| { | |
| "best_metric": 0.3040466010570526, | |
| "best_model_checkpoint": "./enko_xlsr_13p_run1/checkpoint-77020", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 77020, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0003, | |
| "loss": 4.7797, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00029803972817564033, | |
| "loss": 1.2377, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0002960794563512807, | |
| "loss": 0.9452, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.000294119184526921, | |
| "loss": 0.8622, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0002921589127025614, | |
| "loss": 0.82, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0002901986408782018, | |
| "loss": 0.7742, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00028823836905384213, | |
| "loss": 0.75, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.0002862780972294825, | |
| "loss": 0.7459, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00028431782540512285, | |
| "loss": 0.7282, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00028235755358076316, | |
| "loss": 0.7104, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0002803972817564035, | |
| "loss": 0.692, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0002784370099320439, | |
| "loss": 0.6954, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.00027647673810768424, | |
| "loss": 0.6844, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.0002745164662833246, | |
| "loss": 0.6779, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.00027255619445896496, | |
| "loss": 0.6595, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.44951537251472473, | |
| "eval_runtime": 11.7967, | |
| "eval_samples_per_second": 60.271, | |
| "eval_steps_per_second": 7.544, | |
| "eval_wer": 0.297432521395655, | |
| "step": 7702 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 0.0002705959226346053, | |
| "loss": 0.6426, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 0.0002686356508102457, | |
| "loss": 0.6215, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.00026667537898588604, | |
| "loss": 0.6209, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.0002647151071615264, | |
| "loss": 0.6134, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 0.00026275483533716676, | |
| "loss": 0.6127, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 0.00026079456351280707, | |
| "loss": 0.5987, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.0002588342916884474, | |
| "loss": 0.6028, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 0.0002568740198640878, | |
| "loss": 0.5893, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.00025491374803972815, | |
| "loss": 0.59, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 0.0002529534762153685, | |
| "loss": 0.594, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00025099320439100887, | |
| "loss": 0.5903, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.0002490329325666492, | |
| "loss": 0.5779, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.0002470726607422896, | |
| "loss": 0.582, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 0.00024511238891792995, | |
| "loss": 0.5781, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 0.0002431521170935703, | |
| "loss": 0.5717, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.39816993474960327, | |
| "eval_runtime": 11.9909, | |
| "eval_samples_per_second": 59.295, | |
| "eval_steps_per_second": 7.422, | |
| "eval_wer": 0.25622119815668204, | |
| "step": 15404 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 0.00024119184526921064, | |
| "loss": 0.5632, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 0.000239231573444851, | |
| "loss": 0.5298, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 0.00023727130162049136, | |
| "loss": 0.5391, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 0.00023531102979613172, | |
| "loss": 0.5274, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 0.00023335075797177208, | |
| "loss": 0.5368, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 0.00023139048614741244, | |
| "loss": 0.5191, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.00022943021432305277, | |
| "loss": 0.523, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 0.00022746994249869313, | |
| "loss": 0.529, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 0.0002255096706743335, | |
| "loss": 0.5293, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 0.00022354939884997385, | |
| "loss": 0.5091, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 0.0002215891270256142, | |
| "loss": 0.5258, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 0.00021962885520125454, | |
| "loss": 0.5107, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 0.0002176685833768949, | |
| "loss": 0.5197, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 0.00021570831155253526, | |
| "loss": 0.5162, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 0.00021374803972817562, | |
| "loss": 0.5162, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 0.00021178776790381598, | |
| "loss": 0.5134, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.3769395649433136, | |
| "eval_runtime": 12.4234, | |
| "eval_samples_per_second": 57.231, | |
| "eval_steps_per_second": 7.164, | |
| "eval_wer": 0.23647136273864383, | |
| "step": 23106 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 0.00020982749607945634, | |
| "loss": 0.4797, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 0.00020786722425509668, | |
| "loss": 0.4777, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 0.00020590695243073704, | |
| "loss": 0.4729, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 0.0002039466806063774, | |
| "loss": 0.4749, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 0.00020198640878201776, | |
| "loss": 0.4803, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 0.00020002613695765812, | |
| "loss": 0.4738, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 0.00019806586513329845, | |
| "loss": 0.4782, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 0.0001961055933089388, | |
| "loss": 0.4994, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 0.00019414532148457917, | |
| "loss": 0.4676, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 0.00019218504966021953, | |
| "loss": 0.457, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 0.0001902247778358599, | |
| "loss": 0.469, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 0.00018826450601150025, | |
| "loss": 0.4573, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 0.00018630423418714058, | |
| "loss": 0.4672, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 0.00018434396236278094, | |
| "loss": 0.4736, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 0.0001823836905384213, | |
| "loss": 0.467, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.34988921880722046, | |
| "eval_runtime": 12.3506, | |
| "eval_samples_per_second": 57.568, | |
| "eval_steps_per_second": 7.206, | |
| "eval_wer": 0.22027649769585253, | |
| "step": 30808 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 0.00018042341871406166, | |
| "loss": 0.4541, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 0.00017846314688970202, | |
| "loss": 0.4273, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 0.00017650287506534236, | |
| "loss": 0.4334, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 0.00017454260324098272, | |
| "loss": 0.4203, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 0.00017258233141662308, | |
| "loss": 0.4331, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 0.00017062205959226344, | |
| "loss": 0.4306, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 0.00016866178776790382, | |
| "loss": 0.4251, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 0.00016670151594354418, | |
| "loss": 0.4231, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "learning_rate": 0.0001647412441191845, | |
| "loss": 0.4256, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "learning_rate": 0.00016278097229482488, | |
| "loss": 0.4332, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 0.00016082070047046524, | |
| "loss": 0.4238, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 0.0001588604286461056, | |
| "loss": 0.4251, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 0.00015690015682174596, | |
| "loss": 0.425, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 0.0001549398849973863, | |
| "loss": 0.4208, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 0.00015297961317302665, | |
| "loss": 0.4245, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.000151019341348667, | |
| "loss": 0.4156, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.3391418159008026, | |
| "eval_runtime": 12.5768, | |
| "eval_samples_per_second": 56.533, | |
| "eval_steps_per_second": 7.077, | |
| "eval_wer": 0.21158657011191573, | |
| "step": 38510 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 0.00014905906952430737, | |
| "loss": 0.3845, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 5.13, | |
| "learning_rate": 0.0001470987976999477, | |
| "loss": 0.393, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "learning_rate": 0.00014513852587558806, | |
| "loss": 0.3938, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "learning_rate": 0.00014317825405122842, | |
| "loss": 0.3902, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "learning_rate": 0.00014121798222686878, | |
| "loss": 0.3817, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 5.39, | |
| "learning_rate": 0.00013925771040250914, | |
| "loss": 0.3937, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 0.0001372974385781495, | |
| "loss": 0.3821, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "learning_rate": 0.00013533716675378984, | |
| "loss": 0.3903, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 5.58, | |
| "learning_rate": 0.0001333768949294302, | |
| "loss": 0.381, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 0.00013141662310507056, | |
| "loss": 0.3866, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "learning_rate": 0.00012945635128071092, | |
| "loss": 0.3816, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "learning_rate": 0.00012749607945635128, | |
| "loss": 0.3845, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "learning_rate": 0.0001255358076319916, | |
| "loss": 0.3777, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 5.91, | |
| "learning_rate": 0.00012357553580763197, | |
| "loss": 0.3777, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "learning_rate": 0.00012161526398327234, | |
| "loss": 0.379, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.3327275514602661, | |
| "eval_runtime": 12.31, | |
| "eval_samples_per_second": 57.758, | |
| "eval_steps_per_second": 7.23, | |
| "eval_wer": 0.1998683344305464, | |
| "step": 46212 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 0.00011965499215891269, | |
| "loss": 0.3657, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "learning_rate": 0.00011769472033455305, | |
| "loss": 0.3527, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "learning_rate": 0.00011573444851019341, | |
| "loss": 0.3505, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 6.23, | |
| "learning_rate": 0.00011377417668583375, | |
| "loss": 0.3472, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 6.3, | |
| "learning_rate": 0.00011181390486147411, | |
| "loss": 0.3497, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 6.36, | |
| "learning_rate": 0.00010985363303711446, | |
| "loss": 0.349, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "learning_rate": 0.00010789336121275482, | |
| "loss": 0.3545, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "learning_rate": 0.00010593308938839518, | |
| "loss": 0.3453, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 0.00010397281756403553, | |
| "loss": 0.3501, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 0.00010201254573967589, | |
| "loss": 0.3473, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 6.69, | |
| "learning_rate": 0.00010005227391531626, | |
| "loss": 0.3462, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "learning_rate": 9.80920020909566e-05, | |
| "loss": 0.3385, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 6.82, | |
| "learning_rate": 9.613173026659697e-05, | |
| "loss": 0.3459, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "learning_rate": 9.417145844223731e-05, | |
| "loss": 0.3441, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 6.95, | |
| "learning_rate": 9.221118661787767e-05, | |
| "loss": 0.3475, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.31270119547843933, | |
| "eval_runtime": 12.4561, | |
| "eval_samples_per_second": 57.08, | |
| "eval_steps_per_second": 7.145, | |
| "eval_wer": 0.1947333772218565, | |
| "step": 53914 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 9.025091479351803e-05, | |
| "loss": 0.3379, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "learning_rate": 8.829064296915838e-05, | |
| "loss": 0.3159, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "learning_rate": 8.633037114479874e-05, | |
| "loss": 0.3254, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 7.21, | |
| "learning_rate": 8.43700993204391e-05, | |
| "loss": 0.3195, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "learning_rate": 8.240982749607945e-05, | |
| "loss": 0.3225, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 7.34, | |
| "learning_rate": 8.044955567171981e-05, | |
| "loss": 0.3128, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 7.4, | |
| "learning_rate": 7.848928384736017e-05, | |
| "loss": 0.318, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 7.47, | |
| "learning_rate": 7.652901202300051e-05, | |
| "loss": 0.3165, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 7.53, | |
| "learning_rate": 7.456874019864087e-05, | |
| "loss": 0.3191, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "learning_rate": 7.260846837428123e-05, | |
| "loss": 0.3169, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 7.66, | |
| "learning_rate": 7.064819654992158e-05, | |
| "loss": 0.3118, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 7.73, | |
| "learning_rate": 6.868792472556194e-05, | |
| "loss": 0.3192, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "learning_rate": 6.67276529012023e-05, | |
| "loss": 0.3055, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 7.86, | |
| "learning_rate": 6.476738107684265e-05, | |
| "loss": 0.3134, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 7.92, | |
| "learning_rate": 6.2807109252483e-05, | |
| "loss": 0.3081, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 7.98, | |
| "learning_rate": 6.084683742812336e-05, | |
| "loss": 0.3105, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.30813026428222656, | |
| "eval_runtime": 12.2002, | |
| "eval_samples_per_second": 58.278, | |
| "eval_steps_per_second": 7.295, | |
| "eval_wer": 0.1814351547070441, | |
| "step": 61616 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "learning_rate": 5.888656560376372e-05, | |
| "loss": 0.2986, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 8.11, | |
| "learning_rate": 5.692629377940407e-05, | |
| "loss": 0.2908, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 8.18, | |
| "learning_rate": 5.4966021955044426e-05, | |
| "loss": 0.2835, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 8.24, | |
| "learning_rate": 5.3005750130684786e-05, | |
| "loss": 0.2904, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 8.31, | |
| "learning_rate": 5.104547830632514e-05, | |
| "loss": 0.289, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 8.37, | |
| "learning_rate": 4.908520648196549e-05, | |
| "loss": 0.2905, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 8.44, | |
| "learning_rate": 4.7124934657605846e-05, | |
| "loss": 0.288, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "learning_rate": 4.516466283324621e-05, | |
| "loss": 0.2813, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 8.57, | |
| "learning_rate": 4.3204391008886566e-05, | |
| "loss": 0.2831, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 8.63, | |
| "learning_rate": 4.124411918452692e-05, | |
| "loss": 0.2914, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "learning_rate": 3.928384736016727e-05, | |
| "loss": 0.2807, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 8.76, | |
| "learning_rate": 3.7323575535807626e-05, | |
| "loss": 0.2833, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 8.83, | |
| "learning_rate": 3.5363303711447986e-05, | |
| "loss": 0.2825, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "learning_rate": 3.340303188708834e-05, | |
| "loss": 0.2824, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "learning_rate": 3.14427600627287e-05, | |
| "loss": 0.281, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.3067900538444519, | |
| "eval_runtime": 12.6252, | |
| "eval_samples_per_second": 56.316, | |
| "eval_steps_per_second": 7.049, | |
| "eval_wer": 0.17419354838709677, | |
| "step": 69318 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 2.948248823836905e-05, | |
| "loss": 0.2725, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 2.7522216414009406e-05, | |
| "loss": 0.2696, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 9.15, | |
| "learning_rate": 2.5561944589649766e-05, | |
| "loss": 0.2713, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 9.22, | |
| "learning_rate": 2.360167276529012e-05, | |
| "loss": 0.2634, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 9.28, | |
| "learning_rate": 2.1641400940930475e-05, | |
| "loss": 0.2635, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 9.35, | |
| "learning_rate": 1.968112911657083e-05, | |
| "loss": 0.2658, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "learning_rate": 1.7720857292211185e-05, | |
| "loss": 0.2648, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 9.48, | |
| "learning_rate": 1.576058546785154e-05, | |
| "loss": 0.273, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 9.54, | |
| "learning_rate": 1.3800313643491895e-05, | |
| "loss": 0.2601, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 9.61, | |
| "learning_rate": 1.1840041819132252e-05, | |
| "loss": 0.2598, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 9.67, | |
| "learning_rate": 9.879769994772608e-06, | |
| "loss": 0.2698, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 9.74, | |
| "learning_rate": 7.919498170412963e-06, | |
| "loss": 0.2574, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 9.8, | |
| "learning_rate": 5.959226346053318e-06, | |
| "loss": 0.2646, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 9.87, | |
| "learning_rate": 3.998954521693675e-06, | |
| "loss": 0.2584, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 9.93, | |
| "learning_rate": 2.0386826973340303e-06, | |
| "loss": 0.2623, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 7.841087297438577e-08, | |
| "loss": 0.2584, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.3040466010570526, | |
| "eval_runtime": 12.4236, | |
| "eval_samples_per_second": 57.23, | |
| "eval_steps_per_second": 7.164, | |
| "eval_wer": 0.17129690585911783, | |
| "step": 77020 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 77020, | |
| "total_flos": 1.409382714117005e+20, | |
| "train_loss": 0.4655882824058503, | |
| "train_runtime": 41869.3879, | |
| "train_samples_per_second": 29.431, | |
| "train_steps_per_second": 1.84 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 77020, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 1.409382714117005e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |