| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 39.90057995028997, | |
| "eval_steps": 1000, | |
| "global_step": 12040, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 5.6399999999999995e-05, | |
| "loss": 37.3417, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0001164, | |
| "loss": 12.5064, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 0.00017639999999999998, | |
| "loss": 4.823, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0002364, | |
| "loss": 4.7159, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 0.0002964, | |
| "loss": 4.6828, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.0002975563258232235, | |
| "loss": 4.5391, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 0.0002949566724436741, | |
| "loss": 4.3365, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 0.0002923570190641247, | |
| "loss": 3.2316, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 0.00028975736568457536, | |
| "loss": 2.1607, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 0.00028715771230502596, | |
| "loss": 1.778, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "eval_cer": 0.3050481701797491, | |
| "eval_loss": 1.2772819995880127, | |
| "eval_runtime": 143.2917, | |
| "eval_samples_per_second": 33.77, | |
| "eval_steps_per_second": 2.115, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 0.00028455805892547655, | |
| "loss": 1.6287, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 0.0002819584055459272, | |
| "loss": 1.5046, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 0.0002793587521663778, | |
| "loss": 1.4046, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "learning_rate": 0.0002767590987868284, | |
| "loss": 1.3378, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 0.000274159445407279, | |
| "loss": 1.2868, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.3, | |
| "learning_rate": 0.0002715597920277296, | |
| "loss": 1.2376, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.63, | |
| "learning_rate": 0.00026896013864818023, | |
| "loss": 1.2126, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "learning_rate": 0.0002663604852686308, | |
| "loss": 1.1781, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 6.3, | |
| "learning_rate": 0.0002637608318890814, | |
| "loss": 1.1426, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "learning_rate": 0.00026116117850953207, | |
| "loss": 1.1037, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "eval_cer": 0.18878507694297167, | |
| "eval_loss": 0.7715919017791748, | |
| "eval_runtime": 143.3347, | |
| "eval_samples_per_second": 33.76, | |
| "eval_steps_per_second": 2.114, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "learning_rate": 0.00025856152512998266, | |
| "loss": 1.0971, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 7.29, | |
| "learning_rate": 0.00025596187175043326, | |
| "loss": 1.0693, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 7.62, | |
| "learning_rate": 0.00025336221837088385, | |
| "loss": 1.0398, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 7.95, | |
| "learning_rate": 0.00025076256499133445, | |
| "loss": 1.0282, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "learning_rate": 0.00024816291161178504, | |
| "loss": 0.9997, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.62, | |
| "learning_rate": 0.0002455632582322357, | |
| "loss": 0.995, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 8.95, | |
| "learning_rate": 0.00024296360485268629, | |
| "loss": 0.9721, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 9.28, | |
| "learning_rate": 0.00024036395147313688, | |
| "loss": 0.9639, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 9.61, | |
| "learning_rate": 0.0002377642980935875, | |
| "loss": 0.9543, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 9.94, | |
| "learning_rate": 0.0002351646447140381, | |
| "loss": 0.9529, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 9.94, | |
| "eval_cer": 0.16591232380706064, | |
| "eval_loss": 0.6726131439208984, | |
| "eval_runtime": 144.4701, | |
| "eval_samples_per_second": 33.495, | |
| "eval_steps_per_second": 2.097, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 10.27, | |
| "learning_rate": 0.0002325649913344887, | |
| "loss": 0.9281, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 10.6, | |
| "learning_rate": 0.00022996533795493934, | |
| "loss": 0.9274, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 10.94, | |
| "learning_rate": 0.00022736568457538994, | |
| "loss": 0.9149, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 11.27, | |
| "learning_rate": 0.00022476603119584053, | |
| "loss": 0.9036, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 11.6, | |
| "learning_rate": 0.00022216637781629115, | |
| "loss": 0.8838, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 11.93, | |
| "learning_rate": 0.00021956672443674175, | |
| "loss": 0.8872, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 12.26, | |
| "learning_rate": 0.00021696707105719234, | |
| "loss": 0.8661, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 12.59, | |
| "learning_rate": 0.00021436741767764296, | |
| "loss": 0.8529, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 12.92, | |
| "learning_rate": 0.00021176776429809356, | |
| "loss": 0.8566, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 13.26, | |
| "learning_rate": 0.00020916811091854415, | |
| "loss": 0.8424, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 13.26, | |
| "eval_cer": 0.1511541445751972, | |
| "eval_loss": 0.6138216257095337, | |
| "eval_runtime": 143.2958, | |
| "eval_samples_per_second": 33.769, | |
| "eval_steps_per_second": 2.115, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 13.59, | |
| "learning_rate": 0.0002065684575389948, | |
| "loss": 0.8403, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 13.92, | |
| "learning_rate": 0.0002039688041594454, | |
| "loss": 0.836, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 14.25, | |
| "learning_rate": 0.000201369150779896, | |
| "loss": 0.8119, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 14.58, | |
| "learning_rate": 0.0001987694974003466, | |
| "loss": 0.8298, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 14.91, | |
| "learning_rate": 0.0001961698440207972, | |
| "loss": 0.8019, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 15.24, | |
| "learning_rate": 0.0001935701906412478, | |
| "loss": 0.8, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 15.58, | |
| "learning_rate": 0.00019097053726169842, | |
| "loss": 0.7909, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 15.91, | |
| "learning_rate": 0.00018837088388214902, | |
| "loss": 0.7941, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 16.24, | |
| "learning_rate": 0.0001857712305025996, | |
| "loss": 0.7888, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 16.57, | |
| "learning_rate": 0.00018317157712305026, | |
| "loss": 0.767, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 16.57, | |
| "eval_cer": 0.14326587352903142, | |
| "eval_loss": 0.5884789228439331, | |
| "eval_runtime": 143.5148, | |
| "eval_samples_per_second": 33.718, | |
| "eval_steps_per_second": 2.111, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 16.9, | |
| "learning_rate": 0.00018057192374350086, | |
| "loss": 0.7655, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 17.23, | |
| "learning_rate": 0.00017797227036395145, | |
| "loss": 0.7571, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 17.56, | |
| "learning_rate": 0.00017537261698440207, | |
| "loss": 0.754, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 17.9, | |
| "learning_rate": 0.00017277296360485267, | |
| "loss": 0.764, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 18.23, | |
| "learning_rate": 0.0001701733102253033, | |
| "loss": 0.7461, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 18.56, | |
| "learning_rate": 0.00016757365684575388, | |
| "loss": 0.7363, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 18.89, | |
| "learning_rate": 0.00016497400346620448, | |
| "loss": 0.7315, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 19.22, | |
| "learning_rate": 0.0001623743500866551, | |
| "loss": 0.7193, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 19.55, | |
| "learning_rate": 0.00015977469670710572, | |
| "loss": 0.7253, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 19.88, | |
| "learning_rate": 0.00015717504332755632, | |
| "loss": 0.7201, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 19.88, | |
| "eval_cer": 0.1377699469804733, | |
| "eval_loss": 0.5682498812675476, | |
| "eval_runtime": 145.5022, | |
| "eval_samples_per_second": 33.257, | |
| "eval_steps_per_second": 2.082, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 20.22, | |
| "learning_rate": 0.00015457538994800694, | |
| "loss": 0.7203, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 20.55, | |
| "learning_rate": 0.00015197573656845753, | |
| "loss": 0.7045, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 20.88, | |
| "learning_rate": 0.00014937608318890813, | |
| "loss": 0.7072, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 21.21, | |
| "learning_rate": 0.00014677642980935872, | |
| "loss": 0.6979, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 21.54, | |
| "learning_rate": 0.00014417677642980934, | |
| "loss": 0.6971, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 21.87, | |
| "learning_rate": 0.00014157712305025997, | |
| "loss": 0.702, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 22.2, | |
| "learning_rate": 0.00013897746967071056, | |
| "loss": 0.6767, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 22.54, | |
| "learning_rate": 0.00013637781629116116, | |
| "loss": 0.682, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 22.87, | |
| "learning_rate": 0.00013377816291161178, | |
| "loss": 0.6729, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 23.2, | |
| "learning_rate": 0.00013120450606585787, | |
| "loss": 0.664, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 23.2, | |
| "eval_cer": 0.13325197206776154, | |
| "eval_loss": 0.5583386421203613, | |
| "eval_runtime": 143.6376, | |
| "eval_samples_per_second": 33.689, | |
| "eval_steps_per_second": 2.109, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 23.53, | |
| "learning_rate": 0.0001286048526863085, | |
| "loss": 0.668, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 23.86, | |
| "learning_rate": 0.00012600519930675908, | |
| "loss": 0.67, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 24.19, | |
| "learning_rate": 0.0001234055459272097, | |
| "loss": 0.6608, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 24.52, | |
| "learning_rate": 0.0001208058925476603, | |
| "loss": 0.6543, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 24.86, | |
| "learning_rate": 0.0001182062391681109, | |
| "loss": 0.6648, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 25.19, | |
| "learning_rate": 0.00011560658578856152, | |
| "loss": 0.6473, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 25.52, | |
| "learning_rate": 0.00011300693240901212, | |
| "loss": 0.6546, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 25.85, | |
| "learning_rate": 0.00011040727902946273, | |
| "loss": 0.6397, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 26.18, | |
| "learning_rate": 0.00010780762564991333, | |
| "loss": 0.6395, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 26.51, | |
| "learning_rate": 0.00010520797227036395, | |
| "loss": 0.6296, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 26.51, | |
| "eval_cer": 0.1298170179749127, | |
| "eval_loss": 0.5415648818016052, | |
| "eval_runtime": 142.9992, | |
| "eval_samples_per_second": 33.839, | |
| "eval_steps_per_second": 2.119, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 26.84, | |
| "learning_rate": 0.00010260831889081456, | |
| "loss": 0.6284, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 27.17, | |
| "learning_rate": 0.00010000866551126515, | |
| "loss": 0.6328, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 27.51, | |
| "learning_rate": 9.740901213171576e-05, | |
| "loss": 0.6233, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 27.84, | |
| "learning_rate": 9.480935875216638e-05, | |
| "loss": 0.6226, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 28.17, | |
| "learning_rate": 9.220970537261698e-05, | |
| "loss": 0.6214, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 28.5, | |
| "learning_rate": 8.961005199306758e-05, | |
| "loss": 0.6195, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 28.83, | |
| "learning_rate": 8.701039861351819e-05, | |
| "loss": 0.6138, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 29.16, | |
| "learning_rate": 8.441074523396879e-05, | |
| "loss": 0.6067, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 29.49, | |
| "learning_rate": 8.181109185441941e-05, | |
| "loss": 0.6076, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 29.83, | |
| "learning_rate": 7.921143847487002e-05, | |
| "loss": 0.6021, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 29.83, | |
| "eval_cer": 0.12716604163972586, | |
| "eval_loss": 0.5376999974250793, | |
| "eval_runtime": 142.2332, | |
| "eval_samples_per_second": 34.022, | |
| "eval_steps_per_second": 2.13, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 30.16, | |
| "learning_rate": 7.661178509532061e-05, | |
| "loss": 0.5996, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 30.49, | |
| "learning_rate": 7.401213171577122e-05, | |
| "loss": 0.5989, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 30.82, | |
| "learning_rate": 7.141247833622183e-05, | |
| "loss": 0.5924, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 31.15, | |
| "learning_rate": 6.881282495667244e-05, | |
| "loss": 0.5858, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 31.48, | |
| "learning_rate": 6.621317157712304e-05, | |
| "loss": 0.5846, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 31.81, | |
| "learning_rate": 6.363951473136915e-05, | |
| "loss": 0.5906, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 32.15, | |
| "learning_rate": 6.103986135181975e-05, | |
| "loss": 0.5823, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 32.48, | |
| "learning_rate": 5.8466204506065855e-05, | |
| "loss": 0.5853, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 32.81, | |
| "learning_rate": 5.586655112651646e-05, | |
| "loss": 0.5759, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 33.14, | |
| "learning_rate": 5.3266897746967065e-05, | |
| "loss": 0.568, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 33.14, | |
| "eval_cer": 0.1245797232639338, | |
| "eval_loss": 0.5240569710731506, | |
| "eval_runtime": 143.5422, | |
| "eval_samples_per_second": 33.711, | |
| "eval_steps_per_second": 2.111, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 33.47, | |
| "learning_rate": 5.066724436741767e-05, | |
| "loss": 0.574, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 33.8, | |
| "learning_rate": 4.806759098786828e-05, | |
| "loss": 0.5682, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 34.13, | |
| "learning_rate": 4.546793760831888e-05, | |
| "loss": 0.5636, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 34.47, | |
| "learning_rate": 4.28682842287695e-05, | |
| "loss": 0.5651, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 34.8, | |
| "learning_rate": 4.02686308492201e-05, | |
| "loss": 0.5693, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 35.13, | |
| "learning_rate": 3.766897746967071e-05, | |
| "loss": 0.5562, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 35.46, | |
| "learning_rate": 3.5069324090121316e-05, | |
| "loss": 0.5564, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 35.79, | |
| "learning_rate": 3.2469670710571924e-05, | |
| "loss": 0.5613, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 36.12, | |
| "learning_rate": 2.9870017331022526e-05, | |
| "loss": 0.5519, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 36.45, | |
| "learning_rate": 2.7270363951473134e-05, | |
| "loss": 0.5519, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 36.45, | |
| "eval_cer": 0.12275313591103065, | |
| "eval_loss": 0.5184260606765747, | |
| "eval_runtime": 142.3486, | |
| "eval_samples_per_second": 33.994, | |
| "eval_steps_per_second": 2.129, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 36.79, | |
| "learning_rate": 2.4670710571923742e-05, | |
| "loss": 0.5605, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 37.12, | |
| "learning_rate": 2.207105719237435e-05, | |
| "loss": 0.5512, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 37.45, | |
| "learning_rate": 1.9471403812824955e-05, | |
| "loss": 0.5539, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 37.78, | |
| "learning_rate": 1.687175043327556e-05, | |
| "loss": 0.5467, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 38.11, | |
| "learning_rate": 1.4272097053726168e-05, | |
| "loss": 0.5464, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 38.44, | |
| "learning_rate": 1.1672443674176775e-05, | |
| "loss": 0.5457, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 38.77, | |
| "learning_rate": 9.072790294627383e-06, | |
| "loss": 0.5445, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 39.11, | |
| "learning_rate": 6.473136915077989e-06, | |
| "loss": 0.5395, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 39.44, | |
| "learning_rate": 3.8734835355285955e-06, | |
| "loss": 0.542, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 39.77, | |
| "learning_rate": 1.2738301559792027e-06, | |
| "loss": 0.5395, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 39.77, | |
| "eval_cer": 0.12272080693133325, | |
| "eval_loss": 0.5156292915344238, | |
| "eval_runtime": 142.2326, | |
| "eval_samples_per_second": 34.022, | |
| "eval_steps_per_second": 2.13, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 39.9, | |
| "step": 12040, | |
| "total_flos": 1.9557952498596368e+20, | |
| "train_loss": 1.365207207163307, | |
| "train_runtime": 27512.8703, | |
| "train_samples_per_second": 56.137, | |
| "train_steps_per_second": 0.438 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 12040, | |
| "num_train_epochs": 40, | |
| "save_steps": 1000, | |
| "total_flos": 1.9557952498596368e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |