| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 5.0, |
| "eval_steps": 500, |
| "global_step": 1485045, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.016834506698450217, |
| "grad_norm": 0.43832775950431824, |
| "learning_rate": 5.610298078649668e-05, |
| "loss": 0.5078, |
| "mean_token_accuracy": 0.877261368727684, |
| "step": 5000 |
| }, |
| { |
| "epoch": 0.033669013396900434, |
| "grad_norm": 0.08571778982877731, |
| "learning_rate": 0.00011220596157299336, |
| "loss": 0.4657, |
| "mean_token_accuracy": 0.8832919163525105, |
| "step": 10000 |
| }, |
| { |
| "epoch": 0.05050352009535065, |
| "grad_norm": 0.07018748670816422, |
| "learning_rate": 0.00016830894235949004, |
| "loss": 0.4491, |
| "mean_token_accuracy": 0.8868078841388226, |
| "step": 15000 |
| }, |
| { |
| "epoch": 0.06733802679380087, |
| "grad_norm": 0.19401602447032928, |
| "learning_rate": 0.00022440070030526127, |
| "loss": 0.4456, |
| "mean_token_accuracy": 0.8877104536533356, |
| "step": 20000 |
| }, |
| { |
| "epoch": 0.08417253349225108, |
| "grad_norm": 0.1359330117702484, |
| "learning_rate": 0.00028050368109175795, |
| "loss": 0.4431, |
| "mean_token_accuracy": 0.8878662613034248, |
| "step": 25000 |
| }, |
| { |
| "epoch": 0.1010070401907013, |
| "grad_norm": 0.12549470365047455, |
| "learning_rate": 0.00033658421619680375, |
| "loss": 0.44, |
| "mean_token_accuracy": 0.8886842037558556, |
| "step": 30000 |
| }, |
| { |
| "epoch": 0.11784154688915151, |
| "grad_norm": 0.050832249224185944, |
| "learning_rate": 0.0003926871969833004, |
| "loss": 0.4419, |
| "mean_token_accuracy": 0.8879277787983417, |
| "step": 35000 |
| }, |
| { |
| "epoch": 0.13467605358760174, |
| "grad_norm": 0.2055901736021042, |
| "learning_rate": 0.0004487789549290717, |
| "loss": 0.4407, |
| "mean_token_accuracy": 0.8881710756301879, |
| "step": 40000 |
| }, |
| { |
| "epoch": 0.15151056028605195, |
| "grad_norm": 0.19902074337005615, |
| "learning_rate": 0.0004999998880131883, |
| "loss": 0.4412, |
| "mean_token_accuracy": 0.8881018778681755, |
| "step": 45000 |
| }, |
| { |
| "epoch": 0.16834506698450216, |
| "grad_norm": 0.05337480455636978, |
| "learning_rate": 0.000499982457079418, |
| "loss": 0.4355, |
| "mean_token_accuracy": 0.8895625862658024, |
| "step": 50000 |
| }, |
| { |
| "epoch": 0.18517957368295238, |
| "grad_norm": 0.09246748685836792, |
| "learning_rate": 0.0004999353124630852, |
| "loss": 0.4348, |
| "mean_token_accuracy": 0.8891964164555073, |
| "step": 55000 |
| }, |
| { |
| "epoch": 0.2020140803814026, |
| "grad_norm": 0.19823187589645386, |
| "learning_rate": 0.0004998584602833202, |
| "loss": 0.4319, |
| "mean_token_accuracy": 0.8902513480305672, |
| "step": 60000 |
| }, |
| { |
| "epoch": 0.2188485870798528, |
| "grad_norm": 0.25061216950416565, |
| "learning_rate": 0.0004997519339582924, |
| "loss": 0.43, |
| "mean_token_accuracy": 0.8906152650892735, |
| "step": 65000 |
| }, |
| { |
| "epoch": 0.23568309377830302, |
| "grad_norm": 0.0704982578754425, |
| "learning_rate": 0.0004996157337417631, |
| "loss": 0.4399, |
| "mean_token_accuracy": 0.8881455532193184, |
| "step": 70000 |
| }, |
| { |
| "epoch": 0.25251760047675326, |
| "grad_norm": 0.2511500120162964, |
| "learning_rate": 0.0004994499119661194, |
| "loss": 0.4241, |
| "mean_token_accuracy": 0.8919311520159244, |
| "step": 75000 |
| }, |
| { |
| "epoch": 0.26935210717520347, |
| "grad_norm": 0.11100377887487411, |
| "learning_rate": 0.000499254379887611, |
| "loss": 0.4234, |
| "mean_token_accuracy": 0.8915335232317447, |
| "step": 80000 |
| }, |
| { |
| "epoch": 0.2861866138736537, |
| "grad_norm": 0.26038476824760437, |
| "learning_rate": 0.0004990292211820036, |
| "loss": 0.4268, |
| "mean_token_accuracy": 0.8911488522231579, |
| "step": 85000 |
| }, |
| { |
| "epoch": 0.3030211205721039, |
| "grad_norm": 0.19242601096630096, |
| "learning_rate": 0.0004987745704505701, |
| "loss": 0.4178, |
| "mean_token_accuracy": 0.8933170215010643, |
| "step": 90000 |
| }, |
| { |
| "epoch": 0.3198556272705541, |
| "grad_norm": 0.23704983294010162, |
| "learning_rate": 0.0004984904934310164, |
| "loss": 0.4234, |
| "mean_token_accuracy": 0.8918876139044761, |
| "step": 95000 |
| }, |
| { |
| "epoch": 0.3366901339690043, |
| "grad_norm": 0.3718920052051544, |
| "learning_rate": 0.0004981766649175129, |
| "loss": 0.4239, |
| "mean_token_accuracy": 0.8917478860974312, |
| "step": 100000 |
| }, |
| { |
| "epoch": 0.35352464066745454, |
| "grad_norm": 0.32682642340660095, |
| "learning_rate": 0.0004978332662476849, |
| "loss": 0.4289, |
| "mean_token_accuracy": 0.8902536273121834, |
| "step": 105000 |
| }, |
| { |
| "epoch": 0.37035914736590475, |
| "grad_norm": 0.1268782764673233, |
| "learning_rate": 0.0004974605531122731, |
| "loss": 0.425, |
| "mean_token_accuracy": 0.8918057443618774, |
| "step": 110000 |
| }, |
| { |
| "epoch": 0.38719365406435496, |
| "grad_norm": 0.09693191200494766, |
| "learning_rate": 0.0004970584383322471, |
| "loss": 0.4146, |
| "mean_token_accuracy": 0.8939243561327458, |
| "step": 115000 |
| }, |
| { |
| "epoch": 0.4040281607628052, |
| "grad_norm": 0.15526463091373444, |
| "learning_rate": 0.0004966269696842552, |
| "loss": 0.4121, |
| "mean_token_accuracy": 0.8940373718321324, |
| "step": 120000 |
| }, |
| { |
| "epoch": 0.4208626674612554, |
| "grad_norm": 0.09245841205120087, |
| "learning_rate": 0.0004961661033130288, |
| "loss": 0.4213, |
| "mean_token_accuracy": 0.892039551794529, |
| "step": 125000 |
| }, |
| { |
| "epoch": 0.4376971741597056, |
| "grad_norm": 0.2160220742225647, |
| "learning_rate": 0.0004956761793230503, |
| "loss": 0.417, |
| "mean_token_accuracy": 0.8929712912201881, |
| "step": 130000 |
| }, |
| { |
| "epoch": 0.4545316808581558, |
| "grad_norm": 0.19442440569400787, |
| "learning_rate": 0.0004951568637765692, |
| "loss": 0.4167, |
| "mean_token_accuracy": 0.8933151067972184, |
| "step": 135000 |
| }, |
| { |
| "epoch": 0.47136618755660603, |
| "grad_norm": 0.30567029118537903, |
| "learning_rate": 0.0004946084086378884, |
| "loss": 0.4217, |
| "mean_token_accuracy": 0.8917706328332424, |
| "step": 140000 |
| }, |
| { |
| "epoch": 0.48820069425505624, |
| "grad_norm": 0.05811581388115883, |
| "learning_rate": 0.0004940309975288746, |
| "loss": 0.4244, |
| "mean_token_accuracy": 0.8912954099059105, |
| "step": 145000 |
| }, |
| { |
| "epoch": 0.5050352009535065, |
| "grad_norm": 0.11588570475578308, |
| "learning_rate": 0.0004934247164774502, |
| "loss": 0.4131, |
| "mean_token_accuracy": 0.8940740741789341, |
| "step": 150000 |
| }, |
| { |
| "epoch": 0.5218697076519567, |
| "grad_norm": 0.227882981300354, |
| "learning_rate": 0.0004927892648582125, |
| "loss": 0.4088, |
| "mean_token_accuracy": 0.8952134732306004, |
| "step": 155000 |
| }, |
| { |
| "epoch": 0.5387042143504069, |
| "grad_norm": 0.44141674041748047, |
| "learning_rate": 0.0004921250908311196, |
| "loss": 0.4152, |
| "mean_token_accuracy": 0.893320640039444, |
| "step": 160000 |
| }, |
| { |
| "epoch": 0.5555387210488572, |
| "grad_norm": 0.22222481667995453, |
| "learning_rate": 0.0004914322905998705, |
| "loss": 0.4142, |
| "mean_token_accuracy": 0.8936157721459865, |
| "step": 165000 |
| }, |
| { |
| "epoch": 0.5723732277473074, |
| "grad_norm": 0.4054740369319916, |
| "learning_rate": 0.0004907105219481063, |
| "loss": 0.4041, |
| "mean_token_accuracy": 0.8960370869517327, |
| "step": 170000 |
| }, |
| { |
| "epoch": 0.5892077344457576, |
| "grad_norm": 0.1149463877081871, |
| "learning_rate": 0.0004899602951620096, |
| "loss": 0.4062, |
| "mean_token_accuracy": 0.8956535562157631, |
| "step": 175000 |
| }, |
| { |
| "epoch": 0.6060422411442078, |
| "grad_norm": 0.2042626589536667, |
| "learning_rate": 0.0004891813992064691, |
| "loss": 0.4031, |
| "mean_token_accuracy": 0.8964809771299362, |
| "step": 180000 |
| }, |
| { |
| "epoch": 0.622876747842658, |
| "grad_norm": 0.47134169936180115, |
| "learning_rate": 0.0004883744025752487, |
| "loss": 0.4072, |
| "mean_token_accuracy": 0.8952170867323875, |
| "step": 185000 |
| }, |
| { |
| "epoch": 0.6397112545411082, |
| "grad_norm": 0.23691359162330627, |
| "learning_rate": 0.0004875387552102876, |
| "loss": 0.4077, |
| "mean_token_accuracy": 0.895208574271202, |
| "step": 190000 |
| }, |
| { |
| "epoch": 0.6565457612395584, |
| "grad_norm": 0.06782840192317963, |
| "learning_rate": 0.000486675225003022, |
| "loss": 0.4059, |
| "mean_token_accuracy": 0.8954858981251717, |
| "step": 195000 |
| }, |
| { |
| "epoch": 0.6733802679380086, |
| "grad_norm": 0.10574975609779358, |
| "learning_rate": 0.0004857835858446407, |
| "loss": 0.413, |
| "mean_token_accuracy": 0.8937437287509441, |
| "step": 200000 |
| }, |
| { |
| "epoch": 0.6902147746364589, |
| "grad_norm": 0.15213517844676971, |
| "learning_rate": 0.0004848633831255964, |
| "loss": 0.4078, |
| "mean_token_accuracy": 0.8951162541925907, |
| "step": 205000 |
| }, |
| { |
| "epoch": 0.7070492813349091, |
| "grad_norm": 0.5716095566749573, |
| "learning_rate": 0.00048391583036695317, |
| "loss": 0.41, |
| "mean_token_accuracy": 0.894683657681942, |
| "step": 210000 |
| }, |
| { |
| "epoch": 0.7238837880333593, |
| "grad_norm": 0.14124555885791779, |
| "learning_rate": 0.00048294029842265964, |
| "loss": 0.4127, |
| "mean_token_accuracy": 0.8939531687438488, |
| "step": 215000 |
| }, |
| { |
| "epoch": 0.7407182947318095, |
| "grad_norm": 0.12345622479915619, |
| "learning_rate": 0.0004819372935177282, |
| "loss": 0.4177, |
| "mean_token_accuracy": 0.892583994436264, |
| "step": 220000 |
| }, |
| { |
| "epoch": 0.7575528014302597, |
| "grad_norm": 0.08433724194765091, |
| "learning_rate": 0.00048090632440000395, |
| "loss": 0.4037, |
| "mean_token_accuracy": 0.896090100646019, |
| "step": 225000 |
| }, |
| { |
| "epoch": 0.7743873081287099, |
| "grad_norm": 0.33282095193862915, |
| "learning_rate": 0.0004798479095235695, |
| "loss": 0.4071, |
| "mean_token_accuracy": 0.8953932453036308, |
| "step": 230000 |
| }, |
| { |
| "epoch": 0.7912218148271601, |
| "grad_norm": 0.44560015201568604, |
| "learning_rate": 0.0004787626145192989, |
| "loss": 0.402, |
| "mean_token_accuracy": 0.8966847031533718, |
| "step": 235000 |
| }, |
| { |
| "epoch": 0.8080563215256104, |
| "grad_norm": 0.217354878783226, |
| "learning_rate": 0.00047764969963780247, |
| "loss": 0.4092, |
| "mean_token_accuracy": 0.8946428496301174, |
| "step": 240000 |
| }, |
| { |
| "epoch": 0.8248908282240606, |
| "grad_norm": 0.26110517978668213, |
| "learning_rate": 0.00047650995682050826, |
| "loss": 0.3985, |
| "mean_token_accuracy": 0.8973696547925473, |
| "step": 245000 |
| }, |
| { |
| "epoch": 0.8417253349225108, |
| "grad_norm": 0.2720124423503876, |
| "learning_rate": 0.0004753435376784538, |
| "loss": 0.4047, |
| "mean_token_accuracy": 0.8958372500419617, |
| "step": 250000 |
| }, |
| { |
| "epoch": 0.858559841620961, |
| "grad_norm": 0.4335656464099884, |
| "learning_rate": 0.00047415011395479146, |
| "loss": 0.3946, |
| "mean_token_accuracy": 0.8981286253273487, |
| "step": 255000 |
| }, |
| { |
| "epoch": 0.8753943483194112, |
| "grad_norm": 0.07155684381723404, |
| "learning_rate": 0.00047293030494337707, |
| "loss": 0.4082, |
| "mean_token_accuracy": 0.894729559969902, |
| "step": 260000 |
| }, |
| { |
| "epoch": 0.8922288550178614, |
| "grad_norm": 0.08361949026584625, |
| "learning_rate": 0.0004716835153144877, |
| "loss": 0.3991, |
| "mean_token_accuracy": 0.8968888149738312, |
| "step": 265000 |
| }, |
| { |
| "epoch": 0.9090633617163116, |
| "grad_norm": 0.3156299591064453, |
| "learning_rate": 0.00047041037615928215, |
| "loss": 0.4018, |
| "mean_token_accuracy": 0.896518034273386, |
| "step": 270000 |
| }, |
| { |
| "epoch": 0.9258978684147618, |
| "grad_norm": 0.25520145893096924, |
| "learning_rate": 0.00046911130133339647, |
| "loss": 0.3966, |
| "mean_token_accuracy": 0.8977987481534481, |
| "step": 275000 |
| }, |
| { |
| "epoch": 0.9427323751132121, |
| "grad_norm": 0.07610014826059341, |
| "learning_rate": 0.0004677859254134599, |
| "loss": 0.3985, |
| "mean_token_accuracy": 0.8972881227195263, |
| "step": 280000 |
| }, |
| { |
| "epoch": 0.9595668818116623, |
| "grad_norm": 0.11759631335735321, |
| "learning_rate": 0.0004664354818889862, |
| "loss": 0.41, |
| "mean_token_accuracy": 0.8942620950162411, |
| "step": 285000 |
| }, |
| { |
| "epoch": 0.9764013885101125, |
| "grad_norm": 0.28973931074142456, |
| "learning_rate": 0.0004650585094614364, |
| "loss": 0.3957, |
| "mean_token_accuracy": 0.8978256018638611, |
| "step": 290000 |
| }, |
| { |
| "epoch": 0.9932358952085627, |
| "grad_norm": 0.22663962841033936, |
| "learning_rate": 0.00046365597496245093, |
| "loss": 0.3976, |
| "mean_token_accuracy": 0.8974625573575497, |
| "step": 295000 |
| }, |
| { |
| "epoch": 1.010070401907013, |
| "grad_norm": 0.21033737063407898, |
| "learning_rate": 0.00046222833327006955, |
| "loss": 0.3973, |
| "mean_token_accuracy": 0.8973039018332958, |
| "step": 300000 |
| }, |
| { |
| "epoch": 1.0269049086054631, |
| "grad_norm": 0.29906806349754333, |
| "learning_rate": 0.00046077547599502895, |
| "loss": 0.3872, |
| "mean_token_accuracy": 0.8994352421820164, |
| "step": 305000 |
| }, |
| { |
| "epoch": 1.0437394153039135, |
| "grad_norm": 0.06926748156547546, |
| "learning_rate": 0.00045929787394779565, |
| "loss": 0.389, |
| "mean_token_accuracy": 0.8988563313424587, |
| "step": 310000 |
| }, |
| { |
| "epoch": 1.0605739220023636, |
| "grad_norm": 0.15812338888645172, |
| "learning_rate": 0.0004577945049982541, |
| "loss": 0.3898, |
| "mean_token_accuracy": 0.8987668350994586, |
| "step": 315000 |
| }, |
| { |
| "epoch": 1.0774084287008139, |
| "grad_norm": 0.4079231023788452, |
| "learning_rate": 0.00045626704364846093, |
| "loss": 0.3904, |
| "mean_token_accuracy": 0.898627924913168, |
| "step": 320000 |
| }, |
| { |
| "epoch": 1.094242935399264, |
| "grad_norm": 0.2908093333244324, |
| "learning_rate": 0.00045471507499745666, |
| "loss": 0.3924, |
| "mean_token_accuracy": 0.8983583591341973, |
| "step": 325000 |
| }, |
| { |
| "epoch": 1.1110774420977143, |
| "grad_norm": 0.18580128252506256, |
| "learning_rate": 0.0004531391012448105, |
| "loss": 0.3797, |
| "mean_token_accuracy": 0.9014196767151356, |
| "step": 330000 |
| }, |
| { |
| "epoch": 1.1279119487961644, |
| "grad_norm": 0.0843733623623848, |
| "learning_rate": 0.00045153835626076663, |
| "loss": 0.387, |
| "mean_token_accuracy": 0.8993220765590668, |
| "step": 335000 |
| }, |
| { |
| "epoch": 1.1447464554946147, |
| "grad_norm": 0.24164117872714996, |
| "learning_rate": 0.00044991463838454926, |
| "loss": 0.3907, |
| "mean_token_accuracy": 0.8987605724155903, |
| "step": 340000 |
| }, |
| { |
| "epoch": 1.1615809621930648, |
| "grad_norm": 0.383212685585022, |
| "learning_rate": 0.00044826619061076204, |
| "loss": 0.3945, |
| "mean_token_accuracy": 0.8978823225557804, |
| "step": 345000 |
| }, |
| { |
| "epoch": 1.1784154688915152, |
| "grad_norm": 0.35025081038475037, |
| "learning_rate": 0.00044659518715394576, |
| "loss": 0.3881, |
| "mean_token_accuracy": 0.899101534640789, |
| "step": 350000 |
| }, |
| { |
| "epoch": 1.1952499755899653, |
| "grad_norm": 0.33375903964042664, |
| "learning_rate": 0.0004448994783759701, |
| "loss": 0.3971, |
| "mean_token_accuracy": 0.8972333360612392, |
| "step": 355000 |
| }, |
| { |
| "epoch": 1.2120844822884156, |
| "grad_norm": 0.23400317132472992, |
| "learning_rate": 0.00044318128653991003, |
| "loss": 0.4016, |
| "mean_token_accuracy": 0.8961461176872253, |
| "step": 360000 |
| }, |
| { |
| "epoch": 1.2289189889868657, |
| "grad_norm": 0.19517074525356293, |
| "learning_rate": 0.0004414401421667221, |
| "loss": 0.3908, |
| "mean_token_accuracy": 0.8985273965477943, |
| "step": 365000 |
| }, |
| { |
| "epoch": 1.245753495685316, |
| "grad_norm": 0.08583275973796844, |
| "learning_rate": 0.00043967589694662347, |
| "loss": 0.3825, |
| "mean_token_accuracy": 0.9004765003800392, |
| "step": 370000 |
| }, |
| { |
| "epoch": 1.2625880023837661, |
| "grad_norm": 0.2532650828361511, |
| "learning_rate": 0.0004378894663279621, |
| "loss": 0.3912, |
| "mean_token_accuracy": 0.8984145608305931, |
| "step": 375000 |
| }, |
| { |
| "epoch": 1.2794225090822164, |
| "grad_norm": 0.1332409381866455, |
| "learning_rate": 0.0004360807119116439, |
| "loss": 0.381, |
| "mean_token_accuracy": 0.9012842674553394, |
| "step": 380000 |
| }, |
| { |
| "epoch": 1.2962570157806665, |
| "grad_norm": 0.06636536121368408, |
| "learning_rate": 0.00043424984860204, |
| "loss": 0.3906, |
| "mean_token_accuracy": 0.8986146732866764, |
| "step": 385000 |
| }, |
| { |
| "epoch": 1.3130915224791169, |
| "grad_norm": 0.2454979419708252, |
| "learning_rate": 0.0004323970939303554, |
| "loss": 0.3876, |
| "mean_token_accuracy": 0.8994460028350353, |
| "step": 390000 |
| }, |
| { |
| "epoch": 1.329926029177567, |
| "grad_norm": 0.5010497570037842, |
| "learning_rate": 0.00043052304521672895, |
| "loss": 0.3912, |
| "mean_token_accuracy": 0.8984551406502723, |
| "step": 395000 |
| }, |
| { |
| "epoch": 1.3467605358760173, |
| "grad_norm": 0.05747854709625244, |
| "learning_rate": 0.00042862679360435355, |
| "loss": 0.3829, |
| "mean_token_accuracy": 0.9005064392387867, |
| "step": 400000 |
| }, |
| { |
| "epoch": 1.3635950425744674, |
| "grad_norm": 0.43650171160697937, |
| "learning_rate": 0.0004267096959124643, |
| "loss": 0.3798, |
| "mean_token_accuracy": 0.9010796148121357, |
| "step": 405000 |
| }, |
| { |
| "epoch": 1.3804295492729177, |
| "grad_norm": 0.11899413913488388, |
| "learning_rate": 0.00042477160273012726, |
| "loss": 0.3859, |
| "mean_token_accuracy": 0.8996428173482418, |
| "step": 410000 |
| }, |
| { |
| "epoch": 1.3972640559713678, |
| "grad_norm": 0.08395178616046906, |
| "learning_rate": 0.0004228131383188119, |
| "loss": 0.3812, |
| "mean_token_accuracy": 0.9006698013544082, |
| "step": 415000 |
| }, |
| { |
| "epoch": 1.4140985626698181, |
| "grad_norm": 0.06149202585220337, |
| "learning_rate": 0.00042083295537210423, |
| "loss": 0.3858, |
| "mean_token_accuracy": 0.8997336489975453, |
| "step": 420000 |
| }, |
| { |
| "epoch": 1.4309330693682683, |
| "grad_norm": 0.07448065280914307, |
| "learning_rate": 0.0004188340673762989, |
| "loss": 0.3818, |
| "mean_token_accuracy": 0.9008088066935539, |
| "step": 425000 |
| }, |
| { |
| "epoch": 1.4477675760667186, |
| "grad_norm": 0.08532486110925674, |
| "learning_rate": 0.0004168139173690566, |
| "loss": 0.3862, |
| "mean_token_accuracy": 0.899811751973629, |
| "step": 430000 |
| }, |
| { |
| "epoch": 1.4646020827651687, |
| "grad_norm": 0.3341102600097656, |
| "learning_rate": 0.0004147735296466977, |
| "loss": 0.3853, |
| "mean_token_accuracy": 0.8999572498977184, |
| "step": 435000 |
| }, |
| { |
| "epoch": 1.481436589463619, |
| "grad_norm": 0.30423060059547424, |
| "learning_rate": 0.00041271479075397316, |
| "loss": 0.3818, |
| "mean_token_accuracy": 0.9004268211662769, |
| "step": 440000 |
| }, |
| { |
| "epoch": 1.498271096162069, |
| "grad_norm": 0.09514279663562775, |
| "learning_rate": 0.000410635891328664, |
| "loss": 0.3836, |
| "mean_token_accuracy": 0.9004043155133724, |
| "step": 445000 |
| }, |
| { |
| "epoch": 1.5151056028605194, |
| "grad_norm": 0.28618961572647095, |
| "learning_rate": 0.00040853874170894915, |
| "loss": 0.3825, |
| "mean_token_accuracy": 0.9006597127854824, |
| "step": 450000 |
| }, |
| { |
| "epoch": 1.5319401095589695, |
| "grad_norm": 0.4465504586696625, |
| "learning_rate": 0.00040642191245385137, |
| "loss": 0.39, |
| "mean_token_accuracy": 0.8988076066374778, |
| "step": 455000 |
| }, |
| { |
| "epoch": 1.5487746162574199, |
| "grad_norm": 0.4574894309043884, |
| "learning_rate": 0.0004042869197445462, |
| "loss": 0.3782, |
| "mean_token_accuracy": 0.9019242323577404, |
| "step": 460000 |
| }, |
| { |
| "epoch": 1.5656091229558702, |
| "grad_norm": 0.19813834130764008, |
| "learning_rate": 0.00040213359567242024, |
| "loss": 0.3834, |
| "mean_token_accuracy": 0.9002681724607945, |
| "step": 465000 |
| }, |
| { |
| "epoch": 1.5824436296543203, |
| "grad_norm": 0.19863300025463104, |
| "learning_rate": 0.0003999621960813577, |
| "loss": 0.3717, |
| "mean_token_accuracy": 0.9033599689304829, |
| "step": 470000 |
| }, |
| { |
| "epoch": 1.5992781363527704, |
| "grad_norm": 0.26589977741241455, |
| "learning_rate": 0.0003977729789628577, |
| "loss": 0.3869, |
| "mean_token_accuracy": 0.8996208263397216, |
| "step": 475000 |
| }, |
| { |
| "epoch": 1.6161126430512207, |
| "grad_norm": 0.18696311116218567, |
| "learning_rate": 0.0003955662044253815, |
| "loss": 0.3781, |
| "mean_token_accuracy": 0.9017717719852925, |
| "step": 480000 |
| }, |
| { |
| "epoch": 1.632947149749671, |
| "grad_norm": 0.1331593096256256, |
| "learning_rate": 0.00039334168795853336, |
| "loss": 0.3723, |
| "mean_token_accuracy": 0.9028745406925678, |
| "step": 485000 |
| }, |
| { |
| "epoch": 1.6497816564481211, |
| "grad_norm": 0.19751402735710144, |
| "learning_rate": 0.00039110058384058576, |
| "loss": 0.376, |
| "mean_token_accuracy": 0.9021617008984089, |
| "step": 490000 |
| }, |
| { |
| "epoch": 1.6666161631465712, |
| "grad_norm": 0.10486247390508652, |
| "learning_rate": 0.00038884226165995285, |
| "loss": 0.3832, |
| "mean_token_accuracy": 0.9003694749832153, |
| "step": 495000 |
| }, |
| { |
| "epoch": 1.6834506698450216, |
| "grad_norm": 0.3321278393268585, |
| "learning_rate": 0.0003865678932363376, |
| "loss": 0.3842, |
| "mean_token_accuracy": 0.8999432280719281, |
| "step": 500000 |
| }, |
| { |
| "epoch": 1.7002851765434719, |
| "grad_norm": 0.46075284481048584, |
| "learning_rate": 0.0003842772987087549, |
| "loss": 0.3799, |
| "mean_token_accuracy": 0.9012987192094326, |
| "step": 505000 |
| }, |
| { |
| "epoch": 1.717119683241922, |
| "grad_norm": 0.2548496127128601, |
| "learning_rate": 0.0003819707502306668, |
| "loss": 0.3787, |
| "mean_token_accuracy": 0.9015080606997014, |
| "step": 510000 |
| }, |
| { |
| "epoch": 1.733954189940372, |
| "grad_norm": 0.06035539507865906, |
| "learning_rate": 0.0003796485218510794, |
| "loss": 0.3763, |
| "mean_token_accuracy": 0.9021021844089031, |
| "step": 515000 |
| }, |
| { |
| "epoch": 1.7507886966388224, |
| "grad_norm": 0.1969989836215973, |
| "learning_rate": 0.0003773108894819822, |
| "loss": 0.3666, |
| "mean_token_accuracy": 0.9045280769824982, |
| "step": 520000 |
| }, |
| { |
| "epoch": 1.7676232033372727, |
| "grad_norm": 0.34597352147102356, |
| "learning_rate": 0.00037495765863063325, |
| "loss": 0.3803, |
| "mean_token_accuracy": 0.900929649567604, |
| "step": 525000 |
| }, |
| { |
| "epoch": 1.7844577100357228, |
| "grad_norm": 0.13883334398269653, |
| "learning_rate": 0.0003725905255412233, |
| "loss": 0.3747, |
| "mean_token_accuracy": 0.9023768192231655, |
| "step": 530000 |
| }, |
| { |
| "epoch": 1.801292216734173, |
| "grad_norm": 0.11342539638280869, |
| "learning_rate": 0.0003702078767489786, |
| "loss": 0.3758, |
| "mean_token_accuracy": 0.9022041162252427, |
| "step": 535000 |
| }, |
| { |
| "epoch": 1.8181267234326233, |
| "grad_norm": 0.2931898832321167, |
| "learning_rate": 0.00036781142082031014, |
| "loss": 0.37, |
| "mean_token_accuracy": 0.9033766336143016, |
| "step": 540000 |
| }, |
| { |
| "epoch": 1.8349612301310736, |
| "grad_norm": 0.1063842698931694, |
| "learning_rate": 0.00036540048364393056, |
| "loss": 0.3762, |
| "mean_token_accuracy": 0.9015979734063149, |
| "step": 545000 |
| }, |
| { |
| "epoch": 1.8517957368295237, |
| "grad_norm": 0.3918764293193817, |
| "learning_rate": 0.00036297631622986395, |
| "loss": 0.3737, |
| "mean_token_accuracy": 0.9024290130019188, |
| "step": 550000 |
| }, |
| { |
| "epoch": 1.8686302435279738, |
| "grad_norm": 0.4090248942375183, |
| "learning_rate": 0.00036053921474521407, |
| "loss": 0.3675, |
| "mean_token_accuracy": 0.9041390944600105, |
| "step": 555000 |
| }, |
| { |
| "epoch": 1.8854647502264241, |
| "grad_norm": 0.08517559617757797, |
| "learning_rate": 0.000358088493334252, |
| "loss": 0.3682, |
| "mean_token_accuracy": 0.903897952413559, |
| "step": 560000 |
| }, |
| { |
| "epoch": 1.9022992569248744, |
| "grad_norm": 0.3155289590358734, |
| "learning_rate": 0.00035562443537994805, |
| "loss": 0.3746, |
| "mean_token_accuracy": 0.9024747136831284, |
| "step": 565000 |
| }, |
| { |
| "epoch": 1.9191337636233246, |
| "grad_norm": 0.41365504264831543, |
| "learning_rate": 0.00035314831945614124, |
| "loss": 0.3721, |
| "mean_token_accuracy": 0.903032359880209, |
| "step": 570000 |
| }, |
| { |
| "epoch": 1.9359682703217747, |
| "grad_norm": 0.06398110091686249, |
| "learning_rate": 0.00035065994812357105, |
| "loss": 0.3615, |
| "mean_token_accuracy": 0.9055964763224125, |
| "step": 575000 |
| }, |
| { |
| "epoch": 1.952802777020225, |
| "grad_norm": 0.29439544677734375, |
| "learning_rate": 0.0003481596170342461, |
| "loss": 0.371, |
| "mean_token_accuracy": 0.9032617596328258, |
| "step": 580000 |
| }, |
| { |
| "epoch": 1.9696372837186753, |
| "grad_norm": 0.3945983350276947, |
| "learning_rate": 0.0003456476232611551, |
| "loss": 0.3695, |
| "mean_token_accuracy": 0.9035523663759232, |
| "step": 585000 |
| }, |
| { |
| "epoch": 1.9864717904171254, |
| "grad_norm": 0.3393994867801666, |
| "learning_rate": 0.0003431242652629706, |
| "loss": 0.3733, |
| "mean_token_accuracy": 0.9026642459452152, |
| "step": 590000 |
| }, |
| { |
| "epoch": 2.0033062971155755, |
| "grad_norm": 0.4142962098121643, |
| "learning_rate": 0.00034059035102251346, |
| "loss": 0.3625, |
| "mean_token_accuracy": 0.9053196219623089, |
| "step": 595000 |
| }, |
| { |
| "epoch": 2.020140803814026, |
| "grad_norm": 0.43296197056770325, |
| "learning_rate": 0.00033804414684384483, |
| "loss": 0.344, |
| "mean_token_accuracy": 0.9094575625896454, |
| "step": 600000 |
| }, |
| { |
| "epoch": 2.036975310512476, |
| "grad_norm": 0.29096800088882446, |
| "learning_rate": 0.0003354884981837081, |
| "loss": 0.3584, |
| "mean_token_accuracy": 0.9054997138738632, |
| "step": 605000 |
| }, |
| { |
| "epoch": 2.0538098172109263, |
| "grad_norm": 0.16588035225868225, |
| "learning_rate": 0.00033292269233928235, |
| "loss": 0.3578, |
| "mean_token_accuracy": 0.9057940164029599, |
| "step": 610000 |
| }, |
| { |
| "epoch": 2.0706443239093764, |
| "grad_norm": 0.43818554282188416, |
| "learning_rate": 0.0003303470341628387, |
| "loss": 0.3529, |
| "mean_token_accuracy": 0.9069875589847565, |
| "step": 615000 |
| }, |
| { |
| "epoch": 2.087478830607827, |
| "grad_norm": 0.2641706168651581, |
| "learning_rate": 0.00032776182967723765, |
| "loss": 0.3558, |
| "mean_token_accuracy": 0.9062513264656067, |
| "step": 620000 |
| }, |
| { |
| "epoch": 2.104313337306277, |
| "grad_norm": 0.4995701313018799, |
| "learning_rate": 0.0003251673860395698, |
| "loss": 0.3455, |
| "mean_token_accuracy": 0.9089033137738705, |
| "step": 625000 |
| }, |
| { |
| "epoch": 2.121147844004727, |
| "grad_norm": 0.07126510143280029, |
| "learning_rate": 0.0003225634897483936, |
| "loss": 0.3557, |
| "mean_token_accuracy": 0.9063798456907273, |
| "step": 630000 |
| }, |
| { |
| "epoch": 2.137982350703177, |
| "grad_norm": 0.7193754315376282, |
| "learning_rate": 0.00031995149193819473, |
| "loss": 0.3531, |
| "mean_token_accuracy": 0.9070294671416282, |
| "step": 635000 |
| }, |
| { |
| "epoch": 2.1548168574016278, |
| "grad_norm": 0.1252709925174713, |
| "learning_rate": 0.00031733065786681284, |
| "loss": 0.3558, |
| "mean_token_accuracy": 0.9064443079888821, |
| "step": 640000 |
| }, |
| { |
| "epoch": 2.171651364100078, |
| "grad_norm": 0.3576796054840088, |
| "learning_rate": 0.0003147028741093347, |
| "loss": 0.3506, |
| "mean_token_accuracy": 0.9079141332268715, |
| "step": 645000 |
| }, |
| { |
| "epoch": 2.188485870798528, |
| "grad_norm": 0.0793839767575264, |
| "learning_rate": 0.00031206740585541066, |
| "loss": 0.3543, |
| "mean_token_accuracy": 0.9068539317965507, |
| "step": 650000 |
| }, |
| { |
| "epoch": 2.205320377496978, |
| "grad_norm": 0.6826508641242981, |
| "learning_rate": 0.0003094229773017376, |
| "loss": 0.3551, |
| "mean_token_accuracy": 0.9066114868998527, |
| "step": 655000 |
| }, |
| { |
| "epoch": 2.2221548841954286, |
| "grad_norm": 0.13850310444831848, |
| "learning_rate": 0.0003067725448153842, |
| "loss": 0.3451, |
| "mean_token_accuracy": 0.9091179150283336, |
| "step": 660000 |
| }, |
| { |
| "epoch": 2.2389893908938787, |
| "grad_norm": 0.43323540687561035, |
| "learning_rate": 0.0003041153669865286, |
| "loss": 0.3539, |
| "mean_token_accuracy": 0.9069650263249874, |
| "step": 665000 |
| }, |
| { |
| "epoch": 2.255823897592329, |
| "grad_norm": 0.15886716544628143, |
| "learning_rate": 0.0003014517595236635, |
| "loss": 0.3484, |
| "mean_token_accuracy": 0.9081727708339691, |
| "step": 670000 |
| }, |
| { |
| "epoch": 2.272658404290779, |
| "grad_norm": 0.4662713408470154, |
| "learning_rate": 0.0002987815041508871, |
| "loss": 0.3504, |
| "mean_token_accuracy": 0.9076543515741825, |
| "step": 675000 |
| }, |
| { |
| "epoch": 2.2894929109892295, |
| "grad_norm": 0.32600313425064087, |
| "learning_rate": 0.00029610598643571687, |
| "loss": 0.3448, |
| "mean_token_accuracy": 0.9092563287019729, |
| "step": 680000 |
| }, |
| { |
| "epoch": 2.3063274176876796, |
| "grad_norm": 0.07614242285490036, |
| "learning_rate": 0.0002934255276491496, |
| "loss": 0.3503, |
| "mean_token_accuracy": 0.9077876783370972, |
| "step": 685000 |
| }, |
| { |
| "epoch": 2.3231619243861297, |
| "grad_norm": 0.1248120367527008, |
| "learning_rate": 0.00029073829756738904, |
| "loss": 0.3463, |
| "mean_token_accuracy": 0.9089062856316567, |
| "step": 690000 |
| }, |
| { |
| "epoch": 2.3399964310845798, |
| "grad_norm": 0.058201707899570465, |
| "learning_rate": 0.0002880478399859927, |
| "loss": 0.3428, |
| "mean_token_accuracy": 0.9097757088422775, |
| "step": 695000 |
| }, |
| { |
| "epoch": 2.3568309377830303, |
| "grad_norm": 0.16684022545814514, |
| "learning_rate": 0.00028535178411815044, |
| "loss": 0.3507, |
| "mean_token_accuracy": 0.9075722501158714, |
| "step": 700000 |
| }, |
| { |
| "epoch": 2.3736654444814804, |
| "grad_norm": 0.34332218766212463, |
| "learning_rate": 0.0002826515263015769, |
| "loss": 0.3489, |
| "mean_token_accuracy": 0.9079888770639897, |
| "step": 705000 |
| }, |
| { |
| "epoch": 2.3904999511799305, |
| "grad_norm": 0.054960548877716064, |
| "learning_rate": 0.0002799479287942707, |
| "loss": 0.3482, |
| "mean_token_accuracy": 0.9081453487992287, |
| "step": 710000 |
| }, |
| { |
| "epoch": 2.4073344578783806, |
| "grad_norm": 0.26335740089416504, |
| "learning_rate": 0.0002772407730699143, |
| "loss": 0.3396, |
| "mean_token_accuracy": 0.9104099054038525, |
| "step": 715000 |
| }, |
| { |
| "epoch": 2.424168964576831, |
| "grad_norm": 0.6300459504127502, |
| "learning_rate": 0.00027452983817729626, |
| "loss": 0.3446, |
| "mean_token_accuracy": 0.9092252513706685, |
| "step": 720000 |
| }, |
| { |
| "epoch": 2.4410034712752813, |
| "grad_norm": 0.16192521154880524, |
| "learning_rate": 0.00027181653079211054, |
| "loss": 0.3475, |
| "mean_token_accuracy": 0.9086122831404209, |
| "step": 725000 |
| }, |
| { |
| "epoch": 2.4578379779737314, |
| "grad_norm": 0.18668274581432343, |
| "learning_rate": 0.0002691000876743106, |
| "loss": 0.3468, |
| "mean_token_accuracy": 0.9085833285450935, |
| "step": 730000 |
| }, |
| { |
| "epoch": 2.4746724846721815, |
| "grad_norm": 0.0640130266547203, |
| "learning_rate": 0.00026638191835775063, |
| "loss": 0.3408, |
| "mean_token_accuracy": 0.9105134047031402, |
| "step": 735000 |
| }, |
| { |
| "epoch": 2.491506991370632, |
| "grad_norm": 0.42377156019210815, |
| "learning_rate": 0.00026366180264878337, |
| "loss": 0.3408, |
| "mean_token_accuracy": 0.9101354426503181, |
| "step": 740000 |
| }, |
| { |
| "epoch": 2.508341498069082, |
| "grad_norm": 0.5703943371772766, |
| "learning_rate": 0.00026094060844017333, |
| "loss": 0.3424, |
| "mean_token_accuracy": 0.9096509917736053, |
| "step": 745000 |
| }, |
| { |
| "epoch": 2.5251760047675322, |
| "grad_norm": 0.35781392455101013, |
| "learning_rate": 0.00025821702499192583, |
| "loss": 0.342, |
| "mean_token_accuracy": 0.9097659908413886, |
| "step": 750000 |
| }, |
| { |
| "epoch": 2.542010511465983, |
| "grad_norm": 0.46854162216186523, |
| "learning_rate": 0.00025549300995690973, |
| "loss": 0.3389, |
| "mean_token_accuracy": 0.9106320847570896, |
| "step": 755000 |
| }, |
| { |
| "epoch": 2.558845018164433, |
| "grad_norm": 0.7103300094604492, |
| "learning_rate": 0.00025276834227837297, |
| "loss": 0.3415, |
| "mean_token_accuracy": 0.909944889754057, |
| "step": 760000 |
| }, |
| { |
| "epoch": 2.575679524862883, |
| "grad_norm": 1.0073944330215454, |
| "learning_rate": 0.00025004280045479263, |
| "loss": 0.3447, |
| "mean_token_accuracy": 0.9091400749325752, |
| "step": 765000 |
| }, |
| { |
| "epoch": 2.592514031561333, |
| "grad_norm": 0.12742860615253448, |
| "learning_rate": 0.000247317253543907, |
| "loss": 0.3466, |
| "mean_token_accuracy": 0.9088619047820569, |
| "step": 770000 |
| }, |
| { |
| "epoch": 2.609348538259783, |
| "grad_norm": 0.5487165451049805, |
| "learning_rate": 0.0002445925706081293, |
| "loss": 0.3372, |
| "mean_token_accuracy": 0.9113640814960003, |
| "step": 775000 |
| }, |
| { |
| "epoch": 2.6261830449582337, |
| "grad_norm": 0.08162357658147812, |
| "learning_rate": 0.00024186853014774928, |
| "loss": 0.34, |
| "mean_token_accuracy": 0.9104940860927105, |
| "step": 780000 |
| }, |
| { |
| "epoch": 2.643017551656684, |
| "grad_norm": 0.3550926148891449, |
| "learning_rate": 0.0002391454558154401, |
| "loss": 0.3407, |
| "mean_token_accuracy": 0.910196255683899, |
| "step": 785000 |
| }, |
| { |
| "epoch": 2.659852058355134, |
| "grad_norm": 0.3737865388393402, |
| "learning_rate": 0.00023642367114908592, |
| "loss": 0.3339, |
| "mean_token_accuracy": 0.9117865429580212, |
| "step": 790000 |
| }, |
| { |
| "epoch": 2.6766865650535845, |
| "grad_norm": 0.11166583746671677, |
| "learning_rate": 0.00023370349953334127, |
| "loss": 0.3368, |
| "mean_token_accuracy": 0.9109085723817348, |
| "step": 795000 |
| }, |
| { |
| "epoch": 2.6935210717520346, |
| "grad_norm": 0.38489484786987305, |
| "learning_rate": 0.00023098526416120868, |
| "loss": 0.3423, |
| "mean_token_accuracy": 0.9094887253403664, |
| "step": 800000 |
| }, |
| { |
| "epoch": 2.7103555784504847, |
| "grad_norm": 0.15662796795368195, |
| "learning_rate": 0.00022826928799563856, |
| "loss": 0.3313, |
| "mean_token_accuracy": 0.9125971922457218, |
| "step": 805000 |
| }, |
| { |
| "epoch": 2.727190085148935, |
| "grad_norm": 0.37235721945762634, |
| "learning_rate": 0.00022555535111500437, |
| "loss": 0.3379, |
| "mean_token_accuracy": 0.9107684759140015, |
| "step": 810000 |
| }, |
| { |
| "epoch": 2.744024591847385, |
| "grad_norm": 0.12951402366161346, |
| "learning_rate": 0.0002228459457584836, |
| "loss": 0.3357, |
| "mean_token_accuracy": 0.9114405186533928, |
| "step": 815000 |
| }, |
| { |
| "epoch": 2.7608590985458354, |
| "grad_norm": 0.11648359149694443, |
| "learning_rate": 0.00022013922276207525, |
| "loss": 0.335, |
| "mean_token_accuracy": 0.9110855512082576, |
| "step": 820000 |
| }, |
| { |
| "epoch": 2.7776936052442855, |
| "grad_norm": 0.09333443641662598, |
| "learning_rate": 0.00021743496504147974, |
| "loss": 0.3322, |
| "mean_token_accuracy": 0.9121826777040959, |
| "step": 825000 |
| }, |
| { |
| "epoch": 2.7945281119427356, |
| "grad_norm": 0.07768220454454422, |
| "learning_rate": 0.00021473511781076193, |
| "loss": 0.3343, |
| "mean_token_accuracy": 0.9116938260793686, |
| "step": 830000 |
| }, |
| { |
| "epoch": 2.811362618641186, |
| "grad_norm": 0.37164732813835144, |
| "learning_rate": 0.00021203892161615075, |
| "loss": 0.3365, |
| "mean_token_accuracy": 0.9111581486284733, |
| "step": 835000 |
| }, |
| { |
| "epoch": 2.8281971253396363, |
| "grad_norm": 0.3348078727722168, |
| "learning_rate": 0.00020934777548558411, |
| "loss": 0.3247, |
| "mean_token_accuracy": 0.9142998357057571, |
| "step": 840000 |
| }, |
| { |
| "epoch": 2.8450316320380864, |
| "grad_norm": 0.4442961812019348, |
| "learning_rate": 0.00020666199636030407, |
| "loss": 0.3366, |
| "mean_token_accuracy": 0.9112150843858718, |
| "step": 845000 |
| }, |
| { |
| "epoch": 2.8618661387365365, |
| "grad_norm": 0.13652917742729187, |
| "learning_rate": 0.0002039802924893582, |
| "loss": 0.3377, |
| "mean_token_accuracy": 0.9109872903048992, |
| "step": 850000 |
| }, |
| { |
| "epoch": 2.8787006454349866, |
| "grad_norm": 0.06294421851634979, |
| "learning_rate": 0.0002013045933529269, |
| "loss": 0.3209, |
| "mean_token_accuracy": 0.9151487466335296, |
| "step": 855000 |
| }, |
| { |
| "epoch": 2.895535152133437, |
| "grad_norm": 0.10965436697006226, |
| "learning_rate": 0.00019863414628997953, |
| "loss": 0.3237, |
| "mean_token_accuracy": 0.9145387828052044, |
| "step": 860000 |
| }, |
| { |
| "epoch": 2.9123696588318873, |
| "grad_norm": 0.0993124321103096, |
| "learning_rate": 0.0001959708693110222, |
| "loss": 0.3263, |
| "mean_token_accuracy": 0.9138095196843147, |
| "step": 865000 |
| }, |
| { |
| "epoch": 2.9292041655303374, |
| "grad_norm": 0.3972514867782593, |
| "learning_rate": 0.00019331294709624576, |
| "loss": 0.3337, |
| "mean_token_accuracy": 0.9120419765353203, |
| "step": 870000 |
| }, |
| { |
| "epoch": 2.946038672228788, |
| "grad_norm": 0.48021000623703003, |
| "learning_rate": 0.00019066282205831424, |
| "loss": 0.3259, |
| "mean_token_accuracy": 0.9137129559576511, |
| "step": 875000 |
| }, |
| { |
| "epoch": 2.962873178927238, |
| "grad_norm": 0.25525978207588196, |
| "learning_rate": 0.00018801868784193055, |
| "loss": 0.3304, |
| "mean_token_accuracy": 0.9126883270680904, |
| "step": 880000 |
| }, |
| { |
| "epoch": 2.979707685625688, |
| "grad_norm": 0.3702300190925598, |
| "learning_rate": 0.00018538297418971828, |
| "loss": 0.3269, |
| "mean_token_accuracy": 0.9134419854700565, |
| "step": 885000 |
| }, |
| { |
| "epoch": 2.996542192324138, |
| "grad_norm": 0.30407169461250305, |
| "learning_rate": 0.0001827544097035622, |
| "loss": 0.3253, |
| "mean_token_accuracy": 0.9139238992393017, |
| "step": 890000 |
| }, |
| { |
| "epoch": 3.0133766990225888, |
| "grad_norm": 0.3653494417667389, |
| "learning_rate": 0.00018013331139422713, |
| "loss": 0.3065, |
| "mean_token_accuracy": 0.9179954954862595, |
| "step": 895000 |
| }, |
| { |
| "epoch": 3.030211205721039, |
| "grad_norm": 0.4035376012325287, |
| "learning_rate": 0.00017752103932214578, |
| "loss": 0.3056, |
| "mean_token_accuracy": 0.9181765572071076, |
| "step": 900000 |
| }, |
| { |
| "epoch": 3.047045712419489, |
| "grad_norm": 0.07252100110054016, |
| "learning_rate": 0.00017491737872618493, |
| "loss": 0.3158, |
| "mean_token_accuracy": 0.9155188336074352, |
| "step": 905000 |
| }, |
| { |
| "epoch": 3.063880219117939, |
| "grad_norm": 0.06896106898784637, |
| "learning_rate": 0.0001723221207138564, |
| "loss": 0.3093, |
| "mean_token_accuracy": 0.917340937447548, |
| "step": 910000 |
| }, |
| { |
| "epoch": 3.0807147258163896, |
| "grad_norm": 0.8066933751106262, |
| "learning_rate": 0.0001697371283024099, |
| "loss": 0.306, |
| "mean_token_accuracy": 0.918012701535225, |
| "step": 915000 |
| }, |
| { |
| "epoch": 3.0975492325148397, |
| "grad_norm": 0.14338815212249756, |
| "learning_rate": 0.00016716063953168587, |
| "loss": 0.3087, |
| "mean_token_accuracy": 0.9175435605287552, |
| "step": 920000 |
| }, |
| { |
| "epoch": 3.11438373921329, |
| "grad_norm": 0.2732289135456085, |
| "learning_rate": 0.0001645950219836876, |
| "loss": 0.3045, |
| "mean_token_accuracy": 0.9187468481361866, |
| "step": 925000 |
| }, |
| { |
| "epoch": 3.13121824591174, |
| "grad_norm": 0.18121260404586792, |
| "learning_rate": 0.00016203903727026982, |
| "loss": 0.3088, |
| "mean_token_accuracy": 0.9177974644362926, |
| "step": 930000 |
| }, |
| { |
| "epoch": 3.1480527526101905, |
| "grad_norm": 0.657842218875885, |
| "learning_rate": 0.00015949350350320942, |
| "loss": 0.3053, |
| "mean_token_accuracy": 0.9184000668883323, |
| "step": 935000 |
| }, |
| { |
| "epoch": 3.1648872593086406, |
| "grad_norm": 0.07914489507675171, |
| "learning_rate": 0.00015695922918946499, |
| "loss": 0.3024, |
| "mean_token_accuracy": 0.9192089639782905, |
| "step": 940000 |
| }, |
| { |
| "epoch": 3.1817217660070907, |
| "grad_norm": 0.7750961780548096, |
| "learning_rate": 0.00015443499730521384, |
| "loss": 0.3066, |
| "mean_token_accuracy": 0.9179883925318718, |
| "step": 945000 |
| }, |
| { |
| "epoch": 3.1985562727055408, |
| "grad_norm": 0.3910292983055115, |
| "learning_rate": 0.00015192212437361126, |
| "loss": 0.3072, |
| "mean_token_accuracy": 0.9181103789508342, |
| "step": 950000 |
| }, |
| { |
| "epoch": 3.2153907794039913, |
| "grad_norm": 0.2133556455373764, |
| "learning_rate": 0.0001494219073927941, |
| "loss": 0.304, |
| "mean_token_accuracy": 0.9188019089162349, |
| "step": 955000 |
| }, |
| { |
| "epoch": 3.2322252861024414, |
| "grad_norm": 0.1028456836938858, |
| "learning_rate": 0.0001469331389239014, |
| "loss": 0.3086, |
| "mean_token_accuracy": 0.9175772596240044, |
| "step": 960000 |
| }, |
| { |
| "epoch": 3.2490597928008915, |
| "grad_norm": 0.08258257061243057, |
| "learning_rate": 0.0001444566161854202, |
| "loss": 0.3006, |
| "mean_token_accuracy": 0.9199309858679772, |
| "step": 965000 |
| }, |
| { |
| "epoch": 3.2658942994993416, |
| "grad_norm": 0.4385560154914856, |
| "learning_rate": 0.0001419926334215865, |
| "loss": 0.3069, |
| "mean_token_accuracy": 0.9181264399349689, |
| "step": 970000 |
| }, |
| { |
| "epoch": 3.282728806197792, |
| "grad_norm": 0.06624668091535568, |
| "learning_rate": 0.00013954099426403933, |
| "loss": 0.2979, |
| "mean_token_accuracy": 0.9204157708227635, |
| "step": 975000 |
| }, |
| { |
| "epoch": 3.2995633128962423, |
| "grad_norm": 0.12009251117706299, |
| "learning_rate": 0.00013710345731043452, |
| "loss": 0.3101, |
| "mean_token_accuracy": 0.9169986567556858, |
| "step": 980000 |
| }, |
| { |
| "epoch": 3.3163978195946924, |
| "grad_norm": 0.05784051492810249, |
| "learning_rate": 0.0001346788448630482, |
| "loss": 0.3051, |
| "mean_token_accuracy": 0.9185231245040893, |
| "step": 985000 |
| }, |
| { |
| "epoch": 3.3332323262931425, |
| "grad_norm": 0.32644739747047424, |
| "learning_rate": 0.00013226841510668264, |
| "loss": 0.2997, |
| "mean_token_accuracy": 0.9198516924262047, |
| "step": 990000 |
| }, |
| { |
| "epoch": 3.350066832991593, |
| "grad_norm": 0.1706833839416504, |
| "learning_rate": 0.00012987101153340202, |
| "loss": 0.299, |
| "mean_token_accuracy": 0.920125166630745, |
| "step": 995000 |
| }, |
| { |
| "epoch": 3.366901339690043, |
| "grad_norm": 0.09666553884744644, |
| "learning_rate": 0.00012748836188645674, |
| "loss": 0.2971, |
| "mean_token_accuracy": 0.9206091440320014, |
| "step": 1000000 |
| }, |
| { |
| "epoch": 3.3837358463884932, |
| "grad_norm": 0.33264678716659546, |
| "learning_rate": 0.0001251202682711683, |
| "loss": 0.2999, |
| "mean_token_accuracy": 0.9198792561650276, |
| "step": 1005000 |
| }, |
| { |
| "epoch": 3.4005703530869433, |
| "grad_norm": 0.1303383708000183, |
| "learning_rate": 0.00012276701204893368, |
| "loss": 0.3052, |
| "mean_token_accuracy": 0.9185852432012558, |
| "step": 1010000 |
| }, |
| { |
| "epoch": 3.417404859785394, |
| "grad_norm": 0.3467031419277191, |
| "learning_rate": 0.00012042887281826751, |
| "loss": 0.2994, |
| "mean_token_accuracy": 0.9200311563968658, |
| "step": 1015000 |
| }, |
| { |
| "epoch": 3.434239366483844, |
| "grad_norm": 0.09201651811599731, |
| "learning_rate": 0.000118105665205647, |
| "loss": 0.305, |
| "mean_token_accuracy": 0.9185415825903416, |
| "step": 1020000 |
| }, |
| { |
| "epoch": 3.451073873182294, |
| "grad_norm": 0.7123315334320068, |
| "learning_rate": 0.00011579859469914544, |
| "loss": 0.2927, |
| "mean_token_accuracy": 0.9218706982314586, |
| "step": 1025000 |
| }, |
| { |
| "epoch": 3.467908379880744, |
| "grad_norm": 0.5221385359764099, |
| "learning_rate": 0.00011350746912599078, |
| "loss": 0.2965, |
| "mean_token_accuracy": 0.9209587279498577, |
| "step": 1030000 |
| }, |
| { |
| "epoch": 3.4847428865791947, |
| "grad_norm": 0.3537195026874542, |
| "learning_rate": 0.00011123301422639786, |
| "loss": 0.2918, |
| "mean_token_accuracy": 0.9218817425847053, |
| "step": 1035000 |
| }, |
| { |
| "epoch": 3.501577393277645, |
| "grad_norm": 0.05311386287212372, |
| "learning_rate": 0.00010897458991708217, |
| "loss": 0.294, |
| "mean_token_accuracy": 0.9215122280180454, |
| "step": 1040000 |
| }, |
| { |
| "epoch": 3.518411899976095, |
| "grad_norm": 0.13352271914482117, |
| "learning_rate": 0.00010673292132473498, |
| "loss": 0.2958, |
| "mean_token_accuracy": 0.9209851972579957, |
| "step": 1045000 |
| }, |
| { |
| "epoch": 3.5352464066745455, |
| "grad_norm": 0.10409609228372574, |
| "learning_rate": 0.00010450783140330408, |
| "loss": 0.2888, |
| "mean_token_accuracy": 0.9227346060037613, |
| "step": 1050000 |
| }, |
| { |
| "epoch": 3.5520809133729956, |
| "grad_norm": 0.17737708985805511, |
| "learning_rate": 0.00010230047472862142, |
| "loss": 0.2982, |
| "mean_token_accuracy": 0.9203281805872917, |
| "step": 1055000 |
| }, |
| { |
| "epoch": 3.5689154200714457, |
| "grad_norm": 0.08944541215896606, |
| "learning_rate": 0.00010011066674562761, |
| "loss": 0.2952, |
| "mean_token_accuracy": 0.9211396588087082, |
| "step": 1060000 |
| }, |
| { |
| "epoch": 3.585749926769896, |
| "grad_norm": 0.645271897315979, |
| "learning_rate": 9.793866763298673e-05, |
| "loss": 0.2904, |
| "mean_token_accuracy": 0.9224357854366303, |
| "step": 1065000 |
| }, |
| { |
| "epoch": 3.602584433468346, |
| "grad_norm": 0.06127174198627472, |
| "learning_rate": 9.578430631949656e-05, |
| "loss": 0.2955, |
| "mean_token_accuracy": 0.9209574770987033, |
| "step": 1070000 |
| }, |
| { |
| "epoch": 3.6194189401667964, |
| "grad_norm": 0.07396731525659561, |
| "learning_rate": 9.364870068065631e-05, |
| "loss": 0.2874, |
| "mean_token_accuracy": 0.9232437397360802, |
| "step": 1075000 |
| }, |
| { |
| "epoch": 3.6362534468652465, |
| "grad_norm": 0.10846959054470062, |
| "learning_rate": 9.153167168071757e-05, |
| "loss": 0.2925, |
| "mean_token_accuracy": 0.9220283030092716, |
| "step": 1080000 |
| }, |
| { |
| "epoch": 3.6530879535636966, |
| "grad_norm": 0.08513388782739639, |
| "learning_rate": 8.943347085122172e-05, |
| "loss": 0.2905, |
| "mean_token_accuracy": 0.9222006689667702, |
| "step": 1085000 |
| }, |
| { |
| "epoch": 3.669922460262147, |
| "grad_norm": 0.1992253065109253, |
| "learning_rate": 8.73543474866696e-05, |
| "loss": 0.2957, |
| "mean_token_accuracy": 0.9210433046281338, |
| "step": 1090000 |
| }, |
| { |
| "epoch": 3.6867569669605973, |
| "grad_norm": 0.28429633378982544, |
| "learning_rate": 8.52945486149028e-05, |
| "loss": 0.2983, |
| "mean_token_accuracy": 0.9202738110303879, |
| "step": 1095000 |
| }, |
| { |
| "epoch": 3.7035914736590474, |
| "grad_norm": 0.09360179305076599, |
| "learning_rate": 8.325391273284766e-05, |
| "loss": 0.2936, |
| "mean_token_accuracy": 0.9216144227623939, |
| "step": 1100000 |
| }, |
| { |
| "epoch": 3.7204259803574975, |
| "grad_norm": 0.06703831255435944, |
| "learning_rate": 8.123390095196376e-05, |
| "loss": 0.2928, |
| "mean_token_accuracy": 0.9216986531615258, |
| "step": 1105000 |
| }, |
| { |
| "epoch": 3.7372604870559476, |
| "grad_norm": 0.08141499012708664, |
| "learning_rate": 7.923313640930005e-05, |
| "loss": 0.2939, |
| "mean_token_accuracy": 0.9216276071667672, |
| "step": 1110000 |
| }, |
| { |
| "epoch": 3.754094993754398, |
| "grad_norm": 0.656367301940918, |
| "learning_rate": 7.725306351550637e-05, |
| "loss": 0.291, |
| "mean_token_accuracy": 0.9220448384821415, |
| "step": 1115000 |
| }, |
| { |
| "epoch": 3.7709295004528482, |
| "grad_norm": 0.5644939541816711, |
| "learning_rate": 7.52939052810331e-05, |
| "loss": 0.2842, |
| "mean_token_accuracy": 0.9237679554998874, |
| "step": 1120000 |
| }, |
| { |
| "epoch": 3.7877640071512984, |
| "grad_norm": 0.7148278951644897, |
| "learning_rate": 7.335511035424089e-05, |
| "loss": 0.2883, |
| "mean_token_accuracy": 0.9227848765075207, |
| "step": 1125000 |
| }, |
| { |
| "epoch": 3.804598513849749, |
| "grad_norm": 0.27569320797920227, |
| "learning_rate": 7.143692161840876e-05, |
| "loss": 0.2928, |
| "mean_token_accuracy": 0.9220670664846897, |
| "step": 1130000 |
| }, |
| { |
| "epoch": 3.821433020548199, |
| "grad_norm": 0.3058241903781891, |
| "learning_rate": 6.954033440156137e-05, |
| "loss": 0.2897, |
| "mean_token_accuracy": 0.9225042996287346, |
| "step": 1135000 |
| }, |
| { |
| "epoch": 3.838267527246649, |
| "grad_norm": 0.07972966134548187, |
| "learning_rate": 6.766518822203652e-05, |
| "loss": 0.2853, |
| "mean_token_accuracy": 0.9237662140965461, |
| "step": 1140000 |
| }, |
| { |
| "epoch": 3.855102033945099, |
| "grad_norm": 0.09462402015924454, |
| "learning_rate": 6.581170587244034e-05, |
| "loss": 0.2882, |
| "mean_token_accuracy": 0.9229260991454125, |
| "step": 1145000 |
| }, |
| { |
| "epoch": 3.8719365406435493, |
| "grad_norm": 0.08020074665546417, |
| "learning_rate": 6.398010757142466e-05, |
| "loss": 0.2829, |
| "mean_token_accuracy": 0.9245367368340492, |
| "step": 1150000 |
| }, |
| { |
| "epoch": 3.888771047342, |
| "grad_norm": 1.1085759401321411, |
| "learning_rate": 6.217025111925742e-05, |
| "loss": 0.2924, |
| "mean_token_accuracy": 0.9220222983837127, |
| "step": 1155000 |
| }, |
| { |
| "epoch": 3.90560555404045, |
| "grad_norm": 0.2981172204017639, |
| "learning_rate": 6.0383430963287275e-05, |
| "loss": 0.2829, |
| "mean_token_accuracy": 0.9242312931597233, |
| "step": 1160000 |
| }, |
| { |
| "epoch": 3.9224400607389, |
| "grad_norm": 0.17800085246562958, |
| "learning_rate": 5.8618779989759284e-05, |
| "loss": 0.2848, |
| "mean_token_accuracy": 0.9239188097476959, |
| "step": 1165000 |
| }, |
| { |
| "epoch": 3.9392745674373506, |
| "grad_norm": 0.38490796089172363, |
| "learning_rate": 5.687686768122574e-05, |
| "loss": 0.2828, |
| "mean_token_accuracy": 0.9243268824994564, |
| "step": 1170000 |
| }, |
| { |
| "epoch": 3.9561090741358007, |
| "grad_norm": 0.26299554109573364, |
| "learning_rate": 5.5158242622539826e-05, |
| "loss": 0.2854, |
| "mean_token_accuracy": 0.92382511267066, |
| "step": 1175000 |
| }, |
| { |
| "epoch": 3.972943580834251, |
| "grad_norm": 0.6461980938911438, |
| "learning_rate": 5.346242115356267e-05, |
| "loss": 0.2894, |
| "mean_token_accuracy": 0.9226486649751663, |
| "step": 1180000 |
| }, |
| { |
| "epoch": 3.989778087532701, |
| "grad_norm": 0.09252765774726868, |
| "learning_rate": 5.1789618716369305e-05, |
| "loss": 0.2858, |
| "mean_token_accuracy": 0.9233844766557217, |
| "step": 1185000 |
| }, |
| { |
| "epoch": 4.006612594231151, |
| "grad_norm": 0.108070008456707, |
| "learning_rate": 5.0140375765096005e-05, |
| "loss": 0.281, |
| "mean_token_accuracy": 0.924592745500803, |
| "step": 1190000 |
| }, |
| { |
| "epoch": 4.023447100929602, |
| "grad_norm": 0.15629445016384125, |
| "learning_rate": 4.851553387946539e-05, |
| "loss": 0.266, |
| "mean_token_accuracy": 0.9284300253868103, |
| "step": 1195000 |
| }, |
| { |
| "epoch": 4.040281607628052, |
| "grad_norm": 0.38527366518974304, |
| "learning_rate": 4.6914303514949934e-05, |
| "loss": 0.2665, |
| "mean_token_accuracy": 0.9280487103879452, |
| "step": 1200000 |
| }, |
| { |
| "epoch": 4.057116114326502, |
| "grad_norm": 0.4543330669403076, |
| "learning_rate": 4.53368893452808e-05, |
| "loss": 0.2727, |
| "mean_token_accuracy": 0.926523428016901, |
| "step": 1205000 |
| }, |
| { |
| "epoch": 4.073950621024952, |
| "grad_norm": 0.7975602746009827, |
| "learning_rate": 4.37841098741327e-05, |
| "loss": 0.2673, |
| "mean_token_accuracy": 0.9280878017485141, |
| "step": 1210000 |
| }, |
| { |
| "epoch": 4.090785127723402, |
| "grad_norm": 0.32045167684555054, |
| "learning_rate": 4.225643825784287e-05, |
| "loss": 0.2705, |
| "mean_token_accuracy": 0.9270674463570118, |
| "step": 1215000 |
| }, |
| { |
| "epoch": 4.1076196344218525, |
| "grad_norm": 0.2954208254814148, |
| "learning_rate": 4.075223618231558e-05, |
| "loss": 0.2675, |
| "mean_token_accuracy": 0.9278540238678455, |
| "step": 1220000 |
| }, |
| { |
| "epoch": 4.124454141120303, |
| "grad_norm": 0.07819996029138565, |
| "learning_rate": 3.927379555579019e-05, |
| "loss": 0.2671, |
| "mean_token_accuracy": 0.9281146086335182, |
| "step": 1225000 |
| }, |
| { |
| "epoch": 4.141288647818753, |
| "grad_norm": 0.6219560503959656, |
| "learning_rate": 3.782009368010972e-05, |
| "loss": 0.2624, |
| "mean_token_accuracy": 0.9293306515336036, |
| "step": 1230000 |
| }, |
| { |
| "epoch": 4.158123154517203, |
| "grad_norm": 0.05652535706758499, |
| "learning_rate": 3.639188491319606e-05, |
| "loss": 0.2696, |
| "mean_token_accuracy": 0.9273426880776883, |
| "step": 1235000 |
| }, |
| { |
| "epoch": 4.174957661215654, |
| "grad_norm": 0.17151544988155365, |
| "learning_rate": 3.498848913948891e-05, |
| "loss": 0.2644, |
| "mean_token_accuracy": 0.9285072940289975, |
| "step": 1240000 |
| }, |
| { |
| "epoch": 4.1917921679141035, |
| "grad_norm": 0.1508161425590515, |
| "learning_rate": 3.361092290728229e-05, |
| "loss": 0.2686, |
| "mean_token_accuracy": 0.9276358473837376, |
| "step": 1245000 |
| }, |
| { |
| "epoch": 4.208626674612554, |
| "grad_norm": 0.38865068554878235, |
| "learning_rate": 3.2259066610106025e-05, |
| "loss": 0.2713, |
| "mean_token_accuracy": 0.927123271459341, |
| "step": 1250000 |
| }, |
| { |
| "epoch": 4.225461181311004, |
| "grad_norm": 0.08867581933736801, |
| "learning_rate": 3.0933343571826395e-05, |
| "loss": 0.2653, |
| "mean_token_accuracy": 0.9284939861536026, |
| "step": 1255000 |
| }, |
| { |
| "epoch": 4.242295688009454, |
| "grad_norm": 0.1444486528635025, |
| "learning_rate": 2.963338070385993e-05, |
| "loss": 0.2636, |
| "mean_token_accuracy": 0.9287917876064777, |
| "step": 1260000 |
| }, |
| { |
| "epoch": 4.259130194707905, |
| "grad_norm": 0.4002508819103241, |
| "learning_rate": 2.835934812843635e-05, |
| "loss": 0.2684, |
| "mean_token_accuracy": 0.9277478049635887, |
| "step": 1265000 |
| }, |
| { |
| "epoch": 4.275964701406354, |
| "grad_norm": 0.07395654916763306, |
| "learning_rate": 2.711215387192731e-05, |
| "loss": 0.261, |
| "mean_token_accuracy": 0.9295833923220634, |
| "step": 1270000 |
| }, |
| { |
| "epoch": 4.292799208104805, |
| "grad_norm": 0.4482879042625427, |
| "learning_rate": 2.589118946364538e-05, |
| "loss": 0.266, |
| "mean_token_accuracy": 0.9281874049901963, |
| "step": 1275000 |
| }, |
| { |
| "epoch": 4.3096337148032555, |
| "grad_norm": 0.3122960925102234, |
| "learning_rate": 2.469685219863721e-05, |
| "loss": 0.2628, |
| "mean_token_accuracy": 0.929184170460701, |
| "step": 1280000 |
| }, |
| { |
| "epoch": 4.326468221501705, |
| "grad_norm": 0.0817110687494278, |
| "learning_rate": 2.352905306097905e-05, |
| "loss": 0.2653, |
| "mean_token_accuracy": 0.9284079388618469, |
| "step": 1285000 |
| }, |
| { |
| "epoch": 4.343302728200156, |
| "grad_norm": 0.14607825875282288, |
| "learning_rate": 2.2388623531292035e-05, |
| "loss": 0.2641, |
| "mean_token_accuracy": 0.9289733718693256, |
| "step": 1290000 |
| }, |
| { |
| "epoch": 4.360137234898605, |
| "grad_norm": 0.10098662972450256, |
| "learning_rate": 2.1275006377590278e-05, |
| "loss": 0.2699, |
| "mean_token_accuracy": 0.9274944821059704, |
| "step": 1295000 |
| }, |
| { |
| "epoch": 4.376971741597056, |
| "grad_norm": 0.055902060121297836, |
| "learning_rate": 2.0188564831848276e-05, |
| "loss": 0.2637, |
| "mean_token_accuracy": 0.9289655669987201, |
| "step": 1300000 |
| }, |
| { |
| "epoch": 4.3938062482955065, |
| "grad_norm": 0.1107214167714119, |
| "learning_rate": 1.9129427977945563e-05, |
| "loss": 0.2659, |
| "mean_token_accuracy": 0.9283799425303936, |
| "step": 1305000 |
| }, |
| { |
| "epoch": 4.410640754993956, |
| "grad_norm": 0.2062639594078064, |
| "learning_rate": 1.80979253276016e-05, |
| "loss": 0.2601, |
| "mean_token_accuracy": 0.9298893382549286, |
| "step": 1310000 |
| }, |
| { |
| "epoch": 4.427475261692407, |
| "grad_norm": 0.11259380728006363, |
| "learning_rate": 1.7093766592534782e-05, |
| "loss": 0.2642, |
| "mean_token_accuracy": 0.9288688370406628, |
| "step": 1315000 |
| }, |
| { |
| "epoch": 4.444309768390857, |
| "grad_norm": 0.42972731590270996, |
| "learning_rate": 1.611728025299511e-05, |
| "loss": 0.2694, |
| "mean_token_accuracy": 0.9274199161589146, |
| "step": 1320000 |
| }, |
| { |
| "epoch": 4.461144275089307, |
| "grad_norm": 0.11186912655830383, |
| "learning_rate": 1.5168395301277254e-05, |
| "loss": 0.2631, |
| "mean_token_accuracy": 0.9291341427564621, |
| "step": 1325000 |
| }, |
| { |
| "epoch": 4.477978781787757, |
| "grad_norm": 0.10467754304409027, |
| "learning_rate": 1.4247785537743446e-05, |
| "loss": 0.2571, |
| "mean_token_accuracy": 0.9306377294719219, |
| "step": 1330000 |
| }, |
| { |
| "epoch": 4.494813288486207, |
| "grad_norm": 0.12057580053806305, |
| "learning_rate": 1.3354999283167608e-05, |
| "loss": 0.2601, |
| "mean_token_accuracy": 0.9296907308280468, |
| "step": 1335000 |
| }, |
| { |
| "epoch": 4.511647795184658, |
| "grad_norm": 0.1404503881931305, |
| "learning_rate": 1.2490329640000437e-05, |
| "loss": 0.2604, |
| "mean_token_accuracy": 0.9297159347653389, |
| "step": 1340000 |
| }, |
| { |
| "epoch": 4.528482301883108, |
| "grad_norm": 0.3633385896682739, |
| "learning_rate": 1.1653879342634205e-05, |
| "loss": 0.2639, |
| "mean_token_accuracy": 0.9288859981119633, |
| "step": 1345000 |
| }, |
| { |
| "epoch": 4.545316808581558, |
| "grad_norm": 0.601957380771637, |
| "learning_rate": 1.08455889216674e-05, |
| "loss": 0.2644, |
| "mean_token_accuracy": 0.928759328609705, |
| "step": 1350000 |
| }, |
| { |
| "epoch": 4.562151315280008, |
| "grad_norm": 0.12074960023164749, |
| "learning_rate": 1.006603094687561e-05, |
| "loss": 0.2613, |
| "mean_token_accuracy": 0.9296342330694198, |
| "step": 1355000 |
| }, |
| { |
| "epoch": 4.578985821978459, |
| "grad_norm": 0.17096629738807678, |
| "learning_rate": 9.314821506245658e-06, |
| "loss": 0.266, |
| "mean_token_accuracy": 0.9286546633422375, |
| "step": 1360000 |
| }, |
| { |
| "epoch": 4.595820328676909, |
| "grad_norm": 0.7128485441207886, |
| "learning_rate": 8.592208704526372e-06, |
| "loss": 0.262, |
| "mean_token_accuracy": 0.9295457917332649, |
| "step": 1365000 |
| }, |
| { |
| "epoch": 4.612654835375359, |
| "grad_norm": 0.07192352414131165, |
| "learning_rate": 7.898142431776002e-06, |
| "loss": 0.2614, |
| "mean_token_accuracy": 0.9294094405591488, |
| "step": 1370000 |
| }, |
| { |
| "epoch": 4.629489342073809, |
| "grad_norm": 0.3203010857105255, |
| "learning_rate": 7.233113034463307e-06, |
| "loss": 0.2633, |
| "mean_token_accuracy": 0.929205379140377, |
| "step": 1375000 |
| }, |
| { |
| "epoch": 4.646323848772259, |
| "grad_norm": 0.09548963606357574, |
| "learning_rate": 6.596791644979006e-06, |
| "loss": 0.26, |
| "mean_token_accuracy": 0.9300438407242299, |
| "step": 1380000 |
| }, |
| { |
| "epoch": 4.66315835547071, |
| "grad_norm": 0.29013264179229736, |
| "learning_rate": 5.9893898329385355e-06, |
| "loss": 0.262, |
| "mean_token_accuracy": 0.9296431491851807, |
| "step": 1385000 |
| }, |
| { |
| "epoch": 4.6799928621691596, |
| "grad_norm": 0.6831434369087219, |
| "learning_rate": 5.4109797658530025e-06, |
| "loss": 0.2616, |
| "mean_token_accuracy": 0.9296812532305717, |
| "step": 1390000 |
| }, |
| { |
| "epoch": 4.69682736886761, |
| "grad_norm": 0.09422193467617035, |
| "learning_rate": 4.861523164941983e-06, |
| "loss": 0.2619, |
| "mean_token_accuracy": 0.9292743100106716, |
| "step": 1395000 |
| }, |
| { |
| "epoch": 4.713661875566061, |
| "grad_norm": 0.5069900155067444, |
| "learning_rate": 4.341305137475077e-06, |
| "loss": 0.2601, |
| "mean_token_accuracy": 0.9299272497832776, |
| "step": 1400000 |
| }, |
| { |
| "epoch": 4.73049638226451, |
| "grad_norm": 0.10779443383216858, |
| "learning_rate": 3.8502746696087445e-06, |
| "loss": 0.2567, |
| "mean_token_accuracy": 0.9307436456203461, |
| "step": 1405000 |
| }, |
| { |
| "epoch": 4.747330888962961, |
| "grad_norm": 0.06721177697181702, |
| "learning_rate": 3.3884006385650756e-06, |
| "loss": 0.2635, |
| "mean_token_accuracy": 0.9288554300010204, |
| "step": 1410000 |
| }, |
| { |
| "epoch": 4.764165395661411, |
| "grad_norm": 0.10563109815120697, |
| "learning_rate": 2.9559227059363924e-06, |
| "loss": 0.2621, |
| "mean_token_accuracy": 0.9295997213721275, |
| "step": 1415000 |
| }, |
| { |
| "epoch": 4.780999902359861, |
| "grad_norm": 0.41121697425842285, |
| "learning_rate": 2.55279693498886e-06, |
| "loss": 0.2619, |
| "mean_token_accuracy": 0.9292697467625141, |
| "step": 1420000 |
| }, |
| { |
| "epoch": 4.797834409058312, |
| "grad_norm": 0.3246116638183594, |
| "learning_rate": 2.179071222490281e-06, |
| "loss": 0.2605, |
| "mean_token_accuracy": 0.929894913315773, |
| "step": 1425000 |
| }, |
| { |
| "epoch": 4.814668915756761, |
| "grad_norm": 0.09750587493181229, |
| "learning_rate": 1.8348559080686166e-06, |
| "loss": 0.2633, |
| "mean_token_accuracy": 0.9291049893498421, |
| "step": 1430000 |
| }, |
| { |
| "epoch": 4.831503422455212, |
| "grad_norm": 0.11214178055524826, |
| "learning_rate": 1.5199940890220088e-06, |
| "loss": 0.2596, |
| "mean_token_accuracy": 0.9300582256436348, |
| "step": 1435000 |
| }, |
| { |
| "epoch": 4.848337929153662, |
| "grad_norm": 0.4015677571296692, |
| "learning_rate": 1.2347209752872068e-06, |
| "loss": 0.2595, |
| "mean_token_accuracy": 0.9302269735574722, |
| "step": 1440000 |
| }, |
| { |
| "epoch": 4.865172435852112, |
| "grad_norm": 0.37538087368011475, |
| "learning_rate": 9.790045251671097e-07, |
| "loss": 0.2577, |
| "mean_token_accuracy": 0.9305346738100052, |
| "step": 1445000 |
| }, |
| { |
| "epoch": 4.882006942550563, |
| "grad_norm": 0.11590627580881119, |
| "learning_rate": 7.529174040500908e-07, |
| "loss": 0.2597, |
| "mean_token_accuracy": 0.9299580462992191, |
| "step": 1450000 |
| }, |
| { |
| "epoch": 4.898841449249012, |
| "grad_norm": 0.15194453299045563, |
| "learning_rate": 5.563596306569496e-07, |
| "loss": 0.2629, |
| "mean_token_accuracy": 0.9291830613553524, |
| "step": 1455000 |
| }, |
| { |
| "epoch": 4.915675955947463, |
| "grad_norm": 0.2619645297527313, |
| "learning_rate": 3.894814021693227e-07, |
| "loss": 0.2616, |
| "mean_token_accuracy": 0.9297183241307736, |
| "step": 1460000 |
| }, |
| { |
| "epoch": 4.932510462645913, |
| "grad_norm": 0.3074095547199249, |
| "learning_rate": 2.522847508969195e-07, |
| "loss": 0.2601, |
| "mean_token_accuracy": 0.9300595594525337, |
| "step": 1465000 |
| }, |
| { |
| "epoch": 4.949344969344363, |
| "grad_norm": 0.1385357677936554, |
| "learning_rate": 1.4473106658496882e-07, |
| "loss": 0.2648, |
| "mean_token_accuracy": 0.9286121795237064, |
| "step": 1470000 |
| }, |
| { |
| "epoch": 4.9661794760428135, |
| "grad_norm": 0.2064802348613739, |
| "learning_rate": 6.686355136054089e-08, |
| "loss": 0.2624, |
| "mean_token_accuracy": 0.9292844385802745, |
| "step": 1475000 |
| }, |
| { |
| "epoch": 4.983013982741264, |
| "grad_norm": 0.42331379652023315, |
| "learning_rate": 1.869145693253471e-08, |
| "loss": 0.2589, |
| "mean_token_accuracy": 0.9300985207080841, |
| "step": 1480000 |
| }, |
| { |
| "epoch": 4.999848489439714, |
| "grad_norm": 0.07902824133634567, |
| "learning_rate": 2.2050679424068244e-10, |
| "loss": 0.2633, |
| "mean_token_accuracy": 0.9292208172976971, |
| "step": 1485000 |
| } |
| ], |
| "logging_steps": 5000, |
| "max_steps": 1485045, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "save_steps": 5000, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 6.0592925279854264e+19, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
