| { | |
| "best_global_step": 28000, | |
| "best_metric": 0.18110816386678455, | |
| "best_model_checkpoint": "./distil-whisper/checkpoint-28000", | |
| "epoch": 30.0, | |
| "eval_steps": 1000, | |
| "global_step": 51210, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05858230814294083, | |
| "grad_norm": 27.49039077758789, | |
| "learning_rate": 1.9e-05, | |
| "loss": 1.7236, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.11716461628588166, | |
| "grad_norm": 18.32018280029297, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 1.796, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.1757469244288225, | |
| "grad_norm": 12.191054344177246, | |
| "learning_rate": 5.9e-05, | |
| "loss": 1.8559, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.23432923257176333, | |
| "grad_norm": 20.974699020385742, | |
| "learning_rate": 7.900000000000001e-05, | |
| "loss": 1.9804, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29291154071470415, | |
| "grad_norm": 37.665748596191406, | |
| "learning_rate": 9.88e-05, | |
| "loss": 2.2308, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.351493848857645, | |
| "grad_norm": 22.89726448059082, | |
| "learning_rate": 9.917896759542318e-05, | |
| "loss": 2.0073, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.4100761570005858, | |
| "grad_norm": 28.912738800048828, | |
| "learning_rate": 9.830552886714997e-05, | |
| "loss": 2.2137, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.46865846514352666, | |
| "grad_norm": 26.01645851135254, | |
| "learning_rate": 9.743209013887676e-05, | |
| "loss": 2.3668, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.5272407732864675, | |
| "grad_norm": 38.67082214355469, | |
| "learning_rate": 9.655865141060355e-05, | |
| "loss": 2.1304, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.5858230814294083, | |
| "grad_norm": 42.85211181640625, | |
| "learning_rate": 9.568521268233034e-05, | |
| "loss": 2.1291, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.5858230814294083, | |
| "eval_loss": 0.09121495485305786, | |
| "eval_runtime": 150.358, | |
| "eval_samples_per_second": 3.325, | |
| "eval_steps_per_second": 0.419, | |
| "eval_wer": 0.19776009431181846, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6444053895723492, | |
| "grad_norm": 41.63334274291992, | |
| "learning_rate": 9.481177395405713e-05, | |
| "loss": 2.176, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.70298769771529, | |
| "grad_norm": 35.67658996582031, | |
| "learning_rate": 9.393833522578391e-05, | |
| "loss": 2.3521, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.7615700058582309, | |
| "grad_norm": 23.11754608154297, | |
| "learning_rate": 9.30648964975107e-05, | |
| "loss": 2.0235, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.8201523140011716, | |
| "grad_norm": 57.13395690917969, | |
| "learning_rate": 9.219145776923749e-05, | |
| "loss": 2.0512, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.8787346221441125, | |
| "grad_norm": 29.889575958251953, | |
| "learning_rate": 9.131801904096428e-05, | |
| "loss": 2.2161, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.9373169302870533, | |
| "grad_norm": 20.80169105529785, | |
| "learning_rate": 9.044458031269108e-05, | |
| "loss": 2.4303, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.9958992384299942, | |
| "grad_norm": 43.849361419677734, | |
| "learning_rate": 8.957114158441786e-05, | |
| "loss": 2.162, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.054481546572935, | |
| "grad_norm": 19.961891174316406, | |
| "learning_rate": 8.869770285614464e-05, | |
| "loss": 1.8381, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.1130638547158758, | |
| "grad_norm": 25.00478744506836, | |
| "learning_rate": 8.782426412787143e-05, | |
| "loss": 1.8102, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.1716461628588166, | |
| "grad_norm": 23.878835678100586, | |
| "learning_rate": 8.695082539959822e-05, | |
| "loss": 1.7057, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.1716461628588166, | |
| "eval_loss": 0.09117516130208969, | |
| "eval_runtime": 145.9897, | |
| "eval_samples_per_second": 3.425, | |
| "eval_steps_per_second": 0.432, | |
| "eval_wer": 0.2002652519893899, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.2302284710017575, | |
| "grad_norm": 36.49687957763672, | |
| "learning_rate": 8.607738667132501e-05, | |
| "loss": 1.7356, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.2888107791446983, | |
| "grad_norm": 23.07692527770996, | |
| "learning_rate": 8.52039479430518e-05, | |
| "loss": 1.6766, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.3473930872876392, | |
| "grad_norm": 23.606229782104492, | |
| "learning_rate": 8.433050921477858e-05, | |
| "loss": 1.7865, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.40597539543058, | |
| "grad_norm": 12.508922576904297, | |
| "learning_rate": 8.345707048650537e-05, | |
| "loss": 1.7503, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.4645577035735209, | |
| "grad_norm": 15.673484802246094, | |
| "learning_rate": 8.258363175823216e-05, | |
| "loss": 1.7208, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.5231400117164617, | |
| "grad_norm": 33.59520721435547, | |
| "learning_rate": 8.171019302995895e-05, | |
| "loss": 1.7469, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.5817223198594026, | |
| "grad_norm": 15.388129234313965, | |
| "learning_rate": 8.083675430168574e-05, | |
| "loss": 1.7075, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.6403046280023434, | |
| "grad_norm": 11.93837833404541, | |
| "learning_rate": 7.996331557341253e-05, | |
| "loss": 1.696, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.698886936145284, | |
| "grad_norm": 25.31481170654297, | |
| "learning_rate": 7.908987684513932e-05, | |
| "loss": 1.7738, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.757469244288225, | |
| "grad_norm": 38.03368377685547, | |
| "learning_rate": 7.821643811686611e-05, | |
| "loss": 1.7162, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.757469244288225, | |
| "eval_loss": 0.09119272977113724, | |
| "eval_runtime": 150.3416, | |
| "eval_samples_per_second": 3.326, | |
| "eval_steps_per_second": 0.419, | |
| "eval_wer": 0.20601237842617154, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.8160515524311658, | |
| "grad_norm": 15.58292293548584, | |
| "learning_rate": 7.73429993885929e-05, | |
| "loss": 1.6062, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 1.8746338605741066, | |
| "grad_norm": 73.762451171875, | |
| "learning_rate": 7.646956066031969e-05, | |
| "loss": 1.6992, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.9332161687170475, | |
| "grad_norm": 26.838842391967773, | |
| "learning_rate": 7.559612193204648e-05, | |
| "loss": 1.7229, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 1.9917984768599881, | |
| "grad_norm": 22.824892044067383, | |
| "learning_rate": 7.472268320377325e-05, | |
| "loss": 1.7185, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.050380785002929, | |
| "grad_norm": 13.238216400146484, | |
| "learning_rate": 7.384924447550004e-05, | |
| "loss": 1.4583, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.10896309314587, | |
| "grad_norm": 32.13274383544922, | |
| "learning_rate": 7.297580574722683e-05, | |
| "loss": 1.3799, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.1675454012888107, | |
| "grad_norm": 13.44641399383545, | |
| "learning_rate": 7.210236701895362e-05, | |
| "loss": 1.3713, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.2261277094317515, | |
| "grad_norm": 19.05823516845703, | |
| "learning_rate": 7.122892829068041e-05, | |
| "loss": 1.4489, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.2847100175746924, | |
| "grad_norm": 16.24590492248535, | |
| "learning_rate": 7.03554895624072e-05, | |
| "loss": 1.3858, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.3432923257176332, | |
| "grad_norm": 14.049396514892578, | |
| "learning_rate": 6.948205083413399e-05, | |
| "loss": 1.4996, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.3432923257176332, | |
| "eval_loss": 0.09011241793632507, | |
| "eval_runtime": 149.4815, | |
| "eval_samples_per_second": 3.345, | |
| "eval_steps_per_second": 0.421, | |
| "eval_wer": 0.20468611847922194, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.401874633860574, | |
| "grad_norm": 16.26442527770996, | |
| "learning_rate": 6.860861210586078e-05, | |
| "loss": 1.4692, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 2.460456942003515, | |
| "grad_norm": 10.367189407348633, | |
| "learning_rate": 6.773517337758757e-05, | |
| "loss": 1.3203, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 2.519039250146456, | |
| "grad_norm": 21.82206153869629, | |
| "learning_rate": 6.686173464931436e-05, | |
| "loss": 1.5161, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 2.5776215582893967, | |
| "grad_norm": 10.586897850036621, | |
| "learning_rate": 6.598829592104115e-05, | |
| "loss": 1.3843, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 2.6362038664323375, | |
| "grad_norm": 18.108095169067383, | |
| "learning_rate": 6.511485719276793e-05, | |
| "loss": 1.446, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.6947861745752784, | |
| "grad_norm": 17.231735229492188, | |
| "learning_rate": 6.424141846449472e-05, | |
| "loss": 1.4083, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 2.753368482718219, | |
| "grad_norm": 8.833962440490723, | |
| "learning_rate": 6.33679797362215e-05, | |
| "loss": 1.4668, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 2.81195079086116, | |
| "grad_norm": 14.335036277770996, | |
| "learning_rate": 6.24945410079483e-05, | |
| "loss": 1.3906, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 2.870533099004101, | |
| "grad_norm": Infinity, | |
| "learning_rate": 6.162110227967508e-05, | |
| "loss": 1.4241, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 2.9291154071470418, | |
| "grad_norm": 25.987884521484375, | |
| "learning_rate": 6.075639793868461e-05, | |
| "loss": 1.3942, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.9291154071470418, | |
| "eval_loss": 0.0883156806230545, | |
| "eval_runtime": 147.2451, | |
| "eval_samples_per_second": 3.396, | |
| "eval_steps_per_second": 0.428, | |
| "eval_wer": 0.19510757441791923, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.9876977152899826, | |
| "grad_norm": 17.336523056030273, | |
| "learning_rate": 5.988295921041139e-05, | |
| "loss": 1.4093, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 3.0462800234329235, | |
| "grad_norm": 21.30254364013672, | |
| "learning_rate": 5.900952048213818e-05, | |
| "loss": 1.3729, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 3.104862331575864, | |
| "grad_norm": 15.282761573791504, | |
| "learning_rate": 5.813608175386497e-05, | |
| "loss": 1.3091, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 3.1634446397188047, | |
| "grad_norm": 11.485124588012695, | |
| "learning_rate": 5.726264302559175e-05, | |
| "loss": 1.1577, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 3.2220269478617456, | |
| "grad_norm": 17.49385643005371, | |
| "learning_rate": 5.638920429731854e-05, | |
| "loss": 1.2737, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.2806092560046864, | |
| "grad_norm": 19.74750518798828, | |
| "learning_rate": 5.551576556904533e-05, | |
| "loss": 1.1946, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 3.3391915641476273, | |
| "grad_norm": 9.402506828308105, | |
| "learning_rate": 5.4642326840772115e-05, | |
| "loss": 1.2035, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 3.397773872290568, | |
| "grad_norm": 13.279162406921387, | |
| "learning_rate": 5.3768888112498904e-05, | |
| "loss": 1.2891, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 3.456356180433509, | |
| "grad_norm": 18.554702758789062, | |
| "learning_rate": 5.28954493842257e-05, | |
| "loss": 1.1315, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 3.51493848857645, | |
| "grad_norm": 10.541516304016113, | |
| "learning_rate": 5.202201065595249e-05, | |
| "loss": 1.2285, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.51493848857645, | |
| "eval_loss": 0.08758817613124847, | |
| "eval_runtime": 148.7151, | |
| "eval_samples_per_second": 3.362, | |
| "eval_steps_per_second": 0.424, | |
| "eval_wer": 0.1956970232832302, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.5735207967193907, | |
| "grad_norm": 17.15789031982422, | |
| "learning_rate": 5.114857192767928e-05, | |
| "loss": 1.278, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 3.6321031048623316, | |
| "grad_norm": 13.666048049926758, | |
| "learning_rate": 5.027513319940606e-05, | |
| "loss": 1.164, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 3.6906854130052724, | |
| "grad_norm": 14.286330223083496, | |
| "learning_rate": 4.940169447113285e-05, | |
| "loss": 1.2317, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 3.7492677211482133, | |
| "grad_norm": 19.070871353149414, | |
| "learning_rate": 4.852825574285964e-05, | |
| "loss": 1.131, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 3.807850029291154, | |
| "grad_norm": 13.184505462646484, | |
| "learning_rate": 4.7654817014586425e-05, | |
| "loss": 1.2665, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.866432337434095, | |
| "grad_norm": 16.988956451416016, | |
| "learning_rate": 4.6781378286313215e-05, | |
| "loss": 1.2145, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 3.925014645577036, | |
| "grad_norm": 14.714631080627441, | |
| "learning_rate": 4.590793955804001e-05, | |
| "loss": 1.2228, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 3.9835969537199767, | |
| "grad_norm": 15.682711601257324, | |
| "learning_rate": 4.5034500829766794e-05, | |
| "loss": 1.2137, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 4.042179261862917, | |
| "grad_norm": 14.943408012390137, | |
| "learning_rate": 4.4161062101493584e-05, | |
| "loss": 1.2088, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 4.100761570005858, | |
| "grad_norm": 9.292410850524902, | |
| "learning_rate": 4.3287623373220374e-05, | |
| "loss": 1.0637, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.100761570005858, | |
| "eval_loss": 0.08726315200328827, | |
| "eval_runtime": 149.7325, | |
| "eval_samples_per_second": 3.339, | |
| "eval_steps_per_second": 0.421, | |
| "eval_wer": 0.19201296787503683, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.159343878148799, | |
| "grad_norm": 9.017548561096191, | |
| "learning_rate": 4.241418464494716e-05, | |
| "loss": 1.0308, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 4.21792618629174, | |
| "grad_norm": 16.15174102783203, | |
| "learning_rate": 4.1540745916673946e-05, | |
| "loss": 1.0464, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 4.2765084944346805, | |
| "grad_norm": 7.939199447631836, | |
| "learning_rate": 4.0667307188400736e-05, | |
| "loss": 1.1491, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 4.335090802577621, | |
| "grad_norm": 11.129817008972168, | |
| "learning_rate": 3.9793868460127526e-05, | |
| "loss": 1.0738, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 4.393673110720562, | |
| "grad_norm": 8.50700855255127, | |
| "learning_rate": 3.8920429731854315e-05, | |
| "loss": 1.0828, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.452255418863503, | |
| "grad_norm": 11.484732627868652, | |
| "learning_rate": 3.80469910035811e-05, | |
| "loss": 1.0418, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 4.510837727006444, | |
| "grad_norm": 8.929669380187988, | |
| "learning_rate": 3.717355227530789e-05, | |
| "loss": 1.1051, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 4.569420035149385, | |
| "grad_norm": 12.994172096252441, | |
| "learning_rate": 3.630011354703468e-05, | |
| "loss": 1.0193, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 4.628002343292326, | |
| "grad_norm": 9.806758880615234, | |
| "learning_rate": 3.542667481876147e-05, | |
| "loss": 1.1699, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 4.6865846514352665, | |
| "grad_norm": 10.537009239196777, | |
| "learning_rate": 3.455323609048826e-05, | |
| "loss": 1.1144, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.6865846514352665, | |
| "eval_loss": 0.08649158477783203, | |
| "eval_runtime": 148.2812, | |
| "eval_samples_per_second": 3.372, | |
| "eval_steps_per_second": 0.425, | |
| "eval_wer": 0.1927497789566755, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.745166959578207, | |
| "grad_norm": 30.78623390197754, | |
| "learning_rate": 3.367979736221504e-05, | |
| "loss": 1.0902, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 4.803749267721148, | |
| "grad_norm": 9.64354419708252, | |
| "learning_rate": 3.280635863394183e-05, | |
| "loss": 0.9932, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 4.862331575864089, | |
| "grad_norm": 11.149614334106445, | |
| "learning_rate": 3.193291990566862e-05, | |
| "loss": 1.0389, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 4.92091388400703, | |
| "grad_norm": 10.836565971374512, | |
| "learning_rate": 3.10594811773954e-05, | |
| "loss": 1.0626, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 4.979496192149971, | |
| "grad_norm": 11.348654747009277, | |
| "learning_rate": 3.01860424491222e-05, | |
| "loss": 1.0934, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 5.038078500292912, | |
| "grad_norm": 8.341979026794434, | |
| "learning_rate": 2.9312603720848985e-05, | |
| "loss": 1.0845, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 5.0966608084358525, | |
| "grad_norm": 9.784319877624512, | |
| "learning_rate": 2.8439164992575774e-05, | |
| "loss": 0.96, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 5.155243116578793, | |
| "grad_norm": 11.3285551071167, | |
| "learning_rate": 2.756572626430256e-05, | |
| "loss": 0.9211, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 5.213825424721734, | |
| "grad_norm": 8.36048698425293, | |
| "learning_rate": 2.6692287536029347e-05, | |
| "loss": 0.9652, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 5.272407732864675, | |
| "grad_norm": 8.087980270385742, | |
| "learning_rate": 2.5818848807756137e-05, | |
| "loss": 1.0164, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 5.272407732864675, | |
| "eval_loss": 0.08577600121498108, | |
| "eval_runtime": 149.9869, | |
| "eval_samples_per_second": 3.334, | |
| "eval_steps_per_second": 0.42, | |
| "eval_wer": 0.19230769230769232, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 5.330990041007616, | |
| "grad_norm": 14.56843090057373, | |
| "learning_rate": 2.4954144466765657e-05, | |
| "loss": 0.9776, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 5.389572349150557, | |
| "grad_norm": 10.220062255859375, | |
| "learning_rate": 2.4080705738492446e-05, | |
| "loss": 0.9681, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 5.448154657293498, | |
| "grad_norm": 11.614462852478027, | |
| "learning_rate": 2.3207267010219233e-05, | |
| "loss": 0.9691, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 5.506736965436438, | |
| "grad_norm": 6.580599784851074, | |
| "learning_rate": 2.2333828281946022e-05, | |
| "loss": 0.9492, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 5.565319273579379, | |
| "grad_norm": 10.284950256347656, | |
| "learning_rate": 2.1460389553672812e-05, | |
| "loss": 1.0092, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 5.62390158172232, | |
| "grad_norm": 10.794511795043945, | |
| "learning_rate": 2.0586950825399598e-05, | |
| "loss": 0.9039, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 5.682483889865261, | |
| "grad_norm": 12.07039737701416, | |
| "learning_rate": 1.9713512097126388e-05, | |
| "loss": 1.0114, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 5.741066198008202, | |
| "grad_norm": 15.622093200683594, | |
| "learning_rate": 1.8840073368853174e-05, | |
| "loss": 0.9762, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 5.799648506151143, | |
| "grad_norm": 7.460862636566162, | |
| "learning_rate": 1.7966634640579964e-05, | |
| "loss": 0.974, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 5.8582308142940835, | |
| "grad_norm": 9.551807403564453, | |
| "learning_rate": 1.7093195912306753e-05, | |
| "loss": 0.9812, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 5.8582308142940835, | |
| "eval_loss": 0.08563477545976639, | |
| "eval_runtime": 148.6427, | |
| "eval_samples_per_second": 3.364, | |
| "eval_steps_per_second": 0.424, | |
| "eval_wer": 0.1940760389036251, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 5.916813122437024, | |
| "grad_norm": 10.54980754852295, | |
| "learning_rate": 1.621975718403354e-05, | |
| "loss": 0.9981, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 5.975395430579965, | |
| "grad_norm": 8.396002769470215, | |
| "learning_rate": 1.534631845576033e-05, | |
| "loss": 0.911, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 6.033977738722906, | |
| "grad_norm": 6.294841289520264, | |
| "learning_rate": 1.4472879727487117e-05, | |
| "loss": 0.8886, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 6.092560046865847, | |
| "grad_norm": 10.276989936828613, | |
| "learning_rate": 1.3599440999213905e-05, | |
| "loss": 0.8764, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 6.151142355008787, | |
| "grad_norm": 9.27648639678955, | |
| "learning_rate": 1.2726002270940693e-05, | |
| "loss": 0.8728, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 6.209724663151728, | |
| "grad_norm": 10.15081787109375, | |
| "learning_rate": 1.1852563542667483e-05, | |
| "loss": 0.9008, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 6.268306971294669, | |
| "grad_norm": 7.828310012817383, | |
| "learning_rate": 1.0979124814394271e-05, | |
| "loss": 0.8703, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 6.3268892794376095, | |
| "grad_norm": 8.747062683105469, | |
| "learning_rate": 1.0105686086121059e-05, | |
| "loss": 0.9343, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 6.38547158758055, | |
| "grad_norm": 7.199453830718994, | |
| "learning_rate": 9.232247357847849e-06, | |
| "loss": 0.9144, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 6.444053895723491, | |
| "grad_norm": 7.865077972412109, | |
| "learning_rate": 8.358808629574635e-06, | |
| "loss": 0.8927, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 6.444053895723491, | |
| "eval_loss": 0.08493725210428238, | |
| "eval_runtime": 152.514, | |
| "eval_samples_per_second": 3.278, | |
| "eval_steps_per_second": 0.413, | |
| "eval_wer": 0.20173887415266725, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 6.502636203866432, | |
| "grad_norm": 8.560958862304688, | |
| "learning_rate": 7.485369901301424e-06, | |
| "loss": 0.8994, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 6.561218512009373, | |
| "grad_norm": 8.679183006286621, | |
| "learning_rate": 6.611931173028213e-06, | |
| "loss": 0.9121, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 6.619800820152314, | |
| "grad_norm": 7.480246067047119, | |
| "learning_rate": 5.7384924447550014e-06, | |
| "loss": 0.9015, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 6.678383128295255, | |
| "grad_norm": 8.197961807250977, | |
| "learning_rate": 4.865053716481789e-06, | |
| "loss": 0.8958, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 6.7369654364381955, | |
| "grad_norm": 9.989555358886719, | |
| "learning_rate": 3.991614988208577e-06, | |
| "loss": 0.8816, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 6.795547744581136, | |
| "grad_norm": 11.78941535949707, | |
| "learning_rate": 3.118176259935366e-06, | |
| "loss": 0.8848, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 6.854130052724077, | |
| "grad_norm": 8.92465591430664, | |
| "learning_rate": 2.244737531662154e-06, | |
| "loss": 0.8758, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 6.912712360867018, | |
| "grad_norm": 5.927036762237549, | |
| "learning_rate": 1.3712988033889424e-06, | |
| "loss": 0.8421, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 6.971294669009959, | |
| "grad_norm": 6.0556960105896, | |
| "learning_rate": 4.978600751157307e-07, | |
| "loss": 0.8566, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 7.0298769771529, | |
| "grad_norm": 8.540233612060547, | |
| "learning_rate": 3.064574532287266e-05, | |
| "loss": 0.8936, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 7.0298769771529, | |
| "eval_loss": 0.08440528064966202, | |
| "eval_runtime": 151.0062, | |
| "eval_samples_per_second": 3.311, | |
| "eval_steps_per_second": 0.417, | |
| "eval_wer": 0.19613910993221337, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 7.088459285295841, | |
| "grad_norm": 8.703925132751465, | |
| "learning_rate": 3.0042245021122513e-05, | |
| "loss": 0.9237, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 7.147041593438781, | |
| "grad_norm": 11.286332130432129, | |
| "learning_rate": 2.9438744719372363e-05, | |
| "loss": 0.917, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 7.205623901581722, | |
| "grad_norm": 8.638763427734375, | |
| "learning_rate": 2.884127942063971e-05, | |
| "loss": 0.88, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 7.264206209724663, | |
| "grad_norm": 9.697504043579102, | |
| "learning_rate": 2.8237779118889563e-05, | |
| "loss": 0.9342, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 7.322788517867604, | |
| "grad_norm": 8.06286334991455, | |
| "learning_rate": 2.7634278817139407e-05, | |
| "loss": 0.8979, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 7.381370826010545, | |
| "grad_norm": 15.282951354980469, | |
| "learning_rate": 2.703077851538926e-05, | |
| "loss": 0.9201, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 7.439953134153486, | |
| "grad_norm": 11.369593620300293, | |
| "learning_rate": 2.6427278213639107e-05, | |
| "loss": 0.9538, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 7.4985354422964265, | |
| "grad_norm": 10.077091217041016, | |
| "learning_rate": 2.5823777911888958e-05, | |
| "loss": 0.9807, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 7.557117750439367, | |
| "grad_norm": 9.917128562927246, | |
| "learning_rate": 2.5220277610138805e-05, | |
| "loss": 0.939, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 7.615700058582308, | |
| "grad_norm": 10.68909740447998, | |
| "learning_rate": 2.4616777308388655e-05, | |
| "loss": 0.8718, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 7.615700058582308, | |
| "eval_loss": 0.08539500832557678, | |
| "eval_runtime": 149.7688, | |
| "eval_samples_per_second": 3.338, | |
| "eval_steps_per_second": 0.421, | |
| "eval_wer": 0.19790745652814618, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 7.674282366725249, | |
| "grad_norm": 8.260842323303223, | |
| "learning_rate": 2.4013277006638505e-05, | |
| "loss": 0.9048, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 7.73286467486819, | |
| "grad_norm": 16.34197235107422, | |
| "learning_rate": 2.3409776704888352e-05, | |
| "loss": 0.9262, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 7.791446983011131, | |
| "grad_norm": 8.024565696716309, | |
| "learning_rate": 2.2806276403138202e-05, | |
| "loss": 0.9952, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 7.850029291154072, | |
| "grad_norm": 7.884005069732666, | |
| "learning_rate": 2.220277610138805e-05, | |
| "loss": 0.9656, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 7.9086115992970125, | |
| "grad_norm": 8.73161506652832, | |
| "learning_rate": 2.15992757996379e-05, | |
| "loss": 0.9205, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 7.967193907439953, | |
| "grad_norm": 9.00133991241455, | |
| "learning_rate": 2.099577549788775e-05, | |
| "loss": 0.9347, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 8.025776215582894, | |
| "grad_norm": 6.850646495819092, | |
| "learning_rate": 2.03922751961376e-05, | |
| "loss": 0.8468, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 8.084358523725834, | |
| "grad_norm": 11.7725248336792, | |
| "learning_rate": 1.978877489438745e-05, | |
| "loss": 0.8948, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 8.142940831868776, | |
| "grad_norm": 6.4703474044799805, | |
| "learning_rate": 1.9185274592637298e-05, | |
| "loss": 0.8727, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 8.201523140011716, | |
| "grad_norm": 15.587645530700684, | |
| "learning_rate": 1.8581774290887148e-05, | |
| "loss": 0.9019, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 8.201523140011716, | |
| "eval_loss": 0.08469171822071075, | |
| "eval_runtime": 147.0496, | |
| "eval_samples_per_second": 3.4, | |
| "eval_steps_per_second": 0.428, | |
| "eval_wer": 0.18538166814028884, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 8.260105448154658, | |
| "grad_norm": 7.585418701171875, | |
| "learning_rate": 1.7978273989136995e-05, | |
| "loss": 0.8818, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 8.318687756297598, | |
| "grad_norm": 9.436836242675781, | |
| "learning_rate": 1.7374773687386845e-05, | |
| "loss": 0.8864, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 8.37727006444054, | |
| "grad_norm": 12.12936019897461, | |
| "learning_rate": 1.6771273385636692e-05, | |
| "loss": 0.8744, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 8.43585237258348, | |
| "grad_norm": 11.584985733032227, | |
| "learning_rate": 1.6167773083886542e-05, | |
| "loss": 0.8542, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 8.494434680726421, | |
| "grad_norm": 7.6883440017700195, | |
| "learning_rate": 1.5564272782136393e-05, | |
| "loss": 0.8714, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 8.553016988869361, | |
| "grad_norm": 13.686609268188477, | |
| "learning_rate": 1.496077248038624e-05, | |
| "loss": 0.8726, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 8.611599297012303, | |
| "grad_norm": 10.200602531433105, | |
| "learning_rate": 1.436330718165359e-05, | |
| "loss": 0.8839, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 8.670181605155243, | |
| "grad_norm": 6.929018020629883, | |
| "learning_rate": 1.3759806879903441e-05, | |
| "loss": 0.8628, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 8.728763913298184, | |
| "grad_norm": 9.72988224029541, | |
| "learning_rate": 1.3156306578153291e-05, | |
| "loss": 0.8672, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 8.787346221441124, | |
| "grad_norm": 7.273561477661133, | |
| "learning_rate": 1.255280627640314e-05, | |
| "loss": 0.8293, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 8.787346221441124, | |
| "eval_loss": 0.08473628014326096, | |
| "eval_runtime": 147.7929, | |
| "eval_samples_per_second": 3.383, | |
| "eval_steps_per_second": 0.426, | |
| "eval_wer": 0.19834954317712938, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 8.845928529584066, | |
| "grad_norm": 6.900521278381348, | |
| "learning_rate": 1.1949305974652989e-05, | |
| "loss": 0.8797, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 8.904510837727006, | |
| "grad_norm": 13.14035701751709, | |
| "learning_rate": 1.1345805672902837e-05, | |
| "loss": 0.8691, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 8.963093145869948, | |
| "grad_norm": 6.80872106552124, | |
| "learning_rate": 1.0742305371152686e-05, | |
| "loss": 0.859, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 9.021675454012888, | |
| "grad_norm": 5.7985520362854, | |
| "learning_rate": 1.0138805069402535e-05, | |
| "loss": 0.8905, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 9.08025776215583, | |
| "grad_norm": 7.384444236755371, | |
| "learning_rate": 9.535304767652383e-06, | |
| "loss": 0.7981, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 9.13884007029877, | |
| "grad_norm": 6.441751956939697, | |
| "learning_rate": 8.931804465902233e-06, | |
| "loss": 0.9026, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 9.197422378441711, | |
| "grad_norm": 5.471485614776611, | |
| "learning_rate": 8.328304164152082e-06, | |
| "loss": 0.8357, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 9.256004686584651, | |
| "grad_norm": 6.093921661376953, | |
| "learning_rate": 7.724803862401932e-06, | |
| "loss": 0.8427, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 9.314586994727593, | |
| "grad_norm": 5.414072036743164, | |
| "learning_rate": 7.121303560651781e-06, | |
| "loss": 0.8235, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 9.373169302870533, | |
| "grad_norm": 5.2771897315979, | |
| "learning_rate": 6.5178032589016296e-06, | |
| "loss": 0.8363, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 9.373169302870533, | |
| "eval_loss": 0.08421996235847473, | |
| "eval_runtime": 150.6901, | |
| "eval_samples_per_second": 3.318, | |
| "eval_steps_per_second": 0.418, | |
| "eval_wer": 0.19820218096080164, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 9.431751611013475, | |
| "grad_norm": 5.505492687225342, | |
| "learning_rate": 5.914302957151479e-06, | |
| "loss": 0.7991, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 9.490333919156415, | |
| "grad_norm": 6.690750598907471, | |
| "learning_rate": 5.310802655401328e-06, | |
| "loss": 0.8071, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 9.548916227299356, | |
| "grad_norm": 6.654877185821533, | |
| "learning_rate": 4.707302353651177e-06, | |
| "loss": 0.834, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 9.607498535442296, | |
| "grad_norm": 9.937077522277832, | |
| "learning_rate": 4.1038020519010266e-06, | |
| "loss": 0.8055, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 9.666080843585238, | |
| "grad_norm": 6.015642166137695, | |
| "learning_rate": 3.500301750150875e-06, | |
| "loss": 0.8734, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 9.724663151728178, | |
| "grad_norm": 12.434464454650879, | |
| "learning_rate": 2.896801448400724e-06, | |
| "loss": 0.8544, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 9.783245459871118, | |
| "grad_norm": 6.330708980560303, | |
| "learning_rate": 2.2933011466505732e-06, | |
| "loss": 0.798, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 9.84182776801406, | |
| "grad_norm": 5.820682048797607, | |
| "learning_rate": 1.6898008449004227e-06, | |
| "loss": 0.7989, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 9.900410076157002, | |
| "grad_norm": 8.209725379943848, | |
| "learning_rate": 1.0863005431502715e-06, | |
| "loss": 0.8454, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 9.958992384299941, | |
| "grad_norm": 10.676623344421387, | |
| "learning_rate": 4.828002414001208e-07, | |
| "loss": 0.8034, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 9.958992384299941, | |
| "eval_loss": 0.08400186896324158, | |
| "eval_runtime": 148.9273, | |
| "eval_samples_per_second": 3.357, | |
| "eval_steps_per_second": 0.423, | |
| "eval_wer": 0.197465369879163, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 10.017574692442881, | |
| "grad_norm": 8.057076454162598, | |
| "learning_rate": 3.8124720274503957e-05, | |
| "loss": 0.8553, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 10.076157000585823, | |
| "grad_norm": 6.9262895584106445, | |
| "learning_rate": 3.7751752946441895e-05, | |
| "loss": 0.8579, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 10.134739308728763, | |
| "grad_norm": 7.844761371612549, | |
| "learning_rate": 3.737878561837983e-05, | |
| "loss": 0.8764, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 10.193321616871705, | |
| "grad_norm": 17.013307571411133, | |
| "learning_rate": 3.700954796359839e-05, | |
| "loss": 0.9679, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 10.251903925014645, | |
| "grad_norm": 10.550113677978516, | |
| "learning_rate": 3.663658063553633e-05, | |
| "loss": 0.8919, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 10.310486233157587, | |
| "grad_norm": 7.075886249542236, | |
| "learning_rate": 3.6263613307474266e-05, | |
| "loss": 0.9601, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 10.369068541300527, | |
| "grad_norm": 10.027517318725586, | |
| "learning_rate": 3.5890645979412204e-05, | |
| "loss": 0.8596, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 10.427650849443468, | |
| "grad_norm": 5.811990261077881, | |
| "learning_rate": 3.551767865135014e-05, | |
| "loss": 0.9023, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 10.486233157586408, | |
| "grad_norm": 12.586297988891602, | |
| "learning_rate": 3.514471132328808e-05, | |
| "loss": 0.93, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 10.54481546572935, | |
| "grad_norm": 9.313389778137207, | |
| "learning_rate": 3.477174399522602e-05, | |
| "loss": 0.8462, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 10.54481546572935, | |
| "eval_loss": 0.08548491448163986, | |
| "eval_runtime": 156.5497, | |
| "eval_samples_per_second": 3.194, | |
| "eval_steps_per_second": 0.402, | |
| "eval_wer": 0.19525493663424698, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 10.60339777387229, | |
| "grad_norm": 9.245261192321777, | |
| "learning_rate": 3.439877666716396e-05, | |
| "loss": 0.9351, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 10.661980082015232, | |
| "grad_norm": 7.513446807861328, | |
| "learning_rate": 3.4025809339101895e-05, | |
| "loss": 0.8722, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 10.720562390158172, | |
| "grad_norm": 12.897968292236328, | |
| "learning_rate": 3.365284201103984e-05, | |
| "loss": 0.9576, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 10.779144698301113, | |
| "grad_norm": 9.550192832946777, | |
| "learning_rate": 3.327987468297777e-05, | |
| "loss": 0.9241, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 10.837727006444053, | |
| "grad_norm": 6.644899368286133, | |
| "learning_rate": 3.290690735491571e-05, | |
| "loss": 0.9291, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 10.896309314586995, | |
| "grad_norm": 9.285797119140625, | |
| "learning_rate": 3.253394002685365e-05, | |
| "loss": 0.8971, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 10.954891622729935, | |
| "grad_norm": 8.316353797912598, | |
| "learning_rate": 3.2160972698791586e-05, | |
| "loss": 0.8892, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 11.013473930872877, | |
| "grad_norm": 17.100173950195312, | |
| "learning_rate": 3.1788005370729524e-05, | |
| "loss": 0.8732, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 11.072056239015817, | |
| "grad_norm": 19.12342643737793, | |
| "learning_rate": 3.141503804266747e-05, | |
| "loss": 0.8663, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 11.130638547158759, | |
| "grad_norm": 7.627189636230469, | |
| "learning_rate": 3.10420707146054e-05, | |
| "loss": 0.8824, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 11.130638547158759, | |
| "eval_loss": 0.08482780307531357, | |
| "eval_runtime": 146.6345, | |
| "eval_samples_per_second": 3.41, | |
| "eval_steps_per_second": 0.43, | |
| "eval_wer": 0.19304450338933096, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 11.189220855301699, | |
| "grad_norm": 7.1925950050354, | |
| "learning_rate": 3.066910338654334e-05, | |
| "loss": 0.8375, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 11.24780316344464, | |
| "grad_norm": 8.55908489227295, | |
| "learning_rate": 3.0296136058481277e-05, | |
| "loss": 0.8335, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 11.30638547158758, | |
| "grad_norm": 15.089740753173828, | |
| "learning_rate": 2.9923168730419215e-05, | |
| "loss": 0.9117, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 11.364967779730522, | |
| "grad_norm": 8.451448440551758, | |
| "learning_rate": 2.9550201402357153e-05, | |
| "loss": 0.8586, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 11.423550087873462, | |
| "grad_norm": 7.994997501373291, | |
| "learning_rate": 2.9177234074295095e-05, | |
| "loss": 0.8459, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 11.482132396016404, | |
| "grad_norm": 8.656350135803223, | |
| "learning_rate": 2.8804266746233033e-05, | |
| "loss": 0.9268, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 11.540714704159344, | |
| "grad_norm": 10.245903968811035, | |
| "learning_rate": 2.843129941817097e-05, | |
| "loss": 0.8657, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 11.599297012302285, | |
| "grad_norm": 8.758448600769043, | |
| "learning_rate": 2.806206176338953e-05, | |
| "loss": 0.8875, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 11.657879320445225, | |
| "grad_norm": 6.345497131347656, | |
| "learning_rate": 2.7689094435327466e-05, | |
| "loss": 0.9115, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 11.716461628588167, | |
| "grad_norm": 9.602129936218262, | |
| "learning_rate": 2.7316127107265404e-05, | |
| "loss": 0.8591, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 11.716461628588167, | |
| "eval_loss": 0.0848940759897232, | |
| "eval_runtime": 147.8483, | |
| "eval_samples_per_second": 3.382, | |
| "eval_steps_per_second": 0.426, | |
| "eval_wer": 0.18376068376068377, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 11.775043936731107, | |
| "grad_norm": 10.147231101989746, | |
| "learning_rate": 2.6943159779203343e-05, | |
| "loss": 0.8628, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 11.833626244874049, | |
| "grad_norm": 7.136846542358398, | |
| "learning_rate": 2.657019245114128e-05, | |
| "loss": 0.9034, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 11.892208553016989, | |
| "grad_norm": 11.765522003173828, | |
| "learning_rate": 2.6197225123079222e-05, | |
| "loss": 0.8635, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 11.95079086115993, | |
| "grad_norm": 7.662530422210693, | |
| "learning_rate": 2.582425779501716e-05, | |
| "loss": 0.8561, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 12.00937316930287, | |
| "grad_norm": 10.559505462646484, | |
| "learning_rate": 2.5451290466955095e-05, | |
| "loss": 0.8319, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 12.067955477445812, | |
| "grad_norm": 6.208855152130127, | |
| "learning_rate": 2.5078323138893034e-05, | |
| "loss": 0.7679, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 12.126537785588752, | |
| "grad_norm": 10.877766609191895, | |
| "learning_rate": 2.4705355810830972e-05, | |
| "loss": 0.8484, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 12.185120093731694, | |
| "grad_norm": 8.098186492919922, | |
| "learning_rate": 2.433238848276891e-05, | |
| "loss": 0.8644, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 12.243702401874634, | |
| "grad_norm": 7.471461772918701, | |
| "learning_rate": 2.3959421154706848e-05, | |
| "loss": 0.8372, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 12.302284710017574, | |
| "grad_norm": 10.694164276123047, | |
| "learning_rate": 2.3586453826644786e-05, | |
| "loss": 0.8339, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 12.302284710017574, | |
| "eval_loss": 0.08417258411645889, | |
| "eval_runtime": 146.3017, | |
| "eval_samples_per_second": 3.418, | |
| "eval_steps_per_second": 0.431, | |
| "eval_wer": 0.18626584143825523, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 12.360867018160516, | |
| "grad_norm": 8.07205581665039, | |
| "learning_rate": 2.3213486498582724e-05, | |
| "loss": 0.7901, | |
| "step": 21100 | |
| }, | |
| { | |
| "epoch": 12.419449326303456, | |
| "grad_norm": 10.554586410522461, | |
| "learning_rate": 2.2840519170520663e-05, | |
| "loss": 0.8117, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 12.478031634446397, | |
| "grad_norm": 9.55418872833252, | |
| "learning_rate": 2.24675518424586e-05, | |
| "loss": 0.7946, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 12.536613942589337, | |
| "grad_norm": 8.738641738891602, | |
| "learning_rate": 2.2094584514396542e-05, | |
| "loss": 0.8026, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 12.595196250732279, | |
| "grad_norm": 9.077950477600098, | |
| "learning_rate": 2.1721617186334477e-05, | |
| "loss": 0.8102, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 12.653778558875219, | |
| "grad_norm": 7.797760009765625, | |
| "learning_rate": 2.1348649858272415e-05, | |
| "loss": 0.8608, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 12.71236086701816, | |
| "grad_norm": 7.361778736114502, | |
| "learning_rate": 2.0975682530210357e-05, | |
| "loss": 0.8448, | |
| "step": 21700 | |
| }, | |
| { | |
| "epoch": 12.7709431751611, | |
| "grad_norm": 5.896770477294922, | |
| "learning_rate": 2.0602715202148292e-05, | |
| "loss": 0.8394, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 12.829525483304042, | |
| "grad_norm": 11.283666610717773, | |
| "learning_rate": 2.0233477547366852e-05, | |
| "loss": 0.8326, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 12.888107791446982, | |
| "grad_norm": 8.800780296325684, | |
| "learning_rate": 1.986051021930479e-05, | |
| "loss": 0.8573, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 12.888107791446982, | |
| "eval_loss": 0.08360177278518677, | |
| "eval_runtime": 148.5908, | |
| "eval_samples_per_second": 3.365, | |
| "eval_steps_per_second": 0.424, | |
| "eval_wer": 0.19260241674034778, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 12.946690099589924, | |
| "grad_norm": 7.05850887298584, | |
| "learning_rate": 1.948754289124273e-05, | |
| "loss": 0.7926, | |
| "step": 22100 | |
| }, | |
| { | |
| "epoch": 13.005272407732864, | |
| "grad_norm": 7.339128017425537, | |
| "learning_rate": 1.9114575563180667e-05, | |
| "loss": 0.8611, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 13.063854715875806, | |
| "grad_norm": 7.789575576782227, | |
| "learning_rate": 1.8741608235118605e-05, | |
| "loss": 0.8006, | |
| "step": 22300 | |
| }, | |
| { | |
| "epoch": 13.122437024018746, | |
| "grad_norm": 10.398506164550781, | |
| "learning_rate": 1.8368640907056543e-05, | |
| "loss": 0.8397, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 13.181019332161688, | |
| "grad_norm": 8.361679077148438, | |
| "learning_rate": 1.799567357899448e-05, | |
| "loss": 0.8027, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 13.239601640304627, | |
| "grad_norm": 8.431894302368164, | |
| "learning_rate": 1.762270625093242e-05, | |
| "loss": 0.824, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 13.29818394844757, | |
| "grad_norm": 5.742968559265137, | |
| "learning_rate": 1.7249738922870357e-05, | |
| "loss": 0.792, | |
| "step": 22700 | |
| }, | |
| { | |
| "epoch": 13.35676625659051, | |
| "grad_norm": 13.589301109313965, | |
| "learning_rate": 1.6876771594808296e-05, | |
| "loss": 0.7701, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 13.415348564733451, | |
| "grad_norm": 6.760063171386719, | |
| "learning_rate": 1.6503804266746234e-05, | |
| "loss": 0.7756, | |
| "step": 22900 | |
| }, | |
| { | |
| "epoch": 13.473930872876391, | |
| "grad_norm": 8.249403953552246, | |
| "learning_rate": 1.6130836938684172e-05, | |
| "loss": 0.7445, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 13.473930872876391, | |
| "eval_loss": 0.08392482995986938, | |
| "eval_runtime": 146.723, | |
| "eval_samples_per_second": 3.408, | |
| "eval_steps_per_second": 0.429, | |
| "eval_wer": 0.18420277040966695, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 13.532513181019333, | |
| "grad_norm": 6.522210597991943, | |
| "learning_rate": 1.575786961062211e-05, | |
| "loss": 0.7743, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 13.591095489162273, | |
| "grad_norm": 6.639892101287842, | |
| "learning_rate": 1.538490228256005e-05, | |
| "loss": 0.8147, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 13.649677797305214, | |
| "grad_norm": 6.0167999267578125, | |
| "learning_rate": 1.5011934954497987e-05, | |
| "loss": 0.781, | |
| "step": 23300 | |
| }, | |
| { | |
| "epoch": 13.708260105448154, | |
| "grad_norm": 9.794026374816895, | |
| "learning_rate": 1.4638967626435926e-05, | |
| "loss": 0.771, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 13.766842413591096, | |
| "grad_norm": 8.01543140411377, | |
| "learning_rate": 1.4266000298373863e-05, | |
| "loss": 0.7812, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 13.825424721734036, | |
| "grad_norm": 10.331818580627441, | |
| "learning_rate": 1.3893032970311801e-05, | |
| "loss": 0.7864, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 13.884007029876978, | |
| "grad_norm": 8.055398941040039, | |
| "learning_rate": 1.3520065642249741e-05, | |
| "loss": 0.7991, | |
| "step": 23700 | |
| }, | |
| { | |
| "epoch": 13.942589338019918, | |
| "grad_norm": 8.330449104309082, | |
| "learning_rate": 1.3147098314187677e-05, | |
| "loss": 0.8088, | |
| "step": 23800 | |
| }, | |
| { | |
| "epoch": 14.00117164616286, | |
| "grad_norm": 6.869382381439209, | |
| "learning_rate": 1.2774130986125616e-05, | |
| "loss": 0.76, | |
| "step": 23900 | |
| }, | |
| { | |
| "epoch": 14.0597539543058, | |
| "grad_norm": 6.649117469787598, | |
| "learning_rate": 1.2401163658063554e-05, | |
| "loss": 0.783, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 14.0597539543058, | |
| "eval_loss": 0.08357907831668854, | |
| "eval_runtime": 147.3445, | |
| "eval_samples_per_second": 3.393, | |
| "eval_steps_per_second": 0.428, | |
| "eval_wer": 0.18420277040966695, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 14.118336262448741, | |
| "grad_norm": 8.647706031799316, | |
| "learning_rate": 1.2028196330001492e-05, | |
| "loss": 0.7496, | |
| "step": 24100 | |
| }, | |
| { | |
| "epoch": 14.176918570591681, | |
| "grad_norm": 8.944561004638672, | |
| "learning_rate": 1.165522900193943e-05, | |
| "loss": 0.7662, | |
| "step": 24200 | |
| }, | |
| { | |
| "epoch": 14.235500878734623, | |
| "grad_norm": 8.365220069885254, | |
| "learning_rate": 1.1282261673877368e-05, | |
| "loss": 0.7645, | |
| "step": 24300 | |
| }, | |
| { | |
| "epoch": 14.294083186877563, | |
| "grad_norm": 9.97271728515625, | |
| "learning_rate": 1.0909294345815308e-05, | |
| "loss": 0.74, | |
| "step": 24400 | |
| }, | |
| { | |
| "epoch": 14.352665495020505, | |
| "grad_norm": 5.968284606933594, | |
| "learning_rate": 1.0536327017753245e-05, | |
| "loss": 0.7875, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 14.411247803163445, | |
| "grad_norm": 6.4041008949279785, | |
| "learning_rate": 1.0163359689691183e-05, | |
| "loss": 0.7557, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 14.469830111306386, | |
| "grad_norm": 6.7281036376953125, | |
| "learning_rate": 9.790392361629123e-06, | |
| "loss": 0.7478, | |
| "step": 24700 | |
| }, | |
| { | |
| "epoch": 14.528412419449326, | |
| "grad_norm": 9.13178539276123, | |
| "learning_rate": 9.41742503356706e-06, | |
| "loss": 0.7528, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 14.586994727592266, | |
| "grad_norm": 7.79683780670166, | |
| "learning_rate": 9.044457705504997e-06, | |
| "loss": 0.805, | |
| "step": 24900 | |
| }, | |
| { | |
| "epoch": 14.645577035735208, | |
| "grad_norm": 6.039112567901611, | |
| "learning_rate": 8.671490377442937e-06, | |
| "loss": 0.7263, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 14.645577035735208, | |
| "eval_loss": 0.08391948789358139, | |
| "eval_runtime": 147.8059, | |
| "eval_samples_per_second": 3.383, | |
| "eval_steps_per_second": 0.426, | |
| "eval_wer": 0.18243442381373415, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 14.70415934387815, | |
| "grad_norm": 7.50616979598999, | |
| "learning_rate": 8.298523049380874e-06, | |
| "loss": 0.7722, | |
| "step": 25100 | |
| }, | |
| { | |
| "epoch": 14.76274165202109, | |
| "grad_norm": 6.403426647186279, | |
| "learning_rate": 7.925555721318812e-06, | |
| "loss": 0.7742, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 14.82132396016403, | |
| "grad_norm": 7.445984363555908, | |
| "learning_rate": 7.556318066537371e-06, | |
| "loss": 0.7505, | |
| "step": 25300 | |
| }, | |
| { | |
| "epoch": 14.879906268306971, | |
| "grad_norm": 7.770444869995117, | |
| "learning_rate": 7.1833507384753095e-06, | |
| "loss": 0.7524, | |
| "step": 25400 | |
| }, | |
| { | |
| "epoch": 14.938488576449911, | |
| "grad_norm": 6.477992057800293, | |
| "learning_rate": 6.8103834104132485e-06, | |
| "loss": 0.7477, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 14.997070884592853, | |
| "grad_norm": 10.539923667907715, | |
| "learning_rate": 6.437416082351186e-06, | |
| "loss": 0.7285, | |
| "step": 25600 | |
| }, | |
| { | |
| "epoch": 15.055653192735793, | |
| "grad_norm": 6.943923473358154, | |
| "learning_rate": 6.064448754289125e-06, | |
| "loss": 0.6959, | |
| "step": 25700 | |
| }, | |
| { | |
| "epoch": 15.114235500878735, | |
| "grad_norm": 4.48841667175293, | |
| "learning_rate": 5.691481426227062e-06, | |
| "loss": 0.7265, | |
| "step": 25800 | |
| }, | |
| { | |
| "epoch": 15.172817809021675, | |
| "grad_norm": 5.900568008422852, | |
| "learning_rate": 5.318514098165001e-06, | |
| "loss": 0.765, | |
| "step": 25900 | |
| }, | |
| { | |
| "epoch": 15.231400117164617, | |
| "grad_norm": 5.62386417388916, | |
| "learning_rate": 4.9455467701029394e-06, | |
| "loss": 0.7634, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 15.231400117164617, | |
| "eval_loss": 0.08351606130599976, | |
| "eval_runtime": 146.6151, | |
| "eval_samples_per_second": 3.41, | |
| "eval_steps_per_second": 0.43, | |
| "eval_wer": 0.1825817860300619, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 15.289982425307556, | |
| "grad_norm": 6.403947353363037, | |
| "learning_rate": 4.572579442040878e-06, | |
| "loss": 0.728, | |
| "step": 26100 | |
| }, | |
| { | |
| "epoch": 15.348564733450498, | |
| "grad_norm": 6.236737251281738, | |
| "learning_rate": 4.199612113978816e-06, | |
| "loss": 0.7462, | |
| "step": 26200 | |
| }, | |
| { | |
| "epoch": 15.407147041593438, | |
| "grad_norm": 5.554813385009766, | |
| "learning_rate": 3.826644785916754e-06, | |
| "loss": 0.7403, | |
| "step": 26300 | |
| }, | |
| { | |
| "epoch": 15.46572934973638, | |
| "grad_norm": 4.909285068511963, | |
| "learning_rate": 3.453677457854692e-06, | |
| "loss": 0.7653, | |
| "step": 26400 | |
| }, | |
| { | |
| "epoch": 15.52431165787932, | |
| "grad_norm": 5.502344608306885, | |
| "learning_rate": 3.0807101297926304e-06, | |
| "loss": 0.7319, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 15.582893966022262, | |
| "grad_norm": 7.525850772857666, | |
| "learning_rate": 2.7077428017305685e-06, | |
| "loss": 0.7295, | |
| "step": 26600 | |
| }, | |
| { | |
| "epoch": 15.641476274165202, | |
| "grad_norm": 7.245991230010986, | |
| "learning_rate": 2.3347754736685067e-06, | |
| "loss": 0.7382, | |
| "step": 26700 | |
| }, | |
| { | |
| "epoch": 15.700058582308143, | |
| "grad_norm": 5.762548923492432, | |
| "learning_rate": 1.961808145606445e-06, | |
| "loss": 0.7253, | |
| "step": 26800 | |
| }, | |
| { | |
| "epoch": 15.758640890451083, | |
| "grad_norm": 6.127166271209717, | |
| "learning_rate": 1.5888408175443833e-06, | |
| "loss": 0.7399, | |
| "step": 26900 | |
| }, | |
| { | |
| "epoch": 15.817223198594025, | |
| "grad_norm": 6.195973873138428, | |
| "learning_rate": 1.2158734894823213e-06, | |
| "loss": 0.7379, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 15.817223198594025, | |
| "eval_loss": 0.08341451734304428, | |
| "eval_runtime": 146.5011, | |
| "eval_samples_per_second": 3.413, | |
| "eval_steps_per_second": 0.43, | |
| "eval_wer": 0.18287651046271736, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 15.875805506736965, | |
| "grad_norm": 8.746485710144043, | |
| "learning_rate": 8.429061614202597e-07, | |
| "loss": 0.7027, | |
| "step": 27100 | |
| }, | |
| { | |
| "epoch": 15.934387814879907, | |
| "grad_norm": 7.414266109466553, | |
| "learning_rate": 4.699388333581979e-07, | |
| "loss": 0.7047, | |
| "step": 27200 | |
| }, | |
| { | |
| "epoch": 15.992970123022847, | |
| "grad_norm": 7.6758832931518555, | |
| "learning_rate": 1.0070117857675669e-07, | |
| "loss": 0.7324, | |
| "step": 27300 | |
| }, | |
| { | |
| "epoch": 16.05155243116579, | |
| "grad_norm": 12.168989181518555, | |
| "learning_rate": 2.7443730771306743e-05, | |
| "loss": 0.7807, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 16.11013473930873, | |
| "grad_norm": 5.686388969421387, | |
| "learning_rate": 2.717385437469639e-05, | |
| "loss": 0.8098, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 16.16871704745167, | |
| "grad_norm": 6.920952320098877, | |
| "learning_rate": 2.6903977978086036e-05, | |
| "loss": 0.7786, | |
| "step": 27600 | |
| }, | |
| { | |
| "epoch": 16.22729935559461, | |
| "grad_norm": 8.017998695373535, | |
| "learning_rate": 2.6634101581475683e-05, | |
| "loss": 0.8257, | |
| "step": 27700 | |
| }, | |
| { | |
| "epoch": 16.285881663737552, | |
| "grad_norm": 6.6050262451171875, | |
| "learning_rate": 2.6364225184865333e-05, | |
| "loss": 0.7671, | |
| "step": 27800 | |
| }, | |
| { | |
| "epoch": 16.344463971880494, | |
| "grad_norm": 8.146703720092773, | |
| "learning_rate": 2.609434878825498e-05, | |
| "loss": 0.7733, | |
| "step": 27900 | |
| }, | |
| { | |
| "epoch": 16.403046280023432, | |
| "grad_norm": 11.652145385742188, | |
| "learning_rate": 2.5824472391644626e-05, | |
| "loss": 0.7902, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 16.403046280023432, | |
| "eval_loss": 0.08416531980037689, | |
| "eval_runtime": 148.3349, | |
| "eval_samples_per_second": 3.371, | |
| "eval_steps_per_second": 0.425, | |
| "eval_wer": 0.18110816386678455, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 16.461628588166374, | |
| "grad_norm": 9.010910034179688, | |
| "learning_rate": 2.5554595995034276e-05, | |
| "loss": 0.7898, | |
| "step": 28100 | |
| }, | |
| { | |
| "epoch": 16.520210896309315, | |
| "grad_norm": 11.093189239501953, | |
| "learning_rate": 2.5284719598423923e-05, | |
| "loss": 0.7682, | |
| "step": 28200 | |
| }, | |
| { | |
| "epoch": 16.578793204452257, | |
| "grad_norm": 7.964006423950195, | |
| "learning_rate": 2.501484320181357e-05, | |
| "loss": 0.7481, | |
| "step": 28300 | |
| }, | |
| { | |
| "epoch": 16.637375512595195, | |
| "grad_norm": 10.423765182495117, | |
| "learning_rate": 2.474496680520322e-05, | |
| "loss": 0.7515, | |
| "step": 28400 | |
| }, | |
| { | |
| "epoch": 16.695957820738137, | |
| "grad_norm": 6.754664897918701, | |
| "learning_rate": 2.4475090408592866e-05, | |
| "loss": 0.8024, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 16.75454012888108, | |
| "grad_norm": 10.269820213317871, | |
| "learning_rate": 2.4205214011982513e-05, | |
| "loss": 0.7533, | |
| "step": 28600 | |
| }, | |
| { | |
| "epoch": 16.81312243702402, | |
| "grad_norm": 14.783397674560547, | |
| "learning_rate": 2.3935337615372163e-05, | |
| "loss": 0.7937, | |
| "step": 28700 | |
| }, | |
| { | |
| "epoch": 16.87170474516696, | |
| "grad_norm": 8.369039535522461, | |
| "learning_rate": 2.366546121876181e-05, | |
| "loss": 0.8246, | |
| "step": 28800 | |
| }, | |
| { | |
| "epoch": 16.9302870533099, | |
| "grad_norm": 8.320894241333008, | |
| "learning_rate": 2.3395584822151456e-05, | |
| "loss": 0.7703, | |
| "step": 28900 | |
| }, | |
| { | |
| "epoch": 16.988869361452842, | |
| "grad_norm": 10.491236686706543, | |
| "learning_rate": 2.3125708425541102e-05, | |
| "loss": 0.8261, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 16.988869361452842, | |
| "eval_loss": 0.0840698629617691, | |
| "eval_runtime": 145.4812, | |
| "eval_samples_per_second": 3.437, | |
| "eval_steps_per_second": 0.433, | |
| "eval_wer": 0.18493958149130563, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 17.047451669595784, | |
| "grad_norm": 6.7529377937316895, | |
| "learning_rate": 2.2855832028930752e-05, | |
| "loss": 0.8091, | |
| "step": 29100 | |
| }, | |
| { | |
| "epoch": 17.106033977738722, | |
| "grad_norm": 6.652218341827393, | |
| "learning_rate": 2.25859556323204e-05, | |
| "loss": 0.7695, | |
| "step": 29200 | |
| }, | |
| { | |
| "epoch": 17.164616285881664, | |
| "grad_norm": 7.654794692993164, | |
| "learning_rate": 2.2316079235710046e-05, | |
| "loss": 0.7926, | |
| "step": 29300 | |
| }, | |
| { | |
| "epoch": 17.223198594024606, | |
| "grad_norm": 8.277422904968262, | |
| "learning_rate": 2.2046202839099696e-05, | |
| "loss": 0.7702, | |
| "step": 29400 | |
| }, | |
| { | |
| "epoch": 17.281780902167544, | |
| "grad_norm": 15.797304153442383, | |
| "learning_rate": 2.1776326442489342e-05, | |
| "loss": 0.7597, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 17.340363210310485, | |
| "grad_norm": 6.769285202026367, | |
| "learning_rate": 2.150645004587899e-05, | |
| "loss": 0.7588, | |
| "step": 29600 | |
| }, | |
| { | |
| "epoch": 17.398945518453427, | |
| "grad_norm": 8.328302383422852, | |
| "learning_rate": 2.123657364926864e-05, | |
| "loss": 0.7719, | |
| "step": 29700 | |
| }, | |
| { | |
| "epoch": 17.45752782659637, | |
| "grad_norm": 5.7514190673828125, | |
| "learning_rate": 2.0966697252658282e-05, | |
| "loss": 0.7985, | |
| "step": 29800 | |
| }, | |
| { | |
| "epoch": 17.516110134739307, | |
| "grad_norm": 5.553383827209473, | |
| "learning_rate": 2.069682085604793e-05, | |
| "loss": 0.7602, | |
| "step": 29900 | |
| }, | |
| { | |
| "epoch": 17.57469244288225, | |
| "grad_norm": 7.833782196044922, | |
| "learning_rate": 2.042694445943758e-05, | |
| "loss": 0.7531, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 17.57469244288225, | |
| "eval_loss": 0.08400005102157593, | |
| "eval_runtime": 147.1645, | |
| "eval_samples_per_second": 3.398, | |
| "eval_steps_per_second": 0.428, | |
| "eval_wer": 0.18670792808723843, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 17.63327475102519, | |
| "grad_norm": 6.075071334838867, | |
| "learning_rate": 2.0157068062827225e-05, | |
| "loss": 0.7471, | |
| "step": 30100 | |
| }, | |
| { | |
| "epoch": 17.691857059168132, | |
| "grad_norm": 6.894543647766113, | |
| "learning_rate": 1.9887191666216872e-05, | |
| "loss": 0.7926, | |
| "step": 30200 | |
| }, | |
| { | |
| "epoch": 17.75043936731107, | |
| "grad_norm": 11.549782752990723, | |
| "learning_rate": 1.9617315269606522e-05, | |
| "loss": 0.7308, | |
| "step": 30300 | |
| }, | |
| { | |
| "epoch": 17.809021675454012, | |
| "grad_norm": 7.361614227294922, | |
| "learning_rate": 1.934743887299617e-05, | |
| "loss": 0.7545, | |
| "step": 30400 | |
| }, | |
| { | |
| "epoch": 17.867603983596954, | |
| "grad_norm": 12.995938301086426, | |
| "learning_rate": 1.9077562476385815e-05, | |
| "loss": 0.7809, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 17.926186291739896, | |
| "grad_norm": 4.978572368621826, | |
| "learning_rate": 1.8807686079775465e-05, | |
| "loss": 0.7587, | |
| "step": 30600 | |
| }, | |
| { | |
| "epoch": 17.984768599882834, | |
| "grad_norm": 6.543401718139648, | |
| "learning_rate": 1.853780968316511e-05, | |
| "loss": 0.7432, | |
| "step": 30700 | |
| }, | |
| { | |
| "epoch": 18.043350908025776, | |
| "grad_norm": 8.726702690124512, | |
| "learning_rate": 1.8267933286554758e-05, | |
| "loss": 0.7305, | |
| "step": 30800 | |
| }, | |
| { | |
| "epoch": 18.101933216168717, | |
| "grad_norm": 6.32004976272583, | |
| "learning_rate": 1.7998056889944405e-05, | |
| "loss": 0.7454, | |
| "step": 30900 | |
| }, | |
| { | |
| "epoch": 18.16051552431166, | |
| "grad_norm": 5.896217346191406, | |
| "learning_rate": 1.7728180493334055e-05, | |
| "loss": 0.7166, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 18.16051552431166, | |
| "eval_loss": 0.0838567316532135, | |
| "eval_runtime": 147.3922, | |
| "eval_samples_per_second": 3.392, | |
| "eval_steps_per_second": 0.427, | |
| "eval_wer": 0.1905393457117595, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 18.219097832454597, | |
| "grad_norm": 6.72420597076416, | |
| "learning_rate": 1.74583040967237e-05, | |
| "loss": 0.7028, | |
| "step": 31100 | |
| }, | |
| { | |
| "epoch": 18.27768014059754, | |
| "grad_norm": 5.8699140548706055, | |
| "learning_rate": 1.7188427700113348e-05, | |
| "loss": 0.7548, | |
| "step": 31200 | |
| }, | |
| { | |
| "epoch": 18.33626244874048, | |
| "grad_norm": 6.714815139770508, | |
| "learning_rate": 1.6921250067469098e-05, | |
| "loss": 0.7619, | |
| "step": 31300 | |
| }, | |
| { | |
| "epoch": 18.394844756883423, | |
| "grad_norm": 9.205123901367188, | |
| "learning_rate": 1.6651373670858748e-05, | |
| "loss": 0.7146, | |
| "step": 31400 | |
| }, | |
| { | |
| "epoch": 18.45342706502636, | |
| "grad_norm": 7.359625339508057, | |
| "learning_rate": 1.6381497274248395e-05, | |
| "loss": 0.7425, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 18.512009373169303, | |
| "grad_norm": 12.034134864807129, | |
| "learning_rate": 1.6114319641604145e-05, | |
| "loss": 0.7325, | |
| "step": 31600 | |
| }, | |
| { | |
| "epoch": 18.570591681312244, | |
| "grad_norm": 12.600486755371094, | |
| "learning_rate": 1.584444324499379e-05, | |
| "loss": 0.7628, | |
| "step": 31700 | |
| }, | |
| { | |
| "epoch": 18.629173989455186, | |
| "grad_norm": 5.2866530418396, | |
| "learning_rate": 1.557456684838344e-05, | |
| "loss": 0.7636, | |
| "step": 31800 | |
| }, | |
| { | |
| "epoch": 18.687756297598124, | |
| "grad_norm": 5.383816719055176, | |
| "learning_rate": 1.5304690451773088e-05, | |
| "loss": 0.727, | |
| "step": 31900 | |
| }, | |
| { | |
| "epoch": 18.746338605741066, | |
| "grad_norm": 8.094857215881348, | |
| "learning_rate": 1.5034814055162735e-05, | |
| "loss": 0.7976, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 18.746338605741066, | |
| "eval_loss": 0.08411071449518204, | |
| "eval_runtime": 145.6438, | |
| "eval_samples_per_second": 3.433, | |
| "eval_steps_per_second": 0.433, | |
| "eval_wer": 0.18376068376068377, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 18.804920913884008, | |
| "grad_norm": 11.049666404724121, | |
| "learning_rate": 1.4764937658552383e-05, | |
| "loss": 0.7523, | |
| "step": 32100 | |
| }, | |
| { | |
| "epoch": 18.86350322202695, | |
| "grad_norm": 5.7121782302856445, | |
| "learning_rate": 1.4495061261942031e-05, | |
| "loss": 0.7462, | |
| "step": 32200 | |
| }, | |
| { | |
| "epoch": 18.922085530169888, | |
| "grad_norm": 6.261909008026123, | |
| "learning_rate": 1.4225184865331678e-05, | |
| "loss": 0.6894, | |
| "step": 32300 | |
| }, | |
| { | |
| "epoch": 18.98066783831283, | |
| "grad_norm": 9.238897323608398, | |
| "learning_rate": 1.3955308468721326e-05, | |
| "loss": 0.7625, | |
| "step": 32400 | |
| }, | |
| { | |
| "epoch": 19.03925014645577, | |
| "grad_norm": 5.913239002227783, | |
| "learning_rate": 1.3685432072110973e-05, | |
| "loss": 0.7324, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 19.097832454598713, | |
| "grad_norm": 6.142359733581543, | |
| "learning_rate": 1.3415555675500621e-05, | |
| "loss": 0.7275, | |
| "step": 32600 | |
| }, | |
| { | |
| "epoch": 19.15641476274165, | |
| "grad_norm": 8.006820678710938, | |
| "learning_rate": 1.314567927889027e-05, | |
| "loss": 0.7335, | |
| "step": 32700 | |
| }, | |
| { | |
| "epoch": 19.214997070884593, | |
| "grad_norm": 10.434711456298828, | |
| "learning_rate": 1.2875802882279916e-05, | |
| "loss": 0.7093, | |
| "step": 32800 | |
| }, | |
| { | |
| "epoch": 19.273579379027534, | |
| "grad_norm": 7.119588375091553, | |
| "learning_rate": 1.2605926485669564e-05, | |
| "loss": 0.7213, | |
| "step": 32900 | |
| }, | |
| { | |
| "epoch": 19.332161687170476, | |
| "grad_norm": 6.306182861328125, | |
| "learning_rate": 1.2336050089059211e-05, | |
| "loss": 0.7008, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 19.332161687170476, | |
| "eval_loss": 0.08348561823368073, | |
| "eval_runtime": 148.0399, | |
| "eval_samples_per_second": 3.377, | |
| "eval_steps_per_second": 0.426, | |
| "eval_wer": 0.18641320365458297, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 19.390743995313414, | |
| "grad_norm": 7.711204528808594, | |
| "learning_rate": 1.206617369244886e-05, | |
| "loss": 0.7326, | |
| "step": 33100 | |
| }, | |
| { | |
| "epoch": 19.449326303456356, | |
| "grad_norm": 6.137068271636963, | |
| "learning_rate": 1.1796297295838508e-05, | |
| "loss": 0.6976, | |
| "step": 33200 | |
| }, | |
| { | |
| "epoch": 19.507908611599298, | |
| "grad_norm": 6.798486709594727, | |
| "learning_rate": 1.1526420899228154e-05, | |
| "loss": 0.7118, | |
| "step": 33300 | |
| }, | |
| { | |
| "epoch": 19.566490919742236, | |
| "grad_norm": 6.276034355163574, | |
| "learning_rate": 1.12565445026178e-05, | |
| "loss": 0.7112, | |
| "step": 33400 | |
| }, | |
| { | |
| "epoch": 19.625073227885178, | |
| "grad_norm": 7.021731853485107, | |
| "learning_rate": 1.0986668106007449e-05, | |
| "loss": 0.6984, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 19.68365553602812, | |
| "grad_norm": 5.709362983703613, | |
| "learning_rate": 1.0716791709397096e-05, | |
| "loss": 0.7376, | |
| "step": 33600 | |
| }, | |
| { | |
| "epoch": 19.74223784417106, | |
| "grad_norm": 7.312982559204102, | |
| "learning_rate": 1.0446915312786744e-05, | |
| "loss": 0.7448, | |
| "step": 33700 | |
| }, | |
| { | |
| "epoch": 19.800820152314, | |
| "grad_norm": 5.710330486297607, | |
| "learning_rate": 1.0177038916176392e-05, | |
| "loss": 0.7524, | |
| "step": 33800 | |
| }, | |
| { | |
| "epoch": 19.85940246045694, | |
| "grad_norm": 6.241232395172119, | |
| "learning_rate": 9.907162519566039e-06, | |
| "loss": 0.6942, | |
| "step": 33900 | |
| }, | |
| { | |
| "epoch": 19.917984768599883, | |
| "grad_norm": 6.5843024253845215, | |
| "learning_rate": 9.637286122955687e-06, | |
| "loss": 0.707, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 19.917984768599883, | |
| "eval_loss": 0.08329460024833679, | |
| "eval_runtime": 148.0997, | |
| "eval_samples_per_second": 3.376, | |
| "eval_steps_per_second": 0.425, | |
| "eval_wer": 0.18715001473622164, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 19.976567076742825, | |
| "grad_norm": 7.8604302406311035, | |
| "learning_rate": 9.367409726345334e-06, | |
| "loss": 0.7127, | |
| "step": 34100 | |
| }, | |
| { | |
| "epoch": 20.035149384885763, | |
| "grad_norm": 6.885805130004883, | |
| "learning_rate": 9.097533329734982e-06, | |
| "loss": 0.6805, | |
| "step": 34200 | |
| }, | |
| { | |
| "epoch": 20.093731693028705, | |
| "grad_norm": 5.695769786834717, | |
| "learning_rate": 8.82765693312463e-06, | |
| "loss": 0.7139, | |
| "step": 34300 | |
| }, | |
| { | |
| "epoch": 20.152314001171646, | |
| "grad_norm": 6.634642124176025, | |
| "learning_rate": 8.557780536514277e-06, | |
| "loss": 0.6917, | |
| "step": 34400 | |
| }, | |
| { | |
| "epoch": 20.210896309314588, | |
| "grad_norm": 5.601979732513428, | |
| "learning_rate": 8.287904139903925e-06, | |
| "loss": 0.6576, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 20.269478617457526, | |
| "grad_norm": 4.387629508972168, | |
| "learning_rate": 8.018027743293572e-06, | |
| "loss": 0.7389, | |
| "step": 34600 | |
| }, | |
| { | |
| "epoch": 20.328060925600468, | |
| "grad_norm": 5.184898853302002, | |
| "learning_rate": 7.748151346683219e-06, | |
| "loss": 0.6803, | |
| "step": 34700 | |
| }, | |
| { | |
| "epoch": 20.38664323374341, | |
| "grad_norm": 9.722779273986816, | |
| "learning_rate": 7.478274950072867e-06, | |
| "loss": 0.7095, | |
| "step": 34800 | |
| }, | |
| { | |
| "epoch": 20.44522554188635, | |
| "grad_norm": 6.964946269989014, | |
| "learning_rate": 7.208398553462514e-06, | |
| "loss": 0.7372, | |
| "step": 34900 | |
| }, | |
| { | |
| "epoch": 20.50380785002929, | |
| "grad_norm": 9.976523399353027, | |
| "learning_rate": 6.938522156852162e-06, | |
| "loss": 0.6865, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 20.50380785002929, | |
| "eval_loss": 0.08354520797729492, | |
| "eval_runtime": 147.4678, | |
| "eval_samples_per_second": 3.391, | |
| "eval_steps_per_second": 0.427, | |
| "eval_wer": 0.1843501326259947, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 20.56239015817223, | |
| "grad_norm": 6.862576007843018, | |
| "learning_rate": 6.668645760241809e-06, | |
| "loss": 0.6942, | |
| "step": 35100 | |
| }, | |
| { | |
| "epoch": 20.620972466315173, | |
| "grad_norm": 10.506976127624512, | |
| "learning_rate": 6.3987693636314575e-06, | |
| "loss": 0.6863, | |
| "step": 35200 | |
| }, | |
| { | |
| "epoch": 20.679554774458115, | |
| "grad_norm": 6.648481845855713, | |
| "learning_rate": 6.128892967021105e-06, | |
| "loss": 0.6987, | |
| "step": 35300 | |
| }, | |
| { | |
| "epoch": 20.738137082601053, | |
| "grad_norm": 7.607233047485352, | |
| "learning_rate": 5.8590165704107524e-06, | |
| "loss": 0.689, | |
| "step": 35400 | |
| }, | |
| { | |
| "epoch": 20.796719390743995, | |
| "grad_norm": 6.110599040985107, | |
| "learning_rate": 5.589140173800399e-06, | |
| "loss": 0.7004, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 20.855301698886937, | |
| "grad_norm": 5.632733345031738, | |
| "learning_rate": 5.319263777190047e-06, | |
| "loss": 0.7299, | |
| "step": 35600 | |
| }, | |
| { | |
| "epoch": 20.91388400702988, | |
| "grad_norm": 5.644909381866455, | |
| "learning_rate": 5.052086144545798e-06, | |
| "loss": 0.6749, | |
| "step": 35700 | |
| }, | |
| { | |
| "epoch": 20.972466315172817, | |
| "grad_norm": 5.2816386222839355, | |
| "learning_rate": 4.782209747935446e-06, | |
| "loss": 0.671, | |
| "step": 35800 | |
| }, | |
| { | |
| "epoch": 21.03104862331576, | |
| "grad_norm": 4.735965728759766, | |
| "learning_rate": 4.512333351325093e-06, | |
| "loss": 0.7137, | |
| "step": 35900 | |
| }, | |
| { | |
| "epoch": 21.0896309314587, | |
| "grad_norm": 6.491783618927002, | |
| "learning_rate": 4.242456954714741e-06, | |
| "loss": 0.6927, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 21.0896309314587, | |
| "eval_loss": 0.08344998210668564, | |
| "eval_runtime": 147.4098, | |
| "eval_samples_per_second": 3.392, | |
| "eval_steps_per_second": 0.427, | |
| "eval_wer": 0.18818155025051578, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 21.148213239601642, | |
| "grad_norm": 6.068084239959717, | |
| "learning_rate": 3.972580558104388e-06, | |
| "loss": 0.7155, | |
| "step": 36100 | |
| }, | |
| { | |
| "epoch": 21.20679554774458, | |
| "grad_norm": 4.509976863861084, | |
| "learning_rate": 3.702704161494036e-06, | |
| "loss": 0.6615, | |
| "step": 36200 | |
| }, | |
| { | |
| "epoch": 21.26537785588752, | |
| "grad_norm": 6.346072673797607, | |
| "learning_rate": 3.4328277648836835e-06, | |
| "loss": 0.6682, | |
| "step": 36300 | |
| }, | |
| { | |
| "epoch": 21.323960164030463, | |
| "grad_norm": 6.095388889312744, | |
| "learning_rate": 3.1629513682733306e-06, | |
| "loss": 0.6588, | |
| "step": 36400 | |
| }, | |
| { | |
| "epoch": 21.382542472173405, | |
| "grad_norm": 5.079260349273682, | |
| "learning_rate": 2.8930749716629784e-06, | |
| "loss": 0.67, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 21.441124780316343, | |
| "grad_norm": 4.323718547821045, | |
| "learning_rate": 2.623198575052626e-06, | |
| "loss": 0.6606, | |
| "step": 36600 | |
| }, | |
| { | |
| "epoch": 21.499707088459285, | |
| "grad_norm": 5.659894943237305, | |
| "learning_rate": 2.3533221784422738e-06, | |
| "loss": 0.69, | |
| "step": 36700 | |
| }, | |
| { | |
| "epoch": 21.558289396602227, | |
| "grad_norm": 7.08272647857666, | |
| "learning_rate": 2.083445781831921e-06, | |
| "loss": 0.6506, | |
| "step": 36800 | |
| }, | |
| { | |
| "epoch": 21.61687170474517, | |
| "grad_norm": 4.9979119300842285, | |
| "learning_rate": 1.8135693852215685e-06, | |
| "loss": 0.7215, | |
| "step": 36900 | |
| }, | |
| { | |
| "epoch": 21.675454012888107, | |
| "grad_norm": 6.872857093811035, | |
| "learning_rate": 1.5436929886112161e-06, | |
| "loss": 0.7014, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 21.675454012888107, | |
| "eval_loss": 0.08349551260471344, | |
| "eval_runtime": 147.5946, | |
| "eval_samples_per_second": 3.388, | |
| "eval_steps_per_second": 0.427, | |
| "eval_wer": 0.1861184792219275, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 21.73403632103105, | |
| "grad_norm": 6.6599273681640625, | |
| "learning_rate": 1.2738165920008636e-06, | |
| "loss": 0.6701, | |
| "step": 37100 | |
| }, | |
| { | |
| "epoch": 21.79261862917399, | |
| "grad_norm": 4.662840366363525, | |
| "learning_rate": 1.0039401953905112e-06, | |
| "loss": 0.7006, | |
| "step": 37200 | |
| }, | |
| { | |
| "epoch": 21.851200937316932, | |
| "grad_norm": 6.501893043518066, | |
| "learning_rate": 7.340637987801587e-07, | |
| "loss": 0.66, | |
| "step": 37300 | |
| }, | |
| { | |
| "epoch": 21.90978324545987, | |
| "grad_norm": 5.708708763122559, | |
| "learning_rate": 4.6418740216980626e-07, | |
| "loss": 0.6819, | |
| "step": 37400 | |
| }, | |
| { | |
| "epoch": 21.968365553602812, | |
| "grad_norm": 4.885653018951416, | |
| "learning_rate": 1.943110055594538e-07, | |
| "loss": 0.7, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 22.026947861745754, | |
| "grad_norm": 5.802590370178223, | |
| "learning_rate": 1.5496103185816508e-05, | |
| "loss": 0.7875, | |
| "step": 37600 | |
| }, | |
| { | |
| "epoch": 22.085530169888692, | |
| "grad_norm": 6.833991527557373, | |
| "learning_rate": 1.5268219315436854e-05, | |
| "loss": 0.6936, | |
| "step": 37700 | |
| }, | |
| { | |
| "epoch": 22.144112478031634, | |
| "grad_norm": 9.980413436889648, | |
| "learning_rate": 1.5042614283760998e-05, | |
| "loss": 0.6935, | |
| "step": 37800 | |
| }, | |
| { | |
| "epoch": 22.202694786174575, | |
| "grad_norm": 6.407867908477783, | |
| "learning_rate": 1.4814730413381343e-05, | |
| "loss": 0.669, | |
| "step": 37900 | |
| }, | |
| { | |
| "epoch": 22.261277094317517, | |
| "grad_norm": 7.097990989685059, | |
| "learning_rate": 1.4586846543001687e-05, | |
| "loss": 0.6951, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 22.261277094317517, | |
| "eval_loss": 0.08332780748605728, | |
| "eval_runtime": 147.6911, | |
| "eval_samples_per_second": 3.385, | |
| "eval_steps_per_second": 0.427, | |
| "eval_wer": 0.1874447391688771, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 22.319859402460455, | |
| "grad_norm": 7.611706733703613, | |
| "learning_rate": 1.4358962672622033e-05, | |
| "loss": 0.6751, | |
| "step": 38100 | |
| }, | |
| { | |
| "epoch": 22.378441710603397, | |
| "grad_norm": 11.156392097473145, | |
| "learning_rate": 1.4131078802242378e-05, | |
| "loss": 0.7132, | |
| "step": 38200 | |
| }, | |
| { | |
| "epoch": 22.43702401874634, | |
| "grad_norm": 7.024628639221191, | |
| "learning_rate": 1.3903194931862724e-05, | |
| "loss": 0.7599, | |
| "step": 38300 | |
| }, | |
| { | |
| "epoch": 22.49560632688928, | |
| "grad_norm": 6.373144626617432, | |
| "learning_rate": 1.3675311061483068e-05, | |
| "loss": 0.6951, | |
| "step": 38400 | |
| }, | |
| { | |
| "epoch": 22.55418863503222, | |
| "grad_norm": 5.750803470611572, | |
| "learning_rate": 1.3447427191103415e-05, | |
| "loss": 0.7078, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 22.61277094317516, | |
| "grad_norm": 6.302127361297607, | |
| "learning_rate": 1.3219543320723759e-05, | |
| "loss": 0.7153, | |
| "step": 38600 | |
| }, | |
| { | |
| "epoch": 22.671353251318102, | |
| "grad_norm": 6.580202102661133, | |
| "learning_rate": 1.2991659450344103e-05, | |
| "loss": 0.6923, | |
| "step": 38700 | |
| }, | |
| { | |
| "epoch": 22.729935559461044, | |
| "grad_norm": 5.348507404327393, | |
| "learning_rate": 1.2763775579964451e-05, | |
| "loss": 0.6998, | |
| "step": 38800 | |
| }, | |
| { | |
| "epoch": 22.788517867603982, | |
| "grad_norm": 7.927694797515869, | |
| "learning_rate": 1.2535891709584796e-05, | |
| "loss": 0.7093, | |
| "step": 38900 | |
| }, | |
| { | |
| "epoch": 22.847100175746924, | |
| "grad_norm": 4.928946495056152, | |
| "learning_rate": 1.2308007839205142e-05, | |
| "loss": 0.6848, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 22.847100175746924, | |
| "eval_loss": 0.08338670432567596, | |
| "eval_runtime": 146.104, | |
| "eval_samples_per_second": 3.422, | |
| "eval_steps_per_second": 0.431, | |
| "eval_wer": 0.1927497789566755, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 22.905682483889866, | |
| "grad_norm": 9.807655334472656, | |
| "learning_rate": 1.2080123968825486e-05, | |
| "loss": 0.6954, | |
| "step": 39100 | |
| }, | |
| { | |
| "epoch": 22.964264792032807, | |
| "grad_norm": 4.682498931884766, | |
| "learning_rate": 1.1852240098445833e-05, | |
| "loss": 0.7273, | |
| "step": 39200 | |
| }, | |
| { | |
| "epoch": 23.022847100175746, | |
| "grad_norm": 6.8382086753845215, | |
| "learning_rate": 1.1624356228066179e-05, | |
| "loss": 0.721, | |
| "step": 39300 | |
| }, | |
| { | |
| "epoch": 23.081429408318687, | |
| "grad_norm": 7.560245037078857, | |
| "learning_rate": 1.1396472357686523e-05, | |
| "loss": 0.689, | |
| "step": 39400 | |
| }, | |
| { | |
| "epoch": 23.14001171646163, | |
| "grad_norm": 7.024509906768799, | |
| "learning_rate": 1.116858848730687e-05, | |
| "loss": 0.6714, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 23.19859402460457, | |
| "grad_norm": 8.46679973602295, | |
| "learning_rate": 1.0940704616927216e-05, | |
| "loss": 0.6729, | |
| "step": 39600 | |
| }, | |
| { | |
| "epoch": 23.25717633274751, | |
| "grad_norm": 7.632561683654785, | |
| "learning_rate": 1.071282074654756e-05, | |
| "loss": 0.7032, | |
| "step": 39700 | |
| }, | |
| { | |
| "epoch": 23.31575864089045, | |
| "grad_norm": 9.298177719116211, | |
| "learning_rate": 1.0484936876167906e-05, | |
| "loss": 0.6915, | |
| "step": 39800 | |
| }, | |
| { | |
| "epoch": 23.374340949033392, | |
| "grad_norm": 6.186316013336182, | |
| "learning_rate": 1.0259331844492047e-05, | |
| "loss": 0.6625, | |
| "step": 39900 | |
| }, | |
| { | |
| "epoch": 23.432923257176334, | |
| "grad_norm": 4.553198337554932, | |
| "learning_rate": 1.0031447974112393e-05, | |
| "loss": 0.7096, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 23.432923257176334, | |
| "eval_loss": 0.08342915773391724, | |
| "eval_runtime": 146.9046, | |
| "eval_samples_per_second": 3.404, | |
| "eval_steps_per_second": 0.429, | |
| "eval_wer": 0.19363395225464192, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 23.491505565319272, | |
| "grad_norm": 5.817235946655273, | |
| "learning_rate": 9.803564103732739e-06, | |
| "loss": 0.6733, | |
| "step": 40100 | |
| }, | |
| { | |
| "epoch": 23.550087873462214, | |
| "grad_norm": 5.511366844177246, | |
| "learning_rate": 9.575680233353083e-06, | |
| "loss": 0.728, | |
| "step": 40200 | |
| }, | |
| { | |
| "epoch": 23.608670181605156, | |
| "grad_norm": 7.099341869354248, | |
| "learning_rate": 9.34779636297343e-06, | |
| "loss": 0.7007, | |
| "step": 40300 | |
| }, | |
| { | |
| "epoch": 23.667252489748098, | |
| "grad_norm": 4.963494777679443, | |
| "learning_rate": 9.119912492593774e-06, | |
| "loss": 0.6804, | |
| "step": 40400 | |
| }, | |
| { | |
| "epoch": 23.725834797891036, | |
| "grad_norm": 7.063295364379883, | |
| "learning_rate": 8.89202862221412e-06, | |
| "loss": 0.7092, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 23.784417106033978, | |
| "grad_norm": 6.879157543182373, | |
| "learning_rate": 8.664144751834466e-06, | |
| "loss": 0.6753, | |
| "step": 40600 | |
| }, | |
| { | |
| "epoch": 23.84299941417692, | |
| "grad_norm": 5.31470251083374, | |
| "learning_rate": 8.43626088145481e-06, | |
| "loss": 0.6876, | |
| "step": 40700 | |
| }, | |
| { | |
| "epoch": 23.90158172231986, | |
| "grad_norm": 6.351736068725586, | |
| "learning_rate": 8.208377011075157e-06, | |
| "loss": 0.6949, | |
| "step": 40800 | |
| }, | |
| { | |
| "epoch": 23.9601640304628, | |
| "grad_norm": 5.326944351196289, | |
| "learning_rate": 7.980493140695503e-06, | |
| "loss": 0.705, | |
| "step": 40900 | |
| }, | |
| { | |
| "epoch": 24.01874633860574, | |
| "grad_norm": 6.150857448577881, | |
| "learning_rate": 7.752609270315848e-06, | |
| "loss": 0.6952, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 24.01874633860574, | |
| "eval_loss": 0.08346738666296005, | |
| "eval_runtime": 145.9289, | |
| "eval_samples_per_second": 3.426, | |
| "eval_steps_per_second": 0.432, | |
| "eval_wer": 0.19333922782198645, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 24.077328646748683, | |
| "grad_norm": 5.961891174316406, | |
| "learning_rate": 7.524725399936193e-06, | |
| "loss": 0.6498, | |
| "step": 41100 | |
| }, | |
| { | |
| "epoch": 24.135910954891624, | |
| "grad_norm": 4.94498348236084, | |
| "learning_rate": 7.296841529556539e-06, | |
| "loss": 0.6827, | |
| "step": 41200 | |
| }, | |
| { | |
| "epoch": 24.194493263034563, | |
| "grad_norm": 6.914667129516602, | |
| "learning_rate": 7.068957659176884e-06, | |
| "loss": 0.6647, | |
| "step": 41300 | |
| }, | |
| { | |
| "epoch": 24.253075571177504, | |
| "grad_norm": 5.37884521484375, | |
| "learning_rate": 6.84107378879723e-06, | |
| "loss": 0.6447, | |
| "step": 41400 | |
| }, | |
| { | |
| "epoch": 24.311657879320446, | |
| "grad_norm": 7.762170791625977, | |
| "learning_rate": 6.613189918417574e-06, | |
| "loss": 0.6362, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 24.370240187463388, | |
| "grad_norm": 5.311378479003906, | |
| "learning_rate": 6.38530604803792e-06, | |
| "loss": 0.7053, | |
| "step": 41600 | |
| }, | |
| { | |
| "epoch": 24.428822495606326, | |
| "grad_norm": 5.185734748840332, | |
| "learning_rate": 6.157422177658266e-06, | |
| "loss": 0.6724, | |
| "step": 41700 | |
| }, | |
| { | |
| "epoch": 24.487404803749268, | |
| "grad_norm": 6.12027645111084, | |
| "learning_rate": 5.929538307278611e-06, | |
| "loss": 0.6947, | |
| "step": 41800 | |
| }, | |
| { | |
| "epoch": 24.54598711189221, | |
| "grad_norm": 11.165681838989258, | |
| "learning_rate": 5.701654436898956e-06, | |
| "loss": 0.6999, | |
| "step": 41900 | |
| }, | |
| { | |
| "epoch": 24.604569420035148, | |
| "grad_norm": 6.521625995635986, | |
| "learning_rate": 5.473770566519302e-06, | |
| "loss": 0.692, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 24.604569420035148, | |
| "eval_loss": 0.0832965150475502, | |
| "eval_runtime": 145.9791, | |
| "eval_samples_per_second": 3.425, | |
| "eval_steps_per_second": 0.432, | |
| "eval_wer": 0.19304450338933096, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 24.66315172817809, | |
| "grad_norm": 7.7173333168029785, | |
| "learning_rate": 5.248165534843444e-06, | |
| "loss": 0.6641, | |
| "step": 42100 | |
| }, | |
| { | |
| "epoch": 24.72173403632103, | |
| "grad_norm": 5.125652313232422, | |
| "learning_rate": 5.02028166446379e-06, | |
| "loss": 0.6576, | |
| "step": 42200 | |
| }, | |
| { | |
| "epoch": 24.780316344463973, | |
| "grad_norm": 5.99462366104126, | |
| "learning_rate": 4.792397794084135e-06, | |
| "loss": 0.6733, | |
| "step": 42300 | |
| }, | |
| { | |
| "epoch": 24.83889865260691, | |
| "grad_norm": 5.6660614013671875, | |
| "learning_rate": 4.5645139237044806e-06, | |
| "loss": 0.7033, | |
| "step": 42400 | |
| }, | |
| { | |
| "epoch": 24.897480960749853, | |
| "grad_norm": 7.414560317993164, | |
| "learning_rate": 4.336630053324826e-06, | |
| "loss": 0.6756, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 24.956063268892795, | |
| "grad_norm": 6.082986354827881, | |
| "learning_rate": 4.108746182945172e-06, | |
| "loss": 0.6713, | |
| "step": 42600 | |
| }, | |
| { | |
| "epoch": 25.014645577035736, | |
| "grad_norm": 5.000583648681641, | |
| "learning_rate": 3.8808623125655165e-06, | |
| "loss": 0.6821, | |
| "step": 42700 | |
| }, | |
| { | |
| "epoch": 25.073227885178675, | |
| "grad_norm": 6.196038246154785, | |
| "learning_rate": 3.6529784421858623e-06, | |
| "loss": 0.6795, | |
| "step": 42800 | |
| }, | |
| { | |
| "epoch": 25.131810193321616, | |
| "grad_norm": 7.409509181976318, | |
| "learning_rate": 3.4250945718062076e-06, | |
| "loss": 0.6388, | |
| "step": 42900 | |
| }, | |
| { | |
| "epoch": 25.190392501464558, | |
| "grad_norm": 6.713177680969238, | |
| "learning_rate": 3.1972107014265533e-06, | |
| "loss": 0.6552, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 25.190392501464558, | |
| "eval_loss": 0.0831432044506073, | |
| "eval_runtime": 144.9356, | |
| "eval_samples_per_second": 3.45, | |
| "eval_steps_per_second": 0.435, | |
| "eval_wer": 0.18670792808723843, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 25.2489748096075, | |
| "grad_norm": 4.68192720413208, | |
| "learning_rate": 2.9693268310468986e-06, | |
| "loss": 0.6323, | |
| "step": 43100 | |
| }, | |
| { | |
| "epoch": 25.307557117750438, | |
| "grad_norm": 4.701879501342773, | |
| "learning_rate": 2.741442960667244e-06, | |
| "loss": 0.6499, | |
| "step": 43200 | |
| }, | |
| { | |
| "epoch": 25.36613942589338, | |
| "grad_norm": 5.518495559692383, | |
| "learning_rate": 2.5135590902875897e-06, | |
| "loss": 0.6513, | |
| "step": 43300 | |
| }, | |
| { | |
| "epoch": 25.42472173403632, | |
| "grad_norm": 4.257356643676758, | |
| "learning_rate": 2.285675219907935e-06, | |
| "loss": 0.6957, | |
| "step": 43400 | |
| }, | |
| { | |
| "epoch": 25.483304042179263, | |
| "grad_norm": 4.883972644805908, | |
| "learning_rate": 2.0577913495282803e-06, | |
| "loss": 0.6552, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 25.5418863503222, | |
| "grad_norm": 4.6188201904296875, | |
| "learning_rate": 1.8299074791486259e-06, | |
| "loss": 0.6683, | |
| "step": 43600 | |
| }, | |
| { | |
| "epoch": 25.600468658465143, | |
| "grad_norm": 8.4751558303833, | |
| "learning_rate": 1.6020236087689714e-06, | |
| "loss": 0.636, | |
| "step": 43700 | |
| }, | |
| { | |
| "epoch": 25.659050966608085, | |
| "grad_norm": 5.713025093078613, | |
| "learning_rate": 1.374139738389317e-06, | |
| "loss": 0.6833, | |
| "step": 43800 | |
| }, | |
| { | |
| "epoch": 25.717633274751027, | |
| "grad_norm": Infinity, | |
| "learning_rate": 1.1462558680096623e-06, | |
| "loss": 0.6807, | |
| "step": 43900 | |
| }, | |
| { | |
| "epoch": 25.776215582893965, | |
| "grad_norm": 7.257518291473389, | |
| "learning_rate": 9.206508363338042e-07, | |
| "loss": 0.6641, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 25.776215582893965, | |
| "eval_loss": 0.0832269936800003, | |
| "eval_runtime": 145.3966, | |
| "eval_samples_per_second": 3.439, | |
| "eval_steps_per_second": 0.433, | |
| "eval_wer": 0.1874447391688771, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 25.834797891036906, | |
| "grad_norm": 5.92907190322876, | |
| "learning_rate": 6.927669659541498e-07, | |
| "loss": 0.6462, | |
| "step": 44100 | |
| }, | |
| { | |
| "epoch": 25.89338019917985, | |
| "grad_norm": 4.766348361968994, | |
| "learning_rate": 4.6488309557449525e-07, | |
| "loss": 0.6934, | |
| "step": 44200 | |
| }, | |
| { | |
| "epoch": 25.95196250732279, | |
| "grad_norm": 6.461400032043457, | |
| "learning_rate": 2.3699922519484072e-07, | |
| "loss": 0.6662, | |
| "step": 44300 | |
| }, | |
| { | |
| "epoch": 26.010544815465728, | |
| "grad_norm": 5.860207557678223, | |
| "learning_rate": 1.3472687832774603e-05, | |
| "loss": 0.6097, | |
| "step": 44400 | |
| }, | |
| { | |
| "epoch": 26.06912712360867, | |
| "grad_norm": 5.674122333526611, | |
| "learning_rate": 1.3275488069414319e-05, | |
| "loss": 0.6638, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 26.12770943175161, | |
| "grad_norm": 7.24979829788208, | |
| "learning_rate": 1.3078288306054034e-05, | |
| "loss": 0.641, | |
| "step": 44600 | |
| }, | |
| { | |
| "epoch": 26.186291739894553, | |
| "grad_norm": 7.524514198303223, | |
| "learning_rate": 1.288108854269375e-05, | |
| "loss": 0.6722, | |
| "step": 44700 | |
| }, | |
| { | |
| "epoch": 26.24487404803749, | |
| "grad_norm": 5.542699337005615, | |
| "learning_rate": 1.2683888779333466e-05, | |
| "loss": 0.6386, | |
| "step": 44800 | |
| }, | |
| { | |
| "epoch": 26.303456356180433, | |
| "grad_norm": 7.757282257080078, | |
| "learning_rate": 1.2486689015973181e-05, | |
| "loss": 0.6976, | |
| "step": 44900 | |
| }, | |
| { | |
| "epoch": 26.362038664323375, | |
| "grad_norm": 5.828668117523193, | |
| "learning_rate": 1.2289489252612897e-05, | |
| "loss": 0.6921, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 26.362038664323375, | |
| "eval_loss": 0.08329325169324875, | |
| "eval_runtime": 154.0108, | |
| "eval_samples_per_second": 3.247, | |
| "eval_steps_per_second": 0.409, | |
| "eval_wer": 0.18803418803418803, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 26.420620972466317, | |
| "grad_norm": 6.481083869934082, | |
| "learning_rate": 1.2092289489252613e-05, | |
| "loss": 0.6643, | |
| "step": 45100 | |
| }, | |
| { | |
| "epoch": 26.479203280609255, | |
| "grad_norm": 6.508504390716553, | |
| "learning_rate": 1.1895089725892329e-05, | |
| "loss": 0.6986, | |
| "step": 45200 | |
| }, | |
| { | |
| "epoch": 26.537785588752197, | |
| "grad_norm": 5.0908894538879395, | |
| "learning_rate": 1.1697889962532044e-05, | |
| "loss": 0.6741, | |
| "step": 45300 | |
| }, | |
| { | |
| "epoch": 26.59636789689514, | |
| "grad_norm": 6.918355464935303, | |
| "learning_rate": 1.1500690199171762e-05, | |
| "loss": 0.6774, | |
| "step": 45400 | |
| }, | |
| { | |
| "epoch": 26.65495020503808, | |
| "grad_norm": 5.500540733337402, | |
| "learning_rate": 1.1303490435811477e-05, | |
| "loss": 0.6861, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 26.71353251318102, | |
| "grad_norm": 5.732079029083252, | |
| "learning_rate": 1.1106290672451193e-05, | |
| "loss": 0.6836, | |
| "step": 45600 | |
| }, | |
| { | |
| "epoch": 26.77211482132396, | |
| "grad_norm": 7.091628074645996, | |
| "learning_rate": 1.0909090909090909e-05, | |
| "loss": 0.7212, | |
| "step": 45700 | |
| }, | |
| { | |
| "epoch": 26.830697129466902, | |
| "grad_norm": 6.164028644561768, | |
| "learning_rate": 1.0711891145730625e-05, | |
| "loss": 0.6832, | |
| "step": 45800 | |
| }, | |
| { | |
| "epoch": 26.889279437609844, | |
| "grad_norm": 6.679256916046143, | |
| "learning_rate": 1.051469138237034e-05, | |
| "loss": 0.6886, | |
| "step": 45900 | |
| }, | |
| { | |
| "epoch": 26.947861745752782, | |
| "grad_norm": 6.822471618652344, | |
| "learning_rate": 1.0317491619010058e-05, | |
| "loss": 0.6894, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 26.947861745752782, | |
| "eval_loss": 0.08317731320858002, | |
| "eval_runtime": 148.6991, | |
| "eval_samples_per_second": 3.362, | |
| "eval_steps_per_second": 0.424, | |
| "eval_wer": 0.18552903035661655, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 27.006444053895724, | |
| "grad_norm": 5.017155170440674, | |
| "learning_rate": 1.0120291855649773e-05, | |
| "loss": 0.7585, | |
| "step": 46100 | |
| }, | |
| { | |
| "epoch": 27.065026362038665, | |
| "grad_norm": 7.552450180053711, | |
| "learning_rate": 9.92309209228949e-06, | |
| "loss": 0.6377, | |
| "step": 46200 | |
| }, | |
| { | |
| "epoch": 27.123608670181603, | |
| "grad_norm": 5.625948905944824, | |
| "learning_rate": 9.725892328929207e-06, | |
| "loss": 0.6606, | |
| "step": 46300 | |
| }, | |
| { | |
| "epoch": 27.182190978324545, | |
| "grad_norm": 6.0872039794921875, | |
| "learning_rate": 9.528692565568922e-06, | |
| "loss": 0.6853, | |
| "step": 46400 | |
| }, | |
| { | |
| "epoch": 27.240773286467487, | |
| "grad_norm": 7.785630226135254, | |
| "learning_rate": 9.331492802208638e-06, | |
| "loss": 0.6862, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 27.29935559461043, | |
| "grad_norm": 7.472136974334717, | |
| "learning_rate": 9.134293038848354e-06, | |
| "loss": 0.6528, | |
| "step": 46600 | |
| }, | |
| { | |
| "epoch": 27.357937902753367, | |
| "grad_norm": 5.214349269866943, | |
| "learning_rate": 8.937093275488071e-06, | |
| "loss": 0.6734, | |
| "step": 46700 | |
| }, | |
| { | |
| "epoch": 27.41652021089631, | |
| "grad_norm": 5.109072208404541, | |
| "learning_rate": 8.739893512127787e-06, | |
| "loss": 0.6936, | |
| "step": 46800 | |
| }, | |
| { | |
| "epoch": 27.47510251903925, | |
| "grad_norm": 5.236739158630371, | |
| "learning_rate": 8.542693748767503e-06, | |
| "loss": 0.6711, | |
| "step": 46900 | |
| }, | |
| { | |
| "epoch": 27.533684827182192, | |
| "grad_norm": 6.8590264320373535, | |
| "learning_rate": 8.345493985407218e-06, | |
| "loss": 0.7041, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 27.533684827182192, | |
| "eval_loss": 0.08270065486431122, | |
| "eval_runtime": 146.6093, | |
| "eval_samples_per_second": 3.41, | |
| "eval_steps_per_second": 0.43, | |
| "eval_wer": 0.18552903035661655, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 27.59226713532513, | |
| "grad_norm": 6.453911781311035, | |
| "learning_rate": 8.148294222046934e-06, | |
| "loss": 0.7099, | |
| "step": 47100 | |
| }, | |
| { | |
| "epoch": 27.650849443468072, | |
| "grad_norm": 9.15201473236084, | |
| "learning_rate": 7.95109445868665e-06, | |
| "loss": 0.6753, | |
| "step": 47200 | |
| }, | |
| { | |
| "epoch": 27.709431751611014, | |
| "grad_norm": 10.103302955627441, | |
| "learning_rate": 7.753894695326366e-06, | |
| "loss": 0.615, | |
| "step": 47300 | |
| }, | |
| { | |
| "epoch": 27.768014059753956, | |
| "grad_norm": 6.687941551208496, | |
| "learning_rate": 7.556694931966082e-06, | |
| "loss": 0.6812, | |
| "step": 47400 | |
| }, | |
| { | |
| "epoch": 27.826596367896894, | |
| "grad_norm": 6.050097942352295, | |
| "learning_rate": 7.359495168605799e-06, | |
| "loss": 0.6329, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 27.885178676039835, | |
| "grad_norm": 6.162942886352539, | |
| "learning_rate": 7.162295405245514e-06, | |
| "loss": 0.6429, | |
| "step": 47600 | |
| }, | |
| { | |
| "epoch": 27.943760984182777, | |
| "grad_norm": 5.811770915985107, | |
| "learning_rate": 6.96509564188523e-06, | |
| "loss": 0.7052, | |
| "step": 47700 | |
| }, | |
| { | |
| "epoch": 28.00234329232572, | |
| "grad_norm": 4.816370010375977, | |
| "learning_rate": 6.767895878524947e-06, | |
| "loss": 0.6578, | |
| "step": 47800 | |
| }, | |
| { | |
| "epoch": 28.060925600468657, | |
| "grad_norm": 4.412487506866455, | |
| "learning_rate": 6.570696115164662e-06, | |
| "loss": 0.6302, | |
| "step": 47900 | |
| }, | |
| { | |
| "epoch": 28.1195079086116, | |
| "grad_norm": 5.892630100250244, | |
| "learning_rate": 6.373496351804378e-06, | |
| "loss": 0.6452, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 28.1195079086116, | |
| "eval_loss": 0.08297573775053024, | |
| "eval_runtime": 146.2177, | |
| "eval_samples_per_second": 3.42, | |
| "eval_steps_per_second": 0.431, | |
| "eval_wer": 0.18818155025051578, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 28.17809021675454, | |
| "grad_norm": 4.816265106201172, | |
| "learning_rate": 6.1782685860776975e-06, | |
| "loss": 0.6636, | |
| "step": 48100 | |
| }, | |
| { | |
| "epoch": 28.236672524897482, | |
| "grad_norm": 6.337242126464844, | |
| "learning_rate": 5.981068822717413e-06, | |
| "loss": 0.6501, | |
| "step": 48200 | |
| }, | |
| { | |
| "epoch": 28.29525483304042, | |
| "grad_norm": 7.430861473083496, | |
| "learning_rate": 5.783869059357129e-06, | |
| "loss": 0.6455, | |
| "step": 48300 | |
| }, | |
| { | |
| "epoch": 28.353837141183362, | |
| "grad_norm": 10.299432754516602, | |
| "learning_rate": 5.586669295996845e-06, | |
| "loss": 0.6569, | |
| "step": 48400 | |
| }, | |
| { | |
| "epoch": 28.412419449326304, | |
| "grad_norm": 5.967062950134277, | |
| "learning_rate": 5.389469532636561e-06, | |
| "loss": 0.651, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 28.471001757469246, | |
| "grad_norm": 5.434126853942871, | |
| "learning_rate": 5.192269769276277e-06, | |
| "loss": 0.6563, | |
| "step": 48600 | |
| }, | |
| { | |
| "epoch": 28.529584065612184, | |
| "grad_norm": 4.6583075523376465, | |
| "learning_rate": 4.995070005915993e-06, | |
| "loss": 0.6512, | |
| "step": 48700 | |
| }, | |
| { | |
| "epoch": 28.588166373755126, | |
| "grad_norm": 6.697512149810791, | |
| "learning_rate": 4.797870242555709e-06, | |
| "loss": 0.664, | |
| "step": 48800 | |
| }, | |
| { | |
| "epoch": 28.646748681898067, | |
| "grad_norm": 8.926551818847656, | |
| "learning_rate": 4.600670479195425e-06, | |
| "loss": 0.6354, | |
| "step": 48900 | |
| }, | |
| { | |
| "epoch": 28.70533099004101, | |
| "grad_norm": 5.536263942718506, | |
| "learning_rate": 4.403470715835141e-06, | |
| "loss": 0.6682, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 28.70533099004101, | |
| "eval_loss": 0.08284977823495865, | |
| "eval_runtime": 146.4815, | |
| "eval_samples_per_second": 3.413, | |
| "eval_steps_per_second": 0.43, | |
| "eval_wer": 0.18626584143825523, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 28.763913298183947, | |
| "grad_norm": 8.664471626281738, | |
| "learning_rate": 4.206270952474857e-06, | |
| "loss": 0.6994, | |
| "step": 49100 | |
| }, | |
| { | |
| "epoch": 28.82249560632689, | |
| "grad_norm": 6.604372978210449, | |
| "learning_rate": 4.009071189114573e-06, | |
| "loss": 0.6788, | |
| "step": 49200 | |
| }, | |
| { | |
| "epoch": 28.88107791446983, | |
| "grad_norm": 5.231504917144775, | |
| "learning_rate": 3.811871425754289e-06, | |
| "loss": 0.6644, | |
| "step": 49300 | |
| }, | |
| { | |
| "epoch": 28.939660222612773, | |
| "grad_norm": 5.522902011871338, | |
| "learning_rate": 3.6146716623940048e-06, | |
| "loss": 0.6543, | |
| "step": 49400 | |
| }, | |
| { | |
| "epoch": 28.99824253075571, | |
| "grad_norm": 4.572765350341797, | |
| "learning_rate": 3.4174718990337218e-06, | |
| "loss": 0.6769, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 29.056824838898653, | |
| "grad_norm": 5.370816707611084, | |
| "learning_rate": 3.2202721356734375e-06, | |
| "loss": 0.6169, | |
| "step": 49600 | |
| }, | |
| { | |
| "epoch": 29.115407147041594, | |
| "grad_norm": 4.599956035614014, | |
| "learning_rate": 3.0230723723131532e-06, | |
| "loss": 0.6698, | |
| "step": 49700 | |
| }, | |
| { | |
| "epoch": 29.173989455184536, | |
| "grad_norm": 4.932507514953613, | |
| "learning_rate": 2.8258726089528694e-06, | |
| "loss": 0.6294, | |
| "step": 49800 | |
| }, | |
| { | |
| "epoch": 29.232571763327474, | |
| "grad_norm": 8.667673110961914, | |
| "learning_rate": 2.628672845592585e-06, | |
| "loss": 0.6688, | |
| "step": 49900 | |
| }, | |
| { | |
| "epoch": 29.291154071470416, | |
| "grad_norm": 4.365273952484131, | |
| "learning_rate": 2.4314730822323017e-06, | |
| "loss": 0.6357, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 29.291154071470416, | |
| "eval_loss": 0.0829482451081276, | |
| "eval_runtime": 146.3076, | |
| "eval_samples_per_second": 3.417, | |
| "eval_steps_per_second": 0.431, | |
| "eval_wer": 0.18773946360153257, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 29.349736379613358, | |
| "grad_norm": 5.5835652351379395, | |
| "learning_rate": 2.234273318872018e-06, | |
| "loss": 0.633, | |
| "step": 50100 | |
| }, | |
| { | |
| "epoch": 29.408318687756296, | |
| "grad_norm": 5.577920436859131, | |
| "learning_rate": 2.0390455531453363e-06, | |
| "loss": 0.6456, | |
| "step": 50200 | |
| }, | |
| { | |
| "epoch": 29.466900995899238, | |
| "grad_norm": 6.7036566734313965, | |
| "learning_rate": 1.8418457897850522e-06, | |
| "loss": 0.6582, | |
| "step": 50300 | |
| }, | |
| { | |
| "epoch": 29.52548330404218, | |
| "grad_norm": 8.638516426086426, | |
| "learning_rate": 1.6446460264247684e-06, | |
| "loss": 0.6469, | |
| "step": 50400 | |
| }, | |
| { | |
| "epoch": 29.58406561218512, | |
| "grad_norm": 7.847275257110596, | |
| "learning_rate": 1.4474462630644845e-06, | |
| "loss": 0.6314, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 29.64264792032806, | |
| "grad_norm": 4.795884609222412, | |
| "learning_rate": 1.2502464997042004e-06, | |
| "loss": 0.6355, | |
| "step": 50600 | |
| }, | |
| { | |
| "epoch": 29.701230228471, | |
| "grad_norm": 5.026218414306641, | |
| "learning_rate": 1.0530467363439164e-06, | |
| "loss": 0.6476, | |
| "step": 50700 | |
| }, | |
| { | |
| "epoch": 29.759812536613943, | |
| "grad_norm": 6.32857608795166, | |
| "learning_rate": 8.558469729836324e-07, | |
| "loss": 0.6515, | |
| "step": 50800 | |
| }, | |
| { | |
| "epoch": 29.818394844756885, | |
| "grad_norm": 8.374032974243164, | |
| "learning_rate": 6.586472096233485e-07, | |
| "loss": 0.6615, | |
| "step": 50900 | |
| }, | |
| { | |
| "epoch": 29.876977152899823, | |
| "grad_norm": 8.55248737335205, | |
| "learning_rate": 4.6144744626306455e-07, | |
| "loss": 0.6645, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 29.876977152899823, | |
| "eval_loss": 0.08314584940671921, | |
| "eval_runtime": 145.2309, | |
| "eval_samples_per_second": 3.443, | |
| "eval_steps_per_second": 0.434, | |
| "eval_wer": 0.18980253463012084, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 29.935559461042764, | |
| "grad_norm": 4.774900436401367, | |
| "learning_rate": 2.6424768290278054e-07, | |
| "loss": 0.6359, | |
| "step": 51100 | |
| }, | |
| { | |
| "epoch": 29.994141769185706, | |
| "grad_norm": 4.431262493133545, | |
| "learning_rate": 6.704791954249655e-08, | |
| "loss": 0.6618, | |
| "step": 51200 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 51210, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.6610187993088e+19, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |