0799, | |
| "step": 145000 | |
| }, | |
| { | |
| "epoch": 7.18, | |
| "learning_rate": 2.8518517061657986e-05, | |
| "loss": 0.0803, | |
| "step": 146000 | |
| }, | |
| { | |
| "epoch": 7.23, | |
| "learning_rate": 2.843984659258531e-05, | |
| "loss": 0.0808, | |
| "step": 147000 | |
| }, | |
| { | |
| "epoch": 7.28, | |
| "learning_rate": 2.8361412134919855e-05, | |
| "loss": 0.1072, | |
| "step": 148000 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "learning_rate": 2.8282741665847187e-05, | |
| "loss": 0.0799, | |
| "step": 149000 | |
| }, | |
| { | |
| "epoch": 7.38, | |
| "learning_rate": 2.8204071196774513e-05, | |
| "loss": 0.0821, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "learning_rate": 2.812540072770184e-05, | |
| "loss": 0.0838, | |
| "step": 151000 | |
| }, | |
| { | |
| "epoch": 7.47, | |
| "learning_rate": 2.804673025862917e-05, | |
| "loss": 0.0843, | |
| "step": 152000 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "learning_rate": 2.7968059789556495e-05, | |
| "loss": 0.0831, | |
| "step": 153000 | |
| }, | |
| { | |
| "epoch": 7.57, | |
| "learning_rate": 2.7889389320483827e-05, | |
| "loss": 0.0831, | |
| "step": 154000 | |
| }, | |
| { | |
| "epoch": 7.62, | |
| "learning_rate": 2.7810797521880228e-05, | |
| "loss": 0.0836, | |
| "step": 155000 | |
| }, | |
| { | |
| "epoch": 7.67, | |
| "learning_rate": 2.7732127052807553e-05, | |
| "loss": 0.0838, | |
| "step": 156000 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "learning_rate": 2.7653456583734882e-05, | |
| "loss": 0.0825, | |
| "step": 157000 | |
| }, | |
| { | |
| "epoch": 7.77, | |
| "learning_rate": 2.7574786114662214e-05, | |
| "loss": 0.0848, | |
| "step": 158000 | |
| }, | |
| { | |
| "epoch": 7.82, | |
| "learning_rate": 2.749611564558954e-05, | |
| "loss": 0.0865, | |
| "step": 159000 | |
| }, | |
| { | |
| "epoch": 7.87, | |
| "learning_rate": 2.7417445176516868e-05, | |
| "loss": 0.0847, | |
| "step": 160000 | |
| }, | |
| { | |
| "epoch": 7.87, | |
| "eval_cer": 0.1614900513208385, | |
| "eval_loss": 0.4246867895126343, | |
| "eval_runtime": 4767.9424, | |
| "eval_samples_per_second": 15.166, | |
| "eval_steps_per_second": 0.474, | |
| "eval_wer": 0.2072768669852908, | |
| "step": 160000 | |
| } | |
| ], | |
| "max_steps": 508450, | |
| "num_train_epochs": 25, | |
| "total_flos": 2.727768050247581e+21, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |