| { | |
| "best_metric": 2.4236774644604986, | |
| "best_model_checkpoint": "./final-whisper-for-initial-publish/checkpoint-3000", | |
| "epoch": 7.413509060955519, | |
| "eval_steps": 500, | |
| "global_step": 4500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04118616144975288, | |
| "grad_norm": 67.08393096923828, | |
| "learning_rate": 4.2000000000000006e-07, | |
| "loss": 2.8367, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.08237232289950576, | |
| "grad_norm": 17.447174072265625, | |
| "learning_rate": 9.200000000000001e-07, | |
| "loss": 2.0282, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.12355848434925865, | |
| "grad_norm": 12.4807710647583, | |
| "learning_rate": 1.42e-06, | |
| "loss": 1.3152, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.16474464579901152, | |
| "grad_norm": 8.900789260864258, | |
| "learning_rate": 1.9200000000000003e-06, | |
| "loss": 0.7248, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.20593080724876442, | |
| "grad_norm": 7.897016525268555, | |
| "learning_rate": 2.42e-06, | |
| "loss": 0.5731, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.2471169686985173, | |
| "grad_norm": 9.260579109191895, | |
| "learning_rate": 2.92e-06, | |
| "loss": 0.474, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.2883031301482702, | |
| "grad_norm": 6.941812515258789, | |
| "learning_rate": 3.4200000000000007e-06, | |
| "loss": 0.4304, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.32948929159802304, | |
| "grad_norm": 6.919547080993652, | |
| "learning_rate": 3.920000000000001e-06, | |
| "loss": 0.3514, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.37067545304777594, | |
| "grad_norm": 6.444908618927002, | |
| "learning_rate": 4.42e-06, | |
| "loss": 0.273, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.41186161449752884, | |
| "grad_norm": 5.828249454498291, | |
| "learning_rate": 4.92e-06, | |
| "loss": 0.1655, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.45304777594728174, | |
| "grad_norm": 2.492833137512207, | |
| "learning_rate": 5.420000000000001e-06, | |
| "loss": 0.0894, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.4942339373970346, | |
| "grad_norm": 3.24698805809021, | |
| "learning_rate": 5.92e-06, | |
| "loss": 0.076, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5354200988467874, | |
| "grad_norm": 4.378988265991211, | |
| "learning_rate": 6.42e-06, | |
| "loss": 0.0763, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.5766062602965404, | |
| "grad_norm": 3.146955728530884, | |
| "learning_rate": 6.92e-06, | |
| "loss": 0.0679, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.6177924217462932, | |
| "grad_norm": 3.8782553672790527, | |
| "learning_rate": 7.420000000000001e-06, | |
| "loss": 0.0524, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.6589785831960461, | |
| "grad_norm": 6.15730094909668, | |
| "learning_rate": 7.92e-06, | |
| "loss": 0.0554, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.700164744645799, | |
| "grad_norm": 6.1213812828063965, | |
| "learning_rate": 8.42e-06, | |
| "loss": 0.0535, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.7413509060955519, | |
| "grad_norm": 4.010768890380859, | |
| "learning_rate": 8.920000000000001e-06, | |
| "loss": 0.0465, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.7825370675453048, | |
| "grad_norm": 3.509321451187134, | |
| "learning_rate": 9.42e-06, | |
| "loss": 0.0436, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.8237232289950577, | |
| "grad_norm": 2.744082450866699, | |
| "learning_rate": 9.920000000000002e-06, | |
| "loss": 0.0468, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8237232289950577, | |
| "eval_loss": 0.04186907038092613, | |
| "eval_runtime": 1739.4189, | |
| "eval_samples_per_second": 1.395, | |
| "eval_steps_per_second": 0.175, | |
| "eval_wer": 10.859939408063388, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8649093904448105, | |
| "grad_norm": 2.581739664077759, | |
| "learning_rate": 9.953333333333333e-06, | |
| "loss": 0.0387, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.9060955518945635, | |
| "grad_norm": 1.7439745664596558, | |
| "learning_rate": 9.89777777777778e-06, | |
| "loss": 0.0372, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.9472817133443163, | |
| "grad_norm": 3.0741066932678223, | |
| "learning_rate": 9.842222222222223e-06, | |
| "loss": 0.0344, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.9884678747940692, | |
| "grad_norm": 2.640840530395508, | |
| "learning_rate": 9.786666666666667e-06, | |
| "loss": 0.0331, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.029654036243822, | |
| "grad_norm": 0.908171534538269, | |
| "learning_rate": 9.731111111111113e-06, | |
| "loss": 0.0267, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.0708401976935749, | |
| "grad_norm": 1.0449379682540894, | |
| "learning_rate": 9.675555555555555e-06, | |
| "loss": 0.0203, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.112026359143328, | |
| "grad_norm": 2.090158700942993, | |
| "learning_rate": 9.620000000000001e-06, | |
| "loss": 0.0185, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 1.1532125205930808, | |
| "grad_norm": 3.4184322357177734, | |
| "learning_rate": 9.564444444444445e-06, | |
| "loss": 0.0226, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.1943986820428336, | |
| "grad_norm": 2.6463236808776855, | |
| "learning_rate": 9.508888888888889e-06, | |
| "loss": 0.0155, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 1.2355848434925865, | |
| "grad_norm": 0.7438070774078369, | |
| "learning_rate": 9.453333333333335e-06, | |
| "loss": 0.0184, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.2767710049423393, | |
| "grad_norm": 1.7319138050079346, | |
| "learning_rate": 9.397777777777779e-06, | |
| "loss": 0.0161, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.3179571663920924, | |
| "grad_norm": 1.9704327583312988, | |
| "learning_rate": 9.342222222222223e-06, | |
| "loss": 0.0137, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.3591433278418452, | |
| "grad_norm": 1.808190107345581, | |
| "learning_rate": 9.286666666666667e-06, | |
| "loss": 0.0169, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 1.400329489291598, | |
| "grad_norm": 2.1106882095336914, | |
| "learning_rate": 9.231111111111111e-06, | |
| "loss": 0.0141, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.441515650741351, | |
| "grad_norm": 1.5317939519882202, | |
| "learning_rate": 9.175555555555557e-06, | |
| "loss": 0.0114, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 1.4827018121911038, | |
| "grad_norm": 2.758979082107544, | |
| "learning_rate": 9.12e-06, | |
| "loss": 0.0123, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.5238879736408566, | |
| "grad_norm": 2.9624338150024414, | |
| "learning_rate": 9.064444444444447e-06, | |
| "loss": 0.014, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 1.5650741350906094, | |
| "grad_norm": 2.520521402359009, | |
| "learning_rate": 9.008888888888889e-06, | |
| "loss": 0.0107, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.6062602965403623, | |
| "grad_norm": 1.0015959739685059, | |
| "learning_rate": 8.953333333333335e-06, | |
| "loss": 0.0111, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 1.6474464579901154, | |
| "grad_norm": 2.2675178050994873, | |
| "learning_rate": 8.897777777777779e-06, | |
| "loss": 0.0124, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.6474464579901154, | |
| "eval_loss": 0.02125450409948826, | |
| "eval_runtime": 1730.2385, | |
| "eval_samples_per_second": 1.403, | |
| "eval_steps_per_second": 0.176, | |
| "eval_wer": 5.243532975996271, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.6886326194398682, | |
| "grad_norm": 1.1241651773452759, | |
| "learning_rate": 8.842222222222223e-06, | |
| "loss": 0.0108, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 1.729818780889621, | |
| "grad_norm": 0.9443243145942688, | |
| "learning_rate": 8.786666666666668e-06, | |
| "loss": 0.01, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.771004942339374, | |
| "grad_norm": 1.7507250308990479, | |
| "learning_rate": 8.73111111111111e-06, | |
| "loss": 0.0108, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 1.812191103789127, | |
| "grad_norm": 0.8334024548530579, | |
| "learning_rate": 8.675555555555556e-06, | |
| "loss": 0.0108, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.8533772652388798, | |
| "grad_norm": 2.0746381282806396, | |
| "learning_rate": 8.62e-06, | |
| "loss": 0.009, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 1.8945634266886326, | |
| "grad_norm": 1.5471493005752563, | |
| "learning_rate": 8.564444444444445e-06, | |
| "loss": 0.0124, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.9357495881383855, | |
| "grad_norm": 1.4362578392028809, | |
| "learning_rate": 8.50888888888889e-06, | |
| "loss": 0.009, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 1.9769357495881383, | |
| "grad_norm": 1.5219734907150269, | |
| "learning_rate": 8.453333333333334e-06, | |
| "loss": 0.0108, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.018121911037891, | |
| "grad_norm": 0.7635548710823059, | |
| "learning_rate": 8.397777777777778e-06, | |
| "loss": 0.0069, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 2.059308072487644, | |
| "grad_norm": 0.622107982635498, | |
| "learning_rate": 8.342222222222222e-06, | |
| "loss": 0.0045, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.100494233937397, | |
| "grad_norm": 3.157203197479248, | |
| "learning_rate": 8.286666666666668e-06, | |
| "loss": 0.0055, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 2.1416803953871497, | |
| "grad_norm": 0.8712319731712341, | |
| "learning_rate": 8.231111111111112e-06, | |
| "loss": 0.0025, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.182866556836903, | |
| "grad_norm": 0.7438367605209351, | |
| "learning_rate": 8.175555555555556e-06, | |
| "loss": 0.0038, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 2.224052718286656, | |
| "grad_norm": 0.4088328778743744, | |
| "learning_rate": 8.120000000000002e-06, | |
| "loss": 0.004, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.2652388797364087, | |
| "grad_norm": 0.34847044944763184, | |
| "learning_rate": 8.064444444444444e-06, | |
| "loss": 0.003, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 2.3064250411861615, | |
| "grad_norm": 0.08477868884801865, | |
| "learning_rate": 8.00888888888889e-06, | |
| "loss": 0.0024, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.3476112026359144, | |
| "grad_norm": 0.5055556297302246, | |
| "learning_rate": 7.953333333333334e-06, | |
| "loss": 0.004, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 2.3887973640856672, | |
| "grad_norm": 0.27648621797561646, | |
| "learning_rate": 7.897777777777778e-06, | |
| "loss": 0.0037, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 2.42998352553542, | |
| "grad_norm": 2.3948631286621094, | |
| "learning_rate": 7.842222222222224e-06, | |
| "loss": 0.0022, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 2.471169686985173, | |
| "grad_norm": 0.3471105694770813, | |
| "learning_rate": 7.786666666666666e-06, | |
| "loss": 0.0034, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.471169686985173, | |
| "eval_loss": 0.017010504379868507, | |
| "eval_runtime": 1743.3725, | |
| "eval_samples_per_second": 1.392, | |
| "eval_steps_per_second": 0.174, | |
| "eval_wer": 3.821952924726171, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.5123558484349258, | |
| "grad_norm": 0.14169706404209137, | |
| "learning_rate": 7.731111111111112e-06, | |
| "loss": 0.0043, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 2.5535420098846786, | |
| "grad_norm": 0.4921598732471466, | |
| "learning_rate": 7.675555555555556e-06, | |
| "loss": 0.0024, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 2.594728171334432, | |
| "grad_norm": 0.20753732323646545, | |
| "learning_rate": 7.620000000000001e-06, | |
| "loss": 0.004, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 2.6359143327841847, | |
| "grad_norm": 0.10961499810218811, | |
| "learning_rate": 7.564444444444446e-06, | |
| "loss": 0.004, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.6771004942339376, | |
| "grad_norm": 1.0841516256332397, | |
| "learning_rate": 7.50888888888889e-06, | |
| "loss": 0.0046, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 2.7182866556836904, | |
| "grad_norm": 0.12003188580274582, | |
| "learning_rate": 7.453333333333334e-06, | |
| "loss": 0.0023, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 2.7594728171334433, | |
| "grad_norm": 0.055499833077192307, | |
| "learning_rate": 7.3977777777777786e-06, | |
| "loss": 0.0035, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 2.800658978583196, | |
| "grad_norm": 2.9141526222229004, | |
| "learning_rate": 7.342222222222223e-06, | |
| "loss": 0.0023, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.841845140032949, | |
| "grad_norm": 0.025740938261151314, | |
| "learning_rate": 7.2866666666666675e-06, | |
| "loss": 0.0024, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 2.883031301482702, | |
| "grad_norm": 0.6345125436782837, | |
| "learning_rate": 7.231111111111112e-06, | |
| "loss": 0.004, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 2.9242174629324547, | |
| "grad_norm": 0.5243808627128601, | |
| "learning_rate": 7.1755555555555556e-06, | |
| "loss": 0.0024, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 2.9654036243822075, | |
| "grad_norm": 0.44930708408355713, | |
| "learning_rate": 7.1200000000000004e-06, | |
| "loss": 0.0025, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.0065897858319603, | |
| "grad_norm": 1.6802186965942383, | |
| "learning_rate": 7.0644444444444445e-06, | |
| "loss": 0.0044, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 3.047775947281713, | |
| "grad_norm": 0.2793377637863159, | |
| "learning_rate": 7.008888888888889e-06, | |
| "loss": 0.0015, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 3.088962108731466, | |
| "grad_norm": 0.07968737185001373, | |
| "learning_rate": 6.953333333333334e-06, | |
| "loss": 0.0015, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 3.130148270181219, | |
| "grad_norm": 0.04705784469842911, | |
| "learning_rate": 6.897777777777779e-06, | |
| "loss": 0.001, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.171334431630972, | |
| "grad_norm": 0.1313919574022293, | |
| "learning_rate": 6.842222222222222e-06, | |
| "loss": 0.0008, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 3.212520593080725, | |
| "grad_norm": 0.4405129551887512, | |
| "learning_rate": 6.786666666666667e-06, | |
| "loss": 0.0011, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 3.253706754530478, | |
| "grad_norm": 1.2785202264785767, | |
| "learning_rate": 6.731111111111111e-06, | |
| "loss": 0.0017, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 3.2948929159802307, | |
| "grad_norm": 0.04460221529006958, | |
| "learning_rate": 6.675555555555556e-06, | |
| "loss": 0.0008, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.2948929159802307, | |
| "eval_loss": 0.015330803580582142, | |
| "eval_runtime": 1719.9199, | |
| "eval_samples_per_second": 1.411, | |
| "eval_steps_per_second": 0.177, | |
| "eval_wer": 3.5733706206789404, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.3360790774299836, | |
| "grad_norm": 0.10665885359048843, | |
| "learning_rate": 6.620000000000001e-06, | |
| "loss": 0.0022, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 3.3772652388797364, | |
| "grad_norm": 4.365774631500244, | |
| "learning_rate": 6.564444444444446e-06, | |
| "loss": 0.0013, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 3.4184514003294892, | |
| "grad_norm": 0.2422148883342743, | |
| "learning_rate": 6.508888888888889e-06, | |
| "loss": 0.0011, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 3.459637561779242, | |
| "grad_norm": 0.14427553117275238, | |
| "learning_rate": 6.453333333333334e-06, | |
| "loss": 0.0009, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.500823723228995, | |
| "grad_norm": 0.0224087405949831, | |
| "learning_rate": 6.397777777777778e-06, | |
| "loss": 0.0005, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 3.5420098846787478, | |
| "grad_norm": 0.052274756133556366, | |
| "learning_rate": 6.342222222222223e-06, | |
| "loss": 0.0012, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 3.583196046128501, | |
| "grad_norm": 0.12983651459217072, | |
| "learning_rate": 6.286666666666668e-06, | |
| "loss": 0.0011, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 3.624382207578254, | |
| "grad_norm": 0.17136234045028687, | |
| "learning_rate": 6.231111111111111e-06, | |
| "loss": 0.0004, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.6655683690280068, | |
| "grad_norm": 0.02376146800816059, | |
| "learning_rate": 6.175555555555556e-06, | |
| "loss": 0.0006, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 3.7067545304777596, | |
| "grad_norm": 0.020898204296827316, | |
| "learning_rate": 6.120000000000001e-06, | |
| "loss": 0.0004, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 3.7479406919275124, | |
| "grad_norm": 0.011864184401929379, | |
| "learning_rate": 6.064444444444445e-06, | |
| "loss": 0.0014, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 3.7891268533772653, | |
| "grad_norm": 0.07441503554582596, | |
| "learning_rate": 6.00888888888889e-06, | |
| "loss": 0.0005, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.830313014827018, | |
| "grad_norm": 0.09472518414258957, | |
| "learning_rate": 5.9533333333333345e-06, | |
| "loss": 0.0008, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 3.871499176276771, | |
| "grad_norm": 0.5632221698760986, | |
| "learning_rate": 5.897777777777778e-06, | |
| "loss": 0.001, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 3.912685337726524, | |
| "grad_norm": 0.02372151054441929, | |
| "learning_rate": 5.8422222222222226e-06, | |
| "loss": 0.0008, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 3.9538714991762767, | |
| "grad_norm": 0.324942022562027, | |
| "learning_rate": 5.7866666666666674e-06, | |
| "loss": 0.0004, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.9950576606260295, | |
| "grad_norm": 0.022629514336586, | |
| "learning_rate": 5.7311111111111115e-06, | |
| "loss": 0.0003, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 4.036243822075782, | |
| "grad_norm": 0.012355553917586803, | |
| "learning_rate": 5.675555555555556e-06, | |
| "loss": 0.0005, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 4.077429983525535, | |
| "grad_norm": 0.021213717758655548, | |
| "learning_rate": 5.620000000000001e-06, | |
| "loss": 0.0005, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 4.118616144975288, | |
| "grad_norm": 0.01894320733845234, | |
| "learning_rate": 5.5644444444444444e-06, | |
| "loss": 0.0003, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.118616144975288, | |
| "eval_loss": 0.013398992829024792, | |
| "eval_runtime": 1739.4064, | |
| "eval_samples_per_second": 1.395, | |
| "eval_steps_per_second": 0.175, | |
| "eval_wer": 2.9130738755534837, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.159802306425041, | |
| "grad_norm": 0.01304696872830391, | |
| "learning_rate": 5.508888888888889e-06, | |
| "loss": 0.0005, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 4.200988467874794, | |
| "grad_norm": 0.020342445001006126, | |
| "learning_rate": 5.453333333333334e-06, | |
| "loss": 0.0005, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 4.242174629324547, | |
| "grad_norm": 0.02242002822458744, | |
| "learning_rate": 5.397777777777778e-06, | |
| "loss": 0.0007, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 4.283360790774299, | |
| "grad_norm": 0.010535928420722485, | |
| "learning_rate": 5.342222222222223e-06, | |
| "loss": 0.0005, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 4.324546952224052, | |
| "grad_norm": 0.3926488161087036, | |
| "learning_rate": 5.286666666666666e-06, | |
| "loss": 0.0008, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 4.365733113673806, | |
| "grad_norm": 0.016406048089265823, | |
| "learning_rate": 5.231111111111111e-06, | |
| "loss": 0.0002, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 4.406919275123559, | |
| "grad_norm": 0.03407042473554611, | |
| "learning_rate": 5.175555555555556e-06, | |
| "loss": 0.0002, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 4.448105436573312, | |
| "grad_norm": 0.019738828763365746, | |
| "learning_rate": 5.12e-06, | |
| "loss": 0.0003, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 4.4892915980230645, | |
| "grad_norm": 0.03806024044752121, | |
| "learning_rate": 5.064444444444445e-06, | |
| "loss": 0.0005, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 4.530477759472817, | |
| "grad_norm": 0.008825665339827538, | |
| "learning_rate": 5.00888888888889e-06, | |
| "loss": 0.0005, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 4.57166392092257, | |
| "grad_norm": 0.1027652770280838, | |
| "learning_rate": 4.953333333333334e-06, | |
| "loss": 0.0003, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 4.612850082372323, | |
| "grad_norm": 0.014739478006958961, | |
| "learning_rate": 4.897777777777778e-06, | |
| "loss": 0.0002, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 4.654036243822076, | |
| "grad_norm": 0.02095695212483406, | |
| "learning_rate": 4.842222222222223e-06, | |
| "loss": 0.0008, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 4.695222405271829, | |
| "grad_norm": 0.017346713691949844, | |
| "learning_rate": 4.786666666666667e-06, | |
| "loss": 0.0003, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 4.736408566721582, | |
| "grad_norm": 0.022175878286361694, | |
| "learning_rate": 4.731111111111112e-06, | |
| "loss": 0.0002, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 4.7775947281713345, | |
| "grad_norm": 0.008032923564314842, | |
| "learning_rate": 4.675555555555556e-06, | |
| "loss": 0.0003, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 4.818780889621087, | |
| "grad_norm": 0.006342317909002304, | |
| "learning_rate": 4.620000000000001e-06, | |
| "loss": 0.0007, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 4.85996705107084, | |
| "grad_norm": 0.022994259372353554, | |
| "learning_rate": 4.564444444444445e-06, | |
| "loss": 0.0005, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 4.901153212520593, | |
| "grad_norm": 0.07509420812129974, | |
| "learning_rate": 4.50888888888889e-06, | |
| "loss": 0.0005, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 4.942339373970346, | |
| "grad_norm": 0.17356319725513458, | |
| "learning_rate": 4.453333333333334e-06, | |
| "loss": 0.0003, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.942339373970346, | |
| "eval_loss": 0.012571546249091625, | |
| "eval_runtime": 1705.0805, | |
| "eval_samples_per_second": 1.423, | |
| "eval_steps_per_second": 0.178, | |
| "eval_wer": 2.4236774644604986, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.983525535420099, | |
| "grad_norm": 0.03659577667713165, | |
| "learning_rate": 5.4163636363636365e-06, | |
| "loss": 0.001, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 5.0247116968698515, | |
| "grad_norm": 0.08530838042497635, | |
| "learning_rate": 5.370909090909091e-06, | |
| "loss": 0.0004, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 5.065897858319604, | |
| "grad_norm": 0.013511723838746548, | |
| "learning_rate": 5.325454545454546e-06, | |
| "loss": 0.0005, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 5.107084019769357, | |
| "grad_norm": 0.11328202486038208, | |
| "learning_rate": 5.28e-06, | |
| "loss": 0.0003, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 5.14827018121911, | |
| "grad_norm": 0.01356989424675703, | |
| "learning_rate": 5.234545454545456e-06, | |
| "loss": 0.0001, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 5.189456342668863, | |
| "grad_norm": 0.011790513060986996, | |
| "learning_rate": 5.1890909090909095e-06, | |
| "loss": 0.0002, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 5.230642504118616, | |
| "grad_norm": 0.022032860666513443, | |
| "learning_rate": 5.143636363636364e-06, | |
| "loss": 0.0008, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 5.2718286655683695, | |
| "grad_norm": 0.004786133300513029, | |
| "learning_rate": 5.1e-06, | |
| "loss": 0.0007, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 5.313014827018122, | |
| "grad_norm": 0.033955205231904984, | |
| "learning_rate": 5.054545454545455e-06, | |
| "loss": 0.0014, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 5.354200988467875, | |
| "grad_norm": 0.019066009670495987, | |
| "learning_rate": 5.009090909090909e-06, | |
| "loss": 0.0014, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 5.395387149917628, | |
| "grad_norm": 0.3937516212463379, | |
| "learning_rate": 4.963636363636364e-06, | |
| "loss": 0.0004, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 5.436573311367381, | |
| "grad_norm": 0.022780615836381912, | |
| "learning_rate": 4.918181818181819e-06, | |
| "loss": 0.0004, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 5.477759472817134, | |
| "grad_norm": 0.0338226743042469, | |
| "learning_rate": 4.872727272727273e-06, | |
| "loss": 0.0002, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 5.518945634266887, | |
| "grad_norm": 0.012667668052017689, | |
| "learning_rate": 4.827272727272728e-06, | |
| "loss": 0.0003, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 5.560131795716639, | |
| "grad_norm": 0.02727661095559597, | |
| "learning_rate": 4.7818181818181825e-06, | |
| "loss": 0.0002, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 5.601317957166392, | |
| "grad_norm": 0.06680746376514435, | |
| "learning_rate": 4.736363636363637e-06, | |
| "loss": 0.0002, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 5.642504118616145, | |
| "grad_norm": 0.04453982040286064, | |
| "learning_rate": 4.690909090909092e-06, | |
| "loss": 0.0001, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 5.683690280065898, | |
| "grad_norm": 0.01748238503932953, | |
| "learning_rate": 4.645454545454545e-06, | |
| "loss": 0.0002, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 5.724876441515651, | |
| "grad_norm": 0.029951492324471474, | |
| "learning_rate": 4.600000000000001e-06, | |
| "loss": 0.0003, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 5.766062602965404, | |
| "grad_norm": 0.012844923883676529, | |
| "learning_rate": 4.554545454545455e-06, | |
| "loss": 0.0002, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.766062602965404, | |
| "eval_loss": 0.013792040757834911, | |
| "eval_runtime": 1631.8767, | |
| "eval_samples_per_second": 1.487, | |
| "eval_steps_per_second": 0.186, | |
| "eval_wer": 2.6800279655092054, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.8072487644151565, | |
| "grad_norm": 0.24390824139118195, | |
| "learning_rate": 4.50909090909091e-06, | |
| "loss": 0.0003, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 5.848434925864909, | |
| "grad_norm": 0.032755326479673386, | |
| "learning_rate": 4.463636363636364e-06, | |
| "loss": 0.0007, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 5.889621087314662, | |
| "grad_norm": 0.04037921503186226, | |
| "learning_rate": 4.418181818181818e-06, | |
| "loss": 0.0002, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 5.930807248764415, | |
| "grad_norm": 0.02575298771262169, | |
| "learning_rate": 4.372727272727273e-06, | |
| "loss": 0.0014, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 5.971993410214168, | |
| "grad_norm": 0.011575842276215553, | |
| "learning_rate": 4.327272727272728e-06, | |
| "loss": 0.0002, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 6.013179571663921, | |
| "grad_norm": 0.005709837190806866, | |
| "learning_rate": 4.281818181818182e-06, | |
| "loss": 0.0013, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 6.0543657331136735, | |
| "grad_norm": 0.06958606094121933, | |
| "learning_rate": 4.236363636363637e-06, | |
| "loss": 0.0001, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 6.095551894563426, | |
| "grad_norm": 0.00584538234397769, | |
| "learning_rate": 4.190909090909091e-06, | |
| "loss": 0.0002, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 6.136738056013179, | |
| "grad_norm": 0.004265847150236368, | |
| "learning_rate": 4.145454545454546e-06, | |
| "loss": 0.0001, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 6.177924217462932, | |
| "grad_norm": 0.01833544485270977, | |
| "learning_rate": 4.1e-06, | |
| "loss": 0.0002, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 6.219110378912685, | |
| "grad_norm": 0.005921401083469391, | |
| "learning_rate": 4.054545454545455e-06, | |
| "loss": 0.0007, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 6.260296540362438, | |
| "grad_norm": 0.007460064720362425, | |
| "learning_rate": 4.009090909090909e-06, | |
| "loss": 0.0001, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 6.3014827018121915, | |
| "grad_norm": 0.005449911579489708, | |
| "learning_rate": 3.963636363636364e-06, | |
| "loss": 0.0004, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 6.342668863261944, | |
| "grad_norm": 0.008822810836136341, | |
| "learning_rate": 3.918181818181819e-06, | |
| "loss": 0.0001, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 6.383855024711697, | |
| "grad_norm": 0.0073766824789345264, | |
| "learning_rate": 3.872727272727273e-06, | |
| "loss": 0.0001, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 6.42504118616145, | |
| "grad_norm": 0.009812358766794205, | |
| "learning_rate": 3.827272727272728e-06, | |
| "loss": 0.0001, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 6.466227347611203, | |
| "grad_norm": 0.010365708731114864, | |
| "learning_rate": 3.781818181818182e-06, | |
| "loss": 0.0001, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 6.507413509060956, | |
| "grad_norm": 0.005115933250635862, | |
| "learning_rate": 3.736363636363637e-06, | |
| "loss": 0.0001, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 6.548599670510709, | |
| "grad_norm": 0.0084151616320014, | |
| "learning_rate": 3.690909090909091e-06, | |
| "loss": 0.0001, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 6.589785831960461, | |
| "grad_norm": 0.004395967349410057, | |
| "learning_rate": 3.645454545454546e-06, | |
| "loss": 0.0001, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.589785831960461, | |
| "eval_loss": 0.012646839953958988, | |
| "eval_runtime": 1618.9324, | |
| "eval_samples_per_second": 1.499, | |
| "eval_steps_per_second": 0.188, | |
| "eval_wer": 2.547968616484114, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.630971993410214, | |
| "grad_norm": 0.00942137185484171, | |
| "learning_rate": 3.6000000000000003e-06, | |
| "loss": 0.0001, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 6.672158154859967, | |
| "grad_norm": 0.006176659371703863, | |
| "learning_rate": 3.554545454545455e-06, | |
| "loss": 0.0001, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 6.71334431630972, | |
| "grad_norm": 0.008395605720579624, | |
| "learning_rate": 3.509090909090909e-06, | |
| "loss": 0.0002, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 6.754530477759473, | |
| "grad_norm": 0.0078582176938653, | |
| "learning_rate": 3.463636363636364e-06, | |
| "loss": 0.0001, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 6.795716639209226, | |
| "grad_norm": 0.01093814056366682, | |
| "learning_rate": 3.4181818181818182e-06, | |
| "loss": 0.0001, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 6.8369028006589785, | |
| "grad_norm": 0.016128424555063248, | |
| "learning_rate": 3.3727272727272732e-06, | |
| "loss": 0.0003, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 6.878088962108731, | |
| "grad_norm": 0.006562087684869766, | |
| "learning_rate": 3.3272727272727274e-06, | |
| "loss": 0.0001, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 6.919275123558484, | |
| "grad_norm": 0.0189774502068758, | |
| "learning_rate": 3.281818181818182e-06, | |
| "loss": 0.0001, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 6.960461285008237, | |
| "grad_norm": 0.7424684762954712, | |
| "learning_rate": 3.236363636363636e-06, | |
| "loss": 0.0001, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 7.00164744645799, | |
| "grad_norm": 0.007785412948578596, | |
| "learning_rate": 3.190909090909091e-06, | |
| "loss": 0.0003, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 7.042833607907743, | |
| "grad_norm": 0.015975123271346092, | |
| "learning_rate": 3.145454545454546e-06, | |
| "loss": 0.0001, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 7.0840197693574956, | |
| "grad_norm": 0.004259432666003704, | |
| "learning_rate": 3.1000000000000004e-06, | |
| "loss": 0.0001, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 7.125205930807248, | |
| "grad_norm": 0.0031003232579678297, | |
| "learning_rate": 3.054545454545455e-06, | |
| "loss": 0.0002, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 7.166392092257001, | |
| "grad_norm": 0.0041422671638429165, | |
| "learning_rate": 3.009090909090909e-06, | |
| "loss": 0.0001, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 7.207578253706755, | |
| "grad_norm": 0.002572175581008196, | |
| "learning_rate": 2.963636363636364e-06, | |
| "loss": 0.0001, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 7.248764415156508, | |
| "grad_norm": 0.009961551055312157, | |
| "learning_rate": 2.9181818181818183e-06, | |
| "loss": 0.0001, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 7.289950576606261, | |
| "grad_norm": 0.011908295564353466, | |
| "learning_rate": 2.872727272727273e-06, | |
| "loss": 0.0001, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 7.3311367380560135, | |
| "grad_norm": 0.009862988255918026, | |
| "learning_rate": 2.8272727272727275e-06, | |
| "loss": 0.0001, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 7.372322899505766, | |
| "grad_norm": 0.007197460159659386, | |
| "learning_rate": 2.781818181818182e-06, | |
| "loss": 0.0001, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 7.413509060955519, | |
| "grad_norm": 0.02329566702246666, | |
| "learning_rate": 2.7363636363636363e-06, | |
| "loss": 0.0001, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 7.413509060955519, | |
| "eval_loss": 0.012732533738017082, | |
| "eval_runtime": 1644.4784, | |
| "eval_samples_per_second": 1.476, | |
| "eval_steps_per_second": 0.185, | |
| "eval_wer": 2.6722597685077294, | |
| "step": 4500 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 6000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 2.077006843183104e+19, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |