| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.0, |
| "eval_steps": 100, |
| "global_step": 921, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.03257328990228013, |
| "grad_norm": 4.319990634918213, |
| "learning_rate": 7.741935483870967e-07, |
| "loss": 1.3255, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.06514657980456026, |
| "grad_norm": 3.764126777648926, |
| "learning_rate": 1.6344086021505377e-06, |
| "loss": 1.2843, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.09771986970684039, |
| "grad_norm": 4.150228023529053, |
| "learning_rate": 2.4946236559139784e-06, |
| "loss": 1.2342, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.13029315960912052, |
| "grad_norm": 1.6511605978012085, |
| "learning_rate": 3.3548387096774194e-06, |
| "loss": 0.9679, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.16286644951140064, |
| "grad_norm": 0.7357141971588135, |
| "learning_rate": 4.21505376344086e-06, |
| "loss": 0.7937, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.19543973941368079, |
| "grad_norm": 0.598514974117279, |
| "learning_rate": 5.075268817204301e-06, |
| "loss": 0.6905, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.2280130293159609, |
| "grad_norm": 0.6065505743026733, |
| "learning_rate": 5.935483870967741e-06, |
| "loss": 0.6328, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.26058631921824105, |
| "grad_norm": 0.5301859378814697, |
| "learning_rate": 6.7956989247311825e-06, |
| "loss": 0.5722, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.2931596091205212, |
| "grad_norm": 0.5616416335105896, |
| "learning_rate": 7.655913978494624e-06, |
| "loss": 0.5132, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.3257328990228013, |
| "grad_norm": 0.7361940145492554, |
| "learning_rate": 7.998963539365396e-06, |
| "loss": 0.4687, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.3257328990228013, |
| "eval_loss": 0.4614461660385132, |
| "eval_runtime": 33.8621, |
| "eval_samples_per_second": 7.619, |
| "eval_steps_per_second": 3.81, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.3583061889250814, |
| "grad_norm": 0.6860578656196594, |
| "learning_rate": 7.992631558305708e-06, |
| "loss": 0.4247, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.39087947882736157, |
| "grad_norm": 0.6234003901481628, |
| "learning_rate": 7.980552510965744e-06, |
| "loss": 0.38, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.4234527687296417, |
| "grad_norm": 0.7827787399291992, |
| "learning_rate": 7.962743784145322e-06, |
| "loss": 0.3475, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.4560260586319218, |
| "grad_norm": 1.0882517099380493, |
| "learning_rate": 7.939231012048832e-06, |
| "loss": 0.2904, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.48859934853420195, |
| "grad_norm": 0.9678266048431396, |
| "learning_rate": 7.910048039386883e-06, |
| "loss": 0.2646, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.5211726384364821, |
| "grad_norm": 1.0340553522109985, |
| "learning_rate": 7.875236872659625e-06, |
| "loss": 0.2408, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.5537459283387622, |
| "grad_norm": 0.8255159258842468, |
| "learning_rate": 7.834847619691822e-06, |
| "loss": 0.1985, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.5863192182410424, |
| "grad_norm": 0.9649443626403809, |
| "learning_rate": 7.788938417506746e-06, |
| "loss": 0.1981, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.6188925081433225, |
| "grad_norm": 0.7817046642303467, |
| "learning_rate": 7.737575348642683e-06, |
| "loss": 0.1947, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.6514657980456026, |
| "grad_norm": 0.7640238404273987, |
| "learning_rate": 7.680832346032546e-06, |
| "loss": 0.1752, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.6514657980456026, |
| "eval_loss": 0.1747174710035324, |
| "eval_runtime": 33.2887, |
| "eval_samples_per_second": 7.75, |
| "eval_steps_per_second": 3.875, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.6840390879478827, |
| "grad_norm": 0.701442301273346, |
| "learning_rate": 7.618791086583449e-06, |
| "loss": 0.1768, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.7166123778501629, |
| "grad_norm": 0.8486807942390442, |
| "learning_rate": 7.5515408736095e-06, |
| "loss": 0.1667, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.749185667752443, |
| "grad_norm": 0.9864675998687744, |
| "learning_rate": 7.4791785082869865e-06, |
| "loss": 0.1505, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.7817589576547231, |
| "grad_norm": 0.9136099815368652, |
| "learning_rate": 7.401808150317004e-06, |
| "loss": 0.1518, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.8143322475570033, |
| "grad_norm": 0.6785700917243958, |
| "learning_rate": 7.3195411679961005e-06, |
| "loss": 0.1485, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.8469055374592834, |
| "grad_norm": 0.9049308896064758, |
| "learning_rate": 7.232495977910721e-06, |
| "loss": 0.144, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.8794788273615635, |
| "grad_norm": 0.8609828352928162, |
| "learning_rate": 7.140797874486235e-06, |
| "loss": 0.1471, |
| "step": 270 |
| }, |
| { |
| "epoch": 0.9120521172638436, |
| "grad_norm": 0.8438140749931335, |
| "learning_rate": 7.044578849635858e-06, |
| "loss": 0.1424, |
| "step": 280 |
| }, |
| { |
| "epoch": 0.9446254071661238, |
| "grad_norm": 0.7648624181747437, |
| "learning_rate": 6.943977402769108e-06, |
| "loss": 0.1385, |
| "step": 290 |
| }, |
| { |
| "epoch": 0.9771986970684039, |
| "grad_norm": 0.6212359666824341, |
| "learning_rate": 6.839138341433236e-06, |
| "loss": 0.1428, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.9771986970684039, |
| "eval_loss": 0.1377791315317154, |
| "eval_runtime": 33.2415, |
| "eval_samples_per_second": 7.761, |
| "eval_steps_per_second": 3.881, |
| "step": 300 |
| }, |
| { |
| "epoch": 1.009771986970684, |
| "grad_norm": 0.5962893962860107, |
| "learning_rate": 6.730212572874616e-06, |
| "loss": 0.1364, |
| "step": 310 |
| }, |
| { |
| "epoch": 1.0423452768729642, |
| "grad_norm": 0.7699111700057983, |
| "learning_rate": 6.6173568868201244e-06, |
| "loss": 0.1332, |
| "step": 320 |
| }, |
| { |
| "epoch": 1.0749185667752443, |
| "grad_norm": 0.6283778548240662, |
| "learning_rate": 6.500733729791155e-06, |
| "loss": 0.1254, |
| "step": 330 |
| }, |
| { |
| "epoch": 1.1074918566775245, |
| "grad_norm": 0.623935341835022, |
| "learning_rate": 6.380510971275159e-06, |
| "loss": 0.13, |
| "step": 340 |
| }, |
| { |
| "epoch": 1.1400651465798046, |
| "grad_norm": 0.6022305488586426, |
| "learning_rate": 6.256861662091248e-06, |
| "loss": 0.138, |
| "step": 350 |
| }, |
| { |
| "epoch": 1.1726384364820848, |
| "grad_norm": 0.716827392578125, |
| "learning_rate": 6.1299637852977185e-06, |
| "loss": 0.1363, |
| "step": 360 |
| }, |
| { |
| "epoch": 1.205211726384365, |
| "grad_norm": 0.7210624814033508, |
| "learning_rate": 6e-06, |
| "loss": 0.1371, |
| "step": 370 |
| }, |
| { |
| "epoch": 1.237785016286645, |
| "grad_norm": 0.5957107543945312, |
| "learning_rate": 5.867157378427834e-06, |
| "loss": 0.1308, |
| "step": 380 |
| }, |
| { |
| "epoch": 1.2703583061889252, |
| "grad_norm": 0.6254829168319702, |
| "learning_rate": 5.731627136660109e-06, |
| "loss": 0.1298, |
| "step": 390 |
| }, |
| { |
| "epoch": 1.3029315960912053, |
| "grad_norm": 0.5790061950683594, |
| "learning_rate": 5.593604359384966e-06, |
| "loss": 0.1312, |
| "step": 400 |
| }, |
| { |
| "epoch": 1.3029315960912053, |
| "eval_loss": 0.13056354224681854, |
| "eval_runtime": 33.286, |
| "eval_samples_per_second": 7.751, |
| "eval_steps_per_second": 3.876, |
| "step": 400 |
| }, |
| { |
| "epoch": 1.3355048859934853, |
| "grad_norm": 0.6293231844902039, |
| "learning_rate": 5.453287719091365e-06, |
| "loss": 0.1342, |
| "step": 410 |
| }, |
| { |
| "epoch": 1.3680781758957654, |
| "grad_norm": 0.7954932451248169, |
| "learning_rate": 5.3108791900962995e-06, |
| "loss": 0.1312, |
| "step": 420 |
| }, |
| { |
| "epoch": 1.4006514657980456, |
| "grad_norm": 0.5331751108169556, |
| "learning_rate": 5.1665837578192995e-06, |
| "loss": 0.1319, |
| "step": 430 |
| }, |
| { |
| "epoch": 1.4332247557003257, |
| "grad_norm": 0.7147168517112732, |
| "learning_rate": 5.020609123722704e-06, |
| "loss": 0.1307, |
| "step": 440 |
| }, |
| { |
| "epoch": 1.4657980456026058, |
| "grad_norm": 0.6740135550498962, |
| "learning_rate": 4.873165406342406e-06, |
| "loss": 0.133, |
| "step": 450 |
| }, |
| { |
| "epoch": 1.498371335504886, |
| "grad_norm": 0.6303773522377014, |
| "learning_rate": 4.724464838839423e-06, |
| "loss": 0.1203, |
| "step": 460 |
| }, |
| { |
| "epoch": 1.5309446254071661, |
| "grad_norm": 0.6301987171173096, |
| "learning_rate": 4.574721463507632e-06, |
| "loss": 0.1174, |
| "step": 470 |
| }, |
| { |
| "epoch": 1.5635179153094463, |
| "grad_norm": 0.4884192943572998, |
| "learning_rate": 4.424150823677419e-06, |
| "loss": 0.1221, |
| "step": 480 |
| }, |
| { |
| "epoch": 1.5960912052117264, |
| "grad_norm": 0.8104236721992493, |
| "learning_rate": 4.272969653458684e-06, |
| "loss": 0.1248, |
| "step": 490 |
| }, |
| { |
| "epoch": 1.6286644951140063, |
| "grad_norm": 0.6095107197761536, |
| "learning_rate": 4.1213955657698496e-06, |
| "loss": 0.1221, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.6286644951140063, |
| "eval_loss": 0.12636995315551758, |
| "eval_runtime": 33.9806, |
| "eval_samples_per_second": 7.593, |
| "eval_steps_per_second": 3.796, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.6612377850162865, |
| "grad_norm": 0.5424883961677551, |
| "learning_rate": 3.969646739101868e-06, |
| "loss": 0.1196, |
| "step": 510 |
| }, |
| { |
| "epoch": 1.6938110749185666, |
| "grad_norm": 0.8408002257347107, |
| "learning_rate": 3.817941603468146e-06, |
| "loss": 0.1249, |
| "step": 520 |
| }, |
| { |
| "epoch": 1.7263843648208468, |
| "grad_norm": 0.6321083307266235, |
| "learning_rate": 3.666498525992426e-06, |
| "loss": 0.1258, |
| "step": 530 |
| }, |
| { |
| "epoch": 1.758957654723127, |
| "grad_norm": 0.705021321773529, |
| "learning_rate": 3.5155354965871845e-06, |
| "loss": 0.1259, |
| "step": 540 |
| }, |
| { |
| "epoch": 1.791530944625407, |
| "grad_norm": 0.5481059551239014, |
| "learning_rate": 3.3652698141749946e-06, |
| "loss": 0.1273, |
| "step": 550 |
| }, |
| { |
| "epoch": 1.8241042345276872, |
| "grad_norm": 0.5651680827140808, |
| "learning_rate": 3.215917773904527e-06, |
| "loss": 0.1218, |
| "step": 560 |
| }, |
| { |
| "epoch": 1.8566775244299674, |
| "grad_norm": 0.9432061314582825, |
| "learning_rate": 3.067694355811374e-06, |
| "loss": 0.1267, |
| "step": 570 |
| }, |
| { |
| "epoch": 1.8892508143322475, |
| "grad_norm": 0.7050531506538391, |
| "learning_rate": 2.9208129153719025e-06, |
| "loss": 0.1286, |
| "step": 580 |
| }, |
| { |
| "epoch": 1.9218241042345277, |
| "grad_norm": 0.6143736243247986, |
| "learning_rate": 2.7754848763955035e-06, |
| "loss": 0.1236, |
| "step": 590 |
| }, |
| { |
| "epoch": 1.9543973941368078, |
| "grad_norm": 0.5079554319381714, |
| "learning_rate": 2.631919426697325e-06, |
| "loss": 0.1281, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.9543973941368078, |
| "eval_loss": 0.12330074608325958, |
| "eval_runtime": 34.0218, |
| "eval_samples_per_second": 7.583, |
| "eval_steps_per_second": 3.792, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.986970684039088, |
| "grad_norm": 0.6454936861991882, |
| "learning_rate": 2.490323216989543e-06, |
| "loss": 0.127, |
| "step": 610 |
| }, |
| { |
| "epoch": 2.019543973941368, |
| "grad_norm": 0.5652497410774231, |
| "learning_rate": 2.35090006342457e-06, |
| "loss": 0.1288, |
| "step": 620 |
| }, |
| { |
| "epoch": 2.0521172638436482, |
| "grad_norm": 0.5710923671722412, |
| "learning_rate": 2.2138506542184e-06, |
| "loss": 0.1151, |
| "step": 630 |
| }, |
| { |
| "epoch": 2.0846905537459284, |
| "grad_norm": 0.5721701383590698, |
| "learning_rate": 2.079372260776338e-06, |
| "loss": 0.1187, |
| "step": 640 |
| }, |
| { |
| "epoch": 2.1172638436482085, |
| "grad_norm": 0.6961202025413513, |
| "learning_rate": 1.9476584537369703e-06, |
| "loss": 0.1236, |
| "step": 650 |
| }, |
| { |
| "epoch": 2.1498371335504887, |
| "grad_norm": 0.6558916568756104, |
| "learning_rate": 1.8188988243430696e-06, |
| "loss": 0.1118, |
| "step": 660 |
| }, |
| { |
| "epoch": 2.182410423452769, |
| "grad_norm": 0.6206583976745605, |
| "learning_rate": 1.6932787115405317e-06, |
| "loss": 0.1207, |
| "step": 670 |
| }, |
| { |
| "epoch": 2.214983713355049, |
| "grad_norm": 0.5952445864677429, |
| "learning_rate": 1.5709789351981212e-06, |
| "loss": 0.1208, |
| "step": 680 |
| }, |
| { |
| "epoch": 2.247557003257329, |
| "grad_norm": 0.6680878400802612, |
| "learning_rate": 1.4521755358320865e-06, |
| "loss": 0.1222, |
| "step": 690 |
| }, |
| { |
| "epoch": 2.2801302931596092, |
| "grad_norm": 0.5293084383010864, |
| "learning_rate": 1.3370395212102378e-06, |
| "loss": 0.1136, |
| "step": 700 |
| }, |
| { |
| "epoch": 2.2801302931596092, |
| "eval_loss": 0.1221059188246727, |
| "eval_runtime": 34.0931, |
| "eval_samples_per_second": 7.568, |
| "eval_steps_per_second": 3.784, |
| "step": 700 |
| }, |
| { |
| "epoch": 2.3127035830618894, |
| "grad_norm": 0.6006386280059814, |
| "learning_rate": 1.2257366202002693e-06, |
| "loss": 0.1241, |
| "step": 710 |
| }, |
| { |
| "epoch": 2.3452768729641695, |
| "grad_norm": 0.6543139815330505, |
| "learning_rate": 1.1184270442166187e-06, |
| "loss": 0.1265, |
| "step": 720 |
| }, |
| { |
| "epoch": 2.3778501628664497, |
| "grad_norm": 0.694369375705719, |
| "learning_rate": 1.015265256609258e-06, |
| "loss": 0.121, |
| "step": 730 |
| }, |
| { |
| "epoch": 2.41042345276873, |
| "grad_norm": 0.5214750170707703, |
| "learning_rate": 9.163997503263364e-07, |
| "loss": 0.1215, |
| "step": 740 |
| }, |
| { |
| "epoch": 2.44299674267101, |
| "grad_norm": 0.8416522145271301, |
| "learning_rate": 8.219728341707424e-07, |
| "loss": 0.1229, |
| "step": 750 |
| }, |
| { |
| "epoch": 2.47557003257329, |
| "grad_norm": 0.5506000518798828, |
| "learning_rate": 7.32120427958232e-07, |
| "loss": 0.1176, |
| "step": 760 |
| }, |
| { |
| "epoch": 2.5081433224755703, |
| "grad_norm": 0.6719224452972412, |
| "learning_rate": 6.469718668719827e-07, |
| "loss": 0.1201, |
| "step": 770 |
| }, |
| { |
| "epoch": 2.5407166123778504, |
| "grad_norm": 0.7880155444145203, |
| "learning_rate": 5.66649715295179e-07, |
| "loss": 0.1107, |
| "step": 780 |
| }, |
| { |
| "epoch": 2.5732899022801305, |
| "grad_norm": 0.7107143402099609, |
| "learning_rate": 4.91269590389623e-07, |
| "loss": 0.1237, |
| "step": 790 |
| }, |
| { |
| "epoch": 2.6058631921824107, |
| "grad_norm": 0.5582589507102966, |
| "learning_rate": 4.2093999567428893e-07, |
| "loss": 0.1077, |
| "step": 800 |
| }, |
| { |
| "epoch": 2.6058631921824107, |
| "eval_loss": 0.12127382308244705, |
| "eval_runtime": 34.3481, |
| "eval_samples_per_second": 7.511, |
| "eval_steps_per_second": 3.756, |
| "step": 800 |
| }, |
| { |
| "epoch": 2.6384364820846904, |
| "grad_norm": 0.626492977142334, |
| "learning_rate": 3.557621648433935e-07, |
| "loss": 0.114, |
| "step": 810 |
| }, |
| { |
| "epoch": 2.6710097719869705, |
| "grad_norm": 0.658108651638031, |
| "learning_rate": 2.958299160487789e-07, |
| "loss": 0.1213, |
| "step": 820 |
| }, |
| { |
| "epoch": 2.7035830618892507, |
| "grad_norm": 0.5767033696174622, |
| "learning_rate": 2.412295168563667e-07, |
| "loss": 0.1177, |
| "step": 830 |
| }, |
| { |
| "epoch": 2.736156351791531, |
| "grad_norm": 0.6014522910118103, |
| "learning_rate": 1.9203956007105292e-07, |
| "loss": 0.1205, |
| "step": 840 |
| }, |
| { |
| "epoch": 2.768729641693811, |
| "grad_norm": 0.5462273955345154, |
| "learning_rate": 1.4833085060880347e-07, |
| "loss": 0.1195, |
| "step": 850 |
| }, |
| { |
| "epoch": 2.801302931596091, |
| "grad_norm": 0.6881915926933289, |
| "learning_rate": 1.1016630357877276e-07, |
| "loss": 0.1217, |
| "step": 860 |
| }, |
| { |
| "epoch": 2.8338762214983713, |
| "grad_norm": 0.7328351140022278, |
| "learning_rate": 7.760085372215508e-08, |
| "loss": 0.1251, |
| "step": 870 |
| }, |
| { |
| "epoch": 2.8664495114006514, |
| "grad_norm": 0.6310160756111145, |
| "learning_rate": 5.0681376338121527e-08, |
| "loss": 0.1217, |
| "step": 880 |
| }, |
| { |
| "epoch": 2.8990228013029316, |
| "grad_norm": 0.6388233304023743, |
| "learning_rate": 2.944661981066243e-08, |
| "loss": 0.1276, |
| "step": 890 |
| }, |
| { |
| "epoch": 2.9315960912052117, |
| "grad_norm": 0.6657443046569824, |
| "learning_rate": 1.3927149833463659e-08, |
| "loss": 0.1149, |
| "step": 900 |
| }, |
| { |
| "epoch": 2.9315960912052117, |
| "eval_loss": 0.12107826769351959, |
| "eval_runtime": 34.4059, |
| "eval_samples_per_second": 7.499, |
| "eval_steps_per_second": 3.749, |
| "step": 900 |
| }, |
| { |
| "epoch": 2.964169381107492, |
| "grad_norm": 0.7934221029281616, |
| "learning_rate": 4.145305413089772e-09, |
| "loss": 0.1124, |
| "step": 910 |
| }, |
| { |
| "epoch": 2.996742671009772, |
| "grad_norm": 0.6388999819755554, |
| "learning_rate": 1.1516671381528454e-10, |
| "loss": 0.124, |
| "step": 920 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 921, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 200, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 9.388320321760666e+17, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
