{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 5.0, "eval_steps": 500, "global_step": 4835, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.010346611484738748, "grad_norm": 6.139026641845703, "learning_rate": 1.8595041322314052e-07, "loss": 0.9459, "step": 10 }, { "epoch": 0.020693222969477496, "grad_norm": 5.49014139175415, "learning_rate": 3.925619834710744e-07, "loss": 0.9235, "step": 20 }, { "epoch": 0.031039834454216243, "grad_norm": 3.730520009994507, "learning_rate": 5.991735537190084e-07, "loss": 0.8539, "step": 30 }, { "epoch": 0.04138644593895499, "grad_norm": 2.3603694438934326, "learning_rate": 8.057851239669421e-07, "loss": 0.7718, "step": 40 }, { "epoch": 0.05173305742369374, "grad_norm": 2.0935893058776855, "learning_rate": 1.012396694214876e-06, "loss": 0.68, "step": 50 }, { "epoch": 0.062079668908432487, "grad_norm": 1.742537021636963, "learning_rate": 1.21900826446281e-06, "loss": 0.6411, "step": 60 }, { "epoch": 0.07242628039317124, "grad_norm": 1.5265495777130127, "learning_rate": 1.425619834710744e-06, "loss": 0.598, "step": 70 }, { "epoch": 0.08277289187790998, "grad_norm": 1.4529454708099365, "learning_rate": 1.632231404958678e-06, "loss": 0.5599, "step": 80 }, { "epoch": 0.09311950336264874, "grad_norm": 1.4188324213027954, "learning_rate": 1.8388429752066117e-06, "loss": 0.5307, "step": 90 }, { "epoch": 0.10346611484738748, "grad_norm": 1.5109107494354248, "learning_rate": 2.0454545454545457e-06, "loss": 0.525, "step": 100 }, { "epoch": 0.11381272633212623, "grad_norm": 1.3689301013946533, "learning_rate": 2.2520661157024796e-06, "loss": 0.5239, "step": 110 }, { "epoch": 0.12415933781686497, "grad_norm": 1.3892041444778442, "learning_rate": 2.4586776859504134e-06, "loss": 0.5, "step": 120 }, { "epoch": 0.13450594930160373, "grad_norm": 1.3815661668777466, "learning_rate": 2.6652892561983473e-06, "loss": 0.4901, "step": 130 }, { "epoch": 0.14485256078634248, "grad_norm": 1.5329068899154663, "learning_rate": 2.871900826446281e-06, "loss": 0.4787, "step": 140 }, { "epoch": 0.1551991722710812, "grad_norm": 1.2414484024047852, "learning_rate": 3.078512396694215e-06, "loss": 0.4835, "step": 150 }, { "epoch": 0.16554578375581996, "grad_norm": 1.4333281517028809, "learning_rate": 3.2851239669421493e-06, "loss": 0.4706, "step": 160 }, { "epoch": 0.17589239524055872, "grad_norm": 1.2515960931777954, "learning_rate": 3.491735537190083e-06, "loss": 0.4713, "step": 170 }, { "epoch": 0.18623900672529747, "grad_norm": 1.5042508840560913, "learning_rate": 3.6983471074380165e-06, "loss": 0.4646, "step": 180 }, { "epoch": 0.1965856182100362, "grad_norm": 1.4549144506454468, "learning_rate": 3.904958677685951e-06, "loss": 0.4496, "step": 190 }, { "epoch": 0.20693222969477496, "grad_norm": 1.3692591190338135, "learning_rate": 4.111570247933885e-06, "loss": 0.4545, "step": 200 }, { "epoch": 0.2172788411795137, "grad_norm": 1.3498661518096924, "learning_rate": 4.3181818181818185e-06, "loss": 0.451, "step": 210 }, { "epoch": 0.22762545266425246, "grad_norm": 1.5415807962417603, "learning_rate": 4.524793388429753e-06, "loss": 0.4517, "step": 220 }, { "epoch": 0.23797206414899122, "grad_norm": 1.339154601097107, "learning_rate": 4.731404958677686e-06, "loss": 0.4397, "step": 230 }, { "epoch": 0.24831867563372995, "grad_norm": 1.3699935674667358, "learning_rate": 4.9380165289256205e-06, "loss": 0.4405, "step": 240 }, { "epoch": 0.2586652871184687, "grad_norm": 1.2297754287719727, "learning_rate": 5.144628099173554e-06, "loss": 0.4355, "step": 250 }, { "epoch": 0.26901189860320746, "grad_norm": 1.222337007522583, "learning_rate": 5.351239669421488e-06, "loss": 0.4435, "step": 260 }, { "epoch": 0.2793585100879462, "grad_norm": 1.474328637123108, "learning_rate": 5.557851239669422e-06, "loss": 0.4285, "step": 270 }, { "epoch": 0.28970512157268496, "grad_norm": 1.3817209005355835, "learning_rate": 5.764462809917356e-06, "loss": 0.4343, "step": 280 }, { "epoch": 0.3000517330574237, "grad_norm": 1.2763681411743164, "learning_rate": 5.971074380165289e-06, "loss": 0.435, "step": 290 }, { "epoch": 0.3103983445421624, "grad_norm": 1.4916797876358032, "learning_rate": 6.1776859504132236e-06, "loss": 0.4334, "step": 300 }, { "epoch": 0.3207449560269012, "grad_norm": 1.3628357648849487, "learning_rate": 6.384297520661158e-06, "loss": 0.4273, "step": 310 }, { "epoch": 0.33109156751163993, "grad_norm": 1.3091474771499634, "learning_rate": 6.590909090909091e-06, "loss": 0.4245, "step": 320 }, { "epoch": 0.3414381789963787, "grad_norm": 1.4376888275146484, "learning_rate": 6.7975206611570255e-06, "loss": 0.428, "step": 330 }, { "epoch": 0.35178479048111744, "grad_norm": 1.4433401823043823, "learning_rate": 7.00413223140496e-06, "loss": 0.4306, "step": 340 }, { "epoch": 0.3621314019658562, "grad_norm": 1.3882417678833008, "learning_rate": 7.210743801652893e-06, "loss": 0.4292, "step": 350 }, { "epoch": 0.37247801345059495, "grad_norm": 1.3860538005828857, "learning_rate": 7.4173553719008275e-06, "loss": 0.4292, "step": 360 }, { "epoch": 0.3828246249353337, "grad_norm": 1.1911530494689941, "learning_rate": 7.623966942148761e-06, "loss": 0.4303, "step": 370 }, { "epoch": 0.3931712364200724, "grad_norm": 1.315438151359558, "learning_rate": 7.830578512396695e-06, "loss": 0.4127, "step": 380 }, { "epoch": 0.40351784790481116, "grad_norm": 1.3872874975204468, "learning_rate": 8.03719008264463e-06, "loss": 0.4305, "step": 390 }, { "epoch": 0.4138644593895499, "grad_norm": 1.2904235124588013, "learning_rate": 8.243801652892562e-06, "loss": 0.4189, "step": 400 }, { "epoch": 0.42421107087428866, "grad_norm": 1.256445050239563, "learning_rate": 8.450413223140496e-06, "loss": 0.4201, "step": 410 }, { "epoch": 0.4345576823590274, "grad_norm": 1.3195158243179321, "learning_rate": 8.65702479338843e-06, "loss": 0.4214, "step": 420 }, { "epoch": 0.4449042938437662, "grad_norm": 1.4306244850158691, "learning_rate": 8.863636363636365e-06, "loss": 0.4198, "step": 430 }, { "epoch": 0.45525090532850493, "grad_norm": 1.2186839580535889, "learning_rate": 9.070247933884299e-06, "loss": 0.4163, "step": 440 }, { "epoch": 0.4655975168132437, "grad_norm": 1.2255494594573975, "learning_rate": 9.276859504132232e-06, "loss": 0.4103, "step": 450 }, { "epoch": 0.47594412829798244, "grad_norm": 1.1726787090301514, "learning_rate": 9.483471074380166e-06, "loss": 0.4087, "step": 460 }, { "epoch": 0.48629073978272114, "grad_norm": 1.4234670400619507, "learning_rate": 9.6900826446281e-06, "loss": 0.4098, "step": 470 }, { "epoch": 0.4966373512674599, "grad_norm": 1.1076370477676392, "learning_rate": 9.896694214876033e-06, "loss": 0.4058, "step": 480 }, { "epoch": 0.5069839627521987, "grad_norm": 1.0841426849365234, "learning_rate": 9.999967416245537e-06, "loss": 0.3993, "step": 490 }, { "epoch": 0.5173305742369374, "grad_norm": 1.409942865371704, "learning_rate": 9.999706748757903e-06, "loss": 0.4133, "step": 500 }, { "epoch": 0.5276771857216761, "grad_norm": 1.2055038213729858, "learning_rate": 9.999185427372232e-06, "loss": 0.4043, "step": 510 }, { "epoch": 0.5380237972064149, "grad_norm": 1.4699640274047852, "learning_rate": 9.998403479267005e-06, "loss": 0.4075, "step": 520 }, { "epoch": 0.5483704086911536, "grad_norm": 1.1911301612854004, "learning_rate": 9.99736094520818e-06, "loss": 0.4032, "step": 530 }, { "epoch": 0.5587170201758924, "grad_norm": 1.1372137069702148, "learning_rate": 9.996057879547059e-06, "loss": 0.4073, "step": 540 }, { "epoch": 0.5690636316606311, "grad_norm": 1.1768054962158203, "learning_rate": 9.994494350217452e-06, "loss": 0.4035, "step": 550 }, { "epoch": 0.5794102431453699, "grad_norm": 1.343031406402588, "learning_rate": 9.992670438732146e-06, "loss": 0.4105, "step": 560 }, { "epoch": 0.5897568546301086, "grad_norm": 1.0798144340515137, "learning_rate": 9.990586240178642e-06, "loss": 0.4018, "step": 570 }, { "epoch": 0.6001034661148474, "grad_norm": 1.3839797973632812, "learning_rate": 9.988241863214212e-06, "loss": 0.4068, "step": 580 }, { "epoch": 0.6104500775995861, "grad_norm": 1.1565276384353638, "learning_rate": 9.98563743006022e-06, "loss": 0.4057, "step": 590 }, { "epoch": 0.6207966890843248, "grad_norm": 1.14547860622406, "learning_rate": 9.982773076495762e-06, "loss": 0.3988, "step": 600 }, { "epoch": 0.6311433005690636, "grad_norm": 1.0178049802780151, "learning_rate": 9.97964895185058e-06, "loss": 0.3975, "step": 610 }, { "epoch": 0.6414899120538023, "grad_norm": 1.0690793991088867, "learning_rate": 9.97626521899728e-06, "loss": 0.4026, "step": 620 }, { "epoch": 0.6518365235385412, "grad_norm": 1.1290843486785889, "learning_rate": 9.972622054342842e-06, "loss": 0.3923, "step": 630 }, { "epoch": 0.6621831350232799, "grad_norm": 1.1035066843032837, "learning_rate": 9.968719647819414e-06, "loss": 0.4012, "step": 640 }, { "epoch": 0.6725297465080187, "grad_norm": 1.1912881135940552, "learning_rate": 9.964558202874428e-06, "loss": 0.4001, "step": 650 }, { "epoch": 0.6828763579927574, "grad_norm": 1.020052433013916, "learning_rate": 9.960137936459975e-06, "loss": 0.3962, "step": 660 }, { "epoch": 0.6932229694774961, "grad_norm": 1.0658594369888306, "learning_rate": 9.955459079021505e-06, "loss": 0.3984, "step": 670 }, { "epoch": 0.7035695809622349, "grad_norm": 1.0133702754974365, "learning_rate": 9.95052187448581e-06, "loss": 0.3924, "step": 680 }, { "epoch": 0.7139161924469736, "grad_norm": 0.9635597467422485, "learning_rate": 9.94532658024831e-06, "loss": 0.3928, "step": 690 }, { "epoch": 0.7242628039317124, "grad_norm": 1.1104060411453247, "learning_rate": 9.939873467159627e-06, "loss": 0.395, "step": 700 }, { "epoch": 0.7346094154164511, "grad_norm": 1.020979404449463, "learning_rate": 9.934162819511473e-06, "loss": 0.4048, "step": 710 }, { "epoch": 0.7449560269011899, "grad_norm": 0.934624969959259, "learning_rate": 9.928194935021821e-06, "loss": 0.4007, "step": 720 }, { "epoch": 0.7553026383859286, "grad_norm": 1.048291563987732, "learning_rate": 9.921970124819392e-06, "loss": 0.3921, "step": 730 }, { "epoch": 0.7656492498706674, "grad_norm": 1.120107889175415, "learning_rate": 9.915488713427427e-06, "loss": 0.39, "step": 740 }, { "epoch": 0.7759958613554061, "grad_norm": 0.9637139439582825, "learning_rate": 9.908751038746773e-06, "loss": 0.3998, "step": 750 }, { "epoch": 0.7863424728401448, "grad_norm": 0.9837620854377747, "learning_rate": 9.901757452038268e-06, "loss": 0.3861, "step": 760 }, { "epoch": 0.7966890843248836, "grad_norm": 1.0107730627059937, "learning_rate": 9.894508317904418e-06, "loss": 0.3918, "step": 770 }, { "epoch": 0.8070356958096223, "grad_norm": 1.0242232084274292, "learning_rate": 9.887004014270408e-06, "loss": 0.391, "step": 780 }, { "epoch": 0.8173823072943611, "grad_norm": 0.9810847640037537, "learning_rate": 9.879244932364375e-06, "loss": 0.3908, "step": 790 }, { "epoch": 0.8277289187790998, "grad_norm": 0.9811373353004456, "learning_rate": 9.87123147669704e-06, "loss": 0.3947, "step": 800 }, { "epoch": 0.8380755302638386, "grad_norm": 0.956855058670044, "learning_rate": 9.862964065040596e-06, "loss": 0.3867, "step": 810 }, { "epoch": 0.8484221417485773, "grad_norm": 0.9348600506782532, "learning_rate": 9.854443128406931e-06, "loss": 0.3864, "step": 820 }, { "epoch": 0.8587687532333161, "grad_norm": 1.0585615634918213, "learning_rate": 9.845669111025175e-06, "loss": 0.3874, "step": 830 }, { "epoch": 0.8691153647180548, "grad_norm": 1.041060447692871, "learning_rate": 9.83664247031852e-06, "loss": 0.3841, "step": 840 }, { "epoch": 0.8794619762027935, "grad_norm": 1.0687028169631958, "learning_rate": 9.827363676880383e-06, "loss": 0.385, "step": 850 }, { "epoch": 0.8898085876875323, "grad_norm": 1.1607829332351685, "learning_rate": 9.817833214449873e-06, "loss": 0.3821, "step": 860 }, { "epoch": 0.900155199172271, "grad_norm": 0.9849240183830261, "learning_rate": 9.808051579886566e-06, "loss": 0.3819, "step": 870 }, { "epoch": 0.9105018106570099, "grad_norm": 1.220577597618103, "learning_rate": 9.798019283144607e-06, "loss": 0.3886, "step": 880 }, { "epoch": 0.9208484221417486, "grad_norm": 1.0422636270523071, "learning_rate": 9.787736847246121e-06, "loss": 0.3856, "step": 890 }, { "epoch": 0.9311950336264874, "grad_norm": 0.875372588634491, "learning_rate": 9.777204808253949e-06, "loss": 0.3851, "step": 900 }, { "epoch": 0.9415416451112261, "grad_norm": 0.918285608291626, "learning_rate": 9.766423715243697e-06, "loss": 0.3826, "step": 910 }, { "epoch": 0.9518882565959649, "grad_norm": 1.1889722347259521, "learning_rate": 9.755394130275116e-06, "loss": 0.3834, "step": 920 }, { "epoch": 0.9622348680807036, "grad_norm": 0.9664179682731628, "learning_rate": 9.74411662836279e-06, "loss": 0.3799, "step": 930 }, { "epoch": 0.9725814795654423, "grad_norm": 1.064846396446228, "learning_rate": 9.73259179744617e-06, "loss": 0.3816, "step": 940 }, { "epoch": 0.9829280910501811, "grad_norm": 1.0509345531463623, "learning_rate": 9.720820238358918e-06, "loss": 0.3838, "step": 950 }, { "epoch": 0.9932747025349198, "grad_norm": 0.9682555794715881, "learning_rate": 9.70880256479758e-06, "loss": 0.3785, "step": 960 }, { "epoch": 1.0031039834454216, "grad_norm": 0.9066140651702881, "learning_rate": 9.696539403289589e-06, "loss": 0.3685, "step": 970 }, { "epoch": 1.0134505949301604, "grad_norm": 0.9822126030921936, "learning_rate": 9.684031393160614e-06, "loss": 0.3446, "step": 980 }, { "epoch": 1.023797206414899, "grad_norm": 0.9022179245948792, "learning_rate": 9.671279186501221e-06, "loss": 0.3405, "step": 990 }, { "epoch": 1.0341438178996378, "grad_norm": 0.9742788076400757, "learning_rate": 9.658283448132875e-06, "loss": 0.3348, "step": 1000 }, { "epoch": 1.0444904293843766, "grad_norm": 0.9199923276901245, "learning_rate": 9.645044855573286e-06, "loss": 0.3425, "step": 1010 }, { "epoch": 1.0548370408691154, "grad_norm": 0.9245515465736389, "learning_rate": 9.631564099001085e-06, "loss": 0.344, "step": 1020 }, { "epoch": 1.065183652353854, "grad_norm": 0.865315854549408, "learning_rate": 9.617841881219841e-06, "loss": 0.3481, "step": 1030 }, { "epoch": 1.0755302638385928, "grad_norm": 0.8868467211723328, "learning_rate": 9.603878917621422e-06, "loss": 0.3458, "step": 1040 }, { "epoch": 1.0858768753233317, "grad_norm": 0.9886205196380615, "learning_rate": 9.5896759361487e-06, "loss": 0.3413, "step": 1050 }, { "epoch": 1.0962234868080705, "grad_norm": 0.9479049444198608, "learning_rate": 9.5752336772576e-06, "loss": 0.3371, "step": 1060 }, { "epoch": 1.106570098292809, "grad_norm": 0.9921103715896606, "learning_rate": 9.560552893878499e-06, "loss": 0.3431, "step": 1070 }, { "epoch": 1.1169167097775479, "grad_norm": 0.9266170263290405, "learning_rate": 9.545634351376965e-06, "loss": 0.3412, "step": 1080 }, { "epoch": 1.1272633212622867, "grad_norm": 0.9916461110115051, "learning_rate": 9.530478827513867e-06, "loss": 0.3429, "step": 1090 }, { "epoch": 1.1376099327470253, "grad_norm": 0.90711510181427, "learning_rate": 9.51508711240482e-06, "loss": 0.3461, "step": 1100 }, { "epoch": 1.147956544231764, "grad_norm": 0.9366750121116638, "learning_rate": 9.499460008478996e-06, "loss": 0.3392, "step": 1110 }, { "epoch": 1.1583031557165029, "grad_norm": 1.0261372327804565, "learning_rate": 9.483598330437285e-06, "loss": 0.3434, "step": 1120 }, { "epoch": 1.1686497672012415, "grad_norm": 0.9353191256523132, "learning_rate": 9.46750290520983e-06, "loss": 0.344, "step": 1130 }, { "epoch": 1.1789963786859803, "grad_norm": 0.9214949011802673, "learning_rate": 9.451174571912913e-06, "loss": 0.3404, "step": 1140 }, { "epoch": 1.189342990170719, "grad_norm": 0.9317702054977417, "learning_rate": 9.434614181805203e-06, "loss": 0.3453, "step": 1150 }, { "epoch": 1.199689601655458, "grad_norm": 0.8729979395866394, "learning_rate": 9.417822598243382e-06, "loss": 0.3505, "step": 1160 }, { "epoch": 1.2100362131401965, "grad_norm": 0.9862754940986633, "learning_rate": 9.400800696637136e-06, "loss": 0.3416, "step": 1170 }, { "epoch": 1.2203828246249353, "grad_norm": 0.9723279476165771, "learning_rate": 9.383549364403515e-06, "loss": 0.343, "step": 1180 }, { "epoch": 1.2307294361096741, "grad_norm": 0.8829843997955322, "learning_rate": 9.366069500920662e-06, "loss": 0.3375, "step": 1190 }, { "epoch": 1.241076047594413, "grad_norm": 0.9550700783729553, "learning_rate": 9.34836201748094e-06, "loss": 0.3423, "step": 1200 }, { "epoch": 1.2514226590791515, "grad_norm": 0.9076991081237793, "learning_rate": 9.330427837243408e-06, "loss": 0.3369, "step": 1210 }, { "epoch": 1.2617692705638903, "grad_norm": 0.888508677482605, "learning_rate": 9.312267895185697e-06, "loss": 0.3405, "step": 1220 }, { "epoch": 1.2721158820486291, "grad_norm": 0.9597383141517639, "learning_rate": 9.293883138055275e-06, "loss": 0.351, "step": 1230 }, { "epoch": 1.2824624935333677, "grad_norm": 0.9085503816604614, "learning_rate": 9.275274524320075e-06, "loss": 0.3405, "step": 1240 }, { "epoch": 1.2928091050181065, "grad_norm": 0.9536021947860718, "learning_rate": 9.25644302411854e-06, "loss": 0.3408, "step": 1250 }, { "epoch": 1.3031557165028453, "grad_norm": 0.9334466457366943, "learning_rate": 9.237389619209037e-06, "loss": 0.3447, "step": 1260 }, { "epoch": 1.313502327987584, "grad_norm": 0.9981200098991394, "learning_rate": 9.218115302918676e-06, "loss": 0.337, "step": 1270 }, { "epoch": 1.3238489394723227, "grad_norm": 0.9816781282424927, "learning_rate": 9.198621080091525e-06, "loss": 0.35, "step": 1280 }, { "epoch": 1.3341955509570615, "grad_norm": 0.9127581119537354, "learning_rate": 9.178907967036229e-06, "loss": 0.3432, "step": 1290 }, { "epoch": 1.3445421624418004, "grad_norm": 0.8629497289657593, "learning_rate": 9.15897699147301e-06, "loss": 0.3353, "step": 1300 }, { "epoch": 1.3548887739265392, "grad_norm": 0.9100907444953918, "learning_rate": 9.138829192480106e-06, "loss": 0.3358, "step": 1310 }, { "epoch": 1.3652353854112778, "grad_norm": 0.953328549861908, "learning_rate": 9.118465620439594e-06, "loss": 0.3368, "step": 1320 }, { "epoch": 1.3755819968960166, "grad_norm": 0.9011934399604797, "learning_rate": 9.097887336982625e-06, "loss": 0.3438, "step": 1330 }, { "epoch": 1.3859286083807554, "grad_norm": 0.9293943047523499, "learning_rate": 9.077095414934076e-06, "loss": 0.3449, "step": 1340 }, { "epoch": 1.396275219865494, "grad_norm": 0.8752354979515076, "learning_rate": 9.056090938256633e-06, "loss": 0.3418, "step": 1350 }, { "epoch": 1.4066218313502328, "grad_norm": 0.978129506111145, "learning_rate": 9.034875001994258e-06, "loss": 0.3369, "step": 1360 }, { "epoch": 1.4169684428349716, "grad_norm": 0.8396763801574707, "learning_rate": 9.013448712215127e-06, "loss": 0.341, "step": 1370 }, { "epoch": 1.4273150543197102, "grad_norm": 0.8739905953407288, "learning_rate": 8.991813185953942e-06, "loss": 0.3427, "step": 1380 }, { "epoch": 1.437661665804449, "grad_norm": 0.8390309810638428, "learning_rate": 8.969969551153706e-06, "loss": 0.3435, "step": 1390 }, { "epoch": 1.4480082772891878, "grad_norm": 0.9313223958015442, "learning_rate": 8.947918946606924e-06, "loss": 0.3348, "step": 1400 }, { "epoch": 1.4583548887739266, "grad_norm": 0.8645867109298706, "learning_rate": 8.925662521896224e-06, "loss": 0.3377, "step": 1410 }, { "epoch": 1.4687015002586654, "grad_norm": 0.9142380356788635, "learning_rate": 8.90320143733443e-06, "loss": 0.336, "step": 1420 }, { "epoch": 1.479048111743404, "grad_norm": 0.9506648182868958, "learning_rate": 8.88053686390407e-06, "loss": 0.3414, "step": 1430 }, { "epoch": 1.4893947232281428, "grad_norm": 0.855963945388794, "learning_rate": 8.857669983196326e-06, "loss": 0.3293, "step": 1440 }, { "epoch": 1.4997413347128816, "grad_norm": 0.9154329299926758, "learning_rate": 8.834601987349436e-06, "loss": 0.3355, "step": 1450 }, { "epoch": 1.5100879461976202, "grad_norm": 0.8513219356536865, "learning_rate": 8.811334078986542e-06, "loss": 0.3289, "step": 1460 }, { "epoch": 1.520434557682359, "grad_norm": 0.9121130108833313, "learning_rate": 8.787867471152992e-06, "loss": 0.3413, "step": 1470 }, { "epoch": 1.5307811691670978, "grad_norm": 0.8203767538070679, "learning_rate": 8.764203387253102e-06, "loss": 0.3342, "step": 1480 }, { "epoch": 1.5411277806518364, "grad_norm": 0.8152233362197876, "learning_rate": 8.740343060986366e-06, "loss": 0.339, "step": 1490 }, { "epoch": 1.5514743921365752, "grad_norm": 0.881718635559082, "learning_rate": 8.716287736283158e-06, "loss": 0.3342, "step": 1500 }, { "epoch": 1.561821003621314, "grad_norm": 0.851697564125061, "learning_rate": 8.692038667239857e-06, "loss": 0.338, "step": 1510 }, { "epoch": 1.5721676151060526, "grad_norm": 0.9294906258583069, "learning_rate": 8.66759711805348e-06, "loss": 0.3359, "step": 1520 }, { "epoch": 1.5825142265907917, "grad_norm": 0.8974320888519287, "learning_rate": 8.642964362955781e-06, "loss": 0.3349, "step": 1530 }, { "epoch": 1.5928608380755302, "grad_norm": 0.9681616425514221, "learning_rate": 8.618141686146803e-06, "loss": 0.3335, "step": 1540 }, { "epoch": 1.603207449560269, "grad_norm": 0.8356232643127441, "learning_rate": 8.593130381727938e-06, "loss": 0.3332, "step": 1550 }, { "epoch": 1.6135540610450079, "grad_norm": 0.8890557289123535, "learning_rate": 8.567931753634462e-06, "loss": 0.3317, "step": 1560 }, { "epoch": 1.6239006725297465, "grad_norm": 0.9087671637535095, "learning_rate": 8.542547115567553e-06, "loss": 0.3416, "step": 1570 }, { "epoch": 1.6342472840144853, "grad_norm": 0.7936503887176514, "learning_rate": 8.516977790925799e-06, "loss": 0.3366, "step": 1580 }, { "epoch": 1.644593895499224, "grad_norm": 0.849400520324707, "learning_rate": 8.491225112736209e-06, "loss": 0.3394, "step": 1590 }, { "epoch": 1.6549405069839627, "grad_norm": 0.833982527256012, "learning_rate": 8.465290423584718e-06, "loss": 0.3373, "step": 1600 }, { "epoch": 1.6652871184687015, "grad_norm": 0.8153632879257202, "learning_rate": 8.439175075546191e-06, "loss": 0.3413, "step": 1610 }, { "epoch": 1.6756337299534403, "grad_norm": 0.8862062692642212, "learning_rate": 8.412880430113931e-06, "loss": 0.3334, "step": 1620 }, { "epoch": 1.6859803414381789, "grad_norm": 0.7980442643165588, "learning_rate": 8.386407858128707e-06, "loss": 0.331, "step": 1630 }, { "epoch": 1.696326952922918, "grad_norm": 0.8371018171310425, "learning_rate": 8.359758739707275e-06, "loss": 0.3409, "step": 1640 }, { "epoch": 1.7066735644076565, "grad_norm": 0.8669474720954895, "learning_rate": 8.33293446417044e-06, "loss": 0.3325, "step": 1650 }, { "epoch": 1.717020175892395, "grad_norm": 0.8137986660003662, "learning_rate": 8.305936429970622e-06, "loss": 0.3355, "step": 1660 }, { "epoch": 1.7273667873771341, "grad_norm": 0.8193359375, "learning_rate": 8.278766044618936e-06, "loss": 0.3323, "step": 1670 }, { "epoch": 1.7377133988618727, "grad_norm": 0.859048068523407, "learning_rate": 8.251424724611832e-06, "loss": 0.3314, "step": 1680 }, { "epoch": 1.7480600103466115, "grad_norm": 0.924875795841217, "learning_rate": 8.22391389535724e-06, "loss": 0.3377, "step": 1690 }, { "epoch": 1.7584066218313503, "grad_norm": 0.8729901313781738, "learning_rate": 8.196234991100256e-06, "loss": 0.3298, "step": 1700 }, { "epoch": 1.768753233316089, "grad_norm": 0.8952008485794067, "learning_rate": 8.168389454848366e-06, "loss": 0.3324, "step": 1710 }, { "epoch": 1.7790998448008277, "grad_norm": 0.8685171604156494, "learning_rate": 8.140378738296233e-06, "loss": 0.3356, "step": 1720 }, { "epoch": 1.7894464562855665, "grad_norm": 0.7909799218177795, "learning_rate": 8.112204301749988e-06, "loss": 0.3331, "step": 1730 }, { "epoch": 1.7997930677703051, "grad_norm": 0.7982770204544067, "learning_rate": 8.083867614051125e-06, "loss": 0.3422, "step": 1740 }, { "epoch": 1.810139679255044, "grad_norm": 0.8534769415855408, "learning_rate": 8.055370152499909e-06, "loss": 0.3345, "step": 1750 }, { "epoch": 1.8204862907397827, "grad_norm": 0.8252643346786499, "learning_rate": 8.026713402778362e-06, "loss": 0.3394, "step": 1760 }, { "epoch": 1.8308329022245213, "grad_norm": 0.8682839274406433, "learning_rate": 7.99789885887281e-06, "loss": 0.3329, "step": 1770 }, { "epoch": 1.8411795137092604, "grad_norm": 0.9072077870368958, "learning_rate": 7.968928022996e-06, "loss": 0.3349, "step": 1780 }, { "epoch": 1.851526125193999, "grad_norm": 0.802394688129425, "learning_rate": 7.939802405508772e-06, "loss": 0.3374, "step": 1790 }, { "epoch": 1.8618727366787378, "grad_norm": 0.8025832772254944, "learning_rate": 7.910523524841329e-06, "loss": 0.3329, "step": 1800 }, { "epoch": 1.8722193481634766, "grad_norm": 0.8657801151275635, "learning_rate": 7.88109290741407e-06, "loss": 0.3379, "step": 1810 }, { "epoch": 1.8825659596482152, "grad_norm": 0.9007260203361511, "learning_rate": 7.851512087558016e-06, "loss": 0.3338, "step": 1820 }, { "epoch": 1.892912571132954, "grad_norm": 0.7836032509803772, "learning_rate": 7.821782607434815e-06, "loss": 0.3278, "step": 1830 }, { "epoch": 1.9032591826176928, "grad_norm": 0.8562266230583191, "learning_rate": 7.791906016956348e-06, "loss": 0.3352, "step": 1840 }, { "epoch": 1.9136057941024314, "grad_norm": 0.8395200371742249, "learning_rate": 7.761883873703919e-06, "loss": 0.337, "step": 1850 }, { "epoch": 1.9239524055871702, "grad_norm": 0.8483885526657104, "learning_rate": 7.731717742847059e-06, "loss": 0.335, "step": 1860 }, { "epoch": 1.934299017071909, "grad_norm": 0.8115087151527405, "learning_rate": 7.701409197061927e-06, "loss": 0.334, "step": 1870 }, { "epoch": 1.9446456285566476, "grad_norm": 0.8322962522506714, "learning_rate": 7.670959816449313e-06, "loss": 0.3291, "step": 1880 }, { "epoch": 1.9549922400413866, "grad_norm": 0.8189446926116943, "learning_rate": 7.640371188452274e-06, "loss": 0.3291, "step": 1890 }, { "epoch": 1.9653388515261252, "grad_norm": 0.8812534809112549, "learning_rate": 7.609644907773365e-06, "loss": 0.3283, "step": 1900 }, { "epoch": 1.9756854630108638, "grad_norm": 0.8735312223434448, "learning_rate": 7.578782576291501e-06, "loss": 0.3291, "step": 1910 }, { "epoch": 1.9860320744956028, "grad_norm": 0.8245293498039246, "learning_rate": 7.547785802978449e-06, "loss": 0.3291, "step": 1920 }, { "epoch": 1.9963786859803414, "grad_norm": 0.7975995540618896, "learning_rate": 7.516656203814945e-06, "loss": 0.3334, "step": 1930 }, { "epoch": 2.0062079668908432, "grad_norm": 0.8532512187957764, "learning_rate": 7.485395401706447e-06, "loss": 0.2926, "step": 1940 }, { "epoch": 2.016554578375582, "grad_norm": 0.8140570521354675, "learning_rate": 7.454005026398525e-06, "loss": 0.2638, "step": 1950 }, { "epoch": 2.026901189860321, "grad_norm": 0.8759050965309143, "learning_rate": 7.422486714391896e-06, "loss": 0.268, "step": 1960 }, { "epoch": 2.0372478013450595, "grad_norm": 0.8712063431739807, "learning_rate": 7.390842108857111e-06, "loss": 0.2633, "step": 1970 }, { "epoch": 2.047594412829798, "grad_norm": 0.8876409530639648, "learning_rate": 7.359072859548884e-06, "loss": 0.2695, "step": 1980 }, { "epoch": 2.057941024314537, "grad_norm": 0.8743641972541809, "learning_rate": 7.327180622720087e-06, "loss": 0.2661, "step": 1990 }, { "epoch": 2.0682876357992757, "grad_norm": 0.834858238697052, "learning_rate": 7.29516706103541e-06, "loss": 0.2681, "step": 2000 }, { "epoch": 2.0786342472840147, "grad_norm": 0.8777759671211243, "learning_rate": 7.263033843484664e-06, "loss": 0.2639, "step": 2010 }, { "epoch": 2.0889808587687533, "grad_norm": 0.8941816091537476, "learning_rate": 7.2307826452957855e-06, "loss": 0.2648, "step": 2020 }, { "epoch": 2.099327470253492, "grad_norm": 0.9012478590011597, "learning_rate": 7.198415147847491e-06, "loss": 0.2664, "step": 2030 }, { "epoch": 2.109674081738231, "grad_norm": 0.8317115902900696, "learning_rate": 7.165933038581627e-06, "loss": 0.2703, "step": 2040 }, { "epoch": 2.1200206932229695, "grad_norm": 0.8598811030387878, "learning_rate": 7.1333380109151916e-06, "loss": 0.2694, "step": 2050 }, { "epoch": 2.130367304707708, "grad_norm": 0.7854169607162476, "learning_rate": 7.100631764152054e-06, "loss": 0.2737, "step": 2060 }, { "epoch": 2.140713916192447, "grad_norm": 0.8262772560119629, "learning_rate": 7.067816003394359e-06, "loss": 0.2687, "step": 2070 }, { "epoch": 2.1510605276771857, "grad_norm": 0.8128632307052612, "learning_rate": 7.034892439453639e-06, "loss": 0.2682, "step": 2080 }, { "epoch": 2.1614071391619243, "grad_norm": 0.8013981580734253, "learning_rate": 7.001862788761617e-06, "loss": 0.2643, "step": 2090 }, { "epoch": 2.1717537506466633, "grad_norm": 0.8377342820167542, "learning_rate": 6.96872877328073e-06, "loss": 0.2709, "step": 2100 }, { "epoch": 2.182100362131402, "grad_norm": 0.8179459571838379, "learning_rate": 6.935492120414347e-06, "loss": 0.2652, "step": 2110 }, { "epoch": 2.192446973616141, "grad_norm": 0.8413109183311462, "learning_rate": 6.902154562916722e-06, "loss": 0.2695, "step": 2120 }, { "epoch": 2.2027935851008795, "grad_norm": 0.8183903098106384, "learning_rate": 6.86871783880265e-06, "loss": 0.2725, "step": 2130 }, { "epoch": 2.213140196585618, "grad_norm": 0.8539458513259888, "learning_rate": 6.835183691256866e-06, "loss": 0.268, "step": 2140 }, { "epoch": 2.223486808070357, "grad_norm": 0.9063196778297424, "learning_rate": 6.801553868543163e-06, "loss": 0.2742, "step": 2150 }, { "epoch": 2.2338334195550957, "grad_norm": 0.816985011100769, "learning_rate": 6.767830123913241e-06, "loss": 0.2713, "step": 2160 }, { "epoch": 2.2441800310398343, "grad_norm": 0.8586083650588989, "learning_rate": 6.734014215515321e-06, "loss": 0.2746, "step": 2170 }, { "epoch": 2.2545266425245734, "grad_norm": 0.8248201608657837, "learning_rate": 6.700107906302463e-06, "loss": 0.2667, "step": 2180 }, { "epoch": 2.264873254009312, "grad_norm": 0.8500617146492004, "learning_rate": 6.6661129639406785e-06, "loss": 0.2673, "step": 2190 }, { "epoch": 2.2752198654940505, "grad_norm": 0.8682879209518433, "learning_rate": 6.6320311607167565e-06, "loss": 0.267, "step": 2200 }, { "epoch": 2.2855664769787896, "grad_norm": 0.8802211284637451, "learning_rate": 6.597864273445882e-06, "loss": 0.2701, "step": 2210 }, { "epoch": 2.295913088463528, "grad_norm": 0.8036802411079407, "learning_rate": 6.563614083378996e-06, "loss": 0.2712, "step": 2220 }, { "epoch": 2.306259699948267, "grad_norm": 0.8377422094345093, "learning_rate": 6.5292823761099355e-06, "loss": 0.2694, "step": 2230 }, { "epoch": 2.3166063114330058, "grad_norm": 0.8472813963890076, "learning_rate": 6.494870941482336e-06, "loss": 0.2709, "step": 2240 }, { "epoch": 2.3269529229177444, "grad_norm": 0.8234586715698242, "learning_rate": 6.460381573496337e-06, "loss": 0.2664, "step": 2250 }, { "epoch": 2.337299534402483, "grad_norm": 0.8779090642929077, "learning_rate": 6.425816070215032e-06, "loss": 0.2702, "step": 2260 }, { "epoch": 2.347646145887222, "grad_norm": 0.8126579523086548, "learning_rate": 6.391176233670744e-06, "loss": 0.2699, "step": 2270 }, { "epoch": 2.3579927573719606, "grad_norm": 0.796710193157196, "learning_rate": 6.356463869771077e-06, "loss": 0.2676, "step": 2280 }, { "epoch": 2.3683393688566996, "grad_norm": 0.8544085025787354, "learning_rate": 6.3216807882047585e-06, "loss": 0.2699, "step": 2290 }, { "epoch": 2.378685980341438, "grad_norm": 0.7962964773178101, "learning_rate": 6.286828802347307e-06, "loss": 0.267, "step": 2300 }, { "epoch": 2.3890325918261768, "grad_norm": 0.8767339587211609, "learning_rate": 6.251909729166478e-06, "loss": 0.2646, "step": 2310 }, { "epoch": 2.399379203310916, "grad_norm": 0.8584967255592346, "learning_rate": 6.216925389127552e-06, "loss": 0.2645, "step": 2320 }, { "epoch": 2.4097258147956544, "grad_norm": 0.8135905265808105, "learning_rate": 6.18187760609842e-06, "loss": 0.2699, "step": 2330 }, { "epoch": 2.420072426280393, "grad_norm": 0.8041039705276489, "learning_rate": 6.146768207254498e-06, "loss": 0.2712, "step": 2340 }, { "epoch": 2.430419037765132, "grad_norm": 0.8718064427375793, "learning_rate": 6.111599022983473e-06, "loss": 0.2726, "step": 2350 }, { "epoch": 2.4407656492498706, "grad_norm": 0.7888765931129456, "learning_rate": 6.076371886789875e-06, "loss": 0.2713, "step": 2360 }, { "epoch": 2.451112260734609, "grad_norm": 0.870827317237854, "learning_rate": 6.0410886351994864e-06, "loss": 0.2672, "step": 2370 }, { "epoch": 2.4614588722193482, "grad_norm": 0.8175879716873169, "learning_rate": 6.005751107663609e-06, "loss": 0.2606, "step": 2380 }, { "epoch": 2.471805483704087, "grad_norm": 0.9149560332298279, "learning_rate": 5.970361146463149e-06, "loss": 0.2707, "step": 2390 }, { "epoch": 2.482152095188826, "grad_norm": 0.85008704662323, "learning_rate": 5.934920596612588e-06, "loss": 0.2703, "step": 2400 }, { "epoch": 2.4924987066735644, "grad_norm": 0.7932586669921875, "learning_rate": 5.89943130576378e-06, "loss": 0.266, "step": 2410 }, { "epoch": 2.502845318158303, "grad_norm": 0.8065005540847778, "learning_rate": 5.863895124109642e-06, "loss": 0.2677, "step": 2420 }, { "epoch": 2.513191929643042, "grad_norm": 0.8142856955528259, "learning_rate": 5.8283139042876865e-06, "loss": 0.2699, "step": 2430 }, { "epoch": 2.5235385411277806, "grad_norm": 0.8366583585739136, "learning_rate": 5.792689501283436e-06, "loss": 0.262, "step": 2440 }, { "epoch": 2.5338851526125197, "grad_norm": 0.800625205039978, "learning_rate": 5.757023772333721e-06, "loss": 0.2641, "step": 2450 }, { "epoch": 2.5442317640972583, "grad_norm": 0.7906954884529114, "learning_rate": 5.721318576829848e-06, "loss": 0.2675, "step": 2460 }, { "epoch": 2.554578375581997, "grad_norm": 0.7770724296569824, "learning_rate": 5.685575776220671e-06, "loss": 0.2664, "step": 2470 }, { "epoch": 2.5649249870667354, "grad_norm": 0.8194068074226379, "learning_rate": 5.649797233915539e-06, "loss": 0.2702, "step": 2480 }, { "epoch": 2.5752715985514745, "grad_norm": 0.881852388381958, "learning_rate": 5.6139848151871575e-06, "loss": 0.2681, "step": 2490 }, { "epoch": 2.585618210036213, "grad_norm": 0.8650614023208618, "learning_rate": 5.578140387074335e-06, "loss": 0.2676, "step": 2500 }, { "epoch": 2.595964821520952, "grad_norm": 0.8367857933044434, "learning_rate": 5.542265818284652e-06, "loss": 0.2702, "step": 2510 }, { "epoch": 2.6063114330056907, "grad_norm": 0.8549493551254272, "learning_rate": 5.506362979097042e-06, "loss": 0.2694, "step": 2520 }, { "epoch": 2.6166580444904293, "grad_norm": 0.8078193068504333, "learning_rate": 5.470433741264281e-06, "loss": 0.266, "step": 2530 }, { "epoch": 2.627004655975168, "grad_norm": 0.7991025447845459, "learning_rate": 5.4344799779154045e-06, "loss": 0.2674, "step": 2540 }, { "epoch": 2.637351267459907, "grad_norm": 0.8104662895202637, "learning_rate": 5.398503563458064e-06, "loss": 0.2654, "step": 2550 }, { "epoch": 2.6476978789446455, "grad_norm": 0.8707109093666077, "learning_rate": 5.362506373480794e-06, "loss": 0.2665, "step": 2560 }, { "epoch": 2.6580444904293845, "grad_norm": 0.8271766304969788, "learning_rate": 5.326490284655238e-06, "loss": 0.2719, "step": 2570 }, { "epoch": 2.668391101914123, "grad_norm": 0.8149173855781555, "learning_rate": 5.290457174638314e-06, "loss": 0.2676, "step": 2580 }, { "epoch": 2.6787377133988617, "grad_norm": 0.8067767024040222, "learning_rate": 5.254408921974312e-06, "loss": 0.2726, "step": 2590 }, { "epoch": 2.6890843248836007, "grad_norm": 0.8474196791648865, "learning_rate": 5.218347405996973e-06, "loss": 0.2681, "step": 2600 }, { "epoch": 2.6994309363683393, "grad_norm": 0.8074548244476318, "learning_rate": 5.1822745067315e-06, "loss": 0.267, "step": 2610 }, { "epoch": 2.7097775478530783, "grad_norm": 0.8411250114440918, "learning_rate": 5.146192104796556e-06, "loss": 0.2651, "step": 2620 }, { "epoch": 2.720124159337817, "grad_norm": 0.8280041217803955, "learning_rate": 5.110102081306208e-06, "loss": 0.2758, "step": 2630 }, { "epoch": 2.7304707708225555, "grad_norm": 0.8358988165855408, "learning_rate": 5.074006317771873e-06, "loss": 0.2614, "step": 2640 }, { "epoch": 2.740817382307294, "grad_norm": 0.8451923131942749, "learning_rate": 5.037906696004209e-06, "loss": 0.262, "step": 2650 }, { "epoch": 2.751163993792033, "grad_norm": 0.8164005279541016, "learning_rate": 5.0018050980150244e-06, "loss": 0.2644, "step": 2660 }, { "epoch": 2.7615106052767717, "grad_norm": 0.8651638031005859, "learning_rate": 4.965703405919154e-06, "loss": 0.2702, "step": 2670 }, { "epoch": 2.7718572167615108, "grad_norm": 0.8426215648651123, "learning_rate": 4.929603501836336e-06, "loss": 0.2597, "step": 2680 }, { "epoch": 2.7822038282462493, "grad_norm": 0.8239193558692932, "learning_rate": 4.8935072677931e-06, "loss": 0.2637, "step": 2690 }, { "epoch": 2.792550439730988, "grad_norm": 0.8287822008132935, "learning_rate": 4.857416585624635e-06, "loss": 0.2643, "step": 2700 }, { "epoch": 2.802897051215727, "grad_norm": 0.8135037422180176, "learning_rate": 4.821333336876691e-06, "loss": 0.2623, "step": 2710 }, { "epoch": 2.8132436627004656, "grad_norm": 0.8966447710990906, "learning_rate": 4.785259402707489e-06, "loss": 0.2633, "step": 2720 }, { "epoch": 2.8235902741852046, "grad_norm": 0.836792528629303, "learning_rate": 4.749196663789642e-06, "loss": 0.2683, "step": 2730 }, { "epoch": 2.833936885669943, "grad_norm": 0.806907057762146, "learning_rate": 4.713147000212112e-06, "loss": 0.2698, "step": 2740 }, { "epoch": 2.8442834971546818, "grad_norm": 0.8352254629135132, "learning_rate": 4.677112291382191e-06, "loss": 0.2673, "step": 2750 }, { "epoch": 2.8546301086394203, "grad_norm": 0.8209269642829895, "learning_rate": 4.641094415927529e-06, "loss": 0.2672, "step": 2760 }, { "epoch": 2.8649767201241594, "grad_norm": 0.7578653693199158, "learning_rate": 4.605095251598184e-06, "loss": 0.2649, "step": 2770 }, { "epoch": 2.875323331608898, "grad_norm": 0.8572791814804077, "learning_rate": 4.569116675168729e-06, "loss": 0.2642, "step": 2780 }, { "epoch": 2.885669943093637, "grad_norm": 0.8002661466598511, "learning_rate": 4.53316056234041e-06, "loss": 0.2669, "step": 2790 }, { "epoch": 2.8960165545783756, "grad_norm": 0.8416422009468079, "learning_rate": 4.4972287876433675e-06, "loss": 0.2684, "step": 2800 }, { "epoch": 2.906363166063114, "grad_norm": 0.7953474521636963, "learning_rate": 4.461323224338895e-06, "loss": 0.2616, "step": 2810 }, { "epoch": 2.916709777547853, "grad_norm": 0.9063292145729065, "learning_rate": 4.425445744321785e-06, "loss": 0.2654, "step": 2820 }, { "epoch": 2.927056389032592, "grad_norm": 0.8513032793998718, "learning_rate": 4.389598218022742e-06, "loss": 0.2707, "step": 2830 }, { "epoch": 2.937403000517331, "grad_norm": 0.8203792572021484, "learning_rate": 4.353782514310872e-06, "loss": 0.2672, "step": 2840 }, { "epoch": 2.9477496120020694, "grad_norm": 0.8411531448364258, "learning_rate": 4.318000500396242e-06, "loss": 0.264, "step": 2850 }, { "epoch": 2.958096223486808, "grad_norm": 0.8241459727287292, "learning_rate": 4.28225404173254e-06, "loss": 0.2692, "step": 2860 }, { "epoch": 2.9684428349715466, "grad_norm": 0.8481301665306091, "learning_rate": 4.24654500191983e-06, "loss": 0.2617, "step": 2870 }, { "epoch": 2.9787894464562856, "grad_norm": 0.8325083255767822, "learning_rate": 4.210875242607381e-06, "loss": 0.271, "step": 2880 }, { "epoch": 2.989136057941024, "grad_norm": 0.7716158032417297, "learning_rate": 4.175246623396619e-06, "loss": 0.2606, "step": 2890 }, { "epoch": 2.9994826694257632, "grad_norm": 0.8174447417259216, "learning_rate": 4.139661001744178e-06, "loss": 0.261, "step": 2900 }, { "epoch": 3.009311950336265, "grad_norm": 0.8203701972961426, "learning_rate": 4.104120232865072e-06, "loss": 0.1988, "step": 2910 }, { "epoch": 3.0196585618210037, "grad_norm": 0.7871333956718445, "learning_rate": 4.06862616963596e-06, "loss": 0.1842, "step": 2920 }, { "epoch": 3.0300051733057423, "grad_norm": 0.846492350101471, "learning_rate": 4.033180662498557e-06, "loss": 0.1912, "step": 2930 }, { "epoch": 3.0403517847904813, "grad_norm": 0.8526950478553772, "learning_rate": 3.997785559363163e-06, "loss": 0.1851, "step": 2940 }, { "epoch": 3.05069839627522, "grad_norm": 0.8476978540420532, "learning_rate": 3.9624427055123285e-06, "loss": 0.1853, "step": 2950 }, { "epoch": 3.0610450077599585, "grad_norm": 0.8506020307540894, "learning_rate": 3.927153943504644e-06, "loss": 0.1863, "step": 2960 }, { "epoch": 3.0713916192446975, "grad_norm": 0.8641788959503174, "learning_rate": 3.891921113078684e-06, "loss": 0.1849, "step": 2970 }, { "epoch": 3.081738230729436, "grad_norm": 0.8302586674690247, "learning_rate": 3.856746051057096e-06, "loss": 0.1823, "step": 2980 }, { "epoch": 3.0920848422141747, "grad_norm": 0.8413335084915161, "learning_rate": 3.8216305912508425e-06, "loss": 0.1819, "step": 2990 }, { "epoch": 3.1024314536989137, "grad_norm": 0.8483896851539612, "learning_rate": 3.786576564363588e-06, "loss": 0.1851, "step": 3000 }, { "epoch": 3.1127780651836523, "grad_norm": 0.864549458026886, "learning_rate": 3.7515857978962666e-06, "loss": 0.1879, "step": 3010 }, { "epoch": 3.1231246766683913, "grad_norm": 0.8734090328216553, "learning_rate": 3.7166601160518025e-06, "loss": 0.185, "step": 3020 }, { "epoch": 3.13347128815313, "grad_norm": 0.8504880666732788, "learning_rate": 3.681801339640012e-06, "loss": 0.1866, "step": 3030 }, { "epoch": 3.1438178996378685, "grad_norm": 0.8266710638999939, "learning_rate": 3.64701128598267e-06, "loss": 0.1865, "step": 3040 }, { "epoch": 3.1541645111226075, "grad_norm": 0.8905497789382935, "learning_rate": 3.612291768818772e-06, "loss": 0.1906, "step": 3050 }, { "epoch": 3.164511122607346, "grad_norm": 0.8969372510910034, "learning_rate": 3.5776445982099774e-06, "loss": 0.1829, "step": 3060 }, { "epoch": 3.1748577340920847, "grad_norm": 0.8385536670684814, "learning_rate": 3.54307158044624e-06, "loss": 0.1871, "step": 3070 }, { "epoch": 3.1852043455768237, "grad_norm": 0.8617509007453918, "learning_rate": 3.508574517951642e-06, "loss": 0.189, "step": 3080 }, { "epoch": 3.1955509570615623, "grad_norm": 0.8500669002532959, "learning_rate": 3.474155209190425e-06, "loss": 0.1864, "step": 3090 }, { "epoch": 3.205897568546301, "grad_norm": 0.843228816986084, "learning_rate": 3.439815448573231e-06, "loss": 0.1874, "step": 3100 }, { "epoch": 3.21624418003104, "grad_norm": 0.8573629856109619, "learning_rate": 3.405557026363554e-06, "loss": 0.1838, "step": 3110 }, { "epoch": 3.2265907915157785, "grad_norm": 0.9198394417762756, "learning_rate": 3.3713817285844005e-06, "loss": 0.1896, "step": 3120 }, { "epoch": 3.236937403000517, "grad_norm": 0.8321995735168457, "learning_rate": 3.337291336925183e-06, "loss": 0.1869, "step": 3130 }, { "epoch": 3.247284014485256, "grad_norm": 0.820503830909729, "learning_rate": 3.3032876286488342e-06, "loss": 0.1852, "step": 3140 }, { "epoch": 3.2576306259699948, "grad_norm": 0.887699544429779, "learning_rate": 3.269372376499148e-06, "loss": 0.1874, "step": 3150 }, { "epoch": 3.2679772374547333, "grad_norm": 0.8699961304664612, "learning_rate": 3.23554734860836e-06, "loss": 0.1882, "step": 3160 }, { "epoch": 3.2783238489394724, "grad_norm": 0.8927399516105652, "learning_rate": 3.2018143084049718e-06, "loss": 0.1865, "step": 3170 }, { "epoch": 3.288670460424211, "grad_norm": 0.9565382599830627, "learning_rate": 3.1681750145218094e-06, "loss": 0.1878, "step": 3180 }, { "epoch": 3.29901707190895, "grad_norm": 0.8601614236831665, "learning_rate": 3.134631220704348e-06, "loss": 0.1833, "step": 3190 }, { "epoch": 3.3093636833936886, "grad_norm": 0.8467817902565002, "learning_rate": 3.101184675719274e-06, "loss": 0.1883, "step": 3200 }, { "epoch": 3.319710294878427, "grad_norm": 0.832239031791687, "learning_rate": 3.0678371232633232e-06, "loss": 0.1863, "step": 3210 }, { "epoch": 3.330056906363166, "grad_norm": 0.8744021654129028, "learning_rate": 3.0345903018723677e-06, "loss": 0.1868, "step": 3220 }, { "epoch": 3.340403517847905, "grad_norm": 0.8571172952651978, "learning_rate": 3.001445944830782e-06, "loss": 0.1905, "step": 3230 }, { "epoch": 3.3507501293326434, "grad_norm": 0.8718222975730896, "learning_rate": 2.9684057800810844e-06, "loss": 0.1842, "step": 3240 }, { "epoch": 3.3610967408173824, "grad_norm": 0.8766878247261047, "learning_rate": 2.9354715301338477e-06, "loss": 0.1929, "step": 3250 }, { "epoch": 3.371443352302121, "grad_norm": 0.8255637288093567, "learning_rate": 2.9026449119778978e-06, "loss": 0.1842, "step": 3260 }, { "epoch": 3.3817899637868596, "grad_norm": 0.8655493855476379, "learning_rate": 2.8699276369908042e-06, "loss": 0.1864, "step": 3270 }, { "epoch": 3.3921365752715986, "grad_norm": 0.8465432524681091, "learning_rate": 2.8373214108496574e-06, "loss": 0.186, "step": 3280 }, { "epoch": 3.402483186756337, "grad_norm": 0.855995237827301, "learning_rate": 2.8048279334421468e-06, "loss": 0.1855, "step": 3290 }, { "epoch": 3.4128297982410762, "grad_norm": 0.8721122741699219, "learning_rate": 2.772448898777932e-06, "loss": 0.1874, "step": 3300 }, { "epoch": 3.423176409725815, "grad_norm": 0.8814127445220947, "learning_rate": 2.740185994900339e-06, "loss": 0.1842, "step": 3310 }, { "epoch": 3.4335230212105534, "grad_norm": 0.8728169798851013, "learning_rate": 2.7080409037983484e-06, "loss": 0.1856, "step": 3320 }, { "epoch": 3.4438696326952924, "grad_norm": 0.8483415842056274, "learning_rate": 2.6760153013189115e-06, "loss": 0.183, "step": 3330 }, { "epoch": 3.454216244180031, "grad_norm": 0.9060035943984985, "learning_rate": 2.6441108570795717e-06, "loss": 0.1867, "step": 3340 }, { "epoch": 3.4645628556647696, "grad_norm": 0.8824280500411987, "learning_rate": 2.6123292343814345e-06, "loss": 0.1853, "step": 3350 }, { "epoch": 3.4749094671495087, "grad_norm": 0.8550218343734741, "learning_rate": 2.5806720901224474e-06, "loss": 0.1856, "step": 3360 }, { "epoch": 3.4852560786342472, "grad_norm": 0.9069307446479797, "learning_rate": 2.549141074711019e-06, "loss": 0.1845, "step": 3370 }, { "epoch": 3.495602690118986, "grad_norm": 0.891136646270752, "learning_rate": 2.5177378319799707e-06, "loss": 0.1826, "step": 3380 }, { "epoch": 3.505949301603725, "grad_norm": 0.8266403079032898, "learning_rate": 2.4864639991008526e-06, "loss": 0.1849, "step": 3390 }, { "epoch": 3.5162959130884635, "grad_norm": 0.8583815693855286, "learning_rate": 2.4553212064985776e-06, "loss": 0.1867, "step": 3400 }, { "epoch": 3.5266425245732025, "grad_norm": 0.8285118937492371, "learning_rate": 2.42431107776643e-06, "loss": 0.1866, "step": 3410 }, { "epoch": 3.536989136057941, "grad_norm": 0.8632713556289673, "learning_rate": 2.3934352295814094e-06, "loss": 0.1885, "step": 3420 }, { "epoch": 3.5473357475426797, "grad_norm": 0.8700264692306519, "learning_rate": 2.3626952716199647e-06, "loss": 0.1863, "step": 3430 }, { "epoch": 3.5576823590274183, "grad_norm": 0.8530793786048889, "learning_rate": 2.332092806474061e-06, "loss": 0.1859, "step": 3440 }, { "epoch": 3.5680289705121573, "grad_norm": 0.8744956254959106, "learning_rate": 2.301629429567638e-06, "loss": 0.1835, "step": 3450 }, { "epoch": 3.578375581996896, "grad_norm": 0.919748067855835, "learning_rate": 2.2713067290734262e-06, "loss": 0.1802, "step": 3460 }, { "epoch": 3.588722193481635, "grad_norm": 0.8674561977386475, "learning_rate": 2.2411262858301613e-06, "loss": 0.1872, "step": 3470 }, { "epoch": 3.5990688049663735, "grad_norm": 0.86203533411026, "learning_rate": 2.21108967326016e-06, "loss": 0.1839, "step": 3480 }, { "epoch": 3.609415416451112, "grad_norm": 0.8447973132133484, "learning_rate": 2.181198457287295e-06, "loss": 0.1851, "step": 3490 }, { "epoch": 3.619762027935851, "grad_norm": 0.8653585910797119, "learning_rate": 2.1514541962553533e-06, "loss": 0.1835, "step": 3500 }, { "epoch": 3.6301086394205897, "grad_norm": 0.8519960045814514, "learning_rate": 2.1218584408467996e-06, "loss": 0.1843, "step": 3510 }, { "epoch": 3.6404552509053287, "grad_norm": 0.8715120553970337, "learning_rate": 2.092412734001932e-06, "loss": 0.1837, "step": 3520 }, { "epoch": 3.6508018623900673, "grad_norm": 0.8701560497283936, "learning_rate": 2.06311861083844e-06, "loss": 0.1826, "step": 3530 }, { "epoch": 3.661148473874806, "grad_norm": 0.8274082541465759, "learning_rate": 2.03397759857137e-06, "loss": 0.1832, "step": 3540 }, { "epoch": 3.6714950853595445, "grad_norm": 0.8705015778541565, "learning_rate": 2.0049912164335155e-06, "loss": 0.1845, "step": 3550 }, { "epoch": 3.6818416968442835, "grad_norm": 0.8589960932731628, "learning_rate": 1.9761609755962064e-06, "loss": 0.1819, "step": 3560 }, { "epoch": 3.692188308329022, "grad_norm": 0.8815021514892578, "learning_rate": 1.947488379090527e-06, "loss": 0.1858, "step": 3570 }, { "epoch": 3.702534919813761, "grad_norm": 0.8352708220481873, "learning_rate": 1.9189749217289576e-06, "loss": 0.1856, "step": 3580 }, { "epoch": 3.7128815312984997, "grad_norm": 0.8597686886787415, "learning_rate": 1.890622090027443e-06, "loss": 0.1871, "step": 3590 }, { "epoch": 3.7232281427832383, "grad_norm": 0.8759775757789612, "learning_rate": 1.8624313621278984e-06, "loss": 0.1818, "step": 3600 }, { "epoch": 3.7335747542679774, "grad_norm": 0.8692873120307922, "learning_rate": 1.8344042077211438e-06, "loss": 0.1861, "step": 3610 }, { "epoch": 3.743921365752716, "grad_norm": 0.8584245443344116, "learning_rate": 1.8065420879702888e-06, "loss": 0.1836, "step": 3620 }, { "epoch": 3.754267977237455, "grad_norm": 0.861050546169281, "learning_rate": 1.7788464554345468e-06, "loss": 0.1877, "step": 3630 }, { "epoch": 3.7646145887221936, "grad_norm": 0.879095196723938, "learning_rate": 1.7513187539935188e-06, "loss": 0.1861, "step": 3640 }, { "epoch": 3.774961200206932, "grad_norm": 0.8849213719367981, "learning_rate": 1.7239604187719127e-06, "loss": 0.1833, "step": 3650 }, { "epoch": 3.7853078116916707, "grad_norm": 0.9280983209609985, "learning_rate": 1.6967728760647267e-06, "loss": 0.1829, "step": 3660 }, { "epoch": 3.7956544231764098, "grad_norm": 0.8512122631072998, "learning_rate": 1.6697575432628842e-06, "loss": 0.181, "step": 3670 }, { "epoch": 3.8060010346611484, "grad_norm": 0.8548404574394226, "learning_rate": 1.6429158287793512e-06, "loss": 0.1804, "step": 3680 }, { "epoch": 3.8163476461458874, "grad_norm": 0.919730007648468, "learning_rate": 1.6162491319757029e-06, "loss": 0.1848, "step": 3690 }, { "epoch": 3.826694257630626, "grad_norm": 0.9120166301727295, "learning_rate": 1.589758843089172e-06, "loss": 0.1811, "step": 3700 }, { "epoch": 3.8370408691153646, "grad_norm": 0.9141139984130859, "learning_rate": 1.5634463431601655e-06, "loss": 0.1815, "step": 3710 }, { "epoch": 3.8473874806001036, "grad_norm": 0.8539378046989441, "learning_rate": 1.5373130039602753e-06, "loss": 0.1826, "step": 3720 }, { "epoch": 3.857734092084842, "grad_norm": 0.8497625589370728, "learning_rate": 1.5113601879207536e-06, "loss": 0.1826, "step": 3730 }, { "epoch": 3.8680807035695812, "grad_norm": 0.8764408230781555, "learning_rate": 1.4855892480614903e-06, "loss": 0.1834, "step": 3740 }, { "epoch": 3.87842731505432, "grad_norm": 0.8945449590682983, "learning_rate": 1.460001527920467e-06, "loss": 0.1848, "step": 3750 }, { "epoch": 3.8887739265390584, "grad_norm": 0.890354335308075, "learning_rate": 1.4345983614837239e-06, "loss": 0.1805, "step": 3760 }, { "epoch": 3.899120538023797, "grad_norm": 0.8389227986335754, "learning_rate": 1.4093810731158058e-06, "loss": 0.1804, "step": 3770 }, { "epoch": 3.909467149508536, "grad_norm": 0.8829708695411682, "learning_rate": 1.3843509774907222e-06, "loss": 0.1847, "step": 3780 }, { "epoch": 3.9198137609932746, "grad_norm": 0.8937520384788513, "learning_rate": 1.359509379523402e-06, "loss": 0.1835, "step": 3790 }, { "epoch": 3.9301603724780136, "grad_norm": 0.8619959354400635, "learning_rate": 1.3348575743016735e-06, "loss": 0.1801, "step": 3800 }, { "epoch": 3.9405069839627522, "grad_norm": 0.8984729051589966, "learning_rate": 1.3103968470187384e-06, "loss": 0.1788, "step": 3810 }, { "epoch": 3.950853595447491, "grad_norm": 0.9100914597511292, "learning_rate": 1.286128472906173e-06, "loss": 0.1822, "step": 3820 }, { "epoch": 3.96120020693223, "grad_norm": 0.8695194721221924, "learning_rate": 1.2620537171674418e-06, "loss": 0.1858, "step": 3830 }, { "epoch": 3.9715468184169684, "grad_norm": 0.950613796710968, "learning_rate": 1.2381738349119443e-06, "loss": 0.1849, "step": 3840 }, { "epoch": 3.981893429901707, "grad_norm": 0.8430088758468628, "learning_rate": 1.2144900710895757e-06, "loss": 0.1793, "step": 3850 }, { "epoch": 3.992240041386446, "grad_norm": 0.8936494588851929, "learning_rate": 1.1910036604258268e-06, "loss": 0.1822, "step": 3860 }, { "epoch": 4.002069322296948, "grad_norm": 0.7884753346443176, "learning_rate": 1.1677158273574047e-06, "loss": 0.1719, "step": 3870 }, { "epoch": 4.0124159337816865, "grad_norm": 0.778055727481842, "learning_rate": 1.1446277859684146e-06, "loss": 0.1306, "step": 3880 }, { "epoch": 4.022762545266425, "grad_norm": 0.8549284934997559, "learning_rate": 1.1217407399270459e-06, "loss": 0.1255, "step": 3890 }, { "epoch": 4.033109156751164, "grad_norm": 0.8523449301719666, "learning_rate": 1.0990558824228336e-06, "loss": 0.1266, "step": 3900 }, { "epoch": 4.043455768235903, "grad_norm": 0.816981852054596, "learning_rate": 1.0765743961044445e-06, "loss": 0.125, "step": 3910 }, { "epoch": 4.053802379720642, "grad_norm": 0.8329877853393555, "learning_rate": 1.0542974530180327e-06, "loss": 0.1256, "step": 3920 }, { "epoch": 4.06414899120538, "grad_norm": 0.8189502954483032, "learning_rate": 1.0322262145461203e-06, "loss": 0.1249, "step": 3930 }, { "epoch": 4.074495602690119, "grad_norm": 0.8544175624847412, "learning_rate": 1.0103618313470642e-06, "loss": 0.1259, "step": 3940 }, { "epoch": 4.0848422141748575, "grad_norm": 0.8433131575584412, "learning_rate": 9.88705443295057e-07, "loss": 0.1265, "step": 3950 }, { "epoch": 4.095188825659596, "grad_norm": 0.84726482629776, "learning_rate": 9.672581794207141e-07, "loss": 0.1262, "step": 3960 }, { "epoch": 4.1055354371443356, "grad_norm": 0.8315642476081848, "learning_rate": 9.460211578521966e-07, "loss": 0.1247, "step": 3970 }, { "epoch": 4.115882048629074, "grad_norm": 0.8212100267410278, "learning_rate": 9.249954857569326e-07, "loss": 0.1243, "step": 3980 }, { "epoch": 4.126228660113813, "grad_norm": 0.9180471897125244, "learning_rate": 9.041822592838873e-07, "loss": 0.1237, "step": 3990 }, { "epoch": 4.136575271598551, "grad_norm": 0.8375534415245056, "learning_rate": 8.835825635064266e-07, "loss": 0.124, "step": 4000 }, { "epoch": 4.14692188308329, "grad_norm": 0.8534741997718811, "learning_rate": 8.631974723657344e-07, "loss": 0.1242, "step": 4010 }, { "epoch": 4.157268494568029, "grad_norm": 0.8509218692779541, "learning_rate": 8.430280486148368e-07, "loss": 0.1222, "step": 4020 }, { "epoch": 4.167615106052768, "grad_norm": 0.8550025820732117, "learning_rate": 8.230753437631889e-07, "loss": 0.125, "step": 4030 }, { "epoch": 4.177961717537507, "grad_norm": 0.8948724865913391, "learning_rate": 8.033403980218596e-07, "loss": 0.1241, "step": 4040 }, { "epoch": 4.188308329022245, "grad_norm": 0.858658492565155, "learning_rate": 7.838242402492973e-07, "loss": 0.1237, "step": 4050 }, { "epoch": 4.198654940506984, "grad_norm": 0.8689480423927307, "learning_rate": 7.645278878976952e-07, "loss": 0.1272, "step": 4060 }, { "epoch": 4.209001551991722, "grad_norm": 0.8573870062828064, "learning_rate": 7.454523469599484e-07, "loss": 0.1254, "step": 4070 }, { "epoch": 4.219348163476462, "grad_norm": 0.9156485199928284, "learning_rate": 7.265986119172036e-07, "loss": 0.1227, "step": 4080 }, { "epoch": 4.2296947749612, "grad_norm": 0.9127551317214966, "learning_rate": 7.079676656870138e-07, "loss": 0.1272, "step": 4090 }, { "epoch": 4.240041386445939, "grad_norm": 0.8755181431770325, "learning_rate": 6.895604795720989e-07, "loss": 0.1273, "step": 4100 }, { "epoch": 4.250387997930678, "grad_norm": 0.8257960677146912, "learning_rate": 6.713780132097053e-07, "loss": 0.1248, "step": 4110 }, { "epoch": 4.260734609415416, "grad_norm": 0.8848944902420044, "learning_rate": 6.534212145215751e-07, "loss": 0.1234, "step": 4120 }, { "epoch": 4.271081220900156, "grad_norm": 0.8648582100868225, "learning_rate": 6.356910196645271e-07, "loss": 0.1251, "step": 4130 }, { "epoch": 4.281427832384894, "grad_norm": 0.8881234526634216, "learning_rate": 6.181883529816585e-07, "loss": 0.1275, "step": 4140 }, { "epoch": 4.291774443869633, "grad_norm": 0.8552895784378052, "learning_rate": 6.009141269541424e-07, "loss": 0.1242, "step": 4150 }, { "epoch": 4.302121055354371, "grad_norm": 0.8629468679428101, "learning_rate": 5.838692421536696e-07, "loss": 0.1228, "step": 4160 }, { "epoch": 4.31246766683911, "grad_norm": 0.8863692283630371, "learning_rate": 5.67054587195488e-07, "loss": 0.1232, "step": 4170 }, { "epoch": 4.322814278323849, "grad_norm": 0.8393474221229553, "learning_rate": 5.504710386920869e-07, "loss": 0.1225, "step": 4180 }, { "epoch": 4.333160889808588, "grad_norm": 0.8772891759872437, "learning_rate": 5.341194612074824e-07, "loss": 0.1248, "step": 4190 }, { "epoch": 4.343507501293327, "grad_norm": 0.8435797095298767, "learning_rate": 5.180007072121562e-07, "loss": 0.1233, "step": 4200 }, { "epoch": 4.353854112778065, "grad_norm": 0.860095739364624, "learning_rate": 5.021156170386021e-07, "loss": 0.1266, "step": 4210 }, { "epoch": 4.364200724262804, "grad_norm": 0.8484596610069275, "learning_rate": 4.86465018837532e-07, "loss": 0.1223, "step": 4220 }, { "epoch": 4.374547335747542, "grad_norm": 0.8837497234344482, "learning_rate": 4.71049728534681e-07, "loss": 0.1232, "step": 4230 }, { "epoch": 4.384893947232282, "grad_norm": 0.8328167200088501, "learning_rate": 4.5587054978828814e-07, "loss": 0.1285, "step": 4240 }, { "epoch": 4.3952405587170205, "grad_norm": 0.8719879984855652, "learning_rate": 4.4092827394718485e-07, "loss": 0.124, "step": 4250 }, { "epoch": 4.405587170201759, "grad_norm": 0.8502692580223083, "learning_rate": 4.2622368000955207e-07, "loss": 0.1219, "step": 4260 }, { "epoch": 4.415933781686498, "grad_norm": 0.8925228118896484, "learning_rate": 4.1175753458229495e-07, "loss": 0.1266, "step": 4270 }, { "epoch": 4.426280393171236, "grad_norm": 0.8601601719856262, "learning_rate": 3.9753059184108547e-07, "loss": 0.1229, "step": 4280 }, { "epoch": 4.436627004655975, "grad_norm": 0.8607948422431946, "learning_rate": 3.8354359349104143e-07, "loss": 0.1254, "step": 4290 }, { "epoch": 4.446973616140714, "grad_norm": 0.8845228552818298, "learning_rate": 3.697972687280599e-07, "loss": 0.1245, "step": 4300 }, { "epoch": 4.457320227625453, "grad_norm": 0.9021503329277039, "learning_rate": 3.5629233420079733e-07, "loss": 0.1242, "step": 4310 }, { "epoch": 4.4676668391101915, "grad_norm": 0.8827393054962158, "learning_rate": 3.430294939733131e-07, "loss": 0.1265, "step": 4320 }, { "epoch": 4.47801345059493, "grad_norm": 0.9011342525482178, "learning_rate": 3.3000943948836183e-07, "loss": 0.1245, "step": 4330 }, { "epoch": 4.488360062079669, "grad_norm": 0.8301414251327515, "learning_rate": 3.1723284953134594e-07, "loss": 0.125, "step": 4340 }, { "epoch": 4.498706673564408, "grad_norm": 0.8322227597236633, "learning_rate": 3.047003901949258e-07, "loss": 0.1231, "step": 4350 }, { "epoch": 4.509053285049147, "grad_norm": 0.8764147758483887, "learning_rate": 2.9241271484429736e-07, "loss": 0.1229, "step": 4360 }, { "epoch": 4.519399896533885, "grad_norm": 0.8860657215118408, "learning_rate": 2.803704640831284e-07, "loss": 0.1232, "step": 4370 }, { "epoch": 4.529746508018624, "grad_norm": 0.9050237536430359, "learning_rate": 2.685742657201601e-07, "loss": 0.1221, "step": 4380 }, { "epoch": 4.5400931195033625, "grad_norm": 0.8230318427085876, "learning_rate": 2.570247347364779e-07, "loss": 0.1221, "step": 4390 }, { "epoch": 4.550439730988101, "grad_norm": 0.8647773265838623, "learning_rate": 2.4572247325345135e-07, "loss": 0.1234, "step": 4400 }, { "epoch": 4.56078634247284, "grad_norm": 0.833219051361084, "learning_rate": 2.34668070501341e-07, "loss": 0.1233, "step": 4410 }, { "epoch": 4.571132953957579, "grad_norm": 0.8522652983665466, "learning_rate": 2.2386210278858124e-07, "loss": 0.1212, "step": 4420 }, { "epoch": 4.581479565442318, "grad_norm": 0.9003979563713074, "learning_rate": 2.1330513347173398e-07, "loss": 0.1255, "step": 4430 }, { "epoch": 4.591826176927056, "grad_norm": 0.8572881817817688, "learning_rate": 2.0299771292612014e-07, "loss": 0.1231, "step": 4440 }, { "epoch": 4.602172788411795, "grad_norm": 0.8578411340713501, "learning_rate": 1.9294037851712465e-07, "loss": 0.1236, "step": 4450 }, { "epoch": 4.612519399896534, "grad_norm": 0.8606471419334412, "learning_rate": 1.83133654572184e-07, "loss": 0.1249, "step": 4460 }, { "epoch": 4.622866011381273, "grad_norm": 0.8600118160247803, "learning_rate": 1.7357805235344694e-07, "loss": 0.1231, "step": 4470 }, { "epoch": 4.6332126228660115, "grad_norm": 0.848580539226532, "learning_rate": 1.6427407003112517e-07, "loss": 0.1212, "step": 4480 }, { "epoch": 4.64355923435075, "grad_norm": 0.8755767941474915, "learning_rate": 1.5522219265751925e-07, "loss": 0.1208, "step": 4490 }, { "epoch": 4.653905845835489, "grad_norm": 0.8640385866165161, "learning_rate": 1.464228921417321e-07, "loss": 0.1202, "step": 4500 }, { "epoch": 4.664252457320227, "grad_norm": 0.8280548453330994, "learning_rate": 1.3787662722506256e-07, "loss": 0.1206, "step": 4510 }, { "epoch": 4.674599068804966, "grad_norm": 0.8286001682281494, "learning_rate": 1.295838434570973e-07, "loss": 0.1211, "step": 4520 }, { "epoch": 4.684945680289705, "grad_norm": 0.8765918612480164, "learning_rate": 1.2154497317247439e-07, "loss": 0.1238, "step": 4530 }, { "epoch": 4.695292291774444, "grad_norm": 0.9076476693153381, "learning_rate": 1.137604354683497e-07, "loss": 0.1261, "step": 4540 }, { "epoch": 4.7056389032591825, "grad_norm": 0.8170851469039917, "learning_rate": 1.0623063618254548e-07, "loss": 0.1222, "step": 4550 }, { "epoch": 4.715985514743921, "grad_norm": 0.8892994523048401, "learning_rate": 9.895596787239114e-08, "loss": 0.1244, "step": 4560 }, { "epoch": 4.726332126228661, "grad_norm": 0.864942729473114, "learning_rate": 9.193680979426189e-08, "loss": 0.1227, "step": 4570 }, { "epoch": 4.736678737713399, "grad_norm": 0.873085618019104, "learning_rate": 8.517352788380173e-08, "loss": 0.1241, "step": 4580 }, { "epoch": 4.747025349198138, "grad_norm": 0.8733391761779785, "learning_rate": 7.866647473685041e-08, "loss": 0.1222, "step": 4590 }, { "epoch": 4.757371960682876, "grad_norm": 0.8656638264656067, "learning_rate": 7.241598959105645e-08, "loss": 0.1196, "step": 4600 }, { "epoch": 4.767718572167615, "grad_norm": 0.8693123459815979, "learning_rate": 6.642239830819574e-08, "loss": 0.123, "step": 4610 }, { "epoch": 4.7780651836523536, "grad_norm": 0.8679444789886475, "learning_rate": 6.068601335718127e-08, "loss": 0.1236, "step": 4620 }, { "epoch": 4.788411795137092, "grad_norm": 0.9099794030189514, "learning_rate": 5.520713379777276e-08, "loss": 0.1237, "step": 4630 }, { "epoch": 4.798758406621832, "grad_norm": 0.8581921458244324, "learning_rate": 4.9986045264984185e-08, "loss": 0.1235, "step": 4640 }, { "epoch": 4.80910501810657, "grad_norm": 0.8875443339347839, "learning_rate": 4.5023019954197334e-08, "loss": 0.121, "step": 4650 }, { "epoch": 4.819451629591309, "grad_norm": 0.9026442170143127, "learning_rate": 4.031831660696484e-08, "loss": 0.1232, "step": 4660 }, { "epoch": 4.829798241076047, "grad_norm": 0.8439733982086182, "learning_rate": 3.5872180497526497e-08, "loss": 0.1204, "step": 4670 }, { "epoch": 4.840144852560786, "grad_norm": 0.8329970240592957, "learning_rate": 3.1684843420017316e-08, "loss": 0.1251, "step": 4680 }, { "epoch": 4.8504914640455254, "grad_norm": 0.8968592286109924, "learning_rate": 2.775652367638826e-08, "loss": 0.1207, "step": 4690 }, { "epoch": 4.860838075530264, "grad_norm": 0.8602431416511536, "learning_rate": 2.4087426065020902e-08, "loss": 0.1223, "step": 4700 }, { "epoch": 4.871184687015003, "grad_norm": 0.8694334626197815, "learning_rate": 2.0677741870053227e-08, "loss": 0.1247, "step": 4710 }, { "epoch": 4.881531298499741, "grad_norm": 0.8522930145263672, "learning_rate": 1.7527648851406453e-08, "loss": 0.1245, "step": 4720 }, { "epoch": 4.89187790998448, "grad_norm": 0.8731361627578735, "learning_rate": 1.4637311235516926e-08, "loss": 0.121, "step": 4730 }, { "epoch": 4.902224521469218, "grad_norm": 0.8769620656967163, "learning_rate": 1.2006879706776276e-08, "loss": 0.1225, "step": 4740 }, { "epoch": 4.912571132953958, "grad_norm": 0.8584715127944946, "learning_rate": 9.636491399673264e-09, "loss": 0.1194, "step": 4750 }, { "epoch": 4.9229177444386965, "grad_norm": 0.8418834805488586, "learning_rate": 7.526269891646176e-09, "loss": 0.122, "step": 4760 }, { "epoch": 4.933264355923435, "grad_norm": 0.8689851760864258, "learning_rate": 5.676325196640187e-09, "loss": 0.1208, "step": 4770 }, { "epoch": 4.943610967408174, "grad_norm": 0.8356772661209106, "learning_rate": 4.086753759370288e-09, "loss": 0.1204, "step": 4780 }, { "epoch": 4.953957578892912, "grad_norm": 0.9326891303062439, "learning_rate": 2.757638450295308e-09, "loss": 0.125, "step": 4790 }, { "epoch": 4.964304190377652, "grad_norm": 0.892158031463623, "learning_rate": 1.6890485612963691e-09, "loss": 0.1217, "step": 4800 }, { "epoch": 4.97465080186239, "grad_norm": 0.8835722208023071, "learning_rate": 8.810398020653311e-10, "loss": 0.1231, "step": 4810 }, { "epoch": 4.984997413347129, "grad_norm": 0.8562841415405273, "learning_rate": 3.3365429719933727e-10, "loss": 0.1219, "step": 4820 }, { "epoch": 4.9953440248318675, "grad_norm": 0.8547988533973694, "learning_rate": 4.692058400479482e-11, "loss": 0.1233, "step": 4830 }, { "epoch": 5.0, "step": 4835, "total_flos": 7.821379912118514e+19, "train_loss": 0.2716523540056972, "train_runtime": 48548.948, "train_samples_per_second": 6.369, "train_steps_per_second": 0.1 } ], "logging_steps": 10, "max_steps": 4835, "num_input_tokens_seen": 0, "num_train_epochs": 5, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 7.821379912118514e+19, "train_batch_size": 4, "trial_name": null, "trial_params": null }