|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 8.0, |
|
"global_step": 11448, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.07, |
|
"learning_rate": 2.0000000000000003e-06, |
|
"loss": 86.0421, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"learning_rate": 4.000000000000001e-06, |
|
"loss": 51.6706, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"learning_rate": 6e-06, |
|
"loss": 41.1349, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.28, |
|
"learning_rate": 8.000000000000001e-06, |
|
"loss": 36.5061, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"learning_rate": 1e-05, |
|
"loss": 33.1858, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.42, |
|
"learning_rate": 1.2e-05, |
|
"loss": 30.5206, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.49, |
|
"learning_rate": 1.4000000000000001e-05, |
|
"loss": 28.0073, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 0.56, |
|
"learning_rate": 1.6000000000000003e-05, |
|
"loss": 26.1939, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 0.63, |
|
"learning_rate": 1.8e-05, |
|
"loss": 24.3465, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"learning_rate": 2e-05, |
|
"loss": 22.5759, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.77, |
|
"learning_rate": 2.2000000000000003e-05, |
|
"loss": 20.9294, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 0.84, |
|
"learning_rate": 2.4e-05, |
|
"loss": 19.3762, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 0.91, |
|
"learning_rate": 2.6000000000000002e-05, |
|
"loss": 17.72, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 0.98, |
|
"learning_rate": 2.8000000000000003e-05, |
|
"loss": 15.7901, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 1.05, |
|
"learning_rate": 3e-05, |
|
"loss": 14.0008, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 1.12, |
|
"learning_rate": 3.2000000000000005e-05, |
|
"loss": 12.3777, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 1.19, |
|
"learning_rate": 3.4000000000000007e-05, |
|
"loss": 10.7261, |
|
"step": 1700 |
|
}, |
|
{ |
|
"epoch": 1.26, |
|
"learning_rate": 3.6e-05, |
|
"loss": 9.1024, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 1.33, |
|
"learning_rate": 3.8e-05, |
|
"loss": 7.4676, |
|
"step": 1900 |
|
}, |
|
{ |
|
"epoch": 1.4, |
|
"learning_rate": 4e-05, |
|
"loss": 6.6044, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 1.4, |
|
"eval_bleu": 3.8045, |
|
"eval_em": 0.0, |
|
"eval_gen_len": 158.8473, |
|
"eval_loss": 6.171305179595947, |
|
"eval_runtime": 1556.9786, |
|
"eval_samples_per_second": 0.774, |
|
"eval_steps_per_second": 0.097, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 1.47, |
|
"learning_rate": 4.2e-05, |
|
"loss": 6.0941, |
|
"step": 2100 |
|
}, |
|
{ |
|
"epoch": 1.54, |
|
"learning_rate": 4.4000000000000006e-05, |
|
"loss": 5.6741, |
|
"step": 2200 |
|
}, |
|
{ |
|
"epoch": 1.61, |
|
"learning_rate": 4.600000000000001e-05, |
|
"loss": 5.4757, |
|
"step": 2300 |
|
}, |
|
{ |
|
"epoch": 1.68, |
|
"learning_rate": 4.8e-05, |
|
"loss": 5.242, |
|
"step": 2400 |
|
}, |
|
{ |
|
"epoch": 1.75, |
|
"learning_rate": 5e-05, |
|
"loss": 5.0108, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 1.82, |
|
"learning_rate": 4.944121591417077e-05, |
|
"loss": 4.8595, |
|
"step": 2600 |
|
}, |
|
{ |
|
"epoch": 1.89, |
|
"learning_rate": 4.888243182834153e-05, |
|
"loss": 4.695, |
|
"step": 2700 |
|
}, |
|
{ |
|
"epoch": 1.96, |
|
"learning_rate": 4.8323647742512295e-05, |
|
"loss": 4.5706, |
|
"step": 2800 |
|
}, |
|
{ |
|
"epoch": 2.03, |
|
"learning_rate": 4.776486365668306e-05, |
|
"loss": 4.2498, |
|
"step": 2900 |
|
}, |
|
{ |
|
"epoch": 2.1, |
|
"learning_rate": 4.720607957085382e-05, |
|
"loss": 4.1223, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 2.17, |
|
"learning_rate": 4.664729548502459e-05, |
|
"loss": 4.0181, |
|
"step": 3100 |
|
}, |
|
{ |
|
"epoch": 2.24, |
|
"learning_rate": 4.6088511399195353e-05, |
|
"loss": 3.8722, |
|
"step": 3200 |
|
}, |
|
{ |
|
"epoch": 2.31, |
|
"learning_rate": 4.552972731336611e-05, |
|
"loss": 3.7786, |
|
"step": 3300 |
|
}, |
|
{ |
|
"epoch": 2.38, |
|
"learning_rate": 4.497094322753688e-05, |
|
"loss": 3.6403, |
|
"step": 3400 |
|
}, |
|
{ |
|
"epoch": 2.45, |
|
"learning_rate": 4.4412159141707646e-05, |
|
"loss": 3.5437, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 2.52, |
|
"learning_rate": 4.385337505587841e-05, |
|
"loss": 3.389, |
|
"step": 3600 |
|
}, |
|
{ |
|
"epoch": 2.59, |
|
"learning_rate": 4.329459097004918e-05, |
|
"loss": 3.2395, |
|
"step": 3700 |
|
}, |
|
{ |
|
"epoch": 2.66, |
|
"learning_rate": 4.2735806884219945e-05, |
|
"loss": 3.1786, |
|
"step": 3800 |
|
}, |
|
{ |
|
"epoch": 2.73, |
|
"learning_rate": 4.2177022798390704e-05, |
|
"loss": 3.0657, |
|
"step": 3900 |
|
}, |
|
{ |
|
"epoch": 2.8, |
|
"learning_rate": 4.161823871256147e-05, |
|
"loss": 3.032, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 2.8, |
|
"eval_bleu": 27.701, |
|
"eval_em": 0.0, |
|
"eval_gen_len": 33.9568, |
|
"eval_loss": 2.904534339904785, |
|
"eval_runtime": 297.5953, |
|
"eval_samples_per_second": 4.049, |
|
"eval_steps_per_second": 0.507, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 2.87, |
|
"learning_rate": 4.105945462673223e-05, |
|
"loss": 2.8755, |
|
"step": 4100 |
|
}, |
|
{ |
|
"epoch": 2.94, |
|
"learning_rate": 4.0500670540903e-05, |
|
"loss": 2.8396, |
|
"step": 4200 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"learning_rate": 3.994188645507376e-05, |
|
"loss": 2.7454, |
|
"step": 4300 |
|
}, |
|
{ |
|
"epoch": 3.07, |
|
"learning_rate": 3.938310236924452e-05, |
|
"loss": 2.5218, |
|
"step": 4400 |
|
}, |
|
{ |
|
"epoch": 3.14, |
|
"learning_rate": 3.882431828341529e-05, |
|
"loss": 2.4895, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 3.21, |
|
"learning_rate": 3.8265534197586055e-05, |
|
"loss": 2.4554, |
|
"step": 4600 |
|
}, |
|
{ |
|
"epoch": 3.28, |
|
"learning_rate": 3.7706750111756815e-05, |
|
"loss": 2.3573, |
|
"step": 4700 |
|
}, |
|
{ |
|
"epoch": 3.35, |
|
"learning_rate": 3.714796602592758e-05, |
|
"loss": 2.2979, |
|
"step": 4800 |
|
}, |
|
{ |
|
"epoch": 3.42, |
|
"learning_rate": 3.658918194009835e-05, |
|
"loss": 2.1874, |
|
"step": 4900 |
|
}, |
|
{ |
|
"epoch": 3.49, |
|
"learning_rate": 3.603039785426911e-05, |
|
"loss": 2.1803, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 3.56, |
|
"learning_rate": 3.5471613768439874e-05, |
|
"loss": 2.1553, |
|
"step": 5100 |
|
}, |
|
{ |
|
"epoch": 3.63, |
|
"learning_rate": 3.491282968261064e-05, |
|
"loss": 2.0567, |
|
"step": 5200 |
|
}, |
|
{ |
|
"epoch": 3.7, |
|
"learning_rate": 3.4354045596781406e-05, |
|
"loss": 2.0147, |
|
"step": 5300 |
|
}, |
|
{ |
|
"epoch": 3.77, |
|
"learning_rate": 3.379526151095217e-05, |
|
"loss": 1.9817, |
|
"step": 5400 |
|
}, |
|
{ |
|
"epoch": 3.84, |
|
"learning_rate": 3.323647742512294e-05, |
|
"loss": 1.8843, |
|
"step": 5500 |
|
}, |
|
{ |
|
"epoch": 3.91, |
|
"learning_rate": 3.26776933392937e-05, |
|
"loss": 1.8849, |
|
"step": 5600 |
|
}, |
|
{ |
|
"epoch": 3.98, |
|
"learning_rate": 3.2118909253464465e-05, |
|
"loss": 1.8232, |
|
"step": 5700 |
|
}, |
|
{ |
|
"epoch": 4.05, |
|
"learning_rate": 3.156012516763523e-05, |
|
"loss": 1.6714, |
|
"step": 5800 |
|
}, |
|
{ |
|
"epoch": 4.12, |
|
"learning_rate": 3.100134108180599e-05, |
|
"loss": 1.6047, |
|
"step": 5900 |
|
}, |
|
{ |
|
"epoch": 4.19, |
|
"learning_rate": 3.0442556995976757e-05, |
|
"loss": 1.5893, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 4.19, |
|
"eval_bleu": 57.7509, |
|
"eval_em": 0.0033, |
|
"eval_gen_len": 44.132, |
|
"eval_loss": 1.7326730489730835, |
|
"eval_runtime": 426.2734, |
|
"eval_samples_per_second": 2.827, |
|
"eval_steps_per_second": 0.354, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 4.26, |
|
"learning_rate": 2.9883772910147524e-05, |
|
"loss": 1.5864, |
|
"step": 6100 |
|
}, |
|
{ |
|
"epoch": 4.33, |
|
"learning_rate": 2.9324988824318283e-05, |
|
"loss": 1.5608, |
|
"step": 6200 |
|
}, |
|
{ |
|
"epoch": 4.4, |
|
"learning_rate": 2.876620473848905e-05, |
|
"loss": 1.5144, |
|
"step": 6300 |
|
}, |
|
{ |
|
"epoch": 4.47, |
|
"learning_rate": 2.8207420652659816e-05, |
|
"loss": 1.4582, |
|
"step": 6400 |
|
}, |
|
{ |
|
"epoch": 4.54, |
|
"learning_rate": 2.7648636566830576e-05, |
|
"loss": 1.4793, |
|
"step": 6500 |
|
}, |
|
{ |
|
"epoch": 4.61, |
|
"learning_rate": 2.7089852481001342e-05, |
|
"loss": 1.472, |
|
"step": 6600 |
|
}, |
|
{ |
|
"epoch": 4.68, |
|
"learning_rate": 2.653106839517211e-05, |
|
"loss": 1.4424, |
|
"step": 6700 |
|
}, |
|
{ |
|
"epoch": 4.75, |
|
"learning_rate": 2.597228430934287e-05, |
|
"loss": 1.3779, |
|
"step": 6800 |
|
}, |
|
{ |
|
"epoch": 4.82, |
|
"learning_rate": 2.5413500223513638e-05, |
|
"loss": 1.3611, |
|
"step": 6900 |
|
}, |
|
{ |
|
"epoch": 4.89, |
|
"learning_rate": 2.48547161376844e-05, |
|
"loss": 1.3311, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 4.96, |
|
"learning_rate": 2.4295932051855164e-05, |
|
"loss": 1.3164, |
|
"step": 7100 |
|
}, |
|
{ |
|
"epoch": 5.03, |
|
"learning_rate": 2.373714796602593e-05, |
|
"loss": 1.2119, |
|
"step": 7200 |
|
}, |
|
{ |
|
"epoch": 5.1, |
|
"learning_rate": 2.3178363880196693e-05, |
|
"loss": 1.1122, |
|
"step": 7300 |
|
}, |
|
{ |
|
"epoch": 5.17, |
|
"learning_rate": 2.2619579794367456e-05, |
|
"loss": 1.1198, |
|
"step": 7400 |
|
}, |
|
{ |
|
"epoch": 5.24, |
|
"learning_rate": 2.206079570853822e-05, |
|
"loss": 1.0416, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 5.31, |
|
"learning_rate": 2.1502011622708985e-05, |
|
"loss": 1.1042, |
|
"step": 7600 |
|
}, |
|
{ |
|
"epoch": 5.38, |
|
"learning_rate": 2.0943227536879752e-05, |
|
"loss": 1.0715, |
|
"step": 7700 |
|
}, |
|
{ |
|
"epoch": 5.45, |
|
"learning_rate": 2.0384443451050515e-05, |
|
"loss": 1.0815, |
|
"step": 7800 |
|
}, |
|
{ |
|
"epoch": 5.52, |
|
"learning_rate": 1.982565936522128e-05, |
|
"loss": 1.0445, |
|
"step": 7900 |
|
}, |
|
{ |
|
"epoch": 5.59, |
|
"learning_rate": 1.9266875279392044e-05, |
|
"loss": 1.0512, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 5.59, |
|
"eval_bleu": 71.1272, |
|
"eval_em": 0.0241, |
|
"eval_gen_len": 46.0672, |
|
"eval_loss": 1.2382431030273438, |
|
"eval_runtime": 358.1089, |
|
"eval_samples_per_second": 3.365, |
|
"eval_steps_per_second": 0.422, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 5.66, |
|
"learning_rate": 1.8708091193562807e-05, |
|
"loss": 1.0201, |
|
"step": 8100 |
|
}, |
|
{ |
|
"epoch": 5.73, |
|
"learning_rate": 1.8149307107733573e-05, |
|
"loss": 1.0377, |
|
"step": 8200 |
|
}, |
|
{ |
|
"epoch": 5.8, |
|
"learning_rate": 1.7590523021904336e-05, |
|
"loss": 0.986, |
|
"step": 8300 |
|
}, |
|
{ |
|
"epoch": 5.87, |
|
"learning_rate": 1.70317389360751e-05, |
|
"loss": 1.0244, |
|
"step": 8400 |
|
}, |
|
{ |
|
"epoch": 5.94, |
|
"learning_rate": 1.6472954850245866e-05, |
|
"loss": 0.9654, |
|
"step": 8500 |
|
}, |
|
{ |
|
"epoch": 6.01, |
|
"learning_rate": 1.5914170764416632e-05, |
|
"loss": 0.9454, |
|
"step": 8600 |
|
}, |
|
{ |
|
"epoch": 6.08, |
|
"learning_rate": 1.5355386678587395e-05, |
|
"loss": 0.8179, |
|
"step": 8700 |
|
}, |
|
{ |
|
"epoch": 6.15, |
|
"learning_rate": 1.479660259275816e-05, |
|
"loss": 0.8433, |
|
"step": 8800 |
|
}, |
|
{ |
|
"epoch": 6.22, |
|
"learning_rate": 1.4237818506928924e-05, |
|
"loss": 0.8235, |
|
"step": 8900 |
|
}, |
|
{ |
|
"epoch": 6.29, |
|
"learning_rate": 1.3679034421099687e-05, |
|
"loss": 0.832, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 6.36, |
|
"learning_rate": 1.312025033527045e-05, |
|
"loss": 0.8019, |
|
"step": 9100 |
|
}, |
|
{ |
|
"epoch": 6.43, |
|
"learning_rate": 1.2561466249441217e-05, |
|
"loss": 0.806, |
|
"step": 9200 |
|
}, |
|
{ |
|
"epoch": 6.5, |
|
"learning_rate": 1.2002682163611981e-05, |
|
"loss": 0.7985, |
|
"step": 9300 |
|
}, |
|
{ |
|
"epoch": 6.57, |
|
"learning_rate": 1.1443898077782746e-05, |
|
"loss": 0.801, |
|
"step": 9400 |
|
}, |
|
{ |
|
"epoch": 6.64, |
|
"learning_rate": 1.0885113991953509e-05, |
|
"loss": 0.7758, |
|
"step": 9500 |
|
}, |
|
{ |
|
"epoch": 6.71, |
|
"learning_rate": 1.0326329906124274e-05, |
|
"loss": 0.804, |
|
"step": 9600 |
|
}, |
|
{ |
|
"epoch": 6.78, |
|
"learning_rate": 9.767545820295038e-06, |
|
"loss": 0.7846, |
|
"step": 9700 |
|
}, |
|
{ |
|
"epoch": 6.85, |
|
"learning_rate": 9.208761734465803e-06, |
|
"loss": 0.7736, |
|
"step": 9800 |
|
}, |
|
{ |
|
"epoch": 6.92, |
|
"learning_rate": 8.649977648636568e-06, |
|
"loss": 0.7625, |
|
"step": 9900 |
|
}, |
|
{ |
|
"epoch": 6.99, |
|
"learning_rate": 8.09119356280733e-06, |
|
"loss": 0.7515, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 6.99, |
|
"eval_bleu": 75.7356, |
|
"eval_em": 0.0432, |
|
"eval_gen_len": 47.2896, |
|
"eval_loss": 1.057088017463684, |
|
"eval_runtime": 378.6301, |
|
"eval_samples_per_second": 3.183, |
|
"eval_steps_per_second": 0.399, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 7.06, |
|
"learning_rate": 7.532409476978096e-06, |
|
"loss": 0.6703, |
|
"step": 10100 |
|
}, |
|
{ |
|
"epoch": 7.13, |
|
"learning_rate": 6.973625391148861e-06, |
|
"loss": 0.6731, |
|
"step": 10200 |
|
}, |
|
{ |
|
"epoch": 7.2, |
|
"learning_rate": 6.414841305319625e-06, |
|
"loss": 0.6584, |
|
"step": 10300 |
|
}, |
|
{ |
|
"epoch": 7.27, |
|
"learning_rate": 5.856057219490389e-06, |
|
"loss": 0.6758, |
|
"step": 10400 |
|
}, |
|
{ |
|
"epoch": 7.34, |
|
"learning_rate": 5.297273133661153e-06, |
|
"loss": 0.6801, |
|
"step": 10500 |
|
}, |
|
{ |
|
"epoch": 7.41, |
|
"learning_rate": 4.738489047831918e-06, |
|
"loss": 0.6556, |
|
"step": 10600 |
|
}, |
|
{ |
|
"epoch": 7.48, |
|
"learning_rate": 4.1797049620026825e-06, |
|
"loss": 0.6413, |
|
"step": 10700 |
|
}, |
|
{ |
|
"epoch": 7.55, |
|
"learning_rate": 3.6209208761734468e-06, |
|
"loss": 0.6548, |
|
"step": 10800 |
|
}, |
|
{ |
|
"epoch": 7.62, |
|
"learning_rate": 3.062136790344211e-06, |
|
"loss": 0.6631, |
|
"step": 10900 |
|
}, |
|
{ |
|
"epoch": 7.69, |
|
"learning_rate": 2.5033527045149757e-06, |
|
"loss": 0.6458, |
|
"step": 11000 |
|
}, |
|
{ |
|
"epoch": 7.76, |
|
"learning_rate": 1.94456861868574e-06, |
|
"loss": 0.629, |
|
"step": 11100 |
|
}, |
|
{ |
|
"epoch": 7.83, |
|
"learning_rate": 1.3857845328565042e-06, |
|
"loss": 0.643, |
|
"step": 11200 |
|
}, |
|
{ |
|
"epoch": 7.9, |
|
"learning_rate": 8.270004470272687e-07, |
|
"loss": 0.6475, |
|
"step": 11300 |
|
}, |
|
{ |
|
"epoch": 7.97, |
|
"learning_rate": 2.682163611980331e-07, |
|
"loss": 0.633, |
|
"step": 11400 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"step": 11448, |
|
"total_flos": 7528840539566280.0, |
|
"train_loss": 0.15130291744847968, |
|
"train_runtime": 1637.6947, |
|
"train_samples_per_second": 111.816, |
|
"train_steps_per_second": 6.99 |
|
} |
|
], |
|
"max_steps": 11448, |
|
"num_train_epochs": 8, |
|
"total_flos": 7528840539566280.0, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|