| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.6666666666666666, | |
| "eval_steps": 500, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.2757867574691772, | |
| "learning_rate": 0.000499986292341378, | |
| "loss": 5.9015, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7096672654151917, | |
| "learning_rate": 0.0004999451708687113, | |
| "loss": 3.9877, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.6350783109664917, | |
| "learning_rate": 0.0004998766400914329, | |
| "loss": 3.7232, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.5088619589805603, | |
| "learning_rate": 0.0004997807075247146, | |
| "loss": 3.552, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.4739762842655182, | |
| "learning_rate": 0.0004996573836886434, | |
| "loss": 3.5127, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.628942608833313, | |
| "learning_rate": 0.0004995066821070679, | |
| "loss": 3.3798, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.5002753138542175, | |
| "learning_rate": 0.0004993286193061145, | |
| "loss": 3.3999, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.4983285963535309, | |
| "learning_rate": 0.0004991232148123761, | |
| "loss": 3.3398, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.4371359050273895, | |
| "learning_rate": 0.00049889049115077, | |
| "loss": 3.3405, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.5125981569290161, | |
| "learning_rate": 0.0004986304738420684, | |
| "loss": 3.3053, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.47835057973861694, | |
| "learning_rate": 0.000498343191400099, | |
| "loss": 3.2388, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.4769553244113922, | |
| "learning_rate": 0.0004980286753286195, | |
| "loss": 3.2004, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.44102832674980164, | |
| "learning_rate": 0.0004976869601178609, | |
| "loss": 3.2301, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.44598281383514404, | |
| "learning_rate": 0.0004973180832407472, | |
| "loss": 3.2284, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.5305857062339783, | |
| "learning_rate": 0.0004969220851487844, | |
| "loss": 3.1978, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.5015985369682312, | |
| "learning_rate": 0.0004964990092676262, | |
| "loss": 3.198, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.49595481157302856, | |
| "learning_rate": 0.0004960489019923105, | |
| "loss": 3.2582, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.4686504602432251, | |
| "learning_rate": 0.0004955718126821722, | |
| "loss": 3.1974, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.4878421425819397, | |
| "learning_rate": 0.0004950677936554305, | |
| "loss": 3.1467, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.476806640625, | |
| "learning_rate": 0.0004945369001834514, | |
| "loss": 3.1321, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.5151681900024414, | |
| "learning_rate": 0.0004939791904846869, | |
| "loss": 3.1478, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.4726090133190155, | |
| "learning_rate": 0.0004933947257182901, | |
| "loss": 3.1757, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.4868369698524475, | |
| "learning_rate": 0.000492783569977409, | |
| "loss": 3.1573, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.4568879306316376, | |
| "learning_rate": 0.0004921457902821578, | |
| "loss": 3.2129, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.45109352469444275, | |
| "learning_rate": 0.0004914814565722671, | |
| "loss": 3.1417, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.5096346139907837, | |
| "learning_rate": 0.0004907906416994146, | |
| "loss": 3.1397, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.44307172298431396, | |
| "learning_rate": 0.0004900734214192358, | |
| "loss": 3.1851, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.469777375459671, | |
| "learning_rate": 0.0004893298743830168, | |
| "loss": 3.1117, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.4518013596534729, | |
| "learning_rate": 0.0004885600821290692, | |
| "loss": 3.0941, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.4657439589500427, | |
| "learning_rate": 0.0004877641290737884, | |
| "loss": 3.1936, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.4461677372455597, | |
| "learning_rate": 0.00048694210250239646, | |
| "loss": 3.1158, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.454722136259079, | |
| "learning_rate": 0.0004860940925593703, | |
| "loss": 3.0921, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.4858105480670929, | |
| "learning_rate": 0.0004852201922385564, | |
| "loss": 3.1601, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.4848730266094208, | |
| "learning_rate": 0.0004843204973729729, | |
| "loss": 3.1165, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.5170342922210693, | |
| "learning_rate": 0.00048339510662430044, | |
| "loss": 3.0897, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.4945329427719116, | |
| "learning_rate": 0.00048244412147206283, | |
| "loss": 3.0792, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.4752379357814789, | |
| "learning_rate": 0.0004814676462024987, | |
| "loss": 3.035, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.4935299754142761, | |
| "learning_rate": 0.00048046578789712516, | |
| "loss": 3.1129, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.47018569707870483, | |
| "learning_rate": 0.00047943865642099525, | |
| "loss": 3.1255, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.45771998167037964, | |
| "learning_rate": 0.0004783863644106502, | |
| "loss": 3.0546, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.4871847927570343, | |
| "learning_rate": 0.00047730902726176715, | |
| "loss": 3.1176, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.5023790001869202, | |
| "learning_rate": 0.0004762067631165049, | |
| "loss": 3.1396, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.5331537127494812, | |
| "learning_rate": 0.0004750796928505484, | |
| "loss": 3.0541, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.46206673979759216, | |
| "learning_rate": 0.0004739279400598532, | |
| "loss": 3.1429, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.5062265992164612, | |
| "learning_rate": 0.00047275163104709196, | |
| "loss": 3.0668, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.7376482486724854, | |
| "learning_rate": 0.00047155089480780364, | |
| "loss": 3.1377, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.5349658727645874, | |
| "learning_rate": 0.00047032586301624804, | |
| "loss": 3.0838, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.49428579211235046, | |
| "learning_rate": 0.0004690766700109659, | |
| "loss": 3.1257, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.4835917353630066, | |
| "learning_rate": 0.0004678034527800474, | |
| "loss": 3.0709, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.5094427466392517, | |
| "learning_rate": 0.00046650635094610973, | |
| "loss": 3.0692, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.5574087500572205, | |
| "learning_rate": 0.0004651855067509859, | |
| "loss": 3.0667, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.5003999471664429, | |
| "learning_rate": 0.0004638410650401267, | |
| "loss": 3.0311, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.5170697569847107, | |
| "learning_rate": 0.000462473173246716, | |
| "loss": 3.1082, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.5004311800003052, | |
| "learning_rate": 0.00046108198137550377, | |
| "loss": 3.0825, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.49480965733528137, | |
| "learning_rate": 0.00045966764198635603, | |
| "loss": 3.0888, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.49261385202407837, | |
| "learning_rate": 0.00045823031017752484, | |
| "loss": 3.1568, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.4577986001968384, | |
| "learning_rate": 0.00045677014356864043, | |
| "loss": 3.0953, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.4814681112766266, | |
| "learning_rate": 0.000455287302283426, | |
| "loss": 3.0801, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.563042402267456, | |
| "learning_rate": 0.00045378194893213854, | |
| "loss": 3.0719, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.5507716536521912, | |
| "learning_rate": 0.0004522542485937369, | |
| "loss": 3.0724, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.5448073148727417, | |
| "learning_rate": 0.0004507043687977787, | |
| "loss": 3.0294, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.47773197293281555, | |
| "learning_rate": 0.00044913247950604905, | |
| "loss": 3.0682, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.4662880301475525, | |
| "learning_rate": 0.0004475387530939226, | |
| "loss": 3.0106, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.4735540747642517, | |
| "learning_rate": 0.00044592336433145995, | |
| "loss": 3.1345, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.4746765196323395, | |
| "learning_rate": 0.0004442864903642427, | |
| "loss": 3.0772, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.45668378472328186, | |
| "learning_rate": 0.0004426283106939473, | |
| "loss": 3.1174, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.4704893231391907, | |
| "learning_rate": 0.0004409490071586606, | |
| "loss": 3.0816, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.5290801525115967, | |
| "learning_rate": 0.0004392487639129391, | |
| "loss": 3.1089, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.47564446926116943, | |
| "learning_rate": 0.0004375277674076149, | |
| "loss": 3.082, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.4995107352733612, | |
| "learning_rate": 0.00043578620636934855, | |
| "loss": 3.1992, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.48856520652770996, | |
| "learning_rate": 0.00043402427177993366, | |
| "loss": 3.0598, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.4874037802219391, | |
| "learning_rate": 0.00043224215685535287, | |
| "loss": 3.0753, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.45193690061569214, | |
| "learning_rate": 0.00043044005702459054, | |
| "loss": 2.9601, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.46975770592689514, | |
| "learning_rate": 0.00042861816990820087, | |
| "loss": 3.0511, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.5442278981208801, | |
| "learning_rate": 0.00042677669529663686, | |
| "loss": 3.0898, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.5437372922897339, | |
| "learning_rate": 0.00042491583512834137, | |
| "loss": 3.0768, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.4911404848098755, | |
| "learning_rate": 0.0004230357934676017, | |
| "loss": 3.1239, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.4861234128475189, | |
| "learning_rate": 0.0004211367764821722, | |
| "loss": 3.0473, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.49302661418914795, | |
| "learning_rate": 0.0004192189924206652, | |
| "loss": 3.0874, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.4864264726638794, | |
| "learning_rate": 0.0004172826515897146, | |
| "loss": 3.0753, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.48708102107048035, | |
| "learning_rate": 0.00041532796633091297, | |
| "loss": 3.1403, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.5046274662017822, | |
| "learning_rate": 0.0004133551509975264, | |
| "loss": 3.0785, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.5404755473136902, | |
| "learning_rate": 0.00041136442193098765, | |
| "loss": 3.0104, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.5012314915657043, | |
| "learning_rate": 0.00040935599743717243, | |
| "loss": 3.0538, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.5044212937355042, | |
| "learning_rate": 0.0004073300977624594, | |
| "loss": 3.0975, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.5059216618537903, | |
| "learning_rate": 0.0004052869450695776, | |
| "loss": 3.1057, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.4804486930370331, | |
| "learning_rate": 0.00040322676341324415, | |
| "loss": 3.0492, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.5673746466636658, | |
| "learning_rate": 0.00040114977871559375, | |
| "loss": 2.9521, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.4696684181690216, | |
| "learning_rate": 0.00039905621874140396, | |
| "loss": 3.0136, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.5002209544181824, | |
| "learning_rate": 0.0003969463130731183, | |
| "loss": 3.0378, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.48504123091697693, | |
| "learning_rate": 0.0003948202930856697, | |
| "loss": 3.0458, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.48299267888069153, | |
| "learning_rate": 0.000392678391921108, | |
| "loss": 3.0699, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.5440784096717834, | |
| "learning_rate": 0.00039052084446303264, | |
| "loss": 3.044, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.498516321182251, | |
| "learning_rate": 0.000388347887310836, | |
| "loss": 3.0524, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.49369409680366516, | |
| "learning_rate": 0.00038615975875375683, | |
| "loss": 3.0511, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.5007414817810059, | |
| "learning_rate": 0.00038395669874474915, | |
| "loss": 3.0369, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.4814367890357971, | |
| "learning_rate": 0.00038173894887416946, | |
| "loss": 3.0965, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.5295039415359497, | |
| "learning_rate": 0.00037950675234328256, | |
| "loss": 3.0571, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.5096908211708069, | |
| "learning_rate": 0.00037726035393759286, | |
| "loss": 2.9924, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.49976101517677307, | |
| "learning_rate": 0.000375, | |
| "loss": 3.0523, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.486150860786438, | |
| "learning_rate": 0.0003727259384037852, | |
| "loss": 3.0453, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.4886365532875061, | |
| "learning_rate": 0.0003704384185254288, | |
| "loss": 3.0791, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.4869343638420105, | |
| "learning_rate": 0.00036813769121726354, | |
| "loss": 3.0962, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.47605258226394653, | |
| "learning_rate": 0.00036582400877996547, | |
| "loss": 3.087, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.49207133054733276, | |
| "learning_rate": 0.00036349762493488667, | |
| "loss": 3.0406, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.5010517835617065, | |
| "learning_rate": 0.00036115879479623185, | |
| "loss": 3.0638, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.48435714840888977, | |
| "learning_rate": 0.0003588077748430819, | |
| "loss": 3.0325, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.5021737813949585, | |
| "learning_rate": 0.0003564448228912682, | |
| "loss": 3.0115, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.46703141927719116, | |
| "learning_rate": 0.00035407019806510035, | |
| "loss": 2.9539, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.521730363368988, | |
| "learning_rate": 0.0003516841607689501, | |
| "loss": 3.0114, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.4748888611793518, | |
| "learning_rate": 0.00034928697265869515, | |
| "loss": 3.0226, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.4812580645084381, | |
| "learning_rate": 0.00034687889661302575, | |
| "loss": 3.0065, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.5008811354637146, | |
| "learning_rate": 0.00034446019670461683, | |
| "loss": 3.027, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.49032077193260193, | |
| "learning_rate": 0.00034203113817116957, | |
| "loss": 2.9933, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.47988900542259216, | |
| "learning_rate": 0.00033959198738632503, | |
| "loss": 3.0474, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 0.49220654368400574, | |
| "learning_rate": 0.0003371430118304538, | |
| "loss": 3.0226, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 0.49873360991477966, | |
| "learning_rate": 0.0003346844800613229, | |
| "loss": 3.0082, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 0.5125187039375305, | |
| "learning_rate": 0.0003322166616846458, | |
| "loss": 3.0468, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.509082555770874, | |
| "learning_rate": 0.0003297398273245175, | |
| "loss": 3.1134, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.45824429392814636, | |
| "learning_rate": 0.00032725424859373687, | |
| "loss": 3.0516, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.49980753660202026, | |
| "learning_rate": 0.0003247601980640217, | |
| "loss": 2.9742, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.48861539363861084, | |
| "learning_rate": 0.0003222579492361179, | |
| "loss": 3.004, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.5454481840133667, | |
| "learning_rate": 0.00031974777650980735, | |
| "loss": 2.9608, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.4744034707546234, | |
| "learning_rate": 0.0003172299551538164, | |
| "loss": 3.0384, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.5317044854164124, | |
| "learning_rate": 0.00031470476127563017, | |
| "loss": 3.0209, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.49222180247306824, | |
| "learning_rate": 0.0003121724717912138, | |
| "loss": 3.0135, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.4853774309158325, | |
| "learning_rate": 0.00030963336439464523, | |
| "loss": 3.0235, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.4858599603176117, | |
| "learning_rate": 0.00030708771752766396, | |
| "loss": 3.0256, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.4980061948299408, | |
| "learning_rate": 0.0003045358103491357, | |
| "loss": 2.9718, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.49256446957588196, | |
| "learning_rate": 0.0003019779227044398, | |
| "loss": 3.0288, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.5088287591934204, | |
| "learning_rate": 0.00029941433509478153, | |
| "loss": 3.0268, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.5106491446495056, | |
| "learning_rate": 0.0002968453286464312, | |
| "loss": 3.0032, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.4999896287918091, | |
| "learning_rate": 0.0002942711850798959, | |
| "loss": 2.9407, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.4718291163444519, | |
| "learning_rate": 0.00029169218667902556, | |
| "loss": 2.9313, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.485334187746048, | |
| "learning_rate": 0.00028910861626005774, | |
| "loss": 2.9993, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.47278451919555664, | |
| "learning_rate": 0.00028652075714060294, | |
| "loss": 3.0594, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.4961351454257965, | |
| "learning_rate": 0.0002839288931085761, | |
| "loss": 2.9736, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.48107197880744934, | |
| "learning_rate": 0.00028133330839107606, | |
| "loss": 2.9818, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.5018859505653381, | |
| "learning_rate": 0.0002787342876232167, | |
| "loss": 2.9993, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 0.5005733966827393, | |
| "learning_rate": 0.0002761321158169134, | |
| "loss": 2.971, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 0.45412546396255493, | |
| "learning_rate": 0.0002735270783296286, | |
| "loss": 2.9343, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 0.49526095390319824, | |
| "learning_rate": 0.00027091946083307894, | |
| "loss": 3.0247, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.5004629492759705, | |
| "learning_rate": 0.00026830954928190793, | |
| "loss": 2.9386, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.5008641481399536, | |
| "learning_rate": 0.0002656976298823284, | |
| "loss": 2.9768, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.4870932698249817, | |
| "learning_rate": 0.000263083989060736, | |
| "loss": 2.9825, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.47083091735839844, | |
| "learning_rate": 0.0002604689134322999, | |
| "loss": 2.9694, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.4780152440071106, | |
| "learning_rate": 0.00025785268976953206, | |
| "loss": 3.0139, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.47919341921806335, | |
| "learning_rate": 0.00025523560497083924, | |
| "loss": 2.9558, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.4744253158569336, | |
| "learning_rate": 0.00025261794602906147, | |
| "loss": 2.9823, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.47770825028419495, | |
| "learning_rate": 0.00025, | |
| "loss": 2.976, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.535241961479187, | |
| "learning_rate": 0.00024738205397093865, | |
| "loss": 2.9404, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.5206143260002136, | |
| "learning_rate": 0.00024476439502916077, | |
| "loss": 2.9697, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.4685532748699188, | |
| "learning_rate": 0.00024214731023046793, | |
| "loss": 3.031, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.45759207010269165, | |
| "learning_rate": 0.00023953108656770016, | |
| "loss": 2.9371, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.47409725189208984, | |
| "learning_rate": 0.00023691601093926405, | |
| "loss": 2.9492, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.4723811149597168, | |
| "learning_rate": 0.00023430237011767165, | |
| "loss": 3.0055, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.5121801495552063, | |
| "learning_rate": 0.00023169045071809213, | |
| "loss": 2.9654, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.47900766134262085, | |
| "learning_rate": 0.00022908053916692117, | |
| "loss": 2.9665, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.47504156827926636, | |
| "learning_rate": 0.00022647292167037142, | |
| "loss": 2.9543, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.48130717873573303, | |
| "learning_rate": 0.00022386788418308668, | |
| "loss": 2.9948, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.4968855381011963, | |
| "learning_rate": 0.00022126571237678338, | |
| "loss": 3.0497, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.45493414998054504, | |
| "learning_rate": 0.00021866669160892392, | |
| "loss": 3.0211, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.46341362595558167, | |
| "learning_rate": 0.00021607110689142393, | |
| "loss": 2.9249, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.4598373472690582, | |
| "learning_rate": 0.00021347924285939712, | |
| "loss": 2.9306, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.5339600443840027, | |
| "learning_rate": 0.00021089138373994224, | |
| "loss": 2.9457, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.4766974449157715, | |
| "learning_rate": 0.00020830781332097445, | |
| "loss": 2.9573, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.45860227942466736, | |
| "learning_rate": 0.0002057288149201042, | |
| "loss": 2.9067, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.4715401530265808, | |
| "learning_rate": 0.0002031546713535688, | |
| "loss": 2.9373, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.4715852439403534, | |
| "learning_rate": 0.00020058566490521845, | |
| "loss": 2.9486, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.4740593731403351, | |
| "learning_rate": 0.0001980220772955602, | |
| "loss": 2.9672, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.5071026086807251, | |
| "learning_rate": 0.00019546418965086444, | |
| "loss": 2.9791, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.46667274832725525, | |
| "learning_rate": 0.00019291228247233605, | |
| "loss": 2.9517, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.5175219178199768, | |
| "learning_rate": 0.00019036663560535483, | |
| "loss": 2.9765, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.4796634018421173, | |
| "learning_rate": 0.00018782752820878634, | |
| "loss": 2.9258, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.4871932566165924, | |
| "learning_rate": 0.0001852952387243698, | |
| "loss": 2.9403, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.4641266167163849, | |
| "learning_rate": 0.00018277004484618358, | |
| "loss": 2.9058, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.4585517346858978, | |
| "learning_rate": 0.0001802522234901927, | |
| "loss": 2.9802, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.4525165855884552, | |
| "learning_rate": 0.00017774205076388205, | |
| "loss": 3.0241, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.4651874303817749, | |
| "learning_rate": 0.00017523980193597836, | |
| "loss": 2.9783, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.4655090868473053, | |
| "learning_rate": 0.00017274575140626317, | |
| "loss": 2.9115, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.44796276092529297, | |
| "learning_rate": 0.0001702601726754825, | |
| "loss": 2.9357, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.5139617323875427, | |
| "learning_rate": 0.00016778333831535418, | |
| "loss": 2.9124, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.4756507873535156, | |
| "learning_rate": 0.00016531551993867715, | |
| "loss": 2.9252, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.47321391105651855, | |
| "learning_rate": 0.00016285698816954626, | |
| "loss": 2.9801, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.4512488842010498, | |
| "learning_rate": 0.00016040801261367493, | |
| "loss": 3.0134, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.5047795176506042, | |
| "learning_rate": 0.00015796886182883053, | |
| "loss": 2.9549, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.4670908451080322, | |
| "learning_rate": 0.00015553980329538326, | |
| "loss": 2.9534, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.4596426784992218, | |
| "learning_rate": 0.00015312110338697426, | |
| "loss": 2.9802, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.45904773473739624, | |
| "learning_rate": 0.00015071302734130488, | |
| "loss": 2.9435, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.48870381712913513, | |
| "learning_rate": 0.00014831583923105, | |
| "loss": 2.9378, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.4705340564250946, | |
| "learning_rate": 0.00014592980193489974, | |
| "loss": 2.9716, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.45324671268463135, | |
| "learning_rate": 0.00014355517710873183, | |
| "loss": 2.9729, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.4573574662208557, | |
| "learning_rate": 0.00014119222515691815, | |
| "loss": 2.9286, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.48345670104026794, | |
| "learning_rate": 0.00013884120520376818, | |
| "loss": 3.0016, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.4590473175048828, | |
| "learning_rate": 0.00013650237506511331, | |
| "loss": 2.8464, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.7705239057540894, | |
| "learning_rate": 0.00013417599122003462, | |
| "loss": 3.007, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.4770728647708893, | |
| "learning_rate": 0.00013186230878273653, | |
| "loss": 2.9381, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.44512978196144104, | |
| "learning_rate": 0.00012956158147457115, | |
| "loss": 2.9813, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.4624728262424469, | |
| "learning_rate": 0.00012727406159621478, | |
| "loss": 2.9069, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.47390344738960266, | |
| "learning_rate": 0.00012500000000000006, | |
| "loss": 2.9868, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 3000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 1000, | |
| "total_flos": 2.85480780300288e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |