xls-r-300m-fr / trainer_state.json
AlexN's picture
End of training
0e9e665
raw
history blame
23.8 kB
{
"best_metric": 0.23875188827514648,
"best_model_checkpoint": "./checkpoint-13500",
"epoch": 2.0,
"global_step": 13822,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 6.533333333333333e-06,
"loss": 17.2403,
"step": 100
},
{
"epoch": 0.03,
"learning_rate": 1.32e-05,
"loss": 10.2311,
"step": 200
},
{
"epoch": 0.04,
"learning_rate": 1.9800000000000004e-05,
"loss": 7.834,
"step": 300
},
{
"epoch": 0.06,
"learning_rate": 2.646666666666667e-05,
"loss": 6.0656,
"step": 400
},
{
"epoch": 0.07,
"learning_rate": 3.313333333333333e-05,
"loss": 4.3748,
"step": 500
},
{
"epoch": 0.07,
"eval_loss": 3.878422975540161,
"eval_runtime": 285.8223,
"eval_samples_per_second": 20.264,
"eval_steps_per_second": 0.318,
"eval_wer": 1.0,
"step": 500
},
{
"epoch": 0.09,
"learning_rate": 3.9800000000000005e-05,
"loss": 3.2923,
"step": 600
},
{
"epoch": 0.1,
"learning_rate": 4.646666666666667e-05,
"loss": 2.9475,
"step": 700
},
{
"epoch": 0.12,
"learning_rate": 5.3133333333333335e-05,
"loss": 2.8639,
"step": 800
},
{
"epoch": 0.13,
"learning_rate": 5.9800000000000003e-05,
"loss": 2.8265,
"step": 900
},
{
"epoch": 0.14,
"learning_rate": 6.646666666666667e-05,
"loss": 2.8068,
"step": 1000
},
{
"epoch": 0.14,
"eval_loss": 2.828850746154785,
"eval_runtime": 292.3877,
"eval_samples_per_second": 19.809,
"eval_steps_per_second": 0.311,
"eval_wer": 0.9826485059793412,
"step": 1000
},
{
"epoch": 0.16,
"learning_rate": 7.306666666666668e-05,
"loss": 2.779,
"step": 1100
},
{
"epoch": 0.17,
"learning_rate": 7.973333333333334e-05,
"loss": 2.6402,
"step": 1200
},
{
"epoch": 0.19,
"learning_rate": 8.64e-05,
"loss": 2.1119,
"step": 1300
},
{
"epoch": 0.2,
"learning_rate": 9.306666666666667e-05,
"loss": 1.7965,
"step": 1400
},
{
"epoch": 0.22,
"learning_rate": 9.973333333333334e-05,
"loss": 1.6698,
"step": 1500
},
{
"epoch": 0.22,
"eval_loss": 0.881136417388916,
"eval_runtime": 297.1806,
"eval_samples_per_second": 19.49,
"eval_steps_per_second": 0.306,
"eval_wer": 0.7127472384241911,
"step": 1500
},
{
"epoch": 0.23,
"learning_rate": 9.92209056971271e-05,
"loss": 1.5882,
"step": 1600
},
{
"epoch": 0.25,
"learning_rate": 9.840934913163448e-05,
"loss": 1.5172,
"step": 1700
},
{
"epoch": 0.26,
"learning_rate": 9.759779256614186e-05,
"loss": 1.4579,
"step": 1800
},
{
"epoch": 0.27,
"learning_rate": 9.678623600064926e-05,
"loss": 1.3829,
"step": 1900
},
{
"epoch": 0.29,
"learning_rate": 9.597467943515663e-05,
"loss": 1.3488,
"step": 2000
},
{
"epoch": 0.29,
"eval_loss": 0.516592800617218,
"eval_runtime": 301.2842,
"eval_samples_per_second": 19.224,
"eval_steps_per_second": 0.302,
"eval_wer": 0.5369024731988661,
"step": 2000
},
{
"epoch": 0.3,
"learning_rate": 9.516312286966402e-05,
"loss": 1.2981,
"step": 2100
},
{
"epoch": 0.32,
"learning_rate": 9.43515663041714e-05,
"loss": 1.2845,
"step": 2200
},
{
"epoch": 0.33,
"learning_rate": 9.354000973867879e-05,
"loss": 1.2459,
"step": 2300
},
{
"epoch": 0.35,
"learning_rate": 9.272845317318618e-05,
"loss": 1.2255,
"step": 2400
},
{
"epoch": 0.36,
"learning_rate": 9.191689660769356e-05,
"loss": 1.2239,
"step": 2500
},
{
"epoch": 0.36,
"eval_loss": 0.4104757010936737,
"eval_runtime": 299.1395,
"eval_samples_per_second": 19.362,
"eval_steps_per_second": 0.304,
"eval_wer": 0.474111245071524,
"step": 2500
},
{
"epoch": 0.38,
"learning_rate": 9.110534004220094e-05,
"loss": 1.2024,
"step": 2600
},
{
"epoch": 0.39,
"learning_rate": 9.030189904236326e-05,
"loss": 1.1851,
"step": 2700
},
{
"epoch": 0.41,
"learning_rate": 8.949034247687063e-05,
"loss": 1.1768,
"step": 2800
},
{
"epoch": 0.42,
"learning_rate": 8.867878591137803e-05,
"loss": 1.1641,
"step": 2900
},
{
"epoch": 0.43,
"learning_rate": 8.786722934588541e-05,
"loss": 1.1537,
"step": 3000
},
{
"epoch": 0.43,
"eval_loss": 0.35850802063941956,
"eval_runtime": 299.662,
"eval_samples_per_second": 19.328,
"eval_steps_per_second": 0.304,
"eval_wer": 0.4448499462348073,
"step": 3000
},
{
"epoch": 0.45,
"learning_rate": 8.70556727803928e-05,
"loss": 1.1449,
"step": 3100
},
{
"epoch": 0.46,
"learning_rate": 8.624411621490018e-05,
"loss": 1.1379,
"step": 3200
},
{
"epoch": 0.48,
"learning_rate": 8.543255964940758e-05,
"loss": 1.1331,
"step": 3300
},
{
"epoch": 0.49,
"learning_rate": 8.462100308391495e-05,
"loss": 1.1205,
"step": 3400
},
{
"epoch": 0.51,
"learning_rate": 8.380944651842234e-05,
"loss": 1.1184,
"step": 3500
},
{
"epoch": 0.51,
"eval_loss": 0.333638072013855,
"eval_runtime": 297.0402,
"eval_samples_per_second": 19.499,
"eval_steps_per_second": 0.306,
"eval_wer": 0.42922545537489004,
"step": 3500
},
{
"epoch": 0.52,
"learning_rate": 8.299788995292971e-05,
"loss": 1.1014,
"step": 3600
},
{
"epoch": 0.54,
"learning_rate": 8.218633338743711e-05,
"loss": 1.1114,
"step": 3700
},
{
"epoch": 0.55,
"learning_rate": 8.13747768219445e-05,
"loss": 1.117,
"step": 3800
},
{
"epoch": 0.56,
"learning_rate": 8.056322025645188e-05,
"loss": 1.102,
"step": 3900
},
{
"epoch": 0.58,
"learning_rate": 7.975166369095926e-05,
"loss": 1.0968,
"step": 4000
},
{
"epoch": 0.58,
"eval_loss": 0.31949570775032043,
"eval_runtime": 296.0172,
"eval_samples_per_second": 19.566,
"eval_steps_per_second": 0.307,
"eval_wer": 0.4180162273127179,
"step": 4000
},
{
"epoch": 0.59,
"learning_rate": 7.894822269112158e-05,
"loss": 1.0942,
"step": 4100
},
{
"epoch": 0.61,
"learning_rate": 7.813666612562897e-05,
"loss": 1.0859,
"step": 4200
},
{
"epoch": 0.62,
"learning_rate": 7.732510956013635e-05,
"loss": 1.0767,
"step": 4300
},
{
"epoch": 0.64,
"learning_rate": 7.652166856029866e-05,
"loss": 1.0766,
"step": 4400
},
{
"epoch": 0.65,
"learning_rate": 7.571011199480604e-05,
"loss": 1.0737,
"step": 4500
},
{
"epoch": 0.65,
"eval_loss": 0.30754634737968445,
"eval_runtime": 296.2378,
"eval_samples_per_second": 19.552,
"eval_steps_per_second": 0.307,
"eval_wer": 0.41408973899442797,
"step": 4500
},
{
"epoch": 0.67,
"learning_rate": 7.489855542931342e-05,
"loss": 1.0807,
"step": 4600
},
{
"epoch": 0.68,
"learning_rate": 7.40869988638208e-05,
"loss": 1.071,
"step": 4700
},
{
"epoch": 0.69,
"learning_rate": 7.32754422983282e-05,
"loss": 1.0613,
"step": 4800
},
{
"epoch": 0.71,
"learning_rate": 7.246388573283557e-05,
"loss": 1.0635,
"step": 4900
},
{
"epoch": 0.72,
"learning_rate": 7.165232916734297e-05,
"loss": 1.0677,
"step": 5000
},
{
"epoch": 0.72,
"eval_loss": 0.30150118470191956,
"eval_runtime": 297.4742,
"eval_samples_per_second": 19.471,
"eval_steps_per_second": 0.306,
"eval_wer": 0.4089250219948516,
"step": 5000
},
{
"epoch": 0.74,
"learning_rate": 7.084077260185034e-05,
"loss": 1.0707,
"step": 5100
},
{
"epoch": 0.75,
"learning_rate": 7.002921603635774e-05,
"loss": 1.0617,
"step": 5200
},
{
"epoch": 0.77,
"learning_rate": 6.921765947086512e-05,
"loss": 1.0566,
"step": 5300
},
{
"epoch": 0.78,
"learning_rate": 6.84061029053725e-05,
"loss": 1.0518,
"step": 5400
},
{
"epoch": 0.8,
"learning_rate": 6.760266190553481e-05,
"loss": 1.0462,
"step": 5500
},
{
"epoch": 0.8,
"eval_loss": 0.297052800655365,
"eval_runtime": 296.1592,
"eval_samples_per_second": 19.557,
"eval_steps_per_second": 0.307,
"eval_wer": 0.4077193782788621,
"step": 5500
},
{
"epoch": 0.81,
"learning_rate": 6.679110534004221e-05,
"loss": 1.0514,
"step": 5600
},
{
"epoch": 0.82,
"learning_rate": 6.597954877454959e-05,
"loss": 1.0446,
"step": 5700
},
{
"epoch": 0.84,
"learning_rate": 6.516799220905698e-05,
"loss": 1.0358,
"step": 5800
},
{
"epoch": 0.85,
"learning_rate": 6.435643564356436e-05,
"loss": 1.0364,
"step": 5900
},
{
"epoch": 0.87,
"learning_rate": 6.354487907807174e-05,
"loss": 1.0392,
"step": 6000
},
{
"epoch": 0.87,
"eval_loss": 0.2870033383369446,
"eval_runtime": 295.9814,
"eval_samples_per_second": 19.569,
"eval_steps_per_second": 0.307,
"eval_wer": 0.3997034768157972,
"step": 6000
},
{
"epoch": 0.88,
"learning_rate": 6.273332251257913e-05,
"loss": 1.0375,
"step": 6100
},
{
"epoch": 0.9,
"learning_rate": 6.192176594708652e-05,
"loss": 1.0408,
"step": 6200
},
{
"epoch": 0.91,
"learning_rate": 6.11102093815939e-05,
"loss": 1.0382,
"step": 6300
},
{
"epoch": 0.93,
"learning_rate": 6.0298652816101284e-05,
"loss": 1.0335,
"step": 6400
},
{
"epoch": 0.94,
"learning_rate": 5.948709625060867e-05,
"loss": 1.0178,
"step": 6500
},
{
"epoch": 0.94,
"eval_loss": 0.28046590089797974,
"eval_runtime": 297.8045,
"eval_samples_per_second": 19.449,
"eval_steps_per_second": 0.306,
"eval_wer": 0.39629834794225943,
"step": 6500
},
{
"epoch": 0.95,
"learning_rate": 5.867553968511605e-05,
"loss": 1.0312,
"step": 6600
},
{
"epoch": 0.97,
"learning_rate": 5.786398311962344e-05,
"loss": 1.033,
"step": 6700
},
{
"epoch": 0.98,
"learning_rate": 5.7052426554130825e-05,
"loss": 1.0289,
"step": 6800
},
{
"epoch": 1.0,
"learning_rate": 5.624086998863821e-05,
"loss": 1.027,
"step": 6900
},
{
"epoch": 1.01,
"learning_rate": 5.54293134231456e-05,
"loss": 0.992,
"step": 7000
},
{
"epoch": 1.01,
"eval_loss": 0.2747785747051239,
"eval_runtime": 298.4841,
"eval_samples_per_second": 19.405,
"eval_steps_per_second": 0.305,
"eval_wer": 0.39352862589201343,
"step": 7000
},
{
"epoch": 1.03,
"learning_rate": 5.461775685765298e-05,
"loss": 1.0025,
"step": 7100
},
{
"epoch": 1.04,
"learning_rate": 5.3806200292160366e-05,
"loss": 1.0122,
"step": 7200
},
{
"epoch": 1.06,
"learning_rate": 5.299464372666775e-05,
"loss": 1.018,
"step": 7300
},
{
"epoch": 1.07,
"learning_rate": 5.218308716117514e-05,
"loss": 0.9936,
"step": 7400
},
{
"epoch": 1.09,
"learning_rate": 5.137153059568252e-05,
"loss": 1.0197,
"step": 7500
},
{
"epoch": 1.09,
"eval_loss": 0.26907604932785034,
"eval_runtime": 298.796,
"eval_samples_per_second": 19.384,
"eval_steps_per_second": 0.305,
"eval_wer": 0.3884453713056796,
"step": 7500
},
{
"epoch": 1.1,
"learning_rate": 5.055997403018991e-05,
"loss": 1.008,
"step": 7600
},
{
"epoch": 1.11,
"learning_rate": 4.97484174646973e-05,
"loss": 1.0028,
"step": 7700
},
{
"epoch": 1.13,
"learning_rate": 4.893686089920468e-05,
"loss": 0.9929,
"step": 7800
},
{
"epoch": 1.14,
"learning_rate": 4.8125304333712064e-05,
"loss": 0.995,
"step": 7900
},
{
"epoch": 1.16,
"learning_rate": 4.731374776821945e-05,
"loss": 1.0056,
"step": 8000
},
{
"epoch": 1.16,
"eval_loss": 0.26817116141319275,
"eval_runtime": 298.1504,
"eval_samples_per_second": 19.426,
"eval_steps_per_second": 0.305,
"eval_wer": 0.3888526833718922,
"step": 8000
},
{
"epoch": 1.17,
"learning_rate": 4.6510306768381754e-05,
"loss": 0.9971,
"step": 8100
},
{
"epoch": 1.19,
"learning_rate": 4.5698750202889145e-05,
"loss": 0.9976,
"step": 8200
},
{
"epoch": 1.2,
"learning_rate": 4.488719363739653e-05,
"loss": 1.0014,
"step": 8300
},
{
"epoch": 1.22,
"learning_rate": 4.407563707190391e-05,
"loss": 0.9835,
"step": 8400
},
{
"epoch": 1.23,
"learning_rate": 4.3272196072066225e-05,
"loss": 0.9826,
"step": 8500
},
{
"epoch": 1.23,
"eval_loss": 0.26473307609558105,
"eval_runtime": 299.0161,
"eval_samples_per_second": 19.37,
"eval_steps_per_second": 0.304,
"eval_wer": 0.38675095311023494,
"step": 8500
},
{
"epoch": 1.24,
"learning_rate": 4.2460639506573615e-05,
"loss": 0.9838,
"step": 8600
},
{
"epoch": 1.26,
"learning_rate": 4.1649082941081e-05,
"loss": 0.9836,
"step": 8700
},
{
"epoch": 1.27,
"learning_rate": 4.0845641941243305e-05,
"loss": 0.9824,
"step": 8800
},
{
"epoch": 1.29,
"learning_rate": 4.003408537575069e-05,
"loss": 0.9715,
"step": 8900
},
{
"epoch": 1.3,
"learning_rate": 3.922252881025807e-05,
"loss": 0.9815,
"step": 9000
},
{
"epoch": 1.3,
"eval_loss": 0.26034271717071533,
"eval_runtime": 299.6782,
"eval_samples_per_second": 19.327,
"eval_steps_per_second": 0.304,
"eval_wer": 0.3831828994102121,
"step": 9000
},
{
"epoch": 1.32,
"learning_rate": 3.841097224476546e-05,
"loss": 0.9757,
"step": 9100
},
{
"epoch": 1.33,
"learning_rate": 3.7599415679272846e-05,
"loss": 0.9689,
"step": 9200
},
{
"epoch": 1.35,
"learning_rate": 3.678785911378023e-05,
"loss": 0.9778,
"step": 9300
},
{
"epoch": 1.36,
"learning_rate": 3.5976302548287614e-05,
"loss": 0.9794,
"step": 9400
},
{
"epoch": 1.37,
"learning_rate": 3.5164745982795e-05,
"loss": 0.9717,
"step": 9500
},
{
"epoch": 1.37,
"eval_loss": 0.25609487295150757,
"eval_runtime": 299.6976,
"eval_samples_per_second": 19.326,
"eval_steps_per_second": 0.304,
"eval_wer": 0.3807064420476392,
"step": 9500
},
{
"epoch": 1.39,
"learning_rate": 3.435318941730239e-05,
"loss": 0.9752,
"step": 9600
},
{
"epoch": 1.4,
"learning_rate": 3.354163285180977e-05,
"loss": 0.965,
"step": 9700
},
{
"epoch": 1.42,
"learning_rate": 3.2730076286317155e-05,
"loss": 0.9522,
"step": 9800
},
{
"epoch": 1.43,
"learning_rate": 3.191851972082454e-05,
"loss": 0.9718,
"step": 9900
},
{
"epoch": 1.45,
"learning_rate": 3.110696315533193e-05,
"loss": 0.9605,
"step": 10000
},
{
"epoch": 1.45,
"eval_loss": 0.25231894850730896,
"eval_runtime": 297.5796,
"eval_samples_per_second": 19.464,
"eval_steps_per_second": 0.306,
"eval_wer": 0.3782951546156603,
"step": 10000
},
{
"epoch": 1.46,
"learning_rate": 3.0295406589839315e-05,
"loss": 0.9635,
"step": 10100
},
{
"epoch": 1.48,
"learning_rate": 2.94838500243467e-05,
"loss": 0.9632,
"step": 10200
},
{
"epoch": 1.49,
"learning_rate": 2.8672293458854082e-05,
"loss": 0.9548,
"step": 10300
},
{
"epoch": 1.5,
"learning_rate": 2.7868852459016392e-05,
"loss": 0.9554,
"step": 10400
},
{
"epoch": 1.52,
"learning_rate": 2.7057295893523783e-05,
"loss": 0.96,
"step": 10500
},
{
"epoch": 1.52,
"eval_loss": 0.24939315021038055,
"eval_runtime": 300.036,
"eval_samples_per_second": 19.304,
"eval_steps_per_second": 0.303,
"eval_wer": 0.3787513441298185,
"step": 10500
},
{
"epoch": 1.53,
"learning_rate": 2.6245739328031166e-05,
"loss": 0.9611,
"step": 10600
},
{
"epoch": 1.55,
"learning_rate": 2.543418276253855e-05,
"loss": 0.9594,
"step": 10700
},
{
"epoch": 1.56,
"learning_rate": 2.4622626197045937e-05,
"loss": 0.9589,
"step": 10800
},
{
"epoch": 1.58,
"learning_rate": 2.381106963155332e-05,
"loss": 0.9441,
"step": 10900
},
{
"epoch": 1.59,
"learning_rate": 2.2999513066060704e-05,
"loss": 0.9442,
"step": 11000
},
{
"epoch": 1.59,
"eval_loss": 0.24783751368522644,
"eval_runtime": 297.7741,
"eval_samples_per_second": 19.451,
"eval_steps_per_second": 0.306,
"eval_wer": 0.3760142070448695,
"step": 11000
},
{
"epoch": 1.61,
"learning_rate": 2.218795650056809e-05,
"loss": 0.9496,
"step": 11100
},
{
"epoch": 1.62,
"learning_rate": 2.1376399935075474e-05,
"loss": 0.9486,
"step": 11200
},
{
"epoch": 1.64,
"learning_rate": 2.056484336958286e-05,
"loss": 0.9558,
"step": 11300
},
{
"epoch": 1.65,
"learning_rate": 1.9753286804090245e-05,
"loss": 0.9486,
"step": 11400
},
{
"epoch": 1.66,
"learning_rate": 1.8941730238597632e-05,
"loss": 0.9564,
"step": 11500
},
{
"epoch": 1.66,
"eval_loss": 0.2454409897327423,
"eval_runtime": 296.0265,
"eval_samples_per_second": 19.566,
"eval_steps_per_second": 0.307,
"eval_wer": 0.3733096549252175,
"step": 11500
},
{
"epoch": 1.68,
"learning_rate": 1.8130173673105015e-05,
"loss": 0.9427,
"step": 11600
},
{
"epoch": 1.69,
"learning_rate": 1.73186171076124e-05,
"loss": 0.9423,
"step": 11700
},
{
"epoch": 1.71,
"learning_rate": 1.6507060542119786e-05,
"loss": 0.9503,
"step": 11800
},
{
"epoch": 1.72,
"learning_rate": 1.5695503976627173e-05,
"loss": 0.9383,
"step": 11900
},
{
"epoch": 1.74,
"learning_rate": 1.4883947411134558e-05,
"loss": 0.9436,
"step": 12000
},
{
"epoch": 1.74,
"eval_loss": 0.24390804767608643,
"eval_runtime": 295.4584,
"eval_samples_per_second": 19.603,
"eval_steps_per_second": 0.308,
"eval_wer": 0.37467822346769203,
"step": 12000
},
{
"epoch": 1.75,
"learning_rate": 1.4072390845641942e-05,
"loss": 0.9491,
"step": 12100
},
{
"epoch": 1.77,
"learning_rate": 1.3260834280149325e-05,
"loss": 0.9419,
"step": 12200
},
{
"epoch": 1.78,
"learning_rate": 1.2449277714656712e-05,
"loss": 0.9517,
"step": 12300
},
{
"epoch": 1.79,
"learning_rate": 1.1637721149164097e-05,
"loss": 0.9367,
"step": 12400
},
{
"epoch": 1.81,
"learning_rate": 1.0826164583671483e-05,
"loss": 0.938,
"step": 12500
},
{
"epoch": 1.81,
"eval_loss": 0.24111612141132355,
"eval_runtime": 296.9314,
"eval_samples_per_second": 19.506,
"eval_steps_per_second": 0.306,
"eval_wer": 0.37159894424712436,
"step": 12500
},
{
"epoch": 1.82,
"learning_rate": 1.0014608018178868e-05,
"loss": 0.9337,
"step": 12600
},
{
"epoch": 1.84,
"learning_rate": 9.203051452686253e-06,
"loss": 0.9284,
"step": 12700
},
{
"epoch": 1.85,
"learning_rate": 8.391494887193638e-06,
"loss": 0.938,
"step": 12800
},
{
"epoch": 1.87,
"learning_rate": 7.579938321701023e-06,
"loss": 0.9365,
"step": 12900
},
{
"epoch": 1.88,
"learning_rate": 6.768381756208409e-06,
"loss": 0.9353,
"step": 13000
},
{
"epoch": 1.88,
"eval_loss": 0.23965783417224884,
"eval_runtime": 296.6078,
"eval_samples_per_second": 19.527,
"eval_steps_per_second": 0.307,
"eval_wer": 0.3697904786731402,
"step": 13000
},
{
"epoch": 1.9,
"learning_rate": 5.956825190715793e-06,
"loss": 0.9413,
"step": 13100
},
{
"epoch": 1.91,
"learning_rate": 5.153384190878104e-06,
"loss": 0.9356,
"step": 13200
},
{
"epoch": 1.92,
"learning_rate": 4.34182762538549e-06,
"loss": 0.9209,
"step": 13300
},
{
"epoch": 1.94,
"learning_rate": 3.530271059892875e-06,
"loss": 0.9362,
"step": 13400
},
{
"epoch": 1.95,
"learning_rate": 2.71871449440026e-06,
"loss": 0.9271,
"step": 13500
},
{
"epoch": 1.95,
"eval_loss": 0.23875188827514648,
"eval_runtime": 296.9414,
"eval_samples_per_second": 19.506,
"eval_steps_per_second": 0.306,
"eval_wer": 0.3680797679950471,
"step": 13500
},
{
"epoch": 1.97,
"learning_rate": 1.907157928907645e-06,
"loss": 0.9288,
"step": 13600
},
{
"epoch": 1.98,
"learning_rate": 1.0956013634150302e-06,
"loss": 0.9345,
"step": 13700
},
{
"epoch": 2.0,
"learning_rate": 2.840447979224152e-07,
"loss": 0.9326,
"step": 13800
},
{
"epoch": 2.0,
"step": 13822,
"total_flos": 1.2600843645735263e+20,
"train_loss": 1.442369053426242,
"train_runtime": 53680.5392,
"train_samples_per_second": 16.478,
"train_steps_per_second": 0.257
}
],
"max_steps": 13822,
"num_train_epochs": 2,
"total_flos": 1.2600843645735263e+20,
"trial_name": null,
"trial_params": null
}