| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.096121848201715, | |
| "eval_steps": 500, | |
| "global_step": 16000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0004957330602491893, | |
| "loss": 2.0441, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0004914661204983786, | |
| "loss": 1.713, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00048719918074756785, | |
| "loss": 1.6007, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00048293224099675715, | |
| "loss": 1.5181, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00047866530124594645, | |
| "loss": 1.4493, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_bleu": 3.9997520454971136, | |
| "eval_loss": 1.3540712594985962, | |
| "eval_runtime": 10.3322, | |
| "eval_samples_per_second": 193.569, | |
| "eval_steps_per_second": 1.549, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0004743983614951357, | |
| "loss": 1.4096, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.000470131421744325, | |
| "loss": 1.3729, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0004658644819935143, | |
| "loss": 1.3389, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.00046159754224270354, | |
| "loss": 1.3166, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.00045733060249189284, | |
| "loss": 1.2915, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_bleu": 5.393565315342523, | |
| "eval_loss": 1.2113043069839478, | |
| "eval_runtime": 8.5631, | |
| "eval_samples_per_second": 233.562, | |
| "eval_steps_per_second": 1.868, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00045306366274108214, | |
| "loss": 1.2669, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0004487967229902714, | |
| "loss": 1.2492, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0004445297832394607, | |
| "loss": 1.2371, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00044026284348864993, | |
| "loss": 1.2231, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00043599590373783923, | |
| "loss": 1.2059, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_bleu": 5.838067918496751, | |
| "eval_loss": 1.1367273330688477, | |
| "eval_runtime": 9.2404, | |
| "eval_samples_per_second": 216.442, | |
| "eval_steps_per_second": 1.732, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0004317289639870285, | |
| "loss": 1.1931, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0004274620242362178, | |
| "loss": 1.1761, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0004231950844854071, | |
| "loss": 1.1792, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0004189281447345963, | |
| "loss": 1.1614, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0004146612049837856, | |
| "loss": 1.1573, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_bleu": 6.242227124133382, | |
| "eval_loss": 1.0900604724884033, | |
| "eval_runtime": 8.3953, | |
| "eval_samples_per_second": 238.228, | |
| "eval_steps_per_second": 1.906, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0004103942652329749, | |
| "loss": 1.1449, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00040612732548216417, | |
| "loss": 1.1379, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00040186038573135347, | |
| "loss": 1.1281, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.00039759344598054277, | |
| "loss": 1.1226, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.00039332650622973206, | |
| "loss": 1.1121, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_bleu": 6.627148578471305, | |
| "eval_loss": 1.0542418956756592, | |
| "eval_runtime": 9.2835, | |
| "eval_samples_per_second": 215.436, | |
| "eval_steps_per_second": 1.723, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0003890595664789213, | |
| "loss": 1.1076, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.0003847926267281106, | |
| "loss": 1.1001, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.0003805256869772999, | |
| "loss": 1.0996, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.00037625874722648915, | |
| "loss": 1.088, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.00037199180747567845, | |
| "loss": 1.0867, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "eval_bleu": 6.879580310573366, | |
| "eval_loss": 1.025155782699585, | |
| "eval_runtime": 8.5706, | |
| "eval_samples_per_second": 233.355, | |
| "eval_steps_per_second": 1.867, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.00036772486772486775, | |
| "loss": 1.0784, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.000363457927974057, | |
| "loss": 1.0794, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.0003591909882232463, | |
| "loss": 1.0736, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0003549240484724356, | |
| "loss": 1.0684, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.0003506571087216249, | |
| "loss": 1.0623, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_bleu": 7.039305128716119, | |
| "eval_loss": 1.0067821741104126, | |
| "eval_runtime": 9.3223, | |
| "eval_samples_per_second": 214.539, | |
| "eval_steps_per_second": 1.716, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.00034639016897081414, | |
| "loss": 1.0609, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.00034212322922000344, | |
| "loss": 1.0579, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.00033785628946919274, | |
| "loss": 1.0517, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.000333589349718382, | |
| "loss": 1.054, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0003293224099675713, | |
| "loss": 1.0408, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_bleu": 7.2660174031875915, | |
| "eval_loss": 0.9882246255874634, | |
| "eval_runtime": 8.4135, | |
| "eval_samples_per_second": 237.713, | |
| "eval_steps_per_second": 1.902, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 0.0003250554702167606, | |
| "loss": 1.0317, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.0003207885304659498, | |
| "loss": 1.0344, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 0.0003165215907151391, | |
| "loss": 1.0274, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 0.0003122546509643284, | |
| "loss": 1.0216, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 0.0003079877112135177, | |
| "loss": 1.0203, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_bleu": 7.055297387993883, | |
| "eval_loss": 0.9723305702209473, | |
| "eval_runtime": 9.3412, | |
| "eval_samples_per_second": 214.105, | |
| "eval_steps_per_second": 1.713, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.0003037207714627069, | |
| "loss": 1.0217, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.0002994538317118962, | |
| "loss": 1.0184, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.0002951868919610855, | |
| "loss": 1.015, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.00029091995221027477, | |
| "loss": 1.011, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 0.00028665301245946406, | |
| "loss": 1.0054, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "eval_bleu": 7.455525040183996, | |
| "eval_loss": 0.9624391198158264, | |
| "eval_runtime": 8.4826, | |
| "eval_samples_per_second": 235.778, | |
| "eval_steps_per_second": 1.886, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.00028238607270865336, | |
| "loss": 1.0062, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0002781191329578426, | |
| "loss": 1.0081, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 0.0002738521932070319, | |
| "loss": 1.0007, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.0002695852534562212, | |
| "loss": 0.9973, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 0.0002653183137054105, | |
| "loss": 0.9977, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "eval_bleu": 7.426000467701662, | |
| "eval_loss": 0.9526358842849731, | |
| "eval_runtime": 9.3659, | |
| "eval_samples_per_second": 213.542, | |
| "eval_steps_per_second": 1.708, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.00026105137395459975, | |
| "loss": 0.9982, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.00025678443420378905, | |
| "loss": 0.9956, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 0.00025251749445297835, | |
| "loss": 0.9941, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 0.0002482505547021676, | |
| "loss": 0.9881, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 0.0002439836149513569, | |
| "loss": 0.9931, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "eval_bleu": 7.523072732517067, | |
| "eval_loss": 0.9395684003829956, | |
| "eval_runtime": 9.0823, | |
| "eval_samples_per_second": 220.209, | |
| "eval_steps_per_second": 1.762, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.0002397166752005462, | |
| "loss": 0.9834, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 0.00023544973544973544, | |
| "loss": 0.9821, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 0.00023118279569892471, | |
| "loss": 0.9859, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 0.00022691585594811401, | |
| "loss": 0.9762, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 0.0002226489161973033, | |
| "loss": 0.9804, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "eval_bleu": 7.43758431482021, | |
| "eval_loss": 0.9323887825012207, | |
| "eval_runtime": 9.0344, | |
| "eval_samples_per_second": 221.376, | |
| "eval_steps_per_second": 1.771, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00021838197644649259, | |
| "loss": 0.9794, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 0.00021411503669568186, | |
| "loss": 0.9737, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 0.00020984809694487113, | |
| "loss": 0.9696, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 0.00020558115719406043, | |
| "loss": 0.9735, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 0.0002013142174432497, | |
| "loss": 0.9691, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "eval_bleu": 7.52270860836675, | |
| "eval_loss": 0.9264442324638367, | |
| "eval_runtime": 9.3377, | |
| "eval_samples_per_second": 214.186, | |
| "eval_steps_per_second": 1.713, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.000197047277692439, | |
| "loss": 0.9754, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 0.00019278033794162827, | |
| "loss": 0.9688, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 0.00018851339819081755, | |
| "loss": 0.968, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.00018424645844000685, | |
| "loss": 0.9628, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 0.00017997951868919612, | |
| "loss": 0.9645, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_bleu": 7.685904345266889, | |
| "eval_loss": 0.9192501306533813, | |
| "eval_runtime": 8.7414, | |
| "eval_samples_per_second": 228.796, | |
| "eval_steps_per_second": 1.83, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 0.0001757125789383854, | |
| "loss": 0.9674, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 0.00017144563918757466, | |
| "loss": 0.9659, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.00016717869943676394, | |
| "loss": 0.9646, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 0.00016291175968595324, | |
| "loss": 0.9477, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 0.0001586448199351425, | |
| "loss": 0.9509, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "eval_bleu": 7.647325378973634, | |
| "eval_loss": 0.9144095778465271, | |
| "eval_runtime": 9.1538, | |
| "eval_samples_per_second": 218.489, | |
| "eval_steps_per_second": 1.748, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 0.00024078341013824886, | |
| "loss": 0.9532, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 0.00023758320532514082, | |
| "loss": 0.949, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 0.00023438300051203275, | |
| "loss": 0.9506, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 0.00023118279569892471, | |
| "loss": 0.9539, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 0.00022798259088581668, | |
| "loss": 0.9485, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "eval_bleu": 7.654834398763605, | |
| "eval_loss": 0.9117684364318848, | |
| "eval_runtime": 11.8365, | |
| "eval_samples_per_second": 168.969, | |
| "eval_steps_per_second": 1.352, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 0.00022478238607270864, | |
| "loss": 0.9495, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 0.00022158218125960063, | |
| "loss": 0.9484, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 0.00021838197644649259, | |
| "loss": 0.9486, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 0.00021518177163338455, | |
| "loss": 0.9445, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 0.0002119815668202765, | |
| "loss": 0.9437, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "eval_bleu": 7.606584348844775, | |
| "eval_loss": 0.9072746634483337, | |
| "eval_runtime": 9.3821, | |
| "eval_samples_per_second": 213.171, | |
| "eval_steps_per_second": 1.705, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.00020878136200716847, | |
| "loss": 0.9406, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 0.00020558115719406043, | |
| "loss": 0.9418, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 0.0002023809523809524, | |
| "loss": 0.9402, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 0.00019918074756784435, | |
| "loss": 0.94, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 0.00019598054275473631, | |
| "loss": 0.9393, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "eval_bleu": 7.714033503089272, | |
| "eval_loss": 0.9019351601600647, | |
| "eval_runtime": 9.2531, | |
| "eval_samples_per_second": 216.144, | |
| "eval_steps_per_second": 1.729, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 0.00019278033794162827, | |
| "loss": 0.9378, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 0.00018958013312852024, | |
| "loss": 0.9345, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 0.0001863799283154122, | |
| "loss": 0.936, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 0.00018317972350230416, | |
| "loss": 0.9389, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 0.00017997951868919612, | |
| "loss": 0.9336, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "eval_bleu": 7.80952091204729, | |
| "eval_loss": 0.897022545337677, | |
| "eval_runtime": 9.0616, | |
| "eval_samples_per_second": 220.711, | |
| "eval_steps_per_second": 1.766, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 0.00017677931387608805, | |
| "loss": 0.9376, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 0.00017357910906298001, | |
| "loss": 0.9344, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 0.00017037890424987198, | |
| "loss": 0.9358, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 0.00016717869943676394, | |
| "loss": 0.9309, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.0001639784946236559, | |
| "loss": 0.9368, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "eval_bleu": 7.937668751101428, | |
| "eval_loss": 0.8936744928359985, | |
| "eval_runtime": 9.5691, | |
| "eval_samples_per_second": 209.006, | |
| "eval_steps_per_second": 1.672, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 0.00016077828981054789, | |
| "loss": 0.9337, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 0.00015757808499743985, | |
| "loss": 0.9266, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 0.0001543778801843318, | |
| "loss": 0.9276, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 0.00015117767537122377, | |
| "loss": 0.9266, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 0.00014797747055811573, | |
| "loss": 0.925, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "eval_bleu": 7.842526661794731, | |
| "eval_loss": 0.8898113965988159, | |
| "eval_runtime": 9.4133, | |
| "eval_samples_per_second": 212.466, | |
| "eval_steps_per_second": 1.7, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 0.0001447772657450077, | |
| "loss": 0.9161, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 0.00014157706093189965, | |
| "loss": 0.9263, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 0.0001383768561187916, | |
| "loss": 0.9293, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 0.00013517665130568357, | |
| "loss": 0.924, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 0.00013197644649257554, | |
| "loss": 0.921, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "eval_bleu": 7.900750479934787, | |
| "eval_loss": 0.8863839507102966, | |
| "eval_runtime": 9.7323, | |
| "eval_samples_per_second": 205.501, | |
| "eval_steps_per_second": 1.644, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 0.0001287762416794675, | |
| "loss": 0.9213, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.00012557603686635946, | |
| "loss": 0.9219, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 0.00012237583205325142, | |
| "loss": 0.9141, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 0.00011917562724014337, | |
| "loss": 0.9125, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 0.00011597542242703534, | |
| "loss": 0.9177, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "eval_bleu": 7.913356299834283, | |
| "eval_loss": 0.8835927844047546, | |
| "eval_runtime": 9.3736, | |
| "eval_samples_per_second": 213.364, | |
| "eval_steps_per_second": 1.707, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 0.00019022017409114184, | |
| "loss": 0.9129, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 0.0001876600102406554, | |
| "loss": 0.9121, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 0.00018509984639016898, | |
| "loss": 0.9123, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 0.00018253968253968252, | |
| "loss": 0.9168, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.00017997951868919612, | |
| "loss": 0.9151, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "eval_bleu": 7.864673119868297, | |
| "eval_loss": 0.8821397423744202, | |
| "eval_runtime": 11.1353, | |
| "eval_samples_per_second": 179.609, | |
| "eval_steps_per_second": 1.437, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 0.0001774193548387097, | |
| "loss": 0.9112, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 0.00017485919098822326, | |
| "loss": 0.9175, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 0.00017229902713773683, | |
| "loss": 0.9131, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 0.0001697388632872504, | |
| "loss": 0.9116, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 0.00016717869943676394, | |
| "loss": 0.9104, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "eval_bleu": 8.083046367100428, | |
| "eval_loss": 0.8790320158004761, | |
| "eval_runtime": 9.2669, | |
| "eval_samples_per_second": 215.821, | |
| "eval_steps_per_second": 1.727, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 0.0001646185355862775, | |
| "loss": 0.9083, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 0.0001620583717357911, | |
| "loss": 0.9084, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 0.00015949820788530467, | |
| "loss": 0.9155, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 0.00015693804403481824, | |
| "loss": 0.9059, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 0.0001543778801843318, | |
| "loss": 0.9035, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "eval_bleu": 8.095903306480848, | |
| "eval_loss": 0.8766345381736755, | |
| "eval_runtime": 9.4138, | |
| "eval_samples_per_second": 212.454, | |
| "eval_steps_per_second": 1.7, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 0.00015181771633384535, | |
| "loss": 0.9053, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 0.00014925755248335892, | |
| "loss": 0.9047, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 0.00014669738863287251, | |
| "loss": 0.9017, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 0.00014413722478238608, | |
| "loss": 0.9073, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 0.00014157706093189965, | |
| "loss": 0.8992, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "eval_bleu": 8.01784311685486, | |
| "eval_loss": 0.8740718960762024, | |
| "eval_runtime": 9.3262, | |
| "eval_samples_per_second": 214.449, | |
| "eval_steps_per_second": 1.716, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 0.00013901689708141322, | |
| "loss": 0.904, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 0.0001364567332309268, | |
| "loss": 0.9043, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 0.00013389656938044033, | |
| "loss": 0.9038, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 0.0001313364055299539, | |
| "loss": 0.9061, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 0.0001287762416794675, | |
| "loss": 0.8986, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "eval_bleu": 8.038360794483163, | |
| "eval_loss": 0.8720167875289917, | |
| "eval_runtime": 9.7469, | |
| "eval_samples_per_second": 205.193, | |
| "eval_steps_per_second": 1.642, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 0.00012621607782898107, | |
| "loss": 0.8972, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 0.00012365591397849463, | |
| "loss": 0.9032, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 0.00012109575012800819, | |
| "loss": 0.9009, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 0.00011853558627752177, | |
| "loss": 0.9021, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 0.00011597542242703534, | |
| "loss": 0.894, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "eval_bleu": 8.091339671473314, | |
| "eval_loss": 0.8682767152786255, | |
| "eval_runtime": 9.833, | |
| "eval_samples_per_second": 203.397, | |
| "eval_steps_per_second": 1.627, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 0.0001134152585765489, | |
| "loss": 0.9011, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 0.00011085509472606247, | |
| "loss": 0.9016, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 0.00010829493087557605, | |
| "loss": 0.8951, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 0.0001057347670250896, | |
| "loss": 0.8981, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "learning_rate": 0.00010317460317460317, | |
| "loss": 0.8932, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "eval_bleu": 8.099679129372298, | |
| "eval_loss": 0.8663304448127747, | |
| "eval_runtime": 9.7444, | |
| "eval_samples_per_second": 205.246, | |
| "eval_steps_per_second": 1.642, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 0.00010061443932411675, | |
| "loss": 0.8991, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 9.805427547363031e-05, | |
| "loss": 0.8937, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 9.549411162314388e-05, | |
| "loss": 0.8877, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 9.293394777265746e-05, | |
| "loss": 0.8916, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 9.037378392217102e-05, | |
| "loss": 0.8889, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "eval_bleu": 8.108828400885466, | |
| "eval_loss": 0.8641292452812195, | |
| "eval_runtime": 9.0966, | |
| "eval_samples_per_second": 219.862, | |
| "eval_steps_per_second": 1.759, | |
| "step": 16000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 19530, | |
| "num_train_epochs": 5, | |
| "save_steps": 2000, | |
| "total_flos": 4.675529716059341e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |