| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 99.9957805907173, | |
| "global_step": 11800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.00019800000000000002, | |
| "loss": 5.6106, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.000398, | |
| "loss": 2.7044, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 0.000598, | |
| "loss": 0.8324, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 0.0007980000000000001, | |
| "loss": 0.345, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 0.000998, | |
| "loss": 0.2548, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "eval_loss": 0.2469930201768875, | |
| "eval_runtime": 294.3209, | |
| "eval_samples_per_second": 21.959, | |
| "eval_steps_per_second": 0.306, | |
| "eval_wer": 0.36629738582545746, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 0.0009998106236074637, | |
| "loss": 0.2203, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "learning_rate": 0.0009992349703990264, | |
| "loss": 0.1774, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 0.0009982734659650546, | |
| "loss": 0.1643, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.62, | |
| "learning_rate": 0.0009969268534379132, | |
| "loss": 0.1516, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "learning_rate": 0.00099519617359424, | |
| "loss": 0.1435, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "eval_loss": 0.20002000033855438, | |
| "eval_runtime": 295.0975, | |
| "eval_samples_per_second": 21.901, | |
| "eval_steps_per_second": 0.305, | |
| "eval_wer": 0.2791095533162254, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.32, | |
| "learning_rate": 0.0009930827640505444, | |
| "loss": 0.1393, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 10.17, | |
| "learning_rate": 0.0009905882582293853, | |
| "loss": 0.1323, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "learning_rate": 0.0009877145840969208, | |
| "loss": 0.1317, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 11.86, | |
| "learning_rate": 0.0009844639626728146, | |
| "loss": 0.1156, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 12.71, | |
| "learning_rate": 0.000980838906313641, | |
| "loss": 0.1158, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 12.71, | |
| "eval_loss": 0.20303700864315033, | |
| "eval_runtime": 301.1246, | |
| "eval_samples_per_second": 21.463, | |
| "eval_steps_per_second": 0.299, | |
| "eval_wer": 0.26521457929106423, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 13.56, | |
| "learning_rate": 0.0009768422167711244, | |
| "loss": 0.1174, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 14.41, | |
| "learning_rate": 0.0009724769830267044, | |
| "loss": 0.1118, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 15.25, | |
| "learning_rate": 0.0009677465789041117, | |
| "loss": 0.1115, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 16.1, | |
| "learning_rate": 0.0009626546604617881, | |
| "loss": 0.1149, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 16.95, | |
| "learning_rate": 0.0009572051631671743, | |
| "loss": 0.1094, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 16.95, | |
| "eval_loss": 0.20957913994789124, | |
| "eval_runtime": 294.9247, | |
| "eval_samples_per_second": 21.914, | |
| "eval_steps_per_second": 0.305, | |
| "eval_wer": 0.26045774133650457, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 17.79, | |
| "learning_rate": 0.0009514022988550459, | |
| "loss": 0.1049, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 18.64, | |
| "learning_rate": 0.0009452505524722506, | |
| "loss": 0.1031, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 19.49, | |
| "learning_rate": 0.0009387546786113578, | |
| "loss": 0.0991, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 20.34, | |
| "learning_rate": 0.0009319196978359077, | |
| "loss": 0.1015, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 21.19, | |
| "learning_rate": 0.0009247508928000911, | |
| "loss": 0.1004, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 21.19, | |
| "eval_loss": 0.21501973271369934, | |
| "eval_runtime": 295.0309, | |
| "eval_samples_per_second": 21.906, | |
| "eval_steps_per_second": 0.305, | |
| "eval_wer": 0.2477311134756212, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 22.03, | |
| "learning_rate": 0.0009172538041658665, | |
| "loss": 0.0949, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 22.88, | |
| "learning_rate": 0.0009094342263206646, | |
| "loss": 0.0869, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 23.73, | |
| "learning_rate": 0.0009012982028989929, | |
| "loss": 0.0971, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 24.57, | |
| "learning_rate": 0.000892852022111403, | |
| "loss": 0.0907, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 25.42, | |
| "learning_rate": 0.0008841022118844243, | |
| "loss": 0.0945, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 25.42, | |
| "eval_loss": 0.20718325674533844, | |
| "eval_runtime": 295.2378, | |
| "eval_samples_per_second": 21.891, | |
| "eval_steps_per_second": 0.305, | |
| "eval_wer": 0.2369030481316893, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 26.27, | |
| "learning_rate": 0.0008750555348152298, | |
| "loss": 0.0884, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 27.12, | |
| "learning_rate": 0.0008657189829449246, | |
| "loss": 0.0894, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 27.96, | |
| "learning_rate": 0.0008560997723545024, | |
| "loss": 0.0851, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 28.81, | |
| "learning_rate": 0.0008462053375876455, | |
| "loss": 0.09, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 29.66, | |
| "learning_rate": 0.0008360433259046764, | |
| "loss": 0.0844, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 29.66, | |
| "eval_loss": 0.1981162428855896, | |
| "eval_runtime": 296.5443, | |
| "eval_samples_per_second": 21.794, | |
| "eval_steps_per_second": 0.303, | |
| "eval_wer": 0.23275124658363064, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 30.51, | |
| "learning_rate": 0.0008256215913721052, | |
| "loss": 0.084, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 31.35, | |
| "learning_rate": 0.0008149481887923395, | |
| "loss": 0.081, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 32.2, | |
| "learning_rate": 0.0008040313674782489, | |
| "loss": 0.0812, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 33.05, | |
| "learning_rate": 0.0007928795648773933, | |
| "loss": 0.0836, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 33.89, | |
| "learning_rate": 0.0007815014000508474, | |
| "loss": 0.0877, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 33.89, | |
| "eval_loss": 0.20411191880702972, | |
| "eval_runtime": 294.5281, | |
| "eval_samples_per_second": 21.944, | |
| "eval_steps_per_second": 0.306, | |
| "eval_wer": 0.24249441906073313, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 34.74, | |
| "learning_rate": 0.0007699056670116568, | |
| "loss": 0.0842, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 35.59, | |
| "learning_rate": 0.0007581013279280784, | |
| "loss": 0.0802, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 36.44, | |
| "learning_rate": 0.0007460975061968531, | |
| "loss": 0.0812, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 37.29, | |
| "learning_rate": 0.0007339034793918703, | |
| "loss": 0.0754, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 38.14, | |
| "learning_rate": 0.00072152867209367, | |
| "loss": 0.0741, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 38.14, | |
| "eval_loss": 0.235269233584404, | |
| "eval_runtime": 301.8823, | |
| "eval_samples_per_second": 21.409, | |
| "eval_steps_per_second": 0.298, | |
| "eval_wer": 0.24213974254657736, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 38.98, | |
| "learning_rate": 0.0007089826486053256, | |
| "loss": 0.0752, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 39.83, | |
| "learning_rate": 0.0006962751055603378, | |
| "loss": 0.0725, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 40.68, | |
| "learning_rate": 0.0006834158644282511, | |
| "loss": 0.0712, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 41.52, | |
| "learning_rate": 0.000670414863923788, | |
| "loss": 0.0673, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 42.37, | |
| "learning_rate": 0.0006572821523253649, | |
| "loss": 0.0676, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 42.37, | |
| "eval_loss": 0.20924021303653717, | |
| "eval_runtime": 294.7332, | |
| "eval_samples_per_second": 21.928, | |
| "eval_steps_per_second": 0.305, | |
| "eval_wer": 0.22127641818447352, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 43.22, | |
| "learning_rate": 0.000644027879708929, | |
| "loss": 0.0691, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 44.07, | |
| "learning_rate": 0.0006306622901031152, | |
| "loss": 0.0621, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 44.91, | |
| "learning_rate": 0.000617195713571792, | |
| "loss": 0.0648, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 45.76, | |
| "learning_rate": 0.0006036385582301083, | |
| "loss": 0.0682, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 46.61, | |
| "learning_rate": 0.0005900013022002183, | |
| "loss": 0.0623, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 46.61, | |
| "eval_loss": 0.22174060344696045, | |
| "eval_runtime": 298.483, | |
| "eval_samples_per_second": 21.653, | |
| "eval_steps_per_second": 0.302, | |
| "eval_wer": 0.2250318165696522, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 47.46, | |
| "learning_rate": 0.0005762944855128968, | |
| "loss": 0.0702, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 48.3, | |
| "learning_rate": 0.0005625287019613087, | |
| "loss": 0.0597, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 49.15, | |
| "learning_rate": 0.0005487145909132243, | |
| "loss": 0.0557, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 0.0005348628290880112, | |
| "loss": 0.0547, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 50.84, | |
| "learning_rate": 0.0005209841223047587, | |
| "loss": 0.0574, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 50.84, | |
| "eval_loss": 0.21524770557880402, | |
| "eval_runtime": 297.0196, | |
| "eval_samples_per_second": 21.76, | |
| "eval_steps_per_second": 0.303, | |
| "eval_wer": 0.2179382862865369, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 51.69, | |
| "learning_rate": 0.00050708919720791, | |
| "loss": 0.0571, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 52.54, | |
| "learning_rate": 0.0004931887929768004, | |
| "loss": 0.0557, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 53.39, | |
| "learning_rate": 0.0004792936530255048, | |
| "loss": 0.0513, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 54.24, | |
| "learning_rate": 0.00046541451669941626, | |
| "loss": 0.0542, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 55.08, | |
| "learning_rate": 0.00045156211097496503, | |
| "loss": 0.0583, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 55.08, | |
| "eval_loss": 0.22067895531654358, | |
| "eval_runtime": 297.4094, | |
| "eval_samples_per_second": 21.731, | |
| "eval_steps_per_second": 0.303, | |
| "eval_wer": 0.21860591266612422, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 55.93, | |
| "learning_rate": 0.00043774714216890067, | |
| "loss": 0.0516, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 56.78, | |
| "learning_rate": 0.000423980287663539, | |
| "loss": 0.0552, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 57.62, | |
| "learning_rate": 0.00041027218765437467, | |
| "loss": 0.0545, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 58.47, | |
| "learning_rate": 0.0003966334369264325, | |
| "loss": 0.0497, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 59.32, | |
| "learning_rate": 0.00038307457666571714, | |
| "loss": 0.0488, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 59.32, | |
| "eval_loss": 0.22246450185775757, | |
| "eval_runtime": 291.806, | |
| "eval_samples_per_second": 22.148, | |
| "eval_steps_per_second": 0.308, | |
| "eval_wer": 0.21593540714777493, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 60.17, | |
| "learning_rate": 0.000369606086312089, | |
| "loss": 0.0467, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 61.02, | |
| "learning_rate": 0.00035623837545986186, | |
| "loss": 0.0496, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 61.86, | |
| "learning_rate": 0.0003429817758123833, | |
| "loss": 0.0477, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 62.71, | |
| "learning_rate": 0.00032984653319681614, | |
| "loss": 0.0491, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 63.56, | |
| "learning_rate": 0.0003168427996452925, | |
| "loss": 0.0456, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 63.56, | |
| "eval_loss": 0.22932520508766174, | |
| "eval_runtime": 293.0052, | |
| "eval_samples_per_second": 22.058, | |
| "eval_steps_per_second": 0.307, | |
| "eval_wer": 0.20310446266508106, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 64.41, | |
| "learning_rate": 0.0003039806255485599, | |
| "loss": 0.042, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 65.25, | |
| "learning_rate": 0.0002912699518881855, | |
| "loss": 0.0431, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 66.1, | |
| "learning_rate": 0.00027872060255331935, | |
| "loss": 0.042, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 66.95, | |
| "learning_rate": 0.0002663422767479589, | |
| "loss": 0.0422, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 67.79, | |
| "learning_rate": 0.0002541445414945791, | |
| "loss": 0.041, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 67.79, | |
| "eval_loss": 0.22772204875946045, | |
| "eval_runtime": 292.1439, | |
| "eval_samples_per_second": 22.123, | |
| "eval_steps_per_second": 0.308, | |
| "eval_wer": 0.20128935344557802, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 68.64, | |
| "learning_rate": 0.000242136824239924, | |
| "loss": 0.042, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 69.49, | |
| "learning_rate": 0.00023032840556867502, | |
| "loss": 0.0404, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 70.34, | |
| "learning_rate": 0.00021872841203062537, | |
| "loss": 0.0413, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 71.19, | |
| "learning_rate": 0.0002074585300038035, | |
| "loss": 0.0385, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 72.03, | |
| "learning_rate": 0.00019629981023833903, | |
| "loss": 0.0379, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 72.03, | |
| "eval_loss": 0.22867247462272644, | |
| "eval_runtime": 297.2969, | |
| "eval_samples_per_second": 21.739, | |
| "eval_steps_per_second": 0.303, | |
| "eval_wer": 0.199077841063195, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 72.88, | |
| "learning_rate": 0.00018537581579734625, | |
| "loss": 0.037, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 73.73, | |
| "learning_rate": 0.00017469498967238802, | |
| "loss": 0.0362, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 74.57, | |
| "learning_rate": 0.00016426558691388015, | |
| "loss": 0.0348, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 75.42, | |
| "learning_rate": 0.0001540956682508869, | |
| "loss": 0.0367, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 76.27, | |
| "learning_rate": 0.00014419309386110528, | |
| "loss": 0.0381, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 76.27, | |
| "eval_loss": 0.22330239415168762, | |
| "eval_runtime": 290.8488, | |
| "eval_samples_per_second": 22.221, | |
| "eval_steps_per_second": 0.309, | |
| "eval_wer": 0.19536416932674053, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 77.12, | |
| "learning_rate": 0.00013456551729585032, | |
| "loss": 0.0343, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 77.96, | |
| "learning_rate": 0.00012522037956474042, | |
| "loss": 0.033, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 78.81, | |
| "learning_rate": 0.00011616490338465268, | |
| "loss": 0.032, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 79.66, | |
| "learning_rate": 0.00010740608759739317, | |
| "loss": 0.031, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 80.51, | |
| "learning_rate": 9.895070176039717e-05, | |
| "loss": 0.0308, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 80.51, | |
| "eval_loss": 0.21947245299816132, | |
| "eval_runtime": 295.835, | |
| "eval_samples_per_second": 21.847, | |
| "eval_steps_per_second": 0.304, | |
| "eval_wer": 0.18349293776470343, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 81.35, | |
| "learning_rate": 9.080528091463946e-05, | |
| "loss": 0.0309, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 82.2, | |
| "learning_rate": 8.297612053379883e-05, | |
| "loss": 0.0302, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 83.05, | |
| "learning_rate": 7.546927165858153e-05, | |
| "loss": 0.0296, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 83.89, | |
| "learning_rate": 6.82905362199625e-05, | |
| "loss": 0.0289, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 84.74, | |
| "learning_rate": 6.14454625549593e-05, | |
| "loss": 0.0291, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 84.74, | |
| "eval_loss": 0.22659379243850708, | |
| "eval_runtime": 301.2419, | |
| "eval_samples_per_second": 21.455, | |
| "eval_steps_per_second": 0.299, | |
| "eval_wer": 0.18251236151968456, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 85.59, | |
| "learning_rate": 5.493934111840604e-05, | |
| "loss": 0.029, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 86.44, | |
| "learning_rate": 4.87772003940401e-05, | |
| "loss": 0.0276, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 87.29, | |
| "learning_rate": 4.296380300806219e-05, | |
| "loss": 0.0285, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 88.14, | |
| "learning_rate": 3.750364204817463e-05, | |
| "loss": 0.0287, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 88.98, | |
| "learning_rate": 3.245018204574429e-05, | |
| "loss": 0.0266, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 88.98, | |
| "eval_loss": 0.22852544486522675, | |
| "eval_runtime": 296.9937, | |
| "eval_samples_per_second": 21.761, | |
| "eval_steps_per_second": 0.303, | |
| "eval_wer": 0.1800713525693184, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 89.83, | |
| "learning_rate": 2.7705245284789292e-05, | |
| "loss": 0.0269, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 90.68, | |
| "learning_rate": 2.3325338060476542e-05, | |
| "loss": 0.0262, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 91.52, | |
| "learning_rate": 1.9313845537448972e-05, | |
| "loss": 0.0257, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 92.37, | |
| "learning_rate": 1.567386813814875e-05, | |
| "loss": 0.0271, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 93.22, | |
| "learning_rate": 1.2408219146547828e-05, | |
| "loss": 0.0266, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 93.22, | |
| "eval_loss": 0.229219451546669, | |
| "eval_runtime": 299.2804, | |
| "eval_samples_per_second": 21.595, | |
| "eval_steps_per_second": 0.301, | |
| "eval_wer": 0.1801130792180426, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 94.07, | |
| "learning_rate": 9.519422533802668e-06, | |
| "loss": 0.027, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 94.91, | |
| "learning_rate": 7.009711007514796e-06, | |
| "loss": 0.0266, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 95.76, | |
| "learning_rate": 4.8810242861045405e-06, | |
| "loss": 0.0258, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 96.61, | |
| "learning_rate": 3.1350075996317563e-06, | |
| "loss": 0.0256, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 97.46, | |
| "learning_rate": 1.773010418222143e-06, | |
| "loss": 0.0262, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 97.46, | |
| "eval_loss": 0.22784815728664398, | |
| "eval_runtime": 323.2034, | |
| "eval_samples_per_second": 19.997, | |
| "eval_steps_per_second": 0.278, | |
| "eval_wer": 0.17881955310759218, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 98.3, | |
| "learning_rate": 7.96085409082159e-07, | |
| "loss": 0.0267, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 99.15, | |
| "learning_rate": 2.049876229081038e-07, | |
| "loss": 0.026, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 1.7391031886804242e-10, | |
| "loss": 0.0257, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 11800, | |
| "total_flos": 2.7805901871717807e+20, | |
| "train_loss": 0.14547648023750823, | |
| "train_runtime": 96097.4367, | |
| "train_samples_per_second": 17.722, | |
| "train_steps_per_second": 0.123 | |
| } | |
| ], | |
| "max_steps": 11800, | |
| "num_train_epochs": 100, | |
| "total_flos": 2.7805901871717807e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |