| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.993238166791886, | |
| "eval_steps": 500, | |
| "global_step": 996, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03005259203606311, | |
| "grad_norm": 4.810495218440688, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7926, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06010518407212622, | |
| "grad_norm": 3.155508530359899, | |
| "learning_rate": 5e-06, | |
| "loss": 0.722, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09015777610818933, | |
| "grad_norm": 1.225544271167285, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7085, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.12021036814425244, | |
| "grad_norm": 0.9265761014813488, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6829, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.15026296018031554, | |
| "grad_norm": 1.2106246073988878, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6706, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18031555221637866, | |
| "grad_norm": 1.9380737235516812, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6637, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.21036814425244177, | |
| "grad_norm": 0.8111060604176487, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6536, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.24042073628850488, | |
| "grad_norm": 0.9088296756504328, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6441, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.270473328324568, | |
| "grad_norm": 0.6375258367082703, | |
| "learning_rate": 5e-06, | |
| "loss": 0.636, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3005259203606311, | |
| "grad_norm": 1.2329946577337598, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6293, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3305785123966942, | |
| "grad_norm": 0.5905978981231776, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6255, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.3606311044327573, | |
| "grad_norm": 0.5247850741074309, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6275, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.39068369646882045, | |
| "grad_norm": 0.6903136425418634, | |
| "learning_rate": 5e-06, | |
| "loss": 0.635, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.42073628850488354, | |
| "grad_norm": 0.5428640707055408, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6283, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.4507888805409467, | |
| "grad_norm": 0.7725320921008048, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6265, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.48084147257700977, | |
| "grad_norm": 0.6353293754243331, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6219, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.5108940646130729, | |
| "grad_norm": 0.696998211557625, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6222, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.540946656649136, | |
| "grad_norm": 0.5725933318808639, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6233, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.5709992486851991, | |
| "grad_norm": 0.5406370000400791, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6141, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.6010518407212622, | |
| "grad_norm": 0.5328195355513519, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6135, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6311044327573253, | |
| "grad_norm": 0.5757875631327323, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6177, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6611570247933884, | |
| "grad_norm": 0.511696604236356, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6221, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.6912096168294516, | |
| "grad_norm": 0.5535745253744225, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6093, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.7212622088655146, | |
| "grad_norm": 0.5080323201706937, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6198, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.7513148009015778, | |
| "grad_norm": 0.7440297135028378, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6141, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.7813673929376409, | |
| "grad_norm": 0.5240609426616543, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6001, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.8114199849737039, | |
| "grad_norm": 0.5170330245787976, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6077, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.8414725770097671, | |
| "grad_norm": 0.469645994556069, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6158, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.8715251690458302, | |
| "grad_norm": 0.48882124388027265, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6114, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.9015777610818934, | |
| "grad_norm": 0.5404896464902132, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6075, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9316303531179564, | |
| "grad_norm": 0.5064337097380566, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5973, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.9616829451540195, | |
| "grad_norm": 0.5395666354340096, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6142, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.9917355371900827, | |
| "grad_norm": 0.5443646281574408, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6024, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.9977460555972952, | |
| "eval_loss": 0.602512001991272, | |
| "eval_runtime": 115.6071, | |
| "eval_samples_per_second": 77.547, | |
| "eval_steps_per_second": 0.614, | |
| "step": 332 | |
| }, | |
| { | |
| "epoch": 1.0217881292261457, | |
| "grad_norm": 0.6043688308355755, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5718, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.051840721262209, | |
| "grad_norm": 0.5742469748119633, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5559, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.081893313298272, | |
| "grad_norm": 0.5641150930410421, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5565, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.111945905334335, | |
| "grad_norm": 0.5251750994837489, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5559, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.1419984973703983, | |
| "grad_norm": 0.5594749881743232, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5548, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.1720510894064613, | |
| "grad_norm": 0.5192017193387347, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5616, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.2021036814425243, | |
| "grad_norm": 0.5006444275848907, | |
| "learning_rate": 5e-06, | |
| "loss": 0.553, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.2321562734785876, | |
| "grad_norm": 0.5182596004381461, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5534, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.2622088655146506, | |
| "grad_norm": 0.48792630443547186, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5596, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.2922614575507136, | |
| "grad_norm": 0.4971407424850785, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5559, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.322314049586777, | |
| "grad_norm": 0.4769582306533125, | |
| "learning_rate": 5e-06, | |
| "loss": 0.559, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.35236664162284, | |
| "grad_norm": 0.5696514756007727, | |
| "learning_rate": 5e-06, | |
| "loss": 0.559, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.382419233658903, | |
| "grad_norm": 0.47028266891611087, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5552, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.4124718256949662, | |
| "grad_norm": 0.5142628164984128, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5562, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.4425244177310292, | |
| "grad_norm": 0.4298738340578577, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5444, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.4725770097670925, | |
| "grad_norm": 0.5886177624554544, | |
| "learning_rate": 5e-06, | |
| "loss": 0.553, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.5026296018031555, | |
| "grad_norm": 0.479969456761664, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5615, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.5326821938392188, | |
| "grad_norm": 0.546008279918396, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5588, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.5627347858752816, | |
| "grad_norm": 0.5113023466894115, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5515, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.5927873779113448, | |
| "grad_norm": 0.4638757502509985, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5602, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.622839969947408, | |
| "grad_norm": 0.5396179709473905, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5687, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.6528925619834711, | |
| "grad_norm": 0.5307757961312712, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5522, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.6829451540195342, | |
| "grad_norm": 0.5717833229926992, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5539, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.7129977460555974, | |
| "grad_norm": 0.5148441797108223, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5486, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.7430503380916604, | |
| "grad_norm": 0.4744973222154149, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5515, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.7731029301277235, | |
| "grad_norm": 0.4644006036526441, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5529, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.8031555221637867, | |
| "grad_norm": 0.5742071808595426, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5563, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.8332081141998497, | |
| "grad_norm": 0.524857373914898, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5548, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.8632607062359128, | |
| "grad_norm": 0.5108979951578987, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5517, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.893313298271976, | |
| "grad_norm": 0.5651055404638835, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5566, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.923365890308039, | |
| "grad_norm": 0.4596795240284601, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5565, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.953418482344102, | |
| "grad_norm": 0.5231843841522166, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5543, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.9834710743801653, | |
| "grad_norm": 0.4613490632623934, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5526, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.998497370398197, | |
| "eval_loss": 0.5943902730941772, | |
| "eval_runtime": 116.0841, | |
| "eval_samples_per_second": 77.229, | |
| "eval_steps_per_second": 0.612, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 2.0135236664162286, | |
| "grad_norm": 1.0151560958110475, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5296, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.0435762584522914, | |
| "grad_norm": 0.5753358656630801, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5029, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.0736288504883547, | |
| "grad_norm": 0.535229224584348, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5033, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.103681442524418, | |
| "grad_norm": 0.5308477913499555, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4981, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.1337340345604807, | |
| "grad_norm": 0.5032127296654162, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5025, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.163786626596544, | |
| "grad_norm": 0.5692227883516503, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5052, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.1938392186326072, | |
| "grad_norm": 0.5408850500925536, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4995, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.22389181066867, | |
| "grad_norm": 0.47006049113767784, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5053, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.2539444027047333, | |
| "grad_norm": 0.6577893214388708, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5068, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.2839969947407965, | |
| "grad_norm": 0.5453995235461279, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5025, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.3140495867768593, | |
| "grad_norm": 0.5111820993930344, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5025, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.3441021788129226, | |
| "grad_norm": 0.5019244840316214, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5059, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.374154770848986, | |
| "grad_norm": 0.4860932571878682, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5057, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.4042073628850487, | |
| "grad_norm": 0.5105307181041621, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5104, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.434259954921112, | |
| "grad_norm": 0.4985867132406256, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5078, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.464312546957175, | |
| "grad_norm": 0.5460937499191226, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5091, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.494365138993238, | |
| "grad_norm": 0.4887915411999051, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5084, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.5244177310293012, | |
| "grad_norm": 0.48194363042727245, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5041, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.5544703230653645, | |
| "grad_norm": 0.5346871895488505, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5041, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.5845229151014273, | |
| "grad_norm": 0.5741375180255713, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5044, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.6145755071374905, | |
| "grad_norm": 0.5215520764416209, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5113, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.644628099173554, | |
| "grad_norm": 0.5134970783379779, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5037, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.674680691209617, | |
| "grad_norm": 0.5673341470254984, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5126, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.70473328324568, | |
| "grad_norm": 0.5345905680232906, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5071, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.734785875281743, | |
| "grad_norm": 0.6651222249433006, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5069, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.764838467317806, | |
| "grad_norm": 0.5593066350235659, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5105, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.794891059353869, | |
| "grad_norm": 0.45810869599006515, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5058, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.8249436513899324, | |
| "grad_norm": 0.5097170540621323, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5086, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.8549962434259957, | |
| "grad_norm": 0.61115109062032, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5097, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.8850488354620585, | |
| "grad_norm": 0.5416071063853659, | |
| "learning_rate": 5e-06, | |
| "loss": 0.506, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.9151014274981217, | |
| "grad_norm": 0.5010400411025377, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5084, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 2.945154019534185, | |
| "grad_norm": 0.505134630851122, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5054, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 2.975206611570248, | |
| "grad_norm": 0.5247938588776304, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5067, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 2.993238166791886, | |
| "eval_loss": 0.6004215478897095, | |
| "eval_runtime": 113.1633, | |
| "eval_samples_per_second": 79.222, | |
| "eval_steps_per_second": 0.627, | |
| "step": 996 | |
| }, | |
| { | |
| "epoch": 2.993238166791886, | |
| "step": 996, | |
| "total_flos": 1667918337146880.0, | |
| "train_loss": 0.5658263341490045, | |
| "train_runtime": 16943.5847, | |
| "train_samples_per_second": 30.156, | |
| "train_steps_per_second": 0.059 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 996, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1667918337146880.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |