diff --git "a/trainer_state.json" "b/trainer_state.json" new file mode 100644--- /dev/null +++ "b/trainer_state.json" @@ -0,0 +1,22149 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.9997401507125867, + "eval_steps": 100, + "global_step": 3126, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.00031981450758560037, + "grad_norm": 0.7444530129432678, + "learning_rate": 0.0, + "loss": 11.9102, + "step": 1 + }, + { + "epoch": 0.0006396290151712007, + "grad_norm": 0.7161079049110413, + "learning_rate": 6.382978723404255e-06, + "loss": 11.9171, + "step": 2 + }, + { + "epoch": 0.0009594435227568011, + "grad_norm": 0.7712298035621643, + "learning_rate": 1.276595744680851e-05, + "loss": 11.8957, + "step": 3 + }, + { + "epoch": 0.0012792580303424015, + "grad_norm": 0.7672861218452454, + "learning_rate": 1.9148936170212762e-05, + "loss": 11.8628, + "step": 4 + }, + { + "epoch": 0.0015990725379280018, + "grad_norm": 0.7470305562019348, + "learning_rate": 2.553191489361702e-05, + "loss": 11.814, + "step": 5 + }, + { + "epoch": 0.0019188870455136022, + "grad_norm": 0.7858774065971375, + "learning_rate": 3.1914893617021275e-05, + "loss": 11.7389, + "step": 6 + }, + { + "epoch": 0.0022387015530992023, + "grad_norm": 0.8408784866333008, + "learning_rate": 3.8297872340425525e-05, + "loss": 11.6545, + "step": 7 + }, + { + "epoch": 0.002558516060684803, + "grad_norm": 0.8673219084739685, + "learning_rate": 4.468085106382978e-05, + "loss": 11.5748, + "step": 8 + }, + { + "epoch": 0.002878330568270403, + "grad_norm": 1.0075640678405762, + "learning_rate": 5.106382978723404e-05, + "loss": 11.4413, + "step": 9 + }, + { + "epoch": 0.0031981450758560037, + "grad_norm": 1.108944296836853, + "learning_rate": 5.7446808510638294e-05, + "loss": 11.338, + "step": 10 + }, + { + "epoch": 0.003517959583441604, + "grad_norm": 1.2552160024642944, + "learning_rate": 6.382978723404255e-05, + "loss": 11.1894, + "step": 11 + }, + { + "epoch": 0.0038377740910272044, + "grad_norm": 1.4963423013687134, + "learning_rate": 7.02127659574468e-05, + "loss": 10.9778, + "step": 12 + }, + { + "epoch": 0.0041575885986128045, + "grad_norm": 1.6083056926727295, + "learning_rate": 7.659574468085105e-05, + "loss": 10.8362, + "step": 13 + }, + { + "epoch": 0.004477403106198405, + "grad_norm": 1.708142876625061, + "learning_rate": 8.297872340425531e-05, + "loss": 10.7012, + "step": 14 + }, + { + "epoch": 0.004797217613784005, + "grad_norm": 1.8373939990997314, + "learning_rate": 8.936170212765956e-05, + "loss": 10.5208, + "step": 15 + }, + { + "epoch": 0.005117032121369606, + "grad_norm": 1.8939074277877808, + "learning_rate": 9.574468085106382e-05, + "loss": 10.359, + "step": 16 + }, + { + "epoch": 0.005436846628955206, + "grad_norm": 1.882875919342041, + "learning_rate": 0.00010212765957446807, + "loss": 10.2582, + "step": 17 + }, + { + "epoch": 0.005756661136540806, + "grad_norm": 1.8790690898895264, + "learning_rate": 0.00010851063829787234, + "loss": 10.1316, + "step": 18 + }, + { + "epoch": 0.006076475644126406, + "grad_norm": 1.8662227392196655, + "learning_rate": 0.00011489361702127659, + "loss": 10.0097, + "step": 19 + }, + { + "epoch": 0.006396290151712007, + "grad_norm": 1.7853424549102783, + "learning_rate": 0.00012127659574468084, + "loss": 9.9487, + "step": 20 + }, + { + "epoch": 0.0067161046592976075, + "grad_norm": 1.8109134435653687, + "learning_rate": 0.0001276595744680851, + "loss": 9.7924, + "step": 21 + }, + { + "epoch": 0.007035919166883208, + "grad_norm": 1.7692561149597168, + "learning_rate": 0.00013404255319148935, + "loss": 9.6855, + "step": 22 + }, + { + "epoch": 0.007355733674468808, + "grad_norm": 1.66609787940979, + "learning_rate": 0.0001404255319148936, + "loss": 9.5986, + "step": 23 + }, + { + "epoch": 0.007675548182054409, + "grad_norm": 1.6861979961395264, + "learning_rate": 0.00014680851063829785, + "loss": 9.4258, + "step": 24 + }, + { + "epoch": 0.007995362689640009, + "grad_norm": 1.6129530668258667, + "learning_rate": 0.0001531914893617021, + "loss": 9.3046, + "step": 25 + }, + { + "epoch": 0.008315177197225609, + "grad_norm": 1.5948143005371094, + "learning_rate": 0.00015957446808510637, + "loss": 9.1652, + "step": 26 + }, + { + "epoch": 0.00863499170481121, + "grad_norm": 1.5753145217895508, + "learning_rate": 0.00016595744680851062, + "loss": 9.0456, + "step": 27 + }, + { + "epoch": 0.00895480621239681, + "grad_norm": 1.4742200374603271, + "learning_rate": 0.0001723404255319149, + "loss": 8.9336, + "step": 28 + }, + { + "epoch": 0.00927462071998241, + "grad_norm": 1.4124524593353271, + "learning_rate": 0.00017872340425531912, + "loss": 8.8231, + "step": 29 + }, + { + "epoch": 0.00959443522756801, + "grad_norm": 1.3787480592727661, + "learning_rate": 0.0001851063829787234, + "loss": 8.6996, + "step": 30 + }, + { + "epoch": 0.009914249735153612, + "grad_norm": 1.4922345876693726, + "learning_rate": 0.00019148936170212765, + "loss": 8.563, + "step": 31 + }, + { + "epoch": 0.010234064242739212, + "grad_norm": 2.428828239440918, + "learning_rate": 0.00019787234042553187, + "loss": 8.4677, + "step": 32 + }, + { + "epoch": 0.010553878750324812, + "grad_norm": 1.1586626768112183, + "learning_rate": 0.00020425531914893615, + "loss": 8.3377, + "step": 33 + }, + { + "epoch": 0.010873693257910412, + "grad_norm": 1.167616844177246, + "learning_rate": 0.0002106382978723404, + "loss": 8.2483, + "step": 34 + }, + { + "epoch": 0.011193507765496012, + "grad_norm": 1.0302687883377075, + "learning_rate": 0.00021702127659574468, + "loss": 8.1339, + "step": 35 + }, + { + "epoch": 0.011513322273081612, + "grad_norm": 0.8848469853401184, + "learning_rate": 0.0002234042553191489, + "loss": 8.0624, + "step": 36 + }, + { + "epoch": 0.011833136780667212, + "grad_norm": 0.8625127673149109, + "learning_rate": 0.00022978723404255317, + "loss": 7.9982, + "step": 37 + }, + { + "epoch": 0.012152951288252813, + "grad_norm": 0.7874417901039124, + "learning_rate": 0.00023617021276595742, + "loss": 7.8617, + "step": 38 + }, + { + "epoch": 0.012472765795838415, + "grad_norm": 0.6775516271591187, + "learning_rate": 0.00024255319148936167, + "loss": 7.8872, + "step": 39 + }, + { + "epoch": 0.012792580303424015, + "grad_norm": 0.5808572769165039, + "learning_rate": 0.0002489361702127659, + "loss": 7.7168, + "step": 40 + }, + { + "epoch": 0.013112394811009615, + "grad_norm": 0.6454863548278809, + "learning_rate": 0.0002553191489361702, + "loss": 7.7347, + "step": 41 + }, + { + "epoch": 0.013432209318595215, + "grad_norm": 0.44423365592956543, + "learning_rate": 0.0002617021276595745, + "loss": 7.6476, + "step": 42 + }, + { + "epoch": 0.013752023826180815, + "grad_norm": 0.4490496814250946, + "learning_rate": 0.0002680851063829787, + "loss": 7.5881, + "step": 43 + }, + { + "epoch": 0.014071838333766415, + "grad_norm": 0.3814181685447693, + "learning_rate": 0.000274468085106383, + "loss": 7.5605, + "step": 44 + }, + { + "epoch": 0.014391652841352015, + "grad_norm": 0.499479204416275, + "learning_rate": 0.0002808510638297872, + "loss": 7.4517, + "step": 45 + }, + { + "epoch": 0.014711467348937616, + "grad_norm": 0.5197142362594604, + "learning_rate": 0.0002872340425531915, + "loss": 7.4581, + "step": 46 + }, + { + "epoch": 0.015031281856523216, + "grad_norm": 0.5815995335578918, + "learning_rate": 0.0002936170212765957, + "loss": 7.4529, + "step": 47 + }, + { + "epoch": 0.015351096364108818, + "grad_norm": 0.8801454901695251, + "learning_rate": 0.0003, + "loss": 7.5077, + "step": 48 + }, + { + "epoch": 0.015670910871694418, + "grad_norm": 0.8975103497505188, + "learning_rate": 0.0003063829787234042, + "loss": 7.4058, + "step": 49 + }, + { + "epoch": 0.015990725379280018, + "grad_norm": 0.7290157675743103, + "learning_rate": 0.0003127659574468085, + "loss": 7.4331, + "step": 50 + }, + { + "epoch": 0.016310539886865618, + "grad_norm": 0.6545001268386841, + "learning_rate": 0.00031914893617021275, + "loss": 7.3928, + "step": 51 + }, + { + "epoch": 0.016630354394451218, + "grad_norm": 0.7309556603431702, + "learning_rate": 0.00032553191489361697, + "loss": 7.3888, + "step": 52 + }, + { + "epoch": 0.01695016890203682, + "grad_norm": 0.5015254020690918, + "learning_rate": 0.00033191489361702125, + "loss": 7.3532, + "step": 53 + }, + { + "epoch": 0.01726998340962242, + "grad_norm": 0.44359153509140015, + "learning_rate": 0.00033829787234042547, + "loss": 7.2783, + "step": 54 + }, + { + "epoch": 0.01758979791720802, + "grad_norm": 0.6965314149856567, + "learning_rate": 0.0003446808510638298, + "loss": 7.2682, + "step": 55 + }, + { + "epoch": 0.01790961242479362, + "grad_norm": 0.470643013715744, + "learning_rate": 0.000351063829787234, + "loss": 7.287, + "step": 56 + }, + { + "epoch": 0.01822942693237922, + "grad_norm": 0.7140641212463379, + "learning_rate": 0.00035744680851063825, + "loss": 7.1597, + "step": 57 + }, + { + "epoch": 0.01854924143996482, + "grad_norm": 0.4839196801185608, + "learning_rate": 0.0003638297872340425, + "loss": 7.2013, + "step": 58 + }, + { + "epoch": 0.01886905594755042, + "grad_norm": 0.46513426303863525, + "learning_rate": 0.0003702127659574468, + "loss": 7.1586, + "step": 59 + }, + { + "epoch": 0.01918887045513602, + "grad_norm": 0.529293954372406, + "learning_rate": 0.000376595744680851, + "loss": 7.1579, + "step": 60 + }, + { + "epoch": 0.019508684962721623, + "grad_norm": 0.44061756134033203, + "learning_rate": 0.0003829787234042553, + "loss": 7.0869, + "step": 61 + }, + { + "epoch": 0.019828499470307223, + "grad_norm": 0.37706422805786133, + "learning_rate": 0.0003893617021276595, + "loss": 7.0233, + "step": 62 + }, + { + "epoch": 0.020148313977892823, + "grad_norm": 0.40623822808265686, + "learning_rate": 0.00039574468085106374, + "loss": 7.0858, + "step": 63 + }, + { + "epoch": 0.020468128485478423, + "grad_norm": 0.35807713866233826, + "learning_rate": 0.0004021276595744681, + "loss": 7.1834, + "step": 64 + }, + { + "epoch": 0.020787942993064024, + "grad_norm": 0.7779726386070251, + "learning_rate": 0.0004085106382978723, + "loss": 7.0496, + "step": 65 + }, + { + "epoch": 0.021107757500649624, + "grad_norm": 0.6616011261940002, + "learning_rate": 0.0004148936170212766, + "loss": 7.0094, + "step": 66 + }, + { + "epoch": 0.021427572008235224, + "grad_norm": 0.5114516019821167, + "learning_rate": 0.0004212765957446808, + "loss": 7.1059, + "step": 67 + }, + { + "epoch": 0.021747386515820824, + "grad_norm": 0.8513749837875366, + "learning_rate": 0.0004276595744680851, + "loss": 7.0831, + "step": 68 + }, + { + "epoch": 0.022067201023406424, + "grad_norm": 0.7704946398735046, + "learning_rate": 0.00043404255319148935, + "loss": 7.0131, + "step": 69 + }, + { + "epoch": 0.022387015530992024, + "grad_norm": 0.46812355518341064, + "learning_rate": 0.00044042553191489357, + "loss": 7.0296, + "step": 70 + }, + { + "epoch": 0.022706830038577624, + "grad_norm": 0.6466042995452881, + "learning_rate": 0.0004468085106382978, + "loss": 6.9625, + "step": 71 + }, + { + "epoch": 0.023026644546163225, + "grad_norm": 0.391195684671402, + "learning_rate": 0.0004531914893617021, + "loss": 6.8848, + "step": 72 + }, + { + "epoch": 0.023346459053748825, + "grad_norm": 0.5878070592880249, + "learning_rate": 0.00045957446808510635, + "loss": 6.8603, + "step": 73 + }, + { + "epoch": 0.023666273561334425, + "grad_norm": 0.3662557005882263, + "learning_rate": 0.00046595744680851057, + "loss": 6.915, + "step": 74 + }, + { + "epoch": 0.023986088068920025, + "grad_norm": 0.470843642950058, + "learning_rate": 0.00047234042553191485, + "loss": 6.8751, + "step": 75 + }, + { + "epoch": 0.024305902576505625, + "grad_norm": 0.3192857503890991, + "learning_rate": 0.0004787234042553191, + "loss": 6.8031, + "step": 76 + }, + { + "epoch": 0.024625717084091225, + "grad_norm": 0.3653319478034973, + "learning_rate": 0.00048510638297872335, + "loss": 6.7958, + "step": 77 + }, + { + "epoch": 0.02494553159167683, + "grad_norm": 0.3151434361934662, + "learning_rate": 0.0004914893617021277, + "loss": 6.8712, + "step": 78 + }, + { + "epoch": 0.02526534609926243, + "grad_norm": 0.30310502648353577, + "learning_rate": 0.0004978723404255318, + "loss": 6.7834, + "step": 79 + }, + { + "epoch": 0.02558516060684803, + "grad_norm": 0.3447725772857666, + "learning_rate": 0.0005042553191489361, + "loss": 6.7777, + "step": 80 + }, + { + "epoch": 0.02590497511443363, + "grad_norm": 0.3242495059967041, + "learning_rate": 0.0005106382978723404, + "loss": 6.8031, + "step": 81 + }, + { + "epoch": 0.02622478962201923, + "grad_norm": 0.3238501250743866, + "learning_rate": 0.0005170212765957446, + "loss": 6.6858, + "step": 82 + }, + { + "epoch": 0.02654460412960483, + "grad_norm": 0.3448632061481476, + "learning_rate": 0.000523404255319149, + "loss": 6.7636, + "step": 83 + }, + { + "epoch": 0.02686441863719043, + "grad_norm": 0.3381258547306061, + "learning_rate": 0.0005297872340425531, + "loss": 6.7078, + "step": 84 + }, + { + "epoch": 0.02718423314477603, + "grad_norm": 0.31181225180625916, + "learning_rate": 0.0005361702127659574, + "loss": 6.6718, + "step": 85 + }, + { + "epoch": 0.02750404765236163, + "grad_norm": 0.3640996217727661, + "learning_rate": 0.0005425531914893617, + "loss": 6.6122, + "step": 86 + }, + { + "epoch": 0.02782386215994723, + "grad_norm": 0.3216536045074463, + "learning_rate": 0.000548936170212766, + "loss": 6.6551, + "step": 87 + }, + { + "epoch": 0.02814367666753283, + "grad_norm": 0.34431466460227966, + "learning_rate": 0.0005553191489361701, + "loss": 6.642, + "step": 88 + }, + { + "epoch": 0.02846349117511843, + "grad_norm": 0.3031766712665558, + "learning_rate": 0.0005617021276595744, + "loss": 6.5746, + "step": 89 + }, + { + "epoch": 0.02878330568270403, + "grad_norm": 0.2891135513782501, + "learning_rate": 0.0005680851063829787, + "loss": 6.6154, + "step": 90 + }, + { + "epoch": 0.02910312019028963, + "grad_norm": 0.42556002736091614, + "learning_rate": 0.000574468085106383, + "loss": 6.5932, + "step": 91 + }, + { + "epoch": 0.02942293469787523, + "grad_norm": 0.3020540177822113, + "learning_rate": 0.0005808510638297872, + "loss": 6.5155, + "step": 92 + }, + { + "epoch": 0.02974274920546083, + "grad_norm": 0.2864404022693634, + "learning_rate": 0.0005872340425531914, + "loss": 6.4844, + "step": 93 + }, + { + "epoch": 0.03006256371304643, + "grad_norm": 0.43021467328071594, + "learning_rate": 0.0005936170212765957, + "loss": 6.5811, + "step": 94 + }, + { + "epoch": 0.030382378220632035, + "grad_norm": 0.3243064880371094, + "learning_rate": 0.0006, + "loss": 6.5098, + "step": 95 + }, + { + "epoch": 0.030702192728217635, + "grad_norm": 0.36038529872894287, + "learning_rate": 0.0005999998389604413, + "loss": 6.5266, + "step": 96 + }, + { + "epoch": 0.031022007235803235, + "grad_norm": 0.41445377469062805, + "learning_rate": 0.0005999993558419382, + "loss": 6.517, + "step": 97 + }, + { + "epoch": 0.031341821743388835, + "grad_norm": 0.31521666049957275, + "learning_rate": 0.0005999985506450094, + "loss": 6.5555, + "step": 98 + }, + { + "epoch": 0.031661636250974436, + "grad_norm": 0.3607996106147766, + "learning_rate": 0.0005999974233705192, + "loss": 6.5131, + "step": 99 + }, + { + "epoch": 0.031981450758560036, + "grad_norm": 0.30577611923217773, + "learning_rate": 0.000599995974019678, + "loss": 6.5615, + "step": 100 + }, + { + "epoch": 0.031981450758560036, + "eval_runtime": 50.4466, + "eval_samples_per_second": 37.604, + "eval_steps_per_second": 9.416, + "step": 100 + }, + { + "epoch": 0.032301265266145636, + "grad_norm": 0.39252835512161255, + "learning_rate": 0.0005999942025940418, + "loss": 6.4804, + "step": 101 + }, + { + "epoch": 0.032621079773731236, + "grad_norm": 0.3077745735645294, + "learning_rate": 0.0005999921090955123, + "loss": 6.5259, + "step": 102 + }, + { + "epoch": 0.032940894281316836, + "grad_norm": 0.361301451921463, + "learning_rate": 0.0005999896935263372, + "loss": 6.5051, + "step": 103 + }, + { + "epoch": 0.033260708788902436, + "grad_norm": 0.47079065442085266, + "learning_rate": 0.0005999869558891097, + "loss": 6.563, + "step": 104 + }, + { + "epoch": 0.033580523296488037, + "grad_norm": 0.3291119933128357, + "learning_rate": 0.000599983896186769, + "loss": 6.494, + "step": 105 + }, + { + "epoch": 0.03390033780407364, + "grad_norm": 0.40962809324264526, + "learning_rate": 0.0005999805144226, + "loss": 6.4304, + "step": 106 + }, + { + "epoch": 0.03422015231165924, + "grad_norm": 0.2897939085960388, + "learning_rate": 0.0005999768106002334, + "loss": 6.418, + "step": 107 + }, + { + "epoch": 0.03453996681924484, + "grad_norm": 0.3297034204006195, + "learning_rate": 0.0005999727847236454, + "loss": 6.4372, + "step": 108 + }, + { + "epoch": 0.03485978132683044, + "grad_norm": 0.3491882383823395, + "learning_rate": 0.0005999684367971584, + "loss": 6.3556, + "step": 109 + }, + { + "epoch": 0.03517959583441604, + "grad_norm": 0.33439549803733826, + "learning_rate": 0.0005999637668254403, + "loss": 6.3833, + "step": 110 + }, + { + "epoch": 0.03549941034200164, + "grad_norm": 0.3155950903892517, + "learning_rate": 0.0005999587748135047, + "loss": 6.307, + "step": 111 + }, + { + "epoch": 0.03581922484958724, + "grad_norm": 0.3341004252433777, + "learning_rate": 0.000599953460766711, + "loss": 6.3729, + "step": 112 + }, + { + "epoch": 0.03613903935717284, + "grad_norm": 0.30250081419944763, + "learning_rate": 0.0005999478246907643, + "loss": 6.4178, + "step": 113 + }, + { + "epoch": 0.03645885386475844, + "grad_norm": 0.2828693985939026, + "learning_rate": 0.0005999418665917157, + "loss": 6.306, + "step": 114 + }, + { + "epoch": 0.03677866837234404, + "grad_norm": 0.3406890332698822, + "learning_rate": 0.0005999355864759614, + "loss": 6.4154, + "step": 115 + }, + { + "epoch": 0.03709848287992964, + "grad_norm": 0.296409547328949, + "learning_rate": 0.0005999289843502441, + "loss": 6.3743, + "step": 116 + }, + { + "epoch": 0.03741829738751524, + "grad_norm": 0.31075266003608704, + "learning_rate": 0.0005999220602216517, + "loss": 6.2923, + "step": 117 + }, + { + "epoch": 0.03773811189510084, + "grad_norm": 0.3076505661010742, + "learning_rate": 0.0005999148140976179, + "loss": 6.3513, + "step": 118 + }, + { + "epoch": 0.03805792640268644, + "grad_norm": 0.31625932455062866, + "learning_rate": 0.0005999072459859221, + "loss": 6.3186, + "step": 119 + }, + { + "epoch": 0.03837774091027204, + "grad_norm": 0.4036076068878174, + "learning_rate": 0.0005998993558946892, + "loss": 6.2398, + "step": 120 + }, + { + "epoch": 0.038697555417857646, + "grad_norm": 0.304392546415329, + "learning_rate": 0.0005998911438323904, + "loss": 6.3294, + "step": 121 + }, + { + "epoch": 0.039017369925443246, + "grad_norm": 0.5028601884841919, + "learning_rate": 0.000599882609807842, + "loss": 6.2317, + "step": 122 + }, + { + "epoch": 0.039337184433028846, + "grad_norm": 0.4757259488105774, + "learning_rate": 0.000599873753830206, + "loss": 6.3194, + "step": 123 + }, + { + "epoch": 0.039656998940614446, + "grad_norm": 0.4033927321434021, + "learning_rate": 0.0005998645759089901, + "loss": 6.2807, + "step": 124 + }, + { + "epoch": 0.039976813448200046, + "grad_norm": 0.4541259706020355, + "learning_rate": 0.0005998550760540478, + "loss": 6.2899, + "step": 125 + }, + { + "epoch": 0.04029662795578565, + "grad_norm": 0.3444730341434479, + "learning_rate": 0.000599845254275578, + "loss": 6.2245, + "step": 126 + }, + { + "epoch": 0.04061644246337125, + "grad_norm": 0.3179037272930145, + "learning_rate": 0.0005998351105841257, + "loss": 6.331, + "step": 127 + }, + { + "epoch": 0.04093625697095685, + "grad_norm": 0.3816128969192505, + "learning_rate": 0.0005998246449905807, + "loss": 6.2171, + "step": 128 + }, + { + "epoch": 0.04125607147854245, + "grad_norm": 0.32012107968330383, + "learning_rate": 0.0005998138575061791, + "loss": 6.1823, + "step": 129 + }, + { + "epoch": 0.04157588598612805, + "grad_norm": 0.31248337030410767, + "learning_rate": 0.000599802748142502, + "loss": 6.2566, + "step": 130 + }, + { + "epoch": 0.04189570049371365, + "grad_norm": 0.299917608499527, + "learning_rate": 0.0005997913169114768, + "loss": 6.1719, + "step": 131 + }, + { + "epoch": 0.04221551500129925, + "grad_norm": 0.31560277938842773, + "learning_rate": 0.0005997795638253759, + "loss": 6.1971, + "step": 132 + }, + { + "epoch": 0.04253532950888485, + "grad_norm": 0.3071800172328949, + "learning_rate": 0.0005997674888968171, + "loss": 6.2775, + "step": 133 + }, + { + "epoch": 0.04285514401647045, + "grad_norm": 0.27625685930252075, + "learning_rate": 0.0005997550921387643, + "loss": 6.1602, + "step": 134 + }, + { + "epoch": 0.04317495852405605, + "grad_norm": 0.27132847905158997, + "learning_rate": 0.0005997423735645265, + "loss": 6.153, + "step": 135 + }, + { + "epoch": 0.04349477303164165, + "grad_norm": 0.2841440439224243, + "learning_rate": 0.0005997293331877584, + "loss": 6.1605, + "step": 136 + }, + { + "epoch": 0.04381458753922725, + "grad_norm": 0.3245135247707367, + "learning_rate": 0.0005997159710224602, + "loss": 6.147, + "step": 137 + }, + { + "epoch": 0.04413440204681285, + "grad_norm": 0.3364606499671936, + "learning_rate": 0.0005997022870829771, + "loss": 6.1257, + "step": 138 + }, + { + "epoch": 0.04445421655439845, + "grad_norm": 0.3006080687046051, + "learning_rate": 0.0005996882813840005, + "loss": 6.1045, + "step": 139 + }, + { + "epoch": 0.04477403106198405, + "grad_norm": 0.32453054189682007, + "learning_rate": 0.0005996739539405668, + "loss": 6.1715, + "step": 140 + }, + { + "epoch": 0.04509384556956965, + "grad_norm": 0.34519049525260925, + "learning_rate": 0.0005996593047680579, + "loss": 6.1008, + "step": 141 + }, + { + "epoch": 0.04541366007715525, + "grad_norm": 0.3466300070285797, + "learning_rate": 0.0005996443338822011, + "loss": 6.1801, + "step": 142 + }, + { + "epoch": 0.04573347458474085, + "grad_norm": 0.3183090090751648, + "learning_rate": 0.000599629041299069, + "loss": 6.1171, + "step": 143 + }, + { + "epoch": 0.04605328909232645, + "grad_norm": 0.3680199980735779, + "learning_rate": 0.0005996134270350797, + "loss": 6.1576, + "step": 144 + }, + { + "epoch": 0.04637310359991205, + "grad_norm": 0.31519967317581177, + "learning_rate": 0.0005995974911069968, + "loss": 6.0688, + "step": 145 + }, + { + "epoch": 0.04669291810749765, + "grad_norm": 0.42483389377593994, + "learning_rate": 0.0005995812335319289, + "loss": 6.1205, + "step": 146 + }, + { + "epoch": 0.04701273261508325, + "grad_norm": 0.4271079897880554, + "learning_rate": 0.0005995646543273301, + "loss": 6.1039, + "step": 147 + }, + { + "epoch": 0.04733254712266885, + "grad_norm": 0.39717018604278564, + "learning_rate": 0.0005995477535109998, + "loss": 6.1387, + "step": 148 + }, + { + "epoch": 0.04765236163025445, + "grad_norm": 0.34426459670066833, + "learning_rate": 0.0005995305311010826, + "loss": 6.0979, + "step": 149 + }, + { + "epoch": 0.04797217613784005, + "grad_norm": 0.41630908846855164, + "learning_rate": 0.0005995129871160688, + "loss": 6.0947, + "step": 150 + }, + { + "epoch": 0.04829199064542565, + "grad_norm": 0.29474949836730957, + "learning_rate": 0.000599495121574793, + "loss": 6.1455, + "step": 151 + }, + { + "epoch": 0.04861180515301125, + "grad_norm": 0.3705453872680664, + "learning_rate": 0.0005994769344964359, + "loss": 6.1315, + "step": 152 + }, + { + "epoch": 0.04893161966059685, + "grad_norm": 0.36204794049263, + "learning_rate": 0.0005994584259005232, + "loss": 6.1068, + "step": 153 + }, + { + "epoch": 0.04925143416818245, + "grad_norm": 0.3252790570259094, + "learning_rate": 0.0005994395958069254, + "loss": 6.0512, + "step": 154 + }, + { + "epoch": 0.04957124867576806, + "grad_norm": 0.3430827260017395, + "learning_rate": 0.0005994204442358586, + "loss": 6.066, + "step": 155 + }, + { + "epoch": 0.04989106318335366, + "grad_norm": 0.3288519084453583, + "learning_rate": 0.0005994009712078839, + "loss": 6.0219, + "step": 156 + }, + { + "epoch": 0.05021087769093926, + "grad_norm": 0.3370048701763153, + "learning_rate": 0.0005993811767439074, + "loss": 6.051, + "step": 157 + }, + { + "epoch": 0.05053069219852486, + "grad_norm": 0.3363915979862213, + "learning_rate": 0.0005993610608651804, + "loss": 5.9932, + "step": 158 + }, + { + "epoch": 0.05085050670611046, + "grad_norm": 0.3294038772583008, + "learning_rate": 0.0005993406235932992, + "loss": 6.0723, + "step": 159 + }, + { + "epoch": 0.05117032121369606, + "grad_norm": 0.35423871874809265, + "learning_rate": 0.0005993198649502054, + "loss": 6.039, + "step": 160 + }, + { + "epoch": 0.05149013572128166, + "grad_norm": 0.39059337973594666, + "learning_rate": 0.0005992987849581852, + "loss": 6.0737, + "step": 161 + }, + { + "epoch": 0.05180995022886726, + "grad_norm": 0.3352687656879425, + "learning_rate": 0.00059927738363987, + "loss": 6.0654, + "step": 162 + }, + { + "epoch": 0.05212976473645286, + "grad_norm": 0.27634456753730774, + "learning_rate": 0.0005992556610182364, + "loss": 5.9947, + "step": 163 + }, + { + "epoch": 0.05244957924403846, + "grad_norm": 0.3383750319480896, + "learning_rate": 0.0005992336171166056, + "loss": 5.9721, + "step": 164 + }, + { + "epoch": 0.05276939375162406, + "grad_norm": 0.30598023533821106, + "learning_rate": 0.0005992112519586438, + "loss": 5.9772, + "step": 165 + }, + { + "epoch": 0.05308920825920966, + "grad_norm": 0.30256542563438416, + "learning_rate": 0.0005991885655683624, + "loss": 5.9679, + "step": 166 + }, + { + "epoch": 0.05340902276679526, + "grad_norm": 0.276569128036499, + "learning_rate": 0.0005991655579701171, + "loss": 5.8669, + "step": 167 + }, + { + "epoch": 0.05372883727438086, + "grad_norm": 0.27367281913757324, + "learning_rate": 0.0005991422291886092, + "loss": 5.9485, + "step": 168 + }, + { + "epoch": 0.05404865178196646, + "grad_norm": 0.37032443284988403, + "learning_rate": 0.000599118579248884, + "loss": 5.9011, + "step": 169 + }, + { + "epoch": 0.05436846628955206, + "grad_norm": 0.44173505902290344, + "learning_rate": 0.0005990946081763324, + "loss": 6.0507, + "step": 170 + }, + { + "epoch": 0.05468828079713766, + "grad_norm": 0.45422565937042236, + "learning_rate": 0.0005990703159966894, + "loss": 5.9675, + "step": 171 + }, + { + "epoch": 0.05500809530472326, + "grad_norm": 0.3305465579032898, + "learning_rate": 0.000599045702736035, + "loss": 5.9636, + "step": 172 + }, + { + "epoch": 0.05532790981230886, + "grad_norm": 0.3410186469554901, + "learning_rate": 0.000599020768420794, + "loss": 5.944, + "step": 173 + }, + { + "epoch": 0.05564772431989446, + "grad_norm": 0.3027215898036957, + "learning_rate": 0.000598995513077736, + "loss": 6.028, + "step": 174 + }, + { + "epoch": 0.05596753882748006, + "grad_norm": 0.35722747445106506, + "learning_rate": 0.0005989699367339748, + "loss": 6.0524, + "step": 175 + }, + { + "epoch": 0.05628735333506566, + "grad_norm": 0.3533189594745636, + "learning_rate": 0.0005989440394169692, + "loss": 5.9176, + "step": 176 + }, + { + "epoch": 0.05660716784265126, + "grad_norm": 0.45911741256713867, + "learning_rate": 0.0005989178211545223, + "loss": 5.9705, + "step": 177 + }, + { + "epoch": 0.05692698235023686, + "grad_norm": 0.4044831395149231, + "learning_rate": 0.0005988912819747822, + "loss": 5.8702, + "step": 178 + }, + { + "epoch": 0.05724679685782246, + "grad_norm": 0.3743020296096802, + "learning_rate": 0.0005988644219062412, + "loss": 5.9099, + "step": 179 + }, + { + "epoch": 0.05756661136540806, + "grad_norm": 0.3255034387111664, + "learning_rate": 0.0005988372409777362, + "loss": 5.8467, + "step": 180 + }, + { + "epoch": 0.05788642587299366, + "grad_norm": 0.39103788137435913, + "learning_rate": 0.0005988097392184486, + "loss": 5.9099, + "step": 181 + }, + { + "epoch": 0.05820624038057926, + "grad_norm": 0.34550437331199646, + "learning_rate": 0.000598781916657904, + "loss": 5.9385, + "step": 182 + }, + { + "epoch": 0.05852605488816486, + "grad_norm": 0.28878021240234375, + "learning_rate": 0.0005987537733259729, + "loss": 5.8884, + "step": 183 + }, + { + "epoch": 0.05884586939575046, + "grad_norm": 0.30863630771636963, + "learning_rate": 0.0005987253092528697, + "loss": 5.8551, + "step": 184 + }, + { + "epoch": 0.05916568390333606, + "grad_norm": 0.3268856406211853, + "learning_rate": 0.0005986965244691533, + "loss": 5.8184, + "step": 185 + }, + { + "epoch": 0.05948549841092166, + "grad_norm": 0.2993100881576538, + "learning_rate": 0.0005986674190057274, + "loss": 5.8307, + "step": 186 + }, + { + "epoch": 0.05980531291850726, + "grad_norm": 0.31807687878608704, + "learning_rate": 0.0005986379928938389, + "loss": 5.8082, + "step": 187 + }, + { + "epoch": 0.06012512742609286, + "grad_norm": 0.31077608466148376, + "learning_rate": 0.0005986082461650801, + "loss": 5.8453, + "step": 188 + }, + { + "epoch": 0.06044494193367847, + "grad_norm": 0.31170928478240967, + "learning_rate": 0.0005985781788513867, + "loss": 5.874, + "step": 189 + }, + { + "epoch": 0.06076475644126407, + "grad_norm": 0.3185946047306061, + "learning_rate": 0.000598547790985039, + "loss": 5.876, + "step": 190 + }, + { + "epoch": 0.06108457094884967, + "grad_norm": 0.30993083119392395, + "learning_rate": 0.0005985170825986613, + "loss": 5.8248, + "step": 191 + }, + { + "epoch": 0.06140438545643527, + "grad_norm": 0.3225095570087433, + "learning_rate": 0.000598486053725222, + "loss": 5.8798, + "step": 192 + }, + { + "epoch": 0.06172419996402087, + "grad_norm": 0.34688934683799744, + "learning_rate": 0.0005984547043980338, + "loss": 5.8229, + "step": 193 + }, + { + "epoch": 0.06204401447160647, + "grad_norm": 0.40069252252578735, + "learning_rate": 0.0005984230346507529, + "loss": 5.7936, + "step": 194 + }, + { + "epoch": 0.06236382897919207, + "grad_norm": 0.4236740469932556, + "learning_rate": 0.0005983910445173802, + "loss": 5.8617, + "step": 195 + }, + { + "epoch": 0.06268364348677767, + "grad_norm": 0.4122582972049713, + "learning_rate": 0.00059835873403226, + "loss": 5.809, + "step": 196 + }, + { + "epoch": 0.06300345799436327, + "grad_norm": 0.3633243441581726, + "learning_rate": 0.0005983261032300807, + "loss": 5.7554, + "step": 197 + }, + { + "epoch": 0.06332327250194887, + "grad_norm": 0.3823741674423218, + "learning_rate": 0.0005982931521458747, + "loss": 5.7342, + "step": 198 + }, + { + "epoch": 0.06364308700953447, + "grad_norm": 0.3783508539199829, + "learning_rate": 0.0005982598808150184, + "loss": 5.8786, + "step": 199 + }, + { + "epoch": 0.06396290151712007, + "grad_norm": 0.3433767855167389, + "learning_rate": 0.0005982262892732315, + "loss": 5.857, + "step": 200 + }, + { + "epoch": 0.06396290151712007, + "eval_runtime": 23.5677, + "eval_samples_per_second": 80.491, + "eval_steps_per_second": 20.155, + "step": 200 + }, + { + "epoch": 0.06428271602470567, + "grad_norm": 0.34785598516464233, + "learning_rate": 0.000598192377556578, + "loss": 5.8317, + "step": 201 + }, + { + "epoch": 0.06460253053229127, + "grad_norm": 0.31842753291130066, + "learning_rate": 0.0005981581457014652, + "loss": 5.7882, + "step": 202 + }, + { + "epoch": 0.06492234503987687, + "grad_norm": 0.3305855393409729, + "learning_rate": 0.0005981235937446446, + "loss": 5.8158, + "step": 203 + }, + { + "epoch": 0.06524215954746247, + "grad_norm": 0.31251704692840576, + "learning_rate": 0.0005980887217232107, + "loss": 5.7861, + "step": 204 + }, + { + "epoch": 0.06556197405504807, + "grad_norm": 0.345258504152298, + "learning_rate": 0.0005980535296746023, + "loss": 5.7933, + "step": 205 + }, + { + "epoch": 0.06588178856263367, + "grad_norm": 0.3426864743232727, + "learning_rate": 0.0005980180176366013, + "loss": 5.7857, + "step": 206 + }, + { + "epoch": 0.06620160307021927, + "grad_norm": 0.34403687715530396, + "learning_rate": 0.0005979821856473336, + "loss": 5.7566, + "step": 207 + }, + { + "epoch": 0.06652141757780487, + "grad_norm": 0.3185226619243622, + "learning_rate": 0.0005979460337452681, + "loss": 5.7356, + "step": 208 + }, + { + "epoch": 0.06684123208539047, + "grad_norm": 0.342356413602829, + "learning_rate": 0.0005979095619692172, + "loss": 5.7242, + "step": 209 + }, + { + "epoch": 0.06716104659297607, + "grad_norm": 0.3159177005290985, + "learning_rate": 0.0005978727703583374, + "loss": 5.7321, + "step": 210 + }, + { + "epoch": 0.06748086110056167, + "grad_norm": 0.34413307905197144, + "learning_rate": 0.0005978356589521276, + "loss": 5.766, + "step": 211 + }, + { + "epoch": 0.06780067560814727, + "grad_norm": 0.33023929595947266, + "learning_rate": 0.0005977982277904306, + "loss": 5.7338, + "step": 212 + }, + { + "epoch": 0.06812049011573287, + "grad_norm": 0.321130633354187, + "learning_rate": 0.0005977604769134325, + "loss": 5.6788, + "step": 213 + }, + { + "epoch": 0.06844030462331847, + "grad_norm": 0.3121291399002075, + "learning_rate": 0.0005977224063616625, + "loss": 5.7739, + "step": 214 + }, + { + "epoch": 0.06876011913090407, + "grad_norm": 0.3282662034034729, + "learning_rate": 0.0005976840161759931, + "loss": 5.7303, + "step": 215 + }, + { + "epoch": 0.06907993363848967, + "grad_norm": 0.2961140275001526, + "learning_rate": 0.0005976453063976396, + "loss": 5.7689, + "step": 216 + }, + { + "epoch": 0.06939974814607527, + "grad_norm": 0.3259924650192261, + "learning_rate": 0.000597606277068161, + "loss": 5.7449, + "step": 217 + }, + { + "epoch": 0.06971956265366087, + "grad_norm": 0.36330533027648926, + "learning_rate": 0.000597566928229459, + "loss": 5.7217, + "step": 218 + }, + { + "epoch": 0.07003937716124647, + "grad_norm": 0.3424849808216095, + "learning_rate": 0.0005975272599237784, + "loss": 5.6487, + "step": 219 + }, + { + "epoch": 0.07035919166883207, + "grad_norm": 0.36807698011398315, + "learning_rate": 0.0005974872721937069, + "loss": 5.7082, + "step": 220 + }, + { + "epoch": 0.07067900617641767, + "grad_norm": 0.3388591408729553, + "learning_rate": 0.0005974469650821753, + "loss": 5.7487, + "step": 221 + }, + { + "epoch": 0.07099882068400327, + "grad_norm": 0.36010366678237915, + "learning_rate": 0.0005974063386324571, + "loss": 5.687, + "step": 222 + }, + { + "epoch": 0.07131863519158887, + "grad_norm": 0.3253012001514435, + "learning_rate": 0.0005973653928881688, + "loss": 5.6282, + "step": 223 + }, + { + "epoch": 0.07163844969917448, + "grad_norm": 0.36569592356681824, + "learning_rate": 0.0005973241278932695, + "loss": 5.6524, + "step": 224 + }, + { + "epoch": 0.07195826420676008, + "grad_norm": 0.34154123067855835, + "learning_rate": 0.0005972825436920615, + "loss": 5.6554, + "step": 225 + }, + { + "epoch": 0.07227807871434568, + "grad_norm": 0.3732295632362366, + "learning_rate": 0.0005972406403291893, + "loss": 5.6876, + "step": 226 + }, + { + "epoch": 0.07259789322193128, + "grad_norm": 0.3637785315513611, + "learning_rate": 0.00059719841784964, + "loss": 5.717, + "step": 227 + }, + { + "epoch": 0.07291770772951688, + "grad_norm": 0.33887818455696106, + "learning_rate": 0.0005971558762987439, + "loss": 5.6049, + "step": 228 + }, + { + "epoch": 0.07323752223710248, + "grad_norm": 0.34123507142066956, + "learning_rate": 0.0005971130157221733, + "loss": 5.691, + "step": 229 + }, + { + "epoch": 0.07355733674468808, + "grad_norm": 0.4016885757446289, + "learning_rate": 0.0005970698361659431, + "loss": 5.6689, + "step": 230 + }, + { + "epoch": 0.07387715125227368, + "grad_norm": 0.3807375431060791, + "learning_rate": 0.000597026337676411, + "loss": 5.7384, + "step": 231 + }, + { + "epoch": 0.07419696575985928, + "grad_norm": 0.41762468218803406, + "learning_rate": 0.0005969825203002765, + "loss": 5.6623, + "step": 232 + }, + { + "epoch": 0.07451678026744488, + "grad_norm": 0.3555200397968292, + "learning_rate": 0.0005969383840845822, + "loss": 5.6791, + "step": 233 + }, + { + "epoch": 0.07483659477503048, + "grad_norm": 0.36806872487068176, + "learning_rate": 0.0005968939290767123, + "loss": 5.5886, + "step": 234 + }, + { + "epoch": 0.07515640928261608, + "grad_norm": 0.3818735182285309, + "learning_rate": 0.0005968491553243937, + "loss": 5.6905, + "step": 235 + }, + { + "epoch": 0.07547622379020168, + "grad_norm": 0.3234502375125885, + "learning_rate": 0.0005968040628756955, + "loss": 5.6351, + "step": 236 + }, + { + "epoch": 0.07579603829778728, + "grad_norm": 0.3597162663936615, + "learning_rate": 0.0005967586517790285, + "loss": 5.534, + "step": 237 + }, + { + "epoch": 0.07611585280537288, + "grad_norm": 0.36016765236854553, + "learning_rate": 0.0005967129220831461, + "loss": 5.6743, + "step": 238 + }, + { + "epoch": 0.07643566731295848, + "grad_norm": 0.3662470281124115, + "learning_rate": 0.0005966668738371436, + "loss": 5.6293, + "step": 239 + }, + { + "epoch": 0.07675548182054408, + "grad_norm": 0.3480012118816376, + "learning_rate": 0.0005966205070904582, + "loss": 5.6165, + "step": 240 + }, + { + "epoch": 0.07707529632812969, + "grad_norm": 0.40776678919792175, + "learning_rate": 0.0005965738218928693, + "loss": 5.5772, + "step": 241 + }, + { + "epoch": 0.07739511083571529, + "grad_norm": 0.3601037263870239, + "learning_rate": 0.0005965268182944976, + "loss": 5.6298, + "step": 242 + }, + { + "epoch": 0.07771492534330089, + "grad_norm": 0.3565972149372101, + "learning_rate": 0.0005964794963458063, + "loss": 5.6911, + "step": 243 + }, + { + "epoch": 0.07803473985088649, + "grad_norm": 0.3792625367641449, + "learning_rate": 0.0005964318560976001, + "loss": 5.6314, + "step": 244 + }, + { + "epoch": 0.07835455435847209, + "grad_norm": 0.3629399538040161, + "learning_rate": 0.0005963838976010252, + "loss": 5.5684, + "step": 245 + }, + { + "epoch": 0.07867436886605769, + "grad_norm": 0.35087811946868896, + "learning_rate": 0.0005963356209075701, + "loss": 5.5937, + "step": 246 + }, + { + "epoch": 0.07899418337364329, + "grad_norm": 0.37176039814949036, + "learning_rate": 0.0005962870260690641, + "loss": 5.6965, + "step": 247 + }, + { + "epoch": 0.07931399788122889, + "grad_norm": 0.39871492981910706, + "learning_rate": 0.0005962381131376788, + "loss": 5.5578, + "step": 248 + }, + { + "epoch": 0.07963381238881449, + "grad_norm": 0.3734513521194458, + "learning_rate": 0.0005961888821659268, + "loss": 5.6626, + "step": 249 + }, + { + "epoch": 0.07995362689640009, + "grad_norm": 0.3237352669239044, + "learning_rate": 0.0005961393332066623, + "loss": 5.62, + "step": 250 + }, + { + "epoch": 0.08027344140398569, + "grad_norm": 0.49298253655433655, + "learning_rate": 0.0005960894663130811, + "loss": 5.568, + "step": 251 + }, + { + "epoch": 0.0805932559115713, + "grad_norm": 0.3696758449077606, + "learning_rate": 0.0005960392815387201, + "loss": 5.6076, + "step": 252 + }, + { + "epoch": 0.0809130704191569, + "grad_norm": 0.45616376399993896, + "learning_rate": 0.0005959887789374573, + "loss": 5.5846, + "step": 253 + }, + { + "epoch": 0.0812328849267425, + "grad_norm": 0.4054776132106781, + "learning_rate": 0.0005959379585635124, + "loss": 5.5961, + "step": 254 + }, + { + "epoch": 0.0815526994343281, + "grad_norm": 0.35753971338272095, + "learning_rate": 0.0005958868204714459, + "loss": 5.5791, + "step": 255 + }, + { + "epoch": 0.0818725139419137, + "grad_norm": 0.5398254990577698, + "learning_rate": 0.0005958353647161595, + "loss": 5.592, + "step": 256 + }, + { + "epoch": 0.0821923284494993, + "grad_norm": 0.41130876541137695, + "learning_rate": 0.0005957835913528959, + "loss": 5.5233, + "step": 257 + }, + { + "epoch": 0.0825121429570849, + "grad_norm": 0.3787677586078644, + "learning_rate": 0.0005957315004372391, + "loss": 5.5271, + "step": 258 + }, + { + "epoch": 0.0828319574646705, + "grad_norm": 0.4082469344139099, + "learning_rate": 0.0005956790920251133, + "loss": 5.5147, + "step": 259 + }, + { + "epoch": 0.0831517719722561, + "grad_norm": 0.41386041045188904, + "learning_rate": 0.0005956263661727844, + "loss": 5.6523, + "step": 260 + }, + { + "epoch": 0.0834715864798417, + "grad_norm": 0.35570305585861206, + "learning_rate": 0.0005955733229368586, + "loss": 5.6393, + "step": 261 + }, + { + "epoch": 0.0837914009874273, + "grad_norm": 0.4203128218650818, + "learning_rate": 0.000595519962374283, + "loss": 5.5894, + "step": 262 + }, + { + "epoch": 0.0841112154950129, + "grad_norm": 0.3384525775909424, + "learning_rate": 0.0005954662845423452, + "loss": 5.5424, + "step": 263 + }, + { + "epoch": 0.0844310300025985, + "grad_norm": 0.37563735246658325, + "learning_rate": 0.0005954122894986736, + "loss": 5.5749, + "step": 264 + }, + { + "epoch": 0.0847508445101841, + "grad_norm": 0.3936252295970917, + "learning_rate": 0.0005953579773012374, + "loss": 5.5657, + "step": 265 + }, + { + "epoch": 0.0850706590177697, + "grad_norm": 0.39597007632255554, + "learning_rate": 0.0005953033480083456, + "loss": 5.5147, + "step": 266 + }, + { + "epoch": 0.0853904735253553, + "grad_norm": 0.3897547423839569, + "learning_rate": 0.0005952484016786483, + "loss": 5.5618, + "step": 267 + }, + { + "epoch": 0.0857102880329409, + "grad_norm": 0.353368878364563, + "learning_rate": 0.0005951931383711357, + "loss": 5.5448, + "step": 268 + }, + { + "epoch": 0.0860301025405265, + "grad_norm": 0.40237027406692505, + "learning_rate": 0.0005951375581451382, + "loss": 5.5562, + "step": 269 + }, + { + "epoch": 0.0863499170481121, + "grad_norm": 0.353664368391037, + "learning_rate": 0.0005950816610603266, + "loss": 5.6005, + "step": 270 + }, + { + "epoch": 0.0866697315556977, + "grad_norm": 0.39963239431381226, + "learning_rate": 0.0005950254471767119, + "loss": 5.5739, + "step": 271 + }, + { + "epoch": 0.0869895460632833, + "grad_norm": 0.43650707602500916, + "learning_rate": 0.0005949689165546453, + "loss": 5.5679, + "step": 272 + }, + { + "epoch": 0.0873093605708689, + "grad_norm": 0.43688809871673584, + "learning_rate": 0.0005949120692548177, + "loss": 5.5027, + "step": 273 + }, + { + "epoch": 0.0876291750784545, + "grad_norm": 0.35993266105651855, + "learning_rate": 0.0005948549053382602, + "loss": 5.5567, + "step": 274 + }, + { + "epoch": 0.0879489895860401, + "grad_norm": 0.41433578729629517, + "learning_rate": 0.0005947974248663439, + "loss": 5.574, + "step": 275 + }, + { + "epoch": 0.0882688040936257, + "grad_norm": 0.3552456796169281, + "learning_rate": 0.0005947396279007796, + "loss": 5.5176, + "step": 276 + }, + { + "epoch": 0.0885886186012113, + "grad_norm": 0.4384861886501312, + "learning_rate": 0.0005946815145036181, + "loss": 5.5558, + "step": 277 + }, + { + "epoch": 0.0889084331087969, + "grad_norm": 0.40407803654670715, + "learning_rate": 0.0005946230847372496, + "loss": 5.4679, + "step": 278 + }, + { + "epoch": 0.0892282476163825, + "grad_norm": 0.3628694415092468, + "learning_rate": 0.0005945643386644041, + "loss": 5.528, + "step": 279 + }, + { + "epoch": 0.0895480621239681, + "grad_norm": 0.43523159623146057, + "learning_rate": 0.0005945052763481514, + "loss": 5.4525, + "step": 280 + }, + { + "epoch": 0.0898678766315537, + "grad_norm": 0.44130516052246094, + "learning_rate": 0.0005944458978519006, + "loss": 5.4964, + "step": 281 + }, + { + "epoch": 0.0901876911391393, + "grad_norm": 0.3897377550601959, + "learning_rate": 0.0005943862032394, + "loss": 5.565, + "step": 282 + }, + { + "epoch": 0.0905075056467249, + "grad_norm": 0.39829137921333313, + "learning_rate": 0.000594326192574738, + "loss": 5.4365, + "step": 283 + }, + { + "epoch": 0.0908273201543105, + "grad_norm": 0.49638622999191284, + "learning_rate": 0.0005942658659223415, + "loss": 5.4247, + "step": 284 + }, + { + "epoch": 0.0911471346618961, + "grad_norm": 0.3281007409095764, + "learning_rate": 0.0005942052233469771, + "loss": 5.4456, + "step": 285 + }, + { + "epoch": 0.0914669491694817, + "grad_norm": 0.47785961627960205, + "learning_rate": 0.0005941442649137507, + "loss": 5.516, + "step": 286 + }, + { + "epoch": 0.0917867636770673, + "grad_norm": 0.37340331077575684, + "learning_rate": 0.0005940829906881066, + "loss": 5.4579, + "step": 287 + }, + { + "epoch": 0.0921065781846529, + "grad_norm": 0.38307222723960876, + "learning_rate": 0.0005940214007358293, + "loss": 5.5005, + "step": 288 + }, + { + "epoch": 0.0924263926922385, + "grad_norm": 0.4082242250442505, + "learning_rate": 0.0005939594951230412, + "loss": 5.4514, + "step": 289 + }, + { + "epoch": 0.0927462071998241, + "grad_norm": 0.36588606238365173, + "learning_rate": 0.0005938972739162041, + "loss": 5.5224, + "step": 290 + }, + { + "epoch": 0.0930660217074097, + "grad_norm": 0.36439234018325806, + "learning_rate": 0.0005938347371821183, + "loss": 5.5398, + "step": 291 + }, + { + "epoch": 0.0933858362149953, + "grad_norm": 0.3797212541103363, + "learning_rate": 0.0005937718849879232, + "loss": 5.506, + "step": 292 + }, + { + "epoch": 0.0937056507225809, + "grad_norm": 0.42287692427635193, + "learning_rate": 0.0005937087174010968, + "loss": 5.5028, + "step": 293 + }, + { + "epoch": 0.0940254652301665, + "grad_norm": 0.35854318737983704, + "learning_rate": 0.0005936452344894556, + "loss": 5.4425, + "step": 294 + }, + { + "epoch": 0.0943452797377521, + "grad_norm": 0.5297574996948242, + "learning_rate": 0.0005935814363211546, + "loss": 5.4093, + "step": 295 + }, + { + "epoch": 0.0946650942453377, + "grad_norm": 0.4768528938293457, + "learning_rate": 0.0005935173229646873, + "loss": 5.4448, + "step": 296 + }, + { + "epoch": 0.0949849087529233, + "grad_norm": 0.38121071457862854, + "learning_rate": 0.0005934528944888857, + "loss": 5.4627, + "step": 297 + }, + { + "epoch": 0.0953047232605089, + "grad_norm": 0.39731091260910034, + "learning_rate": 0.0005933881509629201, + "loss": 5.4857, + "step": 298 + }, + { + "epoch": 0.0956245377680945, + "grad_norm": 0.6086567044258118, + "learning_rate": 0.0005933230924562987, + "loss": 5.4831, + "step": 299 + }, + { + "epoch": 0.0959443522756801, + "grad_norm": 0.4189840853214264, + "learning_rate": 0.0005932577190388684, + "loss": 5.5201, + "step": 300 + }, + { + "epoch": 0.0959443522756801, + "eval_runtime": 30.3835, + "eval_samples_per_second": 62.435, + "eval_steps_per_second": 15.633, + "step": 300 + }, + { + "epoch": 0.0962641667832657, + "grad_norm": 0.5163981318473816, + "learning_rate": 0.0005931920307808138, + "loss": 5.4968, + "step": 301 + }, + { + "epoch": 0.0965839812908513, + "grad_norm": 0.3878300189971924, + "learning_rate": 0.0005931260277526574, + "loss": 5.4447, + "step": 302 + }, + { + "epoch": 0.0969037957984369, + "grad_norm": 0.40698179602622986, + "learning_rate": 0.0005930597100252602, + "loss": 5.4007, + "step": 303 + }, + { + "epoch": 0.0972236103060225, + "grad_norm": 0.43470317125320435, + "learning_rate": 0.0005929930776698205, + "loss": 5.4617, + "step": 304 + }, + { + "epoch": 0.0975434248136081, + "grad_norm": 0.34721076488494873, + "learning_rate": 0.0005929261307578747, + "loss": 5.457, + "step": 305 + }, + { + "epoch": 0.0978632393211937, + "grad_norm": 0.40585532784461975, + "learning_rate": 0.0005928588693612969, + "loss": 5.3789, + "step": 306 + }, + { + "epoch": 0.0981830538287793, + "grad_norm": 0.41711103916168213, + "learning_rate": 0.0005927912935522985, + "loss": 5.3513, + "step": 307 + }, + { + "epoch": 0.0985028683363649, + "grad_norm": 0.5503503084182739, + "learning_rate": 0.0005927234034034289, + "loss": 5.4622, + "step": 308 + }, + { + "epoch": 0.09882268284395052, + "grad_norm": 0.4085991382598877, + "learning_rate": 0.0005926551989875746, + "loss": 5.4276, + "step": 309 + }, + { + "epoch": 0.09914249735153612, + "grad_norm": 0.8164439797401428, + "learning_rate": 0.0005925866803779598, + "loss": 5.4348, + "step": 310 + }, + { + "epoch": 0.09946231185912172, + "grad_norm": 0.6821435689926147, + "learning_rate": 0.0005925178476481458, + "loss": 5.4336, + "step": 311 + }, + { + "epoch": 0.09978212636670732, + "grad_norm": 0.48287320137023926, + "learning_rate": 0.0005924487008720313, + "loss": 5.5436, + "step": 312 + }, + { + "epoch": 0.10010194087429292, + "grad_norm": 0.7320705056190491, + "learning_rate": 0.0005923792401238519, + "loss": 5.3847, + "step": 313 + }, + { + "epoch": 0.10042175538187852, + "grad_norm": 0.6495065689086914, + "learning_rate": 0.0005923094654781805, + "loss": 5.4669, + "step": 314 + }, + { + "epoch": 0.10074156988946412, + "grad_norm": 0.9512620568275452, + "learning_rate": 0.0005922393770099271, + "loss": 5.4991, + "step": 315 + }, + { + "epoch": 0.10106138439704972, + "grad_norm": 0.4396529495716095, + "learning_rate": 0.0005921689747943384, + "loss": 5.3925, + "step": 316 + }, + { + "epoch": 0.10138119890463532, + "grad_norm": 0.7574517726898193, + "learning_rate": 0.0005920982589069979, + "loss": 5.3651, + "step": 317 + }, + { + "epoch": 0.10170101341222092, + "grad_norm": 0.6250393986701965, + "learning_rate": 0.0005920272294238261, + "loss": 5.2963, + "step": 318 + }, + { + "epoch": 0.10202082791980652, + "grad_norm": 0.5038833022117615, + "learning_rate": 0.0005919558864210801, + "loss": 5.4219, + "step": 319 + }, + { + "epoch": 0.10234064242739212, + "grad_norm": 0.5173036456108093, + "learning_rate": 0.0005918842299753534, + "loss": 5.3737, + "step": 320 + }, + { + "epoch": 0.10266045693497772, + "grad_norm": 0.5143558979034424, + "learning_rate": 0.0005918122601635763, + "loss": 5.3618, + "step": 321 + }, + { + "epoch": 0.10298027144256332, + "grad_norm": 0.5874578356742859, + "learning_rate": 0.0005917399770630151, + "loss": 5.4165, + "step": 322 + }, + { + "epoch": 0.10330008595014892, + "grad_norm": 1.260666847229004, + "learning_rate": 0.000591667380751273, + "loss": 5.4044, + "step": 323 + }, + { + "epoch": 0.10361990045773452, + "grad_norm": 0.6366116404533386, + "learning_rate": 0.0005915944713062891, + "loss": 5.4607, + "step": 324 + }, + { + "epoch": 0.10393971496532012, + "grad_norm": 0.49358871579170227, + "learning_rate": 0.0005915212488063387, + "loss": 5.3529, + "step": 325 + }, + { + "epoch": 0.10425952947290572, + "grad_norm": 0.4817569851875305, + "learning_rate": 0.0005914477133300333, + "loss": 5.4031, + "step": 326 + }, + { + "epoch": 0.10457934398049132, + "grad_norm": 0.9505804181098938, + "learning_rate": 0.0005913738649563205, + "loss": 5.3761, + "step": 327 + }, + { + "epoch": 0.10489915848807692, + "grad_norm": 0.43332940340042114, + "learning_rate": 0.0005912997037644834, + "loss": 5.4563, + "step": 328 + }, + { + "epoch": 0.10521897299566252, + "grad_norm": 0.4971306324005127, + "learning_rate": 0.0005912252298341416, + "loss": 5.4257, + "step": 329 + }, + { + "epoch": 0.10553878750324812, + "grad_norm": 0.43271008133888245, + "learning_rate": 0.0005911504432452498, + "loss": 5.4006, + "step": 330 + }, + { + "epoch": 0.10585860201083372, + "grad_norm": 0.414468914270401, + "learning_rate": 0.0005910753440780988, + "loss": 5.3333, + "step": 331 + }, + { + "epoch": 0.10617841651841932, + "grad_norm": 0.5585057735443115, + "learning_rate": 0.0005909999324133148, + "loss": 5.3707, + "step": 332 + }, + { + "epoch": 0.10649823102600492, + "grad_norm": 0.43397095799446106, + "learning_rate": 0.0005909242083318596, + "loss": 5.4115, + "step": 333 + }, + { + "epoch": 0.10681804553359052, + "grad_norm": 0.642075777053833, + "learning_rate": 0.0005908481719150303, + "loss": 5.3824, + "step": 334 + }, + { + "epoch": 0.10713786004117612, + "grad_norm": 0.43352431058883667, + "learning_rate": 0.0005907718232444594, + "loss": 5.3222, + "step": 335 + }, + { + "epoch": 0.10745767454876172, + "grad_norm": 0.6455655694007874, + "learning_rate": 0.0005906951624021147, + "loss": 5.3516, + "step": 336 + }, + { + "epoch": 0.10777748905634732, + "grad_norm": 0.4931198060512543, + "learning_rate": 0.0005906181894702987, + "loss": 5.3703, + "step": 337 + }, + { + "epoch": 0.10809730356393292, + "grad_norm": 0.5147075653076172, + "learning_rate": 0.0005905409045316497, + "loss": 5.3452, + "step": 338 + }, + { + "epoch": 0.10841711807151852, + "grad_norm": 0.6032655239105225, + "learning_rate": 0.0005904633076691404, + "loss": 5.3709, + "step": 339 + }, + { + "epoch": 0.10873693257910412, + "grad_norm": 0.6385786533355713, + "learning_rate": 0.0005903853989660787, + "loss": 5.4067, + "step": 340 + }, + { + "epoch": 0.10905674708668972, + "grad_norm": 0.5276573300361633, + "learning_rate": 0.0005903071785061069, + "loss": 5.3551, + "step": 341 + }, + { + "epoch": 0.10937656159427532, + "grad_norm": 0.471971720457077, + "learning_rate": 0.0005902286463732026, + "loss": 5.3763, + "step": 342 + }, + { + "epoch": 0.10969637610186092, + "grad_norm": 0.5554648041725159, + "learning_rate": 0.0005901498026516774, + "loss": 5.3544, + "step": 343 + }, + { + "epoch": 0.11001619060944652, + "grad_norm": 0.46863773465156555, + "learning_rate": 0.0005900706474261778, + "loss": 5.3957, + "step": 344 + }, + { + "epoch": 0.11033600511703212, + "grad_norm": 0.75626140832901, + "learning_rate": 0.0005899911807816844, + "loss": 5.3291, + "step": 345 + }, + { + "epoch": 0.11065581962461772, + "grad_norm": 0.5489574670791626, + "learning_rate": 0.0005899114028035128, + "loss": 5.4232, + "step": 346 + }, + { + "epoch": 0.11097563413220332, + "grad_norm": 0.8476126790046692, + "learning_rate": 0.0005898313135773121, + "loss": 5.3686, + "step": 347 + }, + { + "epoch": 0.11129544863978892, + "grad_norm": 0.5443255305290222, + "learning_rate": 0.0005897509131890658, + "loss": 5.3413, + "step": 348 + }, + { + "epoch": 0.11161526314737452, + "grad_norm": 0.5287749171257019, + "learning_rate": 0.0005896702017250916, + "loss": 5.3828, + "step": 349 + }, + { + "epoch": 0.11193507765496012, + "grad_norm": 0.5031596422195435, + "learning_rate": 0.0005895891792720413, + "loss": 5.3627, + "step": 350 + }, + { + "epoch": 0.11225489216254572, + "grad_norm": 0.5278099179267883, + "learning_rate": 0.0005895078459169, + "loss": 5.3866, + "step": 351 + }, + { + "epoch": 0.11257470667013132, + "grad_norm": 0.47713255882263184, + "learning_rate": 0.0005894262017469872, + "loss": 5.3421, + "step": 352 + }, + { + "epoch": 0.11289452117771692, + "grad_norm": 0.8117437958717346, + "learning_rate": 0.0005893442468499557, + "loss": 5.2562, + "step": 353 + }, + { + "epoch": 0.11321433568530252, + "grad_norm": 0.5438576340675354, + "learning_rate": 0.0005892619813137923, + "loss": 5.3388, + "step": 354 + }, + { + "epoch": 0.11353415019288812, + "grad_norm": 0.9697743058204651, + "learning_rate": 0.0005891794052268167, + "loss": 5.2587, + "step": 355 + }, + { + "epoch": 0.11385396470047372, + "grad_norm": 0.5962640047073364, + "learning_rate": 0.0005890965186776825, + "loss": 5.3395, + "step": 356 + }, + { + "epoch": 0.11417377920805932, + "grad_norm": 0.5466102957725525, + "learning_rate": 0.0005890133217553765, + "loss": 5.2616, + "step": 357 + }, + { + "epoch": 0.11449359371564492, + "grad_norm": 0.6477400660514832, + "learning_rate": 0.0005889298145492185, + "loss": 5.3396, + "step": 358 + }, + { + "epoch": 0.11481340822323052, + "grad_norm": 0.5722964406013489, + "learning_rate": 0.0005888459971488618, + "loss": 5.3749, + "step": 359 + }, + { + "epoch": 0.11513322273081612, + "grad_norm": 0.8127800822257996, + "learning_rate": 0.0005887618696442925, + "loss": 5.3882, + "step": 360 + }, + { + "epoch": 0.11545303723840172, + "grad_norm": 1.206176996231079, + "learning_rate": 0.0005886774321258294, + "loss": 5.3092, + "step": 361 + }, + { + "epoch": 0.11577285174598732, + "grad_norm": 0.678299069404602, + "learning_rate": 0.0005885926846841246, + "loss": 5.3435, + "step": 362 + }, + { + "epoch": 0.11609266625357292, + "grad_norm": 0.5589863061904907, + "learning_rate": 0.0005885076274101627, + "loss": 5.2852, + "step": 363 + }, + { + "epoch": 0.11641248076115852, + "grad_norm": 0.6293617486953735, + "learning_rate": 0.0005884222603952608, + "loss": 5.3361, + "step": 364 + }, + { + "epoch": 0.11673229526874412, + "grad_norm": 0.6327272057533264, + "learning_rate": 0.0005883365837310689, + "loss": 5.2168, + "step": 365 + }, + { + "epoch": 0.11705210977632972, + "grad_norm": 0.7606511116027832, + "learning_rate": 0.0005882505975095689, + "loss": 5.2541, + "step": 366 + }, + { + "epoch": 0.11737192428391532, + "grad_norm": 0.5545910000801086, + "learning_rate": 0.0005881643018230755, + "loss": 5.3372, + "step": 367 + }, + { + "epoch": 0.11769173879150092, + "grad_norm": 0.5674049258232117, + "learning_rate": 0.0005880776967642355, + "loss": 5.1965, + "step": 368 + }, + { + "epoch": 0.11801155329908652, + "grad_norm": 0.9930325746536255, + "learning_rate": 0.0005879907824260281, + "loss": 5.2073, + "step": 369 + }, + { + "epoch": 0.11833136780667212, + "grad_norm": 0.48672759532928467, + "learning_rate": 0.0005879035589017638, + "loss": 5.278, + "step": 370 + }, + { + "epoch": 0.11865118231425772, + "grad_norm": 1.980123519897461, + "learning_rate": 0.0005878160262850859, + "loss": 5.3395, + "step": 371 + }, + { + "epoch": 0.11897099682184333, + "grad_norm": 0.5322059988975525, + "learning_rate": 0.0005877281846699689, + "loss": 5.2661, + "step": 372 + }, + { + "epoch": 0.11929081132942893, + "grad_norm": 0.6024339199066162, + "learning_rate": 0.0005876400341507194, + "loss": 5.2545, + "step": 373 + }, + { + "epoch": 0.11961062583701453, + "grad_norm": 0.6507164835929871, + "learning_rate": 0.0005875515748219757, + "loss": 5.3435, + "step": 374 + }, + { + "epoch": 0.11993044034460013, + "grad_norm": 0.9286210536956787, + "learning_rate": 0.0005874628067787072, + "loss": 5.2892, + "step": 375 + }, + { + "epoch": 0.12025025485218573, + "grad_norm": 1.1855777502059937, + "learning_rate": 0.0005873737301162151, + "loss": 5.2602, + "step": 376 + }, + { + "epoch": 0.12057006935977133, + "grad_norm": 1.3940703868865967, + "learning_rate": 0.000587284344930132, + "loss": 5.2012, + "step": 377 + }, + { + "epoch": 0.12088988386735694, + "grad_norm": 0.7048264741897583, + "learning_rate": 0.0005871946513164213, + "loss": 5.2671, + "step": 378 + }, + { + "epoch": 0.12120969837494254, + "grad_norm": 1.485435962677002, + "learning_rate": 0.000587104649371378, + "loss": 5.2638, + "step": 379 + }, + { + "epoch": 0.12152951288252814, + "grad_norm": 12.656268119812012, + "learning_rate": 0.000587014339191628, + "loss": 5.2315, + "step": 380 + }, + { + "epoch": 0.12184932739011374, + "grad_norm": 1.1891939640045166, + "learning_rate": 0.0005869237208741278, + "loss": 5.3239, + "step": 381 + }, + { + "epoch": 0.12216914189769934, + "grad_norm": 0.7396526336669922, + "learning_rate": 0.0005868327945161651, + "loss": 5.3233, + "step": 382 + }, + { + "epoch": 0.12248895640528494, + "grad_norm": 0.7021541595458984, + "learning_rate": 0.0005867415602153582, + "loss": 5.3364, + "step": 383 + }, + { + "epoch": 0.12280877091287054, + "grad_norm": 0.6708441376686096, + "learning_rate": 0.0005866500180696558, + "loss": 5.2237, + "step": 384 + }, + { + "epoch": 0.12312858542045614, + "grad_norm": 0.8744490742683411, + "learning_rate": 0.0005865581681773374, + "loss": 5.2119, + "step": 385 + }, + { + "epoch": 0.12344839992804174, + "grad_norm": 0.5799617767333984, + "learning_rate": 0.000586466010637013, + "loss": 5.2988, + "step": 386 + }, + { + "epoch": 0.12376821443562734, + "grad_norm": 2.3637585639953613, + "learning_rate": 0.0005863735455476222, + "loss": 5.2371, + "step": 387 + }, + { + "epoch": 0.12408802894321294, + "grad_norm": 1.6760447025299072, + "learning_rate": 0.0005862807730084356, + "loss": 5.3499, + "step": 388 + }, + { + "epoch": 0.12440784345079854, + "grad_norm": 0.5137502551078796, + "learning_rate": 0.0005861876931190534, + "loss": 5.3219, + "step": 389 + }, + { + "epoch": 0.12472765795838414, + "grad_norm": 0.5475123524665833, + "learning_rate": 0.0005860943059794059, + "loss": 5.2467, + "step": 390 + }, + { + "epoch": 0.12504747246596973, + "grad_norm": 0.7181013822555542, + "learning_rate": 0.0005860006116897533, + "loss": 5.193, + "step": 391 + }, + { + "epoch": 0.12536728697355534, + "grad_norm": 0.5430626273155212, + "learning_rate": 0.0005859066103506853, + "loss": 5.1984, + "step": 392 + }, + { + "epoch": 0.12568710148114093, + "grad_norm": 0.720116376876831, + "learning_rate": 0.0005858123020631218, + "loss": 5.2152, + "step": 393 + }, + { + "epoch": 0.12600691598872654, + "grad_norm": 0.7978843450546265, + "learning_rate": 0.0005857176869283118, + "loss": 5.2141, + "step": 394 + }, + { + "epoch": 0.12632673049631213, + "grad_norm": 0.7915864586830139, + "learning_rate": 0.0005856227650478335, + "loss": 5.2109, + "step": 395 + }, + { + "epoch": 0.12664654500389774, + "grad_norm": 0.8192650675773621, + "learning_rate": 0.0005855275365235953, + "loss": 5.2829, + "step": 396 + }, + { + "epoch": 0.12696635951148333, + "grad_norm": 0.4708637595176697, + "learning_rate": 0.0005854320014578338, + "loss": 5.2103, + "step": 397 + }, + { + "epoch": 0.12728617401906894, + "grad_norm": 0.6642940640449524, + "learning_rate": 0.0005853361599531155, + "loss": 5.1862, + "step": 398 + }, + { + "epoch": 0.12760598852665453, + "grad_norm": 1.0393221378326416, + "learning_rate": 0.0005852400121123353, + "loss": 5.2537, + "step": 399 + }, + { + "epoch": 0.12792580303424014, + "grad_norm": 0.6117280125617981, + "learning_rate": 0.0005851435580387175, + "loss": 5.2382, + "step": 400 + }, + { + "epoch": 0.12792580303424014, + "eval_runtime": 21.7498, + "eval_samples_per_second": 87.219, + "eval_steps_per_second": 21.839, + "step": 400 + }, + { + "epoch": 0.12824561754182573, + "grad_norm": 0.5303900837898254, + "learning_rate": 0.0005850467978358146, + "loss": 5.2677, + "step": 401 + }, + { + "epoch": 0.12856543204941134, + "grad_norm": 1.598645806312561, + "learning_rate": 0.0005849497316075084, + "loss": 5.1782, + "step": 402 + }, + { + "epoch": 0.12888524655699693, + "grad_norm": 0.7240434885025024, + "learning_rate": 0.0005848523594580086, + "loss": 5.2694, + "step": 403 + }, + { + "epoch": 0.12920506106458254, + "grad_norm": 1.5253021717071533, + "learning_rate": 0.0005847546814918538, + "loss": 5.2418, + "step": 404 + }, + { + "epoch": 0.12952487557216813, + "grad_norm": 1.6910390853881836, + "learning_rate": 0.0005846566978139108, + "loss": 5.3229, + "step": 405 + }, + { + "epoch": 0.12984469007975374, + "grad_norm": 1.0010775327682495, + "learning_rate": 0.0005845584085293745, + "loss": 5.2101, + "step": 406 + }, + { + "epoch": 0.13016450458733933, + "grad_norm": 0.9314549565315247, + "learning_rate": 0.0005844598137437682, + "loss": 5.1788, + "step": 407 + }, + { + "epoch": 0.13048431909492494, + "grad_norm": 1.2149615287780762, + "learning_rate": 0.0005843609135629427, + "loss": 5.2905, + "step": 408 + }, + { + "epoch": 0.13080413360251053, + "grad_norm": 0.5436043739318848, + "learning_rate": 0.0005842617080930771, + "loss": 5.2487, + "step": 409 + }, + { + "epoch": 0.13112394811009614, + "grad_norm": 1.5292751789093018, + "learning_rate": 0.000584162197440678, + "loss": 5.3127, + "step": 410 + }, + { + "epoch": 0.13144376261768173, + "grad_norm": 1.058875560760498, + "learning_rate": 0.0005840623817125799, + "loss": 5.1979, + "step": 411 + }, + { + "epoch": 0.13176357712526734, + "grad_norm": 0.5317851901054382, + "learning_rate": 0.0005839622610159446, + "loss": 5.2293, + "step": 412 + }, + { + "epoch": 0.13208339163285296, + "grad_norm": 0.9199758172035217, + "learning_rate": 0.0005838618354582612, + "loss": 5.1712, + "step": 413 + }, + { + "epoch": 0.13240320614043855, + "grad_norm": 0.7664812207221985, + "learning_rate": 0.0005837611051473466, + "loss": 5.1657, + "step": 414 + }, + { + "epoch": 0.13272302064802416, + "grad_norm": 12.5884428024292, + "learning_rate": 0.0005836600701913443, + "loss": 5.1841, + "step": 415 + }, + { + "epoch": 0.13304283515560975, + "grad_norm": 1.6279326677322388, + "learning_rate": 0.0005835587306987255, + "loss": 5.2522, + "step": 416 + }, + { + "epoch": 0.13336264966319536, + "grad_norm": 0.728519082069397, + "learning_rate": 0.0005834570867782875, + "loss": 5.2529, + "step": 417 + }, + { + "epoch": 0.13368246417078095, + "grad_norm": 1.019119381904602, + "learning_rate": 0.0005833551385391551, + "loss": 5.2708, + "step": 418 + }, + { + "epoch": 0.13400227867836656, + "grad_norm": 2.4016916751861572, + "learning_rate": 0.0005832528860907798, + "loss": 5.3005, + "step": 419 + }, + { + "epoch": 0.13432209318595215, + "grad_norm": 0.8830029964447021, + "learning_rate": 0.0005831503295429393, + "loss": 5.2097, + "step": 420 + }, + { + "epoch": 0.13464190769353776, + "grad_norm": 0.6238778829574585, + "learning_rate": 0.0005830474690057383, + "loss": 5.3171, + "step": 421 + }, + { + "epoch": 0.13496172220112335, + "grad_norm": 0.5882652401924133, + "learning_rate": 0.0005829443045896072, + "loss": 5.3137, + "step": 422 + }, + { + "epoch": 0.13528153670870896, + "grad_norm": 0.7491113543510437, + "learning_rate": 0.0005828408364053031, + "loss": 5.2184, + "step": 423 + }, + { + "epoch": 0.13560135121629455, + "grad_norm": 0.9572281241416931, + "learning_rate": 0.0005827370645639095, + "loss": 5.2442, + "step": 424 + }, + { + "epoch": 0.13592116572388016, + "grad_norm": 2.5302014350891113, + "learning_rate": 0.0005826329891768351, + "loss": 5.2103, + "step": 425 + }, + { + "epoch": 0.13624098023146575, + "grad_norm": 0.7268570065498352, + "learning_rate": 0.0005825286103558151, + "loss": 5.2221, + "step": 426 + }, + { + "epoch": 0.13656079473905136, + "grad_norm": 0.6264063715934753, + "learning_rate": 0.0005824239282129103, + "loss": 5.2934, + "step": 427 + }, + { + "epoch": 0.13688060924663695, + "grad_norm": 0.8299589157104492, + "learning_rate": 0.0005823189428605072, + "loss": 5.2515, + "step": 428 + }, + { + "epoch": 0.13720042375422256, + "grad_norm": 0.6055958271026611, + "learning_rate": 0.0005822136544113177, + "loss": 5.1055, + "step": 429 + }, + { + "epoch": 0.13752023826180815, + "grad_norm": 0.5393593311309814, + "learning_rate": 0.000582108062978379, + "loss": 5.1831, + "step": 430 + }, + { + "epoch": 0.13784005276939376, + "grad_norm": 0.8411848545074463, + "learning_rate": 0.0005820021686750542, + "loss": 5.1415, + "step": 431 + }, + { + "epoch": 0.13815986727697935, + "grad_norm": 0.5718550086021423, + "learning_rate": 0.0005818959716150306, + "loss": 5.1265, + "step": 432 + }, + { + "epoch": 0.13847968178456496, + "grad_norm": 0.7160180807113647, + "learning_rate": 0.0005817894719123214, + "loss": 5.198, + "step": 433 + }, + { + "epoch": 0.13879949629215055, + "grad_norm": 1.3318077325820923, + "learning_rate": 0.0005816826696812643, + "loss": 5.1493, + "step": 434 + }, + { + "epoch": 0.13911931079973616, + "grad_norm": 0.613623321056366, + "learning_rate": 0.0005815755650365217, + "loss": 5.1647, + "step": 435 + }, + { + "epoch": 0.13943912530732175, + "grad_norm": 0.6599540114402771, + "learning_rate": 0.000581468158093081, + "loss": 5.2035, + "step": 436 + }, + { + "epoch": 0.13975893981490736, + "grad_norm": 0.7448862195014954, + "learning_rate": 0.0005813604489662539, + "loss": 5.2438, + "step": 437 + }, + { + "epoch": 0.14007875432249295, + "grad_norm": 1.0715965032577515, + "learning_rate": 0.0005812524377716766, + "loss": 5.2198, + "step": 438 + }, + { + "epoch": 0.14039856883007856, + "grad_norm": 0.8526812195777893, + "learning_rate": 0.0005811441246253098, + "loss": 5.1601, + "step": 439 + }, + { + "epoch": 0.14071838333766415, + "grad_norm": 0.5853661894798279, + "learning_rate": 0.0005810355096434378, + "loss": 5.1762, + "step": 440 + }, + { + "epoch": 0.14103819784524976, + "grad_norm": 0.6756531596183777, + "learning_rate": 0.0005809265929426696, + "loss": 5.1487, + "step": 441 + }, + { + "epoch": 0.14135801235283535, + "grad_norm": 0.7295665144920349, + "learning_rate": 0.0005808173746399377, + "loss": 5.1239, + "step": 442 + }, + { + "epoch": 0.14167782686042096, + "grad_norm": 0.8424263000488281, + "learning_rate": 0.0005807078548524988, + "loss": 5.0663, + "step": 443 + }, + { + "epoch": 0.14199764136800655, + "grad_norm": 0.6921552419662476, + "learning_rate": 0.0005805980336979327, + "loss": 5.2635, + "step": 444 + }, + { + "epoch": 0.14231745587559216, + "grad_norm": 0.7230246663093567, + "learning_rate": 0.0005804879112941433, + "loss": 5.0964, + "step": 445 + }, + { + "epoch": 0.14263727038317775, + "grad_norm": 2.098560333251953, + "learning_rate": 0.0005803774877593575, + "loss": 5.1766, + "step": 446 + }, + { + "epoch": 0.14295708489076336, + "grad_norm": 0.6903122663497925, + "learning_rate": 0.000580266763212126, + "loss": 5.1974, + "step": 447 + }, + { + "epoch": 0.14327689939834895, + "grad_norm": 0.8592844605445862, + "learning_rate": 0.0005801557377713218, + "loss": 5.2174, + "step": 448 + }, + { + "epoch": 0.14359671390593456, + "grad_norm": 0.5316879153251648, + "learning_rate": 0.0005800444115561422, + "loss": 5.2232, + "step": 449 + }, + { + "epoch": 0.14391652841352015, + "grad_norm": 2.65645170211792, + "learning_rate": 0.000579932784686106, + "loss": 5.1639, + "step": 450 + }, + { + "epoch": 0.14423634292110576, + "grad_norm": 0.9056392312049866, + "learning_rate": 0.000579820857281056, + "loss": 5.1587, + "step": 451 + }, + { + "epoch": 0.14455615742869135, + "grad_norm": 0.5324886441230774, + "learning_rate": 0.0005797086294611569, + "loss": 5.1513, + "step": 452 + }, + { + "epoch": 0.14487597193627696, + "grad_norm": 0.7657493352890015, + "learning_rate": 0.0005795961013468961, + "loss": 5.1705, + "step": 453 + }, + { + "epoch": 0.14519578644386255, + "grad_norm": 0.8256158828735352, + "learning_rate": 0.0005794832730590836, + "loss": 5.142, + "step": 454 + }, + { + "epoch": 0.14551560095144817, + "grad_norm": 0.7251825928688049, + "learning_rate": 0.0005793701447188514, + "loss": 5.0919, + "step": 455 + }, + { + "epoch": 0.14583541545903375, + "grad_norm": 1.3375217914581299, + "learning_rate": 0.0005792567164476539, + "loss": 5.1279, + "step": 456 + }, + { + "epoch": 0.14615522996661937, + "grad_norm": 0.955669105052948, + "learning_rate": 0.0005791429883672672, + "loss": 5.1072, + "step": 457 + }, + { + "epoch": 0.14647504447420495, + "grad_norm": 0.8705189824104309, + "learning_rate": 0.0005790289605997895, + "loss": 5.0961, + "step": 458 + }, + { + "epoch": 0.14679485898179057, + "grad_norm": 0.8156420588493347, + "learning_rate": 0.0005789146332676407, + "loss": 5.1038, + "step": 459 + }, + { + "epoch": 0.14711467348937615, + "grad_norm": 1.357147216796875, + "learning_rate": 0.0005788000064935623, + "loss": 5.121, + "step": 460 + }, + { + "epoch": 0.14743448799696177, + "grad_norm": 0.9629470705986023, + "learning_rate": 0.0005786850804006172, + "loss": 5.128, + "step": 461 + }, + { + "epoch": 0.14775430250454735, + "grad_norm": 0.6666399836540222, + "learning_rate": 0.0005785698551121897, + "loss": 5.1226, + "step": 462 + }, + { + "epoch": 0.14807411701213297, + "grad_norm": 1.41956627368927, + "learning_rate": 0.0005784543307519854, + "loss": 5.1058, + "step": 463 + }, + { + "epoch": 0.14839393151971855, + "grad_norm": 1.3943054676055908, + "learning_rate": 0.000578338507444031, + "loss": 5.1805, + "step": 464 + }, + { + "epoch": 0.14871374602730417, + "grad_norm": 0.7356551885604858, + "learning_rate": 0.0005782223853126739, + "loss": 5.1525, + "step": 465 + }, + { + "epoch": 0.14903356053488975, + "grad_norm": 0.865136444568634, + "learning_rate": 0.0005781059644825824, + "loss": 5.1533, + "step": 466 + }, + { + "epoch": 0.14935337504247537, + "grad_norm": 2.0989904403686523, + "learning_rate": 0.0005779892450787458, + "loss": 5.0514, + "step": 467 + }, + { + "epoch": 0.14967318955006095, + "grad_norm": 0.6888847351074219, + "learning_rate": 0.0005778722272264736, + "loss": 5.1824, + "step": 468 + }, + { + "epoch": 0.14999300405764657, + "grad_norm": 0.6499800682067871, + "learning_rate": 0.0005777549110513959, + "loss": 5.1716, + "step": 469 + }, + { + "epoch": 0.15031281856523215, + "grad_norm": 0.9319367408752441, + "learning_rate": 0.0005776372966794628, + "loss": 5.1806, + "step": 470 + }, + { + "epoch": 0.15063263307281777, + "grad_norm": 1.8618899583816528, + "learning_rate": 0.000577519384236945, + "loss": 5.1982, + "step": 471 + }, + { + "epoch": 0.15095244758040335, + "grad_norm": 1.189143180847168, + "learning_rate": 0.0005774011738504326, + "loss": 5.1916, + "step": 472 + }, + { + "epoch": 0.15127226208798897, + "grad_norm": 2.3902299404144287, + "learning_rate": 0.0005772826656468363, + "loss": 5.184, + "step": 473 + }, + { + "epoch": 0.15159207659557455, + "grad_norm": 1.8020542860031128, + "learning_rate": 0.000577163859753386, + "loss": 5.1572, + "step": 474 + }, + { + "epoch": 0.15191189110316017, + "grad_norm": 1.4632331132888794, + "learning_rate": 0.0005770447562976313, + "loss": 5.185, + "step": 475 + }, + { + "epoch": 0.15223170561074575, + "grad_norm": 1.1336344480514526, + "learning_rate": 0.0005769253554074414, + "loss": 5.1749, + "step": 476 + }, + { + "epoch": 0.15255152011833137, + "grad_norm": 0.8037937879562378, + "learning_rate": 0.0005768056572110047, + "loss": 5.1932, + "step": 477 + }, + { + "epoch": 0.15287133462591695, + "grad_norm": 1.3489552736282349, + "learning_rate": 0.000576685661836829, + "loss": 5.1355, + "step": 478 + }, + { + "epoch": 0.15319114913350257, + "grad_norm": 0.7866030931472778, + "learning_rate": 0.0005765653694137406, + "loss": 5.1238, + "step": 479 + }, + { + "epoch": 0.15351096364108816, + "grad_norm": 1.1267096996307373, + "learning_rate": 0.0005764447800708856, + "loss": 5.0929, + "step": 480 + }, + { + "epoch": 0.15383077814867377, + "grad_norm": 0.7161633968353271, + "learning_rate": 0.0005763238939377278, + "loss": 5.1395, + "step": 481 + }, + { + "epoch": 0.15415059265625938, + "grad_norm": 1.1200228929519653, + "learning_rate": 0.0005762027111440506, + "loss": 5.1155, + "step": 482 + }, + { + "epoch": 0.15447040716384497, + "grad_norm": 0.6353023052215576, + "learning_rate": 0.0005760812318199555, + "loss": 5.1221, + "step": 483 + }, + { + "epoch": 0.15479022167143058, + "grad_norm": 0.7894587516784668, + "learning_rate": 0.000575959456095862, + "loss": 5.0294, + "step": 484 + }, + { + "epoch": 0.15511003617901617, + "grad_norm": 0.7961439490318298, + "learning_rate": 0.0005758373841025085, + "loss": 5.0782, + "step": 485 + }, + { + "epoch": 0.15542985068660178, + "grad_norm": 0.8050485253334045, + "learning_rate": 0.000575715015970951, + "loss": 5.0693, + "step": 486 + }, + { + "epoch": 0.15574966519418737, + "grad_norm": 0.6696105599403381, + "learning_rate": 0.0005755923518325637, + "loss": 5.0867, + "step": 487 + }, + { + "epoch": 0.15606947970177298, + "grad_norm": 0.9413130283355713, + "learning_rate": 0.0005754693918190382, + "loss": 5.1213, + "step": 488 + }, + { + "epoch": 0.15638929420935857, + "grad_norm": 3.0895962715148926, + "learning_rate": 0.0005753461360623842, + "loss": 5.0747, + "step": 489 + }, + { + "epoch": 0.15670910871694418, + "grad_norm": 1.2921384572982788, + "learning_rate": 0.0005752225846949287, + "loss": 5.1617, + "step": 490 + }, + { + "epoch": 0.15702892322452977, + "grad_norm": 1.0387719869613647, + "learning_rate": 0.000575098737849316, + "loss": 5.1026, + "step": 491 + }, + { + "epoch": 0.15734873773211538, + "grad_norm": 2.971237897872925, + "learning_rate": 0.0005749745956585077, + "loss": 5.0975, + "step": 492 + }, + { + "epoch": 0.15766855223970097, + "grad_norm": 0.6466496586799622, + "learning_rate": 0.0005748501582557825, + "loss": 5.1701, + "step": 493 + }, + { + "epoch": 0.15798836674728658, + "grad_norm": 1.539628028869629, + "learning_rate": 0.0005747254257747362, + "loss": 5.1236, + "step": 494 + }, + { + "epoch": 0.15830818125487217, + "grad_norm": 1.0888315439224243, + "learning_rate": 0.0005746003983492811, + "loss": 5.0653, + "step": 495 + }, + { + "epoch": 0.15862799576245779, + "grad_norm": 0.7691898941993713, + "learning_rate": 0.0005744750761136463, + "loss": 5.0547, + "step": 496 + }, + { + "epoch": 0.15894781027004337, + "grad_norm": 1.52078378200531, + "learning_rate": 0.0005743494592023773, + "loss": 5.0785, + "step": 497 + }, + { + "epoch": 0.15926762477762899, + "grad_norm": 0.7059910297393799, + "learning_rate": 0.0005742235477503362, + "loss": 5.0719, + "step": 498 + }, + { + "epoch": 0.15958743928521457, + "grad_norm": 2.7390408515930176, + "learning_rate": 0.000574097341892701, + "loss": 5.0393, + "step": 499 + }, + { + "epoch": 0.15990725379280019, + "grad_norm": 1.0773636102676392, + "learning_rate": 0.0005739708417649659, + "loss": 5.0203, + "step": 500 + }, + { + "epoch": 0.15990725379280019, + "eval_runtime": 23.3522, + "eval_samples_per_second": 81.234, + "eval_steps_per_second": 20.341, + "step": 500 + }, + { + "epoch": 0.16022706830038577, + "grad_norm": 1.03023362159729, + "learning_rate": 0.0005738440475029414, + "loss": 5.1076, + "step": 501 + }, + { + "epoch": 0.16054688280797139, + "grad_norm": 0.901528537273407, + "learning_rate": 0.0005737169592427531, + "loss": 5.1338, + "step": 502 + }, + { + "epoch": 0.16086669731555697, + "grad_norm": 0.9528647065162659, + "learning_rate": 0.0005735895771208427, + "loss": 5.1284, + "step": 503 + }, + { + "epoch": 0.1611865118231426, + "grad_norm": 0.6932551264762878, + "learning_rate": 0.0005734619012739673, + "loss": 5.1317, + "step": 504 + }, + { + "epoch": 0.16150632633072817, + "grad_norm": 0.7791666388511658, + "learning_rate": 0.0005733339318391992, + "loss": 5.1592, + "step": 505 + }, + { + "epoch": 0.1618261408383138, + "grad_norm": 0.8244270086288452, + "learning_rate": 0.0005732056689539262, + "loss": 5.0494, + "step": 506 + }, + { + "epoch": 0.16214595534589937, + "grad_norm": 0.9310250282287598, + "learning_rate": 0.0005730771127558508, + "loss": 5.0739, + "step": 507 + }, + { + "epoch": 0.162465769853485, + "grad_norm": 1.1068477630615234, + "learning_rate": 0.0005729482633829906, + "loss": 5.0765, + "step": 508 + }, + { + "epoch": 0.16278558436107057, + "grad_norm": 0.699282169342041, + "learning_rate": 0.000572819120973678, + "loss": 5.0671, + "step": 509 + }, + { + "epoch": 0.1631053988686562, + "grad_norm": 2.1142284870147705, + "learning_rate": 0.0005726896856665599, + "loss": 5.0517, + "step": 510 + }, + { + "epoch": 0.16342521337624177, + "grad_norm": 0.6684252619743347, + "learning_rate": 0.0005725599576005975, + "loss": 5.0819, + "step": 511 + }, + { + "epoch": 0.1637450278838274, + "grad_norm": 1.1115903854370117, + "learning_rate": 0.0005724299369150665, + "loss": 5.1058, + "step": 512 + }, + { + "epoch": 0.16406484239141297, + "grad_norm": 0.8507381677627563, + "learning_rate": 0.0005722996237495569, + "loss": 5.123, + "step": 513 + }, + { + "epoch": 0.1643846568989986, + "grad_norm": 1.5931049585342407, + "learning_rate": 0.0005721690182439724, + "loss": 5.1304, + "step": 514 + }, + { + "epoch": 0.16470447140658417, + "grad_norm": 0.9648686647415161, + "learning_rate": 0.0005720381205385306, + "loss": 5.0735, + "step": 515 + }, + { + "epoch": 0.1650242859141698, + "grad_norm": 1.018010139465332, + "learning_rate": 0.000571906930773763, + "loss": 5.0704, + "step": 516 + }, + { + "epoch": 0.16534410042175537, + "grad_norm": 0.7228468656539917, + "learning_rate": 0.0005717754490905146, + "loss": 5.0984, + "step": 517 + }, + { + "epoch": 0.165663914929341, + "grad_norm": 0.8981325626373291, + "learning_rate": 0.0005716436756299437, + "loss": 5.0386, + "step": 518 + }, + { + "epoch": 0.16598372943692657, + "grad_norm": 2.6513657569885254, + "learning_rate": 0.000571511610533522, + "loss": 5.0908, + "step": 519 + }, + { + "epoch": 0.1663035439445122, + "grad_norm": 0.7023225426673889, + "learning_rate": 0.0005713792539430339, + "loss": 5.0916, + "step": 520 + }, + { + "epoch": 0.16662335845209778, + "grad_norm": 0.783008337020874, + "learning_rate": 0.0005712466060005774, + "loss": 5.0696, + "step": 521 + }, + { + "epoch": 0.1669431729596834, + "grad_norm": 6.10725736618042, + "learning_rate": 0.0005711136668485626, + "loss": 5.0087, + "step": 522 + }, + { + "epoch": 0.16726298746726898, + "grad_norm": 0.7551060914993286, + "learning_rate": 0.0005709804366297129, + "loss": 5.0455, + "step": 523 + }, + { + "epoch": 0.1675828019748546, + "grad_norm": 1.510067343711853, + "learning_rate": 0.0005708469154870636, + "loss": 5.0981, + "step": 524 + }, + { + "epoch": 0.16790261648244018, + "grad_norm": 1.3365272283554077, + "learning_rate": 0.0005707131035639629, + "loss": 5.0775, + "step": 525 + }, + { + "epoch": 0.1682224309900258, + "grad_norm": 0.7854240536689758, + "learning_rate": 0.0005705790010040707, + "loss": 5.0989, + "step": 526 + }, + { + "epoch": 0.16854224549761138, + "grad_norm": 1.0058971643447876, + "learning_rate": 0.000570444607951359, + "loss": 5.1038, + "step": 527 + }, + { + "epoch": 0.168862060005197, + "grad_norm": 5.390463829040527, + "learning_rate": 0.000570309924550112, + "loss": 5.0268, + "step": 528 + }, + { + "epoch": 0.16918187451278258, + "grad_norm": 0.8939847946166992, + "learning_rate": 0.0005701749509449253, + "loss": 5.0141, + "step": 529 + }, + { + "epoch": 0.1695016890203682, + "grad_norm": 1.540946364402771, + "learning_rate": 0.0005700396872807062, + "loss": 5.0335, + "step": 530 + }, + { + "epoch": 0.16982150352795378, + "grad_norm": 1.3480817079544067, + "learning_rate": 0.0005699041337026734, + "loss": 5.0423, + "step": 531 + }, + { + "epoch": 0.1701413180355394, + "grad_norm": 1.26405930519104, + "learning_rate": 0.0005697682903563568, + "loss": 5.0585, + "step": 532 + }, + { + "epoch": 0.17046113254312498, + "grad_norm": 1.3934255838394165, + "learning_rate": 0.0005696321573875974, + "loss": 5.0539, + "step": 533 + }, + { + "epoch": 0.1707809470507106, + "grad_norm": 1.592209815979004, + "learning_rate": 0.0005694957349425472, + "loss": 5.0526, + "step": 534 + }, + { + "epoch": 0.17110076155829618, + "grad_norm": 1.7538596391677856, + "learning_rate": 0.0005693590231676688, + "loss": 5.0624, + "step": 535 + }, + { + "epoch": 0.1714205760658818, + "grad_norm": 1.1525593996047974, + "learning_rate": 0.0005692220222097357, + "loss": 5.0748, + "step": 536 + }, + { + "epoch": 0.17174039057346738, + "grad_norm": 3.496995687484741, + "learning_rate": 0.0005690847322158317, + "loss": 5.0758, + "step": 537 + }, + { + "epoch": 0.172060205081053, + "grad_norm": 0.8620216250419617, + "learning_rate": 0.0005689471533333508, + "loss": 5.0043, + "step": 538 + }, + { + "epoch": 0.17238001958863858, + "grad_norm": 1.7570174932479858, + "learning_rate": 0.0005688092857099974, + "loss": 4.9572, + "step": 539 + }, + { + "epoch": 0.1726998340962242, + "grad_norm": 0.8333970308303833, + "learning_rate": 0.0005686711294937858, + "loss": 4.9495, + "step": 540 + }, + { + "epoch": 0.17301964860380978, + "grad_norm": 2.8553056716918945, + "learning_rate": 0.0005685326848330402, + "loss": 5.0655, + "step": 541 + }, + { + "epoch": 0.1733394631113954, + "grad_norm": 1.2100130319595337, + "learning_rate": 0.0005683939518763942, + "loss": 5.0077, + "step": 542 + }, + { + "epoch": 0.17365927761898098, + "grad_norm": 3.4622247219085693, + "learning_rate": 0.000568254930772791, + "loss": 5.0815, + "step": 543 + }, + { + "epoch": 0.1739790921265666, + "grad_norm": 0.994434654712677, + "learning_rate": 0.0005681156216714836, + "loss": 5.0311, + "step": 544 + }, + { + "epoch": 0.17429890663415218, + "grad_norm": 1.6225733757019043, + "learning_rate": 0.0005679760247220336, + "loss": 5.032, + "step": 545 + }, + { + "epoch": 0.1746187211417378, + "grad_norm": 1.0976136922836304, + "learning_rate": 0.0005678361400743119, + "loss": 5.0604, + "step": 546 + }, + { + "epoch": 0.17493853564932338, + "grad_norm": 1.1218156814575195, + "learning_rate": 0.0005676959678784982, + "loss": 4.9372, + "step": 547 + }, + { + "epoch": 0.175258350156909, + "grad_norm": 0.8958498239517212, + "learning_rate": 0.000567555508285081, + "loss": 5.0545, + "step": 548 + }, + { + "epoch": 0.17557816466449458, + "grad_norm": 1.1382813453674316, + "learning_rate": 0.0005674147614448574, + "loss": 5.0874, + "step": 549 + }, + { + "epoch": 0.1758979791720802, + "grad_norm": 1.1039139032363892, + "learning_rate": 0.0005672737275089327, + "loss": 5.0855, + "step": 550 + }, + { + "epoch": 0.1762177936796658, + "grad_norm": 0.8152859210968018, + "learning_rate": 0.0005671324066287205, + "loss": 4.9976, + "step": 551 + }, + { + "epoch": 0.1765376081872514, + "grad_norm": 1.2729705572128296, + "learning_rate": 0.0005669907989559426, + "loss": 5.055, + "step": 552 + }, + { + "epoch": 0.176857422694837, + "grad_norm": 17.21373176574707, + "learning_rate": 0.0005668489046426285, + "loss": 5.0797, + "step": 553 + }, + { + "epoch": 0.1771772372024226, + "grad_norm": 0.8813936710357666, + "learning_rate": 0.0005667067238411153, + "loss": 5.0611, + "step": 554 + }, + { + "epoch": 0.1774970517100082, + "grad_norm": 1.3374388217926025, + "learning_rate": 0.0005665642567040483, + "loss": 5.0765, + "step": 555 + }, + { + "epoch": 0.1778168662175938, + "grad_norm": 0.9933357238769531, + "learning_rate": 0.0005664215033843796, + "loss": 5.0047, + "step": 556 + }, + { + "epoch": 0.1781366807251794, + "grad_norm": 1.6311304569244385, + "learning_rate": 0.0005662784640353688, + "loss": 5.0058, + "step": 557 + }, + { + "epoch": 0.178456495232765, + "grad_norm": 1.1833336353302002, + "learning_rate": 0.0005661351388105823, + "loss": 5.0812, + "step": 558 + }, + { + "epoch": 0.1787763097403506, + "grad_norm": 0.8471614718437195, + "learning_rate": 0.0005659915278638939, + "loss": 5.0713, + "step": 559 + }, + { + "epoch": 0.1790961242479362, + "grad_norm": 1.0327038764953613, + "learning_rate": 0.0005658476313494839, + "loss": 4.9695, + "step": 560 + }, + { + "epoch": 0.1794159387555218, + "grad_norm": 2.4016010761260986, + "learning_rate": 0.0005657034494218389, + "loss": 5.0221, + "step": 561 + }, + { + "epoch": 0.1797357532631074, + "grad_norm": 1.574601173400879, + "learning_rate": 0.0005655589822357526, + "loss": 5.0089, + "step": 562 + }, + { + "epoch": 0.180055567770693, + "grad_norm": 1.0182067155838013, + "learning_rate": 0.0005654142299463241, + "loss": 5.0776, + "step": 563 + }, + { + "epoch": 0.1803753822782786, + "grad_norm": 0.9470081925392151, + "learning_rate": 0.0005652691927089593, + "loss": 4.9974, + "step": 564 + }, + { + "epoch": 0.1806951967858642, + "grad_norm": 2.8176822662353516, + "learning_rate": 0.0005651238706793697, + "loss": 4.9552, + "step": 565 + }, + { + "epoch": 0.1810150112934498, + "grad_norm": 1.0728617906570435, + "learning_rate": 0.0005649782640135727, + "loss": 5.0019, + "step": 566 + }, + { + "epoch": 0.1813348258010354, + "grad_norm": 1.1144269704818726, + "learning_rate": 0.000564832372867891, + "loss": 4.952, + "step": 567 + }, + { + "epoch": 0.181654640308621, + "grad_norm": 1.2841871976852417, + "learning_rate": 0.0005646861973989531, + "loss": 5.064, + "step": 568 + }, + { + "epoch": 0.1819744548162066, + "grad_norm": 1.861783742904663, + "learning_rate": 0.0005645397377636922, + "loss": 4.9665, + "step": 569 + }, + { + "epoch": 0.1822942693237922, + "grad_norm": 0.836138129234314, + "learning_rate": 0.0005643929941193474, + "loss": 4.957, + "step": 570 + }, + { + "epoch": 0.1826140838313778, + "grad_norm": 0.7673237919807434, + "learning_rate": 0.000564245966623462, + "loss": 4.964, + "step": 571 + }, + { + "epoch": 0.1829338983389634, + "grad_norm": 1.1661232709884644, + "learning_rate": 0.0005640986554338842, + "loss": 4.9578, + "step": 572 + }, + { + "epoch": 0.183253712846549, + "grad_norm": 0.9701364636421204, + "learning_rate": 0.0005639510607087673, + "loss": 5.0989, + "step": 573 + }, + { + "epoch": 0.1835735273541346, + "grad_norm": 1.9132835865020752, + "learning_rate": 0.0005638031826065679, + "loss": 4.9974, + "step": 574 + }, + { + "epoch": 0.1838933418617202, + "grad_norm": 0.8168975114822388, + "learning_rate": 0.0005636550212860479, + "loss": 5.0153, + "step": 575 + }, + { + "epoch": 0.1842131563693058, + "grad_norm": 0.8442965745925903, + "learning_rate": 0.0005635065769062728, + "loss": 5.0177, + "step": 576 + }, + { + "epoch": 0.1845329708768914, + "grad_norm": 1.1235432624816895, + "learning_rate": 0.0005633578496266121, + "loss": 4.9645, + "step": 577 + }, + { + "epoch": 0.184852785384477, + "grad_norm": 1.4992458820343018, + "learning_rate": 0.0005632088396067389, + "loss": 4.9862, + "step": 578 + }, + { + "epoch": 0.1851725998920626, + "grad_norm": 1.7065036296844482, + "learning_rate": 0.0005630595470066299, + "loss": 5.0417, + "step": 579 + }, + { + "epoch": 0.1854924143996482, + "grad_norm": 1.4709714651107788, + "learning_rate": 0.0005629099719865652, + "loss": 5.0758, + "step": 580 + }, + { + "epoch": 0.1858122289072338, + "grad_norm": 2.2433032989501953, + "learning_rate": 0.0005627601147071282, + "loss": 4.9549, + "step": 581 + }, + { + "epoch": 0.1861320434148194, + "grad_norm": 1.9223910570144653, + "learning_rate": 0.000562609975329205, + "loss": 4.9459, + "step": 582 + }, + { + "epoch": 0.186451857922405, + "grad_norm": 0.8473854064941406, + "learning_rate": 0.0005624595540139851, + "loss": 5.031, + "step": 583 + }, + { + "epoch": 0.1867716724299906, + "grad_norm": 1.0762763023376465, + "learning_rate": 0.0005623088509229602, + "loss": 4.9234, + "step": 584 + }, + { + "epoch": 0.1870914869375762, + "grad_norm": 1.3030319213867188, + "learning_rate": 0.0005621578662179247, + "loss": 4.9436, + "step": 585 + }, + { + "epoch": 0.1874113014451618, + "grad_norm": 0.9178854823112488, + "learning_rate": 0.0005620066000609755, + "loss": 5.0147, + "step": 586 + }, + { + "epoch": 0.1877311159527474, + "grad_norm": 0.9894868731498718, + "learning_rate": 0.0005618550526145113, + "loss": 4.9777, + "step": 587 + }, + { + "epoch": 0.188050930460333, + "grad_norm": 1.1683672666549683, + "learning_rate": 0.0005617032240412329, + "loss": 4.9263, + "step": 588 + }, + { + "epoch": 0.1883707449679186, + "grad_norm": 2.9599194526672363, + "learning_rate": 0.0005615511145041433, + "loss": 4.9893, + "step": 589 + }, + { + "epoch": 0.1886905594755042, + "grad_norm": 1.5975987911224365, + "learning_rate": 0.0005613987241665468, + "loss": 5.0248, + "step": 590 + }, + { + "epoch": 0.1890103739830898, + "grad_norm": 0.806911826133728, + "learning_rate": 0.000561246053192049, + "loss": 4.9316, + "step": 591 + }, + { + "epoch": 0.1893301884906754, + "grad_norm": 1.8664882183074951, + "learning_rate": 0.0005610931017445573, + "loss": 4.9967, + "step": 592 + }, + { + "epoch": 0.189650002998261, + "grad_norm": 1.8052270412445068, + "learning_rate": 0.0005609398699882796, + "loss": 4.9838, + "step": 593 + }, + { + "epoch": 0.1899698175058466, + "grad_norm": 1.2483378648757935, + "learning_rate": 0.0005607863580877253, + "loss": 5.0202, + "step": 594 + }, + { + "epoch": 0.19028963201343221, + "grad_norm": 3.886770009994507, + "learning_rate": 0.0005606325662077042, + "loss": 4.9587, + "step": 595 + }, + { + "epoch": 0.1906094465210178, + "grad_norm": 1.298531174659729, + "learning_rate": 0.0005604784945133271, + "loss": 4.9342, + "step": 596 + }, + { + "epoch": 0.19092926102860341, + "grad_norm": 1.7648839950561523, + "learning_rate": 0.0005603241431700045, + "loss": 5.0132, + "step": 597 + }, + { + "epoch": 0.191249075536189, + "grad_norm": 4.505471706390381, + "learning_rate": 0.0005601695123434477, + "loss": 4.91, + "step": 598 + }, + { + "epoch": 0.19156889004377461, + "grad_norm": 3.2875914573669434, + "learning_rate": 0.000560014602199668, + "loss": 4.9889, + "step": 599 + }, + { + "epoch": 0.1918887045513602, + "grad_norm": 2.551464319229126, + "learning_rate": 0.0005598594129049765, + "loss": 5.0991, + "step": 600 + }, + { + "epoch": 0.1918887045513602, + "eval_runtime": 22.1405, + "eval_samples_per_second": 85.68, + "eval_steps_per_second": 21.454, + "step": 600 + }, + { + "epoch": 0.19220851905894581, + "grad_norm": 2.0192019939422607, + "learning_rate": 0.0005597039446259837, + "loss": 5.1104, + "step": 601 + }, + { + "epoch": 0.1925283335665314, + "grad_norm": 10.809158325195312, + "learning_rate": 0.0005595481975296002, + "loss": 5.0915, + "step": 602 + }, + { + "epoch": 0.19284814807411702, + "grad_norm": 4.00751256942749, + "learning_rate": 0.0005593921717830354, + "loss": 4.9808, + "step": 603 + }, + { + "epoch": 0.1931679625817026, + "grad_norm": 1.6277306079864502, + "learning_rate": 0.0005592358675537983, + "loss": 5.0477, + "step": 604 + }, + { + "epoch": 0.19348777708928822, + "grad_norm": 1.235906958580017, + "learning_rate": 0.0005590792850096965, + "loss": 4.9598, + "step": 605 + }, + { + "epoch": 0.1938075915968738, + "grad_norm": 1.166163682937622, + "learning_rate": 0.0005589224243188365, + "loss": 5.0153, + "step": 606 + }, + { + "epoch": 0.19412740610445942, + "grad_norm": 1.1783647537231445, + "learning_rate": 0.0005587652856496236, + "loss": 4.9265, + "step": 607 + }, + { + "epoch": 0.194447220612045, + "grad_norm": 1.0235730409622192, + "learning_rate": 0.0005586078691707614, + "loss": 4.9112, + "step": 608 + }, + { + "epoch": 0.19476703511963062, + "grad_norm": 1.167936086654663, + "learning_rate": 0.0005584501750512516, + "loss": 5.0016, + "step": 609 + }, + { + "epoch": 0.1950868496272162, + "grad_norm": 0.9465678334236145, + "learning_rate": 0.0005582922034603945, + "loss": 4.954, + "step": 610 + }, + { + "epoch": 0.19540666413480182, + "grad_norm": 1.9766476154327393, + "learning_rate": 0.0005581339545677877, + "loss": 4.9902, + "step": 611 + }, + { + "epoch": 0.1957264786423874, + "grad_norm": 1.0055640935897827, + "learning_rate": 0.0005579754285433269, + "loss": 4.926, + "step": 612 + }, + { + "epoch": 0.19604629314997302, + "grad_norm": 2.2566444873809814, + "learning_rate": 0.0005578166255572048, + "loss": 5.0093, + "step": 613 + }, + { + "epoch": 0.1963661076575586, + "grad_norm": 2.322197198867798, + "learning_rate": 0.0005576575457799122, + "loss": 4.9589, + "step": 614 + }, + { + "epoch": 0.19668592216514422, + "grad_norm": 1.9794822931289673, + "learning_rate": 0.0005574981893822365, + "loss": 4.9489, + "step": 615 + }, + { + "epoch": 0.1970057366727298, + "grad_norm": 1.240434169769287, + "learning_rate": 0.0005573385565352622, + "loss": 4.9461, + "step": 616 + }, + { + "epoch": 0.19732555118031542, + "grad_norm": 1.7943123579025269, + "learning_rate": 0.0005571786474103709, + "loss": 5.0365, + "step": 617 + }, + { + "epoch": 0.19764536568790103, + "grad_norm": 1.8845024108886719, + "learning_rate": 0.0005570184621792405, + "loss": 4.8807, + "step": 618 + }, + { + "epoch": 0.19796518019548662, + "grad_norm": 1.3060119152069092, + "learning_rate": 0.0005568580010138452, + "loss": 4.9825, + "step": 619 + }, + { + "epoch": 0.19828499470307223, + "grad_norm": 1.2993851900100708, + "learning_rate": 0.0005566972640864558, + "loss": 4.9435, + "step": 620 + }, + { + "epoch": 0.19860480921065782, + "grad_norm": 0.948917031288147, + "learning_rate": 0.0005565362515696389, + "loss": 4.9668, + "step": 621 + }, + { + "epoch": 0.19892462371824343, + "grad_norm": 1.9834548234939575, + "learning_rate": 0.0005563749636362572, + "loss": 4.8959, + "step": 622 + }, + { + "epoch": 0.19924443822582902, + "grad_norm": 1.93113374710083, + "learning_rate": 0.0005562134004594687, + "loss": 5.0353, + "step": 623 + }, + { + "epoch": 0.19956425273341463, + "grad_norm": 1.545680284500122, + "learning_rate": 0.0005560515622127276, + "loss": 4.9146, + "step": 624 + }, + { + "epoch": 0.19988406724100022, + "grad_norm": 7.140727996826172, + "learning_rate": 0.0005558894490697824, + "loss": 4.9108, + "step": 625 + }, + { + "epoch": 0.20020388174858583, + "grad_norm": 3.4650886058807373, + "learning_rate": 0.0005557270612046777, + "loss": 4.9677, + "step": 626 + }, + { + "epoch": 0.20052369625617142, + "grad_norm": 1.133226990699768, + "learning_rate": 0.0005555643987917525, + "loss": 4.9308, + "step": 627 + }, + { + "epoch": 0.20084351076375703, + "grad_norm": 1.049751877784729, + "learning_rate": 0.0005554014620056406, + "loss": 4.9232, + "step": 628 + }, + { + "epoch": 0.20116332527134262, + "grad_norm": 2.2143635749816895, + "learning_rate": 0.0005552382510212706, + "loss": 4.9026, + "step": 629 + }, + { + "epoch": 0.20148313977892823, + "grad_norm": 1.0087807178497314, + "learning_rate": 0.0005550747660138653, + "loss": 4.9684, + "step": 630 + }, + { + "epoch": 0.20180295428651382, + "grad_norm": 1.052030086517334, + "learning_rate": 0.0005549110071589418, + "loss": 4.9301, + "step": 631 + }, + { + "epoch": 0.20212276879409943, + "grad_norm": 5.905435085296631, + "learning_rate": 0.0005547469746323109, + "loss": 4.9165, + "step": 632 + }, + { + "epoch": 0.20244258330168502, + "grad_norm": 1.152402639389038, + "learning_rate": 0.0005545826686100776, + "loss": 5.022, + "step": 633 + }, + { + "epoch": 0.20276239780927063, + "grad_norm": 1.8156484365463257, + "learning_rate": 0.0005544180892686403, + "loss": 4.8487, + "step": 634 + }, + { + "epoch": 0.20308221231685622, + "grad_norm": 2.3884832859039307, + "learning_rate": 0.000554253236784691, + "loss": 4.8924, + "step": 635 + }, + { + "epoch": 0.20340202682444183, + "grad_norm": 1.7557746171951294, + "learning_rate": 0.0005540881113352148, + "loss": 4.9559, + "step": 636 + }, + { + "epoch": 0.20372184133202742, + "grad_norm": 1.7006818056106567, + "learning_rate": 0.0005539227130974898, + "loss": 4.9636, + "step": 637 + }, + { + "epoch": 0.20404165583961303, + "grad_norm": 1.1384116411209106, + "learning_rate": 0.0005537570422490871, + "loss": 4.8832, + "step": 638 + }, + { + "epoch": 0.20436147034719862, + "grad_norm": 2.094517469406128, + "learning_rate": 0.0005535910989678706, + "loss": 4.9161, + "step": 639 + }, + { + "epoch": 0.20468128485478423, + "grad_norm": 0.8671731352806091, + "learning_rate": 0.0005534248834319962, + "loss": 4.9199, + "step": 640 + }, + { + "epoch": 0.20500109936236982, + "grad_norm": 4.315222263336182, + "learning_rate": 0.0005532583958199126, + "loss": 4.8911, + "step": 641 + }, + { + "epoch": 0.20532091386995543, + "grad_norm": 1.7062413692474365, + "learning_rate": 0.0005530916363103605, + "loss": 5.0162, + "step": 642 + }, + { + "epoch": 0.20564072837754102, + "grad_norm": 2.088627338409424, + "learning_rate": 0.0005529246050823723, + "loss": 4.9213, + "step": 643 + }, + { + "epoch": 0.20596054288512664, + "grad_norm": 1.0082868337631226, + "learning_rate": 0.0005527573023152722, + "loss": 4.9127, + "step": 644 + }, + { + "epoch": 0.20628035739271222, + "grad_norm": 1.5624102354049683, + "learning_rate": 0.0005525897281886761, + "loss": 4.8707, + "step": 645 + }, + { + "epoch": 0.20660017190029784, + "grad_norm": 7.336634159088135, + "learning_rate": 0.000552421882882491, + "loss": 4.9253, + "step": 646 + }, + { + "epoch": 0.20691998640788342, + "grad_norm": 1.5458581447601318, + "learning_rate": 0.000552253766576915, + "loss": 4.9221, + "step": 647 + }, + { + "epoch": 0.20723980091546904, + "grad_norm": 1.8274946212768555, + "learning_rate": 0.0005520853794524375, + "loss": 5.0246, + "step": 648 + }, + { + "epoch": 0.20755961542305462, + "grad_norm": 2.0848498344421387, + "learning_rate": 0.0005519167216898383, + "loss": 4.9821, + "step": 649 + }, + { + "epoch": 0.20787942993064024, + "grad_norm": 4.351363658905029, + "learning_rate": 0.0005517477934701879, + "loss": 4.8666, + "step": 650 + }, + { + "epoch": 0.20819924443822582, + "grad_norm": 1.857691764831543, + "learning_rate": 0.0005515785949748471, + "loss": 4.9579, + "step": 651 + }, + { + "epoch": 0.20851905894581144, + "grad_norm": 4.572240829467773, + "learning_rate": 0.0005514091263854671, + "loss": 4.8941, + "step": 652 + }, + { + "epoch": 0.20883887345339702, + "grad_norm": 2.392765522003174, + "learning_rate": 0.0005512393878839885, + "loss": 4.8779, + "step": 653 + }, + { + "epoch": 0.20915868796098264, + "grad_norm": 2.6113176345825195, + "learning_rate": 0.0005510693796526425, + "loss": 4.8369, + "step": 654 + }, + { + "epoch": 0.20947850246856822, + "grad_norm": 2.3355824947357178, + "learning_rate": 0.000550899101873949, + "loss": 5.001, + "step": 655 + }, + { + "epoch": 0.20979831697615384, + "grad_norm": 5.363628387451172, + "learning_rate": 0.0005507285547307181, + "loss": 4.9034, + "step": 656 + }, + { + "epoch": 0.21011813148373942, + "grad_norm": 1.4802601337432861, + "learning_rate": 0.0005505577384060485, + "loss": 4.9108, + "step": 657 + }, + { + "epoch": 0.21043794599132504, + "grad_norm": 2.5010576248168945, + "learning_rate": 0.0005503866530833281, + "loss": 4.9582, + "step": 658 + }, + { + "epoch": 0.21075776049891062, + "grad_norm": 2.0544583797454834, + "learning_rate": 0.0005502152989462337, + "loss": 4.9877, + "step": 659 + }, + { + "epoch": 0.21107757500649624, + "grad_norm": 1.4103690385818481, + "learning_rate": 0.0005500436761787306, + "loss": 4.8515, + "step": 660 + }, + { + "epoch": 0.21139738951408182, + "grad_norm": 1.046213150024414, + "learning_rate": 0.0005498717849650724, + "loss": 4.9468, + "step": 661 + }, + { + "epoch": 0.21171720402166744, + "grad_norm": 1.5419740676879883, + "learning_rate": 0.0005496996254898011, + "loss": 4.9278, + "step": 662 + }, + { + "epoch": 0.21203701852925302, + "grad_norm": 3.964923858642578, + "learning_rate": 0.0005495271979377464, + "loss": 4.8956, + "step": 663 + }, + { + "epoch": 0.21235683303683864, + "grad_norm": 1.126347541809082, + "learning_rate": 0.0005493545024940264, + "loss": 4.8877, + "step": 664 + }, + { + "epoch": 0.21267664754442422, + "grad_norm": 1.6360716819763184, + "learning_rate": 0.000549181539344046, + "loss": 4.8964, + "step": 665 + }, + { + "epoch": 0.21299646205200984, + "grad_norm": 2.0658321380615234, + "learning_rate": 0.0005490083086734982, + "loss": 4.9128, + "step": 666 + }, + { + "epoch": 0.21331627655959542, + "grad_norm": 1.8495466709136963, + "learning_rate": 0.000548834810668363, + "loss": 4.9678, + "step": 667 + }, + { + "epoch": 0.21363609106718104, + "grad_norm": 1.0233263969421387, + "learning_rate": 0.0005486610455149069, + "loss": 4.9579, + "step": 668 + }, + { + "epoch": 0.21395590557476662, + "grad_norm": 1.0872044563293457, + "learning_rate": 0.0005484870133996842, + "loss": 4.9317, + "step": 669 + }, + { + "epoch": 0.21427572008235224, + "grad_norm": 2.707123279571533, + "learning_rate": 0.0005483127145095349, + "loss": 4.8766, + "step": 670 + }, + { + "epoch": 0.21459553458993783, + "grad_norm": 1.394896388053894, + "learning_rate": 0.0005481381490315859, + "loss": 4.9976, + "step": 671 + }, + { + "epoch": 0.21491534909752344, + "grad_norm": 1.3538010120391846, + "learning_rate": 0.0005479633171532503, + "loss": 4.8397, + "step": 672 + }, + { + "epoch": 0.21523516360510903, + "grad_norm": 1.1525969505310059, + "learning_rate": 0.0005477882190622269, + "loss": 4.8763, + "step": 673 + }, + { + "epoch": 0.21555497811269464, + "grad_norm": 1.7986615896224976, + "learning_rate": 0.0005476128549465006, + "loss": 4.9002, + "step": 674 + }, + { + "epoch": 0.21587479262028023, + "grad_norm": 2.867746591567993, + "learning_rate": 0.0005474372249943417, + "loss": 4.9621, + "step": 675 + }, + { + "epoch": 0.21619460712786584, + "grad_norm": 0.9375014901161194, + "learning_rate": 0.0005472613293943062, + "loss": 4.9411, + "step": 676 + }, + { + "epoch": 0.21651442163545143, + "grad_norm": 9.238116264343262, + "learning_rate": 0.0005470851683352349, + "loss": 4.8946, + "step": 677 + }, + { + "epoch": 0.21683423614303704, + "grad_norm": 1.578727126121521, + "learning_rate": 0.0005469087420062538, + "loss": 4.9276, + "step": 678 + }, + { + "epoch": 0.21715405065062263, + "grad_norm": 2.0771985054016113, + "learning_rate": 0.0005467320505967739, + "loss": 4.8671, + "step": 679 + }, + { + "epoch": 0.21747386515820824, + "grad_norm": 3.139260768890381, + "learning_rate": 0.0005465550942964903, + "loss": 4.8899, + "step": 680 + }, + { + "epoch": 0.21779367966579383, + "grad_norm": 0.9771436452865601, + "learning_rate": 0.000546377873295383, + "loss": 4.8459, + "step": 681 + }, + { + "epoch": 0.21811349417337944, + "grad_norm": 1.0279310941696167, + "learning_rate": 0.0005462003877837157, + "loss": 4.9454, + "step": 682 + }, + { + "epoch": 0.21843330868096503, + "grad_norm": 2.2688028812408447, + "learning_rate": 0.0005460226379520365, + "loss": 4.9459, + "step": 683 + }, + { + "epoch": 0.21875312318855064, + "grad_norm": 1.255409598350525, + "learning_rate": 0.0005458446239911772, + "loss": 4.8824, + "step": 684 + }, + { + "epoch": 0.21907293769613623, + "grad_norm": 0.9813272356987, + "learning_rate": 0.0005456663460922528, + "loss": 4.9385, + "step": 685 + }, + { + "epoch": 0.21939275220372184, + "grad_norm": 3.1207239627838135, + "learning_rate": 0.000545487804446662, + "loss": 4.8725, + "step": 686 + }, + { + "epoch": 0.21971256671130746, + "grad_norm": 1.4311422109603882, + "learning_rate": 0.0005453089992460868, + "loss": 4.9643, + "step": 687 + }, + { + "epoch": 0.22003238121889304, + "grad_norm": 0.990202009677887, + "learning_rate": 0.0005451299306824917, + "loss": 4.8807, + "step": 688 + }, + { + "epoch": 0.22035219572647866, + "grad_norm": 2.956839084625244, + "learning_rate": 0.0005449505989481243, + "loss": 4.9224, + "step": 689 + }, + { + "epoch": 0.22067201023406424, + "grad_norm": 1.1552083492279053, + "learning_rate": 0.0005447710042355145, + "loss": 4.8737, + "step": 690 + }, + { + "epoch": 0.22099182474164986, + "grad_norm": 0.9768129587173462, + "learning_rate": 0.0005445911467374747, + "loss": 4.9309, + "step": 691 + }, + { + "epoch": 0.22131163924923544, + "grad_norm": 1.2962573766708374, + "learning_rate": 0.0005444110266470995, + "loss": 4.9746, + "step": 692 + }, + { + "epoch": 0.22163145375682106, + "grad_norm": 1.4152376651763916, + "learning_rate": 0.0005442306441577651, + "loss": 5.1138, + "step": 693 + }, + { + "epoch": 0.22195126826440664, + "grad_norm": 1.6053122282028198, + "learning_rate": 0.0005440499994631299, + "loss": 4.9292, + "step": 694 + }, + { + "epoch": 0.22227108277199226, + "grad_norm": 4.396821022033691, + "learning_rate": 0.0005438690927571332, + "loss": 4.8633, + "step": 695 + }, + { + "epoch": 0.22259089727957784, + "grad_norm": 2.8457746505737305, + "learning_rate": 0.000543687924233996, + "loss": 4.8311, + "step": 696 + }, + { + "epoch": 0.22291071178716346, + "grad_norm": 2.485562562942505, + "learning_rate": 0.0005435064940882204, + "loss": 4.9475, + "step": 697 + }, + { + "epoch": 0.22323052629474904, + "grad_norm": 1.6794689893722534, + "learning_rate": 0.0005433248025145894, + "loss": 4.8663, + "step": 698 + }, + { + "epoch": 0.22355034080233466, + "grad_norm": 1.9672435522079468, + "learning_rate": 0.0005431428497081661, + "loss": 4.9456, + "step": 699 + }, + { + "epoch": 0.22387015530992024, + "grad_norm": 1.0523738861083984, + "learning_rate": 0.0005429606358642948, + "loss": 4.8134, + "step": 700 + }, + { + "epoch": 0.22387015530992024, + "eval_runtime": 21.3749, + "eval_samples_per_second": 88.749, + "eval_steps_per_second": 22.222, + "step": 700 + }, + { + "epoch": 0.22418996981750586, + "grad_norm": 1.7386295795440674, + "learning_rate": 0.0005427781611785998, + "loss": 4.8519, + "step": 701 + }, + { + "epoch": 0.22450978432509144, + "grad_norm": 1.7530839443206787, + "learning_rate": 0.0005425954258469852, + "loss": 4.8939, + "step": 702 + }, + { + "epoch": 0.22482959883267706, + "grad_norm": 3.3223302364349365, + "learning_rate": 0.000542412430065635, + "loss": 4.9035, + "step": 703 + }, + { + "epoch": 0.22514941334026264, + "grad_norm": 2.2270140647888184, + "learning_rate": 0.0005422291740310134, + "loss": 4.8525, + "step": 704 + }, + { + "epoch": 0.22546922784784826, + "grad_norm": 2.1718201637268066, + "learning_rate": 0.0005420456579398632, + "loss": 4.8883, + "step": 705 + }, + { + "epoch": 0.22578904235543384, + "grad_norm": 1.2573323249816895, + "learning_rate": 0.0005418618819892067, + "loss": 4.9088, + "step": 706 + }, + { + "epoch": 0.22610885686301946, + "grad_norm": 2.7146825790405273, + "learning_rate": 0.0005416778463763454, + "loss": 4.8544, + "step": 707 + }, + { + "epoch": 0.22642867137060504, + "grad_norm": 1.1185940504074097, + "learning_rate": 0.0005414935512988593, + "loss": 4.8779, + "step": 708 + }, + { + "epoch": 0.22674848587819066, + "grad_norm": 1.2980778217315674, + "learning_rate": 0.0005413089969546071, + "loss": 4.9463, + "step": 709 + }, + { + "epoch": 0.22706830038577624, + "grad_norm": 4.475209712982178, + "learning_rate": 0.0005411241835417256, + "loss": 4.9317, + "step": 710 + }, + { + "epoch": 0.22738811489336186, + "grad_norm": 1.1199473142623901, + "learning_rate": 0.0005409391112586303, + "loss": 4.8803, + "step": 711 + }, + { + "epoch": 0.22770792940094745, + "grad_norm": 1.3532423973083496, + "learning_rate": 0.0005407537803040139, + "loss": 4.8571, + "step": 712 + }, + { + "epoch": 0.22802774390853306, + "grad_norm": 5.268401145935059, + "learning_rate": 0.0005405681908768475, + "loss": 4.8288, + "step": 713 + }, + { + "epoch": 0.22834755841611865, + "grad_norm": 5.336316108703613, + "learning_rate": 0.0005403823431763791, + "loss": 4.9164, + "step": 714 + }, + { + "epoch": 0.22866737292370426, + "grad_norm": 1.28131103515625, + "learning_rate": 0.0005401962374021342, + "loss": 4.8627, + "step": 715 + }, + { + "epoch": 0.22898718743128985, + "grad_norm": 651.6729125976562, + "learning_rate": 0.0005400098737539157, + "loss": 5.0391, + "step": 716 + }, + { + "epoch": 0.22930700193887546, + "grad_norm": 1.2687968015670776, + "learning_rate": 0.0005398232524318029, + "loss": 4.8839, + "step": 717 + }, + { + "epoch": 0.22962681644646105, + "grad_norm": 3.2805144786834717, + "learning_rate": 0.0005396363736361519, + "loss": 4.9147, + "step": 718 + }, + { + "epoch": 0.22994663095404666, + "grad_norm": 1.8791998624801636, + "learning_rate": 0.0005394492375675953, + "loss": 4.8277, + "step": 719 + }, + { + "epoch": 0.23026644546163225, + "grad_norm": 1.3806695938110352, + "learning_rate": 0.0005392618444270417, + "loss": 4.8521, + "step": 720 + }, + { + "epoch": 0.23058625996921786, + "grad_norm": 1.5364006757736206, + "learning_rate": 0.0005390741944156759, + "loss": 4.9476, + "step": 721 + }, + { + "epoch": 0.23090607447680345, + "grad_norm": 104.7022933959961, + "learning_rate": 0.0005388862877349584, + "loss": 4.8401, + "step": 722 + }, + { + "epoch": 0.23122588898438906, + "grad_norm": 4.618757724761963, + "learning_rate": 0.0005386981245866252, + "loss": 4.8651, + "step": 723 + }, + { + "epoch": 0.23154570349197465, + "grad_norm": 1.8429489135742188, + "learning_rate": 0.0005385097051726879, + "loss": 4.8921, + "step": 724 + }, + { + "epoch": 0.23186551799956026, + "grad_norm": 5.0341033935546875, + "learning_rate": 0.0005383210296954328, + "loss": 4.9299, + "step": 725 + }, + { + "epoch": 0.23218533250714585, + "grad_norm": 1.5784238576889038, + "learning_rate": 0.0005381320983574214, + "loss": 4.9627, + "step": 726 + }, + { + "epoch": 0.23250514701473146, + "grad_norm": 1.785085916519165, + "learning_rate": 0.0005379429113614898, + "loss": 4.9823, + "step": 727 + }, + { + "epoch": 0.23282496152231705, + "grad_norm": 0.8675895929336548, + "learning_rate": 0.0005377534689107487, + "loss": 4.8556, + "step": 728 + }, + { + "epoch": 0.23314477602990266, + "grad_norm": 1.8664259910583496, + "learning_rate": 0.0005375637712085829, + "loss": 4.8055, + "step": 729 + }, + { + "epoch": 0.23346459053748825, + "grad_norm": 3.9013876914978027, + "learning_rate": 0.0005373738184586514, + "loss": 4.8622, + "step": 730 + }, + { + "epoch": 0.23378440504507386, + "grad_norm": 1.6758432388305664, + "learning_rate": 0.0005371836108648868, + "loss": 4.8947, + "step": 731 + }, + { + "epoch": 0.23410421955265945, + "grad_norm": 1.4901044368743896, + "learning_rate": 0.0005369931486314953, + "loss": 4.8416, + "step": 732 + }, + { + "epoch": 0.23442403406024506, + "grad_norm": 2.5616910457611084, + "learning_rate": 0.0005368024319629569, + "loss": 4.9829, + "step": 733 + }, + { + "epoch": 0.23474384856783065, + "grad_norm": 1.2935492992401123, + "learning_rate": 0.0005366114610640241, + "loss": 4.8795, + "step": 734 + }, + { + "epoch": 0.23506366307541626, + "grad_norm": 1.8023591041564941, + "learning_rate": 0.000536420236139723, + "loss": 4.8318, + "step": 735 + }, + { + "epoch": 0.23538347758300185, + "grad_norm": 8.306290626525879, + "learning_rate": 0.000536228757395352, + "loss": 4.7995, + "step": 736 + }, + { + "epoch": 0.23570329209058746, + "grad_norm": 1.3580899238586426, + "learning_rate": 0.000536037025036482, + "loss": 4.8862, + "step": 737 + }, + { + "epoch": 0.23602310659817305, + "grad_norm": 0.9650902152061462, + "learning_rate": 0.0005358450392689564, + "loss": 4.8159, + "step": 738 + }, + { + "epoch": 0.23634292110575866, + "grad_norm": 1.8293365240097046, + "learning_rate": 0.0005356528002988907, + "loss": 4.8764, + "step": 739 + }, + { + "epoch": 0.23666273561334425, + "grad_norm": 1.4953889846801758, + "learning_rate": 0.000535460308332672, + "loss": 4.8804, + "step": 740 + }, + { + "epoch": 0.23698255012092986, + "grad_norm": 3.914886951446533, + "learning_rate": 0.0005352675635769589, + "loss": 4.9617, + "step": 741 + }, + { + "epoch": 0.23730236462851545, + "grad_norm": 1.4074091911315918, + "learning_rate": 0.0005350745662386818, + "loss": 4.8882, + "step": 742 + }, + { + "epoch": 0.23762217913610106, + "grad_norm": 1.2808117866516113, + "learning_rate": 0.000534881316525042, + "loss": 4.8522, + "step": 743 + }, + { + "epoch": 0.23794199364368665, + "grad_norm": 7.654114723205566, + "learning_rate": 0.0005346878146435119, + "loss": 4.8444, + "step": 744 + }, + { + "epoch": 0.23826180815127226, + "grad_norm": 1.861201524734497, + "learning_rate": 0.0005344940608018345, + "loss": 4.8796, + "step": 745 + }, + { + "epoch": 0.23858162265885785, + "grad_norm": 1.3476163148880005, + "learning_rate": 0.0005343000552080235, + "loss": 4.8778, + "step": 746 + }, + { + "epoch": 0.23890143716644346, + "grad_norm": 1.498911738395691, + "learning_rate": 0.0005341057980703624, + "loss": 4.837, + "step": 747 + }, + { + "epoch": 0.23922125167402905, + "grad_norm": 2.5194218158721924, + "learning_rate": 0.0005339112895974054, + "loss": 4.8776, + "step": 748 + }, + { + "epoch": 0.23954106618161466, + "grad_norm": 1.918076515197754, + "learning_rate": 0.0005337165299979761, + "loss": 4.8737, + "step": 749 + }, + { + "epoch": 0.23986088068920025, + "grad_norm": 1.0306103229522705, + "learning_rate": 0.0005335215194811678, + "loss": 4.8847, + "step": 750 + }, + { + "epoch": 0.24018069519678586, + "grad_norm": 0.940801203250885, + "learning_rate": 0.0005333262582563434, + "loss": 4.86, + "step": 751 + }, + { + "epoch": 0.24050050970437145, + "grad_norm": 1.8134843111038208, + "learning_rate": 0.0005331307465331346, + "loss": 4.909, + "step": 752 + }, + { + "epoch": 0.24082032421195707, + "grad_norm": 4.539953708648682, + "learning_rate": 0.0005329349845214421, + "loss": 4.8616, + "step": 753 + }, + { + "epoch": 0.24114013871954265, + "grad_norm": 2.6143178939819336, + "learning_rate": 0.0005327389724314357, + "loss": 4.9105, + "step": 754 + }, + { + "epoch": 0.24145995322712827, + "grad_norm": 1.0225307941436768, + "learning_rate": 0.0005325427104735533, + "loss": 4.7909, + "step": 755 + }, + { + "epoch": 0.24177976773471388, + "grad_norm": 2.2887401580810547, + "learning_rate": 0.0005323461988585011, + "loss": 4.8373, + "step": 756 + }, + { + "epoch": 0.24209958224229947, + "grad_norm": 1.2456971406936646, + "learning_rate": 0.0005321494377972534, + "loss": 4.8513, + "step": 757 + }, + { + "epoch": 0.24241939674988508, + "grad_norm": 1.3714308738708496, + "learning_rate": 0.0005319524275010524, + "loss": 4.8683, + "step": 758 + }, + { + "epoch": 0.24273921125747067, + "grad_norm": 6.023526668548584, + "learning_rate": 0.0005317551681814076, + "loss": 4.8772, + "step": 759 + }, + { + "epoch": 0.24305902576505628, + "grad_norm": 1.1236687898635864, + "learning_rate": 0.0005315576600500962, + "loss": 4.9208, + "step": 760 + }, + { + "epoch": 0.24337884027264187, + "grad_norm": 1.1819261312484741, + "learning_rate": 0.0005313599033191622, + "loss": 4.8076, + "step": 761 + }, + { + "epoch": 0.24369865478022748, + "grad_norm": 1.660784363746643, + "learning_rate": 0.0005311618982009168, + "loss": 4.9019, + "step": 762 + }, + { + "epoch": 0.24401846928781307, + "grad_norm": 2.6814041137695312, + "learning_rate": 0.0005309636449079377, + "loss": 4.8988, + "step": 763 + }, + { + "epoch": 0.24433828379539868, + "grad_norm": 0.9590451121330261, + "learning_rate": 0.0005307651436530688, + "loss": 4.8904, + "step": 764 + }, + { + "epoch": 0.24465809830298427, + "grad_norm": 0.9582365155220032, + "learning_rate": 0.0005305663946494208, + "loss": 4.7931, + "step": 765 + }, + { + "epoch": 0.24497791281056988, + "grad_norm": 6.016003131866455, + "learning_rate": 0.0005303673981103698, + "loss": 4.8527, + "step": 766 + }, + { + "epoch": 0.24529772731815547, + "grad_norm": 1.4327377080917358, + "learning_rate": 0.000530168154249558, + "loss": 4.814, + "step": 767 + }, + { + "epoch": 0.24561754182574108, + "grad_norm": 4.536919116973877, + "learning_rate": 0.000529968663280893, + "loss": 4.8601, + "step": 768 + }, + { + "epoch": 0.24593735633332667, + "grad_norm": 1.4784783124923706, + "learning_rate": 0.0005297689254185478, + "loss": 4.9016, + "step": 769 + }, + { + "epoch": 0.24625717084091228, + "grad_norm": 1.948062539100647, + "learning_rate": 0.0005295689408769602, + "loss": 4.8604, + "step": 770 + }, + { + "epoch": 0.24657698534849787, + "grad_norm": 1.7188756465911865, + "learning_rate": 0.0005293687098708332, + "loss": 4.8975, + "step": 771 + }, + { + "epoch": 0.24689679985608348, + "grad_norm": 1.2714964151382446, + "learning_rate": 0.0005291682326151342, + "loss": 4.8675, + "step": 772 + }, + { + "epoch": 0.24721661436366907, + "grad_norm": 1.1173559427261353, + "learning_rate": 0.0005289675093250949, + "loss": 4.9005, + "step": 773 + }, + { + "epoch": 0.24753642887125468, + "grad_norm": 1.3392081260681152, + "learning_rate": 0.0005287665402162112, + "loss": 4.8462, + "step": 774 + }, + { + "epoch": 0.24785624337884027, + "grad_norm": 6.4686479568481445, + "learning_rate": 0.0005285653255042432, + "loss": 4.8446, + "step": 775 + }, + { + "epoch": 0.24817605788642588, + "grad_norm": 14.128937721252441, + "learning_rate": 0.0005283638654052141, + "loss": 4.803, + "step": 776 + }, + { + "epoch": 0.24849587239401147, + "grad_norm": 1.504435658454895, + "learning_rate": 0.000528162160135411, + "loss": 4.8898, + "step": 777 + }, + { + "epoch": 0.24881568690159708, + "grad_norm": 1.6049143075942993, + "learning_rate": 0.000527960209911384, + "loss": 4.8468, + "step": 778 + }, + { + "epoch": 0.24913550140918267, + "grad_norm": 1.4276559352874756, + "learning_rate": 0.0005277580149499465, + "loss": 4.8433, + "step": 779 + }, + { + "epoch": 0.24945531591676828, + "grad_norm": 1.9621052742004395, + "learning_rate": 0.0005275555754681742, + "loss": 4.8524, + "step": 780 + }, + { + "epoch": 0.24977513042435387, + "grad_norm": 2.969590663909912, + "learning_rate": 0.0005273528916834056, + "loss": 4.7982, + "step": 781 + }, + { + "epoch": 0.25009494493193946, + "grad_norm": 2.2482008934020996, + "learning_rate": 0.0005271499638132415, + "loss": 4.8408, + "step": 782 + }, + { + "epoch": 0.25041475943952507, + "grad_norm": 1.6192853450775146, + "learning_rate": 0.0005269467920755446, + "loss": 4.8771, + "step": 783 + }, + { + "epoch": 0.2507345739471107, + "grad_norm": 1.1925208568572998, + "learning_rate": 0.0005267433766884394, + "loss": 4.8214, + "step": 784 + }, + { + "epoch": 0.2510543884546963, + "grad_norm": 1.889891505241394, + "learning_rate": 0.0005265397178703122, + "loss": 4.7838, + "step": 785 + }, + { + "epoch": 0.25137420296228186, + "grad_norm": 1.5606637001037598, + "learning_rate": 0.0005263358158398104, + "loss": 4.7825, + "step": 786 + }, + { + "epoch": 0.25169401746986747, + "grad_norm": 2.185866594314575, + "learning_rate": 0.0005261316708158426, + "loss": 4.8343, + "step": 787 + }, + { + "epoch": 0.2520138319774531, + "grad_norm": 2.1191327571868896, + "learning_rate": 0.0005259272830175784, + "loss": 4.7935, + "step": 788 + }, + { + "epoch": 0.2523336464850387, + "grad_norm": 1.4705206155776978, + "learning_rate": 0.0005257226526644478, + "loss": 4.8694, + "step": 789 + }, + { + "epoch": 0.25265346099262426, + "grad_norm": 3.1378836631774902, + "learning_rate": 0.0005255177799761416, + "loss": 4.8248, + "step": 790 + }, + { + "epoch": 0.25297327550020987, + "grad_norm": 2.2317490577697754, + "learning_rate": 0.0005253126651726102, + "loss": 4.8333, + "step": 791 + }, + { + "epoch": 0.2532930900077955, + "grad_norm": 4.546291351318359, + "learning_rate": 0.0005251073084740646, + "loss": 4.8405, + "step": 792 + }, + { + "epoch": 0.2536129045153811, + "grad_norm": 2.402541160583496, + "learning_rate": 0.0005249017101009747, + "loss": 4.9049, + "step": 793 + }, + { + "epoch": 0.25393271902296666, + "grad_norm": 2.6798481941223145, + "learning_rate": 0.0005246958702740707, + "loss": 4.8472, + "step": 794 + }, + { + "epoch": 0.25425253353055227, + "grad_norm": 1.2594612836837769, + "learning_rate": 0.0005244897892143414, + "loss": 4.7669, + "step": 795 + }, + { + "epoch": 0.2545723480381379, + "grad_norm": 1.449859619140625, + "learning_rate": 0.0005242834671430349, + "loss": 4.8225, + "step": 796 + }, + { + "epoch": 0.2548921625457235, + "grad_norm": 2.1234524250030518, + "learning_rate": 0.0005240769042816581, + "loss": 4.7398, + "step": 797 + }, + { + "epoch": 0.25521197705330906, + "grad_norm": 2.7703492641448975, + "learning_rate": 0.0005238701008519761, + "loss": 4.7053, + "step": 798 + }, + { + "epoch": 0.25553179156089467, + "grad_norm": 1.3402894735336304, + "learning_rate": 0.0005236630570760126, + "loss": 4.7966, + "step": 799 + }, + { + "epoch": 0.2558516060684803, + "grad_norm": 1.2409049272537231, + "learning_rate": 0.0005234557731760489, + "loss": 4.8525, + "step": 800 + }, + { + "epoch": 0.2558516060684803, + "eval_runtime": 22.2695, + "eval_samples_per_second": 85.184, + "eval_steps_per_second": 21.33, + "step": 800 + }, + { + "epoch": 0.2561714205760659, + "grad_norm": 1.5717597007751465, + "learning_rate": 0.0005232482493746247, + "loss": 4.8493, + "step": 801 + }, + { + "epoch": 0.25649123508365146, + "grad_norm": 2.491340398788452, + "learning_rate": 0.0005230404858945369, + "loss": 4.8699, + "step": 802 + }, + { + "epoch": 0.2568110495912371, + "grad_norm": 2.582090377807617, + "learning_rate": 0.0005228324829588396, + "loss": 4.8699, + "step": 803 + }, + { + "epoch": 0.2571308640988227, + "grad_norm": 2.150022268295288, + "learning_rate": 0.0005226242407908441, + "loss": 4.8037, + "step": 804 + }, + { + "epoch": 0.2574506786064083, + "grad_norm": 3.018885374069214, + "learning_rate": 0.0005224157596141189, + "loss": 4.8091, + "step": 805 + }, + { + "epoch": 0.25777049311399386, + "grad_norm": 1.284055233001709, + "learning_rate": 0.0005222070396524886, + "loss": 4.753, + "step": 806 + }, + { + "epoch": 0.2580903076215795, + "grad_norm": 1.3739261627197266, + "learning_rate": 0.0005219980811300342, + "loss": 4.9714, + "step": 807 + }, + { + "epoch": 0.2584101221291651, + "grad_norm": 3.777207612991333, + "learning_rate": 0.0005217888842710931, + "loss": 4.8654, + "step": 808 + }, + { + "epoch": 0.2587299366367507, + "grad_norm": 2.5351736545562744, + "learning_rate": 0.0005215794493002583, + "loss": 4.7445, + "step": 809 + }, + { + "epoch": 0.25904975114433626, + "grad_norm": 1.7035949230194092, + "learning_rate": 0.000521369776442379, + "loss": 4.7995, + "step": 810 + }, + { + "epoch": 0.2593695656519219, + "grad_norm": 1.5396779775619507, + "learning_rate": 0.0005211598659225588, + "loss": 4.8041, + "step": 811 + }, + { + "epoch": 0.2596893801595075, + "grad_norm": 1.5708547830581665, + "learning_rate": 0.0005209497179661573, + "loss": 4.8735, + "step": 812 + }, + { + "epoch": 0.2600091946670931, + "grad_norm": 1.1383682489395142, + "learning_rate": 0.0005207393327987886, + "loss": 4.7766, + "step": 813 + }, + { + "epoch": 0.26032900917467866, + "grad_norm": 3.960334539413452, + "learning_rate": 0.0005205287106463219, + "loss": 4.7891, + "step": 814 + }, + { + "epoch": 0.2606488236822643, + "grad_norm": 2.656538486480713, + "learning_rate": 0.0005203178517348801, + "loss": 4.8682, + "step": 815 + }, + { + "epoch": 0.2609686381898499, + "grad_norm": 2.3352162837982178, + "learning_rate": 0.0005201067562908409, + "loss": 4.7779, + "step": 816 + }, + { + "epoch": 0.2612884526974355, + "grad_norm": 3.2890145778656006, + "learning_rate": 0.0005198954245408359, + "loss": 4.7632, + "step": 817 + }, + { + "epoch": 0.26160826720502106, + "grad_norm": 1.3325610160827637, + "learning_rate": 0.00051968385671175, + "loss": 4.813, + "step": 818 + }, + { + "epoch": 0.2619280817126067, + "grad_norm": 1.4385597705841064, + "learning_rate": 0.000519472053030722, + "loss": 4.7875, + "step": 819 + }, + { + "epoch": 0.2622478962201923, + "grad_norm": 1.27521812915802, + "learning_rate": 0.0005192600137251435, + "loss": 4.7993, + "step": 820 + }, + { + "epoch": 0.2625677107277779, + "grad_norm": 18.648155212402344, + "learning_rate": 0.0005190477390226595, + "loss": 4.8936, + "step": 821 + }, + { + "epoch": 0.26288752523536346, + "grad_norm": 3.521735906600952, + "learning_rate": 0.0005188352291511673, + "loss": 4.7657, + "step": 822 + }, + { + "epoch": 0.2632073397429491, + "grad_norm": 4.632253170013428, + "learning_rate": 0.000518622484338817, + "loss": 4.8452, + "step": 823 + }, + { + "epoch": 0.2635271542505347, + "grad_norm": 14.4740571975708, + "learning_rate": 0.0005184095048140106, + "loss": 4.8457, + "step": 824 + }, + { + "epoch": 0.2638469687581203, + "grad_norm": 2.033447504043579, + "learning_rate": 0.0005181962908054027, + "loss": 4.8633, + "step": 825 + }, + { + "epoch": 0.2641667832657059, + "grad_norm": 2.000201940536499, + "learning_rate": 0.0005179828425418988, + "loss": 4.8149, + "step": 826 + }, + { + "epoch": 0.2644865977732915, + "grad_norm": 3.670712947845459, + "learning_rate": 0.0005177691602526566, + "loss": 4.7743, + "step": 827 + }, + { + "epoch": 0.2648064122808771, + "grad_norm": 12.847811698913574, + "learning_rate": 0.0005175552441670847, + "loss": 4.8032, + "step": 828 + }, + { + "epoch": 0.2651262267884627, + "grad_norm": 4.958078384399414, + "learning_rate": 0.0005173410945148427, + "loss": 4.876, + "step": 829 + }, + { + "epoch": 0.2654460412960483, + "grad_norm": 1.6014165878295898, + "learning_rate": 0.0005171267115258412, + "loss": 4.8556, + "step": 830 + }, + { + "epoch": 0.2657658558036339, + "grad_norm": 1.165225625038147, + "learning_rate": 0.0005169120954302409, + "loss": 4.8061, + "step": 831 + }, + { + "epoch": 0.2660856703112195, + "grad_norm": 1.400700569152832, + "learning_rate": 0.0005166972464584532, + "loss": 4.7478, + "step": 832 + }, + { + "epoch": 0.2664054848188051, + "grad_norm": 1.087833046913147, + "learning_rate": 0.0005164821648411394, + "loss": 4.778, + "step": 833 + }, + { + "epoch": 0.2667252993263907, + "grad_norm": 2.4143247604370117, + "learning_rate": 0.0005162668508092103, + "loss": 4.6853, + "step": 834 + }, + { + "epoch": 0.2670451138339763, + "grad_norm": 6.460576057434082, + "learning_rate": 0.0005160513045938265, + "loss": 4.7761, + "step": 835 + }, + { + "epoch": 0.2673649283415619, + "grad_norm": 1.8703075647354126, + "learning_rate": 0.0005158355264263978, + "loss": 4.8112, + "step": 836 + }, + { + "epoch": 0.2676847428491475, + "grad_norm": 1.099163293838501, + "learning_rate": 0.0005156195165385829, + "loss": 4.7886, + "step": 837 + }, + { + "epoch": 0.2680045573567331, + "grad_norm": 4.774145126342773, + "learning_rate": 0.0005154032751622894, + "loss": 4.8082, + "step": 838 + }, + { + "epoch": 0.2683243718643187, + "grad_norm": 1.7929049730300903, + "learning_rate": 0.0005151868025296736, + "loss": 4.8991, + "step": 839 + }, + { + "epoch": 0.2686441863719043, + "grad_norm": 6.122037410736084, + "learning_rate": 0.0005149700988731397, + "loss": 4.7537, + "step": 840 + }, + { + "epoch": 0.2689640008794899, + "grad_norm": 1.1260855197906494, + "learning_rate": 0.0005147531644253402, + "loss": 4.8963, + "step": 841 + }, + { + "epoch": 0.2692838153870755, + "grad_norm": 1.4437628984451294, + "learning_rate": 0.0005145359994191751, + "loss": 4.8747, + "step": 842 + }, + { + "epoch": 0.2696036298946611, + "grad_norm": 2.4688847064971924, + "learning_rate": 0.0005143186040877923, + "loss": 4.7537, + "step": 843 + }, + { + "epoch": 0.2699234444022467, + "grad_norm": 1.4157013893127441, + "learning_rate": 0.0005141009786645868, + "loss": 4.7986, + "step": 844 + }, + { + "epoch": 0.2702432589098323, + "grad_norm": 2.6778647899627686, + "learning_rate": 0.0005138831233832005, + "loss": 4.8443, + "step": 845 + }, + { + "epoch": 0.2705630734174179, + "grad_norm": 2.8258864879608154, + "learning_rate": 0.0005136650384775221, + "loss": 4.8014, + "step": 846 + }, + { + "epoch": 0.2708828879250035, + "grad_norm": 4.153007507324219, + "learning_rate": 0.0005134467241816872, + "loss": 4.7693, + "step": 847 + }, + { + "epoch": 0.2712027024325891, + "grad_norm": 1.2077422142028809, + "learning_rate": 0.0005132281807300773, + "loss": 4.8968, + "step": 848 + }, + { + "epoch": 0.2715225169401747, + "grad_norm": 0.9899333119392395, + "learning_rate": 0.0005130094083573198, + "loss": 4.7883, + "step": 849 + }, + { + "epoch": 0.2718423314477603, + "grad_norm": 1.7131859064102173, + "learning_rate": 0.0005127904072982884, + "loss": 4.7929, + "step": 850 + }, + { + "epoch": 0.2721621459553459, + "grad_norm": 9.47719955444336, + "learning_rate": 0.0005125711777881016, + "loss": 4.7496, + "step": 851 + }, + { + "epoch": 0.2724819604629315, + "grad_norm": 1.634995937347412, + "learning_rate": 0.0005123517200621238, + "loss": 4.8026, + "step": 852 + }, + { + "epoch": 0.2728017749705171, + "grad_norm": 1.188254475593567, + "learning_rate": 0.0005121320343559641, + "loss": 4.7887, + "step": 853 + }, + { + "epoch": 0.2731215894781027, + "grad_norm": 1.4568613767623901, + "learning_rate": 0.0005119121209054767, + "loss": 4.8787, + "step": 854 + }, + { + "epoch": 0.2734414039856883, + "grad_norm": 1.6773172616958618, + "learning_rate": 0.0005116919799467597, + "loss": 4.8639, + "step": 855 + }, + { + "epoch": 0.2737612184932739, + "grad_norm": 2.126314163208008, + "learning_rate": 0.0005114716117161558, + "loss": 4.8036, + "step": 856 + }, + { + "epoch": 0.2740810330008595, + "grad_norm": 4.216558456420898, + "learning_rate": 0.0005112510164502518, + "loss": 4.782, + "step": 857 + }, + { + "epoch": 0.2744008475084451, + "grad_norm": 1.477610468864441, + "learning_rate": 0.000511030194385878, + "loss": 4.8034, + "step": 858 + }, + { + "epoch": 0.2747206620160307, + "grad_norm": 1.1306517124176025, + "learning_rate": 0.0005108091457601085, + "loss": 4.765, + "step": 859 + }, + { + "epoch": 0.2750404765236163, + "grad_norm": 1.3832201957702637, + "learning_rate": 0.0005105878708102604, + "loss": 4.7923, + "step": 860 + }, + { + "epoch": 0.2753602910312019, + "grad_norm": 2.307032346725464, + "learning_rate": 0.0005103663697738937, + "loss": 4.7305, + "step": 861 + }, + { + "epoch": 0.2756801055387875, + "grad_norm": 1.233642578125, + "learning_rate": 0.0005101446428888115, + "loss": 4.7942, + "step": 862 + }, + { + "epoch": 0.2759999200463731, + "grad_norm": 1.022827386856079, + "learning_rate": 0.0005099226903930589, + "loss": 4.8306, + "step": 863 + }, + { + "epoch": 0.2763197345539587, + "grad_norm": 1.0522570610046387, + "learning_rate": 0.0005097005125249236, + "loss": 4.7521, + "step": 864 + }, + { + "epoch": 0.2766395490615443, + "grad_norm": 1.0753915309906006, + "learning_rate": 0.0005094781095229352, + "loss": 4.7412, + "step": 865 + }, + { + "epoch": 0.2769593635691299, + "grad_norm": 1.8710469007492065, + "learning_rate": 0.0005092554816258644, + "loss": 4.7702, + "step": 866 + }, + { + "epoch": 0.2772791780767155, + "grad_norm": 2.6220498085021973, + "learning_rate": 0.0005090326290727245, + "loss": 4.7748, + "step": 867 + }, + { + "epoch": 0.2775989925843011, + "grad_norm": 1.2614277601242065, + "learning_rate": 0.0005088095521027689, + "loss": 4.7846, + "step": 868 + }, + { + "epoch": 0.2779188070918867, + "grad_norm": 1.269464373588562, + "learning_rate": 0.0005085862509554926, + "loss": 4.7519, + "step": 869 + }, + { + "epoch": 0.2782386215994723, + "grad_norm": 2.22206711769104, + "learning_rate": 0.000508362725870631, + "loss": 4.7544, + "step": 870 + }, + { + "epoch": 0.2785584361070579, + "grad_norm": 2.356398820877075, + "learning_rate": 0.0005081389770881599, + "loss": 4.8267, + "step": 871 + }, + { + "epoch": 0.2788782506146435, + "grad_norm": 1.3920925855636597, + "learning_rate": 0.0005079150048482954, + "loss": 4.7101, + "step": 872 + }, + { + "epoch": 0.2791980651222291, + "grad_norm": 2.786470890045166, + "learning_rate": 0.0005076908093914936, + "loss": 4.7889, + "step": 873 + }, + { + "epoch": 0.2795178796298147, + "grad_norm": 3.574336051940918, + "learning_rate": 0.0005074663909584498, + "loss": 4.7993, + "step": 874 + }, + { + "epoch": 0.2798376941374003, + "grad_norm": 2.7027246952056885, + "learning_rate": 0.000507241749790099, + "loss": 4.7936, + "step": 875 + }, + { + "epoch": 0.2801575086449859, + "grad_norm": 1.3531469106674194, + "learning_rate": 0.0005070168861276155, + "loss": 4.8065, + "step": 876 + }, + { + "epoch": 0.2804773231525715, + "grad_norm": 1.1245033740997314, + "learning_rate": 0.0005067918002124121, + "loss": 4.7677, + "step": 877 + }, + { + "epoch": 0.2807971376601571, + "grad_norm": 4.90814208984375, + "learning_rate": 0.0005065664922861405, + "loss": 4.854, + "step": 878 + }, + { + "epoch": 0.2811169521677427, + "grad_norm": 1.9774717092514038, + "learning_rate": 0.0005063409625906905, + "loss": 4.7747, + "step": 879 + }, + { + "epoch": 0.2814367666753283, + "grad_norm": 58.17668533325195, + "learning_rate": 0.0005061152113681901, + "loss": 4.7818, + "step": 880 + }, + { + "epoch": 0.2817565811829139, + "grad_norm": 1.9991294145584106, + "learning_rate": 0.0005058892388610053, + "loss": 4.802, + "step": 881 + }, + { + "epoch": 0.2820763956904995, + "grad_norm": 1.686437964439392, + "learning_rate": 0.0005056630453117394, + "loss": 4.7922, + "step": 882 + }, + { + "epoch": 0.2823962101980851, + "grad_norm": 2.1209917068481445, + "learning_rate": 0.0005054366309632333, + "loss": 4.8187, + "step": 883 + }, + { + "epoch": 0.2827160247056707, + "grad_norm": 3.691697359085083, + "learning_rate": 0.0005052099960585645, + "loss": 4.7778, + "step": 884 + }, + { + "epoch": 0.2830358392132563, + "grad_norm": 2.5676209926605225, + "learning_rate": 0.0005049831408410478, + "loss": 4.8098, + "step": 885 + }, + { + "epoch": 0.2833556537208419, + "grad_norm": 1.7618777751922607, + "learning_rate": 0.0005047560655542342, + "loss": 4.7762, + "step": 886 + }, + { + "epoch": 0.2836754682284275, + "grad_norm": 1.145598292350769, + "learning_rate": 0.000504528770441911, + "loss": 4.7829, + "step": 887 + }, + { + "epoch": 0.2839952827360131, + "grad_norm": 2.2979626655578613, + "learning_rate": 0.0005043012557481016, + "loss": 4.786, + "step": 888 + }, + { + "epoch": 0.2843150972435987, + "grad_norm": 1.177972674369812, + "learning_rate": 0.0005040735217170653, + "loss": 4.7875, + "step": 889 + }, + { + "epoch": 0.2846349117511843, + "grad_norm": 1.0710322856903076, + "learning_rate": 0.0005038455685932964, + "loss": 4.7515, + "step": 890 + }, + { + "epoch": 0.2849547262587699, + "grad_norm": 6.937191009521484, + "learning_rate": 0.0005036173966215248, + "loss": 4.7293, + "step": 891 + }, + { + "epoch": 0.2852745407663555, + "grad_norm": 1.065787434577942, + "learning_rate": 0.0005033890060467153, + "loss": 4.7242, + "step": 892 + }, + { + "epoch": 0.2855943552739411, + "grad_norm": 1.42545485496521, + "learning_rate": 0.0005031603971140674, + "loss": 4.7538, + "step": 893 + }, + { + "epoch": 0.28591416978152673, + "grad_norm": 2.2558865547180176, + "learning_rate": 0.000502931570069015, + "loss": 4.6528, + "step": 894 + }, + { + "epoch": 0.28623398428911234, + "grad_norm": 1.506136178970337, + "learning_rate": 0.0005027025251572259, + "loss": 4.7736, + "step": 895 + }, + { + "epoch": 0.2865537987966979, + "grad_norm": 1.3734699487686157, + "learning_rate": 0.0005024732626246022, + "loss": 4.7424, + "step": 896 + }, + { + "epoch": 0.2868736133042835, + "grad_norm": 1.7918834686279297, + "learning_rate": 0.0005022437827172795, + "loss": 4.8466, + "step": 897 + }, + { + "epoch": 0.28719342781186913, + "grad_norm": 1.4223873615264893, + "learning_rate": 0.0005020140856816268, + "loss": 4.872, + "step": 898 + }, + { + "epoch": 0.28751324231945474, + "grad_norm": 2.283947467803955, + "learning_rate": 0.0005017841717642461, + "loss": 4.8072, + "step": 899 + }, + { + "epoch": 0.2878330568270403, + "grad_norm": 2.999143123626709, + "learning_rate": 0.0005015540412119721, + "loss": 4.718, + "step": 900 + }, + { + "epoch": 0.2878330568270403, + "eval_runtime": 22.4143, + "eval_samples_per_second": 84.634, + "eval_steps_per_second": 21.192, + "step": 900 + }, + { + "epoch": 0.2881528713346259, + "grad_norm": 1.6161720752716064, + "learning_rate": 0.0005013236942718725, + "loss": 4.7605, + "step": 901 + }, + { + "epoch": 0.28847268584221153, + "grad_norm": 1.6709802150726318, + "learning_rate": 0.0005010931311912473, + "loss": 4.8159, + "step": 902 + }, + { + "epoch": 0.28879250034979714, + "grad_norm": 1.3722419738769531, + "learning_rate": 0.0005008623522176279, + "loss": 4.7999, + "step": 903 + }, + { + "epoch": 0.2891123148573827, + "grad_norm": 1.7038166522979736, + "learning_rate": 0.0005006313575987784, + "loss": 4.7526, + "step": 904 + }, + { + "epoch": 0.2894321293649683, + "grad_norm": 1.4413397312164307, + "learning_rate": 0.0005004001475826935, + "loss": 4.8354, + "step": 905 + }, + { + "epoch": 0.28975194387255393, + "grad_norm": 2.4395699501037598, + "learning_rate": 0.0005001687224175999, + "loss": 4.8284, + "step": 906 + }, + { + "epoch": 0.29007175838013954, + "grad_norm": 6.372836589813232, + "learning_rate": 0.0004999370823519548, + "loss": 4.8145, + "step": 907 + }, + { + "epoch": 0.2903915728877251, + "grad_norm": 4.7980241775512695, + "learning_rate": 0.0004997052276344463, + "loss": 4.8156, + "step": 908 + }, + { + "epoch": 0.2907113873953107, + "grad_norm": 1.41959547996521, + "learning_rate": 0.000499473158513993, + "loss": 4.8342, + "step": 909 + }, + { + "epoch": 0.29103120190289633, + "grad_norm": 1.8412573337554932, + "learning_rate": 0.0004992408752397437, + "loss": 4.7596, + "step": 910 + }, + { + "epoch": 0.29135101641048194, + "grad_norm": 2.1000828742980957, + "learning_rate": 0.0004990083780610769, + "loss": 4.6603, + "step": 911 + }, + { + "epoch": 0.2916708309180675, + "grad_norm": 5.267429828643799, + "learning_rate": 0.000498775667227601, + "loss": 4.7482, + "step": 912 + }, + { + "epoch": 0.2919906454256531, + "grad_norm": 1.9088479280471802, + "learning_rate": 0.0004985427429891536, + "loss": 4.832, + "step": 913 + }, + { + "epoch": 0.29231045993323873, + "grad_norm": 1.2986080646514893, + "learning_rate": 0.0004983096055958014, + "loss": 4.7477, + "step": 914 + }, + { + "epoch": 0.29263027444082435, + "grad_norm": 1.331952452659607, + "learning_rate": 0.0004980762552978403, + "loss": 4.7035, + "step": 915 + }, + { + "epoch": 0.2929500889484099, + "grad_norm": 1.2349594831466675, + "learning_rate": 0.0004978426923457942, + "loss": 4.7414, + "step": 916 + }, + { + "epoch": 0.2932699034559955, + "grad_norm": 1.3519343137741089, + "learning_rate": 0.0004976089169904156, + "loss": 4.6685, + "step": 917 + }, + { + "epoch": 0.29358971796358113, + "grad_norm": 4.408710956573486, + "learning_rate": 0.0004973749294826853, + "loss": 4.7511, + "step": 918 + }, + { + "epoch": 0.29390953247116675, + "grad_norm": 5.034624099731445, + "learning_rate": 0.0004971407300738114, + "loss": 4.6662, + "step": 919 + }, + { + "epoch": 0.2942293469787523, + "grad_norm": 0.9654132723808289, + "learning_rate": 0.0004969063190152297, + "loss": 4.7033, + "step": 920 + }, + { + "epoch": 0.2945491614863379, + "grad_norm": 1.2738070487976074, + "learning_rate": 0.0004966716965586033, + "loss": 4.7343, + "step": 921 + }, + { + "epoch": 0.29486897599392353, + "grad_norm": 1.2482532262802124, + "learning_rate": 0.0004964368629558221, + "loss": 4.7566, + "step": 922 + }, + { + "epoch": 0.29518879050150915, + "grad_norm": 1.2440038919448853, + "learning_rate": 0.0004962018184590028, + "loss": 4.7688, + "step": 923 + }, + { + "epoch": 0.2955086050090947, + "grad_norm": 4.011666297912598, + "learning_rate": 0.0004959665633204885, + "loss": 4.6708, + "step": 924 + }, + { + "epoch": 0.2958284195166803, + "grad_norm": 1.4939723014831543, + "learning_rate": 0.0004957310977928484, + "loss": 4.7259, + "step": 925 + }, + { + "epoch": 0.29614823402426593, + "grad_norm": 2.640099048614502, + "learning_rate": 0.0004954954221288775, + "loss": 4.7754, + "step": 926 + }, + { + "epoch": 0.29646804853185155, + "grad_norm": 1.4164668321609497, + "learning_rate": 0.0004952595365815967, + "loss": 4.8152, + "step": 927 + }, + { + "epoch": 0.2967878630394371, + "grad_norm": 1.5110958814620972, + "learning_rate": 0.0004950234414042519, + "loss": 4.6824, + "step": 928 + }, + { + "epoch": 0.2971076775470227, + "grad_norm": 5.062756538391113, + "learning_rate": 0.0004947871368503143, + "loss": 4.7096, + "step": 929 + }, + { + "epoch": 0.29742749205460833, + "grad_norm": 1.995178461074829, + "learning_rate": 0.0004945506231734796, + "loss": 4.801, + "step": 930 + }, + { + "epoch": 0.29774730656219395, + "grad_norm": 1.7213729619979858, + "learning_rate": 0.0004943139006276683, + "loss": 4.7884, + "step": 931 + }, + { + "epoch": 0.2980671210697795, + "grad_norm": 2.2084274291992188, + "learning_rate": 0.0004940769694670251, + "loss": 4.8286, + "step": 932 + }, + { + "epoch": 0.2983869355773651, + "grad_norm": 2.3480591773986816, + "learning_rate": 0.0004938398299459183, + "loss": 4.7567, + "step": 933 + }, + { + "epoch": 0.29870675008495073, + "grad_norm": 2.518298387527466, + "learning_rate": 0.0004936024823189406, + "loss": 4.6935, + "step": 934 + }, + { + "epoch": 0.29902656459253635, + "grad_norm": 2.0153439044952393, + "learning_rate": 0.0004933649268409073, + "loss": 4.6976, + "step": 935 + }, + { + "epoch": 0.2993463791001219, + "grad_norm": 2.1075024604797363, + "learning_rate": 0.0004931271637668577, + "loss": 4.8394, + "step": 936 + }, + { + "epoch": 0.2996661936077075, + "grad_norm": 2.1927642822265625, + "learning_rate": 0.0004928891933520533, + "loss": 4.7547, + "step": 937 + }, + { + "epoch": 0.29998600811529313, + "grad_norm": 8.683347702026367, + "learning_rate": 0.0004926510158519784, + "loss": 4.7532, + "step": 938 + }, + { + "epoch": 0.30030582262287875, + "grad_norm": 1.6210992336273193, + "learning_rate": 0.0004924126315223396, + "loss": 4.8369, + "step": 939 + }, + { + "epoch": 0.3006256371304643, + "grad_norm": 1.581929087638855, + "learning_rate": 0.0004921740406190659, + "loss": 4.7111, + "step": 940 + }, + { + "epoch": 0.3009454516380499, + "grad_norm": 1.7412323951721191, + "learning_rate": 0.0004919352433983075, + "loss": 4.6758, + "step": 941 + }, + { + "epoch": 0.30126526614563554, + "grad_norm": 9.842538833618164, + "learning_rate": 0.0004916962401164365, + "loss": 4.7227, + "step": 942 + }, + { + "epoch": 0.30158508065322115, + "grad_norm": 4.679938793182373, + "learning_rate": 0.0004914570310300462, + "loss": 4.7232, + "step": 943 + }, + { + "epoch": 0.3019048951608067, + "grad_norm": 2.031250476837158, + "learning_rate": 0.0004912176163959506, + "loss": 4.794, + "step": 944 + }, + { + "epoch": 0.3022247096683923, + "grad_norm": 18.600994110107422, + "learning_rate": 0.0004909779964711848, + "loss": 4.701, + "step": 945 + }, + { + "epoch": 0.30254452417597794, + "grad_norm": 3.319349527359009, + "learning_rate": 0.0004907381715130038, + "loss": 4.6733, + "step": 946 + }, + { + "epoch": 0.30286433868356355, + "grad_norm": 1.7237615585327148, + "learning_rate": 0.000490498141778883, + "loss": 4.7347, + "step": 947 + }, + { + "epoch": 0.3031841531911491, + "grad_norm": 2.477285146713257, + "learning_rate": 0.0004902579075265178, + "loss": 4.7009, + "step": 948 + }, + { + "epoch": 0.3035039676987347, + "grad_norm": 2.068237781524658, + "learning_rate": 0.0004900174690138229, + "loss": 4.7421, + "step": 949 + }, + { + "epoch": 0.30382378220632034, + "grad_norm": 1.9319586753845215, + "learning_rate": 0.0004897768264989323, + "loss": 4.7894, + "step": 950 + }, + { + "epoch": 0.30414359671390595, + "grad_norm": 2.7936956882476807, + "learning_rate": 0.0004895359802401992, + "loss": 4.7879, + "step": 951 + }, + { + "epoch": 0.3044634112214915, + "grad_norm": 1.922629952430725, + "learning_rate": 0.0004892949304961952, + "loss": 4.7193, + "step": 952 + }, + { + "epoch": 0.3047832257290771, + "grad_norm": 10.242497444152832, + "learning_rate": 0.0004890536775257109, + "loss": 4.6534, + "step": 953 + }, + { + "epoch": 0.30510304023666274, + "grad_norm": 12.216218948364258, + "learning_rate": 0.0004888122215877547, + "loss": 4.7337, + "step": 954 + }, + { + "epoch": 0.30542285474424835, + "grad_norm": 3.5213205814361572, + "learning_rate": 0.0004885705629415528, + "loss": 4.7311, + "step": 955 + }, + { + "epoch": 0.3057426692518339, + "grad_norm": 1.6398122310638428, + "learning_rate": 0.0004883287018465494, + "loss": 4.7558, + "step": 956 + }, + { + "epoch": 0.3060624837594195, + "grad_norm": 1.624783992767334, + "learning_rate": 0.00048808663856240596, + "loss": 4.8081, + "step": 957 + }, + { + "epoch": 0.30638229826700514, + "grad_norm": 3.1297478675842285, + "learning_rate": 0.0004878443733490006, + "loss": 4.7092, + "step": 958 + }, + { + "epoch": 0.30670211277459075, + "grad_norm": 1.3655167818069458, + "learning_rate": 0.00048760190646642866, + "loss": 4.6653, + "step": 959 + }, + { + "epoch": 0.3070219272821763, + "grad_norm": 1.8067723512649536, + "learning_rate": 0.000487359238175002, + "loss": 4.743, + "step": 960 + }, + { + "epoch": 0.3073417417897619, + "grad_norm": 2.391735553741455, + "learning_rate": 0.00048711636873524856, + "loss": 4.7218, + "step": 961 + }, + { + "epoch": 0.30766155629734754, + "grad_norm": 1.2850096225738525, + "learning_rate": 0.00048687329840791207, + "loss": 4.7809, + "step": 962 + }, + { + "epoch": 0.30798137080493315, + "grad_norm": 2.633840560913086, + "learning_rate": 0.0004866300274539523, + "loss": 4.7127, + "step": 963 + }, + { + "epoch": 0.30830118531251877, + "grad_norm": 1.6916964054107666, + "learning_rate": 0.0004863865561345442, + "loss": 4.7497, + "step": 964 + }, + { + "epoch": 0.3086209998201043, + "grad_norm": 1.8493642807006836, + "learning_rate": 0.00048614288471107774, + "loss": 4.657, + "step": 965 + }, + { + "epoch": 0.30894081432768994, + "grad_norm": 1.8589093685150146, + "learning_rate": 0.00048589901344515805, + "loss": 4.7827, + "step": 966 + }, + { + "epoch": 0.30926062883527555, + "grad_norm": 2.012005090713501, + "learning_rate": 0.00048565494259860434, + "loss": 4.7079, + "step": 967 + }, + { + "epoch": 0.30958044334286117, + "grad_norm": 1.6338340044021606, + "learning_rate": 0.00048541067243345064, + "loss": 4.725, + "step": 968 + }, + { + "epoch": 0.3099002578504467, + "grad_norm": 1.9878655672073364, + "learning_rate": 0.00048516620321194443, + "loss": 4.7294, + "step": 969 + }, + { + "epoch": 0.31022007235803234, + "grad_norm": 1.7506719827651978, + "learning_rate": 0.0004849215351965474, + "loss": 4.6513, + "step": 970 + }, + { + "epoch": 0.31053988686561795, + "grad_norm": 1.8063925504684448, + "learning_rate": 0.0004846766686499342, + "loss": 4.7487, + "step": 971 + }, + { + "epoch": 0.31085970137320357, + "grad_norm": 6.096024990081787, + "learning_rate": 0.0004844316038349929, + "loss": 4.6987, + "step": 972 + }, + { + "epoch": 0.3111795158807891, + "grad_norm": 1.7704654932022095, + "learning_rate": 0.00048418634101482435, + "loss": 4.6898, + "step": 973 + }, + { + "epoch": 0.31149933038837474, + "grad_norm": 2.966852903366089, + "learning_rate": 0.000483940880452742, + "loss": 4.7395, + "step": 974 + }, + { + "epoch": 0.31181914489596035, + "grad_norm": 2.54073429107666, + "learning_rate": 0.0004836952224122716, + "loss": 4.7734, + "step": 975 + }, + { + "epoch": 0.31213895940354597, + "grad_norm": 1.2422308921813965, + "learning_rate": 0.00048344936715715104, + "loss": 4.781, + "step": 976 + }, + { + "epoch": 0.3124587739111315, + "grad_norm": 26.399932861328125, + "learning_rate": 0.0004832033149513295, + "loss": 4.6701, + "step": 977 + }, + { + "epoch": 0.31277858841871714, + "grad_norm": 1.187759518623352, + "learning_rate": 0.0004829570660589681, + "loss": 4.8006, + "step": 978 + }, + { + "epoch": 0.31309840292630275, + "grad_norm": 2.7032063007354736, + "learning_rate": 0.0004827106207444389, + "loss": 4.6998, + "step": 979 + }, + { + "epoch": 0.31341821743388837, + "grad_norm": 3.3780429363250732, + "learning_rate": 0.00048246397927232483, + "loss": 4.7422, + "step": 980 + }, + { + "epoch": 0.3137380319414739, + "grad_norm": 2.4896981716156006, + "learning_rate": 0.00048221714190741947, + "loss": 4.7418, + "step": 981 + }, + { + "epoch": 0.31405784644905954, + "grad_norm": 2.132725954055786, + "learning_rate": 0.00048197010891472665, + "loss": 4.7145, + "step": 982 + }, + { + "epoch": 0.31437766095664516, + "grad_norm": 2.9203903675079346, + "learning_rate": 0.00048172288055946033, + "loss": 4.7747, + "step": 983 + }, + { + "epoch": 0.31469747546423077, + "grad_norm": 1.7667096853256226, + "learning_rate": 0.0004814754571070442, + "loss": 4.7754, + "step": 984 + }, + { + "epoch": 0.3150172899718163, + "grad_norm": 1.7587201595306396, + "learning_rate": 0.00048122783882311126, + "loss": 4.7568, + "step": 985 + }, + { + "epoch": 0.31533710447940194, + "grad_norm": 2.853320837020874, + "learning_rate": 0.0004809800259735038, + "loss": 4.7479, + "step": 986 + }, + { + "epoch": 0.31565691898698756, + "grad_norm": 1.9687455892562866, + "learning_rate": 0.0004807320188242728, + "loss": 4.6381, + "step": 987 + }, + { + "epoch": 0.31597673349457317, + "grad_norm": 2.0123367309570312, + "learning_rate": 0.0004804838176416782, + "loss": 4.6948, + "step": 988 + }, + { + "epoch": 0.31629654800215873, + "grad_norm": 4.395848274230957, + "learning_rate": 0.000480235422692188, + "loss": 4.8021, + "step": 989 + }, + { + "epoch": 0.31661636250974434, + "grad_norm": 1.7425918579101562, + "learning_rate": 0.0004799868342424784, + "loss": 4.6604, + "step": 990 + }, + { + "epoch": 0.31693617701732996, + "grad_norm": 3.6321957111358643, + "learning_rate": 0.00047973805255943305, + "loss": 4.7472, + "step": 991 + }, + { + "epoch": 0.31725599152491557, + "grad_norm": 3.3034281730651855, + "learning_rate": 0.0004794890779101434, + "loss": 4.7566, + "step": 992 + }, + { + "epoch": 0.31757580603250113, + "grad_norm": 4.606922626495361, + "learning_rate": 0.0004792399105619077, + "loss": 4.7697, + "step": 993 + }, + { + "epoch": 0.31789562054008674, + "grad_norm": 1.7199561595916748, + "learning_rate": 0.0004789905507822314, + "loss": 4.6988, + "step": 994 + }, + { + "epoch": 0.31821543504767236, + "grad_norm": 1.6437046527862549, + "learning_rate": 0.00047874099883882644, + "loss": 4.7349, + "step": 995 + }, + { + "epoch": 0.31853524955525797, + "grad_norm": 1.5925757884979248, + "learning_rate": 0.000478491254999611, + "loss": 4.704, + "step": 996 + }, + { + "epoch": 0.31885506406284353, + "grad_norm": 2.9087681770324707, + "learning_rate": 0.0004782413195327094, + "loss": 4.7409, + "step": 997 + }, + { + "epoch": 0.31917487857042914, + "grad_norm": 1.7927370071411133, + "learning_rate": 0.0004779911927064516, + "loss": 4.6084, + "step": 998 + }, + { + "epoch": 0.31949469307801476, + "grad_norm": 2.6278116703033447, + "learning_rate": 0.000477740874789373, + "loss": 4.7106, + "step": 999 + }, + { + "epoch": 0.31981450758560037, + "grad_norm": 2.0117173194885254, + "learning_rate": 0.0004774903660502142, + "loss": 4.6847, + "step": 1000 + }, + { + "epoch": 0.31981450758560037, + "eval_runtime": 21.1436, + "eval_samples_per_second": 89.72, + "eval_steps_per_second": 22.465, + "step": 1000 + }, + { + "epoch": 0.32013432209318593, + "grad_norm": 1.4243024587631226, + "learning_rate": 0.0004772396667579205, + "loss": 4.7684, + "step": 1001 + }, + { + "epoch": 0.32045413660077154, + "grad_norm": 3.3258867263793945, + "learning_rate": 0.0004769887771816422, + "loss": 4.6635, + "step": 1002 + }, + { + "epoch": 0.32077395110835716, + "grad_norm": 1.7894575595855713, + "learning_rate": 0.0004767376975907334, + "loss": 4.741, + "step": 1003 + }, + { + "epoch": 0.32109376561594277, + "grad_norm": 2.5367367267608643, + "learning_rate": 0.00047648642825475255, + "loss": 4.6813, + "step": 1004 + }, + { + "epoch": 0.32141358012352833, + "grad_norm": 1.4260019063949585, + "learning_rate": 0.0004762349694434615, + "loss": 4.6968, + "step": 1005 + }, + { + "epoch": 0.32173339463111394, + "grad_norm": 1.8931576013565063, + "learning_rate": 0.0004759833214268259, + "loss": 4.6804, + "step": 1006 + }, + { + "epoch": 0.32205320913869956, + "grad_norm": 1.7740689516067505, + "learning_rate": 0.0004757314844750141, + "loss": 4.6599, + "step": 1007 + }, + { + "epoch": 0.3223730236462852, + "grad_norm": 4.1497483253479, + "learning_rate": 0.00047547945885839763, + "loss": 4.7632, + "step": 1008 + }, + { + "epoch": 0.32269283815387073, + "grad_norm": 10.036355972290039, + "learning_rate": 0.00047522724484755054, + "loss": 4.7483, + "step": 1009 + }, + { + "epoch": 0.32301265266145635, + "grad_norm": 5.461592197418213, + "learning_rate": 0.0004749748427132488, + "loss": 4.7028, + "step": 1010 + }, + { + "epoch": 0.32333246716904196, + "grad_norm": 2.9767701625823975, + "learning_rate": 0.00047472225272647084, + "loss": 4.7782, + "step": 1011 + }, + { + "epoch": 0.3236522816766276, + "grad_norm": 3.0752665996551514, + "learning_rate": 0.00047446947515839634, + "loss": 4.7315, + "step": 1012 + }, + { + "epoch": 0.32397209618421313, + "grad_norm": 2.13083553314209, + "learning_rate": 0.0004742165102804067, + "loss": 4.7536, + "step": 1013 + }, + { + "epoch": 0.32429191069179875, + "grad_norm": 3.8241984844207764, + "learning_rate": 0.00047396335836408427, + "loss": 4.8076, + "step": 1014 + }, + { + "epoch": 0.32461172519938436, + "grad_norm": 2.8082728385925293, + "learning_rate": 0.0004737100196812121, + "loss": 4.7087, + "step": 1015 + }, + { + "epoch": 0.32493153970697, + "grad_norm": 1.476997971534729, + "learning_rate": 0.00047345649450377395, + "loss": 4.6592, + "step": 1016 + }, + { + "epoch": 0.32525135421455553, + "grad_norm": 2.348454236984253, + "learning_rate": 0.0004732027831039536, + "loss": 4.5935, + "step": 1017 + }, + { + "epoch": 0.32557116872214115, + "grad_norm": 5.2407426834106445, + "learning_rate": 0.00047294888575413486, + "loss": 4.7312, + "step": 1018 + }, + { + "epoch": 0.32589098322972676, + "grad_norm": 2.1126835346221924, + "learning_rate": 0.0004726948027269013, + "loss": 4.7148, + "step": 1019 + }, + { + "epoch": 0.3262107977373124, + "grad_norm": 6.471402645111084, + "learning_rate": 0.00047244053429503565, + "loss": 4.7975, + "step": 1020 + }, + { + "epoch": 0.32653061224489793, + "grad_norm": 134.11346435546875, + "learning_rate": 0.00047218608073151976, + "loss": 4.6492, + "step": 1021 + }, + { + "epoch": 0.32685042675248355, + "grad_norm": 3.825564384460449, + "learning_rate": 0.0004719314423095342, + "loss": 4.7534, + "step": 1022 + }, + { + "epoch": 0.32717024126006916, + "grad_norm": 1.4359794855117798, + "learning_rate": 0.00047167661930245803, + "loss": 4.6304, + "step": 1023 + }, + { + "epoch": 0.3274900557676548, + "grad_norm": 1.3668348789215088, + "learning_rate": 0.0004714216119838685, + "loss": 4.7066, + "step": 1024 + }, + { + "epoch": 0.32780987027524033, + "grad_norm": 3.1498653888702393, + "learning_rate": 0.00047116642062754074, + "loss": 4.7177, + "step": 1025 + }, + { + "epoch": 0.32812968478282595, + "grad_norm": 1.4201635122299194, + "learning_rate": 0.00047091104550744733, + "loss": 4.7115, + "step": 1026 + }, + { + "epoch": 0.32844949929041156, + "grad_norm": 3.807563304901123, + "learning_rate": 0.00047065548689775844, + "loss": 4.6511, + "step": 1027 + }, + { + "epoch": 0.3287693137979972, + "grad_norm": 3.457021951675415, + "learning_rate": 0.00047039974507284086, + "loss": 4.7461, + "step": 1028 + }, + { + "epoch": 0.32908912830558273, + "grad_norm": 1.5525423288345337, + "learning_rate": 0.0004701438203072584, + "loss": 4.7263, + "step": 1029 + }, + { + "epoch": 0.32940894281316835, + "grad_norm": 1.9328609704971313, + "learning_rate": 0.00046988771287577105, + "loss": 4.7246, + "step": 1030 + }, + { + "epoch": 0.32972875732075396, + "grad_norm": 1.6063733100891113, + "learning_rate": 0.0004696314230533349, + "loss": 4.6872, + "step": 1031 + }, + { + "epoch": 0.3300485718283396, + "grad_norm": 4.702549457550049, + "learning_rate": 0.00046937495111510204, + "loss": 4.6696, + "step": 1032 + }, + { + "epoch": 0.3303683863359252, + "grad_norm": 3.3707308769226074, + "learning_rate": 0.00046911829733642016, + "loss": 4.6759, + "step": 1033 + }, + { + "epoch": 0.33068820084351075, + "grad_norm": 1.9789724349975586, + "learning_rate": 0.0004688614619928318, + "loss": 4.7431, + "step": 1034 + }, + { + "epoch": 0.33100801535109636, + "grad_norm": 1.7036818265914917, + "learning_rate": 0.00046860444536007473, + "loss": 4.7868, + "step": 1035 + }, + { + "epoch": 0.331327829858682, + "grad_norm": 2.3478903770446777, + "learning_rate": 0.0004683472477140811, + "loss": 4.7137, + "step": 1036 + }, + { + "epoch": 0.3316476443662676, + "grad_norm": 2.7520999908447266, + "learning_rate": 0.0004680898693309777, + "loss": 4.674, + "step": 1037 + }, + { + "epoch": 0.33196745887385315, + "grad_norm": 2.8961760997772217, + "learning_rate": 0.0004678323104870852, + "loss": 4.7345, + "step": 1038 + }, + { + "epoch": 0.33228727338143876, + "grad_norm": 2.629613161087036, + "learning_rate": 0.00046757457145891784, + "loss": 4.6158, + "step": 1039 + }, + { + "epoch": 0.3326070878890244, + "grad_norm": 1.4060332775115967, + "learning_rate": 0.00046731665252318366, + "loss": 4.6919, + "step": 1040 + }, + { + "epoch": 0.33292690239661, + "grad_norm": 2.569775104522705, + "learning_rate": 0.0004670585539567837, + "loss": 4.6806, + "step": 1041 + }, + { + "epoch": 0.33324671690419555, + "grad_norm": 2.9161183834075928, + "learning_rate": 0.00046680027603681164, + "loss": 4.6487, + "step": 1042 + }, + { + "epoch": 0.33356653141178116, + "grad_norm": 1.8700696229934692, + "learning_rate": 0.0004665418190405541, + "loss": 4.6109, + "step": 1043 + }, + { + "epoch": 0.3338863459193668, + "grad_norm": 1.7654300928115845, + "learning_rate": 0.0004662831832454895, + "loss": 4.736, + "step": 1044 + }, + { + "epoch": 0.3342061604269524, + "grad_norm": 3.293349504470825, + "learning_rate": 0.00046602436892928875, + "loss": 4.6787, + "step": 1045 + }, + { + "epoch": 0.33452597493453795, + "grad_norm": 8.065296173095703, + "learning_rate": 0.000465765376369814, + "loss": 4.7259, + "step": 1046 + }, + { + "epoch": 0.33484578944212356, + "grad_norm": 4.667914867401123, + "learning_rate": 0.00046550620584511886, + "loss": 4.7166, + "step": 1047 + }, + { + "epoch": 0.3351656039497092, + "grad_norm": 4.0608320236206055, + "learning_rate": 0.00046524685763344803, + "loss": 4.6643, + "step": 1048 + }, + { + "epoch": 0.3354854184572948, + "grad_norm": 1.5849798917770386, + "learning_rate": 0.00046498733201323715, + "loss": 4.6823, + "step": 1049 + }, + { + "epoch": 0.33580523296488035, + "grad_norm": 2.3301053047180176, + "learning_rate": 0.000464727629263112, + "loss": 4.7095, + "step": 1050 + }, + { + "epoch": 0.33612504747246597, + "grad_norm": 6.076179504394531, + "learning_rate": 0.0004644677496618887, + "loss": 4.7116, + "step": 1051 + }, + { + "epoch": 0.3364448619800516, + "grad_norm": 2.0797460079193115, + "learning_rate": 0.00046420769348857343, + "loss": 4.7087, + "step": 1052 + }, + { + "epoch": 0.3367646764876372, + "grad_norm": 28.04334259033203, + "learning_rate": 0.00046394746102236144, + "loss": 4.6294, + "step": 1053 + }, + { + "epoch": 0.33708449099522275, + "grad_norm": 5.230954170227051, + "learning_rate": 0.00046368705254263773, + "loss": 4.641, + "step": 1054 + }, + { + "epoch": 0.33740430550280837, + "grad_norm": 1.595682144165039, + "learning_rate": 0.000463426468328976, + "loss": 4.6678, + "step": 1055 + }, + { + "epoch": 0.337724120010394, + "grad_norm": 1.2185108661651611, + "learning_rate": 0.0004631657086611387, + "loss": 4.6786, + "step": 1056 + }, + { + "epoch": 0.3380439345179796, + "grad_norm": 2.838529348373413, + "learning_rate": 0.0004629047738190767, + "loss": 4.6992, + "step": 1057 + }, + { + "epoch": 0.33836374902556515, + "grad_norm": 1.8879170417785645, + "learning_rate": 0.00046264366408292883, + "loss": 4.6898, + "step": 1058 + }, + { + "epoch": 0.33868356353315077, + "grad_norm": 2.506721258163452, + "learning_rate": 0.0004623823797330216, + "loss": 4.716, + "step": 1059 + }, + { + "epoch": 0.3390033780407364, + "grad_norm": 1.4764541387557983, + "learning_rate": 0.00046212092104986946, + "loss": 4.7193, + "step": 1060 + }, + { + "epoch": 0.339323192548322, + "grad_norm": 2.153808116912842, + "learning_rate": 0.0004618592883141734, + "loss": 4.6283, + "step": 1061 + }, + { + "epoch": 0.33964300705590755, + "grad_norm": 2.1847524642944336, + "learning_rate": 0.00046159748180682166, + "loss": 4.6983, + "step": 1062 + }, + { + "epoch": 0.33996282156349317, + "grad_norm": 1.489572286605835, + "learning_rate": 0.0004613355018088889, + "loss": 4.5809, + "step": 1063 + }, + { + "epoch": 0.3402826360710788, + "grad_norm": 1.6547610759735107, + "learning_rate": 0.00046107334860163605, + "loss": 4.7021, + "step": 1064 + }, + { + "epoch": 0.3406024505786644, + "grad_norm": 1.4357720613479614, + "learning_rate": 0.00046081102246651014, + "loss": 4.7081, + "step": 1065 + }, + { + "epoch": 0.34092226508624995, + "grad_norm": 1.2727382183074951, + "learning_rate": 0.0004605485236851436, + "loss": 4.6501, + "step": 1066 + }, + { + "epoch": 0.34124207959383557, + "grad_norm": 3.398604154586792, + "learning_rate": 0.0004602858525393544, + "loss": 4.6037, + "step": 1067 + }, + { + "epoch": 0.3415618941014212, + "grad_norm": 1.606054425239563, + "learning_rate": 0.00046002300931114555, + "loss": 4.6916, + "step": 1068 + }, + { + "epoch": 0.3418817086090068, + "grad_norm": 1.5120043754577637, + "learning_rate": 0.0004597599942827048, + "loss": 4.7073, + "step": 1069 + }, + { + "epoch": 0.34220152311659235, + "grad_norm": 3.5497753620147705, + "learning_rate": 0.0004594968077364041, + "loss": 4.7797, + "step": 1070 + }, + { + "epoch": 0.34252133762417797, + "grad_norm": 4.054625511169434, + "learning_rate": 0.00045923344995480006, + "loss": 4.6681, + "step": 1071 + }, + { + "epoch": 0.3428411521317636, + "grad_norm": 2.1043620109558105, + "learning_rate": 0.0004589699212206325, + "loss": 4.7648, + "step": 1072 + }, + { + "epoch": 0.3431609666393492, + "grad_norm": 2.3509936332702637, + "learning_rate": 0.0004587062218168253, + "loss": 4.7259, + "step": 1073 + }, + { + "epoch": 0.34348078114693475, + "grad_norm": 2.246786117553711, + "learning_rate": 0.0004584423520264853, + "loss": 4.7338, + "step": 1074 + }, + { + "epoch": 0.34380059565452037, + "grad_norm": 3.060725212097168, + "learning_rate": 0.0004581783121329024, + "loss": 4.7052, + "step": 1075 + }, + { + "epoch": 0.344120410162106, + "grad_norm": 2.1863625049591064, + "learning_rate": 0.00045791410241954894, + "loss": 4.5644, + "step": 1076 + }, + { + "epoch": 0.3444402246696916, + "grad_norm": 1.5245671272277832, + "learning_rate": 0.0004576497231700798, + "loss": 4.6828, + "step": 1077 + }, + { + "epoch": 0.34476003917727716, + "grad_norm": 2.603569269180298, + "learning_rate": 0.0004573851746683317, + "loss": 4.704, + "step": 1078 + }, + { + "epoch": 0.34507985368486277, + "grad_norm": 1.5441813468933105, + "learning_rate": 0.00045712045719832313, + "loss": 4.6255, + "step": 1079 + }, + { + "epoch": 0.3453996681924484, + "grad_norm": 2.1603050231933594, + "learning_rate": 0.00045685557104425397, + "loss": 4.7954, + "step": 1080 + }, + { + "epoch": 0.345719482700034, + "grad_norm": 2.4608540534973145, + "learning_rate": 0.00045659051649050525, + "loss": 4.6407, + "step": 1081 + }, + { + "epoch": 0.34603929720761956, + "grad_norm": 2.1810410022735596, + "learning_rate": 0.00045632529382163883, + "loss": 4.643, + "step": 1082 + }, + { + "epoch": 0.34635911171520517, + "grad_norm": 3.403019666671753, + "learning_rate": 0.00045605990332239684, + "loss": 4.6433, + "step": 1083 + }, + { + "epoch": 0.3466789262227908, + "grad_norm": 3.33785343170166, + "learning_rate": 0.00045579434527770186, + "loss": 4.6416, + "step": 1084 + }, + { + "epoch": 0.3469987407303764, + "grad_norm": 1.7174862623214722, + "learning_rate": 0.0004555286199726561, + "loss": 4.6235, + "step": 1085 + }, + { + "epoch": 0.34731855523796196, + "grad_norm": 2.371502637863159, + "learning_rate": 0.0004552627276925416, + "loss": 4.6496, + "step": 1086 + }, + { + "epoch": 0.34763836974554757, + "grad_norm": 3.605410099029541, + "learning_rate": 0.0004549966687228195, + "loss": 4.623, + "step": 1087 + }, + { + "epoch": 0.3479581842531332, + "grad_norm": 7.701961994171143, + "learning_rate": 0.0004547304433491299, + "loss": 4.6901, + "step": 1088 + }, + { + "epoch": 0.3482779987607188, + "grad_norm": 2.09686017036438, + "learning_rate": 0.00045446405185729154, + "loss": 4.6437, + "step": 1089 + }, + { + "epoch": 0.34859781326830436, + "grad_norm": 1.6308674812316895, + "learning_rate": 0.00045419749453330167, + "loss": 4.6451, + "step": 1090 + }, + { + "epoch": 0.34891762777588997, + "grad_norm": 2.687452554702759, + "learning_rate": 0.00045393077166333524, + "loss": 4.6866, + "step": 1091 + }, + { + "epoch": 0.3492374422834756, + "grad_norm": 2.2837085723876953, + "learning_rate": 0.0004536638835337452, + "loss": 4.6746, + "step": 1092 + }, + { + "epoch": 0.3495572567910612, + "grad_norm": 4.630202770233154, + "learning_rate": 0.00045339683043106214, + "loss": 4.6606, + "step": 1093 + }, + { + "epoch": 0.34987707129864676, + "grad_norm": 2.010613441467285, + "learning_rate": 0.00045312961264199316, + "loss": 4.6116, + "step": 1094 + }, + { + "epoch": 0.35019688580623237, + "grad_norm": 1.7351913452148438, + "learning_rate": 0.0004528622304534225, + "loss": 4.6978, + "step": 1095 + }, + { + "epoch": 0.350516700313818, + "grad_norm": 1.290215253829956, + "learning_rate": 0.00045259468415241117, + "loss": 4.6019, + "step": 1096 + }, + { + "epoch": 0.3508365148214036, + "grad_norm": 5.041443824768066, + "learning_rate": 0.0004523269740261957, + "loss": 4.6958, + "step": 1097 + }, + { + "epoch": 0.35115632932898916, + "grad_norm": 6.838896751403809, + "learning_rate": 0.0004520591003621892, + "loss": 4.6047, + "step": 1098 + }, + { + "epoch": 0.3514761438365748, + "grad_norm": 2.1653571128845215, + "learning_rate": 0.00045179106344798005, + "loss": 4.6874, + "step": 1099 + }, + { + "epoch": 0.3517959583441604, + "grad_norm": 1.5599297285079956, + "learning_rate": 0.00045152286357133157, + "loss": 4.6629, + "step": 1100 + }, + { + "epoch": 0.3517959583441604, + "eval_runtime": 21.9815, + "eval_samples_per_second": 86.3, + "eval_steps_per_second": 21.609, + "step": 1100 + }, + { + "epoch": 0.352115772851746, + "grad_norm": 6.188464164733887, + "learning_rate": 0.0004512545010201828, + "loss": 4.5841, + "step": 1101 + }, + { + "epoch": 0.3524355873593316, + "grad_norm": 3.5884718894958496, + "learning_rate": 0.0004509859760826466, + "loss": 4.5795, + "step": 1102 + }, + { + "epoch": 0.3527554018669172, + "grad_norm": 1.449102520942688, + "learning_rate": 0.0004507172890470108, + "loss": 4.6483, + "step": 1103 + }, + { + "epoch": 0.3530752163745028, + "grad_norm": 1.7918610572814941, + "learning_rate": 0.000450448440201737, + "loss": 4.5962, + "step": 1104 + }, + { + "epoch": 0.3533950308820884, + "grad_norm": 2.100355863571167, + "learning_rate": 0.0004501794298354603, + "loss": 4.6471, + "step": 1105 + }, + { + "epoch": 0.353714845389674, + "grad_norm": 3.562272787094116, + "learning_rate": 0.0004499102582369897, + "loss": 4.5327, + "step": 1106 + }, + { + "epoch": 0.3540346598972596, + "grad_norm": 2.7404353618621826, + "learning_rate": 0.0004496409256953069, + "loss": 4.7471, + "step": 1107 + }, + { + "epoch": 0.3543544744048452, + "grad_norm": 3.045760154724121, + "learning_rate": 0.0004493714324995666, + "loss": 4.7155, + "step": 1108 + }, + { + "epoch": 0.3546742889124308, + "grad_norm": 3.4457881450653076, + "learning_rate": 0.00044910177893909577, + "loss": 4.7447, + "step": 1109 + }, + { + "epoch": 0.3549941034200164, + "grad_norm": 44.27642822265625, + "learning_rate": 0.00044883196530339376, + "loss": 4.6577, + "step": 1110 + }, + { + "epoch": 0.355313917927602, + "grad_norm": 2.2440507411956787, + "learning_rate": 0.0004485619918821318, + "loss": 4.6625, + "step": 1111 + }, + { + "epoch": 0.3556337324351876, + "grad_norm": 3.003131866455078, + "learning_rate": 0.00044829185896515245, + "loss": 4.7614, + "step": 1112 + }, + { + "epoch": 0.3559535469427732, + "grad_norm": 1.8937455415725708, + "learning_rate": 0.0004480215668424696, + "loss": 4.6792, + "step": 1113 + }, + { + "epoch": 0.3562733614503588, + "grad_norm": 2.734799861907959, + "learning_rate": 0.00044775111580426817, + "loss": 4.6599, + "step": 1114 + }, + { + "epoch": 0.3565931759579444, + "grad_norm": 2.1578783988952637, + "learning_rate": 0.00044748050614090343, + "loss": 4.6936, + "step": 1115 + }, + { + "epoch": 0.35691299046553, + "grad_norm": 3.8896563053131104, + "learning_rate": 0.00044720973814290125, + "loss": 4.6234, + "step": 1116 + }, + { + "epoch": 0.3572328049731156, + "grad_norm": 1.502860426902771, + "learning_rate": 0.0004469388121009574, + "loss": 4.6506, + "step": 1117 + }, + { + "epoch": 0.3575526194807012, + "grad_norm": 2.178720235824585, + "learning_rate": 0.00044666772830593714, + "loss": 4.6765, + "step": 1118 + }, + { + "epoch": 0.3578724339882868, + "grad_norm": 2.0360758304595947, + "learning_rate": 0.00044639648704887535, + "loss": 4.6438, + "step": 1119 + }, + { + "epoch": 0.3581922484958724, + "grad_norm": 1.8404183387756348, + "learning_rate": 0.00044612508862097575, + "loss": 4.6518, + "step": 1120 + }, + { + "epoch": 0.358512063003458, + "grad_norm": 1.6645119190216064, + "learning_rate": 0.00044585353331361095, + "loss": 4.6385, + "step": 1121 + }, + { + "epoch": 0.3588318775110436, + "grad_norm": 73.05937957763672, + "learning_rate": 0.000445581821418322, + "loss": 4.6632, + "step": 1122 + }, + { + "epoch": 0.3591516920186292, + "grad_norm": 3.5887668132781982, + "learning_rate": 0.0004453099532268178, + "loss": 4.7021, + "step": 1123 + }, + { + "epoch": 0.3594715065262148, + "grad_norm": 7.575666427612305, + "learning_rate": 0.0004450379290309755, + "loss": 4.5337, + "step": 1124 + }, + { + "epoch": 0.3597913210338004, + "grad_norm": 7.1917853355407715, + "learning_rate": 0.0004447657491228392, + "loss": 4.613, + "step": 1125 + }, + { + "epoch": 0.360111135541386, + "grad_norm": 11.06462287902832, + "learning_rate": 0.0004444934137946207, + "loss": 4.628, + "step": 1126 + }, + { + "epoch": 0.3604309500489716, + "grad_norm": 2.2979581356048584, + "learning_rate": 0.00044422092333869814, + "loss": 4.6527, + "step": 1127 + }, + { + "epoch": 0.3607507645565572, + "grad_norm": 6.938014984130859, + "learning_rate": 0.00044394827804761667, + "loss": 4.7164, + "step": 1128 + }, + { + "epoch": 0.3610705790641428, + "grad_norm": 1.234302043914795, + "learning_rate": 0.0004436754782140875, + "loss": 4.61, + "step": 1129 + }, + { + "epoch": 0.3613903935717284, + "grad_norm": 1.7554782629013062, + "learning_rate": 0.0004434025241309876, + "loss": 4.7045, + "step": 1130 + }, + { + "epoch": 0.361710208079314, + "grad_norm": 2.8521852493286133, + "learning_rate": 0.0004431294160913597, + "loss": 4.581, + "step": 1131 + }, + { + "epoch": 0.3620300225868996, + "grad_norm": 27.637285232543945, + "learning_rate": 0.0004428561543884118, + "loss": 4.5174, + "step": 1132 + }, + { + "epoch": 0.3623498370944852, + "grad_norm": 4.580589771270752, + "learning_rate": 0.0004425827393155169, + "loss": 4.6396, + "step": 1133 + }, + { + "epoch": 0.3626696516020708, + "grad_norm": 2.3599798679351807, + "learning_rate": 0.00044230917116621266, + "loss": 4.5838, + "step": 1134 + }, + { + "epoch": 0.3629894661096564, + "grad_norm": 4.184299468994141, + "learning_rate": 0.00044203545023420085, + "loss": 4.7215, + "step": 1135 + }, + { + "epoch": 0.363309280617242, + "grad_norm": 1.1452866792678833, + "learning_rate": 0.00044176157681334767, + "loss": 4.5952, + "step": 1136 + }, + { + "epoch": 0.3636290951248276, + "grad_norm": 1.9863295555114746, + "learning_rate": 0.0004414875511976827, + "loss": 4.5965, + "step": 1137 + }, + { + "epoch": 0.3639489096324132, + "grad_norm": 4.71095609664917, + "learning_rate": 0.00044121337368139906, + "loss": 4.7035, + "step": 1138 + }, + { + "epoch": 0.3642687241399988, + "grad_norm": 2.205233335494995, + "learning_rate": 0.0004409390445588528, + "loss": 4.6361, + "step": 1139 + }, + { + "epoch": 0.3645885386475844, + "grad_norm": 2.154536008834839, + "learning_rate": 0.0004406645641245631, + "loss": 4.6499, + "step": 1140 + }, + { + "epoch": 0.36490835315517, + "grad_norm": 2.4756979942321777, + "learning_rate": 0.0004403899326732112, + "loss": 4.6074, + "step": 1141 + }, + { + "epoch": 0.3652281676627556, + "grad_norm": 2.0306153297424316, + "learning_rate": 0.00044011515049964073, + "loss": 4.6336, + "step": 1142 + }, + { + "epoch": 0.3655479821703412, + "grad_norm": 1.9566924571990967, + "learning_rate": 0.0004398402178988568, + "loss": 4.7073, + "step": 1143 + }, + { + "epoch": 0.3658677966779268, + "grad_norm": 1.2427000999450684, + "learning_rate": 0.00043956513516602653, + "loss": 4.6696, + "step": 1144 + }, + { + "epoch": 0.3661876111855124, + "grad_norm": 5.675422191619873, + "learning_rate": 0.00043928990259647764, + "loss": 4.5456, + "step": 1145 + }, + { + "epoch": 0.366507425693098, + "grad_norm": 1.4576926231384277, + "learning_rate": 0.00043901452048569913, + "loss": 4.5857, + "step": 1146 + }, + { + "epoch": 0.3668272402006836, + "grad_norm": 1.869260311126709, + "learning_rate": 0.00043873898912934054, + "loss": 4.7149, + "step": 1147 + }, + { + "epoch": 0.3671470547082692, + "grad_norm": 1.1366474628448486, + "learning_rate": 0.00043846330882321146, + "loss": 4.642, + "step": 1148 + }, + { + "epoch": 0.3674668692158548, + "grad_norm": 2.4113426208496094, + "learning_rate": 0.00043818747986328136, + "loss": 4.5432, + "step": 1149 + }, + { + "epoch": 0.3677866837234404, + "grad_norm": 2.6148619651794434, + "learning_rate": 0.0004379115025456795, + "loss": 4.6291, + "step": 1150 + }, + { + "epoch": 0.368106498231026, + "grad_norm": 1.659581184387207, + "learning_rate": 0.0004376353771666942, + "loss": 4.6672, + "step": 1151 + }, + { + "epoch": 0.3684263127386116, + "grad_norm": 1.6792813539505005, + "learning_rate": 0.000437359104022773, + "loss": 4.6768, + "step": 1152 + }, + { + "epoch": 0.3687461272461972, + "grad_norm": 4.113164901733398, + "learning_rate": 0.00043708268341052185, + "loss": 4.5949, + "step": 1153 + }, + { + "epoch": 0.3690659417537828, + "grad_norm": 2.5655832290649414, + "learning_rate": 0.00043680611562670513, + "loss": 4.607, + "step": 1154 + }, + { + "epoch": 0.3693857562613684, + "grad_norm": 5.6711106300354, + "learning_rate": 0.00043652940096824516, + "loss": 4.6602, + "step": 1155 + }, + { + "epoch": 0.369705570768954, + "grad_norm": 1.8205032348632812, + "learning_rate": 0.00043625253973222206, + "loss": 4.6044, + "step": 1156 + }, + { + "epoch": 0.3700253852765396, + "grad_norm": 2.113050937652588, + "learning_rate": 0.00043597553221587316, + "loss": 4.7192, + "step": 1157 + }, + { + "epoch": 0.3703451997841252, + "grad_norm": 2.005631446838379, + "learning_rate": 0.00043569837871659296, + "loss": 4.6259, + "step": 1158 + }, + { + "epoch": 0.3706650142917108, + "grad_norm": 2.0923709869384766, + "learning_rate": 0.0004354210795319327, + "loss": 4.686, + "step": 1159 + }, + { + "epoch": 0.3709848287992964, + "grad_norm": 2.160266637802124, + "learning_rate": 0.00043514363495959985, + "loss": 4.5935, + "step": 1160 + }, + { + "epoch": 0.371304643306882, + "grad_norm": 3.1247971057891846, + "learning_rate": 0.0004348660452974581, + "loss": 4.5756, + "step": 1161 + }, + { + "epoch": 0.3716244578144676, + "grad_norm": 2.206557512283325, + "learning_rate": 0.00043458831084352705, + "loss": 4.6412, + "step": 1162 + }, + { + "epoch": 0.3719442723220532, + "grad_norm": 4.030003070831299, + "learning_rate": 0.00043431043189598125, + "loss": 4.6618, + "step": 1163 + }, + { + "epoch": 0.3722640868296388, + "grad_norm": 9.344891548156738, + "learning_rate": 0.0004340324087531511, + "loss": 4.6623, + "step": 1164 + }, + { + "epoch": 0.3725839013372244, + "grad_norm": 1.3127264976501465, + "learning_rate": 0.00043375424171352133, + "loss": 4.5699, + "step": 1165 + }, + { + "epoch": 0.37290371584481, + "grad_norm": 2.9577102661132812, + "learning_rate": 0.00043347593107573106, + "loss": 4.6555, + "step": 1166 + }, + { + "epoch": 0.37322353035239564, + "grad_norm": 2.46506929397583, + "learning_rate": 0.000433197477138574, + "loss": 4.6507, + "step": 1167 + }, + { + "epoch": 0.3735433448599812, + "grad_norm": 6.051362991333008, + "learning_rate": 0.00043291888020099723, + "loss": 4.5643, + "step": 1168 + }, + { + "epoch": 0.3738631593675668, + "grad_norm": 2.233818531036377, + "learning_rate": 0.0004326401405621019, + "loss": 4.6551, + "step": 1169 + }, + { + "epoch": 0.3741829738751524, + "grad_norm": 2.1978511810302734, + "learning_rate": 0.0004323612585211419, + "loss": 4.6372, + "step": 1170 + }, + { + "epoch": 0.37450278838273804, + "grad_norm": 1.3209125995635986, + "learning_rate": 0.0004320822343775242, + "loss": 4.7265, + "step": 1171 + }, + { + "epoch": 0.3748226028903236, + "grad_norm": 1.2830467224121094, + "learning_rate": 0.00043180306843080836, + "loss": 4.6679, + "step": 1172 + }, + { + "epoch": 0.3751424173979092, + "grad_norm": 2.137458562850952, + "learning_rate": 0.0004315237609807059, + "loss": 4.6848, + "step": 1173 + }, + { + "epoch": 0.3754622319054948, + "grad_norm": 60.69804763793945, + "learning_rate": 0.00043124431232708076, + "loss": 4.6096, + "step": 1174 + }, + { + "epoch": 0.37578204641308044, + "grad_norm": 1.6131093502044678, + "learning_rate": 0.000430964722769948, + "loss": 4.6076, + "step": 1175 + }, + { + "epoch": 0.376101860920666, + "grad_norm": 6.561488628387451, + "learning_rate": 0.0004306849926094742, + "loss": 4.5213, + "step": 1176 + }, + { + "epoch": 0.3764216754282516, + "grad_norm": 2.4109458923339844, + "learning_rate": 0.00043040512214597684, + "loss": 4.6067, + "step": 1177 + }, + { + "epoch": 0.3767414899358372, + "grad_norm": 2.2183072566986084, + "learning_rate": 0.00043012511167992405, + "loss": 4.6192, + "step": 1178 + }, + { + "epoch": 0.37706130444342284, + "grad_norm": 23.891170501708984, + "learning_rate": 0.0004298449615119343, + "loss": 4.5911, + "step": 1179 + }, + { + "epoch": 0.3773811189510084, + "grad_norm": 1.986393928527832, + "learning_rate": 0.0004295646719427758, + "loss": 4.6497, + "step": 1180 + }, + { + "epoch": 0.377700933458594, + "grad_norm": 1.717061996459961, + "learning_rate": 0.00042928424327336667, + "loss": 4.6066, + "step": 1181 + }, + { + "epoch": 0.3780207479661796, + "grad_norm": 1.510114073753357, + "learning_rate": 0.00042900367580477446, + "loss": 4.5451, + "step": 1182 + }, + { + "epoch": 0.37834056247376524, + "grad_norm": 3.5713465213775635, + "learning_rate": 0.0004287229698382154, + "loss": 4.6275, + "step": 1183 + }, + { + "epoch": 0.3786603769813508, + "grad_norm": 2.9515762329101562, + "learning_rate": 0.0004284421256750547, + "loss": 4.6631, + "step": 1184 + }, + { + "epoch": 0.3789801914889364, + "grad_norm": 2.485563039779663, + "learning_rate": 0.0004281611436168059, + "loss": 4.6272, + "step": 1185 + }, + { + "epoch": 0.379300005996522, + "grad_norm": 1.3505284786224365, + "learning_rate": 0.00042788002396513023, + "loss": 4.5989, + "step": 1186 + }, + { + "epoch": 0.37961982050410764, + "grad_norm": 1.6701558828353882, + "learning_rate": 0.00042759876702183706, + "loss": 4.6174, + "step": 1187 + }, + { + "epoch": 0.3799396350116932, + "grad_norm": 1.2788770198822021, + "learning_rate": 0.0004273173730888831, + "loss": 4.5687, + "step": 1188 + }, + { + "epoch": 0.3802594495192788, + "grad_norm": 2.520719289779663, + "learning_rate": 0.00042703584246837206, + "loss": 4.6278, + "step": 1189 + }, + { + "epoch": 0.38057926402686443, + "grad_norm": 8.861785888671875, + "learning_rate": 0.0004267541754625543, + "loss": 4.635, + "step": 1190 + }, + { + "epoch": 0.38089907853445004, + "grad_norm": 4.913196563720703, + "learning_rate": 0.00042647237237382666, + "loss": 4.5658, + "step": 1191 + }, + { + "epoch": 0.3812188930420356, + "grad_norm": 2.777759313583374, + "learning_rate": 0.00042619043350473223, + "loss": 4.7422, + "step": 1192 + }, + { + "epoch": 0.3815387075496212, + "grad_norm": 1.6322238445281982, + "learning_rate": 0.0004259083591579596, + "loss": 4.607, + "step": 1193 + }, + { + "epoch": 0.38185852205720683, + "grad_norm": 2.90915846824646, + "learning_rate": 0.000425626149636343, + "loss": 4.5783, + "step": 1194 + }, + { + "epoch": 0.38217833656479244, + "grad_norm": 1.7546790838241577, + "learning_rate": 0.0004253438052428619, + "loss": 4.5833, + "step": 1195 + }, + { + "epoch": 0.382498151072378, + "grad_norm": 2.3547275066375732, + "learning_rate": 0.00042506132628064016, + "loss": 4.682, + "step": 1196 + }, + { + "epoch": 0.3828179655799636, + "grad_norm": 1.8587547540664673, + "learning_rate": 0.00042477871305294655, + "loss": 4.6728, + "step": 1197 + }, + { + "epoch": 0.38313778008754923, + "grad_norm": 2.6339187622070312, + "learning_rate": 0.0004244959658631938, + "loss": 4.5707, + "step": 1198 + }, + { + "epoch": 0.38345759459513484, + "grad_norm": 1.7075780630111694, + "learning_rate": 0.00042421308501493823, + "loss": 4.5251, + "step": 1199 + }, + { + "epoch": 0.3837774091027204, + "grad_norm": 1.5770255327224731, + "learning_rate": 0.0004239300708118802, + "loss": 4.5879, + "step": 1200 + }, + { + "epoch": 0.3837774091027204, + "eval_runtime": 21.1674, + "eval_samples_per_second": 89.619, + "eval_steps_per_second": 22.44, + "step": 1200 + }, + { + "epoch": 0.384097223610306, + "grad_norm": 1.0582361221313477, + "learning_rate": 0.0004236469235578627, + "loss": 4.6028, + "step": 1201 + }, + { + "epoch": 0.38441703811789163, + "grad_norm": 2.770181894302368, + "learning_rate": 0.0004233636435568719, + "loss": 4.6377, + "step": 1202 + }, + { + "epoch": 0.38473685262547724, + "grad_norm": 1.4466156959533691, + "learning_rate": 0.00042308023111303636, + "loss": 4.6109, + "step": 1203 + }, + { + "epoch": 0.3850566671330628, + "grad_norm": 15.550615310668945, + "learning_rate": 0.00042279668653062686, + "loss": 4.6517, + "step": 1204 + }, + { + "epoch": 0.3853764816406484, + "grad_norm": 2.696341037750244, + "learning_rate": 0.0004225130101140559, + "loss": 4.5965, + "step": 1205 + }, + { + "epoch": 0.38569629614823403, + "grad_norm": 2.713088035583496, + "learning_rate": 0.00042222920216787786, + "loss": 4.6104, + "step": 1206 + }, + { + "epoch": 0.38601611065581964, + "grad_norm": 1.6078623533248901, + "learning_rate": 0.000421945262996788, + "loss": 4.6546, + "step": 1207 + }, + { + "epoch": 0.3863359251634052, + "grad_norm": 9.1212739944458, + "learning_rate": 0.0004216611929056225, + "loss": 4.627, + "step": 1208 + }, + { + "epoch": 0.3866557396709908, + "grad_norm": 5.938100337982178, + "learning_rate": 0.0004213769921993583, + "loss": 4.6713, + "step": 1209 + }, + { + "epoch": 0.38697555417857643, + "grad_norm": 1.561599850654602, + "learning_rate": 0.0004210926611831124, + "loss": 4.4964, + "step": 1210 + }, + { + "epoch": 0.38729536868616204, + "grad_norm": 3.133498430252075, + "learning_rate": 0.0004208082001621417, + "loss": 4.6404, + "step": 1211 + }, + { + "epoch": 0.3876151831937476, + "grad_norm": 1.5157006978988647, + "learning_rate": 0.0004205236094418428, + "loss": 4.6688, + "step": 1212 + }, + { + "epoch": 0.3879349977013332, + "grad_norm": 2.1215054988861084, + "learning_rate": 0.0004202388893277515, + "loss": 4.5977, + "step": 1213 + }, + { + "epoch": 0.38825481220891883, + "grad_norm": 8.695862770080566, + "learning_rate": 0.00041995404012554226, + "loss": 4.5622, + "step": 1214 + }, + { + "epoch": 0.38857462671650445, + "grad_norm": 1.446882963180542, + "learning_rate": 0.0004196690621410285, + "loss": 4.5854, + "step": 1215 + }, + { + "epoch": 0.38889444122409, + "grad_norm": 1.185651183128357, + "learning_rate": 0.0004193839556801617, + "loss": 4.6127, + "step": 1216 + }, + { + "epoch": 0.3892142557316756, + "grad_norm": 1.1880130767822266, + "learning_rate": 0.0004190987210490314, + "loss": 4.6379, + "step": 1217 + }, + { + "epoch": 0.38953407023926123, + "grad_norm": 3.5236897468566895, + "learning_rate": 0.00041881335855386463, + "loss": 4.6355, + "step": 1218 + }, + { + "epoch": 0.38985388474684685, + "grad_norm": 3.3104519844055176, + "learning_rate": 0.00041852786850102557, + "loss": 4.5993, + "step": 1219 + }, + { + "epoch": 0.3901736992544324, + "grad_norm": 1.5470588207244873, + "learning_rate": 0.00041824225119701576, + "loss": 4.6652, + "step": 1220 + }, + { + "epoch": 0.390493513762018, + "grad_norm": 3.452380657196045, + "learning_rate": 0.0004179565069484729, + "loss": 4.5941, + "step": 1221 + }, + { + "epoch": 0.39081332826960363, + "grad_norm": 3.133631467819214, + "learning_rate": 0.0004176706360621713, + "loss": 4.5714, + "step": 1222 + }, + { + "epoch": 0.39113314277718925, + "grad_norm": 1.7067134380340576, + "learning_rate": 0.0004173846388450209, + "loss": 4.6097, + "step": 1223 + }, + { + "epoch": 0.3914529572847748, + "grad_norm": 9.706552505493164, + "learning_rate": 0.0004170985156040677, + "loss": 4.5979, + "step": 1224 + }, + { + "epoch": 0.3917727717923604, + "grad_norm": 9.765279769897461, + "learning_rate": 0.0004168122666464927, + "loss": 4.6548, + "step": 1225 + }, + { + "epoch": 0.39209258629994603, + "grad_norm": 13.520505905151367, + "learning_rate": 0.0004165258922796119, + "loss": 4.6996, + "step": 1226 + }, + { + "epoch": 0.39241240080753165, + "grad_norm": 1.7092180252075195, + "learning_rate": 0.00041623939281087605, + "loss": 4.6495, + "step": 1227 + }, + { + "epoch": 0.3927322153151172, + "grad_norm": 4.84594202041626, + "learning_rate": 0.00041595276854787007, + "loss": 4.5388, + "step": 1228 + }, + { + "epoch": 0.3930520298227028, + "grad_norm": 2.837066888809204, + "learning_rate": 0.00041566601979831287, + "loss": 4.6748, + "step": 1229 + }, + { + "epoch": 0.39337184433028843, + "grad_norm": 15.487530708312988, + "learning_rate": 0.00041537914687005714, + "loss": 4.698, + "step": 1230 + }, + { + "epoch": 0.39369165883787405, + "grad_norm": 2.7634496688842773, + "learning_rate": 0.00041509215007108885, + "loss": 4.6891, + "step": 1231 + }, + { + "epoch": 0.3940114733454596, + "grad_norm": 2.6953999996185303, + "learning_rate": 0.0004148050297095269, + "loss": 4.6608, + "step": 1232 + }, + { + "epoch": 0.3943312878530452, + "grad_norm": 2.3132693767547607, + "learning_rate": 0.00041451778609362286, + "loss": 4.6588, + "step": 1233 + }, + { + "epoch": 0.39465110236063083, + "grad_norm": 1.9645071029663086, + "learning_rate": 0.0004142304195317605, + "loss": 4.5915, + "step": 1234 + }, + { + "epoch": 0.39497091686821645, + "grad_norm": 1.9759736061096191, + "learning_rate": 0.00041394293033245597, + "loss": 4.5558, + "step": 1235 + }, + { + "epoch": 0.39529073137580206, + "grad_norm": 6.738734722137451, + "learning_rate": 0.00041365531880435647, + "loss": 4.6252, + "step": 1236 + }, + { + "epoch": 0.3956105458833876, + "grad_norm": 3.076261043548584, + "learning_rate": 0.0004133675852562413, + "loss": 4.7313, + "step": 1237 + }, + { + "epoch": 0.39593036039097323, + "grad_norm": 24.047224044799805, + "learning_rate": 0.00041307972999702014, + "loss": 4.5665, + "step": 1238 + }, + { + "epoch": 0.39625017489855885, + "grad_norm": 2.6592719554901123, + "learning_rate": 0.00041279175333573345, + "loss": 4.6946, + "step": 1239 + }, + { + "epoch": 0.39656998940614446, + "grad_norm": 2.6919350624084473, + "learning_rate": 0.00041250365558155236, + "loss": 4.7171, + "step": 1240 + }, + { + "epoch": 0.39688980391373, + "grad_norm": 1.3404223918914795, + "learning_rate": 0.0004122154370437776, + "loss": 4.6246, + "step": 1241 + }, + { + "epoch": 0.39720961842131564, + "grad_norm": 2.085395574569702, + "learning_rate": 0.0004119270980318398, + "loss": 4.6182, + "step": 1242 + }, + { + "epoch": 0.39752943292890125, + "grad_norm": 3.020949363708496, + "learning_rate": 0.0004116386388552988, + "loss": 4.6469, + "step": 1243 + }, + { + "epoch": 0.39784924743648686, + "grad_norm": 2.9341249465942383, + "learning_rate": 0.0004113500598238437, + "loss": 4.6394, + "step": 1244 + }, + { + "epoch": 0.3981690619440724, + "grad_norm": 1.2419863939285278, + "learning_rate": 0.000411061361247292, + "loss": 4.6336, + "step": 1245 + }, + { + "epoch": 0.39848887645165804, + "grad_norm": 1.639393925666809, + "learning_rate": 0.00041077254343558955, + "loss": 4.6126, + "step": 1246 + }, + { + "epoch": 0.39880869095924365, + "grad_norm": 2.290335178375244, + "learning_rate": 0.00041048360669881055, + "loss": 4.6729, + "step": 1247 + }, + { + "epoch": 0.39912850546682926, + "grad_norm": 2.977069616317749, + "learning_rate": 0.0004101945513471563, + "loss": 4.6031, + "step": 1248 + }, + { + "epoch": 0.3994483199744148, + "grad_norm": 2.796233654022217, + "learning_rate": 0.000409905377690956, + "loss": 4.635, + "step": 1249 + }, + { + "epoch": 0.39976813448200044, + "grad_norm": 5.325570106506348, + "learning_rate": 0.00040961608604066566, + "loss": 4.6497, + "step": 1250 + }, + { + "epoch": 0.40008794898958605, + "grad_norm": 1.053239345550537, + "learning_rate": 0.0004093266767068677, + "loss": 4.5559, + "step": 1251 + }, + { + "epoch": 0.40040776349717166, + "grad_norm": 1.3334789276123047, + "learning_rate": 0.0004090371500002715, + "loss": 4.559, + "step": 1252 + }, + { + "epoch": 0.4007275780047572, + "grad_norm": 1.8353711366653442, + "learning_rate": 0.00040874750623171176, + "loss": 4.529, + "step": 1253 + }, + { + "epoch": 0.40104739251234284, + "grad_norm": 1.1362560987472534, + "learning_rate": 0.00040845774571214924, + "loss": 4.6226, + "step": 1254 + }, + { + "epoch": 0.40136720701992845, + "grad_norm": 2.0161991119384766, + "learning_rate": 0.0004081678687526701, + "loss": 4.5871, + "step": 1255 + }, + { + "epoch": 0.40168702152751407, + "grad_norm": 2.166210651397705, + "learning_rate": 0.0004078778756644854, + "loss": 4.663, + "step": 1256 + }, + { + "epoch": 0.4020068360350996, + "grad_norm": 1.401020884513855, + "learning_rate": 0.00040758776675893065, + "loss": 4.6182, + "step": 1257 + }, + { + "epoch": 0.40232665054268524, + "grad_norm": 1.8439130783081055, + "learning_rate": 0.00040729754234746613, + "loss": 4.5799, + "step": 1258 + }, + { + "epoch": 0.40264646505027085, + "grad_norm": 1.5867342948913574, + "learning_rate": 0.0004070072027416758, + "loss": 4.6129, + "step": 1259 + }, + { + "epoch": 0.40296627955785647, + "grad_norm": 5.7303900718688965, + "learning_rate": 0.00040671674825326745, + "loss": 4.5891, + "step": 1260 + }, + { + "epoch": 0.403286094065442, + "grad_norm": 1.617805004119873, + "learning_rate": 0.0004064261791940723, + "loss": 4.576, + "step": 1261 + }, + { + "epoch": 0.40360590857302764, + "grad_norm": 1.9642220735549927, + "learning_rate": 0.0004061354958760441, + "loss": 4.6732, + "step": 1262 + }, + { + "epoch": 0.40392572308061325, + "grad_norm": 3.112950086593628, + "learning_rate": 0.00040584469861126, + "loss": 4.5684, + "step": 1263 + }, + { + "epoch": 0.40424553758819887, + "grad_norm": 3.7484447956085205, + "learning_rate": 0.00040555378771191876, + "loss": 4.5692, + "step": 1264 + }, + { + "epoch": 0.4045653520957844, + "grad_norm": 2.3384182453155518, + "learning_rate": 0.0004052627634903417, + "loss": 4.6671, + "step": 1265 + }, + { + "epoch": 0.40488516660337004, + "grad_norm": 2.6692070960998535, + "learning_rate": 0.00040497162625897134, + "loss": 4.6274, + "step": 1266 + }, + { + "epoch": 0.40520498111095565, + "grad_norm": 1.0335692167282104, + "learning_rate": 0.00040468037633037196, + "loss": 4.5424, + "step": 1267 + }, + { + "epoch": 0.40552479561854127, + "grad_norm": 2.268076181411743, + "learning_rate": 0.0004043890140172286, + "loss": 4.5476, + "step": 1268 + }, + { + "epoch": 0.4058446101261268, + "grad_norm": 1.315422534942627, + "learning_rate": 0.00040409753963234675, + "loss": 4.6448, + "step": 1269 + }, + { + "epoch": 0.40616442463371244, + "grad_norm": 2.364116907119751, + "learning_rate": 0.00040380595348865286, + "loss": 4.5256, + "step": 1270 + }, + { + "epoch": 0.40648423914129805, + "grad_norm": 3.5734450817108154, + "learning_rate": 0.00040351425589919257, + "loss": 4.6538, + "step": 1271 + }, + { + "epoch": 0.40680405364888367, + "grad_norm": 2.4196889400482178, + "learning_rate": 0.0004032224471771317, + "loss": 4.6079, + "step": 1272 + }, + { + "epoch": 0.4071238681564692, + "grad_norm": 1.8854360580444336, + "learning_rate": 0.00040293052763575537, + "loss": 4.5651, + "step": 1273 + }, + { + "epoch": 0.40744368266405484, + "grad_norm": 8.096748352050781, + "learning_rate": 0.0004026384975884673, + "loss": 4.5938, + "step": 1274 + }, + { + "epoch": 0.40776349717164045, + "grad_norm": 2.198469877243042, + "learning_rate": 0.00040234635734879036, + "loss": 4.6575, + "step": 1275 + }, + { + "epoch": 0.40808331167922607, + "grad_norm": 4.262251853942871, + "learning_rate": 0.00040205410723036526, + "loss": 4.5719, + "step": 1276 + }, + { + "epoch": 0.4084031261868116, + "grad_norm": 4.23693323135376, + "learning_rate": 0.0004017617475469508, + "loss": 4.5964, + "step": 1277 + }, + { + "epoch": 0.40872294069439724, + "grad_norm": 1.5615571737289429, + "learning_rate": 0.00040146927861242366, + "loss": 4.5331, + "step": 1278 + }, + { + "epoch": 0.40904275520198285, + "grad_norm": 3.6274046897888184, + "learning_rate": 0.00040117670074077747, + "loss": 4.619, + "step": 1279 + }, + { + "epoch": 0.40936256970956847, + "grad_norm": 1.6320995092391968, + "learning_rate": 0.00040088401424612317, + "loss": 4.5858, + "step": 1280 + }, + { + "epoch": 0.409682384217154, + "grad_norm": 2.6233675479888916, + "learning_rate": 0.000400591219442688, + "loss": 4.5414, + "step": 1281 + }, + { + "epoch": 0.41000219872473964, + "grad_norm": 1.7418190240859985, + "learning_rate": 0.0004002983166448155, + "loss": 4.5834, + "step": 1282 + }, + { + "epoch": 0.41032201323232526, + "grad_norm": 2.7616779804229736, + "learning_rate": 0.0004000053061669654, + "loss": 4.5902, + "step": 1283 + }, + { + "epoch": 0.41064182773991087, + "grad_norm": 2.7519638538360596, + "learning_rate": 0.00039971218832371284, + "loss": 4.6328, + "step": 1284 + }, + { + "epoch": 0.41096164224749643, + "grad_norm": 1.2930335998535156, + "learning_rate": 0.0003994189634297483, + "loss": 4.4931, + "step": 1285 + }, + { + "epoch": 0.41128145675508204, + "grad_norm": 1.559300184249878, + "learning_rate": 0.00039912563179987713, + "loss": 4.691, + "step": 1286 + }, + { + "epoch": 0.41160127126266766, + "grad_norm": 1.8185760974884033, + "learning_rate": 0.00039883219374901933, + "loss": 4.5544, + "step": 1287 + }, + { + "epoch": 0.41192108577025327, + "grad_norm": 1.7445274591445923, + "learning_rate": 0.0003985386495922091, + "loss": 4.5618, + "step": 1288 + }, + { + "epoch": 0.41224090027783883, + "grad_norm": 2.098858594894409, + "learning_rate": 0.00039824499964459455, + "loss": 4.5741, + "step": 1289 + }, + { + "epoch": 0.41256071478542444, + "grad_norm": 2.0627784729003906, + "learning_rate": 0.00039795124422143746, + "loss": 4.5533, + "step": 1290 + }, + { + "epoch": 0.41288052929301006, + "grad_norm": 3.861133575439453, + "learning_rate": 0.0003976573836381128, + "loss": 4.6023, + "step": 1291 + }, + { + "epoch": 0.41320034380059567, + "grad_norm": 1.6284693479537964, + "learning_rate": 0.00039736341821010833, + "loss": 4.6094, + "step": 1292 + }, + { + "epoch": 0.41352015830818123, + "grad_norm": 1.714142084121704, + "learning_rate": 0.0003970693482530247, + "loss": 4.569, + "step": 1293 + }, + { + "epoch": 0.41383997281576684, + "grad_norm": 1.4243638515472412, + "learning_rate": 0.00039677517408257424, + "loss": 4.5513, + "step": 1294 + }, + { + "epoch": 0.41415978732335246, + "grad_norm": 1.6191344261169434, + "learning_rate": 0.00039648089601458165, + "loss": 4.5597, + "step": 1295 + }, + { + "epoch": 0.41447960183093807, + "grad_norm": 2.0909595489501953, + "learning_rate": 0.000396186514364983, + "loss": 4.6551, + "step": 1296 + }, + { + "epoch": 0.41479941633852363, + "grad_norm": 1.9448293447494507, + "learning_rate": 0.0003958920294498255, + "loss": 4.621, + "step": 1297 + }, + { + "epoch": 0.41511923084610924, + "grad_norm": 1.9645379781723022, + "learning_rate": 0.00039559744158526735, + "loss": 4.5759, + "step": 1298 + }, + { + "epoch": 0.41543904535369486, + "grad_norm": 14.529280662536621, + "learning_rate": 0.0003953027510875772, + "loss": 4.555, + "step": 1299 + }, + { + "epoch": 0.41575885986128047, + "grad_norm": 1.3694531917572021, + "learning_rate": 0.0003950079582731339, + "loss": 4.63, + "step": 1300 + }, + { + "epoch": 0.41575885986128047, + "eval_runtime": 21.8952, + "eval_samples_per_second": 86.64, + "eval_steps_per_second": 21.694, + "step": 1300 + }, + { + "epoch": 0.41607867436886603, + "grad_norm": 3.6136231422424316, + "learning_rate": 0.0003947130634584261, + "loss": 4.5868, + "step": 1301 + }, + { + "epoch": 0.41639848887645164, + "grad_norm": 1.9168570041656494, + "learning_rate": 0.000394418066960052, + "loss": 4.6024, + "step": 1302 + }, + { + "epoch": 0.41671830338403726, + "grad_norm": 2.9139790534973145, + "learning_rate": 0.00039412296909471914, + "loss": 4.5634, + "step": 1303 + }, + { + "epoch": 0.4170381178916229, + "grad_norm": 2.778794765472412, + "learning_rate": 0.00039382777017924354, + "loss": 4.5892, + "step": 1304 + }, + { + "epoch": 0.4173579323992085, + "grad_norm": 22.92460060119629, + "learning_rate": 0.00039353247053054984, + "loss": 4.581, + "step": 1305 + }, + { + "epoch": 0.41767774690679405, + "grad_norm": 2.5368330478668213, + "learning_rate": 0.0003932370704656711, + "loss": 4.6262, + "step": 1306 + }, + { + "epoch": 0.41799756141437966, + "grad_norm": 1.7910631895065308, + "learning_rate": 0.00039294157030174783, + "loss": 4.596, + "step": 1307 + }, + { + "epoch": 0.4183173759219653, + "grad_norm": 1.2267634868621826, + "learning_rate": 0.00039264597035602807, + "loss": 4.5242, + "step": 1308 + }, + { + "epoch": 0.4186371904295509, + "grad_norm": 2.650428295135498, + "learning_rate": 0.0003923502709458672, + "loss": 4.5955, + "step": 1309 + }, + { + "epoch": 0.41895700493713645, + "grad_norm": 2.0779755115509033, + "learning_rate": 0.00039205447238872706, + "loss": 4.5817, + "step": 1310 + }, + { + "epoch": 0.41927681944472206, + "grad_norm": 2.8054747581481934, + "learning_rate": 0.0003917585750021763, + "loss": 4.5681, + "step": 1311 + }, + { + "epoch": 0.4195966339523077, + "grad_norm": 2.8588578701019287, + "learning_rate": 0.0003914625791038893, + "loss": 4.5764, + "step": 1312 + }, + { + "epoch": 0.4199164484598933, + "grad_norm": 2.2470176219940186, + "learning_rate": 0.00039116648501164665, + "loss": 4.5394, + "step": 1313 + }, + { + "epoch": 0.42023626296747885, + "grad_norm": 1.701889157295227, + "learning_rate": 0.0003908702930433338, + "loss": 4.581, + "step": 1314 + }, + { + "epoch": 0.42055607747506446, + "grad_norm": 1.182225227355957, + "learning_rate": 0.0003905740035169417, + "loss": 4.5506, + "step": 1315 + }, + { + "epoch": 0.4208758919826501, + "grad_norm": 6.077612400054932, + "learning_rate": 0.00039027761675056595, + "loss": 4.5271, + "step": 1316 + }, + { + "epoch": 0.4211957064902357, + "grad_norm": 1.1774729490280151, + "learning_rate": 0.0003899811330624065, + "loss": 4.5342, + "step": 1317 + }, + { + "epoch": 0.42151552099782125, + "grad_norm": 5.805182933807373, + "learning_rate": 0.0003896845527707673, + "loss": 4.5499, + "step": 1318 + }, + { + "epoch": 0.42183533550540686, + "grad_norm": 1.5339539051055908, + "learning_rate": 0.00038938787619405616, + "loss": 4.6104, + "step": 1319 + }, + { + "epoch": 0.4221551500129925, + "grad_norm": 2.2717056274414062, + "learning_rate": 0.00038909110365078413, + "loss": 4.5523, + "step": 1320 + }, + { + "epoch": 0.4224749645205781, + "grad_norm": 1.6891810894012451, + "learning_rate": 0.00038879423545956534, + "loss": 4.4945, + "step": 1321 + }, + { + "epoch": 0.42279477902816365, + "grad_norm": 2.0225696563720703, + "learning_rate": 0.00038849727193911664, + "loss": 4.562, + "step": 1322 + }, + { + "epoch": 0.42311459353574926, + "grad_norm": 2.4896163940429688, + "learning_rate": 0.0003882002134082571, + "loss": 4.5822, + "step": 1323 + }, + { + "epoch": 0.4234344080433349, + "grad_norm": 9.039996147155762, + "learning_rate": 0.000387903060185908, + "loss": 4.5495, + "step": 1324 + }, + { + "epoch": 0.4237542225509205, + "grad_norm": 2.7000205516815186, + "learning_rate": 0.00038760581259109214, + "loss": 4.5874, + "step": 1325 + }, + { + "epoch": 0.42407403705850605, + "grad_norm": 1.9824517965316772, + "learning_rate": 0.0003873084709429336, + "loss": 4.6034, + "step": 1326 + }, + { + "epoch": 0.42439385156609166, + "grad_norm": 3.1104137897491455, + "learning_rate": 0.00038701103556065754, + "loss": 4.6496, + "step": 1327 + }, + { + "epoch": 0.4247136660736773, + "grad_norm": 2.9838178157806396, + "learning_rate": 0.0003867135067635898, + "loss": 4.5984, + "step": 1328 + }, + { + "epoch": 0.4250334805812629, + "grad_norm": 1.6490141153335571, + "learning_rate": 0.0003864158848711562, + "loss": 4.5605, + "step": 1329 + }, + { + "epoch": 0.42535329508884845, + "grad_norm": 2.605144739151001, + "learning_rate": 0.000386118170202883, + "loss": 4.559, + "step": 1330 + }, + { + "epoch": 0.42567310959643406, + "grad_norm": 10.05894947052002, + "learning_rate": 0.00038582036307839557, + "loss": 4.6163, + "step": 1331 + }, + { + "epoch": 0.4259929241040197, + "grad_norm": 3.559980630874634, + "learning_rate": 0.00038552246381741884, + "loss": 4.5294, + "step": 1332 + }, + { + "epoch": 0.4263127386116053, + "grad_norm": 1.5636258125305176, + "learning_rate": 0.0003852244727397766, + "loss": 4.6664, + "step": 1333 + }, + { + "epoch": 0.42663255311919085, + "grad_norm": 1.5836095809936523, + "learning_rate": 0.00038492639016539116, + "loss": 4.5886, + "step": 1334 + }, + { + "epoch": 0.42695236762677646, + "grad_norm": 1.5852701663970947, + "learning_rate": 0.0003846282164142831, + "loss": 4.6733, + "step": 1335 + }, + { + "epoch": 0.4272721821343621, + "grad_norm": 1.7198234796524048, + "learning_rate": 0.00038432995180657094, + "loss": 4.5234, + "step": 1336 + }, + { + "epoch": 0.4275919966419477, + "grad_norm": 7.945826530456543, + "learning_rate": 0.00038403159666247063, + "loss": 4.5379, + "step": 1337 + }, + { + "epoch": 0.42791181114953325, + "grad_norm": 2.3104543685913086, + "learning_rate": 0.0003837331513022954, + "loss": 4.5506, + "step": 1338 + }, + { + "epoch": 0.42823162565711886, + "grad_norm": 2.2884445190429688, + "learning_rate": 0.0003834346160464553, + "loss": 4.5828, + "step": 1339 + }, + { + "epoch": 0.4285514401647045, + "grad_norm": 1.9064819812774658, + "learning_rate": 0.0003831359912154569, + "loss": 4.5653, + "step": 1340 + }, + { + "epoch": 0.4288712546722901, + "grad_norm": 1.7079178094863892, + "learning_rate": 0.0003828372771299029, + "loss": 4.4772, + "step": 1341 + }, + { + "epoch": 0.42919106917987565, + "grad_norm": 3.277074098587036, + "learning_rate": 0.00038253847411049194, + "loss": 4.4587, + "step": 1342 + }, + { + "epoch": 0.42951088368746126, + "grad_norm": 1.3470120429992676, + "learning_rate": 0.000382239582478018, + "loss": 4.631, + "step": 1343 + }, + { + "epoch": 0.4298306981950469, + "grad_norm": 5.253669738769531, + "learning_rate": 0.00038194060255337026, + "loss": 4.6436, + "step": 1344 + }, + { + "epoch": 0.4301505127026325, + "grad_norm": 1.9300462007522583, + "learning_rate": 0.0003816415346575327, + "loss": 4.5796, + "step": 1345 + }, + { + "epoch": 0.43047032721021805, + "grad_norm": 13.181463241577148, + "learning_rate": 0.0003813423791115838, + "loss": 4.4692, + "step": 1346 + }, + { + "epoch": 0.43079014171780367, + "grad_norm": 3.646790027618408, + "learning_rate": 0.00038104313623669604, + "loss": 4.6115, + "step": 1347 + }, + { + "epoch": 0.4311099562253893, + "grad_norm": 1.2707982063293457, + "learning_rate": 0.0003807438063541356, + "loss": 4.5766, + "step": 1348 + }, + { + "epoch": 0.4314297707329749, + "grad_norm": 1.7788244485855103, + "learning_rate": 0.00038044438978526235, + "loss": 4.5987, + "step": 1349 + }, + { + "epoch": 0.43174958524056045, + "grad_norm": 2.680772066116333, + "learning_rate": 0.0003801448868515287, + "loss": 4.5396, + "step": 1350 + }, + { + "epoch": 0.43206939974814607, + "grad_norm": 1.161279320716858, + "learning_rate": 0.00037984529787448047, + "loss": 4.5482, + "step": 1351 + }, + { + "epoch": 0.4323892142557317, + "grad_norm": 3.4628868103027344, + "learning_rate": 0.0003795456231757554, + "loss": 4.5581, + "step": 1352 + }, + { + "epoch": 0.4327090287633173, + "grad_norm": 3.8042807579040527, + "learning_rate": 0.0003792458630770833, + "loss": 4.5563, + "step": 1353 + }, + { + "epoch": 0.43302884327090285, + "grad_norm": 1.346595048904419, + "learning_rate": 0.00037894601790028576, + "loss": 4.6433, + "step": 1354 + }, + { + "epoch": 0.43334865777848847, + "grad_norm": 1.1366397142410278, + "learning_rate": 0.0003786460879672756, + "loss": 4.5367, + "step": 1355 + }, + { + "epoch": 0.4336684722860741, + "grad_norm": 1.7212153673171997, + "learning_rate": 0.0003783460736000569, + "loss": 4.46, + "step": 1356 + }, + { + "epoch": 0.4339882867936597, + "grad_norm": 1.374093770980835, + "learning_rate": 0.0003780459751207241, + "loss": 4.5627, + "step": 1357 + }, + { + "epoch": 0.43430810130124525, + "grad_norm": 1.2930073738098145, + "learning_rate": 0.0003777457928514619, + "loss": 4.5804, + "step": 1358 + }, + { + "epoch": 0.43462791580883087, + "grad_norm": 1.3635977506637573, + "learning_rate": 0.0003774455271145454, + "loss": 4.4743, + "step": 1359 + }, + { + "epoch": 0.4349477303164165, + "grad_norm": 2.5675625801086426, + "learning_rate": 0.0003771451782323388, + "loss": 4.5414, + "step": 1360 + }, + { + "epoch": 0.4352675448240021, + "grad_norm": 2.484837293624878, + "learning_rate": 0.0003768447465272959, + "loss": 4.5197, + "step": 1361 + }, + { + "epoch": 0.43558735933158765, + "grad_norm": 1.443600058555603, + "learning_rate": 0.0003765442323219591, + "loss": 4.5458, + "step": 1362 + }, + { + "epoch": 0.43590717383917327, + "grad_norm": 1.2279186248779297, + "learning_rate": 0.00037624363593895976, + "loss": 4.5437, + "step": 1363 + }, + { + "epoch": 0.4362269883467589, + "grad_norm": 1.2826124429702759, + "learning_rate": 0.00037594295770101716, + "loss": 4.5714, + "step": 1364 + }, + { + "epoch": 0.4365468028543445, + "grad_norm": 2.0542068481445312, + "learning_rate": 0.0003756421979309387, + "loss": 4.5703, + "step": 1365 + }, + { + "epoch": 0.43686661736193005, + "grad_norm": 2.848710060119629, + "learning_rate": 0.00037534135695161904, + "loss": 4.5307, + "step": 1366 + }, + { + "epoch": 0.43718643186951567, + "grad_norm": 1.9498159885406494, + "learning_rate": 0.0003750404350860402, + "loss": 4.5644, + "step": 1367 + }, + { + "epoch": 0.4375062463771013, + "grad_norm": 1.0984961986541748, + "learning_rate": 0.00037473943265727114, + "loss": 4.5832, + "step": 1368 + }, + { + "epoch": 0.4378260608846869, + "grad_norm": 1.7323179244995117, + "learning_rate": 0.000374438349988467, + "loss": 4.5796, + "step": 1369 + }, + { + "epoch": 0.43814587539227245, + "grad_norm": 1.7928155660629272, + "learning_rate": 0.00037413718740286935, + "loss": 4.4952, + "step": 1370 + }, + { + "epoch": 0.43846568989985807, + "grad_norm": 1.052549123764038, + "learning_rate": 0.00037383594522380546, + "loss": 4.5612, + "step": 1371 + }, + { + "epoch": 0.4387855044074437, + "grad_norm": 2.0341806411743164, + "learning_rate": 0.00037353462377468806, + "loss": 4.5977, + "step": 1372 + }, + { + "epoch": 0.4391053189150293, + "grad_norm": 1.56754469871521, + "learning_rate": 0.0003732332233790149, + "loss": 4.4942, + "step": 1373 + }, + { + "epoch": 0.4394251334226149, + "grad_norm": 2.1558115482330322, + "learning_rate": 0.00037293174436036855, + "loss": 4.6851, + "step": 1374 + }, + { + "epoch": 0.43974494793020047, + "grad_norm": 4.067927837371826, + "learning_rate": 0.000372630187042416, + "loss": 4.6448, + "step": 1375 + }, + { + "epoch": 0.4400647624377861, + "grad_norm": 2.1450400352478027, + "learning_rate": 0.0003723285517489084, + "loss": 4.57, + "step": 1376 + }, + { + "epoch": 0.4403845769453717, + "grad_norm": 2.9361460208892822, + "learning_rate": 0.0003720268388036805, + "loss": 4.5305, + "step": 1377 + }, + { + "epoch": 0.4407043914529573, + "grad_norm": 1.263005018234253, + "learning_rate": 0.0003717250485306503, + "loss": 4.6151, + "step": 1378 + }, + { + "epoch": 0.44102420596054287, + "grad_norm": 6.089189529418945, + "learning_rate": 0.00037142318125381915, + "loss": 4.5891, + "step": 1379 + }, + { + "epoch": 0.4413440204681285, + "grad_norm": 10.544564247131348, + "learning_rate": 0.0003711212372972706, + "loss": 4.5569, + "step": 1380 + }, + { + "epoch": 0.4416638349757141, + "grad_norm": 2.998872995376587, + "learning_rate": 0.000370819216985171, + "loss": 4.6647, + "step": 1381 + }, + { + "epoch": 0.4419836494832997, + "grad_norm": 2.028391122817993, + "learning_rate": 0.0003705171206417685, + "loss": 4.5831, + "step": 1382 + }, + { + "epoch": 0.44230346399088527, + "grad_norm": 1.676659107208252, + "learning_rate": 0.0003702149485913926, + "loss": 4.4987, + "step": 1383 + }, + { + "epoch": 0.4426232784984709, + "grad_norm": 1.9588249921798706, + "learning_rate": 0.0003699127011584546, + "loss": 4.6079, + "step": 1384 + }, + { + "epoch": 0.4429430930060565, + "grad_norm": 3.512953519821167, + "learning_rate": 0.0003696103786674463, + "loss": 4.5823, + "step": 1385 + }, + { + "epoch": 0.4432629075136421, + "grad_norm": 3.9054934978485107, + "learning_rate": 0.0003693079814429403, + "loss": 4.62, + "step": 1386 + }, + { + "epoch": 0.44358272202122767, + "grad_norm": 2.096737861633301, + "learning_rate": 0.00036900550980958934, + "loss": 4.4982, + "step": 1387 + }, + { + "epoch": 0.4439025365288133, + "grad_norm": 1.5459403991699219, + "learning_rate": 0.000368702964092126, + "loss": 4.5093, + "step": 1388 + }, + { + "epoch": 0.4442223510363989, + "grad_norm": 4.902289390563965, + "learning_rate": 0.0003684003446153627, + "loss": 4.5377, + "step": 1389 + }, + { + "epoch": 0.4445421655439845, + "grad_norm": 2.3250067234039307, + "learning_rate": 0.0003680976517041905, + "loss": 4.4883, + "step": 1390 + }, + { + "epoch": 0.44486198005157007, + "grad_norm": 2.3967413902282715, + "learning_rate": 0.00036779488568358, + "loss": 4.5004, + "step": 1391 + }, + { + "epoch": 0.4451817945591557, + "grad_norm": 2.3411030769348145, + "learning_rate": 0.00036749204687857955, + "loss": 4.4746, + "step": 1392 + }, + { + "epoch": 0.4455016090667413, + "grad_norm": 7.2375922203063965, + "learning_rate": 0.00036718913561431613, + "loss": 4.5311, + "step": 1393 + }, + { + "epoch": 0.4458214235743269, + "grad_norm": 12.489344596862793, + "learning_rate": 0.0003668861522159945, + "loss": 4.4759, + "step": 1394 + }, + { + "epoch": 0.4461412380819125, + "grad_norm": 7.700294017791748, + "learning_rate": 0.00036658309700889655, + "loss": 4.5503, + "step": 1395 + }, + { + "epoch": 0.4464610525894981, + "grad_norm": 1.733701467514038, + "learning_rate": 0.0003662799703183817, + "loss": 4.5612, + "step": 1396 + }, + { + "epoch": 0.4467808670970837, + "grad_norm": 2.111077308654785, + "learning_rate": 0.00036597677246988564, + "loss": 4.5392, + "step": 1397 + }, + { + "epoch": 0.4471006816046693, + "grad_norm": 3.9190855026245117, + "learning_rate": 0.00036567350378892074, + "loss": 4.6047, + "step": 1398 + }, + { + "epoch": 0.4474204961122549, + "grad_norm": 7.627423286437988, + "learning_rate": 0.00036537016460107545, + "loss": 4.5841, + "step": 1399 + }, + { + "epoch": 0.4477403106198405, + "grad_norm": 1.8391305208206177, + "learning_rate": 0.00036506675523201385, + "loss": 4.6372, + "step": 1400 + }, + { + "epoch": 0.4477403106198405, + "eval_runtime": 21.2469, + "eval_samples_per_second": 89.283, + "eval_steps_per_second": 22.356, + "step": 1400 + }, + { + "epoch": 0.4480601251274261, + "grad_norm": 1.9108299016952515, + "learning_rate": 0.0003647632760074751, + "loss": 4.548, + "step": 1401 + }, + { + "epoch": 0.4483799396350117, + "grad_norm": 3.5664114952087402, + "learning_rate": 0.0003644597272532739, + "loss": 4.5832, + "step": 1402 + }, + { + "epoch": 0.4486997541425973, + "grad_norm": 4.394417762756348, + "learning_rate": 0.00036415610929529913, + "loss": 4.5901, + "step": 1403 + }, + { + "epoch": 0.4490195686501829, + "grad_norm": 1.395249366760254, + "learning_rate": 0.00036385242245951427, + "loss": 4.6247, + "step": 1404 + }, + { + "epoch": 0.4493393831577685, + "grad_norm": 4.6826605796813965, + "learning_rate": 0.0003635486670719564, + "loss": 4.5924, + "step": 1405 + }, + { + "epoch": 0.4496591976653541, + "grad_norm": 2.718470573425293, + "learning_rate": 0.0003632448434587366, + "loss": 4.5286, + "step": 1406 + }, + { + "epoch": 0.4499790121729397, + "grad_norm": 3.0207581520080566, + "learning_rate": 0.00036294095194603905, + "loss": 4.5374, + "step": 1407 + }, + { + "epoch": 0.4502988266805253, + "grad_norm": 1.7212485074996948, + "learning_rate": 0.00036263699286012056, + "loss": 4.5751, + "step": 1408 + }, + { + "epoch": 0.4506186411881109, + "grad_norm": 2.3694756031036377, + "learning_rate": 0.0003623329665273108, + "loss": 4.4721, + "step": 1409 + }, + { + "epoch": 0.4509384556956965, + "grad_norm": 3.130624771118164, + "learning_rate": 0.00036202887327401167, + "loss": 4.5421, + "step": 1410 + }, + { + "epoch": 0.4512582702032821, + "grad_norm": 1.4994175434112549, + "learning_rate": 0.00036172471342669663, + "loss": 4.6084, + "step": 1411 + }, + { + "epoch": 0.4515780847108677, + "grad_norm": 2.9330925941467285, + "learning_rate": 0.000361420487311911, + "loss": 4.5586, + "step": 1412 + }, + { + "epoch": 0.4518978992184533, + "grad_norm": 2.2451465129852295, + "learning_rate": 0.0003611161952562707, + "loss": 4.559, + "step": 1413 + }, + { + "epoch": 0.4522177137260389, + "grad_norm": 1.2707191705703735, + "learning_rate": 0.00036081183758646313, + "loss": 4.5383, + "step": 1414 + }, + { + "epoch": 0.4525375282336245, + "grad_norm": 12.234856605529785, + "learning_rate": 0.00036050741462924563, + "loss": 4.5699, + "step": 1415 + }, + { + "epoch": 0.4528573427412101, + "grad_norm": 5.96612548828125, + "learning_rate": 0.0003602029267114457, + "loss": 4.5376, + "step": 1416 + }, + { + "epoch": 0.4531771572487957, + "grad_norm": 2.2183375358581543, + "learning_rate": 0.000359898374159961, + "loss": 4.5431, + "step": 1417 + }, + { + "epoch": 0.4534969717563813, + "grad_norm": 2.8054633140563965, + "learning_rate": 0.0003595937573017579, + "loss": 4.5747, + "step": 1418 + }, + { + "epoch": 0.4538167862639669, + "grad_norm": 2.3553640842437744, + "learning_rate": 0.00035928907646387234, + "loss": 4.5511, + "step": 1419 + }, + { + "epoch": 0.4541366007715525, + "grad_norm": 2.3520686626434326, + "learning_rate": 0.00035898433197340874, + "loss": 4.5902, + "step": 1420 + }, + { + "epoch": 0.4544564152791381, + "grad_norm": 1.8044930696487427, + "learning_rate": 0.0003586795241575398, + "loss": 4.5776, + "step": 1421 + }, + { + "epoch": 0.4547762297867237, + "grad_norm": 1.8007372617721558, + "learning_rate": 0.00035837465334350637, + "loss": 4.5716, + "step": 1422 + }, + { + "epoch": 0.4550960442943093, + "grad_norm": 1.0908401012420654, + "learning_rate": 0.0003580697198586169, + "loss": 4.5407, + "step": 1423 + }, + { + "epoch": 0.4554158588018949, + "grad_norm": 1.4694181680679321, + "learning_rate": 0.0003577647240302471, + "loss": 4.6537, + "step": 1424 + }, + { + "epoch": 0.4557356733094805, + "grad_norm": 3.163874387741089, + "learning_rate": 0.0003574596661858396, + "loss": 4.4868, + "step": 1425 + }, + { + "epoch": 0.4560554878170661, + "grad_norm": 3.7016303539276123, + "learning_rate": 0.00035715454665290343, + "loss": 4.5075, + "step": 1426 + }, + { + "epoch": 0.4563753023246517, + "grad_norm": 1.5130887031555176, + "learning_rate": 0.0003568493657590142, + "loss": 4.5005, + "step": 1427 + }, + { + "epoch": 0.4566951168322373, + "grad_norm": 5.953606605529785, + "learning_rate": 0.0003565441238318131, + "loss": 4.5842, + "step": 1428 + }, + { + "epoch": 0.4570149313398229, + "grad_norm": 5.794526100158691, + "learning_rate": 0.000356238821199007, + "loss": 4.4569, + "step": 1429 + }, + { + "epoch": 0.4573347458474085, + "grad_norm": 2.003514051437378, + "learning_rate": 0.000355933458188368, + "loss": 4.6042, + "step": 1430 + }, + { + "epoch": 0.4576545603549941, + "grad_norm": 2.4744718074798584, + "learning_rate": 0.00035562803512773284, + "loss": 4.5108, + "step": 1431 + }, + { + "epoch": 0.4579743748625797, + "grad_norm": 1.1637517213821411, + "learning_rate": 0.00035532255234500284, + "loss": 4.5071, + "step": 1432 + }, + { + "epoch": 0.4582941893701653, + "grad_norm": 2.616117238998413, + "learning_rate": 0.0003550170101681434, + "loss": 4.4701, + "step": 1433 + }, + { + "epoch": 0.4586140038777509, + "grad_norm": 1.665345311164856, + "learning_rate": 0.00035471140892518366, + "loss": 4.5298, + "step": 1434 + }, + { + "epoch": 0.4589338183853365, + "grad_norm": 1.419633150100708, + "learning_rate": 0.0003544057489442164, + "loss": 4.5056, + "step": 1435 + }, + { + "epoch": 0.4592536328929221, + "grad_norm": 1.6220885515213013, + "learning_rate": 0.0003541000305533971, + "loss": 4.5873, + "step": 1436 + }, + { + "epoch": 0.4595734474005077, + "grad_norm": 2.549685478210449, + "learning_rate": 0.00035379425408094416, + "loss": 4.6013, + "step": 1437 + }, + { + "epoch": 0.4598932619080933, + "grad_norm": 1.3311550617218018, + "learning_rate": 0.00035348841985513834, + "loss": 4.5457, + "step": 1438 + }, + { + "epoch": 0.4602130764156789, + "grad_norm": 2.6028363704681396, + "learning_rate": 0.00035318252820432236, + "loss": 4.5019, + "step": 1439 + }, + { + "epoch": 0.4605328909232645, + "grad_norm": 1.4333462715148926, + "learning_rate": 0.00035287657945690045, + "loss": 4.5122, + "step": 1440 + }, + { + "epoch": 0.4608527054308501, + "grad_norm": 1.3282840251922607, + "learning_rate": 0.0003525705739413385, + "loss": 4.5591, + "step": 1441 + }, + { + "epoch": 0.4611725199384357, + "grad_norm": 1.5086230039596558, + "learning_rate": 0.000352264511986163, + "loss": 4.4962, + "step": 1442 + }, + { + "epoch": 0.46149233444602134, + "grad_norm": 1.0487247705459595, + "learning_rate": 0.0003519583939199613, + "loss": 4.4663, + "step": 1443 + }, + { + "epoch": 0.4618121489536069, + "grad_norm": 7.277250289916992, + "learning_rate": 0.00035165222007138076, + "loss": 4.5811, + "step": 1444 + }, + { + "epoch": 0.4621319634611925, + "grad_norm": 4.0232157707214355, + "learning_rate": 0.00035134599076912866, + "loss": 4.5198, + "step": 1445 + }, + { + "epoch": 0.4624517779687781, + "grad_norm": 1.7522287368774414, + "learning_rate": 0.00035103970634197193, + "loss": 4.5674, + "step": 1446 + }, + { + "epoch": 0.46277159247636374, + "grad_norm": 12.003565788269043, + "learning_rate": 0.00035073336711873666, + "loss": 4.6262, + "step": 1447 + }, + { + "epoch": 0.4630914069839493, + "grad_norm": 27.776098251342773, + "learning_rate": 0.00035042697342830783, + "loss": 4.57, + "step": 1448 + }, + { + "epoch": 0.4634112214915349, + "grad_norm": 1.4718890190124512, + "learning_rate": 0.00035012052559962853, + "loss": 4.5724, + "step": 1449 + }, + { + "epoch": 0.4637310359991205, + "grad_norm": 1.6918957233428955, + "learning_rate": 0.0003498140239617005, + "loss": 4.5641, + "step": 1450 + }, + { + "epoch": 0.46405085050670614, + "grad_norm": 14.615422248840332, + "learning_rate": 0.0003495074688435829, + "loss": 4.5536, + "step": 1451 + }, + { + "epoch": 0.4643706650142917, + "grad_norm": 2.4006426334381104, + "learning_rate": 0.00034920086057439243, + "loss": 4.5847, + "step": 1452 + }, + { + "epoch": 0.4646904795218773, + "grad_norm": 5.930614948272705, + "learning_rate": 0.0003488941994833028, + "loss": 4.4903, + "step": 1453 + }, + { + "epoch": 0.4650102940294629, + "grad_norm": 2.770338535308838, + "learning_rate": 0.00034858748589954437, + "loss": 4.5087, + "step": 1454 + }, + { + "epoch": 0.46533010853704854, + "grad_norm": 1.4953203201293945, + "learning_rate": 0.0003482807201524042, + "loss": 4.5566, + "step": 1455 + }, + { + "epoch": 0.4656499230446341, + "grad_norm": 1.4480317831039429, + "learning_rate": 0.00034797390257122486, + "loss": 4.5035, + "step": 1456 + }, + { + "epoch": 0.4659697375522197, + "grad_norm": 1.9608365297317505, + "learning_rate": 0.0003476670334854049, + "loss": 4.5778, + "step": 1457 + }, + { + "epoch": 0.4662895520598053, + "grad_norm": 1.4294487237930298, + "learning_rate": 0.00034736011322439796, + "loss": 4.5409, + "step": 1458 + }, + { + "epoch": 0.46660936656739094, + "grad_norm": 1.423373818397522, + "learning_rate": 0.0003470531421177128, + "loss": 4.5103, + "step": 1459 + }, + { + "epoch": 0.4669291810749765, + "grad_norm": 1.8504818677902222, + "learning_rate": 0.00034674612049491276, + "loss": 4.5191, + "step": 1460 + }, + { + "epoch": 0.4672489955825621, + "grad_norm": 7.944705486297607, + "learning_rate": 0.0003464390486856153, + "loss": 4.5549, + "step": 1461 + }, + { + "epoch": 0.4675688100901477, + "grad_norm": 1.6638401746749878, + "learning_rate": 0.0003461319270194919, + "loss": 4.5042, + "step": 1462 + }, + { + "epoch": 0.46788862459773334, + "grad_norm": 2.77482008934021, + "learning_rate": 0.0003458247558262672, + "loss": 4.504, + "step": 1463 + }, + { + "epoch": 0.4682084391053189, + "grad_norm": 5.044100761413574, + "learning_rate": 0.0003455175354357195, + "loss": 4.6068, + "step": 1464 + }, + { + "epoch": 0.4685282536129045, + "grad_norm": 3.7135977745056152, + "learning_rate": 0.0003452102661776798, + "loss": 4.4081, + "step": 1465 + }, + { + "epoch": 0.4688480681204901, + "grad_norm": 2.2570598125457764, + "learning_rate": 0.0003449029483820313, + "loss": 4.4548, + "step": 1466 + }, + { + "epoch": 0.46916788262807574, + "grad_norm": 1.3007919788360596, + "learning_rate": 0.00034459558237870955, + "loss": 4.5403, + "step": 1467 + }, + { + "epoch": 0.4694876971356613, + "grad_norm": 1.646416425704956, + "learning_rate": 0.00034428816849770173, + "loss": 4.5362, + "step": 1468 + }, + { + "epoch": 0.4698075116432469, + "grad_norm": 1.4433326721191406, + "learning_rate": 0.00034398070706904657, + "loss": 4.5197, + "step": 1469 + }, + { + "epoch": 0.4701273261508325, + "grad_norm": 1.4545226097106934, + "learning_rate": 0.0003436731984228336, + "loss": 4.4706, + "step": 1470 + }, + { + "epoch": 0.47044714065841814, + "grad_norm": 1.13702392578125, + "learning_rate": 0.00034336564288920334, + "loss": 4.3837, + "step": 1471 + }, + { + "epoch": 0.4707669551660037, + "grad_norm": 2.1877059936523438, + "learning_rate": 0.0003430580407983465, + "loss": 4.5029, + "step": 1472 + }, + { + "epoch": 0.4710867696735893, + "grad_norm": 1.9040402173995972, + "learning_rate": 0.00034275039248050384, + "loss": 4.448, + "step": 1473 + }, + { + "epoch": 0.4714065841811749, + "grad_norm": 5.9631805419921875, + "learning_rate": 0.00034244269826596543, + "loss": 4.4978, + "step": 1474 + }, + { + "epoch": 0.47172639868876054, + "grad_norm": 3.003028631210327, + "learning_rate": 0.0003421349584850711, + "loss": 4.5614, + "step": 1475 + }, + { + "epoch": 0.4720462131963461, + "grad_norm": 1.8893511295318604, + "learning_rate": 0.0003418271734682093, + "loss": 4.496, + "step": 1476 + }, + { + "epoch": 0.4723660277039317, + "grad_norm": 2.7292380332946777, + "learning_rate": 0.00034151934354581715, + "loss": 4.5546, + "step": 1477 + }, + { + "epoch": 0.4726858422115173, + "grad_norm": 1.8344134092330933, + "learning_rate": 0.00034121146904837995, + "loss": 4.5427, + "step": 1478 + }, + { + "epoch": 0.47300565671910294, + "grad_norm": 2.8705153465270996, + "learning_rate": 0.00034090355030643083, + "loss": 4.5068, + "step": 1479 + }, + { + "epoch": 0.4733254712266885, + "grad_norm": 4.340440273284912, + "learning_rate": 0.00034059558765055047, + "loss": 4.5565, + "step": 1480 + }, + { + "epoch": 0.4736452857342741, + "grad_norm": 2.2909066677093506, + "learning_rate": 0.0003402875814113666, + "loss": 4.4517, + "step": 1481 + }, + { + "epoch": 0.4739651002418597, + "grad_norm": 1.6870324611663818, + "learning_rate": 0.00033997953191955383, + "loss": 4.4569, + "step": 1482 + }, + { + "epoch": 0.47428491474944534, + "grad_norm": 3.3249154090881348, + "learning_rate": 0.0003396714395058333, + "loss": 4.4844, + "step": 1483 + }, + { + "epoch": 0.4746047292570309, + "grad_norm": 6.47374153137207, + "learning_rate": 0.00033936330450097193, + "loss": 4.4912, + "step": 1484 + }, + { + "epoch": 0.4749245437646165, + "grad_norm": 18.88282585144043, + "learning_rate": 0.0003390551272357829, + "loss": 4.5043, + "step": 1485 + }, + { + "epoch": 0.4752443582722021, + "grad_norm": 4.135046482086182, + "learning_rate": 0.00033874690804112397, + "loss": 4.5254, + "step": 1486 + }, + { + "epoch": 0.47556417277978774, + "grad_norm": 73.5289535522461, + "learning_rate": 0.00033843864724789866, + "loss": 4.517, + "step": 1487 + }, + { + "epoch": 0.4758839872873733, + "grad_norm": 1.3389397859573364, + "learning_rate": 0.00033813034518705463, + "loss": 4.5534, + "step": 1488 + }, + { + "epoch": 0.4762038017949589, + "grad_norm": 1.4252009391784668, + "learning_rate": 0.00033782200218958433, + "loss": 4.5173, + "step": 1489 + }, + { + "epoch": 0.47652361630254453, + "grad_norm": 1.5649197101593018, + "learning_rate": 0.00033751361858652375, + "loss": 4.5008, + "step": 1490 + }, + { + "epoch": 0.47684343081013014, + "grad_norm": 1.6209142208099365, + "learning_rate": 0.0003372051947089526, + "loss": 4.5603, + "step": 1491 + }, + { + "epoch": 0.4771632453177157, + "grad_norm": 1.2705239057540894, + "learning_rate": 0.0003368967308879939, + "loss": 4.6102, + "step": 1492 + }, + { + "epoch": 0.4774830598253013, + "grad_norm": 3.956341028213501, + "learning_rate": 0.0003365882274548135, + "loss": 4.447, + "step": 1493 + }, + { + "epoch": 0.47780287433288693, + "grad_norm": 2.7548580169677734, + "learning_rate": 0.00033627968474061966, + "loss": 4.4819, + "step": 1494 + }, + { + "epoch": 0.47812268884047254, + "grad_norm": 3.3609745502471924, + "learning_rate": 0.0003359711030766631, + "loss": 4.5775, + "step": 1495 + }, + { + "epoch": 0.4784425033480581, + "grad_norm": 3.3029165267944336, + "learning_rate": 0.0003356624827942361, + "loss": 4.4915, + "step": 1496 + }, + { + "epoch": 0.4787623178556437, + "grad_norm": 1.7780169248580933, + "learning_rate": 0.00033535382422467255, + "loss": 4.4354, + "step": 1497 + }, + { + "epoch": 0.47908213236322933, + "grad_norm": 1.1975871324539185, + "learning_rate": 0.0003350451276993473, + "loss": 4.5301, + "step": 1498 + }, + { + "epoch": 0.47940194687081494, + "grad_norm": 6.0945611000061035, + "learning_rate": 0.000334736393549676, + "loss": 4.4398, + "step": 1499 + }, + { + "epoch": 0.4797217613784005, + "grad_norm": 5.053747653961182, + "learning_rate": 0.00033442762210711483, + "loss": 4.4644, + "step": 1500 + }, + { + "epoch": 0.4797217613784005, + "eval_runtime": 21.0968, + "eval_samples_per_second": 89.919, + "eval_steps_per_second": 22.515, + "step": 1500 + }, + { + "epoch": 0.4800415758859861, + "grad_norm": 2.686842679977417, + "learning_rate": 0.0003341188137031599, + "loss": 4.4637, + "step": 1501 + }, + { + "epoch": 0.48036139039357173, + "grad_norm": 2.36848783493042, + "learning_rate": 0.0003338099686693469, + "loss": 4.5577, + "step": 1502 + }, + { + "epoch": 0.48068120490115734, + "grad_norm": 3.264366865158081, + "learning_rate": 0.00033350108733725103, + "loss": 4.4832, + "step": 1503 + }, + { + "epoch": 0.4810010194087429, + "grad_norm": 1.8056799173355103, + "learning_rate": 0.00033319217003848644, + "loss": 4.5545, + "step": 1504 + }, + { + "epoch": 0.4813208339163285, + "grad_norm": 1.0576683282852173, + "learning_rate": 0.0003328832171047057, + "loss": 4.4828, + "step": 1505 + }, + { + "epoch": 0.48164064842391413, + "grad_norm": 1.7674775123596191, + "learning_rate": 0.0003325742288675998, + "loss": 4.511, + "step": 1506 + }, + { + "epoch": 0.48196046293149974, + "grad_norm": 3.1068592071533203, + "learning_rate": 0.0003322652056588976, + "loss": 4.5388, + "step": 1507 + }, + { + "epoch": 0.4822802774390853, + "grad_norm": 1.1966476440429688, + "learning_rate": 0.0003319561478103656, + "loss": 4.5276, + "step": 1508 + }, + { + "epoch": 0.4826000919466709, + "grad_norm": 1.421072244644165, + "learning_rate": 0.00033164705565380737, + "loss": 4.4601, + "step": 1509 + }, + { + "epoch": 0.48291990645425653, + "grad_norm": 1.5598156452178955, + "learning_rate": 0.00033133792952106327, + "loss": 4.4802, + "step": 1510 + }, + { + "epoch": 0.48323972096184215, + "grad_norm": 2.181429386138916, + "learning_rate": 0.0003310287697440102, + "loss": 4.4071, + "step": 1511 + }, + { + "epoch": 0.48355953546942776, + "grad_norm": 2.0646767616271973, + "learning_rate": 0.0003307195766545612, + "loss": 4.4171, + "step": 1512 + }, + { + "epoch": 0.4838793499770133, + "grad_norm": 2.699219226837158, + "learning_rate": 0.00033041035058466525, + "loss": 4.543, + "step": 1513 + }, + { + "epoch": 0.48419916448459893, + "grad_norm": 1.093851089477539, + "learning_rate": 0.00033010109186630625, + "loss": 4.4638, + "step": 1514 + }, + { + "epoch": 0.48451897899218455, + "grad_norm": 2.2062931060791016, + "learning_rate": 0.00032979180083150366, + "loss": 4.4553, + "step": 1515 + }, + { + "epoch": 0.48483879349977016, + "grad_norm": 58.853755950927734, + "learning_rate": 0.00032948247781231134, + "loss": 4.4896, + "step": 1516 + }, + { + "epoch": 0.4851586080073557, + "grad_norm": 5.704889297485352, + "learning_rate": 0.0003291731231408175, + "loss": 4.5006, + "step": 1517 + }, + { + "epoch": 0.48547842251494133, + "grad_norm": 1.4435511827468872, + "learning_rate": 0.00032886373714914455, + "loss": 4.5595, + "step": 1518 + }, + { + "epoch": 0.48579823702252695, + "grad_norm": 1.938969373703003, + "learning_rate": 0.00032855432016944835, + "loss": 4.4494, + "step": 1519 + }, + { + "epoch": 0.48611805153011256, + "grad_norm": 6.375471115112305, + "learning_rate": 0.000328244872533918, + "loss": 4.5077, + "step": 1520 + }, + { + "epoch": 0.4864378660376981, + "grad_norm": 1.735550045967102, + "learning_rate": 0.00032793539457477564, + "loss": 4.4728, + "step": 1521 + }, + { + "epoch": 0.48675768054528373, + "grad_norm": 1.5262601375579834, + "learning_rate": 0.00032762588662427585, + "loss": 4.416, + "step": 1522 + }, + { + "epoch": 0.48707749505286935, + "grad_norm": 2.8879239559173584, + "learning_rate": 0.0003273163490147054, + "loss": 4.3939, + "step": 1523 + }, + { + "epoch": 0.48739730956045496, + "grad_norm": 3.057363510131836, + "learning_rate": 0.0003270067820783831, + "loss": 4.4979, + "step": 1524 + }, + { + "epoch": 0.4877171240680405, + "grad_norm": 2.2869110107421875, + "learning_rate": 0.0003266971861476589, + "loss": 4.5236, + "step": 1525 + }, + { + "epoch": 0.48803693857562613, + "grad_norm": 9.431357383728027, + "learning_rate": 0.00032638756155491436, + "loss": 4.4864, + "step": 1526 + }, + { + "epoch": 0.48835675308321175, + "grad_norm": 2.6866185665130615, + "learning_rate": 0.0003260779086325612, + "loss": 4.4922, + "step": 1527 + }, + { + "epoch": 0.48867656759079736, + "grad_norm": 2.019314765930176, + "learning_rate": 0.0003257682277130422, + "loss": 4.4547, + "step": 1528 + }, + { + "epoch": 0.4889963820983829, + "grad_norm": 1.3251731395721436, + "learning_rate": 0.0003254585191288297, + "loss": 4.4668, + "step": 1529 + }, + { + "epoch": 0.48931619660596853, + "grad_norm": 2.1976327896118164, + "learning_rate": 0.0003251487832124259, + "loss": 4.4045, + "step": 1530 + }, + { + "epoch": 0.48963601111355415, + "grad_norm": 1.4416550397872925, + "learning_rate": 0.00032483902029636257, + "loss": 4.49, + "step": 1531 + }, + { + "epoch": 0.48995582562113976, + "grad_norm": 1.4494566917419434, + "learning_rate": 0.00032452923071320006, + "loss": 4.5339, + "step": 1532 + }, + { + "epoch": 0.4902756401287253, + "grad_norm": 4.242152214050293, + "learning_rate": 0.00032421941479552767, + "loss": 4.5034, + "step": 1533 + }, + { + "epoch": 0.49059545463631093, + "grad_norm": 3.1841986179351807, + "learning_rate": 0.00032390957287596275, + "loss": 4.5481, + "step": 1534 + }, + { + "epoch": 0.49091526914389655, + "grad_norm": 1.7304768562316895, + "learning_rate": 0.0003235997052871508, + "loss": 4.5392, + "step": 1535 + }, + { + "epoch": 0.49123508365148216, + "grad_norm": 2.055410861968994, + "learning_rate": 0.00032328981236176465, + "loss": 4.415, + "step": 1536 + }, + { + "epoch": 0.4915548981590677, + "grad_norm": 2.213911771774292, + "learning_rate": 0.00032297989443250445, + "loss": 4.4704, + "step": 1537 + }, + { + "epoch": 0.49187471266665334, + "grad_norm": 1.5110361576080322, + "learning_rate": 0.0003226699518320973, + "loss": 4.5045, + "step": 1538 + }, + { + "epoch": 0.49219452717423895, + "grad_norm": 2.1036298274993896, + "learning_rate": 0.0003223599848932964, + "loss": 4.4653, + "step": 1539 + }, + { + "epoch": 0.49251434168182456, + "grad_norm": 5.652075290679932, + "learning_rate": 0.0003220499939488817, + "loss": 4.4288, + "step": 1540 + }, + { + "epoch": 0.4928341561894101, + "grad_norm": 1.0321521759033203, + "learning_rate": 0.0003217399793316583, + "loss": 4.5447, + "step": 1541 + }, + { + "epoch": 0.49315397069699574, + "grad_norm": 1.8729327917099, + "learning_rate": 0.00032142994137445693, + "loss": 4.4108, + "step": 1542 + }, + { + "epoch": 0.49347378520458135, + "grad_norm": 1.3717520236968994, + "learning_rate": 0.0003211198804101337, + "loss": 4.482, + "step": 1543 + }, + { + "epoch": 0.49379359971216696, + "grad_norm": 2.5830628871917725, + "learning_rate": 0.000320809796771569, + "loss": 4.4513, + "step": 1544 + }, + { + "epoch": 0.4941134142197525, + "grad_norm": 3.476147413253784, + "learning_rate": 0.00032049969079166765, + "loss": 4.3947, + "step": 1545 + }, + { + "epoch": 0.49443322872733814, + "grad_norm": 1.3547031879425049, + "learning_rate": 0.0003201895628033587, + "loss": 4.5383, + "step": 1546 + }, + { + "epoch": 0.49475304323492375, + "grad_norm": 1.6946719884872437, + "learning_rate": 0.00031987941313959433, + "loss": 4.5286, + "step": 1547 + }, + { + "epoch": 0.49507285774250936, + "grad_norm": 4.0507354736328125, + "learning_rate": 0.0003195692421333506, + "loss": 4.474, + "step": 1548 + }, + { + "epoch": 0.4953926722500949, + "grad_norm": 2.9038877487182617, + "learning_rate": 0.0003192590501176261, + "loss": 4.4669, + "step": 1549 + }, + { + "epoch": 0.49571248675768054, + "grad_norm": 1.1318068504333496, + "learning_rate": 0.0003189488374254421, + "loss": 4.4662, + "step": 1550 + }, + { + "epoch": 0.49603230126526615, + "grad_norm": 2.556771993637085, + "learning_rate": 0.00031863860438984193, + "loss": 4.5247, + "step": 1551 + }, + { + "epoch": 0.49635211577285177, + "grad_norm": 2.2349417209625244, + "learning_rate": 0.00031832835134389093, + "loss": 4.5368, + "step": 1552 + }, + { + "epoch": 0.4966719302804373, + "grad_norm": 2.288339614868164, + "learning_rate": 0.0003180180786206759, + "loss": 4.4279, + "step": 1553 + }, + { + "epoch": 0.49699174478802294, + "grad_norm": 1.8688393831253052, + "learning_rate": 0.0003177077865533046, + "loss": 4.4775, + "step": 1554 + }, + { + "epoch": 0.49731155929560855, + "grad_norm": 3.0701799392700195, + "learning_rate": 0.00031739747547490584, + "loss": 4.4638, + "step": 1555 + }, + { + "epoch": 0.49763137380319417, + "grad_norm": 3.4693198204040527, + "learning_rate": 0.0003170871457186286, + "loss": 4.5335, + "step": 1556 + }, + { + "epoch": 0.4979511883107797, + "grad_norm": 2.1603169441223145, + "learning_rate": 0.0003167767976176419, + "loss": 4.3369, + "step": 1557 + }, + { + "epoch": 0.49827100281836534, + "grad_norm": 2.0994441509246826, + "learning_rate": 0.0003164664315051347, + "loss": 4.4924, + "step": 1558 + }, + { + "epoch": 0.49859081732595095, + "grad_norm": 2.3222177028656006, + "learning_rate": 0.00031615604771431514, + "loss": 4.4315, + "step": 1559 + }, + { + "epoch": 0.49891063183353657, + "grad_norm": 5.830596446990967, + "learning_rate": 0.00031584564657841015, + "loss": 4.5201, + "step": 1560 + }, + { + "epoch": 0.4992304463411221, + "grad_norm": 6.539062023162842, + "learning_rate": 0.0003155352284306657, + "loss": 4.4442, + "step": 1561 + }, + { + "epoch": 0.49955026084870774, + "grad_norm": 3.7740273475646973, + "learning_rate": 0.00031522479360434567, + "loss": 4.4715, + "step": 1562 + }, + { + "epoch": 0.49987007535629335, + "grad_norm": 3.043351173400879, + "learning_rate": 0.00031491434243273214, + "loss": 4.5875, + "step": 1563 + }, + { + "epoch": 0.5001898898638789, + "grad_norm": 2.0353150367736816, + "learning_rate": 0.00031460387524912437, + "loss": 4.511, + "step": 1564 + }, + { + "epoch": 0.5005097043714646, + "grad_norm": 2.4786124229431152, + "learning_rate": 0.0003142933923868391, + "loss": 4.3459, + "step": 1565 + }, + { + "epoch": 0.5008295188790501, + "grad_norm": 1.8204888105392456, + "learning_rate": 0.00031398289417920976, + "loss": 4.4785, + "step": 1566 + }, + { + "epoch": 0.5011493333866357, + "grad_norm": 2.904404878616333, + "learning_rate": 0.00031367238095958644, + "loss": 4.5261, + "step": 1567 + }, + { + "epoch": 0.5014691478942214, + "grad_norm": 3.449228525161743, + "learning_rate": 0.00031336185306133523, + "loss": 4.3711, + "step": 1568 + }, + { + "epoch": 0.5017889624018069, + "grad_norm": 1.671663761138916, + "learning_rate": 0.0003130513108178378, + "loss": 4.4228, + "step": 1569 + }, + { + "epoch": 0.5021087769093926, + "grad_norm": 2.128556251525879, + "learning_rate": 0.0003127407545624915, + "loss": 4.6034, + "step": 1570 + }, + { + "epoch": 0.5024285914169782, + "grad_norm": 1.3598664999008179, + "learning_rate": 0.0003124301846287085, + "loss": 4.4778, + "step": 1571 + }, + { + "epoch": 0.5027484059245637, + "grad_norm": 1.345407247543335, + "learning_rate": 0.00031211960134991596, + "loss": 4.4562, + "step": 1572 + }, + { + "epoch": 0.5030682204321494, + "grad_norm": 2.172025203704834, + "learning_rate": 0.00031180900505955496, + "loss": 4.5447, + "step": 1573 + }, + { + "epoch": 0.5033880349397349, + "grad_norm": 2.028595209121704, + "learning_rate": 0.000311498396091081, + "loss": 4.4886, + "step": 1574 + }, + { + "epoch": 0.5037078494473205, + "grad_norm": 1.6336915493011475, + "learning_rate": 0.00031118777477796275, + "loss": 4.452, + "step": 1575 + }, + { + "epoch": 0.5040276639549062, + "grad_norm": 2.2957088947296143, + "learning_rate": 0.0003108771414536825, + "loss": 4.487, + "step": 1576 + }, + { + "epoch": 0.5043474784624917, + "grad_norm": 2.4803295135498047, + "learning_rate": 0.0003105664964517351, + "loss": 4.4282, + "step": 1577 + }, + { + "epoch": 0.5046672929700774, + "grad_norm": 1.4016950130462646, + "learning_rate": 0.0003102558401056282, + "loss": 4.4071, + "step": 1578 + }, + { + "epoch": 0.504987107477663, + "grad_norm": 4.501900672912598, + "learning_rate": 0.00030994517274888155, + "loss": 4.4579, + "step": 1579 + }, + { + "epoch": 0.5053069219852485, + "grad_norm": 2.4082539081573486, + "learning_rate": 0.00030963449471502674, + "loss": 4.4744, + "step": 1580 + }, + { + "epoch": 0.5056267364928342, + "grad_norm": 1.69310462474823, + "learning_rate": 0.0003093238063376068, + "loss": 4.4508, + "step": 1581 + }, + { + "epoch": 0.5059465510004197, + "grad_norm": 3.4210100173950195, + "learning_rate": 0.00030901310795017567, + "loss": 4.5211, + "step": 1582 + }, + { + "epoch": 0.5062663655080054, + "grad_norm": 1.3291157484054565, + "learning_rate": 0.00030870239988629844, + "loss": 4.4627, + "step": 1583 + }, + { + "epoch": 0.506586180015591, + "grad_norm": 1.5633634328842163, + "learning_rate": 0.0003083916824795503, + "loss": 4.4959, + "step": 1584 + }, + { + "epoch": 0.5069059945231765, + "grad_norm": 1.1438454389572144, + "learning_rate": 0.0003080809560635165, + "loss": 4.3895, + "step": 1585 + }, + { + "epoch": 0.5072258090307622, + "grad_norm": 4.223932266235352, + "learning_rate": 0.0003077702209717921, + "loss": 4.5163, + "step": 1586 + }, + { + "epoch": 0.5075456235383478, + "grad_norm": 1.9954129457473755, + "learning_rate": 0.0003074594775379812, + "loss": 4.4116, + "step": 1587 + }, + { + "epoch": 0.5078654380459333, + "grad_norm": 1.7698767185211182, + "learning_rate": 0.00030714872609569733, + "loss": 4.4256, + "step": 1588 + }, + { + "epoch": 0.508185252553519, + "grad_norm": 2.36091947555542, + "learning_rate": 0.0003068379669785622, + "loss": 4.4634, + "step": 1589 + }, + { + "epoch": 0.5085050670611045, + "grad_norm": 9.579581260681152, + "learning_rate": 0.0003065272005202056, + "loss": 4.4265, + "step": 1590 + }, + { + "epoch": 0.5088248815686902, + "grad_norm": 3.1399269104003906, + "learning_rate": 0.00030621642705426586, + "loss": 4.491, + "step": 1591 + }, + { + "epoch": 0.5091446960762758, + "grad_norm": 1.3669244050979614, + "learning_rate": 0.0003059056469143884, + "loss": 4.4296, + "step": 1592 + }, + { + "epoch": 0.5094645105838613, + "grad_norm": 1.3722813129425049, + "learning_rate": 0.0003055948604342257, + "loss": 4.4209, + "step": 1593 + }, + { + "epoch": 0.509784325091447, + "grad_norm": 1.5370477437973022, + "learning_rate": 0.0003052840679474373, + "loss": 4.3651, + "step": 1594 + }, + { + "epoch": 0.5101041395990326, + "grad_norm": 3.1041219234466553, + "learning_rate": 0.0003049732697876891, + "loss": 4.4806, + "step": 1595 + }, + { + "epoch": 0.5104239541066181, + "grad_norm": 2.5296454429626465, + "learning_rate": 0.000304662466288653, + "loss": 4.5163, + "step": 1596 + }, + { + "epoch": 0.5107437686142038, + "grad_norm": 3.568481683731079, + "learning_rate": 0.000304351657784007, + "loss": 4.4307, + "step": 1597 + }, + { + "epoch": 0.5110635831217893, + "grad_norm": 2.123547315597534, + "learning_rate": 0.0003040408446074339, + "loss": 4.5755, + "step": 1598 + }, + { + "epoch": 0.511383397629375, + "grad_norm": 26.742572784423828, + "learning_rate": 0.000303730027092622, + "loss": 4.4115, + "step": 1599 + }, + { + "epoch": 0.5117032121369606, + "grad_norm": 1.5322078466415405, + "learning_rate": 0.00030341920557326385, + "loss": 4.4317, + "step": 1600 + }, + { + "epoch": 0.5117032121369606, + "eval_runtime": 21.2116, + "eval_samples_per_second": 89.432, + "eval_steps_per_second": 22.393, + "step": 1600 + }, + { + "epoch": 0.5120230266445461, + "grad_norm": 3.900994062423706, + "learning_rate": 0.0003031083803830567, + "loss": 4.4324, + "step": 1601 + }, + { + "epoch": 0.5123428411521318, + "grad_norm": 2.0009708404541016, + "learning_rate": 0.0003027975518557016, + "loss": 4.4895, + "step": 1602 + }, + { + "epoch": 0.5126626556597174, + "grad_norm": 1.7859699726104736, + "learning_rate": 0.00030248672032490295, + "loss": 4.4816, + "step": 1603 + }, + { + "epoch": 0.5129824701673029, + "grad_norm": 1.6561046838760376, + "learning_rate": 0.0003021758861243688, + "loss": 4.4773, + "step": 1604 + }, + { + "epoch": 0.5133022846748886, + "grad_norm": 1.787001132965088, + "learning_rate": 0.0003018650495878096, + "loss": 4.452, + "step": 1605 + }, + { + "epoch": 0.5136220991824741, + "grad_norm": 1.6168007850646973, + "learning_rate": 0.0003015542110489387, + "loss": 4.4085, + "step": 1606 + }, + { + "epoch": 0.5139419136900598, + "grad_norm": 46.81372833251953, + "learning_rate": 0.00030124337084147144, + "loss": 4.422, + "step": 1607 + }, + { + "epoch": 0.5142617281976454, + "grad_norm": 2.9768941402435303, + "learning_rate": 0.0003009325292991247, + "loss": 4.4337, + "step": 1608 + }, + { + "epoch": 0.5145815427052309, + "grad_norm": 12.249431610107422, + "learning_rate": 0.0003006216867556175, + "loss": 4.4698, + "step": 1609 + }, + { + "epoch": 0.5149013572128166, + "grad_norm": 2.8322079181671143, + "learning_rate": 0.00030031084354466904, + "loss": 4.4465, + "step": 1610 + }, + { + "epoch": 0.5152211717204022, + "grad_norm": 1.4054012298583984, + "learning_rate": 0.0003, + "loss": 4.5151, + "step": 1611 + }, + { + "epoch": 0.5155409862279877, + "grad_norm": 1.296450138092041, + "learning_rate": 0.00029968915645533085, + "loss": 4.4601, + "step": 1612 + }, + { + "epoch": 0.5158608007355734, + "grad_norm": 1.0465245246887207, + "learning_rate": 0.0002993783132443825, + "loss": 4.4348, + "step": 1613 + }, + { + "epoch": 0.516180615243159, + "grad_norm": 8.50320816040039, + "learning_rate": 0.0002990674707008752, + "loss": 4.4441, + "step": 1614 + }, + { + "epoch": 0.5165004297507446, + "grad_norm": 3.0005431175231934, + "learning_rate": 0.0002987566291585286, + "loss": 4.4732, + "step": 1615 + }, + { + "epoch": 0.5168202442583302, + "grad_norm": 1.9206727743148804, + "learning_rate": 0.00029844578895106127, + "loss": 4.4044, + "step": 1616 + }, + { + "epoch": 0.5171400587659157, + "grad_norm": 1.1312333345413208, + "learning_rate": 0.0002981349504121904, + "loss": 4.4027, + "step": 1617 + }, + { + "epoch": 0.5174598732735014, + "grad_norm": 1.2756153345108032, + "learning_rate": 0.0002978241138756312, + "loss": 4.4184, + "step": 1618 + }, + { + "epoch": 0.517779687781087, + "grad_norm": 1.0122417211532593, + "learning_rate": 0.00029751327967509695, + "loss": 4.4465, + "step": 1619 + }, + { + "epoch": 0.5180995022886725, + "grad_norm": 0.9842408299446106, + "learning_rate": 0.0002972024481442984, + "loss": 4.4122, + "step": 1620 + }, + { + "epoch": 0.5184193167962582, + "grad_norm": 2.5204224586486816, + "learning_rate": 0.00029689161961694323, + "loss": 4.418, + "step": 1621 + }, + { + "epoch": 0.5187391313038437, + "grad_norm": 1.8059693574905396, + "learning_rate": 0.00029658079442673616, + "loss": 4.4842, + "step": 1622 + }, + { + "epoch": 0.5190589458114294, + "grad_norm": 2.1444461345672607, + "learning_rate": 0.000296269972907378, + "loss": 4.4044, + "step": 1623 + }, + { + "epoch": 0.519378760319015, + "grad_norm": 2.454047918319702, + "learning_rate": 0.00029595915539256605, + "loss": 4.4966, + "step": 1624 + }, + { + "epoch": 0.5196985748266005, + "grad_norm": 2.915874481201172, + "learning_rate": 0.000295648342215993, + "loss": 4.485, + "step": 1625 + }, + { + "epoch": 0.5200183893341862, + "grad_norm": 8.692238807678223, + "learning_rate": 0.0002953375337113468, + "loss": 4.4227, + "step": 1626 + }, + { + "epoch": 0.5203382038417718, + "grad_norm": 1.423197865486145, + "learning_rate": 0.00029502673021231096, + "loss": 4.4833, + "step": 1627 + }, + { + "epoch": 0.5206580183493573, + "grad_norm": 1.3001943826675415, + "learning_rate": 0.0002947159320525627, + "loss": 4.4855, + "step": 1628 + }, + { + "epoch": 0.520977832856943, + "grad_norm": 1.3888938426971436, + "learning_rate": 0.0002944051395657744, + "loss": 4.4171, + "step": 1629 + }, + { + "epoch": 0.5212976473645285, + "grad_norm": 15.654441833496094, + "learning_rate": 0.0002940943530856116, + "loss": 4.4707, + "step": 1630 + }, + { + "epoch": 0.5216174618721142, + "grad_norm": 2.021806240081787, + "learning_rate": 0.00029378357294573403, + "loss": 4.4039, + "step": 1631 + }, + { + "epoch": 0.5219372763796998, + "grad_norm": 3.191789388656616, + "learning_rate": 0.0002934727994797944, + "loss": 4.5279, + "step": 1632 + }, + { + "epoch": 0.5222570908872853, + "grad_norm": 1.493067979812622, + "learning_rate": 0.0002931620330214378, + "loss": 4.476, + "step": 1633 + }, + { + "epoch": 0.522576905394871, + "grad_norm": 1.6242133378982544, + "learning_rate": 0.00029285127390430273, + "loss": 4.4378, + "step": 1634 + }, + { + "epoch": 0.5228967199024566, + "grad_norm": 13.364924430847168, + "learning_rate": 0.00029254052246201873, + "loss": 4.4999, + "step": 1635 + }, + { + "epoch": 0.5232165344100421, + "grad_norm": 3.7106330394744873, + "learning_rate": 0.00029222977902820785, + "loss": 4.5261, + "step": 1636 + }, + { + "epoch": 0.5235363489176278, + "grad_norm": 1.4169503450393677, + "learning_rate": 0.0002919190439364835, + "loss": 4.4383, + "step": 1637 + }, + { + "epoch": 0.5238561634252134, + "grad_norm": 3.8853206634521484, + "learning_rate": 0.00029160831752044966, + "loss": 4.5023, + "step": 1638 + }, + { + "epoch": 0.524175977932799, + "grad_norm": 5.8178229331970215, + "learning_rate": 0.00029129760011370156, + "loss": 4.4076, + "step": 1639 + }, + { + "epoch": 0.5244957924403846, + "grad_norm": 6.127602577209473, + "learning_rate": 0.00029098689204982433, + "loss": 4.4955, + "step": 1640 + }, + { + "epoch": 0.5248156069479701, + "grad_norm": 1.5075477361679077, + "learning_rate": 0.00029067619366239327, + "loss": 4.4435, + "step": 1641 + }, + { + "epoch": 0.5251354214555558, + "grad_norm": 1.3636072874069214, + "learning_rate": 0.00029036550528497326, + "loss": 4.4475, + "step": 1642 + }, + { + "epoch": 0.5254552359631414, + "grad_norm": 1.1479662656784058, + "learning_rate": 0.0002900548272511183, + "loss": 4.5011, + "step": 1643 + }, + { + "epoch": 0.5257750504707269, + "grad_norm": 2.4820282459259033, + "learning_rate": 0.00028974415989437176, + "loss": 4.447, + "step": 1644 + }, + { + "epoch": 0.5260948649783126, + "grad_norm": 1.322964072227478, + "learning_rate": 0.0002894335035482649, + "loss": 4.5088, + "step": 1645 + }, + { + "epoch": 0.5264146794858982, + "grad_norm": 1.9006342887878418, + "learning_rate": 0.00028912285854631754, + "loss": 4.4052, + "step": 1646 + }, + { + "epoch": 0.5267344939934838, + "grad_norm": 4.762979507446289, + "learning_rate": 0.0002888122252220372, + "loss": 4.4393, + "step": 1647 + }, + { + "epoch": 0.5270543085010694, + "grad_norm": 1.9398455619812012, + "learning_rate": 0.00028850160390891895, + "loss": 4.4838, + "step": 1648 + }, + { + "epoch": 0.5273741230086549, + "grad_norm": 12.971105575561523, + "learning_rate": 0.000288190994940445, + "loss": 4.4335, + "step": 1649 + }, + { + "epoch": 0.5276939375162406, + "grad_norm": 3.9425482749938965, + "learning_rate": 0.00028788039865008404, + "loss": 4.5124, + "step": 1650 + }, + { + "epoch": 0.5280137520238262, + "grad_norm": 34.72132110595703, + "learning_rate": 0.0002875698153712915, + "loss": 4.4177, + "step": 1651 + }, + { + "epoch": 0.5283335665314118, + "grad_norm": 2.859478712081909, + "learning_rate": 0.0002872592454375086, + "loss": 4.4321, + "step": 1652 + }, + { + "epoch": 0.5286533810389974, + "grad_norm": 2.480025291442871, + "learning_rate": 0.00028694868918216227, + "loss": 4.4677, + "step": 1653 + }, + { + "epoch": 0.528973195546583, + "grad_norm": 3.598733901977539, + "learning_rate": 0.0002866381469386648, + "loss": 4.501, + "step": 1654 + }, + { + "epoch": 0.5292930100541686, + "grad_norm": 2.2065670490264893, + "learning_rate": 0.0002863276190404135, + "loss": 4.4168, + "step": 1655 + }, + { + "epoch": 0.5296128245617542, + "grad_norm": 132.14743041992188, + "learning_rate": 0.0002860171058207902, + "loss": 4.5055, + "step": 1656 + }, + { + "epoch": 0.5299326390693397, + "grad_norm": 5.4599103927612305, + "learning_rate": 0.0002857066076131609, + "loss": 4.4414, + "step": 1657 + }, + { + "epoch": 0.5302524535769254, + "grad_norm": 1.8813872337341309, + "learning_rate": 0.00028539612475087563, + "loss": 4.4436, + "step": 1658 + }, + { + "epoch": 0.530572268084511, + "grad_norm": 1.403524398803711, + "learning_rate": 0.0002850856575672679, + "loss": 4.4515, + "step": 1659 + }, + { + "epoch": 0.5308920825920966, + "grad_norm": 2.691366672515869, + "learning_rate": 0.0002847752063956543, + "loss": 4.4313, + "step": 1660 + }, + { + "epoch": 0.5312118970996822, + "grad_norm": 2.570934534072876, + "learning_rate": 0.00028446477156933425, + "loss": 4.3826, + "step": 1661 + }, + { + "epoch": 0.5315317116072678, + "grad_norm": 2.4571893215179443, + "learning_rate": 0.0002841543534215898, + "loss": 4.4946, + "step": 1662 + }, + { + "epoch": 0.5318515261148534, + "grad_norm": 3.515303134918213, + "learning_rate": 0.0002838439522856849, + "loss": 4.4524, + "step": 1663 + }, + { + "epoch": 0.532171340622439, + "grad_norm": 43.24845886230469, + "learning_rate": 0.00028353356849486526, + "loss": 4.4339, + "step": 1664 + }, + { + "epoch": 0.5324911551300245, + "grad_norm": 2.5286126136779785, + "learning_rate": 0.00028322320238235814, + "loss": 4.3808, + "step": 1665 + }, + { + "epoch": 0.5328109696376102, + "grad_norm": 6.093226909637451, + "learning_rate": 0.00028291285428137146, + "loss": 4.5141, + "step": 1666 + }, + { + "epoch": 0.5331307841451958, + "grad_norm": 3.8247766494750977, + "learning_rate": 0.0002826025245250941, + "loss": 4.4479, + "step": 1667 + }, + { + "epoch": 0.5334505986527814, + "grad_norm": 2.5034408569335938, + "learning_rate": 0.00028229221344669534, + "loss": 4.5013, + "step": 1668 + }, + { + "epoch": 0.533770413160367, + "grad_norm": 2.815786123275757, + "learning_rate": 0.0002819819213793241, + "loss": 4.4723, + "step": 1669 + }, + { + "epoch": 0.5340902276679526, + "grad_norm": 2.6400837898254395, + "learning_rate": 0.00028167164865610907, + "loss": 4.4585, + "step": 1670 + }, + { + "epoch": 0.5344100421755382, + "grad_norm": 3.206653356552124, + "learning_rate": 0.00028136139561015807, + "loss": 4.3819, + "step": 1671 + }, + { + "epoch": 0.5347298566831238, + "grad_norm": 2.5874226093292236, + "learning_rate": 0.00028105116257455786, + "loss": 4.3634, + "step": 1672 + }, + { + "epoch": 0.5350496711907093, + "grad_norm": 2.8991687297821045, + "learning_rate": 0.00028074094988237385, + "loss": 4.4363, + "step": 1673 + }, + { + "epoch": 0.535369485698295, + "grad_norm": 1.9159778356552124, + "learning_rate": 0.00028043075786664934, + "loss": 4.4425, + "step": 1674 + }, + { + "epoch": 0.5356893002058806, + "grad_norm": 1.9093455076217651, + "learning_rate": 0.0002801205868604057, + "loss": 4.3368, + "step": 1675 + }, + { + "epoch": 0.5360091147134662, + "grad_norm": 19.991605758666992, + "learning_rate": 0.0002798104371966414, + "loss": 4.4203, + "step": 1676 + }, + { + "epoch": 0.5363289292210518, + "grad_norm": 3.848237991333008, + "learning_rate": 0.0002795003092083324, + "loss": 4.5209, + "step": 1677 + }, + { + "epoch": 0.5366487437286374, + "grad_norm": 1.4244316816329956, + "learning_rate": 0.000279190203228431, + "loss": 4.3764, + "step": 1678 + }, + { + "epoch": 0.536968558236223, + "grad_norm": 1.945302128791809, + "learning_rate": 0.00027888011958986623, + "loss": 4.5115, + "step": 1679 + }, + { + "epoch": 0.5372883727438086, + "grad_norm": 2.43974232673645, + "learning_rate": 0.00027857005862554307, + "loss": 4.3868, + "step": 1680 + }, + { + "epoch": 0.5376081872513941, + "grad_norm": 1.1911664009094238, + "learning_rate": 0.00027826002066834167, + "loss": 4.51, + "step": 1681 + }, + { + "epoch": 0.5379280017589798, + "grad_norm": 2.185182571411133, + "learning_rate": 0.0002779500060511184, + "loss": 4.3778, + "step": 1682 + }, + { + "epoch": 0.5382478162665654, + "grad_norm": 2.263256549835205, + "learning_rate": 0.00027764001510670354, + "loss": 4.2892, + "step": 1683 + }, + { + "epoch": 0.538567630774151, + "grad_norm": 3.3477792739868164, + "learning_rate": 0.00027733004816790267, + "loss": 4.4173, + "step": 1684 + }, + { + "epoch": 0.5388874452817366, + "grad_norm": 1.1815649271011353, + "learning_rate": 0.00027702010556749556, + "loss": 4.4189, + "step": 1685 + }, + { + "epoch": 0.5392072597893222, + "grad_norm": 1.3073785305023193, + "learning_rate": 0.0002767101876382353, + "loss": 4.4348, + "step": 1686 + }, + { + "epoch": 0.5395270742969078, + "grad_norm": 2.1098945140838623, + "learning_rate": 0.00027640029471284923, + "loss": 4.39, + "step": 1687 + }, + { + "epoch": 0.5398468888044934, + "grad_norm": 1.6169921159744263, + "learning_rate": 0.00027609042712403725, + "loss": 4.449, + "step": 1688 + }, + { + "epoch": 0.5401667033120789, + "grad_norm": 3.421034574508667, + "learning_rate": 0.0002757805852044724, + "loss": 4.4283, + "step": 1689 + }, + { + "epoch": 0.5404865178196646, + "grad_norm": 2.2211074829101562, + "learning_rate": 0.00027547076928679994, + "loss": 4.4231, + "step": 1690 + }, + { + "epoch": 0.5408063323272502, + "grad_norm": 1.597461462020874, + "learning_rate": 0.0002751609797036373, + "loss": 4.3871, + "step": 1691 + }, + { + "epoch": 0.5411261468348358, + "grad_norm": 1.937148928642273, + "learning_rate": 0.00027485121678757406, + "loss": 4.4498, + "step": 1692 + }, + { + "epoch": 0.5414459613424214, + "grad_norm": 3.795621156692505, + "learning_rate": 0.0002745414808711703, + "loss": 4.4031, + "step": 1693 + }, + { + "epoch": 0.541765775850007, + "grad_norm": 2.4538938999176025, + "learning_rate": 0.00027423177228695785, + "loss": 4.3768, + "step": 1694 + }, + { + "epoch": 0.5420855903575926, + "grad_norm": 1.317055106163025, + "learning_rate": 0.00027392209136743875, + "loss": 4.3811, + "step": 1695 + }, + { + "epoch": 0.5424054048651782, + "grad_norm": 1.9118362665176392, + "learning_rate": 0.0002736124384450857, + "loss": 4.4102, + "step": 1696 + }, + { + "epoch": 0.5427252193727637, + "grad_norm": 3.2753493785858154, + "learning_rate": 0.0002733028138523411, + "loss": 4.388, + "step": 1697 + }, + { + "epoch": 0.5430450338803494, + "grad_norm": 2.2327990531921387, + "learning_rate": 0.0002729932179216169, + "loss": 4.3472, + "step": 1698 + }, + { + "epoch": 0.543364848387935, + "grad_norm": 2.5335423946380615, + "learning_rate": 0.0002726836509852946, + "loss": 4.3918, + "step": 1699 + }, + { + "epoch": 0.5436846628955206, + "grad_norm": 1.3557298183441162, + "learning_rate": 0.0002723741133757242, + "loss": 4.3948, + "step": 1700 + }, + { + "epoch": 0.5436846628955206, + "eval_runtime": 21.1884, + "eval_samples_per_second": 89.53, + "eval_steps_per_second": 22.418, + "step": 1700 + }, + { + "epoch": 0.5440044774031062, + "grad_norm": 5.7346272468566895, + "learning_rate": 0.0002720646054252244, + "loss": 4.4382, + "step": 1701 + }, + { + "epoch": 0.5443242919106918, + "grad_norm": 1.1384223699569702, + "learning_rate": 0.000271755127466082, + "loss": 4.4592, + "step": 1702 + }, + { + "epoch": 0.5446441064182774, + "grad_norm": 15.419846534729004, + "learning_rate": 0.0002714456798305516, + "loss": 4.4519, + "step": 1703 + }, + { + "epoch": 0.544963920925863, + "grad_norm": 1.949489951133728, + "learning_rate": 0.0002711362628508554, + "loss": 4.3605, + "step": 1704 + }, + { + "epoch": 0.5452837354334485, + "grad_norm": 1.9095494747161865, + "learning_rate": 0.0002708268768591825, + "loss": 4.4349, + "step": 1705 + }, + { + "epoch": 0.5456035499410342, + "grad_norm": 9.877264022827148, + "learning_rate": 0.0002705175221876887, + "loss": 4.4293, + "step": 1706 + }, + { + "epoch": 0.5459233644486198, + "grad_norm": 1.7485895156860352, + "learning_rate": 0.00027020819916849634, + "loss": 4.4018, + "step": 1707 + }, + { + "epoch": 0.5462431789562054, + "grad_norm": 1.1110365390777588, + "learning_rate": 0.0002698989081336937, + "loss": 4.3829, + "step": 1708 + }, + { + "epoch": 0.546562993463791, + "grad_norm": 3.4032797813415527, + "learning_rate": 0.00026958964941533475, + "loss": 4.3905, + "step": 1709 + }, + { + "epoch": 0.5468828079713766, + "grad_norm": 1.2629104852676392, + "learning_rate": 0.00026928042334543867, + "loss": 4.3711, + "step": 1710 + }, + { + "epoch": 0.5472026224789622, + "grad_norm": 1.0965633392333984, + "learning_rate": 0.0002689712302559898, + "loss": 4.4883, + "step": 1711 + }, + { + "epoch": 0.5475224369865478, + "grad_norm": 1.1227163076400757, + "learning_rate": 0.0002686620704789367, + "loss": 4.4279, + "step": 1712 + }, + { + "epoch": 0.5478422514941333, + "grad_norm": 1.058948278427124, + "learning_rate": 0.00026835294434619263, + "loss": 4.4245, + "step": 1713 + }, + { + "epoch": 0.548162066001719, + "grad_norm": 0.9659677743911743, + "learning_rate": 0.0002680438521896343, + "loss": 4.4369, + "step": 1714 + }, + { + "epoch": 0.5484818805093046, + "grad_norm": 6.858630180358887, + "learning_rate": 0.0002677347943411023, + "loss": 4.4548, + "step": 1715 + }, + { + "epoch": 0.5488016950168902, + "grad_norm": 33.97810745239258, + "learning_rate": 0.0002674257711324002, + "loss": 4.3876, + "step": 1716 + }, + { + "epoch": 0.5491215095244758, + "grad_norm": 1.3881527185440063, + "learning_rate": 0.0002671167828952943, + "loss": 4.4359, + "step": 1717 + }, + { + "epoch": 0.5494413240320614, + "grad_norm": 1.7751132249832153, + "learning_rate": 0.0002668078299615136, + "loss": 4.4027, + "step": 1718 + }, + { + "epoch": 0.549761138539647, + "grad_norm": 3.169660806655884, + "learning_rate": 0.0002664989126627489, + "loss": 4.5096, + "step": 1719 + }, + { + "epoch": 0.5500809530472326, + "grad_norm": 1.979705572128296, + "learning_rate": 0.00026619003133065305, + "loss": 4.5115, + "step": 1720 + }, + { + "epoch": 0.5504007675548183, + "grad_norm": 0.9634051322937012, + "learning_rate": 0.0002658811862968401, + "loss": 4.4056, + "step": 1721 + }, + { + "epoch": 0.5507205820624038, + "grad_norm": 1.1241754293441772, + "learning_rate": 0.0002655723778928851, + "loss": 4.4045, + "step": 1722 + }, + { + "epoch": 0.5510403965699894, + "grad_norm": 2.7725794315338135, + "learning_rate": 0.00026526360645032405, + "loss": 4.4454, + "step": 1723 + }, + { + "epoch": 0.551360211077575, + "grad_norm": 2.7524471282958984, + "learning_rate": 0.0002649548723006527, + "loss": 4.5234, + "step": 1724 + }, + { + "epoch": 0.5516800255851606, + "grad_norm": 2.8122780323028564, + "learning_rate": 0.00026464617577532756, + "loss": 4.5729, + "step": 1725 + }, + { + "epoch": 0.5519998400927462, + "grad_norm": 1.313642978668213, + "learning_rate": 0.00026433751720576385, + "loss": 4.3779, + "step": 1726 + }, + { + "epoch": 0.5523196546003318, + "grad_norm": 1.3509106636047363, + "learning_rate": 0.00026402889692333684, + "loss": 4.3733, + "step": 1727 + }, + { + "epoch": 0.5526394691079174, + "grad_norm": 1.4904330968856812, + "learning_rate": 0.00026372031525938034, + "loss": 4.3392, + "step": 1728 + }, + { + "epoch": 0.5529592836155031, + "grad_norm": 2.434535264968872, + "learning_rate": 0.0002634117725451865, + "loss": 4.3979, + "step": 1729 + }, + { + "epoch": 0.5532790981230886, + "grad_norm": 1.39297616481781, + "learning_rate": 0.00026310326911200616, + "loss": 4.4224, + "step": 1730 + }, + { + "epoch": 0.5535989126306742, + "grad_norm": 3.068152904510498, + "learning_rate": 0.0002627948052910474, + "loss": 4.5193, + "step": 1731 + }, + { + "epoch": 0.5539187271382598, + "grad_norm": 1.2025773525238037, + "learning_rate": 0.00026248638141347614, + "loss": 4.4054, + "step": 1732 + }, + { + "epoch": 0.5542385416458454, + "grad_norm": 2.6135780811309814, + "learning_rate": 0.00026217799781041567, + "loss": 4.4926, + "step": 1733 + }, + { + "epoch": 0.554558356153431, + "grad_norm": 5.590660572052002, + "learning_rate": 0.00026186965481294526, + "loss": 4.3268, + "step": 1734 + }, + { + "epoch": 0.5548781706610166, + "grad_norm": 20.141006469726562, + "learning_rate": 0.0002615613527521014, + "loss": 4.4427, + "step": 1735 + }, + { + "epoch": 0.5551979851686022, + "grad_norm": 1.4918791055679321, + "learning_rate": 0.00026125309195887603, + "loss": 4.4428, + "step": 1736 + }, + { + "epoch": 0.5555177996761879, + "grad_norm": 3.141359806060791, + "learning_rate": 0.0002609448727642172, + "loss": 4.3533, + "step": 1737 + }, + { + "epoch": 0.5558376141837734, + "grad_norm": 4.361044883728027, + "learning_rate": 0.000260636695499028, + "loss": 4.4284, + "step": 1738 + }, + { + "epoch": 0.556157428691359, + "grad_norm": 1.2768278121948242, + "learning_rate": 0.00026032856049416664, + "loss": 4.3838, + "step": 1739 + }, + { + "epoch": 0.5564772431989446, + "grad_norm": 1.246594786643982, + "learning_rate": 0.00026002046808044617, + "loss": 4.4573, + "step": 1740 + }, + { + "epoch": 0.5567970577065302, + "grad_norm": 3.0351738929748535, + "learning_rate": 0.0002597124185886334, + "loss": 4.4799, + "step": 1741 + }, + { + "epoch": 0.5571168722141158, + "grad_norm": 1.5607551336288452, + "learning_rate": 0.0002594044123494496, + "loss": 4.416, + "step": 1742 + }, + { + "epoch": 0.5574366867217014, + "grad_norm": 1.4376870393753052, + "learning_rate": 0.00025909644969356917, + "loss": 4.4087, + "step": 1743 + }, + { + "epoch": 0.557756501229287, + "grad_norm": 2.601006031036377, + "learning_rate": 0.00025878853095162, + "loss": 4.416, + "step": 1744 + }, + { + "epoch": 0.5580763157368727, + "grad_norm": 1.8028939962387085, + "learning_rate": 0.00025848065645418285, + "loss": 4.3547, + "step": 1745 + }, + { + "epoch": 0.5583961302444582, + "grad_norm": 2.2018816471099854, + "learning_rate": 0.00025817282653179064, + "loss": 4.4787, + "step": 1746 + }, + { + "epoch": 0.5587159447520438, + "grad_norm": 3.8400559425354004, + "learning_rate": 0.0002578650415149289, + "loss": 4.4116, + "step": 1747 + }, + { + "epoch": 0.5590357592596295, + "grad_norm": 1.3404124975204468, + "learning_rate": 0.00025755730173403457, + "loss": 4.3548, + "step": 1748 + }, + { + "epoch": 0.559355573767215, + "grad_norm": 7.124460220336914, + "learning_rate": 0.0002572496075194963, + "loss": 4.3851, + "step": 1749 + }, + { + "epoch": 0.5596753882748006, + "grad_norm": 1.7407119274139404, + "learning_rate": 0.00025694195920165344, + "loss": 4.49, + "step": 1750 + }, + { + "epoch": 0.5599952027823862, + "grad_norm": 1.9646064043045044, + "learning_rate": 0.0002566343571107966, + "loss": 4.4129, + "step": 1751 + }, + { + "epoch": 0.5603150172899718, + "grad_norm": 4.31877326965332, + "learning_rate": 0.0002563268015771664, + "loss": 4.4313, + "step": 1752 + }, + { + "epoch": 0.5606348317975575, + "grad_norm": 23.61766242980957, + "learning_rate": 0.00025601929293095344, + "loss": 4.4024, + "step": 1753 + }, + { + "epoch": 0.560954646305143, + "grad_norm": 1.3564687967300415, + "learning_rate": 0.00025571183150229827, + "loss": 4.3634, + "step": 1754 + }, + { + "epoch": 0.5612744608127286, + "grad_norm": 1.7455663681030273, + "learning_rate": 0.00025540441762129045, + "loss": 4.3618, + "step": 1755 + }, + { + "epoch": 0.5615942753203143, + "grad_norm": 1.860573410987854, + "learning_rate": 0.00025509705161796866, + "loss": 4.4205, + "step": 1756 + }, + { + "epoch": 0.5619140898278998, + "grad_norm": 1.115132451057434, + "learning_rate": 0.0002547897338223202, + "loss": 4.3544, + "step": 1757 + }, + { + "epoch": 0.5622339043354854, + "grad_norm": 1.0278851985931396, + "learning_rate": 0.0002544824645642804, + "loss": 4.4366, + "step": 1758 + }, + { + "epoch": 0.562553718843071, + "grad_norm": 4.820397853851318, + "learning_rate": 0.00025417524417373276, + "loss": 4.3838, + "step": 1759 + }, + { + "epoch": 0.5628735333506566, + "grad_norm": 1.4825657606124878, + "learning_rate": 0.00025386807298050817, + "loss": 4.4654, + "step": 1760 + }, + { + "epoch": 0.5631933478582423, + "grad_norm": 2.824028968811035, + "learning_rate": 0.00025356095131438464, + "loss": 4.4438, + "step": 1761 + }, + { + "epoch": 0.5635131623658278, + "grad_norm": 1.9610645771026611, + "learning_rate": 0.0002532538795050872, + "loss": 4.4661, + "step": 1762 + }, + { + "epoch": 0.5638329768734134, + "grad_norm": 1.805777668952942, + "learning_rate": 0.0002529468578822871, + "loss": 4.4179, + "step": 1763 + }, + { + "epoch": 0.564152791380999, + "grad_norm": 1.1127445697784424, + "learning_rate": 0.00025263988677560204, + "loss": 4.3391, + "step": 1764 + }, + { + "epoch": 0.5644726058885846, + "grad_norm": 2.7552614212036133, + "learning_rate": 0.0002523329665145951, + "loss": 4.3162, + "step": 1765 + }, + { + "epoch": 0.5647924203961702, + "grad_norm": 2.029419183731079, + "learning_rate": 0.00025202609742877515, + "loss": 4.3919, + "step": 1766 + }, + { + "epoch": 0.5651122349037558, + "grad_norm": 1.2228567600250244, + "learning_rate": 0.0002517192798475958, + "loss": 4.263, + "step": 1767 + }, + { + "epoch": 0.5654320494113414, + "grad_norm": 1.0710145235061646, + "learning_rate": 0.00025141251410045547, + "loss": 4.3386, + "step": 1768 + }, + { + "epoch": 0.5657518639189271, + "grad_norm": 3.3128814697265625, + "learning_rate": 0.0002511058005166972, + "loss": 4.2852, + "step": 1769 + }, + { + "epoch": 0.5660716784265126, + "grad_norm": 14.066691398620605, + "learning_rate": 0.0002507991394256075, + "loss": 4.3672, + "step": 1770 + }, + { + "epoch": 0.5663914929340982, + "grad_norm": 4.372364521026611, + "learning_rate": 0.00025049253115641713, + "loss": 4.4313, + "step": 1771 + }, + { + "epoch": 0.5667113074416839, + "grad_norm": 2.4273574352264404, + "learning_rate": 0.00025018597603829944, + "loss": 4.4525, + "step": 1772 + }, + { + "epoch": 0.5670311219492694, + "grad_norm": 2.6426570415496826, + "learning_rate": 0.0002498794744003715, + "loss": 4.3693, + "step": 1773 + }, + { + "epoch": 0.567350936456855, + "grad_norm": 1.8511203527450562, + "learning_rate": 0.0002495730265716922, + "loss": 4.3827, + "step": 1774 + }, + { + "epoch": 0.5676707509644406, + "grad_norm": 2.3621106147766113, + "learning_rate": 0.00024926663288126323, + "loss": 4.4039, + "step": 1775 + }, + { + "epoch": 0.5679905654720262, + "grad_norm": 1.2252600193023682, + "learning_rate": 0.00024896029365802807, + "loss": 4.343, + "step": 1776 + }, + { + "epoch": 0.5683103799796119, + "grad_norm": 1.553681492805481, + "learning_rate": 0.0002486540092308713, + "loss": 4.4556, + "step": 1777 + }, + { + "epoch": 0.5686301944871974, + "grad_norm": 3.4369189739227295, + "learning_rate": 0.00024834777992861935, + "loss": 4.3188, + "step": 1778 + }, + { + "epoch": 0.568950008994783, + "grad_norm": 2.4597878456115723, + "learning_rate": 0.0002480416060800387, + "loss": 4.404, + "step": 1779 + }, + { + "epoch": 0.5692698235023687, + "grad_norm": 1.3967732191085815, + "learning_rate": 0.0002477354880138369, + "loss": 4.4033, + "step": 1780 + }, + { + "epoch": 0.5695896380099542, + "grad_norm": 3.743600368499756, + "learning_rate": 0.0002474294260586615, + "loss": 4.3736, + "step": 1781 + }, + { + "epoch": 0.5699094525175398, + "grad_norm": 1.2528659105300903, + "learning_rate": 0.00024712342054309945, + "loss": 4.4283, + "step": 1782 + }, + { + "epoch": 0.5702292670251254, + "grad_norm": 2.025254964828491, + "learning_rate": 0.00024681747179567775, + "loss": 4.3832, + "step": 1783 + }, + { + "epoch": 0.570549081532711, + "grad_norm": 1.6618618965148926, + "learning_rate": 0.0002465115801448617, + "loss": 4.389, + "step": 1784 + }, + { + "epoch": 0.5708688960402967, + "grad_norm": 0.9266157150268555, + "learning_rate": 0.0002462057459190559, + "loss": 4.3977, + "step": 1785 + }, + { + "epoch": 0.5711887105478822, + "grad_norm": 1.2248742580413818, + "learning_rate": 0.0002458999694466029, + "loss": 4.3895, + "step": 1786 + }, + { + "epoch": 0.5715085250554678, + "grad_norm": 2.623033046722412, + "learning_rate": 0.0002455942510557836, + "loss": 4.4049, + "step": 1787 + }, + { + "epoch": 0.5718283395630535, + "grad_norm": 1.2491263151168823, + "learning_rate": 0.0002452885910748163, + "loss": 4.3944, + "step": 1788 + }, + { + "epoch": 0.572148154070639, + "grad_norm": 3.226208209991455, + "learning_rate": 0.0002449829898318566, + "loss": 4.3628, + "step": 1789 + }, + { + "epoch": 0.5724679685782247, + "grad_norm": 1.793782353401184, + "learning_rate": 0.0002446774476549972, + "loss": 4.4122, + "step": 1790 + }, + { + "epoch": 0.5727877830858102, + "grad_norm": 2.082371234893799, + "learning_rate": 0.00024437196487226716, + "loss": 4.3426, + "step": 1791 + }, + { + "epoch": 0.5731075975933958, + "grad_norm": 1.4262444972991943, + "learning_rate": 0.00024406654181163197, + "loss": 4.3733, + "step": 1792 + }, + { + "epoch": 0.5734274121009815, + "grad_norm": 2.1430935859680176, + "learning_rate": 0.000243761178800993, + "loss": 4.3886, + "step": 1793 + }, + { + "epoch": 0.573747226608567, + "grad_norm": 1.2457352876663208, + "learning_rate": 0.00024345587616818692, + "loss": 4.3072, + "step": 1794 + }, + { + "epoch": 0.5740670411161526, + "grad_norm": 17.569589614868164, + "learning_rate": 0.00024315063424098585, + "loss": 4.3812, + "step": 1795 + }, + { + "epoch": 0.5743868556237383, + "grad_norm": 1.19608736038208, + "learning_rate": 0.00024284545334709657, + "loss": 4.5371, + "step": 1796 + }, + { + "epoch": 0.5747066701313238, + "grad_norm": 2.4358391761779785, + "learning_rate": 0.00024254033381416047, + "loss": 4.3266, + "step": 1797 + }, + { + "epoch": 0.5750264846389095, + "grad_norm": 7.1609368324279785, + "learning_rate": 0.00024223527596975284, + "loss": 4.3794, + "step": 1798 + }, + { + "epoch": 0.575346299146495, + "grad_norm": 1.9270249605178833, + "learning_rate": 0.000241930280141383, + "loss": 4.4444, + "step": 1799 + }, + { + "epoch": 0.5756661136540806, + "grad_norm": 1.5177801847457886, + "learning_rate": 0.00024162534665649358, + "loss": 4.4564, + "step": 1800 + }, + { + "epoch": 0.5756661136540806, + "eval_runtime": 21.217, + "eval_samples_per_second": 89.409, + "eval_steps_per_second": 22.388, + "step": 1800 + }, + { + "epoch": 0.5759859281616663, + "grad_norm": 1.427351474761963, + "learning_rate": 0.0002413204758424602, + "loss": 4.3227, + "step": 1801 + }, + { + "epoch": 0.5763057426692518, + "grad_norm": 0.9850735068321228, + "learning_rate": 0.00024101566802659137, + "loss": 4.4405, + "step": 1802 + }, + { + "epoch": 0.5766255571768374, + "grad_norm": 6.984677791595459, + "learning_rate": 0.0002407109235361277, + "loss": 4.3523, + "step": 1803 + }, + { + "epoch": 0.5769453716844231, + "grad_norm": 3.6690211296081543, + "learning_rate": 0.0002404062426982421, + "loss": 4.4188, + "step": 1804 + }, + { + "epoch": 0.5772651861920086, + "grad_norm": 2.4254889488220215, + "learning_rate": 0.00024010162584003905, + "loss": 4.3921, + "step": 1805 + }, + { + "epoch": 0.5775850006995943, + "grad_norm": 1.057378888130188, + "learning_rate": 0.0002397970732885542, + "loss": 4.4946, + "step": 1806 + }, + { + "epoch": 0.5779048152071798, + "grad_norm": 5.068242073059082, + "learning_rate": 0.0002394925853707544, + "loss": 4.3515, + "step": 1807 + }, + { + "epoch": 0.5782246297147654, + "grad_norm": 1.9902795553207397, + "learning_rate": 0.00023918816241353684, + "loss": 4.4122, + "step": 1808 + }, + { + "epoch": 0.5785444442223511, + "grad_norm": 1.0972682237625122, + "learning_rate": 0.0002388838047437293, + "loss": 4.3937, + "step": 1809 + }, + { + "epoch": 0.5788642587299366, + "grad_norm": 1.6081758737564087, + "learning_rate": 0.00023857951268808905, + "loss": 4.3601, + "step": 1810 + }, + { + "epoch": 0.5791840732375222, + "grad_norm": 2.258594274520874, + "learning_rate": 0.00023827528657330331, + "loss": 4.5067, + "step": 1811 + }, + { + "epoch": 0.5795038877451079, + "grad_norm": 3.941608428955078, + "learning_rate": 0.00023797112672598833, + "loss": 4.3747, + "step": 1812 + }, + { + "epoch": 0.5798237022526934, + "grad_norm": 1.4697375297546387, + "learning_rate": 0.0002376670334726891, + "loss": 4.3473, + "step": 1813 + }, + { + "epoch": 0.5801435167602791, + "grad_norm": 1.4181195497512817, + "learning_rate": 0.00023736300713987946, + "loss": 4.4536, + "step": 1814 + }, + { + "epoch": 0.5804633312678646, + "grad_norm": 1.0497387647628784, + "learning_rate": 0.00023705904805396095, + "loss": 4.3939, + "step": 1815 + }, + { + "epoch": 0.5807831457754502, + "grad_norm": 2.2155075073242188, + "learning_rate": 0.00023675515654126327, + "loss": 4.402, + "step": 1816 + }, + { + "epoch": 0.5811029602830359, + "grad_norm": 2.02759051322937, + "learning_rate": 0.00023645133292804352, + "loss": 4.3989, + "step": 1817 + }, + { + "epoch": 0.5814227747906214, + "grad_norm": 1.259806513786316, + "learning_rate": 0.0002361475775404857, + "loss": 4.3436, + "step": 1818 + }, + { + "epoch": 0.581742589298207, + "grad_norm": 2.529588460922241, + "learning_rate": 0.00023584389070470087, + "loss": 4.4211, + "step": 1819 + }, + { + "epoch": 0.5820624038057927, + "grad_norm": 1.4366644620895386, + "learning_rate": 0.000235540272746726, + "loss": 4.3846, + "step": 1820 + }, + { + "epoch": 0.5823822183133782, + "grad_norm": 2.6791019439697266, + "learning_rate": 0.00023523672399252492, + "loss": 4.3954, + "step": 1821 + }, + { + "epoch": 0.5827020328209639, + "grad_norm": 21.338560104370117, + "learning_rate": 0.00023493324476798618, + "loss": 4.4056, + "step": 1822 + }, + { + "epoch": 0.5830218473285494, + "grad_norm": 1.5838208198547363, + "learning_rate": 0.0002346298353989245, + "loss": 4.3634, + "step": 1823 + }, + { + "epoch": 0.583341661836135, + "grad_norm": 3.7914466857910156, + "learning_rate": 0.00023432649621107928, + "loss": 4.4847, + "step": 1824 + }, + { + "epoch": 0.5836614763437207, + "grad_norm": 1.2197026014328003, + "learning_rate": 0.00023402322753011433, + "loss": 4.2956, + "step": 1825 + }, + { + "epoch": 0.5839812908513062, + "grad_norm": 1.6614536046981812, + "learning_rate": 0.0002337200296816184, + "loss": 4.3478, + "step": 1826 + }, + { + "epoch": 0.5843011053588918, + "grad_norm": 3.160306215286255, + "learning_rate": 0.00023341690299110338, + "loss": 4.4108, + "step": 1827 + }, + { + "epoch": 0.5846209198664775, + "grad_norm": 1.0466111898422241, + "learning_rate": 0.0002331138477840054, + "loss": 4.318, + "step": 1828 + }, + { + "epoch": 0.584940734374063, + "grad_norm": 2.481264352798462, + "learning_rate": 0.00023281086438568384, + "loss": 4.3597, + "step": 1829 + }, + { + "epoch": 0.5852605488816487, + "grad_norm": 25.556428909301758, + "learning_rate": 0.0002325079531214204, + "loss": 4.3998, + "step": 1830 + }, + { + "epoch": 0.5855803633892342, + "grad_norm": 2.172438859939575, + "learning_rate": 0.00023220511431642008, + "loss": 4.3622, + "step": 1831 + }, + { + "epoch": 0.5859001778968198, + "grad_norm": 2.3837709426879883, + "learning_rate": 0.00023190234829580943, + "loss": 4.391, + "step": 1832 + }, + { + "epoch": 0.5862199924044055, + "grad_norm": 1.1741242408752441, + "learning_rate": 0.00023159965538463738, + "loss": 4.3177, + "step": 1833 + }, + { + "epoch": 0.586539806911991, + "grad_norm": 1.4240802526474, + "learning_rate": 0.00023129703590787394, + "loss": 4.33, + "step": 1834 + }, + { + "epoch": 0.5868596214195766, + "grad_norm": 1.0304468870162964, + "learning_rate": 0.0002309944901904107, + "loss": 4.3558, + "step": 1835 + }, + { + "epoch": 0.5871794359271623, + "grad_norm": 1.0549205541610718, + "learning_rate": 0.00023069201855705973, + "loss": 4.3712, + "step": 1836 + }, + { + "epoch": 0.5874992504347478, + "grad_norm": 1.878810167312622, + "learning_rate": 0.00023038962133255366, + "loss": 4.3659, + "step": 1837 + }, + { + "epoch": 0.5878190649423335, + "grad_norm": 1.2352280616760254, + "learning_rate": 0.00023008729884154542, + "loss": 4.3454, + "step": 1838 + }, + { + "epoch": 0.588138879449919, + "grad_norm": 1.4102368354797363, + "learning_rate": 0.00022978505140860736, + "loss": 4.4269, + "step": 1839 + }, + { + "epoch": 0.5884586939575046, + "grad_norm": 4.818721771240234, + "learning_rate": 0.00022948287935823153, + "loss": 4.3837, + "step": 1840 + }, + { + "epoch": 0.5887785084650903, + "grad_norm": 2.3673958778381348, + "learning_rate": 0.00022918078301482897, + "loss": 4.3393, + "step": 1841 + }, + { + "epoch": 0.5890983229726758, + "grad_norm": 4.766218662261963, + "learning_rate": 0.00022887876270272938, + "loss": 4.3811, + "step": 1842 + }, + { + "epoch": 0.5894181374802614, + "grad_norm": 2.4529714584350586, + "learning_rate": 0.0002285768187461809, + "loss": 4.3912, + "step": 1843 + }, + { + "epoch": 0.5897379519878471, + "grad_norm": 2.070490598678589, + "learning_rate": 0.00022827495146934964, + "loss": 4.34, + "step": 1844 + }, + { + "epoch": 0.5900577664954326, + "grad_norm": 1.0089727640151978, + "learning_rate": 0.00022797316119631952, + "loss": 4.4047, + "step": 1845 + }, + { + "epoch": 0.5903775810030183, + "grad_norm": 1.5751157999038696, + "learning_rate": 0.00022767144825109153, + "loss": 4.3379, + "step": 1846 + }, + { + "epoch": 0.5906973955106039, + "grad_norm": 3.5401597023010254, + "learning_rate": 0.00022736981295758393, + "loss": 4.3677, + "step": 1847 + }, + { + "epoch": 0.5910172100181894, + "grad_norm": 4.936036109924316, + "learning_rate": 0.00022706825563963148, + "loss": 4.4361, + "step": 1848 + }, + { + "epoch": 0.5913370245257751, + "grad_norm": 10.496859550476074, + "learning_rate": 0.00022676677662098512, + "loss": 4.3945, + "step": 1849 + }, + { + "epoch": 0.5916568390333606, + "grad_norm": 1.452393889427185, + "learning_rate": 0.00022646537622531197, + "loss": 4.3704, + "step": 1850 + }, + { + "epoch": 0.5919766535409462, + "grad_norm": 1.7415666580200195, + "learning_rate": 0.00022616405477619448, + "loss": 4.42, + "step": 1851 + }, + { + "epoch": 0.5922964680485319, + "grad_norm": 3.5396132469177246, + "learning_rate": 0.00022586281259713055, + "loss": 4.3442, + "step": 1852 + }, + { + "epoch": 0.5926162825561174, + "grad_norm": 6.809600353240967, + "learning_rate": 0.00022556165001153295, + "loss": 4.4324, + "step": 1853 + }, + { + "epoch": 0.5929360970637031, + "grad_norm": 2.627774715423584, + "learning_rate": 0.0002252605673427288, + "loss": 4.451, + "step": 1854 + }, + { + "epoch": 0.5932559115712887, + "grad_norm": 3.084240436553955, + "learning_rate": 0.0002249595649139597, + "loss": 4.3921, + "step": 1855 + }, + { + "epoch": 0.5935757260788742, + "grad_norm": 1.3993078470230103, + "learning_rate": 0.0002246586430483809, + "loss": 4.319, + "step": 1856 + }, + { + "epoch": 0.5938955405864599, + "grad_norm": 1.2266440391540527, + "learning_rate": 0.00022435780206906132, + "loss": 4.4772, + "step": 1857 + }, + { + "epoch": 0.5942153550940454, + "grad_norm": 1.9506676197052002, + "learning_rate": 0.00022405704229898278, + "loss": 4.4421, + "step": 1858 + }, + { + "epoch": 0.5945351696016311, + "grad_norm": 1.1255192756652832, + "learning_rate": 0.00022375636406104022, + "loss": 4.3756, + "step": 1859 + }, + { + "epoch": 0.5948549841092167, + "grad_norm": 3.135082483291626, + "learning_rate": 0.00022345576767804085, + "loss": 4.3441, + "step": 1860 + }, + { + "epoch": 0.5951747986168022, + "grad_norm": 22.569377899169922, + "learning_rate": 0.00022315525347270412, + "loss": 4.3939, + "step": 1861 + }, + { + "epoch": 0.5954946131243879, + "grad_norm": 2.510176181793213, + "learning_rate": 0.00022285482176766122, + "loss": 4.3401, + "step": 1862 + }, + { + "epoch": 0.5958144276319735, + "grad_norm": 2.446462392807007, + "learning_rate": 0.00022255447288545453, + "loss": 4.3835, + "step": 1863 + }, + { + "epoch": 0.596134242139559, + "grad_norm": 7.099766731262207, + "learning_rate": 0.00022225420714853798, + "loss": 4.3613, + "step": 1864 + }, + { + "epoch": 0.5964540566471447, + "grad_norm": 2.6582837104797363, + "learning_rate": 0.00022195402487927592, + "loss": 4.3512, + "step": 1865 + }, + { + "epoch": 0.5967738711547302, + "grad_norm": 1.6339704990386963, + "learning_rate": 0.00022165392639994307, + "loss": 4.3388, + "step": 1866 + }, + { + "epoch": 0.5970936856623159, + "grad_norm": 2.6551482677459717, + "learning_rate": 0.00022135391203272441, + "loss": 4.3182, + "step": 1867 + }, + { + "epoch": 0.5974135001699015, + "grad_norm": 1.0081322193145752, + "learning_rate": 0.00022105398209971424, + "loss": 4.3782, + "step": 1868 + }, + { + "epoch": 0.597733314677487, + "grad_norm": 5.9165754318237305, + "learning_rate": 0.00022075413692291678, + "loss": 4.3066, + "step": 1869 + }, + { + "epoch": 0.5980531291850727, + "grad_norm": 9.98748779296875, + "learning_rate": 0.00022045437682424458, + "loss": 4.3654, + "step": 1870 + }, + { + "epoch": 0.5983729436926583, + "grad_norm": 10.65009880065918, + "learning_rate": 0.00022015470212551942, + "loss": 4.3706, + "step": 1871 + }, + { + "epoch": 0.5986927582002438, + "grad_norm": 3.0988478660583496, + "learning_rate": 0.00021985511314847128, + "loss": 4.3891, + "step": 1872 + }, + { + "epoch": 0.5990125727078295, + "grad_norm": 1.9206080436706543, + "learning_rate": 0.00021955561021473765, + "loss": 4.4231, + "step": 1873 + }, + { + "epoch": 0.599332387215415, + "grad_norm": 1.5138871669769287, + "learning_rate": 0.0002192561936458644, + "loss": 4.3312, + "step": 1874 + }, + { + "epoch": 0.5996522017230007, + "grad_norm": 1.5875927209854126, + "learning_rate": 0.00021895686376330396, + "loss": 4.2731, + "step": 1875 + }, + { + "epoch": 0.5999720162305863, + "grad_norm": 2.02584171295166, + "learning_rate": 0.00021865762088841607, + "loss": 4.261, + "step": 1876 + }, + { + "epoch": 0.6002918307381718, + "grad_norm": 2.500817060470581, + "learning_rate": 0.00021835846534246726, + "loss": 4.3411, + "step": 1877 + }, + { + "epoch": 0.6006116452457575, + "grad_norm": 7.956034183502197, + "learning_rate": 0.00021805939744662964, + "loss": 4.3752, + "step": 1878 + }, + { + "epoch": 0.6009314597533431, + "grad_norm": 1.7237507104873657, + "learning_rate": 0.00021776041752198202, + "loss": 4.3721, + "step": 1879 + }, + { + "epoch": 0.6012512742609286, + "grad_norm": 1.4250215291976929, + "learning_rate": 0.00021746152588950809, + "loss": 4.388, + "step": 1880 + }, + { + "epoch": 0.6015710887685143, + "grad_norm": 3.2383265495300293, + "learning_rate": 0.00021716272287009713, + "loss": 4.421, + "step": 1881 + }, + { + "epoch": 0.6018909032760998, + "grad_norm": 22.77423095703125, + "learning_rate": 0.00021686400878454312, + "loss": 4.2464, + "step": 1882 + }, + { + "epoch": 0.6022107177836855, + "grad_norm": 3.5048582553863525, + "learning_rate": 0.0002165653839535447, + "loss": 4.2907, + "step": 1883 + }, + { + "epoch": 0.6025305322912711, + "grad_norm": 1.813532829284668, + "learning_rate": 0.00021626684869770462, + "loss": 4.2947, + "step": 1884 + }, + { + "epoch": 0.6028503467988566, + "grad_norm": 1.8007831573486328, + "learning_rate": 0.00021596840333752934, + "loss": 4.3623, + "step": 1885 + }, + { + "epoch": 0.6031701613064423, + "grad_norm": 10.830092430114746, + "learning_rate": 0.00021567004819342907, + "loss": 4.4534, + "step": 1886 + }, + { + "epoch": 0.6034899758140279, + "grad_norm": 1.379032015800476, + "learning_rate": 0.00021537178358571686, + "loss": 4.3107, + "step": 1887 + }, + { + "epoch": 0.6038097903216134, + "grad_norm": 0.9854781627655029, + "learning_rate": 0.00021507360983460882, + "loss": 4.3697, + "step": 1888 + }, + { + "epoch": 0.6041296048291991, + "grad_norm": 10.340782165527344, + "learning_rate": 0.0002147755272602234, + "loss": 4.2789, + "step": 1889 + }, + { + "epoch": 0.6044494193367846, + "grad_norm": 7.905874729156494, + "learning_rate": 0.00021447753618258116, + "loss": 4.3998, + "step": 1890 + }, + { + "epoch": 0.6047692338443703, + "grad_norm": 9.338372230529785, + "learning_rate": 0.00021417963692160448, + "loss": 4.3027, + "step": 1891 + }, + { + "epoch": 0.6050890483519559, + "grad_norm": 10.630999565124512, + "learning_rate": 0.00021388182979711703, + "loss": 4.2862, + "step": 1892 + }, + { + "epoch": 0.6054088628595414, + "grad_norm": 8.252276420593262, + "learning_rate": 0.0002135841151288438, + "loss": 4.4429, + "step": 1893 + }, + { + "epoch": 0.6057286773671271, + "grad_norm": 10.241751670837402, + "learning_rate": 0.00021328649323641022, + "loss": 4.3186, + "step": 1894 + }, + { + "epoch": 0.6060484918747127, + "grad_norm": 2.167212963104248, + "learning_rate": 0.00021298896443934238, + "loss": 4.3333, + "step": 1895 + }, + { + "epoch": 0.6063683063822982, + "grad_norm": 2.1178698539733887, + "learning_rate": 0.00021269152905706637, + "loss": 4.3651, + "step": 1896 + }, + { + "epoch": 0.6066881208898839, + "grad_norm": 2.2207374572753906, + "learning_rate": 0.00021239418740890786, + "loss": 4.379, + "step": 1897 + }, + { + "epoch": 0.6070079353974694, + "grad_norm": 7.579524040222168, + "learning_rate": 0.000212096939814092, + "loss": 4.3855, + "step": 1898 + }, + { + "epoch": 0.6073277499050551, + "grad_norm": 3.2750277519226074, + "learning_rate": 0.00021179978659174284, + "loss": 4.3483, + "step": 1899 + }, + { + "epoch": 0.6076475644126407, + "grad_norm": 2.997750759124756, + "learning_rate": 0.00021150272806088333, + "loss": 4.3138, + "step": 1900 + }, + { + "epoch": 0.6076475644126407, + "eval_runtime": 21.9528, + "eval_samples_per_second": 86.413, + "eval_steps_per_second": 21.637, + "step": 1900 + }, + { + "epoch": 0.6079673789202262, + "grad_norm": 1.5388489961624146, + "learning_rate": 0.00021120576454043463, + "loss": 4.3193, + "step": 1901 + }, + { + "epoch": 0.6082871934278119, + "grad_norm": 1.5157217979431152, + "learning_rate": 0.00021090889634921585, + "loss": 4.4471, + "step": 1902 + }, + { + "epoch": 0.6086070079353975, + "grad_norm": 2.7461342811584473, + "learning_rate": 0.00021061212380594382, + "loss": 4.3919, + "step": 1903 + }, + { + "epoch": 0.608926822442983, + "grad_norm": 1.7940540313720703, + "learning_rate": 0.00021031544722923266, + "loss": 4.4844, + "step": 1904 + }, + { + "epoch": 0.6092466369505687, + "grad_norm": 1.6713277101516724, + "learning_rate": 0.0002100188669375935, + "loss": 4.3996, + "step": 1905 + }, + { + "epoch": 0.6095664514581542, + "grad_norm": 2.3862862586975098, + "learning_rate": 0.000209722383249434, + "loss": 4.2585, + "step": 1906 + }, + { + "epoch": 0.6098862659657399, + "grad_norm": 2.192728281021118, + "learning_rate": 0.0002094259964830582, + "loss": 4.296, + "step": 1907 + }, + { + "epoch": 0.6102060804733255, + "grad_norm": 1.2656503915786743, + "learning_rate": 0.0002091297069566662, + "loss": 4.2155, + "step": 1908 + }, + { + "epoch": 0.610525894980911, + "grad_norm": 2.4450693130493164, + "learning_rate": 0.00020883351498835335, + "loss": 4.4052, + "step": 1909 + }, + { + "epoch": 0.6108457094884967, + "grad_norm": 1.482854962348938, + "learning_rate": 0.00020853742089611067, + "loss": 4.4438, + "step": 1910 + }, + { + "epoch": 0.6111655239960823, + "grad_norm": 1.7138361930847168, + "learning_rate": 0.00020824142499782368, + "loss": 4.343, + "step": 1911 + }, + { + "epoch": 0.6114853385036678, + "grad_norm": 1.9445782899856567, + "learning_rate": 0.00020794552761127283, + "loss": 4.2304, + "step": 1912 + }, + { + "epoch": 0.6118051530112535, + "grad_norm": 2.1975700855255127, + "learning_rate": 0.0002076497290541328, + "loss": 4.351, + "step": 1913 + }, + { + "epoch": 0.612124967518839, + "grad_norm": 1.674682378768921, + "learning_rate": 0.0002073540296439719, + "loss": 4.277, + "step": 1914 + }, + { + "epoch": 0.6124447820264247, + "grad_norm": 3.3086462020874023, + "learning_rate": 0.00020705842969825225, + "loss": 4.4568, + "step": 1915 + }, + { + "epoch": 0.6127645965340103, + "grad_norm": 1.4060635566711426, + "learning_rate": 0.00020676292953432886, + "loss": 4.2999, + "step": 1916 + }, + { + "epoch": 0.6130844110415958, + "grad_norm": 1.9203327894210815, + "learning_rate": 0.00020646752946945016, + "loss": 4.3272, + "step": 1917 + }, + { + "epoch": 0.6134042255491815, + "grad_norm": 25.126161575317383, + "learning_rate": 0.00020617222982075646, + "loss": 4.2996, + "step": 1918 + }, + { + "epoch": 0.6137240400567671, + "grad_norm": 3.5299510955810547, + "learning_rate": 0.0002058770309052808, + "loss": 4.3422, + "step": 1919 + }, + { + "epoch": 0.6140438545643526, + "grad_norm": 2.4547934532165527, + "learning_rate": 0.00020558193303994797, + "loss": 4.3378, + "step": 1920 + }, + { + "epoch": 0.6143636690719383, + "grad_norm": 1.4542007446289062, + "learning_rate": 0.0002052869365415738, + "loss": 4.3702, + "step": 1921 + }, + { + "epoch": 0.6146834835795238, + "grad_norm": 1.7529739141464233, + "learning_rate": 0.00020499204172686616, + "loss": 4.3304, + "step": 1922 + }, + { + "epoch": 0.6150032980871095, + "grad_norm": 3.305755376815796, + "learning_rate": 0.00020469724891242281, + "loss": 4.32, + "step": 1923 + }, + { + "epoch": 0.6153231125946951, + "grad_norm": 1.4616423845291138, + "learning_rate": 0.00020440255841473252, + "loss": 4.3057, + "step": 1924 + }, + { + "epoch": 0.6156429271022806, + "grad_norm": 2.0238516330718994, + "learning_rate": 0.0002041079705501745, + "loss": 4.3172, + "step": 1925 + }, + { + "epoch": 0.6159627416098663, + "grad_norm": 3.9908838272094727, + "learning_rate": 0.00020381348563501694, + "loss": 4.2927, + "step": 1926 + }, + { + "epoch": 0.6162825561174519, + "grad_norm": 6.180693626403809, + "learning_rate": 0.00020351910398541835, + "loss": 4.4269, + "step": 1927 + }, + { + "epoch": 0.6166023706250375, + "grad_norm": 5.725349426269531, + "learning_rate": 0.00020322482591742576, + "loss": 4.3274, + "step": 1928 + }, + { + "epoch": 0.6169221851326231, + "grad_norm": 2.755397081375122, + "learning_rate": 0.0002029306517469754, + "loss": 4.3282, + "step": 1929 + }, + { + "epoch": 0.6172419996402086, + "grad_norm": 1.05733323097229, + "learning_rate": 0.00020263658178989162, + "loss": 4.3579, + "step": 1930 + }, + { + "epoch": 0.6175618141477943, + "grad_norm": 3.2340614795684814, + "learning_rate": 0.0002023426163618872, + "loss": 4.3141, + "step": 1931 + }, + { + "epoch": 0.6178816286553799, + "grad_norm": 4.539401531219482, + "learning_rate": 0.00020204875577856256, + "loss": 4.346, + "step": 1932 + }, + { + "epoch": 0.6182014431629654, + "grad_norm": 1.9272664785385132, + "learning_rate": 0.00020175500035540545, + "loss": 4.3174, + "step": 1933 + }, + { + "epoch": 0.6185212576705511, + "grad_norm": 2.688950538635254, + "learning_rate": 0.00020146135040779097, + "loss": 4.2532, + "step": 1934 + }, + { + "epoch": 0.6188410721781367, + "grad_norm": 2.4798758029937744, + "learning_rate": 0.0002011678062509807, + "loss": 4.3437, + "step": 1935 + }, + { + "epoch": 0.6191608866857223, + "grad_norm": 6.99690580368042, + "learning_rate": 0.00020087436820012287, + "loss": 4.3775, + "step": 1936 + }, + { + "epoch": 0.6194807011933079, + "grad_norm": 3.2319202423095703, + "learning_rate": 0.0002005810365702517, + "loss": 4.2907, + "step": 1937 + }, + { + "epoch": 0.6198005157008935, + "grad_norm": 1.4424253702163696, + "learning_rate": 0.00020028781167628714, + "loss": 4.3525, + "step": 1938 + }, + { + "epoch": 0.6201203302084791, + "grad_norm": 3.0073513984680176, + "learning_rate": 0.0001999946938330346, + "loss": 4.3052, + "step": 1939 + }, + { + "epoch": 0.6204401447160647, + "grad_norm": 1.3872689008712769, + "learning_rate": 0.0001997016833551845, + "loss": 4.362, + "step": 1940 + }, + { + "epoch": 0.6207599592236502, + "grad_norm": 1.3766533136367798, + "learning_rate": 0.00019940878055731208, + "loss": 4.2848, + "step": 1941 + }, + { + "epoch": 0.6210797737312359, + "grad_norm": 7.695796966552734, + "learning_rate": 0.00019911598575387683, + "loss": 4.2405, + "step": 1942 + }, + { + "epoch": 0.6213995882388215, + "grad_norm": 14.916356086730957, + "learning_rate": 0.00019882329925922245, + "loss": 4.3791, + "step": 1943 + }, + { + "epoch": 0.6217194027464071, + "grad_norm": 1.857189416885376, + "learning_rate": 0.00019853072138757637, + "loss": 4.3308, + "step": 1944 + }, + { + "epoch": 0.6220392172539927, + "grad_norm": 1.6914682388305664, + "learning_rate": 0.00019823825245304918, + "loss": 4.375, + "step": 1945 + }, + { + "epoch": 0.6223590317615783, + "grad_norm": 1.2899153232574463, + "learning_rate": 0.00019794589276963482, + "loss": 4.3765, + "step": 1946 + }, + { + "epoch": 0.6226788462691639, + "grad_norm": 3.3484532833099365, + "learning_rate": 0.00019765364265120962, + "loss": 4.3422, + "step": 1947 + }, + { + "epoch": 0.6229986607767495, + "grad_norm": 2.2809407711029053, + "learning_rate": 0.00019736150241153258, + "loss": 4.2945, + "step": 1948 + }, + { + "epoch": 0.623318475284335, + "grad_norm": 20.591279983520508, + "learning_rate": 0.0001970694723642446, + "loss": 4.2704, + "step": 1949 + }, + { + "epoch": 0.6236382897919207, + "grad_norm": 1.5780494213104248, + "learning_rate": 0.00019677755282286822, + "loss": 4.3213, + "step": 1950 + }, + { + "epoch": 0.6239581042995063, + "grad_norm": 2.9748880863189697, + "learning_rate": 0.00019648574410080743, + "loss": 4.3245, + "step": 1951 + }, + { + "epoch": 0.6242779188070919, + "grad_norm": 1.0901050567626953, + "learning_rate": 0.00019619404651134717, + "loss": 4.3388, + "step": 1952 + }, + { + "epoch": 0.6245977333146775, + "grad_norm": 2.94333815574646, + "learning_rate": 0.0001959024603676532, + "loss": 4.4353, + "step": 1953 + }, + { + "epoch": 0.624917547822263, + "grad_norm": 3.1907095909118652, + "learning_rate": 0.00019561098598277145, + "loss": 4.4397, + "step": 1954 + }, + { + "epoch": 0.6252373623298487, + "grad_norm": 1.8787250518798828, + "learning_rate": 0.000195319623669628, + "loss": 4.3464, + "step": 1955 + }, + { + "epoch": 0.6255571768374343, + "grad_norm": 2.4155571460723877, + "learning_rate": 0.00019502837374102866, + "loss": 4.3334, + "step": 1956 + }, + { + "epoch": 0.6258769913450198, + "grad_norm": 1.8786048889160156, + "learning_rate": 0.00019473723650965832, + "loss": 4.3839, + "step": 1957 + }, + { + "epoch": 0.6261968058526055, + "grad_norm": 2.3666434288024902, + "learning_rate": 0.0001944462122880813, + "loss": 4.3951, + "step": 1958 + }, + { + "epoch": 0.6265166203601911, + "grad_norm": 3.975346565246582, + "learning_rate": 0.00019415530138874, + "loss": 4.3177, + "step": 1959 + }, + { + "epoch": 0.6268364348677767, + "grad_norm": 29.15705680847168, + "learning_rate": 0.0001938645041239558, + "loss": 4.3062, + "step": 1960 + }, + { + "epoch": 0.6271562493753623, + "grad_norm": 2.425487756729126, + "learning_rate": 0.00019357382080592773, + "loss": 4.3544, + "step": 1961 + }, + { + "epoch": 0.6274760638829479, + "grad_norm": 3.139003038406372, + "learning_rate": 0.00019328325174673247, + "loss": 4.3879, + "step": 1962 + }, + { + "epoch": 0.6277958783905335, + "grad_norm": 3.1949918270111084, + "learning_rate": 0.0001929927972583242, + "loss": 4.4004, + "step": 1963 + }, + { + "epoch": 0.6281156928981191, + "grad_norm": 2.4575185775756836, + "learning_rate": 0.00019270245765253382, + "loss": 4.351, + "step": 1964 + }, + { + "epoch": 0.6284355074057046, + "grad_norm": 2.129354953765869, + "learning_rate": 0.0001924122332410694, + "loss": 4.4049, + "step": 1965 + }, + { + "epoch": 0.6287553219132903, + "grad_norm": 3.496196746826172, + "learning_rate": 0.00019212212433551465, + "loss": 4.3782, + "step": 1966 + }, + { + "epoch": 0.6290751364208759, + "grad_norm": 6.099643230438232, + "learning_rate": 0.00019183213124732986, + "loss": 4.4082, + "step": 1967 + }, + { + "epoch": 0.6293949509284615, + "grad_norm": 0.9659833908081055, + "learning_rate": 0.0001915422542878508, + "loss": 4.2431, + "step": 1968 + }, + { + "epoch": 0.6297147654360471, + "grad_norm": 2.8269596099853516, + "learning_rate": 0.00019125249376828824, + "loss": 4.3419, + "step": 1969 + }, + { + "epoch": 0.6300345799436327, + "grad_norm": 1.3621259927749634, + "learning_rate": 0.00019096284999972862, + "loss": 4.2266, + "step": 1970 + }, + { + "epoch": 0.6303543944512183, + "grad_norm": 1.7086843252182007, + "learning_rate": 0.00019067332329313226, + "loss": 4.2819, + "step": 1971 + }, + { + "epoch": 0.6306742089588039, + "grad_norm": 2.8977789878845215, + "learning_rate": 0.0001903839139593343, + "loss": 4.2329, + "step": 1972 + }, + { + "epoch": 0.6309940234663894, + "grad_norm": 1.4791843891143799, + "learning_rate": 0.00019009462230904398, + "loss": 4.3334, + "step": 1973 + }, + { + "epoch": 0.6313138379739751, + "grad_norm": 1.1379327774047852, + "learning_rate": 0.0001898054486528436, + "loss": 4.4114, + "step": 1974 + }, + { + "epoch": 0.6316336524815607, + "grad_norm": 3.3686492443084717, + "learning_rate": 0.00018951639330118953, + "loss": 4.3261, + "step": 1975 + }, + { + "epoch": 0.6319534669891463, + "grad_norm": 1.859269618988037, + "learning_rate": 0.0001892274565644104, + "loss": 4.2585, + "step": 1976 + }, + { + "epoch": 0.6322732814967319, + "grad_norm": 4.202902793884277, + "learning_rate": 0.000188938638752708, + "loss": 4.225, + "step": 1977 + }, + { + "epoch": 0.6325930960043175, + "grad_norm": 1.5296368598937988, + "learning_rate": 0.00018864994017615624, + "loss": 4.2678, + "step": 1978 + }, + { + "epoch": 0.6329129105119031, + "grad_norm": 2.759230136871338, + "learning_rate": 0.0001883613611447011, + "loss": 4.4147, + "step": 1979 + }, + { + "epoch": 0.6332327250194887, + "grad_norm": 1.6064265966415405, + "learning_rate": 0.00018807290196816022, + "loss": 4.266, + "step": 1980 + }, + { + "epoch": 0.6335525395270742, + "grad_norm": 5.738393783569336, + "learning_rate": 0.00018778456295622239, + "loss": 4.2743, + "step": 1981 + }, + { + "epoch": 0.6338723540346599, + "grad_norm": 1.3017499446868896, + "learning_rate": 0.00018749634441844764, + "loss": 4.3081, + "step": 1982 + }, + { + "epoch": 0.6341921685422455, + "grad_norm": 21.420686721801758, + "learning_rate": 0.00018720824666426647, + "loss": 4.3348, + "step": 1983 + }, + { + "epoch": 0.6345119830498311, + "grad_norm": 3.171353578567505, + "learning_rate": 0.00018692027000297986, + "loss": 4.381, + "step": 1984 + }, + { + "epoch": 0.6348317975574167, + "grad_norm": 1.9541981220245361, + "learning_rate": 0.0001866324147437587, + "loss": 4.3963, + "step": 1985 + }, + { + "epoch": 0.6351516120650023, + "grad_norm": 1.0255112648010254, + "learning_rate": 0.00018634468119564342, + "loss": 4.3437, + "step": 1986 + }, + { + "epoch": 0.6354714265725879, + "grad_norm": 2.1944940090179443, + "learning_rate": 0.00018605706966754408, + "loss": 4.2612, + "step": 1987 + }, + { + "epoch": 0.6357912410801735, + "grad_norm": 7.297569274902344, + "learning_rate": 0.00018576958046823944, + "loss": 4.2822, + "step": 1988 + }, + { + "epoch": 0.636111055587759, + "grad_norm": 2.677182912826538, + "learning_rate": 0.0001854822139063772, + "loss": 4.3697, + "step": 1989 + }, + { + "epoch": 0.6364308700953447, + "grad_norm": 6.630171775817871, + "learning_rate": 0.00018519497029047307, + "loss": 4.3418, + "step": 1990 + }, + { + "epoch": 0.6367506846029303, + "grad_norm": 1.100528597831726, + "learning_rate": 0.00018490784992891107, + "loss": 4.376, + "step": 1991 + }, + { + "epoch": 0.6370704991105159, + "grad_norm": 1.4696028232574463, + "learning_rate": 0.00018462085312994278, + "loss": 4.3104, + "step": 1992 + }, + { + "epoch": 0.6373903136181015, + "grad_norm": 1.4809238910675049, + "learning_rate": 0.0001843339802016871, + "loss": 4.4346, + "step": 1993 + }, + { + "epoch": 0.6377101281256871, + "grad_norm": 0.9368327260017395, + "learning_rate": 0.00018404723145212993, + "loss": 4.3128, + "step": 1994 + }, + { + "epoch": 0.6380299426332727, + "grad_norm": 2.330907106399536, + "learning_rate": 0.00018376060718912392, + "loss": 4.325, + "step": 1995 + }, + { + "epoch": 0.6383497571408583, + "grad_norm": 1.8129738569259644, + "learning_rate": 0.00018347410772038807, + "loss": 4.2841, + "step": 1996 + }, + { + "epoch": 0.638669571648444, + "grad_norm": 7.109486103057861, + "learning_rate": 0.00018318773335350723, + "loss": 4.2779, + "step": 1997 + }, + { + "epoch": 0.6389893861560295, + "grad_norm": 3.0293161869049072, + "learning_rate": 0.0001829014843959322, + "loss": 4.3104, + "step": 1998 + }, + { + "epoch": 0.6393092006636151, + "grad_norm": 1.1990113258361816, + "learning_rate": 0.00018261536115497904, + "loss": 4.3464, + "step": 1999 + }, + { + "epoch": 0.6396290151712007, + "grad_norm": 1.508091926574707, + "learning_rate": 0.0001823293639378287, + "loss": 4.3292, + "step": 2000 + }, + { + "epoch": 0.6396290151712007, + "eval_runtime": 21.1981, + "eval_samples_per_second": 89.489, + "eval_steps_per_second": 22.408, + "step": 2000 + }, + { + "epoch": 0.6399488296787863, + "grad_norm": 1.6650848388671875, + "learning_rate": 0.00018204349305152707, + "loss": 4.2386, + "step": 2001 + }, + { + "epoch": 0.6402686441863719, + "grad_norm": 4.6853928565979, + "learning_rate": 0.00018175774880298422, + "loss": 4.4185, + "step": 2002 + }, + { + "epoch": 0.6405884586939575, + "grad_norm": 1.0524141788482666, + "learning_rate": 0.0001814721314989743, + "loss": 4.4103, + "step": 2003 + }, + { + "epoch": 0.6409082732015431, + "grad_norm": 1.3180805444717407, + "learning_rate": 0.0001811866414461354, + "loss": 4.3315, + "step": 2004 + }, + { + "epoch": 0.6412280877091288, + "grad_norm": 5.738834857940674, + "learning_rate": 0.00018090127895096855, + "loss": 4.243, + "step": 2005 + }, + { + "epoch": 0.6415479022167143, + "grad_norm": 1.5634493827819824, + "learning_rate": 0.0001806160443198383, + "loss": 4.3308, + "step": 2006 + }, + { + "epoch": 0.6418677167242999, + "grad_norm": 0.9018966555595398, + "learning_rate": 0.00018033093785897145, + "loss": 4.2621, + "step": 2007 + }, + { + "epoch": 0.6421875312318855, + "grad_norm": 2.196821928024292, + "learning_rate": 0.00018004595987445782, + "loss": 4.3625, + "step": 2008 + }, + { + "epoch": 0.6425073457394711, + "grad_norm": 2.2245755195617676, + "learning_rate": 0.00017976111067224854, + "loss": 4.3808, + "step": 2009 + }, + { + "epoch": 0.6428271602470567, + "grad_norm": 1.9312374591827393, + "learning_rate": 0.00017947639055815713, + "loss": 4.3598, + "step": 2010 + }, + { + "epoch": 0.6431469747546423, + "grad_norm": 6.4608025550842285, + "learning_rate": 0.00017919179983785828, + "loss": 4.3475, + "step": 2011 + }, + { + "epoch": 0.6434667892622279, + "grad_norm": 1.5554989576339722, + "learning_rate": 0.00017890733881688754, + "loss": 4.3772, + "step": 2012 + }, + { + "epoch": 0.6437866037698136, + "grad_norm": 2.9698777198791504, + "learning_rate": 0.0001786230078006417, + "loss": 4.2504, + "step": 2013 + }, + { + "epoch": 0.6441064182773991, + "grad_norm": 2.6870768070220947, + "learning_rate": 0.00017833880709437752, + "loss": 4.3112, + "step": 2014 + }, + { + "epoch": 0.6444262327849847, + "grad_norm": 1.8793983459472656, + "learning_rate": 0.00017805473700321193, + "loss": 4.2448, + "step": 2015 + }, + { + "epoch": 0.6447460472925703, + "grad_norm": 2.0156643390655518, + "learning_rate": 0.00017777079783212215, + "loss": 4.392, + "step": 2016 + }, + { + "epoch": 0.6450658618001559, + "grad_norm": 2.6315724849700928, + "learning_rate": 0.00017748698988594394, + "loss": 4.3411, + "step": 2017 + }, + { + "epoch": 0.6453856763077415, + "grad_norm": 7.722999572753906, + "learning_rate": 0.00017720331346937317, + "loss": 4.2887, + "step": 2018 + }, + { + "epoch": 0.6457054908153271, + "grad_norm": 1.1182135343551636, + "learning_rate": 0.0001769197688869636, + "loss": 4.2463, + "step": 2019 + }, + { + "epoch": 0.6460253053229127, + "grad_norm": 2.5218985080718994, + "learning_rate": 0.0001766363564431281, + "loss": 4.2819, + "step": 2020 + }, + { + "epoch": 0.6463451198304984, + "grad_norm": 0.9212332963943481, + "learning_rate": 0.00017635307644213726, + "loss": 4.2956, + "step": 2021 + }, + { + "epoch": 0.6466649343380839, + "grad_norm": 1.9455363750457764, + "learning_rate": 0.00017606992918811976, + "loss": 4.271, + "step": 2022 + }, + { + "epoch": 0.6469847488456695, + "grad_norm": 1.6982426643371582, + "learning_rate": 0.00017578691498506177, + "loss": 4.2886, + "step": 2023 + }, + { + "epoch": 0.6473045633532551, + "grad_norm": 1.0481082201004028, + "learning_rate": 0.00017550403413680625, + "loss": 4.3683, + "step": 2024 + }, + { + "epoch": 0.6476243778608407, + "grad_norm": 4.87114953994751, + "learning_rate": 0.00017522128694705347, + "loss": 4.2511, + "step": 2025 + }, + { + "epoch": 0.6479441923684263, + "grad_norm": 1.1189478635787964, + "learning_rate": 0.0001749386737193598, + "loss": 4.2944, + "step": 2026 + }, + { + "epoch": 0.6482640068760119, + "grad_norm": 0.9406677484512329, + "learning_rate": 0.00017465619475713813, + "loss": 4.3349, + "step": 2027 + }, + { + "epoch": 0.6485838213835975, + "grad_norm": 1.5378926992416382, + "learning_rate": 0.00017437385036365695, + "loss": 4.333, + "step": 2028 + }, + { + "epoch": 0.6489036358911832, + "grad_norm": 2.2596795558929443, + "learning_rate": 0.00017409164084204037, + "loss": 4.2701, + "step": 2029 + }, + { + "epoch": 0.6492234503987687, + "grad_norm": 3.071223735809326, + "learning_rate": 0.00017380956649526785, + "loss": 4.2832, + "step": 2030 + }, + { + "epoch": 0.6495432649063543, + "grad_norm": 3.295929193496704, + "learning_rate": 0.00017352762762617334, + "loss": 4.286, + "step": 2031 + }, + { + "epoch": 0.64986307941394, + "grad_norm": 2.5586657524108887, + "learning_rate": 0.00017324582453744577, + "loss": 4.283, + "step": 2032 + }, + { + "epoch": 0.6501828939215255, + "grad_norm": 1.5775175094604492, + "learning_rate": 0.00017296415753162786, + "loss": 4.3016, + "step": 2033 + }, + { + "epoch": 0.6505027084291111, + "grad_norm": 1.0437132120132446, + "learning_rate": 0.00017268262691111675, + "loss": 4.3675, + "step": 2034 + }, + { + "epoch": 0.6508225229366967, + "grad_norm": 1.6079857349395752, + "learning_rate": 0.0001724012329781629, + "loss": 4.2957, + "step": 2035 + }, + { + "epoch": 0.6511423374442823, + "grad_norm": 10.98817253112793, + "learning_rate": 0.0001721199760348698, + "loss": 4.3161, + "step": 2036 + }, + { + "epoch": 0.651462151951868, + "grad_norm": 1.3609607219696045, + "learning_rate": 0.00017183885638319426, + "loss": 4.3111, + "step": 2037 + }, + { + "epoch": 0.6517819664594535, + "grad_norm": 5.901271820068359, + "learning_rate": 0.00017155787432494529, + "loss": 4.3277, + "step": 2038 + }, + { + "epoch": 0.6521017809670391, + "grad_norm": 1.7044252157211304, + "learning_rate": 0.00017127703016178445, + "loss": 4.3659, + "step": 2039 + }, + { + "epoch": 0.6524215954746247, + "grad_norm": 3.839336395263672, + "learning_rate": 0.00017099632419522552, + "loss": 4.374, + "step": 2040 + }, + { + "epoch": 0.6527414099822103, + "grad_norm": 2.5221285820007324, + "learning_rate": 0.00017071575672663325, + "loss": 4.2821, + "step": 2041 + }, + { + "epoch": 0.6530612244897959, + "grad_norm": 6.933326721191406, + "learning_rate": 0.0001704353280572243, + "loss": 4.4177, + "step": 2042 + }, + { + "epoch": 0.6533810389973815, + "grad_norm": 5.539220809936523, + "learning_rate": 0.0001701550384880658, + "loss": 4.2902, + "step": 2043 + }, + { + "epoch": 0.6537008535049671, + "grad_norm": 3.834388494491577, + "learning_rate": 0.00016987488832007593, + "loss": 4.332, + "step": 2044 + }, + { + "epoch": 0.6540206680125528, + "grad_norm": 1.0735085010528564, + "learning_rate": 0.00016959487785402313, + "loss": 4.2956, + "step": 2045 + }, + { + "epoch": 0.6543404825201383, + "grad_norm": 1.5837112665176392, + "learning_rate": 0.00016931500739052576, + "loss": 4.3307, + "step": 2046 + }, + { + "epoch": 0.6546602970277239, + "grad_norm": 1.4215750694274902, + "learning_rate": 0.00016903527723005206, + "loss": 4.3729, + "step": 2047 + }, + { + "epoch": 0.6549801115353096, + "grad_norm": 0.9576036930084229, + "learning_rate": 0.0001687556876729193, + "loss": 4.3558, + "step": 2048 + }, + { + "epoch": 0.6552999260428951, + "grad_norm": 1.402064561843872, + "learning_rate": 0.00016847623901929408, + "loss": 4.2334, + "step": 2049 + }, + { + "epoch": 0.6556197405504807, + "grad_norm": 1.5215175151824951, + "learning_rate": 0.00016819693156919167, + "loss": 4.237, + "step": 2050 + }, + { + "epoch": 0.6559395550580663, + "grad_norm": 1.7449076175689697, + "learning_rate": 0.00016791776562247572, + "loss": 4.2177, + "step": 2051 + }, + { + "epoch": 0.6562593695656519, + "grad_norm": 1.0422505140304565, + "learning_rate": 0.0001676387414788581, + "loss": 4.1756, + "step": 2052 + }, + { + "epoch": 0.6565791840732376, + "grad_norm": 1.0781958103179932, + "learning_rate": 0.00016735985943789808, + "loss": 4.2988, + "step": 2053 + }, + { + "epoch": 0.6568989985808231, + "grad_norm": 1.1183533668518066, + "learning_rate": 0.0001670811197990027, + "loss": 4.1867, + "step": 2054 + }, + { + "epoch": 0.6572188130884087, + "grad_norm": 19.493942260742188, + "learning_rate": 0.000166802522861426, + "loss": 4.2072, + "step": 2055 + }, + { + "epoch": 0.6575386275959944, + "grad_norm": 4.562228679656982, + "learning_rate": 0.00016652406892426902, + "loss": 4.3206, + "step": 2056 + }, + { + "epoch": 0.6578584421035799, + "grad_norm": 1.4225586652755737, + "learning_rate": 0.00016624575828647878, + "loss": 4.2823, + "step": 2057 + }, + { + "epoch": 0.6581782566111655, + "grad_norm": 1.0276960134506226, + "learning_rate": 0.0001659675912468489, + "loss": 4.3248, + "step": 2058 + }, + { + "epoch": 0.6584980711187511, + "grad_norm": 2.2550899982452393, + "learning_rate": 0.00016568956810401867, + "loss": 4.2659, + "step": 2059 + }, + { + "epoch": 0.6588178856263367, + "grad_norm": 0.9947875142097473, + "learning_rate": 0.00016541168915647298, + "loss": 4.2402, + "step": 2060 + }, + { + "epoch": 0.6591377001339224, + "grad_norm": 0.8552061915397644, + "learning_rate": 0.00016513395470254194, + "loss": 4.2895, + "step": 2061 + }, + { + "epoch": 0.6594575146415079, + "grad_norm": 4.5843000411987305, + "learning_rate": 0.00016485636504040015, + "loss": 4.3039, + "step": 2062 + }, + { + "epoch": 0.6597773291490935, + "grad_norm": 1.5974023342132568, + "learning_rate": 0.00016457892046806727, + "loss": 4.2506, + "step": 2063 + }, + { + "epoch": 0.6600971436566792, + "grad_norm": 1.1103382110595703, + "learning_rate": 0.00016430162128340693, + "loss": 4.2402, + "step": 2064 + }, + { + "epoch": 0.6604169581642647, + "grad_norm": 1.044456124305725, + "learning_rate": 0.0001640244677841267, + "loss": 4.3088, + "step": 2065 + }, + { + "epoch": 0.6607367726718504, + "grad_norm": 1.8953404426574707, + "learning_rate": 0.00016374746026777794, + "loss": 4.2672, + "step": 2066 + }, + { + "epoch": 0.6610565871794359, + "grad_norm": 1.1261482238769531, + "learning_rate": 0.0001634705990317548, + "loss": 4.3079, + "step": 2067 + }, + { + "epoch": 0.6613764016870215, + "grad_norm": 2.279447555541992, + "learning_rate": 0.00016319388437329482, + "loss": 4.2885, + "step": 2068 + }, + { + "epoch": 0.6616962161946072, + "grad_norm": 1.2991628646850586, + "learning_rate": 0.00016291731658947808, + "loss": 4.2594, + "step": 2069 + }, + { + "epoch": 0.6620160307021927, + "grad_norm": 1.6666761636734009, + "learning_rate": 0.0001626408959772269, + "loss": 4.3335, + "step": 2070 + }, + { + "epoch": 0.6623358452097783, + "grad_norm": 1.6983450651168823, + "learning_rate": 0.00016236462283330578, + "loss": 4.2403, + "step": 2071 + }, + { + "epoch": 0.662655659717364, + "grad_norm": 1.3658196926116943, + "learning_rate": 0.0001620884974543205, + "loss": 4.3243, + "step": 2072 + }, + { + "epoch": 0.6629754742249495, + "grad_norm": 11.376717567443848, + "learning_rate": 0.00016181252013671858, + "loss": 4.2995, + "step": 2073 + }, + { + "epoch": 0.6632952887325352, + "grad_norm": 1.299176573753357, + "learning_rate": 0.00016153669117678848, + "loss": 4.3288, + "step": 2074 + }, + { + "epoch": 0.6636151032401207, + "grad_norm": 1.6570117473602295, + "learning_rate": 0.00016126101087065933, + "loss": 4.3569, + "step": 2075 + }, + { + "epoch": 0.6639349177477063, + "grad_norm": 0.99131840467453, + "learning_rate": 0.00016098547951430082, + "loss": 4.3299, + "step": 2076 + }, + { + "epoch": 0.664254732255292, + "grad_norm": 1.6765660047531128, + "learning_rate": 0.00016071009740352237, + "loss": 4.3393, + "step": 2077 + }, + { + "epoch": 0.6645745467628775, + "grad_norm": 1.6409902572631836, + "learning_rate": 0.0001604348648339736, + "loss": 4.2569, + "step": 2078 + }, + { + "epoch": 0.6648943612704631, + "grad_norm": 1.7353764772415161, + "learning_rate": 0.0001601597821011431, + "loss": 4.3707, + "step": 2079 + }, + { + "epoch": 0.6652141757780488, + "grad_norm": 1.1918566226959229, + "learning_rate": 0.0001598848495003593, + "loss": 4.2434, + "step": 2080 + }, + { + "epoch": 0.6655339902856343, + "grad_norm": 1.164758324623108, + "learning_rate": 0.00015961006732678873, + "loss": 4.316, + "step": 2081 + }, + { + "epoch": 0.66585380479322, + "grad_norm": 1.488234281539917, + "learning_rate": 0.00015933543587543682, + "loss": 4.2669, + "step": 2082 + }, + { + "epoch": 0.6661736193008055, + "grad_norm": 2.990535020828247, + "learning_rate": 0.0001590609554411472, + "loss": 4.3715, + "step": 2083 + }, + { + "epoch": 0.6664934338083911, + "grad_norm": 1.1554975509643555, + "learning_rate": 0.0001587866263186009, + "loss": 4.2782, + "step": 2084 + }, + { + "epoch": 0.6668132483159768, + "grad_norm": 1.1111619472503662, + "learning_rate": 0.0001585124488023173, + "loss": 4.3747, + "step": 2085 + }, + { + "epoch": 0.6671330628235623, + "grad_norm": 2.607558488845825, + "learning_rate": 0.00015823842318665233, + "loss": 4.1682, + "step": 2086 + }, + { + "epoch": 0.6674528773311479, + "grad_norm": 1.0271543264389038, + "learning_rate": 0.00015796454976579901, + "loss": 4.3432, + "step": 2087 + }, + { + "epoch": 0.6677726918387336, + "grad_norm": 1.0558197498321533, + "learning_rate": 0.00015769082883378737, + "loss": 4.2375, + "step": 2088 + }, + { + "epoch": 0.6680925063463191, + "grad_norm": 3.2165932655334473, + "learning_rate": 0.00015741726068448293, + "loss": 4.2687, + "step": 2089 + }, + { + "epoch": 0.6684123208539048, + "grad_norm": 1.5289154052734375, + "learning_rate": 0.0001571438456115881, + "loss": 4.2737, + "step": 2090 + }, + { + "epoch": 0.6687321353614903, + "grad_norm": 4.825890064239502, + "learning_rate": 0.0001568705839086402, + "loss": 4.3358, + "step": 2091 + }, + { + "epoch": 0.6690519498690759, + "grad_norm": 2.0604193210601807, + "learning_rate": 0.00015659747586901243, + "loss": 4.3345, + "step": 2092 + }, + { + "epoch": 0.6693717643766616, + "grad_norm": 1.527177333831787, + "learning_rate": 0.00015632452178591252, + "loss": 4.2679, + "step": 2093 + }, + { + "epoch": 0.6696915788842471, + "grad_norm": 12.681228637695312, + "learning_rate": 0.00015605172195238314, + "loss": 4.2765, + "step": 2094 + }, + { + "epoch": 0.6700113933918327, + "grad_norm": 2.7059664726257324, + "learning_rate": 0.00015577907666130178, + "loss": 4.3767, + "step": 2095 + }, + { + "epoch": 0.6703312078994184, + "grad_norm": 7.503986835479736, + "learning_rate": 0.00015550658620537932, + "loss": 4.3817, + "step": 2096 + }, + { + "epoch": 0.6706510224070039, + "grad_norm": 1.5696512460708618, + "learning_rate": 0.0001552342508771608, + "loss": 4.2039, + "step": 2097 + }, + { + "epoch": 0.6709708369145896, + "grad_norm": 1.8507659435272217, + "learning_rate": 0.00015496207096902457, + "loss": 4.2754, + "step": 2098 + }, + { + "epoch": 0.6712906514221751, + "grad_norm": 1.5928252935409546, + "learning_rate": 0.00015469004677318214, + "loss": 4.305, + "step": 2099 + }, + { + "epoch": 0.6716104659297607, + "grad_norm": 1.1693589687347412, + "learning_rate": 0.000154418178581678, + "loss": 4.3088, + "step": 2100 + }, + { + "epoch": 0.6716104659297607, + "eval_runtime": 21.2011, + "eval_samples_per_second": 89.477, + "eval_steps_per_second": 22.405, + "step": 2100 + }, + { + "epoch": 0.6719302804373464, + "grad_norm": 4.87490177154541, + "learning_rate": 0.00015414646668638897, + "loss": 4.2437, + "step": 2101 + }, + { + "epoch": 0.6722500949449319, + "grad_norm": 1.2594079971313477, + "learning_rate": 0.00015387491137902428, + "loss": 4.2845, + "step": 2102 + }, + { + "epoch": 0.6725699094525175, + "grad_norm": 2.604454755783081, + "learning_rate": 0.00015360351295112468, + "loss": 4.2268, + "step": 2103 + }, + { + "epoch": 0.6728897239601032, + "grad_norm": 3.125438928604126, + "learning_rate": 0.00015333227169406284, + "loss": 4.3158, + "step": 2104 + }, + { + "epoch": 0.6732095384676887, + "grad_norm": 1.85345458984375, + "learning_rate": 0.0001530611878990426, + "loss": 4.2548, + "step": 2105 + }, + { + "epoch": 0.6735293529752744, + "grad_norm": 2.351177215576172, + "learning_rate": 0.00015279026185709865, + "loss": 4.222, + "step": 2106 + }, + { + "epoch": 0.67384916748286, + "grad_norm": 1.6685160398483276, + "learning_rate": 0.0001525194938590966, + "loss": 4.339, + "step": 2107 + }, + { + "epoch": 0.6741689819904455, + "grad_norm": 1.827606201171875, + "learning_rate": 0.0001522488841957319, + "loss": 4.3243, + "step": 2108 + }, + { + "epoch": 0.6744887964980312, + "grad_norm": 1.2187941074371338, + "learning_rate": 0.00015197843315753034, + "loss": 4.3781, + "step": 2109 + }, + { + "epoch": 0.6748086110056167, + "grad_norm": 1.1468909978866577, + "learning_rate": 0.00015170814103484747, + "loss": 4.2843, + "step": 2110 + }, + { + "epoch": 0.6751284255132023, + "grad_norm": 4.3299560546875, + "learning_rate": 0.00015143800811786805, + "loss": 4.288, + "step": 2111 + }, + { + "epoch": 0.675448240020788, + "grad_norm": 2.718714475631714, + "learning_rate": 0.00015116803469660616, + "loss": 4.3275, + "step": 2112 + }, + { + "epoch": 0.6757680545283735, + "grad_norm": 3.339301109313965, + "learning_rate": 0.00015089822106090418, + "loss": 4.3508, + "step": 2113 + }, + { + "epoch": 0.6760878690359592, + "grad_norm": 1.2896249294281006, + "learning_rate": 0.00015062856750043343, + "loss": 4.2493, + "step": 2114 + }, + { + "epoch": 0.6764076835435447, + "grad_norm": 0.7643463015556335, + "learning_rate": 0.00015035907430469304, + "loss": 4.3036, + "step": 2115 + }, + { + "epoch": 0.6767274980511303, + "grad_norm": 3.2152016162872314, + "learning_rate": 0.00015008974176301031, + "loss": 4.3232, + "step": 2116 + }, + { + "epoch": 0.677047312558716, + "grad_norm": 1.41158926486969, + "learning_rate": 0.00014982057016453969, + "loss": 4.2823, + "step": 2117 + }, + { + "epoch": 0.6773671270663015, + "grad_norm": 1.202553153038025, + "learning_rate": 0.00014955155979826302, + "loss": 4.2899, + "step": 2118 + }, + { + "epoch": 0.6776869415738871, + "grad_norm": 1.0611287355422974, + "learning_rate": 0.00014928271095298912, + "loss": 4.1584, + "step": 2119 + }, + { + "epoch": 0.6780067560814728, + "grad_norm": 0.8936102986335754, + "learning_rate": 0.00014901402391735328, + "loss": 4.2912, + "step": 2120 + }, + { + "epoch": 0.6783265705890583, + "grad_norm": 2.928180456161499, + "learning_rate": 0.00014874549897981725, + "loss": 4.2426, + "step": 2121 + }, + { + "epoch": 0.678646385096644, + "grad_norm": 0.9338362216949463, + "learning_rate": 0.00014847713642866835, + "loss": 4.3567, + "step": 2122 + }, + { + "epoch": 0.6789661996042295, + "grad_norm": 5.4628777503967285, + "learning_rate": 0.00014820893655201998, + "loss": 4.2677, + "step": 2123 + }, + { + "epoch": 0.6792860141118151, + "grad_norm": 0.8840541839599609, + "learning_rate": 0.0001479408996378107, + "loss": 4.247, + "step": 2124 + }, + { + "epoch": 0.6796058286194008, + "grad_norm": 3.629621982574463, + "learning_rate": 0.00014767302597380418, + "loss": 4.3792, + "step": 2125 + }, + { + "epoch": 0.6799256431269863, + "grad_norm": 1.2234013080596924, + "learning_rate": 0.0001474053158475889, + "loss": 4.2608, + "step": 2126 + }, + { + "epoch": 0.6802454576345719, + "grad_norm": 1.1823759078979492, + "learning_rate": 0.00014713776954657743, + "loss": 4.251, + "step": 2127 + }, + { + "epoch": 0.6805652721421576, + "grad_norm": 2.4592196941375732, + "learning_rate": 0.00014687038735800693, + "loss": 4.3664, + "step": 2128 + }, + { + "epoch": 0.6808850866497431, + "grad_norm": 1.0833048820495605, + "learning_rate": 0.0001466031695689378, + "loss": 4.2593, + "step": 2129 + }, + { + "epoch": 0.6812049011573288, + "grad_norm": 1.074945330619812, + "learning_rate": 0.0001463361164662546, + "loss": 4.2459, + "step": 2130 + }, + { + "epoch": 0.6815247156649143, + "grad_norm": 1.1903793811798096, + "learning_rate": 0.00014606922833666476, + "loss": 4.2693, + "step": 2131 + }, + { + "epoch": 0.6818445301724999, + "grad_norm": 1.6766566038131714, + "learning_rate": 0.00014580250546669836, + "loss": 4.3149, + "step": 2132 + }, + { + "epoch": 0.6821643446800856, + "grad_norm": 2.5219738483428955, + "learning_rate": 0.0001455359481427085, + "loss": 4.3391, + "step": 2133 + }, + { + "epoch": 0.6824841591876711, + "grad_norm": 2.312596559524536, + "learning_rate": 0.00014526955665087013, + "loss": 4.2328, + "step": 2134 + }, + { + "epoch": 0.6828039736952568, + "grad_norm": 48.63291549682617, + "learning_rate": 0.00014500333127718035, + "loss": 4.3363, + "step": 2135 + }, + { + "epoch": 0.6831237882028424, + "grad_norm": 1.133103609085083, + "learning_rate": 0.00014473727230745833, + "loss": 4.2708, + "step": 2136 + }, + { + "epoch": 0.6834436027104279, + "grad_norm": 0.90025794506073, + "learning_rate": 0.0001444713800273438, + "loss": 4.3179, + "step": 2137 + }, + { + "epoch": 0.6837634172180136, + "grad_norm": 0.846717894077301, + "learning_rate": 0.0001442056547222982, + "loss": 4.3319, + "step": 2138 + }, + { + "epoch": 0.6840832317255992, + "grad_norm": 2.3262157440185547, + "learning_rate": 0.0001439400966776032, + "loss": 4.2759, + "step": 2139 + }, + { + "epoch": 0.6844030462331847, + "grad_norm": 1.3408619165420532, + "learning_rate": 0.00014367470617836117, + "loss": 4.3991, + "step": 2140 + }, + { + "epoch": 0.6847228607407704, + "grad_norm": 4.441734790802002, + "learning_rate": 0.00014340948350949467, + "loss": 4.4114, + "step": 2141 + }, + { + "epoch": 0.6850426752483559, + "grad_norm": 3.942800998687744, + "learning_rate": 0.00014314442895574595, + "loss": 4.3253, + "step": 2142 + }, + { + "epoch": 0.6853624897559416, + "grad_norm": 0.8225685358047485, + "learning_rate": 0.00014287954280167695, + "loss": 4.2311, + "step": 2143 + }, + { + "epoch": 0.6856823042635272, + "grad_norm": 10.745675086975098, + "learning_rate": 0.00014261482533166832, + "loss": 4.1605, + "step": 2144 + }, + { + "epoch": 0.6860021187711127, + "grad_norm": 6.6501874923706055, + "learning_rate": 0.0001423502768299202, + "loss": 4.252, + "step": 2145 + }, + { + "epoch": 0.6863219332786984, + "grad_norm": 1.2363886833190918, + "learning_rate": 0.00014208589758045098, + "loss": 4.2931, + "step": 2146 + }, + { + "epoch": 0.686641747786284, + "grad_norm": 2.7055091857910156, + "learning_rate": 0.00014182168786709755, + "loss": 4.2846, + "step": 2147 + }, + { + "epoch": 0.6869615622938695, + "grad_norm": 1.2917979955673218, + "learning_rate": 0.00014155764797351472, + "loss": 4.3626, + "step": 2148 + }, + { + "epoch": 0.6872813768014552, + "grad_norm": 4.229926586151123, + "learning_rate": 0.0001412937781831747, + "loss": 4.3873, + "step": 2149 + }, + { + "epoch": 0.6876011913090407, + "grad_norm": 4.406290054321289, + "learning_rate": 0.0001410300787793675, + "loss": 4.2838, + "step": 2150 + }, + { + "epoch": 0.6879210058166264, + "grad_norm": 0.8523226380348206, + "learning_rate": 0.00014076655004519997, + "loss": 4.2662, + "step": 2151 + }, + { + "epoch": 0.688240820324212, + "grad_norm": 1.9754862785339355, + "learning_rate": 0.00014050319226359593, + "loss": 4.3747, + "step": 2152 + }, + { + "epoch": 0.6885606348317975, + "grad_norm": 7.5988240242004395, + "learning_rate": 0.00014024000571729526, + "loss": 4.2471, + "step": 2153 + }, + { + "epoch": 0.6888804493393832, + "grad_norm": 1.6670303344726562, + "learning_rate": 0.00013997699068885443, + "loss": 4.2472, + "step": 2154 + }, + { + "epoch": 0.6892002638469688, + "grad_norm": 1.4159529209136963, + "learning_rate": 0.00013971414746064554, + "loss": 4.3069, + "step": 2155 + }, + { + "epoch": 0.6895200783545543, + "grad_norm": 0.7494491338729858, + "learning_rate": 0.00013945147631485634, + "loss": 4.2243, + "step": 2156 + }, + { + "epoch": 0.68983989286214, + "grad_norm": 0.9474825859069824, + "learning_rate": 0.00013918897753348991, + "loss": 4.2411, + "step": 2157 + }, + { + "epoch": 0.6901597073697255, + "grad_norm": 1.8350554704666138, + "learning_rate": 0.00013892665139836392, + "loss": 4.2476, + "step": 2158 + }, + { + "epoch": 0.6904795218773112, + "grad_norm": 1.4637562036514282, + "learning_rate": 0.0001386644981911111, + "loss": 4.3424, + "step": 2159 + }, + { + "epoch": 0.6907993363848968, + "grad_norm": 1.469955563545227, + "learning_rate": 0.00013840251819317832, + "loss": 4.2649, + "step": 2160 + }, + { + "epoch": 0.6911191508924823, + "grad_norm": 1.1518014669418335, + "learning_rate": 0.00013814071168582654, + "loss": 4.2405, + "step": 2161 + }, + { + "epoch": 0.691438965400068, + "grad_norm": 0.897095799446106, + "learning_rate": 0.00013787907895013054, + "loss": 4.2588, + "step": 2162 + }, + { + "epoch": 0.6917587799076536, + "grad_norm": 0.9591032862663269, + "learning_rate": 0.0001376176202669783, + "loss": 4.2272, + "step": 2163 + }, + { + "epoch": 0.6920785944152391, + "grad_norm": 2.1118273735046387, + "learning_rate": 0.00013735633591707117, + "loss": 4.3202, + "step": 2164 + }, + { + "epoch": 0.6923984089228248, + "grad_norm": 1.7391027212142944, + "learning_rate": 0.00013709522618092328, + "loss": 4.2258, + "step": 2165 + }, + { + "epoch": 0.6927182234304103, + "grad_norm": 1.6449528932571411, + "learning_rate": 0.00013683429133886122, + "loss": 4.2519, + "step": 2166 + }, + { + "epoch": 0.693038037937996, + "grad_norm": 1.4094949960708618, + "learning_rate": 0.00013657353167102401, + "loss": 4.3069, + "step": 2167 + }, + { + "epoch": 0.6933578524455816, + "grad_norm": 1.53618586063385, + "learning_rate": 0.00013631294745736227, + "loss": 4.3098, + "step": 2168 + }, + { + "epoch": 0.6936776669531671, + "grad_norm": 1.1365201473236084, + "learning_rate": 0.0001360525389776385, + "loss": 4.2188, + "step": 2169 + }, + { + "epoch": 0.6939974814607528, + "grad_norm": 1.0074905157089233, + "learning_rate": 0.00013579230651142654, + "loss": 4.258, + "step": 2170 + }, + { + "epoch": 0.6943172959683384, + "grad_norm": 21.4325008392334, + "learning_rate": 0.00013553225033811114, + "loss": 4.2978, + "step": 2171 + }, + { + "epoch": 0.6946371104759239, + "grad_norm": 1.105054497718811, + "learning_rate": 0.00013527237073688797, + "loss": 4.3735, + "step": 2172 + }, + { + "epoch": 0.6949569249835096, + "grad_norm": 1.158308506011963, + "learning_rate": 0.00013501266798676283, + "loss": 4.3072, + "step": 2173 + }, + { + "epoch": 0.6952767394910951, + "grad_norm": 0.9725875854492188, + "learning_rate": 0.000134753142366552, + "loss": 4.2638, + "step": 2174 + }, + { + "epoch": 0.6955965539986808, + "grad_norm": 11.04317569732666, + "learning_rate": 0.0001344937941548811, + "loss": 4.3209, + "step": 2175 + }, + { + "epoch": 0.6959163685062664, + "grad_norm": 1.3237169981002808, + "learning_rate": 0.00013423462363018604, + "loss": 4.1653, + "step": 2176 + }, + { + "epoch": 0.6962361830138519, + "grad_norm": 4.11657190322876, + "learning_rate": 0.00013397563107071125, + "loss": 4.2468, + "step": 2177 + }, + { + "epoch": 0.6965559975214376, + "grad_norm": 1.3359613418579102, + "learning_rate": 0.0001337168167545104, + "loss": 4.3698, + "step": 2178 + }, + { + "epoch": 0.6968758120290232, + "grad_norm": 1.8677021265029907, + "learning_rate": 0.000133458180959446, + "loss": 4.3064, + "step": 2179 + }, + { + "epoch": 0.6971956265366087, + "grad_norm": 3.530254602432251, + "learning_rate": 0.00013319972396318828, + "loss": 4.2538, + "step": 2180 + }, + { + "epoch": 0.6975154410441944, + "grad_norm": 1.1801210641860962, + "learning_rate": 0.00013294144604321633, + "loss": 4.3102, + "step": 2181 + }, + { + "epoch": 0.6978352555517799, + "grad_norm": 1.5837007761001587, + "learning_rate": 0.00013268334747681626, + "loss": 4.227, + "step": 2182 + }, + { + "epoch": 0.6981550700593656, + "grad_norm": 0.8676749467849731, + "learning_rate": 0.0001324254285410821, + "loss": 4.3138, + "step": 2183 + }, + { + "epoch": 0.6984748845669512, + "grad_norm": 6.718521595001221, + "learning_rate": 0.0001321676895129149, + "loss": 4.25, + "step": 2184 + }, + { + "epoch": 0.6987946990745367, + "grad_norm": 4.645453929901123, + "learning_rate": 0.0001319101306690222, + "loss": 4.2314, + "step": 2185 + }, + { + "epoch": 0.6991145135821224, + "grad_norm": 0.8738028407096863, + "learning_rate": 0.0001316527522859189, + "loss": 4.2904, + "step": 2186 + }, + { + "epoch": 0.699434328089708, + "grad_norm": 5.937346935272217, + "learning_rate": 0.00013139555463992527, + "loss": 4.2043, + "step": 2187 + }, + { + "epoch": 0.6997541425972935, + "grad_norm": 1.4075615406036377, + "learning_rate": 0.00013113853800716824, + "loss": 4.2601, + "step": 2188 + }, + { + "epoch": 0.7000739571048792, + "grad_norm": 8.532005310058594, + "learning_rate": 0.00013088170266357986, + "loss": 4.2916, + "step": 2189 + }, + { + "epoch": 0.7003937716124647, + "grad_norm": 1.1289187669754028, + "learning_rate": 0.00013062504888489788, + "loss": 4.3012, + "step": 2190 + }, + { + "epoch": 0.7007135861200504, + "grad_norm": 1.6149331331253052, + "learning_rate": 0.0001303685769466651, + "loss": 4.3652, + "step": 2191 + }, + { + "epoch": 0.701033400627636, + "grad_norm": 2.135042428970337, + "learning_rate": 0.00013011228712422898, + "loss": 4.269, + "step": 2192 + }, + { + "epoch": 0.7013532151352215, + "grad_norm": 1.581347942352295, + "learning_rate": 0.0001298561796927417, + "loss": 4.3691, + "step": 2193 + }, + { + "epoch": 0.7016730296428072, + "grad_norm": 3.475996255874634, + "learning_rate": 0.00012960025492715914, + "loss": 4.2961, + "step": 2194 + }, + { + "epoch": 0.7019928441503928, + "grad_norm": 1.6966077089309692, + "learning_rate": 0.0001293445131022416, + "loss": 4.2515, + "step": 2195 + }, + { + "epoch": 0.7023126586579783, + "grad_norm": 1.1822967529296875, + "learning_rate": 0.00012908895449255262, + "loss": 4.1737, + "step": 2196 + }, + { + "epoch": 0.702632473165564, + "grad_norm": 1.8335177898406982, + "learning_rate": 0.0001288335793724592, + "loss": 4.3145, + "step": 2197 + }, + { + "epoch": 0.7029522876731495, + "grad_norm": 1.297350287437439, + "learning_rate": 0.00012857838801613153, + "loss": 4.2761, + "step": 2198 + }, + { + "epoch": 0.7032721021807352, + "grad_norm": 2.0235729217529297, + "learning_rate": 0.000128323380697542, + "loss": 4.1373, + "step": 2199 + }, + { + "epoch": 0.7035919166883208, + "grad_norm": 4.431934356689453, + "learning_rate": 0.0001280685576904658, + "loss": 4.2745, + "step": 2200 + }, + { + "epoch": 0.7035919166883208, + "eval_runtime": 22.1492, + "eval_samples_per_second": 85.647, + "eval_steps_per_second": 21.445, + "step": 2200 + }, + { + "epoch": 0.7039117311959063, + "grad_norm": 1.2721166610717773, + "learning_rate": 0.0001278139192684802, + "loss": 4.2631, + "step": 2201 + }, + { + "epoch": 0.704231545703492, + "grad_norm": 2.1965203285217285, + "learning_rate": 0.00012755946570496427, + "loss": 4.2525, + "step": 2202 + }, + { + "epoch": 0.7045513602110776, + "grad_norm": 1.465809941291809, + "learning_rate": 0.0001273051972730987, + "loss": 4.1846, + "step": 2203 + }, + { + "epoch": 0.7048711747186632, + "grad_norm": 0.8960003852844238, + "learning_rate": 0.00012705111424586512, + "loss": 4.3356, + "step": 2204 + }, + { + "epoch": 0.7051909892262488, + "grad_norm": 1.2165048122406006, + "learning_rate": 0.00012679721689604642, + "loss": 4.2884, + "step": 2205 + }, + { + "epoch": 0.7055108037338343, + "grad_norm": 1.1528109312057495, + "learning_rate": 0.00012654350549622605, + "loss": 4.2858, + "step": 2206 + }, + { + "epoch": 0.70583061824142, + "grad_norm": 0.746715784072876, + "learning_rate": 0.00012628998031878784, + "loss": 4.2189, + "step": 2207 + }, + { + "epoch": 0.7061504327490056, + "grad_norm": 0.8898685574531555, + "learning_rate": 0.00012603664163591573, + "loss": 4.331, + "step": 2208 + }, + { + "epoch": 0.7064702472565911, + "grad_norm": 1.220816731452942, + "learning_rate": 0.00012578348971959324, + "loss": 4.2126, + "step": 2209 + }, + { + "epoch": 0.7067900617641768, + "grad_norm": 2.303783655166626, + "learning_rate": 0.0001255305248416036, + "loss": 4.2043, + "step": 2210 + }, + { + "epoch": 0.7071098762717624, + "grad_norm": 2.597175359725952, + "learning_rate": 0.0001252777472735291, + "loss": 4.3002, + "step": 2211 + }, + { + "epoch": 0.707429690779348, + "grad_norm": 1.139110803604126, + "learning_rate": 0.00012502515728675124, + "loss": 4.2495, + "step": 2212 + }, + { + "epoch": 0.7077495052869336, + "grad_norm": 0.7789515256881714, + "learning_rate": 0.00012477275515244951, + "loss": 4.2906, + "step": 2213 + }, + { + "epoch": 0.7080693197945191, + "grad_norm": 2.000424861907959, + "learning_rate": 0.00012452054114160232, + "loss": 4.3076, + "step": 2214 + }, + { + "epoch": 0.7083891343021048, + "grad_norm": 1.188720703125, + "learning_rate": 0.00012426851552498584, + "loss": 4.2672, + "step": 2215 + }, + { + "epoch": 0.7087089488096904, + "grad_norm": 1.0890549421310425, + "learning_rate": 0.00012401667857317406, + "loss": 4.2493, + "step": 2216 + }, + { + "epoch": 0.7090287633172759, + "grad_norm": 1.1436631679534912, + "learning_rate": 0.0001237650305565385, + "loss": 4.3899, + "step": 2217 + }, + { + "epoch": 0.7093485778248616, + "grad_norm": 0.7897758483886719, + "learning_rate": 0.00012351357174524745, + "loss": 4.3085, + "step": 2218 + }, + { + "epoch": 0.7096683923324472, + "grad_norm": 1.3549960851669312, + "learning_rate": 0.00012326230240926653, + "loss": 4.3382, + "step": 2219 + }, + { + "epoch": 0.7099882068400328, + "grad_norm": 2.543323040008545, + "learning_rate": 0.00012301122281835772, + "loss": 4.3002, + "step": 2220 + }, + { + "epoch": 0.7103080213476184, + "grad_norm": 1.1650097370147705, + "learning_rate": 0.00012276033324207935, + "loss": 4.1195, + "step": 2221 + }, + { + "epoch": 0.710627835855204, + "grad_norm": 1.4357078075408936, + "learning_rate": 0.00012250963394978584, + "loss": 4.3082, + "step": 2222 + }, + { + "epoch": 0.7109476503627896, + "grad_norm": 1.0691293478012085, + "learning_rate": 0.00012225912521062702, + "loss": 4.3223, + "step": 2223 + }, + { + "epoch": 0.7112674648703752, + "grad_norm": 2.8936126232147217, + "learning_rate": 0.00012200880729354847, + "loss": 4.3065, + "step": 2224 + }, + { + "epoch": 0.7115872793779607, + "grad_norm": 1.7941921949386597, + "learning_rate": 0.0001217586804672905, + "loss": 4.2445, + "step": 2225 + }, + { + "epoch": 0.7119070938855464, + "grad_norm": 1.9436687231063843, + "learning_rate": 0.0001215087450003889, + "loss": 4.3569, + "step": 2226 + }, + { + "epoch": 0.712226908393132, + "grad_norm": 5.292397499084473, + "learning_rate": 0.00012125900116117357, + "loss": 4.0625, + "step": 2227 + }, + { + "epoch": 0.7125467229007176, + "grad_norm": 1.2170920372009277, + "learning_rate": 0.0001210094492177686, + "loss": 4.2152, + "step": 2228 + }, + { + "epoch": 0.7128665374083032, + "grad_norm": 1.2025548219680786, + "learning_rate": 0.00012076008943809238, + "loss": 4.2667, + "step": 2229 + }, + { + "epoch": 0.7131863519158887, + "grad_norm": 1.6722089052200317, + "learning_rate": 0.00012051092208985671, + "loss": 4.2841, + "step": 2230 + }, + { + "epoch": 0.7135061664234744, + "grad_norm": 1.8300215005874634, + "learning_rate": 0.00012026194744056684, + "loss": 4.2428, + "step": 2231 + }, + { + "epoch": 0.71382598093106, + "grad_norm": 1.2550748586654663, + "learning_rate": 0.00012001316575752159, + "loss": 4.2557, + "step": 2232 + }, + { + "epoch": 0.7141457954386455, + "grad_norm": 1.046065330505371, + "learning_rate": 0.00011976457730781191, + "loss": 4.3212, + "step": 2233 + }, + { + "epoch": 0.7144656099462312, + "grad_norm": 1.5702720880508423, + "learning_rate": 0.00011951618235832183, + "loss": 4.2367, + "step": 2234 + }, + { + "epoch": 0.7147854244538168, + "grad_norm": 1.0399309396743774, + "learning_rate": 0.00011926798117572722, + "loss": 4.3435, + "step": 2235 + }, + { + "epoch": 0.7151052389614024, + "grad_norm": 1.2550163269042969, + "learning_rate": 0.00011901997402649629, + "loss": 4.2829, + "step": 2236 + }, + { + "epoch": 0.715425053468988, + "grad_norm": 2.1564676761627197, + "learning_rate": 0.00011877216117688875, + "loss": 4.2509, + "step": 2237 + }, + { + "epoch": 0.7157448679765736, + "grad_norm": 2.321582317352295, + "learning_rate": 0.00011852454289295575, + "loss": 4.1779, + "step": 2238 + }, + { + "epoch": 0.7160646824841592, + "grad_norm": 1.0394196510314941, + "learning_rate": 0.00011827711944053962, + "loss": 4.2051, + "step": 2239 + }, + { + "epoch": 0.7163844969917448, + "grad_norm": 0.9859041571617126, + "learning_rate": 0.00011802989108527331, + "loss": 4.3341, + "step": 2240 + }, + { + "epoch": 0.7167043114993303, + "grad_norm": 4.739549160003662, + "learning_rate": 0.00011778285809258052, + "loss": 4.2586, + "step": 2241 + }, + { + "epoch": 0.717024126006916, + "grad_norm": 4.265996932983398, + "learning_rate": 0.00011753602072767514, + "loss": 4.2754, + "step": 2242 + }, + { + "epoch": 0.7173439405145016, + "grad_norm": 1.0207865238189697, + "learning_rate": 0.00011728937925556107, + "loss": 4.1966, + "step": 2243 + }, + { + "epoch": 0.7176637550220872, + "grad_norm": 1.6690582036972046, + "learning_rate": 0.00011704293394103194, + "loss": 4.2261, + "step": 2244 + }, + { + "epoch": 0.7179835695296728, + "grad_norm": 2.1228718757629395, + "learning_rate": 0.00011679668504867051, + "loss": 4.3243, + "step": 2245 + }, + { + "epoch": 0.7183033840372584, + "grad_norm": 0.9624343514442444, + "learning_rate": 0.00011655063284284901, + "loss": 4.1826, + "step": 2246 + }, + { + "epoch": 0.718623198544844, + "grad_norm": 1.5692168474197388, + "learning_rate": 0.0001163047775877283, + "loss": 4.2021, + "step": 2247 + }, + { + "epoch": 0.7189430130524296, + "grad_norm": 1.4274945259094238, + "learning_rate": 0.00011605911954725802, + "loss": 4.2196, + "step": 2248 + }, + { + "epoch": 0.7192628275600151, + "grad_norm": 0.7484304904937744, + "learning_rate": 0.00011581365898517567, + "loss": 4.1319, + "step": 2249 + }, + { + "epoch": 0.7195826420676008, + "grad_norm": 1.4674358367919922, + "learning_rate": 0.0001155683961650071, + "loss": 4.3073, + "step": 2250 + }, + { + "epoch": 0.7199024565751864, + "grad_norm": 5.066298961639404, + "learning_rate": 0.00011532333135006579, + "loss": 4.1771, + "step": 2251 + }, + { + "epoch": 0.720222271082772, + "grad_norm": 1.7535251379013062, + "learning_rate": 0.00011507846480345255, + "loss": 4.2991, + "step": 2252 + }, + { + "epoch": 0.7205420855903576, + "grad_norm": 0.758303165435791, + "learning_rate": 0.00011483379678805551, + "loss": 4.1732, + "step": 2253 + }, + { + "epoch": 0.7208619000979432, + "grad_norm": 2.9513726234436035, + "learning_rate": 0.00011458932756654938, + "loss": 4.2438, + "step": 2254 + }, + { + "epoch": 0.7211817146055288, + "grad_norm": 0.8239002227783203, + "learning_rate": 0.00011434505740139558, + "loss": 4.2282, + "step": 2255 + }, + { + "epoch": 0.7215015291131144, + "grad_norm": 2.2233285903930664, + "learning_rate": 0.00011410098655484194, + "loss": 4.2268, + "step": 2256 + }, + { + "epoch": 0.7218213436206999, + "grad_norm": 2.914273977279663, + "learning_rate": 0.00011385711528892216, + "loss": 4.2424, + "step": 2257 + }, + { + "epoch": 0.7221411581282856, + "grad_norm": 1.124178171157837, + "learning_rate": 0.00011361344386545585, + "loss": 4.3186, + "step": 2258 + }, + { + "epoch": 0.7224609726358712, + "grad_norm": 6.065934181213379, + "learning_rate": 0.00011336997254604769, + "loss": 4.2216, + "step": 2259 + }, + { + "epoch": 0.7227807871434568, + "grad_norm": 16.960329055786133, + "learning_rate": 0.0001131267015920879, + "loss": 4.2733, + "step": 2260 + }, + { + "epoch": 0.7231006016510424, + "grad_norm": 0.7926948070526123, + "learning_rate": 0.0001128836312647514, + "loss": 4.1824, + "step": 2261 + }, + { + "epoch": 0.723420416158628, + "grad_norm": 1.125644564628601, + "learning_rate": 0.00011264076182499787, + "loss": 4.3319, + "step": 2262 + }, + { + "epoch": 0.7237402306662136, + "grad_norm": 1.2370562553405762, + "learning_rate": 0.00011239809353357127, + "loss": 4.3201, + "step": 2263 + }, + { + "epoch": 0.7240600451737992, + "grad_norm": 22.526020050048828, + "learning_rate": 0.00011215562665099941, + "loss": 4.1381, + "step": 2264 + }, + { + "epoch": 0.7243798596813849, + "grad_norm": 11.185654640197754, + "learning_rate": 0.00011191336143759417, + "loss": 4.2814, + "step": 2265 + }, + { + "epoch": 0.7246996741889704, + "grad_norm": 2.1497981548309326, + "learning_rate": 0.00011167129815345048, + "loss": 4.2567, + "step": 2266 + }, + { + "epoch": 0.725019488696556, + "grad_norm": 1.911161184310913, + "learning_rate": 0.0001114294370584471, + "loss": 4.206, + "step": 2267 + }, + { + "epoch": 0.7253393032041416, + "grad_norm": 2.1089258193969727, + "learning_rate": 0.00011118777841224534, + "loss": 4.282, + "step": 2268 + }, + { + "epoch": 0.7256591177117272, + "grad_norm": 0.89681077003479, + "learning_rate": 0.00011094632247428907, + "loss": 4.1651, + "step": 2269 + }, + { + "epoch": 0.7259789322193128, + "grad_norm": 0.9843345284461975, + "learning_rate": 0.00011070506950380483, + "loss": 4.1979, + "step": 2270 + }, + { + "epoch": 0.7262987467268984, + "grad_norm": 0.8795956373214722, + "learning_rate": 0.0001104640197598008, + "loss": 4.2308, + "step": 2271 + }, + { + "epoch": 0.726618561234484, + "grad_norm": 1.8232970237731934, + "learning_rate": 0.00011022317350106774, + "loss": 4.2521, + "step": 2272 + }, + { + "epoch": 0.7269383757420697, + "grad_norm": 1.4000154733657837, + "learning_rate": 0.00010998253098617707, + "loss": 4.2844, + "step": 2273 + }, + { + "epoch": 0.7272581902496552, + "grad_norm": 2.3723652362823486, + "learning_rate": 0.00010974209247348211, + "loss": 4.2448, + "step": 2274 + }, + { + "epoch": 0.7275780047572408, + "grad_norm": 4.57194709777832, + "learning_rate": 0.00010950185822111697, + "loss": 4.1836, + "step": 2275 + }, + { + "epoch": 0.7278978192648264, + "grad_norm": 1.340511441230774, + "learning_rate": 0.00010926182848699613, + "loss": 4.2324, + "step": 2276 + }, + { + "epoch": 0.728217633772412, + "grad_norm": 0.8756692409515381, + "learning_rate": 0.00010902200352881522, + "loss": 4.3927, + "step": 2277 + }, + { + "epoch": 0.7285374482799976, + "grad_norm": 2.3351075649261475, + "learning_rate": 0.00010878238360404934, + "loss": 4.2118, + "step": 2278 + }, + { + "epoch": 0.7288572627875832, + "grad_norm": 1.4256292581558228, + "learning_rate": 0.00010854296896995379, + "loss": 4.326, + "step": 2279 + }, + { + "epoch": 0.7291770772951688, + "grad_norm": 1.524228811264038, + "learning_rate": 0.00010830375988356354, + "loss": 4.2238, + "step": 2280 + }, + { + "epoch": 0.7294968918027545, + "grad_norm": 1.78737473487854, + "learning_rate": 0.00010806475660169243, + "loss": 4.2169, + "step": 2281 + }, + { + "epoch": 0.72981670631034, + "grad_norm": 2.7804081439971924, + "learning_rate": 0.00010782595938093417, + "loss": 4.3032, + "step": 2282 + }, + { + "epoch": 0.7301365208179256, + "grad_norm": 1.6948069334030151, + "learning_rate": 0.00010758736847766033, + "loss": 4.2235, + "step": 2283 + }, + { + "epoch": 0.7304563353255112, + "grad_norm": 3.867884635925293, + "learning_rate": 0.00010734898414802169, + "loss": 4.2197, + "step": 2284 + }, + { + "epoch": 0.7307761498330968, + "grad_norm": 0.8450716137886047, + "learning_rate": 0.00010711080664794676, + "loss": 4.2675, + "step": 2285 + }, + { + "epoch": 0.7310959643406824, + "grad_norm": 1.6160097122192383, + "learning_rate": 0.00010687283623314225, + "loss": 4.2256, + "step": 2286 + }, + { + "epoch": 0.731415778848268, + "grad_norm": 1.0513497591018677, + "learning_rate": 0.00010663507315909255, + "loss": 4.2763, + "step": 2287 + }, + { + "epoch": 0.7317355933558536, + "grad_norm": 3.876077175140381, + "learning_rate": 0.00010639751768105936, + "loss": 4.2419, + "step": 2288 + }, + { + "epoch": 0.7320554078634393, + "grad_norm": 3.397737741470337, + "learning_rate": 0.00010616017005408167, + "loss": 4.2355, + "step": 2289 + }, + { + "epoch": 0.7323752223710248, + "grad_norm": 3.1993489265441895, + "learning_rate": 0.00010592303053297499, + "loss": 4.22, + "step": 2290 + }, + { + "epoch": 0.7326950368786104, + "grad_norm": 2.7461044788360596, + "learning_rate": 0.00010568609937233168, + "loss": 4.2291, + "step": 2291 + }, + { + "epoch": 0.733014851386196, + "grad_norm": 2.497793197631836, + "learning_rate": 0.00010544937682652035, + "loss": 4.2476, + "step": 2292 + }, + { + "epoch": 0.7333346658937816, + "grad_norm": 0.8880087733268738, + "learning_rate": 0.00010521286314968567, + "loss": 4.2024, + "step": 2293 + }, + { + "epoch": 0.7336544804013672, + "grad_norm": 1.373935580253601, + "learning_rate": 0.00010497655859574809, + "loss": 4.2292, + "step": 2294 + }, + { + "epoch": 0.7339742949089528, + "grad_norm": 2.3294129371643066, + "learning_rate": 0.00010474046341840329, + "loss": 4.2534, + "step": 2295 + }, + { + "epoch": 0.7342941094165384, + "grad_norm": 1.4270519018173218, + "learning_rate": 0.00010450457787112246, + "loss": 4.2613, + "step": 2296 + }, + { + "epoch": 0.7346139239241241, + "grad_norm": 1.5154391527175903, + "learning_rate": 0.00010426890220715164, + "loss": 4.1997, + "step": 2297 + }, + { + "epoch": 0.7349337384317096, + "grad_norm": 0.7951825261116028, + "learning_rate": 0.00010403343667951149, + "loss": 4.1306, + "step": 2298 + }, + { + "epoch": 0.7352535529392952, + "grad_norm": 1.782274603843689, + "learning_rate": 0.00010379818154099724, + "loss": 4.2362, + "step": 2299 + }, + { + "epoch": 0.7355733674468808, + "grad_norm": 10.45250415802002, + "learning_rate": 0.00010356313704417794, + "loss": 4.2309, + "step": 2300 + }, + { + "epoch": 0.7355733674468808, + "eval_runtime": 21.1689, + "eval_samples_per_second": 89.613, + "eval_steps_per_second": 22.439, + "step": 2300 + }, + { + "epoch": 0.7358931819544664, + "grad_norm": 6.215102195739746, + "learning_rate": 0.0001033283034413967, + "loss": 4.1588, + "step": 2301 + }, + { + "epoch": 0.736212996462052, + "grad_norm": 1.601189374923706, + "learning_rate": 0.00010309368098477025, + "loss": 4.2849, + "step": 2302 + }, + { + "epoch": 0.7365328109696376, + "grad_norm": 1.6865421533584595, + "learning_rate": 0.00010285926992618855, + "loss": 4.1874, + "step": 2303 + }, + { + "epoch": 0.7368526254772232, + "grad_norm": 1.3950881958007812, + "learning_rate": 0.0001026250705173147, + "loss": 4.1924, + "step": 2304 + }, + { + "epoch": 0.7371724399848089, + "grad_norm": 2.692542552947998, + "learning_rate": 0.00010239108300958432, + "loss": 4.2827, + "step": 2305 + }, + { + "epoch": 0.7374922544923944, + "grad_norm": 1.5706571340560913, + "learning_rate": 0.00010215730765420579, + "loss": 4.2558, + "step": 2306 + }, + { + "epoch": 0.73781206899998, + "grad_norm": 1.6970666646957397, + "learning_rate": 0.00010192374470215969, + "loss": 4.2546, + "step": 2307 + }, + { + "epoch": 0.7381318835075656, + "grad_norm": 1.492335557937622, + "learning_rate": 0.00010169039440419855, + "loss": 4.2776, + "step": 2308 + }, + { + "epoch": 0.7384516980151512, + "grad_norm": 1.0124475955963135, + "learning_rate": 0.00010145725701084643, + "loss": 4.1941, + "step": 2309 + }, + { + "epoch": 0.7387715125227368, + "grad_norm": 1.1548728942871094, + "learning_rate": 0.000101224332772399, + "loss": 4.1421, + "step": 2310 + }, + { + "epoch": 0.7390913270303224, + "grad_norm": 1.2299275398254395, + "learning_rate": 0.00010099162193892303, + "loss": 4.2813, + "step": 2311 + }, + { + "epoch": 0.739411141537908, + "grad_norm": 1.7321922779083252, + "learning_rate": 0.00010075912476025623, + "loss": 4.305, + "step": 2312 + }, + { + "epoch": 0.7397309560454937, + "grad_norm": 1.1434346437454224, + "learning_rate": 0.000100526841486007, + "loss": 4.1581, + "step": 2313 + }, + { + "epoch": 0.7400507705530792, + "grad_norm": 3.048884630203247, + "learning_rate": 0.00010029477236555372, + "loss": 4.24, + "step": 2314 + }, + { + "epoch": 0.7403705850606648, + "grad_norm": 0.8909955620765686, + "learning_rate": 0.00010006291764804523, + "loss": 4.2303, + "step": 2315 + }, + { + "epoch": 0.7406903995682504, + "grad_norm": 1.9216283559799194, + "learning_rate": 9.98312775824001e-05, + "loss": 4.175, + "step": 2316 + }, + { + "epoch": 0.741010214075836, + "grad_norm": 1.0550016164779663, + "learning_rate": 9.959985241730641e-05, + "loss": 4.2451, + "step": 2317 + }, + { + "epoch": 0.7413300285834216, + "grad_norm": 10.487397193908691, + "learning_rate": 9.936864240122164e-05, + "loss": 4.1845, + "step": 2318 + }, + { + "epoch": 0.7416498430910072, + "grad_norm": 1.522336483001709, + "learning_rate": 9.913764778237196e-05, + "loss": 4.1885, + "step": 2319 + }, + { + "epoch": 0.7419696575985928, + "grad_norm": 1.266965627670288, + "learning_rate": 9.890686880875274e-05, + "loss": 4.3285, + "step": 2320 + }, + { + "epoch": 0.7422894721061785, + "grad_norm": 5.041074275970459, + "learning_rate": 9.86763057281273e-05, + "loss": 4.1895, + "step": 2321 + }, + { + "epoch": 0.742609286613764, + "grad_norm": 1.4574302434921265, + "learning_rate": 9.844595878802778e-05, + "loss": 4.1565, + "step": 2322 + }, + { + "epoch": 0.7429291011213496, + "grad_norm": 0.9724259376525879, + "learning_rate": 9.821582823575398e-05, + "loss": 4.2667, + "step": 2323 + }, + { + "epoch": 0.7432489156289352, + "grad_norm": 3.8262031078338623, + "learning_rate": 9.79859143183732e-05, + "loss": 4.3096, + "step": 2324 + }, + { + "epoch": 0.7435687301365208, + "grad_norm": 0.7972651124000549, + "learning_rate": 9.77562172827205e-05, + "loss": 4.2289, + "step": 2325 + }, + { + "epoch": 0.7438885446441064, + "grad_norm": 3.357048273086548, + "learning_rate": 9.752673737539779e-05, + "loss": 4.2417, + "step": 2326 + }, + { + "epoch": 0.744208359151692, + "grad_norm": 2.941267728805542, + "learning_rate": 9.729747484277402e-05, + "loss": 4.1785, + "step": 2327 + }, + { + "epoch": 0.7445281736592776, + "grad_norm": 1.182925820350647, + "learning_rate": 9.706842993098503e-05, + "loss": 4.2338, + "step": 2328 + }, + { + "epoch": 0.7448479881668633, + "grad_norm": 2.1440060138702393, + "learning_rate": 9.683960288593249e-05, + "loss": 4.2075, + "step": 2329 + }, + { + "epoch": 0.7451678026744488, + "grad_norm": 2.2097790241241455, + "learning_rate": 9.661099395328463e-05, + "loss": 4.182, + "step": 2330 + }, + { + "epoch": 0.7454876171820344, + "grad_norm": 4.899297714233398, + "learning_rate": 9.638260337847513e-05, + "loss": 4.1931, + "step": 2331 + }, + { + "epoch": 0.74580743168962, + "grad_norm": 3.723695755004883, + "learning_rate": 9.615443140670357e-05, + "loss": 4.21, + "step": 2332 + }, + { + "epoch": 0.7461272461972056, + "grad_norm": 1.1193941831588745, + "learning_rate": 9.592647828293468e-05, + "loss": 4.1894, + "step": 2333 + }, + { + "epoch": 0.7464470607047913, + "grad_norm": 1.2366178035736084, + "learning_rate": 9.569874425189827e-05, + "loss": 4.1465, + "step": 2334 + }, + { + "epoch": 0.7467668752123768, + "grad_norm": 1.9091682434082031, + "learning_rate": 9.547122955808902e-05, + "loss": 4.3169, + "step": 2335 + }, + { + "epoch": 0.7470866897199624, + "grad_norm": 1.6513155698776245, + "learning_rate": 9.524393444576585e-05, + "loss": 4.2761, + "step": 2336 + }, + { + "epoch": 0.7474065042275481, + "grad_norm": 13.898649215698242, + "learning_rate": 9.501685915895218e-05, + "loss": 4.2007, + "step": 2337 + }, + { + "epoch": 0.7477263187351336, + "grad_norm": 1.9269304275512695, + "learning_rate": 9.479000394143543e-05, + "loss": 4.2057, + "step": 2338 + }, + { + "epoch": 0.7480461332427192, + "grad_norm": 1.2362256050109863, + "learning_rate": 9.456336903676666e-05, + "loss": 4.2716, + "step": 2339 + }, + { + "epoch": 0.7483659477503048, + "grad_norm": 1.4338337182998657, + "learning_rate": 9.433695468826055e-05, + "loss": 4.2628, + "step": 2340 + }, + { + "epoch": 0.7486857622578904, + "grad_norm": 1.8466761112213135, + "learning_rate": 9.411076113899465e-05, + "loss": 4.2567, + "step": 2341 + }, + { + "epoch": 0.7490055767654761, + "grad_norm": 0.7579329609870911, + "learning_rate": 9.388478863180982e-05, + "loss": 4.2421, + "step": 2342 + }, + { + "epoch": 0.7493253912730616, + "grad_norm": 0.9924945831298828, + "learning_rate": 9.365903740930947e-05, + "loss": 4.2834, + "step": 2343 + }, + { + "epoch": 0.7496452057806472, + "grad_norm": 1.0770480632781982, + "learning_rate": 9.343350771385957e-05, + "loss": 4.2424, + "step": 2344 + }, + { + "epoch": 0.7499650202882329, + "grad_norm": 1.3987319469451904, + "learning_rate": 9.320819978758787e-05, + "loss": 4.2143, + "step": 2345 + }, + { + "epoch": 0.7502848347958184, + "grad_norm": 1.738250970840454, + "learning_rate": 9.298311387238449e-05, + "loss": 4.2901, + "step": 2346 + }, + { + "epoch": 0.750604649303404, + "grad_norm": 1.3063344955444336, + "learning_rate": 9.275825020990092e-05, + "loss": 4.247, + "step": 2347 + }, + { + "epoch": 0.7509244638109897, + "grad_norm": 1.4446301460266113, + "learning_rate": 9.25336090415502e-05, + "loss": 4.1459, + "step": 2348 + }, + { + "epoch": 0.7512442783185752, + "grad_norm": 1.3165273666381836, + "learning_rate": 9.230919060850645e-05, + "loss": 4.3115, + "step": 2349 + }, + { + "epoch": 0.7515640928261609, + "grad_norm": 0.8902044892311096, + "learning_rate": 9.208499515170451e-05, + "loss": 4.1914, + "step": 2350 + }, + { + "epoch": 0.7518839073337464, + "grad_norm": 6.7718024253845215, + "learning_rate": 9.186102291184003e-05, + "loss": 4.3121, + "step": 2351 + }, + { + "epoch": 0.752203721841332, + "grad_norm": 0.8847265839576721, + "learning_rate": 9.163727412936895e-05, + "loss": 4.248, + "step": 2352 + }, + { + "epoch": 0.7525235363489177, + "grad_norm": 1.0176126956939697, + "learning_rate": 9.141374904450733e-05, + "loss": 4.2897, + "step": 2353 + }, + { + "epoch": 0.7528433508565032, + "grad_norm": 0.8262820839881897, + "learning_rate": 9.119044789723108e-05, + "loss": 4.1388, + "step": 2354 + }, + { + "epoch": 0.7531631653640888, + "grad_norm": 1.2585965394973755, + "learning_rate": 9.09673709272755e-05, + "loss": 4.3209, + "step": 2355 + }, + { + "epoch": 0.7534829798716745, + "grad_norm": 1.2643582820892334, + "learning_rate": 9.07445183741355e-05, + "loss": 4.1393, + "step": 2356 + }, + { + "epoch": 0.75380279437926, + "grad_norm": 4.159008979797363, + "learning_rate": 9.052189047706484e-05, + "loss": 4.1605, + "step": 2357 + }, + { + "epoch": 0.7541226088868457, + "grad_norm": 3.648601531982422, + "learning_rate": 9.029948747507627e-05, + "loss": 4.1601, + "step": 2358 + }, + { + "epoch": 0.7544424233944312, + "grad_norm": 2.1243815422058105, + "learning_rate": 9.0077309606941e-05, + "loss": 4.2922, + "step": 2359 + }, + { + "epoch": 0.7547622379020168, + "grad_norm": 2.099933385848999, + "learning_rate": 8.985535711118844e-05, + "loss": 4.2453, + "step": 2360 + }, + { + "epoch": 0.7550820524096025, + "grad_norm": 0.9834458231925964, + "learning_rate": 8.963363022610623e-05, + "loss": 4.1078, + "step": 2361 + }, + { + "epoch": 0.755401866917188, + "grad_norm": 1.1493343114852905, + "learning_rate": 8.941212918973952e-05, + "loss": 4.2684, + "step": 2362 + }, + { + "epoch": 0.7557216814247736, + "grad_norm": 1.7550033330917358, + "learning_rate": 8.919085423989135e-05, + "loss": 4.1687, + "step": 2363 + }, + { + "epoch": 0.7560414959323593, + "grad_norm": 0.8992706537246704, + "learning_rate": 8.896980561412196e-05, + "loss": 4.2371, + "step": 2364 + }, + { + "epoch": 0.7563613104399448, + "grad_norm": 6.107719421386719, + "learning_rate": 8.874898354974821e-05, + "loss": 4.2124, + "step": 2365 + }, + { + "epoch": 0.7566811249475305, + "grad_norm": 0.8747105598449707, + "learning_rate": 8.85283882838443e-05, + "loss": 4.2176, + "step": 2366 + }, + { + "epoch": 0.757000939455116, + "grad_norm": 1.7237741947174072, + "learning_rate": 8.830802005324031e-05, + "loss": 4.167, + "step": 2367 + }, + { + "epoch": 0.7573207539627016, + "grad_norm": 3.720590114593506, + "learning_rate": 8.808787909452334e-05, + "loss": 4.1666, + "step": 2368 + }, + { + "epoch": 0.7576405684702873, + "grad_norm": 1.397730827331543, + "learning_rate": 8.786796564403575e-05, + "loss": 4.1681, + "step": 2369 + }, + { + "epoch": 0.7579603829778728, + "grad_norm": 1.5617105960845947, + "learning_rate": 8.764827993787613e-05, + "loss": 4.2455, + "step": 2370 + }, + { + "epoch": 0.7582801974854584, + "grad_norm": 1.774060606956482, + "learning_rate": 8.742882221189844e-05, + "loss": 4.2587, + "step": 2371 + }, + { + "epoch": 0.758600011993044, + "grad_norm": 1.2551524639129639, + "learning_rate": 8.720959270171162e-05, + "loss": 4.1946, + "step": 2372 + }, + { + "epoch": 0.7589198265006296, + "grad_norm": 3.613083839416504, + "learning_rate": 8.699059164268015e-05, + "loss": 4.2526, + "step": 2373 + }, + { + "epoch": 0.7592396410082153, + "grad_norm": 1.587499737739563, + "learning_rate": 8.677181926992271e-05, + "loss": 4.1788, + "step": 2374 + }, + { + "epoch": 0.7595594555158008, + "grad_norm": 1.089231014251709, + "learning_rate": 8.655327581831279e-05, + "loss": 4.2225, + "step": 2375 + }, + { + "epoch": 0.7598792700233864, + "grad_norm": 1.4902892112731934, + "learning_rate": 8.633496152247784e-05, + "loss": 4.2745, + "step": 2376 + }, + { + "epoch": 0.7601990845309721, + "grad_norm": 8.95482349395752, + "learning_rate": 8.611687661679945e-05, + "loss": 4.2438, + "step": 2377 + }, + { + "epoch": 0.7605188990385576, + "grad_norm": 1.874094009399414, + "learning_rate": 8.589902133541323e-05, + "loss": 4.2321, + "step": 2378 + }, + { + "epoch": 0.7608387135461432, + "grad_norm": 1.5763680934906006, + "learning_rate": 8.568139591220764e-05, + "loss": 4.1183, + "step": 2379 + }, + { + "epoch": 0.7611585280537289, + "grad_norm": 2.3884940147399902, + "learning_rate": 8.546400058082492e-05, + "loss": 4.1881, + "step": 2380 + }, + { + "epoch": 0.7614783425613144, + "grad_norm": 1.6015372276306152, + "learning_rate": 8.524683557465987e-05, + "loss": 4.2813, + "step": 2381 + }, + { + "epoch": 0.7617981570689001, + "grad_norm": 16.38766098022461, + "learning_rate": 8.502990112686028e-05, + "loss": 4.3229, + "step": 2382 + }, + { + "epoch": 0.7621179715764856, + "grad_norm": 1.267602562904358, + "learning_rate": 8.481319747032635e-05, + "loss": 4.3058, + "step": 2383 + }, + { + "epoch": 0.7624377860840712, + "grad_norm": 2.3478281497955322, + "learning_rate": 8.459672483771046e-05, + "loss": 4.2514, + "step": 2384 + }, + { + "epoch": 0.7627576005916569, + "grad_norm": 1.0165644884109497, + "learning_rate": 8.438048346141713e-05, + "loss": 4.1699, + "step": 2385 + }, + { + "epoch": 0.7630774150992424, + "grad_norm": 3.3435797691345215, + "learning_rate": 8.416447357360224e-05, + "loss": 4.2959, + "step": 2386 + }, + { + "epoch": 0.763397229606828, + "grad_norm": 1.897871971130371, + "learning_rate": 8.394869540617347e-05, + "loss": 4.2229, + "step": 2387 + }, + { + "epoch": 0.7637170441144137, + "grad_norm": 7.05217170715332, + "learning_rate": 8.373314919078964e-05, + "loss": 4.1647, + "step": 2388 + }, + { + "epoch": 0.7640368586219992, + "grad_norm": 0.8193559050559998, + "learning_rate": 8.35178351588605e-05, + "loss": 4.1605, + "step": 2389 + }, + { + "epoch": 0.7643566731295849, + "grad_norm": 2.585585594177246, + "learning_rate": 8.330275354154672e-05, + "loss": 4.2021, + "step": 2390 + }, + { + "epoch": 0.7646764876371704, + "grad_norm": 0.9105507135391235, + "learning_rate": 8.308790456975905e-05, + "loss": 4.2208, + "step": 2391 + }, + { + "epoch": 0.764996302144756, + "grad_norm": 0.9755898714065552, + "learning_rate": 8.28732884741588e-05, + "loss": 4.2616, + "step": 2392 + }, + { + "epoch": 0.7653161166523417, + "grad_norm": 10.436922073364258, + "learning_rate": 8.265890548515723e-05, + "loss": 4.1767, + "step": 2393 + }, + { + "epoch": 0.7656359311599272, + "grad_norm": 1.1803678274154663, + "learning_rate": 8.244475583291522e-05, + "loss": 4.1535, + "step": 2394 + }, + { + "epoch": 0.7659557456675128, + "grad_norm": 0.8434635996818542, + "learning_rate": 8.223083974734336e-05, + "loss": 4.2617, + "step": 2395 + }, + { + "epoch": 0.7662755601750985, + "grad_norm": 1.2088812589645386, + "learning_rate": 8.201715745810112e-05, + "loss": 4.2795, + "step": 2396 + }, + { + "epoch": 0.766595374682684, + "grad_norm": 0.9391944408416748, + "learning_rate": 8.180370919459728e-05, + "loss": 4.1859, + "step": 2397 + }, + { + "epoch": 0.7669151891902697, + "grad_norm": 1.5034995079040527, + "learning_rate": 8.159049518598924e-05, + "loss": 4.178, + "step": 2398 + }, + { + "epoch": 0.7672350036978552, + "grad_norm": 1.5647478103637695, + "learning_rate": 8.137751566118306e-05, + "loss": 4.1856, + "step": 2399 + }, + { + "epoch": 0.7675548182054408, + "grad_norm": 1.0978316068649292, + "learning_rate": 8.11647708488327e-05, + "loss": 4.31, + "step": 2400 + }, + { + "epoch": 0.7675548182054408, + "eval_runtime": 21.168, + "eval_samples_per_second": 89.617, + "eval_steps_per_second": 22.44, + "step": 2400 + }, + { + "epoch": 0.7678746327130265, + "grad_norm": 1.0744236707687378, + "learning_rate": 8.09522609773405e-05, + "loss": 4.1867, + "step": 2401 + }, + { + "epoch": 0.768194447220612, + "grad_norm": 1.068185806274414, + "learning_rate": 8.073998627485641e-05, + "loss": 4.2493, + "step": 2402 + }, + { + "epoch": 0.7685142617281977, + "grad_norm": 1.1011041402816772, + "learning_rate": 8.052794696927796e-05, + "loss": 4.1355, + "step": 2403 + }, + { + "epoch": 0.7688340762357833, + "grad_norm": 6.319514751434326, + "learning_rate": 8.031614328824998e-05, + "loss": 4.192, + "step": 2404 + }, + { + "epoch": 0.7691538907433688, + "grad_norm": 0.910234272480011, + "learning_rate": 8.010457545916408e-05, + "loss": 4.2297, + "step": 2405 + }, + { + "epoch": 0.7694737052509545, + "grad_norm": 1.1834043264389038, + "learning_rate": 7.989324370915899e-05, + "loss": 4.252, + "step": 2406 + }, + { + "epoch": 0.76979351975854, + "grad_norm": 1.086834192276001, + "learning_rate": 7.968214826511987e-05, + "loss": 4.2314, + "step": 2407 + }, + { + "epoch": 0.7701133342661256, + "grad_norm": 2.584946393966675, + "learning_rate": 7.947128935367813e-05, + "loss": 4.2715, + "step": 2408 + }, + { + "epoch": 0.7704331487737113, + "grad_norm": 3.4537439346313477, + "learning_rate": 7.926066720121134e-05, + "loss": 4.2017, + "step": 2409 + }, + { + "epoch": 0.7707529632812968, + "grad_norm": 2.544417142868042, + "learning_rate": 7.905028203384269e-05, + "loss": 4.2049, + "step": 2410 + }, + { + "epoch": 0.7710727777888825, + "grad_norm": 4.528934955596924, + "learning_rate": 7.884013407744129e-05, + "loss": 4.2494, + "step": 2411 + }, + { + "epoch": 0.7713925922964681, + "grad_norm": 0.8724344372749329, + "learning_rate": 7.863022355762101e-05, + "loss": 4.2863, + "step": 2412 + }, + { + "epoch": 0.7717124068040536, + "grad_norm": 1.3964775800704956, + "learning_rate": 7.842055069974149e-05, + "loss": 4.2273, + "step": 2413 + }, + { + "epoch": 0.7720322213116393, + "grad_norm": 0.8182308673858643, + "learning_rate": 7.82111157289069e-05, + "loss": 4.2452, + "step": 2414 + }, + { + "epoch": 0.7723520358192248, + "grad_norm": 5.355796813964844, + "learning_rate": 7.800191886996578e-05, + "loss": 4.0883, + "step": 2415 + }, + { + "epoch": 0.7726718503268104, + "grad_norm": 1.290724754333496, + "learning_rate": 7.779296034751152e-05, + "loss": 4.1468, + "step": 2416 + }, + { + "epoch": 0.7729916648343961, + "grad_norm": 1.5933973789215088, + "learning_rate": 7.75842403858811e-05, + "loss": 4.271, + "step": 2417 + }, + { + "epoch": 0.7733114793419816, + "grad_norm": 7.741376876831055, + "learning_rate": 7.737575920915574e-05, + "loss": 4.2076, + "step": 2418 + }, + { + "epoch": 0.7736312938495673, + "grad_norm": 11.15859317779541, + "learning_rate": 7.716751704116042e-05, + "loss": 4.1902, + "step": 2419 + }, + { + "epoch": 0.7739511083571529, + "grad_norm": 2.4089276790618896, + "learning_rate": 7.695951410546311e-05, + "loss": 4.33, + "step": 2420 + }, + { + "epoch": 0.7742709228647384, + "grad_norm": 0.7691051363945007, + "learning_rate": 7.67517506253753e-05, + "loss": 4.2373, + "step": 2421 + }, + { + "epoch": 0.7745907373723241, + "grad_norm": 4.013336181640625, + "learning_rate": 7.654422682395106e-05, + "loss": 4.2096, + "step": 2422 + }, + { + "epoch": 0.7749105518799096, + "grad_norm": 1.6960875988006592, + "learning_rate": 7.633694292398745e-05, + "loss": 4.1526, + "step": 2423 + }, + { + "epoch": 0.7752303663874952, + "grad_norm": 12.997591018676758, + "learning_rate": 7.612989914802383e-05, + "loss": 4.2133, + "step": 2424 + }, + { + "epoch": 0.7755501808950809, + "grad_norm": 1.0910037755966187, + "learning_rate": 7.592309571834179e-05, + "loss": 4.1856, + "step": 2425 + }, + { + "epoch": 0.7758699954026664, + "grad_norm": 4.121542930603027, + "learning_rate": 7.5716532856965e-05, + "loss": 4.2158, + "step": 2426 + }, + { + "epoch": 0.7761898099102521, + "grad_norm": 0.9528754353523254, + "learning_rate": 7.551021078565857e-05, + "loss": 4.2032, + "step": 2427 + }, + { + "epoch": 0.7765096244178377, + "grad_norm": 3.85128116607666, + "learning_rate": 7.530412972592928e-05, + "loss": 4.2681, + "step": 2428 + }, + { + "epoch": 0.7768294389254232, + "grad_norm": 0.9954482316970825, + "learning_rate": 7.509828989902525e-05, + "loss": 4.2309, + "step": 2429 + }, + { + "epoch": 0.7771492534330089, + "grad_norm": 1.8523609638214111, + "learning_rate": 7.489269152593543e-05, + "loss": 4.1958, + "step": 2430 + }, + { + "epoch": 0.7774690679405944, + "grad_norm": 36.62735366821289, + "learning_rate": 7.468733482738976e-05, + "loss": 4.3061, + "step": 2431 + }, + { + "epoch": 0.77778888244818, + "grad_norm": 102.16072082519531, + "learning_rate": 7.44822200238584e-05, + "loss": 4.187, + "step": 2432 + }, + { + "epoch": 0.7781086969557657, + "grad_norm": 0.9704092144966125, + "learning_rate": 7.42773473355521e-05, + "loss": 4.2563, + "step": 2433 + }, + { + "epoch": 0.7784285114633512, + "grad_norm": 1.225210189819336, + "learning_rate": 7.407271698242155e-05, + "loss": 4.2152, + "step": 2434 + }, + { + "epoch": 0.7787483259709369, + "grad_norm": 2.3196194171905518, + "learning_rate": 7.386832918415741e-05, + "loss": 4.1914, + "step": 2435 + }, + { + "epoch": 0.7790681404785225, + "grad_norm": 1.0257723331451416, + "learning_rate": 7.366418416018963e-05, + "loss": 4.1834, + "step": 2436 + }, + { + "epoch": 0.779387954986108, + "grad_norm": 0.9030638337135315, + "learning_rate": 7.346028212968778e-05, + "loss": 4.1901, + "step": 2437 + }, + { + "epoch": 0.7797077694936937, + "grad_norm": 4.089504241943359, + "learning_rate": 7.325662331156049e-05, + "loss": 4.1743, + "step": 2438 + }, + { + "epoch": 0.7800275840012793, + "grad_norm": 5.624063968658447, + "learning_rate": 7.305320792445532e-05, + "loss": 4.1331, + "step": 2439 + }, + { + "epoch": 0.7803473985088648, + "grad_norm": 0.7666806578636169, + "learning_rate": 7.285003618675842e-05, + "loss": 4.1826, + "step": 2440 + }, + { + "epoch": 0.7806672130164505, + "grad_norm": 1.7262241840362549, + "learning_rate": 7.264710831659426e-05, + "loss": 4.1571, + "step": 2441 + }, + { + "epoch": 0.780987027524036, + "grad_norm": 2.565640687942505, + "learning_rate": 7.24444245318257e-05, + "loss": 4.1521, + "step": 2442 + }, + { + "epoch": 0.7813068420316217, + "grad_norm": 1.7281955480575562, + "learning_rate": 7.224198505005344e-05, + "loss": 4.2339, + "step": 2443 + }, + { + "epoch": 0.7816266565392073, + "grad_norm": 0.779292106628418, + "learning_rate": 7.203979008861588e-05, + "loss": 4.235, + "step": 2444 + }, + { + "epoch": 0.7819464710467928, + "grad_norm": 0.7449750304222107, + "learning_rate": 7.183783986458906e-05, + "loss": 4.2191, + "step": 2445 + }, + { + "epoch": 0.7822662855543785, + "grad_norm": 2.9117207527160645, + "learning_rate": 7.163613459478595e-05, + "loss": 4.2776, + "step": 2446 + }, + { + "epoch": 0.782586100061964, + "grad_norm": 1.6501144170761108, + "learning_rate": 7.143467449575682e-05, + "loss": 4.161, + "step": 2447 + }, + { + "epoch": 0.7829059145695496, + "grad_norm": 2.053727865219116, + "learning_rate": 7.12334597837887e-05, + "loss": 4.2526, + "step": 2448 + }, + { + "epoch": 0.7832257290771353, + "grad_norm": 1.4463319778442383, + "learning_rate": 7.103249067490502e-05, + "loss": 4.2302, + "step": 2449 + }, + { + "epoch": 0.7835455435847208, + "grad_norm": 2.6692161560058594, + "learning_rate": 7.083176738486578e-05, + "loss": 4.1257, + "step": 2450 + }, + { + "epoch": 0.7838653580923065, + "grad_norm": 0.7135071754455566, + "learning_rate": 7.063129012916671e-05, + "loss": 4.1731, + "step": 2451 + }, + { + "epoch": 0.7841851725998921, + "grad_norm": 3.0395052433013916, + "learning_rate": 7.04310591230397e-05, + "loss": 4.1732, + "step": 2452 + }, + { + "epoch": 0.7845049871074776, + "grad_norm": 17.021873474121094, + "learning_rate": 7.023107458145214e-05, + "loss": 4.2885, + "step": 2453 + }, + { + "epoch": 0.7848248016150633, + "grad_norm": 8.884562492370605, + "learning_rate": 7.003133671910688e-05, + "loss": 4.2402, + "step": 2454 + }, + { + "epoch": 0.7851446161226489, + "grad_norm": 2.4456584453582764, + "learning_rate": 6.983184575044199e-05, + "loss": 4.1983, + "step": 2455 + }, + { + "epoch": 0.7854644306302344, + "grad_norm": 0.9869387745857239, + "learning_rate": 6.963260188963016e-05, + "loss": 4.1878, + "step": 2456 + }, + { + "epoch": 0.7857842451378201, + "grad_norm": 2.917192220687866, + "learning_rate": 6.943360535057926e-05, + "loss": 4.2325, + "step": 2457 + }, + { + "epoch": 0.7861040596454056, + "grad_norm": 2.5623371601104736, + "learning_rate": 6.923485634693109e-05, + "loss": 4.2076, + "step": 2458 + }, + { + "epoch": 0.7864238741529913, + "grad_norm": 1.378758192062378, + "learning_rate": 6.903635509206234e-05, + "loss": 4.2408, + "step": 2459 + }, + { + "epoch": 0.7867436886605769, + "grad_norm": 1.4268980026245117, + "learning_rate": 6.883810179908315e-05, + "loss": 4.1756, + "step": 2460 + }, + { + "epoch": 0.7870635031681624, + "grad_norm": 0.785111129283905, + "learning_rate": 6.86400966808377e-05, + "loss": 4.2163, + "step": 2461 + }, + { + "epoch": 0.7873833176757481, + "grad_norm": 0.8125237226486206, + "learning_rate": 6.844233994990382e-05, + "loss": 4.2076, + "step": 2462 + }, + { + "epoch": 0.7877031321833337, + "grad_norm": 1.186068058013916, + "learning_rate": 6.824483181859231e-05, + "loss": 4.2046, + "step": 2463 + }, + { + "epoch": 0.7880229466909192, + "grad_norm": 1.046829342842102, + "learning_rate": 6.804757249894762e-05, + "loss": 4.1839, + "step": 2464 + }, + { + "epoch": 0.7883427611985049, + "grad_norm": 1.4435135126113892, + "learning_rate": 6.785056220274658e-05, + "loss": 4.1626, + "step": 2465 + }, + { + "epoch": 0.7886625757060904, + "grad_norm": 1.12195885181427, + "learning_rate": 6.765380114149887e-05, + "loss": 4.2092, + "step": 2466 + }, + { + "epoch": 0.7889823902136761, + "grad_norm": 2.4519267082214355, + "learning_rate": 6.745728952644675e-05, + "loss": 4.214, + "step": 2467 + }, + { + "epoch": 0.7893022047212617, + "grad_norm": 1.1661007404327393, + "learning_rate": 6.726102756856422e-05, + "loss": 4.2317, + "step": 2468 + }, + { + "epoch": 0.7896220192288472, + "grad_norm": 2.0803890228271484, + "learning_rate": 6.706501547855787e-05, + "loss": 4.2332, + "step": 2469 + }, + { + "epoch": 0.7899418337364329, + "grad_norm": 1.5335798263549805, + "learning_rate": 6.686925346686544e-05, + "loss": 4.1744, + "step": 2470 + }, + { + "epoch": 0.7902616482440185, + "grad_norm": 4.207418441772461, + "learning_rate": 6.667374174365667e-05, + "loss": 4.2298, + "step": 2471 + }, + { + "epoch": 0.7905814627516041, + "grad_norm": 1.0971169471740723, + "learning_rate": 6.647848051883217e-05, + "loss": 4.1979, + "step": 2472 + }, + { + "epoch": 0.7909012772591897, + "grad_norm": 0.7819692492485046, + "learning_rate": 6.628347000202381e-05, + "loss": 4.1749, + "step": 2473 + }, + { + "epoch": 0.7912210917667752, + "grad_norm": 1.873777985572815, + "learning_rate": 6.608871040259457e-05, + "loss": 4.1794, + "step": 2474 + }, + { + "epoch": 0.7915409062743609, + "grad_norm": 1.569891095161438, + "learning_rate": 6.589420192963754e-05, + "loss": 4.1707, + "step": 2475 + }, + { + "epoch": 0.7918607207819465, + "grad_norm": 3.980315685272217, + "learning_rate": 6.56999447919766e-05, + "loss": 4.2557, + "step": 2476 + }, + { + "epoch": 0.792180535289532, + "grad_norm": 1.2121565341949463, + "learning_rate": 6.550593919816545e-05, + "loss": 4.1849, + "step": 2477 + }, + { + "epoch": 0.7925003497971177, + "grad_norm": 0.7986897826194763, + "learning_rate": 6.531218535648807e-05, + "loss": 4.166, + "step": 2478 + }, + { + "epoch": 0.7928201643047033, + "grad_norm": 1.5680044889450073, + "learning_rate": 6.511868347495793e-05, + "loss": 4.1842, + "step": 2479 + }, + { + "epoch": 0.7931399788122889, + "grad_norm": 0.8249400854110718, + "learning_rate": 6.492543376131817e-05, + "loss": 4.2549, + "step": 2480 + }, + { + "epoch": 0.7934597933198745, + "grad_norm": 11.279062271118164, + "learning_rate": 6.473243642304114e-05, + "loss": 4.2332, + "step": 2481 + }, + { + "epoch": 0.79377960782746, + "grad_norm": 0.731088399887085, + "learning_rate": 6.453969166732808e-05, + "loss": 4.2316, + "step": 2482 + }, + { + "epoch": 0.7940994223350457, + "grad_norm": 1.3943499326705933, + "learning_rate": 6.434719970110923e-05, + "loss": 4.1057, + "step": 2483 + }, + { + "epoch": 0.7944192368426313, + "grad_norm": 0.9094737768173218, + "learning_rate": 6.415496073104344e-05, + "loss": 4.2852, + "step": 2484 + }, + { + "epoch": 0.7947390513502168, + "grad_norm": 1.1480377912521362, + "learning_rate": 6.396297496351791e-05, + "loss": 4.2531, + "step": 2485 + }, + { + "epoch": 0.7950588658578025, + "grad_norm": 0.9361832737922668, + "learning_rate": 6.377124260464804e-05, + "loss": 4.2049, + "step": 2486 + }, + { + "epoch": 0.7953786803653881, + "grad_norm": 0.8900967240333557, + "learning_rate": 6.357976386027697e-05, + "loss": 4.2528, + "step": 2487 + }, + { + "epoch": 0.7956984948729737, + "grad_norm": 2.6632485389709473, + "learning_rate": 6.338853893597584e-05, + "loss": 4.2191, + "step": 2488 + }, + { + "epoch": 0.7960183093805593, + "grad_norm": 14.60572338104248, + "learning_rate": 6.319756803704311e-05, + "loss": 4.2224, + "step": 2489 + }, + { + "epoch": 0.7963381238881448, + "grad_norm": 2.023411989212036, + "learning_rate": 6.300685136850458e-05, + "loss": 4.2276, + "step": 2490 + }, + { + "epoch": 0.7966579383957305, + "grad_norm": 2.3787682056427, + "learning_rate": 6.281638913511324e-05, + "loss": 4.2016, + "step": 2491 + }, + { + "epoch": 0.7969777529033161, + "grad_norm": 2.218717098236084, + "learning_rate": 6.262618154134858e-05, + "loss": 4.1279, + "step": 2492 + }, + { + "epoch": 0.7972975674109016, + "grad_norm": 1.5670682191848755, + "learning_rate": 6.2436228791417e-05, + "loss": 4.2189, + "step": 2493 + }, + { + "epoch": 0.7976173819184873, + "grad_norm": 2.456651210784912, + "learning_rate": 6.224653108925122e-05, + "loss": 4.1966, + "step": 2494 + }, + { + "epoch": 0.7979371964260729, + "grad_norm": 0.934609055519104, + "learning_rate": 6.205708863851019e-05, + "loss": 4.1306, + "step": 2495 + }, + { + "epoch": 0.7982570109336585, + "grad_norm": 0.9249786734580994, + "learning_rate": 6.186790164257866e-05, + "loss": 4.2236, + "step": 2496 + }, + { + "epoch": 0.7985768254412441, + "grad_norm": 1.96321439743042, + "learning_rate": 6.167897030456725e-05, + "loss": 4.2353, + "step": 2497 + }, + { + "epoch": 0.7988966399488296, + "grad_norm": 7.929496765136719, + "learning_rate": 6.149029482731211e-05, + "loss": 4.1733, + "step": 2498 + }, + { + "epoch": 0.7992164544564153, + "grad_norm": 0.7240726351737976, + "learning_rate": 6.13018754133747e-05, + "loss": 4.2346, + "step": 2499 + }, + { + "epoch": 0.7995362689640009, + "grad_norm": 1.0983538627624512, + "learning_rate": 6.111371226504162e-05, + "loss": 4.1988, + "step": 2500 + }, + { + "epoch": 0.7995362689640009, + "eval_runtime": 22.123, + "eval_samples_per_second": 85.748, + "eval_steps_per_second": 21.471, + "step": 2500 + }, + { + "epoch": 0.7998560834715864, + "grad_norm": 0.8758413791656494, + "learning_rate": 6.092580558432409e-05, + "loss": 4.2335, + "step": 2501 + }, + { + "epoch": 0.8001758979791721, + "grad_norm": 0.7572125196456909, + "learning_rate": 6.073815557295827e-05, + "loss": 4.1885, + "step": 2502 + }, + { + "epoch": 0.8004957124867577, + "grad_norm": 1.2751429080963135, + "learning_rate": 6.055076243240467e-05, + "loss": 4.1616, + "step": 2503 + }, + { + "epoch": 0.8008155269943433, + "grad_norm": 2.166659355163574, + "learning_rate": 6.036362636384798e-05, + "loss": 4.1835, + "step": 2504 + }, + { + "epoch": 0.8011353415019289, + "grad_norm": 1.203718900680542, + "learning_rate": 6.017674756819705e-05, + "loss": 4.1501, + "step": 2505 + }, + { + "epoch": 0.8014551560095144, + "grad_norm": 3.1784892082214355, + "learning_rate": 5.9990126246084204e-05, + "loss": 4.1564, + "step": 2506 + }, + { + "epoch": 0.8017749705171001, + "grad_norm": 0.9053913950920105, + "learning_rate": 5.9803762597865745e-05, + "loss": 4.1219, + "step": 2507 + }, + { + "epoch": 0.8020947850246857, + "grad_norm": 1.0492571592330933, + "learning_rate": 5.96176568236209e-05, + "loss": 4.1919, + "step": 2508 + }, + { + "epoch": 0.8024145995322712, + "grad_norm": 1.8547947406768799, + "learning_rate": 5.9431809123152465e-05, + "loss": 4.1982, + "step": 2509 + }, + { + "epoch": 0.8027344140398569, + "grad_norm": 1.3849730491638184, + "learning_rate": 5.924621969598604e-05, + "loss": 4.2983, + "step": 2510 + }, + { + "epoch": 0.8030542285474425, + "grad_norm": 0.7662582993507385, + "learning_rate": 5.906088874136968e-05, + "loss": 4.2386, + "step": 2511 + }, + { + "epoch": 0.8033740430550281, + "grad_norm": 1.7136598825454712, + "learning_rate": 5.887581645827436e-05, + "loss": 4.2079, + "step": 2512 + }, + { + "epoch": 0.8036938575626137, + "grad_norm": 1.2888476848602295, + "learning_rate": 5.869100304539297e-05, + "loss": 4.1803, + "step": 2513 + }, + { + "epoch": 0.8040136720701992, + "grad_norm": 1.3532698154449463, + "learning_rate": 5.850644870114063e-05, + "loss": 4.1267, + "step": 2514 + }, + { + "epoch": 0.8043334865777849, + "grad_norm": 0.6778098940849304, + "learning_rate": 5.832215362365458e-05, + "loss": 4.2169, + "step": 2515 + }, + { + "epoch": 0.8046533010853705, + "grad_norm": 2.293372392654419, + "learning_rate": 5.813811801079325e-05, + "loss": 4.2566, + "step": 2516 + }, + { + "epoch": 0.804973115592956, + "grad_norm": 1.2440351247787476, + "learning_rate": 5.795434206013685e-05, + "loss": 4.1927, + "step": 2517 + }, + { + "epoch": 0.8052929301005417, + "grad_norm": 1.4825369119644165, + "learning_rate": 5.77708259689866e-05, + "loss": 4.2354, + "step": 2518 + }, + { + "epoch": 0.8056127446081273, + "grad_norm": 1.1397002935409546, + "learning_rate": 5.7587569934364896e-05, + "loss": 4.2714, + "step": 2519 + }, + { + "epoch": 0.8059325591157129, + "grad_norm": 1.0114519596099854, + "learning_rate": 5.740457415301486e-05, + "loss": 4.24, + "step": 2520 + }, + { + "epoch": 0.8062523736232985, + "grad_norm": 0.8091397285461426, + "learning_rate": 5.72218388214002e-05, + "loss": 4.1981, + "step": 2521 + }, + { + "epoch": 0.806572188130884, + "grad_norm": 2.4450981616973877, + "learning_rate": 5.703936413570519e-05, + "loss": 4.1968, + "step": 2522 + }, + { + "epoch": 0.8068920026384697, + "grad_norm": 1.204534649848938, + "learning_rate": 5.6857150291833884e-05, + "loss": 4.253, + "step": 2523 + }, + { + "epoch": 0.8072118171460553, + "grad_norm": 0.8494846820831299, + "learning_rate": 5.667519748541064e-05, + "loss": 4.1705, + "step": 2524 + }, + { + "epoch": 0.8075316316536408, + "grad_norm": 0.9812785387039185, + "learning_rate": 5.649350591177946e-05, + "loss": 4.1213, + "step": 2525 + }, + { + "epoch": 0.8078514461612265, + "grad_norm": 0.9506296515464783, + "learning_rate": 5.6312075766003876e-05, + "loss": 4.2578, + "step": 2526 + }, + { + "epoch": 0.8081712606688121, + "grad_norm": 1.0995137691497803, + "learning_rate": 5.613090724286681e-05, + "loss": 4.1417, + "step": 2527 + }, + { + "epoch": 0.8084910751763977, + "grad_norm": 1.0665677785873413, + "learning_rate": 5.595000053687014e-05, + "loss": 4.2409, + "step": 2528 + }, + { + "epoch": 0.8088108896839833, + "grad_norm": 1.1351603269577026, + "learning_rate": 5.576935584223482e-05, + "loss": 4.1775, + "step": 2529 + }, + { + "epoch": 0.8091307041915689, + "grad_norm": 0.9329645037651062, + "learning_rate": 5.55889733529005e-05, + "loss": 4.2066, + "step": 2530 + }, + { + "epoch": 0.8094505186991545, + "grad_norm": 13.807785034179688, + "learning_rate": 5.540885326252531e-05, + "loss": 4.3348, + "step": 2531 + }, + { + "epoch": 0.8097703332067401, + "grad_norm": 0.7993907928466797, + "learning_rate": 5.5228995764485564e-05, + "loss": 4.1509, + "step": 2532 + }, + { + "epoch": 0.8100901477143256, + "grad_norm": 1.684177279472351, + "learning_rate": 5.5049401051875765e-05, + "loss": 4.2405, + "step": 2533 + }, + { + "epoch": 0.8104099622219113, + "grad_norm": 1.5524111986160278, + "learning_rate": 5.487006931750828e-05, + "loss": 4.1425, + "step": 2534 + }, + { + "epoch": 0.8107297767294969, + "grad_norm": 0.9459340572357178, + "learning_rate": 5.469100075391314e-05, + "loss": 4.2372, + "step": 2535 + }, + { + "epoch": 0.8110495912370825, + "grad_norm": 0.6475396156311035, + "learning_rate": 5.451219555333792e-05, + "loss": 4.1664, + "step": 2536 + }, + { + "epoch": 0.8113694057446681, + "grad_norm": 1.0679532289505005, + "learning_rate": 5.4333653907747174e-05, + "loss": 4.0972, + "step": 2537 + }, + { + "epoch": 0.8116892202522537, + "grad_norm": 4.013439655303955, + "learning_rate": 5.4155376008822805e-05, + "loss": 4.1786, + "step": 2538 + }, + { + "epoch": 0.8120090347598393, + "grad_norm": 0.7991141676902771, + "learning_rate": 5.397736204796337e-05, + "loss": 4.2178, + "step": 2539 + }, + { + "epoch": 0.8123288492674249, + "grad_norm": 1.9116607904434204, + "learning_rate": 5.37996122162842e-05, + "loss": 4.1968, + "step": 2540 + }, + { + "epoch": 0.8126486637750105, + "grad_norm": 0.7526158690452576, + "learning_rate": 5.362212670461706e-05, + "loss": 4.2137, + "step": 2541 + }, + { + "epoch": 0.8129684782825961, + "grad_norm": 2.0017004013061523, + "learning_rate": 5.3444905703509687e-05, + "loss": 4.147, + "step": 2542 + }, + { + "epoch": 0.8132882927901817, + "grad_norm": 2.603358030319214, + "learning_rate": 5.3267949403226104e-05, + "loss": 4.2014, + "step": 2543 + }, + { + "epoch": 0.8136081072977673, + "grad_norm": 4.472846508026123, + "learning_rate": 5.3091257993746115e-05, + "loss": 4.1869, + "step": 2544 + }, + { + "epoch": 0.8139279218053529, + "grad_norm": 0.9687552452087402, + "learning_rate": 5.2914831664765045e-05, + "loss": 4.157, + "step": 2545 + }, + { + "epoch": 0.8142477363129385, + "grad_norm": 4.424249649047852, + "learning_rate": 5.2738670605693814e-05, + "loss": 4.235, + "step": 2546 + }, + { + "epoch": 0.8145675508205241, + "grad_norm": 6.971482276916504, + "learning_rate": 5.256277500565823e-05, + "loss": 4.188, + "step": 2547 + }, + { + "epoch": 0.8148873653281097, + "grad_norm": 2.7917723655700684, + "learning_rate": 5.238714505349938e-05, + "loss": 4.1956, + "step": 2548 + }, + { + "epoch": 0.8152071798356954, + "grad_norm": 29.214433670043945, + "learning_rate": 5.221178093777303e-05, + "loss": 4.234, + "step": 2549 + }, + { + "epoch": 0.8155269943432809, + "grad_norm": 1.089066743850708, + "learning_rate": 5.2036682846749645e-05, + "loss": 4.1925, + "step": 2550 + }, + { + "epoch": 0.8158468088508665, + "grad_norm": 1.9935816526412964, + "learning_rate": 5.186185096841402e-05, + "loss": 4.2312, + "step": 2551 + }, + { + "epoch": 0.8161666233584521, + "grad_norm": 1.6591036319732666, + "learning_rate": 5.168728549046508e-05, + "loss": 4.1928, + "step": 2552 + }, + { + "epoch": 0.8164864378660377, + "grad_norm": 0.9029149413108826, + "learning_rate": 5.151298660031587e-05, + "loss": 4.203, + "step": 2553 + }, + { + "epoch": 0.8168062523736233, + "grad_norm": 1.1216133832931519, + "learning_rate": 5.133895448509299e-05, + "loss": 4.1617, + "step": 2554 + }, + { + "epoch": 0.8171260668812089, + "grad_norm": 0.7909253239631653, + "learning_rate": 5.116518933163709e-05, + "loss": 4.1498, + "step": 2555 + }, + { + "epoch": 0.8174458813887945, + "grad_norm": 0.8655667901039124, + "learning_rate": 5.099169132650173e-05, + "loss": 4.1908, + "step": 2556 + }, + { + "epoch": 0.8177656958963802, + "grad_norm": 1.5724070072174072, + "learning_rate": 5.0818460655953894e-05, + "loss": 4.1265, + "step": 2557 + }, + { + "epoch": 0.8180855104039657, + "grad_norm": 1.1621187925338745, + "learning_rate": 5.0645497505973633e-05, + "loss": 4.1459, + "step": 2558 + }, + { + "epoch": 0.8184053249115513, + "grad_norm": 2.53725528717041, + "learning_rate": 5.0472802062253426e-05, + "loss": 4.1624, + "step": 2559 + }, + { + "epoch": 0.8187251394191369, + "grad_norm": 1.8308603763580322, + "learning_rate": 5.0300374510198904e-05, + "loss": 4.1598, + "step": 2560 + }, + { + "epoch": 0.8190449539267225, + "grad_norm": 1.0580358505249023, + "learning_rate": 5.012821503492755e-05, + "loss": 4.1784, + "step": 2561 + }, + { + "epoch": 0.819364768434308, + "grad_norm": 1.0718411207199097, + "learning_rate": 4.9956323821269326e-05, + "loss": 4.1937, + "step": 2562 + }, + { + "epoch": 0.8196845829418937, + "grad_norm": 1.1936836242675781, + "learning_rate": 4.978470105376627e-05, + "loss": 4.2338, + "step": 2563 + }, + { + "epoch": 0.8200043974494793, + "grad_norm": 1.6368927955627441, + "learning_rate": 4.961334691667177e-05, + "loss": 4.1657, + "step": 2564 + }, + { + "epoch": 0.820324211957065, + "grad_norm": 1.0185908079147339, + "learning_rate": 4.9442261593951496e-05, + "loss": 4.2015, + "step": 2565 + }, + { + "epoch": 0.8206440264646505, + "grad_norm": 1.239174723625183, + "learning_rate": 4.9271445269281884e-05, + "loss": 4.2568, + "step": 2566 + }, + { + "epoch": 0.8209638409722361, + "grad_norm": 0.7394360303878784, + "learning_rate": 4.910089812605098e-05, + "loss": 4.1643, + "step": 2567 + }, + { + "epoch": 0.8212836554798217, + "grad_norm": 0.7906307578086853, + "learning_rate": 4.893062034735758e-05, + "loss": 4.1334, + "step": 2568 + }, + { + "epoch": 0.8216034699874073, + "grad_norm": 0.8245829343795776, + "learning_rate": 4.8760612116011464e-05, + "loss": 4.1746, + "step": 2569 + }, + { + "epoch": 0.8219232844949929, + "grad_norm": 0.7398518919944763, + "learning_rate": 4.8590873614532956e-05, + "loss": 4.1711, + "step": 2570 + }, + { + "epoch": 0.8222430990025785, + "grad_norm": 1.1121344566345215, + "learning_rate": 4.842140502515282e-05, + "loss": 4.1624, + "step": 2571 + }, + { + "epoch": 0.8225629135101641, + "grad_norm": 8.878984451293945, + "learning_rate": 4.825220652981211e-05, + "loss": 4.1934, + "step": 2572 + }, + { + "epoch": 0.8228827280177498, + "grad_norm": 0.9753667116165161, + "learning_rate": 4.80832783101617e-05, + "loss": 4.1764, + "step": 2573 + }, + { + "epoch": 0.8232025425253353, + "grad_norm": 0.5970510840415955, + "learning_rate": 4.7914620547562475e-05, + "loss": 4.1509, + "step": 2574 + }, + { + "epoch": 0.8235223570329209, + "grad_norm": 1.0197758674621582, + "learning_rate": 4.7746233423084965e-05, + "loss": 4.1104, + "step": 2575 + }, + { + "epoch": 0.8238421715405065, + "grad_norm": 1.371875524520874, + "learning_rate": 4.757811711750903e-05, + "loss": 4.2038, + "step": 2576 + }, + { + "epoch": 0.8241619860480921, + "grad_norm": 1.1378782987594604, + "learning_rate": 4.741027181132392e-05, + "loss": 4.2413, + "step": 2577 + }, + { + "epoch": 0.8244818005556777, + "grad_norm": 0.8936293125152588, + "learning_rate": 4.724269768472776e-05, + "loss": 4.1801, + "step": 2578 + }, + { + "epoch": 0.8248016150632633, + "grad_norm": 1.3190841674804688, + "learning_rate": 4.707539491762767e-05, + "loss": 4.2803, + "step": 2579 + }, + { + "epoch": 0.8251214295708489, + "grad_norm": 0.9134860634803772, + "learning_rate": 4.690836368963945e-05, + "loss": 4.2787, + "step": 2580 + }, + { + "epoch": 0.8254412440784346, + "grad_norm": 2.821227788925171, + "learning_rate": 4.674160418008728e-05, + "loss": 4.1523, + "step": 2581 + }, + { + "epoch": 0.8257610585860201, + "grad_norm": 1.119240403175354, + "learning_rate": 4.657511656800381e-05, + "loss": 4.2342, + "step": 2582 + }, + { + "epoch": 0.8260808730936057, + "grad_norm": 3.9736719131469727, + "learning_rate": 4.6408901032129476e-05, + "loss": 4.2035, + "step": 2583 + }, + { + "epoch": 0.8264006876011913, + "grad_norm": 1.3923814296722412, + "learning_rate": 4.624295775091282e-05, + "loss": 4.0491, + "step": 2584 + }, + { + "epoch": 0.8267205021087769, + "grad_norm": 1.088600516319275, + "learning_rate": 4.6077286902510144e-05, + "loss": 4.1741, + "step": 2585 + }, + { + "epoch": 0.8270403166163625, + "grad_norm": 3.4332587718963623, + "learning_rate": 4.591188866478513e-05, + "loss": 4.1546, + "step": 2586 + }, + { + "epoch": 0.8273601311239481, + "grad_norm": 1.0963387489318848, + "learning_rate": 4.574676321530891e-05, + "loss": 4.1646, + "step": 2587 + }, + { + "epoch": 0.8276799456315337, + "grad_norm": 3.2208251953125, + "learning_rate": 4.558191073135957e-05, + "loss": 4.269, + "step": 2588 + }, + { + "epoch": 0.8279997601391194, + "grad_norm": 0.8735865950584412, + "learning_rate": 4.541733138992231e-05, + "loss": 4.2057, + "step": 2589 + }, + { + "epoch": 0.8283195746467049, + "grad_norm": 2.1947598457336426, + "learning_rate": 4.525302536768901e-05, + "loss": 4.2577, + "step": 2590 + }, + { + "epoch": 0.8286393891542905, + "grad_norm": 1.4556514024734497, + "learning_rate": 4.5088992841058214e-05, + "loss": 4.2059, + "step": 2591 + }, + { + "epoch": 0.8289592036618761, + "grad_norm": 1.151827096939087, + "learning_rate": 4.4925233986134614e-05, + "loss": 4.2267, + "step": 2592 + }, + { + "epoch": 0.8292790181694617, + "grad_norm": 1.0784119367599487, + "learning_rate": 4.4761748978729305e-05, + "loss": 4.1132, + "step": 2593 + }, + { + "epoch": 0.8295988326770473, + "grad_norm": 0.9911250472068787, + "learning_rate": 4.4598537994359297e-05, + "loss": 4.0945, + "step": 2594 + }, + { + "epoch": 0.8299186471846329, + "grad_norm": 1.0901545286178589, + "learning_rate": 4.443560120824748e-05, + "loss": 4.0978, + "step": 2595 + }, + { + "epoch": 0.8302384616922185, + "grad_norm": 1.0317355394363403, + "learning_rate": 4.427293879532231e-05, + "loss": 4.1454, + "step": 2596 + }, + { + "epoch": 0.8305582761998042, + "grad_norm": 1.0607333183288574, + "learning_rate": 4.411055093021758e-05, + "loss": 4.1176, + "step": 2597 + }, + { + "epoch": 0.8308780907073897, + "grad_norm": 0.852952778339386, + "learning_rate": 4.394843778727247e-05, + "loss": 4.1647, + "step": 2598 + }, + { + "epoch": 0.8311979052149753, + "grad_norm": 2.547391653060913, + "learning_rate": 4.3786599540531164e-05, + "loss": 4.1912, + "step": 2599 + }, + { + "epoch": 0.8315177197225609, + "grad_norm": 0.7389708161354065, + "learning_rate": 4.362503636374277e-05, + "loss": 4.2385, + "step": 2600 + }, + { + "epoch": 0.8315177197225609, + "eval_runtime": 21.2151, + "eval_samples_per_second": 89.417, + "eval_steps_per_second": 22.39, + "step": 2600 + }, + { + "epoch": 0.8318375342301465, + "grad_norm": 0.7786986827850342, + "learning_rate": 4.346374843036104e-05, + "loss": 4.1167, + "step": 2601 + }, + { + "epoch": 0.8321573487377321, + "grad_norm": 1.9430526494979858, + "learning_rate": 4.3302735913544174e-05, + "loss": 4.1328, + "step": 2602 + }, + { + "epoch": 0.8324771632453177, + "grad_norm": 0.8430537581443787, + "learning_rate": 4.314199898615481e-05, + "loss": 4.167, + "step": 2603 + }, + { + "epoch": 0.8327969777529033, + "grad_norm": 1.777001142501831, + "learning_rate": 4.298153782075946e-05, + "loss": 4.1522, + "step": 2604 + }, + { + "epoch": 0.833116792260489, + "grad_norm": 0.758348822593689, + "learning_rate": 4.2821352589628944e-05, + "loss": 4.174, + "step": 2605 + }, + { + "epoch": 0.8334366067680745, + "grad_norm": 0.6821554899215698, + "learning_rate": 4.26614434647377e-05, + "loss": 4.2252, + "step": 2606 + }, + { + "epoch": 0.8337564212756601, + "grad_norm": 1.0476925373077393, + "learning_rate": 4.25018106177635e-05, + "loss": 4.2049, + "step": 2607 + }, + { + "epoch": 0.8340762357832457, + "grad_norm": 1.4836349487304688, + "learning_rate": 4.2342454220087855e-05, + "loss": 4.1105, + "step": 2608 + }, + { + "epoch": 0.8343960502908313, + "grad_norm": 1.9636969566345215, + "learning_rate": 4.21833744427952e-05, + "loss": 4.1703, + "step": 2609 + }, + { + "epoch": 0.834715864798417, + "grad_norm": 9.261558532714844, + "learning_rate": 4.202457145667311e-05, + "loss": 4.2154, + "step": 2610 + }, + { + "epoch": 0.8350356793060025, + "grad_norm": 1.0543816089630127, + "learning_rate": 4.1866045432212214e-05, + "loss": 4.2397, + "step": 2611 + }, + { + "epoch": 0.8353554938135881, + "grad_norm": 1.2306382656097412, + "learning_rate": 4.1707796539605385e-05, + "loss": 4.2202, + "step": 2612 + }, + { + "epoch": 0.8356753083211738, + "grad_norm": 1.2134190797805786, + "learning_rate": 4.154982494874829e-05, + "loss": 4.1067, + "step": 2613 + }, + { + "epoch": 0.8359951228287593, + "grad_norm": 2.670654773712158, + "learning_rate": 4.139213082923862e-05, + "loss": 4.2078, + "step": 2614 + }, + { + "epoch": 0.8363149373363449, + "grad_norm": 1.7149115800857544, + "learning_rate": 4.12347143503764e-05, + "loss": 4.206, + "step": 2615 + }, + { + "epoch": 0.8366347518439305, + "grad_norm": 26.764354705810547, + "learning_rate": 4.107757568116352e-05, + "loss": 4.1932, + "step": 2616 + }, + { + "epoch": 0.8369545663515161, + "grad_norm": 0.9246897101402283, + "learning_rate": 4.092071499030355e-05, + "loss": 4.1713, + "step": 2617 + }, + { + "epoch": 0.8372743808591018, + "grad_norm": 1.910569429397583, + "learning_rate": 4.076413244620177e-05, + "loss": 4.1605, + "step": 2618 + }, + { + "epoch": 0.8375941953666873, + "grad_norm": 4.581338882446289, + "learning_rate": 4.060782821696458e-05, + "loss": 4.1375, + "step": 2619 + }, + { + "epoch": 0.8379140098742729, + "grad_norm": 0.7591927647590637, + "learning_rate": 4.0451802470399805e-05, + "loss": 4.1913, + "step": 2620 + }, + { + "epoch": 0.8382338243818586, + "grad_norm": 0.7027854323387146, + "learning_rate": 4.029605537401623e-05, + "loss": 4.1703, + "step": 2621 + }, + { + "epoch": 0.8385536388894441, + "grad_norm": 0.8651039004325867, + "learning_rate": 4.01405870950235e-05, + "loss": 4.2071, + "step": 2622 + }, + { + "epoch": 0.8388734533970297, + "grad_norm": 0.9622931480407715, + "learning_rate": 3.9985397800331965e-05, + "loss": 4.1492, + "step": 2623 + }, + { + "epoch": 0.8391932679046153, + "grad_norm": 0.7661730051040649, + "learning_rate": 3.983048765655225e-05, + "loss": 4.1584, + "step": 2624 + }, + { + "epoch": 0.8395130824122009, + "grad_norm": 8.227343559265137, + "learning_rate": 3.9675856829995513e-05, + "loss": 4.159, + "step": 2625 + }, + { + "epoch": 0.8398328969197866, + "grad_norm": 0.8512206673622131, + "learning_rate": 3.95215054866729e-05, + "loss": 4.1914, + "step": 2626 + }, + { + "epoch": 0.8401527114273721, + "grad_norm": 5.020411491394043, + "learning_rate": 3.936743379229572e-05, + "loss": 4.1392, + "step": 2627 + }, + { + "epoch": 0.8404725259349577, + "grad_norm": 2.949896812438965, + "learning_rate": 3.921364191227466e-05, + "loss": 4.1941, + "step": 2628 + }, + { + "epoch": 0.8407923404425434, + "grad_norm": 0.8112735748291016, + "learning_rate": 3.9060130011720345e-05, + "loss": 4.252, + "step": 2629 + }, + { + "epoch": 0.8411121549501289, + "grad_norm": 0.9454008936882019, + "learning_rate": 3.890689825544271e-05, + "loss": 4.2193, + "step": 2630 + }, + { + "epoch": 0.8414319694577145, + "grad_norm": 0.7736871242523193, + "learning_rate": 3.875394680795092e-05, + "loss": 4.1434, + "step": 2631 + }, + { + "epoch": 0.8417517839653001, + "grad_norm": 0.7958900332450867, + "learning_rate": 3.8601275833453224e-05, + "loss": 4.1634, + "step": 2632 + }, + { + "epoch": 0.8420715984728857, + "grad_norm": 0.817162036895752, + "learning_rate": 3.844888549585662e-05, + "loss": 4.2027, + "step": 2633 + }, + { + "epoch": 0.8423914129804714, + "grad_norm": 1.2179948091506958, + "learning_rate": 3.829677595876699e-05, + "loss": 4.1246, + "step": 2634 + }, + { + "epoch": 0.8427112274880569, + "grad_norm": 0.9808216094970703, + "learning_rate": 3.814494738548871e-05, + "loss": 4.1706, + "step": 2635 + }, + { + "epoch": 0.8430310419956425, + "grad_norm": 2.6890151500701904, + "learning_rate": 3.799339993902446e-05, + "loss": 4.1071, + "step": 2636 + }, + { + "epoch": 0.8433508565032282, + "grad_norm": 1.4320547580718994, + "learning_rate": 3.784213378207522e-05, + "loss": 4.2075, + "step": 2637 + }, + { + "epoch": 0.8436706710108137, + "grad_norm": 0.8642268776893616, + "learning_rate": 3.769114907703973e-05, + "loss": 4.1979, + "step": 2638 + }, + { + "epoch": 0.8439904855183993, + "grad_norm": 1.293900966644287, + "learning_rate": 3.7540445986014845e-05, + "loss": 4.0239, + "step": 2639 + }, + { + "epoch": 0.844310300025985, + "grad_norm": 1.5481514930725098, + "learning_rate": 3.739002467079488e-05, + "loss": 4.208, + "step": 2640 + }, + { + "epoch": 0.8446301145335705, + "grad_norm": 2.070423126220703, + "learning_rate": 3.723988529287176e-05, + "loss": 4.1942, + "step": 2641 + }, + { + "epoch": 0.8449499290411562, + "grad_norm": 1.3346174955368042, + "learning_rate": 3.709002801343478e-05, + "loss": 4.1619, + "step": 2642 + }, + { + "epoch": 0.8452697435487417, + "grad_norm": 1.2388272285461426, + "learning_rate": 3.6940452993370105e-05, + "loss": 4.2243, + "step": 2643 + }, + { + "epoch": 0.8455895580563273, + "grad_norm": 2.2560555934906006, + "learning_rate": 3.679116039326115e-05, + "loss": 4.1354, + "step": 2644 + }, + { + "epoch": 0.845909372563913, + "grad_norm": 1.7694833278656006, + "learning_rate": 3.664215037338785e-05, + "loss": 4.1872, + "step": 2645 + }, + { + "epoch": 0.8462291870714985, + "grad_norm": 1.034848928451538, + "learning_rate": 3.6493423093727084e-05, + "loss": 4.2059, + "step": 2646 + }, + { + "epoch": 0.8465490015790841, + "grad_norm": 2.749107599258423, + "learning_rate": 3.634497871395207e-05, + "loss": 4.252, + "step": 2647 + }, + { + "epoch": 0.8468688160866698, + "grad_norm": 2.254084587097168, + "learning_rate": 3.6196817393432085e-05, + "loss": 4.1798, + "step": 2648 + }, + { + "epoch": 0.8471886305942553, + "grad_norm": 4.71826171875, + "learning_rate": 3.604893929123284e-05, + "loss": 4.2003, + "step": 2649 + }, + { + "epoch": 0.847508445101841, + "grad_norm": 4.910301685333252, + "learning_rate": 3.590134456611562e-05, + "loss": 4.183, + "step": 2650 + }, + { + "epoch": 0.8478282596094265, + "grad_norm": 1.1861979961395264, + "learning_rate": 3.5754033376537947e-05, + "loss": 4.2284, + "step": 2651 + }, + { + "epoch": 0.8481480741170121, + "grad_norm": 0.7333687543869019, + "learning_rate": 3.560700588065252e-05, + "loss": 4.199, + "step": 2652 + }, + { + "epoch": 0.8484678886245978, + "grad_norm": 1.2275673151016235, + "learning_rate": 3.5460262236307657e-05, + "loss": 4.1722, + "step": 2653 + }, + { + "epoch": 0.8487877031321833, + "grad_norm": 1.9804834127426147, + "learning_rate": 3.531380260104698e-05, + "loss": 4.187, + "step": 2654 + }, + { + "epoch": 0.8491075176397689, + "grad_norm": 0.761823296546936, + "learning_rate": 3.516762713210891e-05, + "loss": 4.2288, + "step": 2655 + }, + { + "epoch": 0.8494273321473546, + "grad_norm": 1.2529851198196411, + "learning_rate": 3.502173598642728e-05, + "loss": 4.2028, + "step": 2656 + }, + { + "epoch": 0.8497471466549401, + "grad_norm": 1.3719714879989624, + "learning_rate": 3.4876129320630196e-05, + "loss": 4.0843, + "step": 2657 + }, + { + "epoch": 0.8500669611625258, + "grad_norm": 0.9276340007781982, + "learning_rate": 3.473080729104062e-05, + "loss": 4.2433, + "step": 2658 + }, + { + "epoch": 0.8503867756701113, + "grad_norm": 0.703091025352478, + "learning_rate": 3.4585770053675876e-05, + "loss": 4.1936, + "step": 2659 + }, + { + "epoch": 0.8507065901776969, + "grad_norm": 1.2949669361114502, + "learning_rate": 3.444101776424738e-05, + "loss": 4.1865, + "step": 2660 + }, + { + "epoch": 0.8510264046852826, + "grad_norm": 1.0420503616333008, + "learning_rate": 3.429655057816099e-05, + "loss": 4.1438, + "step": 2661 + }, + { + "epoch": 0.8513462191928681, + "grad_norm": 0.8074566125869751, + "learning_rate": 3.415236865051606e-05, + "loss": 4.144, + "step": 2662 + }, + { + "epoch": 0.8516660337004537, + "grad_norm": 5.127699851989746, + "learning_rate": 3.4008472136106046e-05, + "loss": 4.1738, + "step": 2663 + }, + { + "epoch": 0.8519858482080394, + "grad_norm": 1.1977909803390503, + "learning_rate": 3.3864861189417636e-05, + "loss": 4.2523, + "step": 2664 + }, + { + "epoch": 0.8523056627156249, + "grad_norm": 1.3616605997085571, + "learning_rate": 3.3721535964631195e-05, + "loss": 4.1554, + "step": 2665 + }, + { + "epoch": 0.8526254772232106, + "grad_norm": 1.021154522895813, + "learning_rate": 3.3578496615620307e-05, + "loss": 4.1876, + "step": 2666 + }, + { + "epoch": 0.8529452917307961, + "grad_norm": 1.1995530128479004, + "learning_rate": 3.343574329595157e-05, + "loss": 4.2393, + "step": 2667 + }, + { + "epoch": 0.8532651062383817, + "grad_norm": 1.744409203529358, + "learning_rate": 3.329327615888461e-05, + "loss": 4.1893, + "step": 2668 + }, + { + "epoch": 0.8535849207459674, + "grad_norm": 0.832141101360321, + "learning_rate": 3.315109535737155e-05, + "loss": 4.1662, + "step": 2669 + }, + { + "epoch": 0.8539047352535529, + "grad_norm": 1.521525263786316, + "learning_rate": 3.300920104405739e-05, + "loss": 4.1429, + "step": 2670 + }, + { + "epoch": 0.8542245497611385, + "grad_norm": 1.068145990371704, + "learning_rate": 3.2867593371279434e-05, + "loss": 4.1991, + "step": 2671 + }, + { + "epoch": 0.8545443642687242, + "grad_norm": 0.8391280174255371, + "learning_rate": 3.272627249106724e-05, + "loss": 4.1018, + "step": 2672 + }, + { + "epoch": 0.8548641787763097, + "grad_norm": 0.9857653379440308, + "learning_rate": 3.258523855514258e-05, + "loss": 4.0746, + "step": 2673 + }, + { + "epoch": 0.8551839932838954, + "grad_norm": 1.142601728439331, + "learning_rate": 3.244449171491896e-05, + "loss": 4.1887, + "step": 2674 + }, + { + "epoch": 0.8555038077914809, + "grad_norm": 1.5742888450622559, + "learning_rate": 3.230403212150179e-05, + "loss": 4.2561, + "step": 2675 + }, + { + "epoch": 0.8558236222990665, + "grad_norm": 0.9153767228126526, + "learning_rate": 3.216385992568813e-05, + "loss": 4.203, + "step": 2676 + }, + { + "epoch": 0.8561434368066522, + "grad_norm": 1.0245417356491089, + "learning_rate": 3.202397527796637e-05, + "loss": 4.1199, + "step": 2677 + }, + { + "epoch": 0.8564632513142377, + "grad_norm": 1.3338567018508911, + "learning_rate": 3.188437832851639e-05, + "loss": 4.1724, + "step": 2678 + }, + { + "epoch": 0.8567830658218234, + "grad_norm": 9.496756553649902, + "learning_rate": 3.1745069227208894e-05, + "loss": 4.1568, + "step": 2679 + }, + { + "epoch": 0.857102880329409, + "grad_norm": 1.2807888984680176, + "learning_rate": 3.160604812360579e-05, + "loss": 4.1655, + "step": 2680 + }, + { + "epoch": 0.8574226948369945, + "grad_norm": 1.5378843545913696, + "learning_rate": 3.146731516695974e-05, + "loss": 4.105, + "step": 2681 + }, + { + "epoch": 0.8577425093445802, + "grad_norm": 1.0192182064056396, + "learning_rate": 3.1328870506214044e-05, + "loss": 4.1627, + "step": 2682 + }, + { + "epoch": 0.8580623238521657, + "grad_norm": 0.8195664286613464, + "learning_rate": 3.119071429000254e-05, + "loss": 4.2441, + "step": 2683 + }, + { + "epoch": 0.8583821383597513, + "grad_norm": 1.5050681829452515, + "learning_rate": 3.105284666664918e-05, + "loss": 4.1244, + "step": 2684 + }, + { + "epoch": 0.858701952867337, + "grad_norm": 0.9606510400772095, + "learning_rate": 3.091526778416833e-05, + "loss": 4.1361, + "step": 2685 + }, + { + "epoch": 0.8590217673749225, + "grad_norm": 0.9771678447723389, + "learning_rate": 3.077797779026428e-05, + "loss": 4.1959, + "step": 2686 + }, + { + "epoch": 0.8593415818825082, + "grad_norm": 1.325870394706726, + "learning_rate": 3.064097683233121e-05, + "loss": 4.1822, + "step": 2687 + }, + { + "epoch": 0.8596613963900938, + "grad_norm": 1.6404540538787842, + "learning_rate": 3.0504265057452815e-05, + "loss": 4.0909, + "step": 2688 + }, + { + "epoch": 0.8599812108976793, + "grad_norm": 0.6895898580551147, + "learning_rate": 3.036784261240255e-05, + "loss": 4.013, + "step": 2689 + }, + { + "epoch": 0.860301025405265, + "grad_norm": 2.376892566680908, + "learning_rate": 3.0231709643643086e-05, + "loss": 4.1956, + "step": 2690 + }, + { + "epoch": 0.8606208399128505, + "grad_norm": 3.4172263145446777, + "learning_rate": 3.0095866297326455e-05, + "loss": 4.1853, + "step": 2691 + }, + { + "epoch": 0.8609406544204361, + "grad_norm": 1.4309022426605225, + "learning_rate": 2.996031271929369e-05, + "loss": 4.1776, + "step": 2692 + }, + { + "epoch": 0.8612604689280218, + "grad_norm": 0.8799441456794739, + "learning_rate": 2.982504905507461e-05, + "loss": 4.1314, + "step": 2693 + }, + { + "epoch": 0.8615802834356073, + "grad_norm": 1.4177411794662476, + "learning_rate": 2.969007544988793e-05, + "loss": 4.0717, + "step": 2694 + }, + { + "epoch": 0.861900097943193, + "grad_norm": 0.7793790698051453, + "learning_rate": 2.9555392048640924e-05, + "loss": 4.1588, + "step": 2695 + }, + { + "epoch": 0.8622199124507786, + "grad_norm": 1.4761580228805542, + "learning_rate": 2.9420998995929267e-05, + "loss": 4.196, + "step": 2696 + }, + { + "epoch": 0.8625397269583641, + "grad_norm": 1.1391254663467407, + "learning_rate": 2.9286896436037076e-05, + "loss": 4.1678, + "step": 2697 + }, + { + "epoch": 0.8628595414659498, + "grad_norm": 1.1287897825241089, + "learning_rate": 2.9153084512936285e-05, + "loss": 4.2073, + "step": 2698 + }, + { + "epoch": 0.8631793559735353, + "grad_norm": 6.10676908493042, + "learning_rate": 2.9019563370287112e-05, + "loss": 4.1614, + "step": 2699 + }, + { + "epoch": 0.8634991704811209, + "grad_norm": 6.598787784576416, + "learning_rate": 2.8886333151437292e-05, + "loss": 4.118, + "step": 2700 + }, + { + "epoch": 0.8634991704811209, + "eval_runtime": 21.161, + "eval_samples_per_second": 89.646, + "eval_steps_per_second": 22.447, + "step": 2700 + }, + { + "epoch": 0.8638189849887066, + "grad_norm": 2.090200901031494, + "learning_rate": 2.875339399942257e-05, + "loss": 4.1775, + "step": 2701 + }, + { + "epoch": 0.8641387994962921, + "grad_norm": 1.294780969619751, + "learning_rate": 2.862074605696605e-05, + "loss": 4.1418, + "step": 2702 + }, + { + "epoch": 0.8644586140038778, + "grad_norm": 1.1893248558044434, + "learning_rate": 2.848838946647801e-05, + "loss": 4.1896, + "step": 2703 + }, + { + "epoch": 0.8647784285114634, + "grad_norm": 1.170419454574585, + "learning_rate": 2.835632437005626e-05, + "loss": 4.0535, + "step": 2704 + }, + { + "epoch": 0.8650982430190489, + "grad_norm": 0.6554948687553406, + "learning_rate": 2.8224550909485344e-05, + "loss": 4.2454, + "step": 2705 + }, + { + "epoch": 0.8654180575266346, + "grad_norm": 1.122994303703308, + "learning_rate": 2.8093069226236865e-05, + "loss": 4.1762, + "step": 2706 + }, + { + "epoch": 0.8657378720342201, + "grad_norm": 8.86845588684082, + "learning_rate": 2.796187946146937e-05, + "loss": 4.1847, + "step": 2707 + }, + { + "epoch": 0.8660576865418057, + "grad_norm": 1.1721240282058716, + "learning_rate": 2.7830981756027636e-05, + "loss": 4.0553, + "step": 2708 + }, + { + "epoch": 0.8663775010493914, + "grad_norm": 0.5891955494880676, + "learning_rate": 2.7700376250443147e-05, + "loss": 4.1238, + "step": 2709 + }, + { + "epoch": 0.8666973155569769, + "grad_norm": 1.2026443481445312, + "learning_rate": 2.757006308493347e-05, + "loss": 4.1576, + "step": 2710 + }, + { + "epoch": 0.8670171300645626, + "grad_norm": 4.224611282348633, + "learning_rate": 2.7440042399402496e-05, + "loss": 4.165, + "step": 2711 + }, + { + "epoch": 0.8673369445721482, + "grad_norm": 0.8235098719596863, + "learning_rate": 2.7310314333440097e-05, + "loss": 4.0855, + "step": 2712 + }, + { + "epoch": 0.8676567590797337, + "grad_norm": 0.7895171642303467, + "learning_rate": 2.7180879026321866e-05, + "loss": 4.1752, + "step": 2713 + }, + { + "epoch": 0.8679765735873194, + "grad_norm": 1.1421613693237305, + "learning_rate": 2.7051736617009277e-05, + "loss": 4.1231, + "step": 2714 + }, + { + "epoch": 0.868296388094905, + "grad_norm": 1.1233763694763184, + "learning_rate": 2.6922887244149126e-05, + "loss": 4.2013, + "step": 2715 + }, + { + "epoch": 0.8686162026024905, + "grad_norm": 3.2594377994537354, + "learning_rate": 2.6794331046073724e-05, + "loss": 4.0775, + "step": 2716 + }, + { + "epoch": 0.8689360171100762, + "grad_norm": 0.7115456461906433, + "learning_rate": 2.6666068160800702e-05, + "loss": 4.2211, + "step": 2717 + }, + { + "epoch": 0.8692558316176617, + "grad_norm": 0.8150424957275391, + "learning_rate": 2.6538098726032675e-05, + "loss": 4.1396, + "step": 2718 + }, + { + "epoch": 0.8695756461252474, + "grad_norm": 1.0282390117645264, + "learning_rate": 2.6410422879157313e-05, + "loss": 4.1756, + "step": 2719 + }, + { + "epoch": 0.869895460632833, + "grad_norm": 0.71715247631073, + "learning_rate": 2.628304075724693e-05, + "loss": 4.1053, + "step": 2720 + }, + { + "epoch": 0.8702152751404185, + "grad_norm": 1.9775540828704834, + "learning_rate": 2.6155952497058643e-05, + "loss": 4.1498, + "step": 2721 + }, + { + "epoch": 0.8705350896480042, + "grad_norm": 0.8759939670562744, + "learning_rate": 2.6029158235033997e-05, + "loss": 4.2499, + "step": 2722 + }, + { + "epoch": 0.8708549041555897, + "grad_norm": 1.5837782621383667, + "learning_rate": 2.5902658107299078e-05, + "loss": 4.2111, + "step": 2723 + }, + { + "epoch": 0.8711747186631753, + "grad_norm": 1.1584738492965698, + "learning_rate": 2.5776452249663847e-05, + "loss": 4.1689, + "step": 2724 + }, + { + "epoch": 0.871494533170761, + "grad_norm": 3.0437989234924316, + "learning_rate": 2.5650540797622687e-05, + "loss": 4.2654, + "step": 2725 + }, + { + "epoch": 0.8718143476783465, + "grad_norm": 2.3175833225250244, + "learning_rate": 2.5524923886353697e-05, + "loss": 4.1203, + "step": 2726 + }, + { + "epoch": 0.8721341621859322, + "grad_norm": 1.0406721830368042, + "learning_rate": 2.5399601650718838e-05, + "loss": 4.2196, + "step": 2727 + }, + { + "epoch": 0.8724539766935178, + "grad_norm": 0.9513920545578003, + "learning_rate": 2.5274574225263776e-05, + "loss": 4.1639, + "step": 2728 + }, + { + "epoch": 0.8727737912011033, + "grad_norm": 1.2580915689468384, + "learning_rate": 2.5149841744217415e-05, + "loss": 4.0963, + "step": 2729 + }, + { + "epoch": 0.873093605708689, + "grad_norm": 0.8016980886459351, + "learning_rate": 2.5025404341492327e-05, + "loss": 4.1399, + "step": 2730 + }, + { + "epoch": 0.8734134202162745, + "grad_norm": 0.8207408785820007, + "learning_rate": 2.4901262150684055e-05, + "loss": 4.1886, + "step": 2731 + }, + { + "epoch": 0.8737332347238601, + "grad_norm": 1.560724139213562, + "learning_rate": 2.4777415305071346e-05, + "loss": 4.2082, + "step": 2732 + }, + { + "epoch": 0.8740530492314458, + "grad_norm": 1.0942937135696411, + "learning_rate": 2.4653863937615813e-05, + "loss": 4.1878, + "step": 2733 + }, + { + "epoch": 0.8743728637390313, + "grad_norm": 1.9375314712524414, + "learning_rate": 2.4530608180961786e-05, + "loss": 4.1725, + "step": 2734 + }, + { + "epoch": 0.874692678246617, + "grad_norm": 4.972708702087402, + "learning_rate": 2.440764816743631e-05, + "loss": 4.1646, + "step": 2735 + }, + { + "epoch": 0.8750124927542026, + "grad_norm": 4.094714641571045, + "learning_rate": 2.428498402904889e-05, + "loss": 4.1496, + "step": 2736 + }, + { + "epoch": 0.8753323072617881, + "grad_norm": 0.7932257652282715, + "learning_rate": 2.416261589749139e-05, + "loss": 4.207, + "step": 2737 + }, + { + "epoch": 0.8756521217693738, + "grad_norm": 2.857753276824951, + "learning_rate": 2.4040543904137942e-05, + "loss": 4.1648, + "step": 2738 + }, + { + "epoch": 0.8759719362769594, + "grad_norm": 3.030517816543579, + "learning_rate": 2.391876818004452e-05, + "loss": 4.157, + "step": 2739 + }, + { + "epoch": 0.8762917507845449, + "grad_norm": 0.9735044836997986, + "learning_rate": 2.3797288855949382e-05, + "loss": 4.0529, + "step": 2740 + }, + { + "epoch": 0.8766115652921306, + "grad_norm": 1.4307125806808472, + "learning_rate": 2.3676106062272126e-05, + "loss": 4.188, + "step": 2741 + }, + { + "epoch": 0.8769313797997161, + "grad_norm": 1.1469011306762695, + "learning_rate": 2.3555219929114454e-05, + "loss": 4.1993, + "step": 2742 + }, + { + "epoch": 0.8772511943073018, + "grad_norm": 1.0959744453430176, + "learning_rate": 2.343463058625932e-05, + "loss": 4.1433, + "step": 2743 + }, + { + "epoch": 0.8775710088148874, + "grad_norm": 1.7874549627304077, + "learning_rate": 2.331433816317102e-05, + "loss": 4.2492, + "step": 2744 + }, + { + "epoch": 0.8778908233224729, + "grad_norm": 3.5088582038879395, + "learning_rate": 2.3194342788995257e-05, + "loss": 4.0575, + "step": 2745 + }, + { + "epoch": 0.8782106378300586, + "grad_norm": 1.305413007736206, + "learning_rate": 2.307464459255851e-05, + "loss": 4.1609, + "step": 2746 + }, + { + "epoch": 0.8785304523376442, + "grad_norm": 0.9300746917724609, + "learning_rate": 2.2955243702368652e-05, + "loss": 4.1378, + "step": 2747 + }, + { + "epoch": 0.8788502668452298, + "grad_norm": 1.9145596027374268, + "learning_rate": 2.2836140246613977e-05, + "loss": 4.1626, + "step": 2748 + }, + { + "epoch": 0.8791700813528154, + "grad_norm": 0.7742648720741272, + "learning_rate": 2.271733435316363e-05, + "loss": 4.1933, + "step": 2749 + }, + { + "epoch": 0.8794898958604009, + "grad_norm": 2.0791635513305664, + "learning_rate": 2.2598826149567352e-05, + "loss": 4.2223, + "step": 2750 + }, + { + "epoch": 0.8798097103679866, + "grad_norm": 15.671849250793457, + "learning_rate": 2.2480615763055032e-05, + "loss": 4.1274, + "step": 2751 + }, + { + "epoch": 0.8801295248755722, + "grad_norm": 2.891798734664917, + "learning_rate": 2.2362703320537156e-05, + "loss": 4.1232, + "step": 2752 + }, + { + "epoch": 0.8804493393831577, + "grad_norm": 1.283813714981079, + "learning_rate": 2.2245088948604095e-05, + "loss": 4.1748, + "step": 2753 + }, + { + "epoch": 0.8807691538907434, + "grad_norm": 1.2309876680374146, + "learning_rate": 2.2127772773526342e-05, + "loss": 4.1488, + "step": 2754 + }, + { + "epoch": 0.881088968398329, + "grad_norm": 1.2558320760726929, + "learning_rate": 2.201075492125415e-05, + "loss": 4.133, + "step": 2755 + }, + { + "epoch": 0.8814087829059146, + "grad_norm": 1.0078858137130737, + "learning_rate": 2.1894035517417486e-05, + "loss": 4.2008, + "step": 2756 + }, + { + "epoch": 0.8817285974135002, + "grad_norm": 0.9881557822227478, + "learning_rate": 2.1777614687326116e-05, + "loss": 4.1111, + "step": 2757 + }, + { + "epoch": 0.8820484119210857, + "grad_norm": 1.1171656847000122, + "learning_rate": 2.166149255596896e-05, + "loss": 4.141, + "step": 2758 + }, + { + "epoch": 0.8823682264286714, + "grad_norm": 0.8496936559677124, + "learning_rate": 2.154566924801453e-05, + "loss": 4.157, + "step": 2759 + }, + { + "epoch": 0.882688040936257, + "grad_norm": 1.4101358652114868, + "learning_rate": 2.1430144887810218e-05, + "loss": 4.1971, + "step": 2760 + }, + { + "epoch": 0.8830078554438425, + "grad_norm": 7.661594867706299, + "learning_rate": 2.131491959938275e-05, + "loss": 4.1623, + "step": 2761 + }, + { + "epoch": 0.8833276699514282, + "grad_norm": 0.632546603679657, + "learning_rate": 2.119999350643764e-05, + "loss": 4.1266, + "step": 2762 + }, + { + "epoch": 0.8836474844590138, + "grad_norm": 1.281146764755249, + "learning_rate": 2.108536673235922e-05, + "loss": 4.0977, + "step": 2763 + }, + { + "epoch": 0.8839672989665994, + "grad_norm": 1.0984668731689453, + "learning_rate": 2.0971039400210453e-05, + "loss": 4.1923, + "step": 2764 + }, + { + "epoch": 0.884287113474185, + "grad_norm": 0.8950844407081604, + "learning_rate": 2.0857011632732755e-05, + "loss": 4.1358, + "step": 2765 + }, + { + "epoch": 0.8846069279817705, + "grad_norm": 1.893420696258545, + "learning_rate": 2.0743283552346067e-05, + "loss": 4.1637, + "step": 2766 + }, + { + "epoch": 0.8849267424893562, + "grad_norm": 2.8926994800567627, + "learning_rate": 2.062985528114852e-05, + "loss": 4.1089, + "step": 2767 + }, + { + "epoch": 0.8852465569969418, + "grad_norm": 1.346966028213501, + "learning_rate": 2.0516726940916372e-05, + "loss": 4.0847, + "step": 2768 + }, + { + "epoch": 0.8855663715045273, + "grad_norm": 0.7872565984725952, + "learning_rate": 2.0403898653103867e-05, + "loss": 4.2901, + "step": 2769 + }, + { + "epoch": 0.885886186012113, + "grad_norm": 1.0383862257003784, + "learning_rate": 2.029137053884311e-05, + "loss": 4.2264, + "step": 2770 + }, + { + "epoch": 0.8862060005196986, + "grad_norm": 0.8865336179733276, + "learning_rate": 2.0179142718943964e-05, + "loss": 4.1907, + "step": 2771 + }, + { + "epoch": 0.8865258150272842, + "grad_norm": 5.493813991546631, + "learning_rate": 2.006721531389388e-05, + "loss": 4.2112, + "step": 2772 + }, + { + "epoch": 0.8868456295348698, + "grad_norm": 0.7308863997459412, + "learning_rate": 1.9955588443857807e-05, + "loss": 4.1664, + "step": 2773 + }, + { + "epoch": 0.8871654440424553, + "grad_norm": 2.379464864730835, + "learning_rate": 1.9844262228678077e-05, + "loss": 4.152, + "step": 2774 + }, + { + "epoch": 0.887485258550041, + "grad_norm": 3.1958096027374268, + "learning_rate": 1.9733236787874053e-05, + "loss": 4.1641, + "step": 2775 + }, + { + "epoch": 0.8878050730576266, + "grad_norm": 1.5092412233352661, + "learning_rate": 1.9622512240642386e-05, + "loss": 4.1753, + "step": 2776 + }, + { + "epoch": 0.8881248875652121, + "grad_norm": 2.852043867111206, + "learning_rate": 1.9512088705856654e-05, + "loss": 4.1901, + "step": 2777 + }, + { + "epoch": 0.8884447020727978, + "grad_norm": 1.1756094694137573, + "learning_rate": 1.9401966302067262e-05, + "loss": 4.165, + "step": 2778 + }, + { + "epoch": 0.8887645165803834, + "grad_norm": 1.4411547183990479, + "learning_rate": 1.9292145147501204e-05, + "loss": 4.1408, + "step": 2779 + }, + { + "epoch": 0.889084331087969, + "grad_norm": 1.3202412128448486, + "learning_rate": 1.91826253600622e-05, + "loss": 4.1129, + "step": 2780 + }, + { + "epoch": 0.8894041455955546, + "grad_norm": 2.877309799194336, + "learning_rate": 1.907340705733036e-05, + "loss": 4.173, + "step": 2781 + }, + { + "epoch": 0.8897239601031401, + "grad_norm": 0.7331299185752869, + "learning_rate": 1.8964490356562155e-05, + "loss": 4.1812, + "step": 2782 + }, + { + "epoch": 0.8900437746107258, + "grad_norm": 1.0688873529434204, + "learning_rate": 1.8855875374690288e-05, + "loss": 4.1218, + "step": 2783 + }, + { + "epoch": 0.8903635891183114, + "grad_norm": 2.4238147735595703, + "learning_rate": 1.8747562228323344e-05, + "loss": 4.1279, + "step": 2784 + }, + { + "epoch": 0.8906834036258969, + "grad_norm": 1.725925087928772, + "learning_rate": 1.863955103374607e-05, + "loss": 4.1675, + "step": 2785 + }, + { + "epoch": 0.8910032181334826, + "grad_norm": 1.66714608669281, + "learning_rate": 1.8531841906918976e-05, + "loss": 4.1463, + "step": 2786 + }, + { + "epoch": 0.8913230326410682, + "grad_norm": 0.784699022769928, + "learning_rate": 1.8424434963478262e-05, + "loss": 4.1571, + "step": 2787 + }, + { + "epoch": 0.8916428471486538, + "grad_norm": 1.3017423152923584, + "learning_rate": 1.8317330318735757e-05, + "loss": 4.1792, + "step": 2788 + }, + { + "epoch": 0.8919626616562394, + "grad_norm": 0.8157433271408081, + "learning_rate": 1.8210528087678577e-05, + "loss": 4.1515, + "step": 2789 + }, + { + "epoch": 0.892282476163825, + "grad_norm": 1.1038450002670288, + "learning_rate": 1.810402838496937e-05, + "loss": 4.1337, + "step": 2790 + }, + { + "epoch": 0.8926022906714106, + "grad_norm": 2.374354600906372, + "learning_rate": 1.799783132494581e-05, + "loss": 4.1146, + "step": 2791 + }, + { + "epoch": 0.8929221051789962, + "grad_norm": 7.83481502532959, + "learning_rate": 1.789193702162086e-05, + "loss": 4.267, + "step": 2792 + }, + { + "epoch": 0.8932419196865817, + "grad_norm": 4.094333171844482, + "learning_rate": 1.7786345588682317e-05, + "loss": 4.1978, + "step": 2793 + }, + { + "epoch": 0.8935617341941674, + "grad_norm": 1.0507333278656006, + "learning_rate": 1.7681057139492792e-05, + "loss": 4.1926, + "step": 2794 + }, + { + "epoch": 0.893881548701753, + "grad_norm": 0.6722399592399597, + "learning_rate": 1.7576071787089672e-05, + "loss": 4.1486, + "step": 2795 + }, + { + "epoch": 0.8942013632093386, + "grad_norm": 0.953079342842102, + "learning_rate": 1.7471389644184897e-05, + "loss": 4.2506, + "step": 2796 + }, + { + "epoch": 0.8945211777169242, + "grad_norm": 2.9529361724853516, + "learning_rate": 1.7367010823164862e-05, + "loss": 4.2109, + "step": 2797 + }, + { + "epoch": 0.8948409922245097, + "grad_norm": 3.412799596786499, + "learning_rate": 1.726293543609053e-05, + "loss": 4.0979, + "step": 2798 + }, + { + "epoch": 0.8951608067320954, + "grad_norm": 0.5511144995689392, + "learning_rate": 1.7159163594696756e-05, + "loss": 4.1403, + "step": 2799 + }, + { + "epoch": 0.895480621239681, + "grad_norm": 0.8664823770523071, + "learning_rate": 1.7055695410392823e-05, + "loss": 4.0889, + "step": 2800 + }, + { + "epoch": 0.895480621239681, + "eval_runtime": 22.1793, + "eval_samples_per_second": 85.53, + "eval_steps_per_second": 21.416, + "step": 2800 + }, + { + "epoch": 0.8958004357472665, + "grad_norm": 0.9239434003829956, + "learning_rate": 1.695253099426177e-05, + "loss": 4.0582, + "step": 2801 + }, + { + "epoch": 0.8961202502548522, + "grad_norm": 2.667048931121826, + "learning_rate": 1.6849670457060605e-05, + "loss": 4.1386, + "step": 2802 + }, + { + "epoch": 0.8964400647624378, + "grad_norm": 0.8980498313903809, + "learning_rate": 1.6747113909220155e-05, + "loss": 4.1541, + "step": 2803 + }, + { + "epoch": 0.8967598792700234, + "grad_norm": 0.9032712578773499, + "learning_rate": 1.6644861460844782e-05, + "loss": 4.0576, + "step": 2804 + }, + { + "epoch": 0.897079693777609, + "grad_norm": 1.0201606750488281, + "learning_rate": 1.6542913221712506e-05, + "loss": 4.2342, + "step": 2805 + }, + { + "epoch": 0.8973995082851945, + "grad_norm": 0.9507570266723633, + "learning_rate": 1.6441269301274572e-05, + "loss": 4.1576, + "step": 2806 + }, + { + "epoch": 0.8977193227927802, + "grad_norm": 0.9103635549545288, + "learning_rate": 1.633992980865556e-05, + "loss": 4.1784, + "step": 2807 + }, + { + "epoch": 0.8980391373003658, + "grad_norm": 1.0239713191986084, + "learning_rate": 1.6238894852653338e-05, + "loss": 4.1477, + "step": 2808 + }, + { + "epoch": 0.8983589518079513, + "grad_norm": 0.843490481376648, + "learning_rate": 1.6138164541738674e-05, + "loss": 4.2171, + "step": 2809 + }, + { + "epoch": 0.898678766315537, + "grad_norm": 0.6031708717346191, + "learning_rate": 1.6037738984055425e-05, + "loss": 4.1281, + "step": 2810 + }, + { + "epoch": 0.8989985808231226, + "grad_norm": 0.8581626415252686, + "learning_rate": 1.5937618287420052e-05, + "loss": 4.1549, + "step": 2811 + }, + { + "epoch": 0.8993183953307082, + "grad_norm": 0.8034144043922424, + "learning_rate": 1.583780255932193e-05, + "loss": 4.1411, + "step": 2812 + }, + { + "epoch": 0.8996382098382938, + "grad_norm": 1.6237574815750122, + "learning_rate": 1.5738291906922883e-05, + "loss": 4.1204, + "step": 2813 + }, + { + "epoch": 0.8999580243458793, + "grad_norm": 0.849178671836853, + "learning_rate": 1.5639086437057314e-05, + "loss": 4.1114, + "step": 2814 + }, + { + "epoch": 0.900277838853465, + "grad_norm": 4.505744457244873, + "learning_rate": 1.5540186256231823e-05, + "loss": 4.1822, + "step": 2815 + }, + { + "epoch": 0.9005976533610506, + "grad_norm": 1.0464649200439453, + "learning_rate": 1.5441591470625414e-05, + "loss": 4.1213, + "step": 2816 + }, + { + "epoch": 0.9009174678686362, + "grad_norm": 0.8497574329376221, + "learning_rate": 1.534330218608918e-05, + "loss": 4.22, + "step": 2817 + }, + { + "epoch": 0.9012372823762218, + "grad_norm": 1.2809534072875977, + "learning_rate": 1.5245318508146175e-05, + "loss": 4.1489, + "step": 2818 + }, + { + "epoch": 0.9015570968838074, + "grad_norm": 0.6939866542816162, + "learning_rate": 1.5147640541991424e-05, + "loss": 4.2266, + "step": 2819 + }, + { + "epoch": 0.901876911391393, + "grad_norm": 0.9401768445968628, + "learning_rate": 1.5050268392491639e-05, + "loss": 4.2173, + "step": 2820 + }, + { + "epoch": 0.9021967258989786, + "grad_norm": 0.993955671787262, + "learning_rate": 1.4953202164185297e-05, + "loss": 4.2446, + "step": 2821 + }, + { + "epoch": 0.9025165404065641, + "grad_norm": 0.832715630531311, + "learning_rate": 1.4856441961282472e-05, + "loss": 4.2171, + "step": 2822 + }, + { + "epoch": 0.9028363549141498, + "grad_norm": 0.9631035327911377, + "learning_rate": 1.4759987887664537e-05, + "loss": 4.1058, + "step": 2823 + }, + { + "epoch": 0.9031561694217354, + "grad_norm": 0.6047245264053345, + "learning_rate": 1.4663840046884423e-05, + "loss": 4.0924, + "step": 2824 + }, + { + "epoch": 0.903475983929321, + "grad_norm": 0.7924666404724121, + "learning_rate": 1.456799854216606e-05, + "loss": 4.0921, + "step": 2825 + }, + { + "epoch": 0.9037957984369066, + "grad_norm": 3.435307741165161, + "learning_rate": 1.447246347640464e-05, + "loss": 4.1721, + "step": 2826 + }, + { + "epoch": 0.9041156129444922, + "grad_norm": 0.631420373916626, + "learning_rate": 1.437723495216635e-05, + "loss": 4.0608, + "step": 2827 + }, + { + "epoch": 0.9044354274520778, + "grad_norm": 1.0624877214431763, + "learning_rate": 1.4282313071688211e-05, + "loss": 4.2288, + "step": 2828 + }, + { + "epoch": 0.9047552419596634, + "grad_norm": 0.6305270195007324, + "learning_rate": 1.4187697936878172e-05, + "loss": 4.1036, + "step": 2829 + }, + { + "epoch": 0.905075056467249, + "grad_norm": 0.6638223528862, + "learning_rate": 1.4093389649314613e-05, + "loss": 4.1841, + "step": 2830 + }, + { + "epoch": 0.9053948709748346, + "grad_norm": 0.8302531242370605, + "learning_rate": 1.399938831024674e-05, + "loss": 4.2903, + "step": 2831 + }, + { + "epoch": 0.9057146854824202, + "grad_norm": 19.70148468017578, + "learning_rate": 1.3905694020594093e-05, + "loss": 4.2677, + "step": 2832 + }, + { + "epoch": 0.9060344999900058, + "grad_norm": 1.5270586013793945, + "learning_rate": 1.3812306880946577e-05, + "loss": 4.0878, + "step": 2833 + }, + { + "epoch": 0.9063543144975914, + "grad_norm": 1.118834376335144, + "learning_rate": 1.3719226991564392e-05, + "loss": 4.1221, + "step": 2834 + }, + { + "epoch": 0.906674129005177, + "grad_norm": 1.0337523221969604, + "learning_rate": 1.3626454452377734e-05, + "loss": 4.202, + "step": 2835 + }, + { + "epoch": 0.9069939435127626, + "grad_norm": 0.6634312272071838, + "learning_rate": 1.3533989362987063e-05, + "loss": 4.2284, + "step": 2836 + }, + { + "epoch": 0.9073137580203482, + "grad_norm": 3.2054545879364014, + "learning_rate": 1.3441831822662441e-05, + "loss": 4.1661, + "step": 2837 + }, + { + "epoch": 0.9076335725279338, + "grad_norm": 0.8443681597709656, + "learning_rate": 1.3349981930344156e-05, + "loss": 4.1059, + "step": 2838 + }, + { + "epoch": 0.9079533870355194, + "grad_norm": 1.3460081815719604, + "learning_rate": 1.3258439784641795e-05, + "loss": 4.192, + "step": 2839 + }, + { + "epoch": 0.908273201543105, + "grad_norm": 0.8404378890991211, + "learning_rate": 1.3167205483834842e-05, + "loss": 4.0957, + "step": 2840 + }, + { + "epoch": 0.9085930160506906, + "grad_norm": 1.1446937322616577, + "learning_rate": 1.307627912587218e-05, + "loss": 4.1216, + "step": 2841 + }, + { + "epoch": 0.9089128305582762, + "grad_norm": 1.0498093366622925, + "learning_rate": 1.2985660808371955e-05, + "loss": 4.2284, + "step": 2842 + }, + { + "epoch": 0.9092326450658618, + "grad_norm": 11.120169639587402, + "learning_rate": 1.2895350628621882e-05, + "loss": 4.0898, + "step": 2843 + }, + { + "epoch": 0.9095524595734474, + "grad_norm": 1.0541397333145142, + "learning_rate": 1.2805348683578598e-05, + "loss": 4.1049, + "step": 2844 + }, + { + "epoch": 0.909872274081033, + "grad_norm": 1.1664717197418213, + "learning_rate": 1.271565506986798e-05, + "loss": 4.1095, + "step": 2845 + }, + { + "epoch": 0.9101920885886186, + "grad_norm": 1.09609055519104, + "learning_rate": 1.2626269883784834e-05, + "loss": 4.1096, + "step": 2846 + }, + { + "epoch": 0.9105119030962042, + "grad_norm": 0.7260861396789551, + "learning_rate": 1.2537193221292763e-05, + "loss": 4.1868, + "step": 2847 + }, + { + "epoch": 0.9108317176037898, + "grad_norm": 0.8366726040840149, + "learning_rate": 1.2448425178024302e-05, + "loss": 4.191, + "step": 2848 + }, + { + "epoch": 0.9111515321113755, + "grad_norm": 0.9127855896949768, + "learning_rate": 1.2359965849280518e-05, + "loss": 4.1588, + "step": 2849 + }, + { + "epoch": 0.911471346618961, + "grad_norm": 1.6013119220733643, + "learning_rate": 1.2271815330031076e-05, + "loss": 4.1997, + "step": 2850 + }, + { + "epoch": 0.9117911611265466, + "grad_norm": 1.4319251775741577, + "learning_rate": 1.218397371491414e-05, + "loss": 4.1363, + "step": 2851 + }, + { + "epoch": 0.9121109756341322, + "grad_norm": 1.0195541381835938, + "learning_rate": 1.2096441098236108e-05, + "loss": 4.1455, + "step": 2852 + }, + { + "epoch": 0.9124307901417178, + "grad_norm": 1.0409489870071411, + "learning_rate": 1.2009217573971907e-05, + "loss": 4.1772, + "step": 2853 + }, + { + "epoch": 0.9127506046493034, + "grad_norm": 23.02406883239746, + "learning_rate": 1.1922303235764363e-05, + "loss": 4.1914, + "step": 2854 + }, + { + "epoch": 0.913070419156889, + "grad_norm": 0.955950915813446, + "learning_rate": 1.1835698176924468e-05, + "loss": 4.1625, + "step": 2855 + }, + { + "epoch": 0.9133902336644746, + "grad_norm": 2.255568027496338, + "learning_rate": 1.1749402490431148e-05, + "loss": 4.1521, + "step": 2856 + }, + { + "epoch": 0.9137100481720603, + "grad_norm": 1.5526041984558105, + "learning_rate": 1.1663416268931192e-05, + "loss": 4.1377, + "step": 2857 + }, + { + "epoch": 0.9140298626796458, + "grad_norm": 3.71274471282959, + "learning_rate": 1.1577739604739155e-05, + "loss": 4.1475, + "step": 2858 + }, + { + "epoch": 0.9143496771872314, + "grad_norm": 1.166366457939148, + "learning_rate": 1.1492372589837261e-05, + "loss": 4.1684, + "step": 2859 + }, + { + "epoch": 0.914669491694817, + "grad_norm": 0.8275213241577148, + "learning_rate": 1.1407315315875365e-05, + "loss": 4.1871, + "step": 2860 + }, + { + "epoch": 0.9149893062024026, + "grad_norm": 0.7931932210922241, + "learning_rate": 1.1322567874170552e-05, + "loss": 4.1604, + "step": 2861 + }, + { + "epoch": 0.9153091207099882, + "grad_norm": 0.6553447842597961, + "learning_rate": 1.1238130355707509e-05, + "loss": 4.0968, + "step": 2862 + }, + { + "epoch": 0.9156289352175738, + "grad_norm": 0.9364184737205505, + "learning_rate": 1.1154002851138122e-05, + "loss": 4.1265, + "step": 2863 + }, + { + "epoch": 0.9159487497251594, + "grad_norm": 0.8897423148155212, + "learning_rate": 1.107018545078141e-05, + "loss": 4.0807, + "step": 2864 + }, + { + "epoch": 0.916268564232745, + "grad_norm": 0.9923008680343628, + "learning_rate": 1.0986678244623526e-05, + "loss": 4.1093, + "step": 2865 + }, + { + "epoch": 0.9165883787403306, + "grad_norm": 0.9756916761398315, + "learning_rate": 1.0903481322317486e-05, + "loss": 4.1813, + "step": 2866 + }, + { + "epoch": 0.9169081932479162, + "grad_norm": 1.5137089490890503, + "learning_rate": 1.0820594773183278e-05, + "loss": 4.1056, + "step": 2867 + }, + { + "epoch": 0.9172280077555018, + "grad_norm": 0.7398456931114197, + "learning_rate": 1.0738018686207683e-05, + "loss": 4.1533, + "step": 2868 + }, + { + "epoch": 0.9175478222630874, + "grad_norm": 2.472379446029663, + "learning_rate": 1.0655753150044155e-05, + "loss": 4.0387, + "step": 2869 + }, + { + "epoch": 0.917867636770673, + "grad_norm": 1.7408876419067383, + "learning_rate": 1.0573798253012778e-05, + "loss": 4.1514, + "step": 2870 + }, + { + "epoch": 0.9181874512782586, + "grad_norm": 2.224121570587158, + "learning_rate": 1.0492154083099968e-05, + "loss": 4.2118, + "step": 2871 + }, + { + "epoch": 0.9185072657858442, + "grad_norm": 0.714945912361145, + "learning_rate": 1.0410820727958712e-05, + "loss": 4.1865, + "step": 2872 + }, + { + "epoch": 0.9188270802934299, + "grad_norm": 0.7081162929534912, + "learning_rate": 1.0329798274908297e-05, + "loss": 4.1695, + "step": 2873 + }, + { + "epoch": 0.9191468948010154, + "grad_norm": 0.8127989768981934, + "learning_rate": 1.0249086810934204e-05, + "loss": 4.1606, + "step": 2874 + }, + { + "epoch": 0.919466709308601, + "grad_norm": 2.152107000350952, + "learning_rate": 1.0168686422687921e-05, + "loss": 4.1271, + "step": 2875 + }, + { + "epoch": 0.9197865238161866, + "grad_norm": 0.8278626203536987, + "learning_rate": 1.008859719648717e-05, + "loss": 4.101, + "step": 2876 + }, + { + "epoch": 0.9201063383237722, + "grad_norm": 2.979034662246704, + "learning_rate": 1.0008819218315434e-05, + "loss": 4.1707, + "step": 2877 + }, + { + "epoch": 0.9204261528313578, + "grad_norm": 1.1299265623092651, + "learning_rate": 9.929352573822203e-06, + "loss": 4.0792, + "step": 2878 + }, + { + "epoch": 0.9207459673389434, + "grad_norm": 0.734524667263031, + "learning_rate": 9.850197348322597e-06, + "loss": 4.1294, + "step": 2879 + }, + { + "epoch": 0.921065781846529, + "grad_norm": 0.9837729930877686, + "learning_rate": 9.771353626797373e-06, + "loss": 4.1301, + "step": 2880 + }, + { + "epoch": 0.9213855963541147, + "grad_norm": 2.679908037185669, + "learning_rate": 9.692821493892988e-06, + "loss": 4.2125, + "step": 2881 + }, + { + "epoch": 0.9217054108617002, + "grad_norm": 1.5348048210144043, + "learning_rate": 9.614601033921266e-06, + "loss": 4.0703, + "step": 2882 + }, + { + "epoch": 0.9220252253692858, + "grad_norm": 1.1841853857040405, + "learning_rate": 9.536692330859497e-06, + "loss": 4.2365, + "step": 2883 + }, + { + "epoch": 0.9223450398768714, + "grad_norm": 1.069101333618164, + "learning_rate": 9.459095468350241e-06, + "loss": 4.1356, + "step": 2884 + }, + { + "epoch": 0.922664854384457, + "grad_norm": 1.2758231163024902, + "learning_rate": 9.381810529701228e-06, + "loss": 4.1726, + "step": 2885 + }, + { + "epoch": 0.9229846688920427, + "grad_norm": 0.7392111420631409, + "learning_rate": 9.30483759788535e-06, + "loss": 4.1963, + "step": 2886 + }, + { + "epoch": 0.9233044833996282, + "grad_norm": 0.8856028318405151, + "learning_rate": 9.228176755540506e-06, + "loss": 4.1671, + "step": 2887 + }, + { + "epoch": 0.9236242979072138, + "grad_norm": 0.9773308634757996, + "learning_rate": 9.151828084969593e-06, + "loss": 4.1572, + "step": 2888 + }, + { + "epoch": 0.9239441124147995, + "grad_norm": 1.8765637874603271, + "learning_rate": 9.075791668140308e-06, + "loss": 4.1989, + "step": 2889 + }, + { + "epoch": 0.924263926922385, + "grad_norm": 1.1740829944610596, + "learning_rate": 9.000067586685089e-06, + "loss": 4.1259, + "step": 2890 + }, + { + "epoch": 0.9245837414299706, + "grad_norm": 3.2066540718078613, + "learning_rate": 8.924655921901135e-06, + "loss": 4.1556, + "step": 2891 + }, + { + "epoch": 0.9249035559375562, + "grad_norm": 0.787665069103241, + "learning_rate": 8.849556754750153e-06, + "loss": 4.1691, + "step": 2892 + }, + { + "epoch": 0.9252233704451418, + "grad_norm": 0.9632423520088196, + "learning_rate": 8.774770165858347e-06, + "loss": 4.1926, + "step": 2893 + }, + { + "epoch": 0.9255431849527275, + "grad_norm": 0.6398853063583374, + "learning_rate": 8.70029623551649e-06, + "loss": 4.1396, + "step": 2894 + }, + { + "epoch": 0.925862999460313, + "grad_norm": 0.5944526195526123, + "learning_rate": 8.626135043679495e-06, + "loss": 4.1491, + "step": 2895 + }, + { + "epoch": 0.9261828139678986, + "grad_norm": 1.4589953422546387, + "learning_rate": 8.552286669966635e-06, + "loss": 4.0778, + "step": 2896 + }, + { + "epoch": 0.9265026284754843, + "grad_norm": 0.8516110181808472, + "learning_rate": 8.47875119366126e-06, + "loss": 4.184, + "step": 2897 + }, + { + "epoch": 0.9268224429830698, + "grad_norm": 2.0680315494537354, + "learning_rate": 8.405528693710883e-06, + "loss": 4.1842, + "step": 2898 + }, + { + "epoch": 0.9271422574906554, + "grad_norm": 1.3118245601654053, + "learning_rate": 8.332619248726957e-06, + "loss": 4.2091, + "step": 2899 + }, + { + "epoch": 0.927462071998241, + "grad_norm": 1.5643266439437866, + "learning_rate": 8.260022936984833e-06, + "loss": 4.1439, + "step": 2900 + }, + { + "epoch": 0.927462071998241, + "eval_runtime": 21.1701, + "eval_samples_per_second": 89.607, + "eval_steps_per_second": 22.437, + "step": 2900 + }, + { + "epoch": 0.9277818865058266, + "grad_norm": 1.3597246408462524, + "learning_rate": 8.187739836423734e-06, + "loss": 4.1601, + "step": 2901 + }, + { + "epoch": 0.9281017010134123, + "grad_norm": 0.6507638096809387, + "learning_rate": 8.115770024646518e-06, + "loss": 4.1278, + "step": 2902 + }, + { + "epoch": 0.9284215155209978, + "grad_norm": 0.916664719581604, + "learning_rate": 8.044113578919842e-06, + "loss": 4.2382, + "step": 2903 + }, + { + "epoch": 0.9287413300285834, + "grad_norm": 3.136094093322754, + "learning_rate": 7.97277057617377e-06, + "loss": 4.1818, + "step": 2904 + }, + { + "epoch": 0.9290611445361691, + "grad_norm": 0.7802349925041199, + "learning_rate": 7.901741093002002e-06, + "loss": 4.1187, + "step": 2905 + }, + { + "epoch": 0.9293809590437546, + "grad_norm": 0.7877249717712402, + "learning_rate": 7.8310252056616e-06, + "loss": 4.1644, + "step": 2906 + }, + { + "epoch": 0.9297007735513402, + "grad_norm": 1.5993577241897583, + "learning_rate": 7.760622990072873e-06, + "loss": 4.1928, + "step": 2907 + }, + { + "epoch": 0.9300205880589258, + "grad_norm": 1.804561972618103, + "learning_rate": 7.690534521819458e-06, + "loss": 4.1438, + "step": 2908 + }, + { + "epoch": 0.9303404025665114, + "grad_norm": 0.757645308971405, + "learning_rate": 7.6207598761481305e-06, + "loss": 4.092, + "step": 2909 + }, + { + "epoch": 0.9306602170740971, + "grad_norm": 6.15857458114624, + "learning_rate": 7.5512991279687684e-06, + "loss": 4.125, + "step": 2910 + }, + { + "epoch": 0.9309800315816826, + "grad_norm": 1.0770012140274048, + "learning_rate": 7.482152351854187e-06, + "loss": 4.136, + "step": 2911 + }, + { + "epoch": 0.9312998460892682, + "grad_norm": 0.8908947110176086, + "learning_rate": 7.413319622040137e-06, + "loss": 4.0903, + "step": 2912 + }, + { + "epoch": 0.9316196605968539, + "grad_norm": 1.2202068567276, + "learning_rate": 7.344801012425306e-06, + "loss": 4.1358, + "step": 2913 + }, + { + "epoch": 0.9319394751044394, + "grad_norm": 0.7125909328460693, + "learning_rate": 7.276596596571016e-06, + "loss": 4.1526, + "step": 2914 + }, + { + "epoch": 0.932259289612025, + "grad_norm": 5.307232856750488, + "learning_rate": 7.208706447701395e-06, + "loss": 4.1704, + "step": 2915 + }, + { + "epoch": 0.9325791041196106, + "grad_norm": 1.0679246187210083, + "learning_rate": 7.141130638703041e-06, + "loss": 4.1413, + "step": 2916 + }, + { + "epoch": 0.9328989186271962, + "grad_norm": 4.103248119354248, + "learning_rate": 7.073869242125152e-06, + "loss": 4.2343, + "step": 2917 + }, + { + "epoch": 0.9332187331347819, + "grad_norm": 1.742585301399231, + "learning_rate": 7.006922330179398e-06, + "loss": 4.2028, + "step": 2918 + }, + { + "epoch": 0.9335385476423674, + "grad_norm": 0.6980456709861755, + "learning_rate": 6.940289974739754e-06, + "loss": 4.0447, + "step": 2919 + }, + { + "epoch": 0.933858362149953, + "grad_norm": 1.4174977540969849, + "learning_rate": 6.8739722473425295e-06, + "loss": 4.0596, + "step": 2920 + }, + { + "epoch": 0.9341781766575387, + "grad_norm": 1.5717507600784302, + "learning_rate": 6.807969219186271e-06, + "loss": 4.1207, + "step": 2921 + }, + { + "epoch": 0.9344979911651242, + "grad_norm": 2.3921797275543213, + "learning_rate": 6.742280961131563e-06, + "loss": 4.2424, + "step": 2922 + }, + { + "epoch": 0.9348178056727098, + "grad_norm": 1.926538109779358, + "learning_rate": 6.676907543701227e-06, + "loss": 4.1812, + "step": 2923 + }, + { + "epoch": 0.9351376201802954, + "grad_norm": 1.7509901523590088, + "learning_rate": 6.611849037079886e-06, + "loss": 4.1716, + "step": 2924 + }, + { + "epoch": 0.935457434687881, + "grad_norm": 1.3256241083145142, + "learning_rate": 6.5471055111142035e-06, + "loss": 4.1333, + "step": 2925 + }, + { + "epoch": 0.9357772491954667, + "grad_norm": 1.4543012380599976, + "learning_rate": 6.4826770353126115e-06, + "loss": 4.1293, + "step": 2926 + }, + { + "epoch": 0.9360970637030522, + "grad_norm": 0.8532224297523499, + "learning_rate": 6.418563678845379e-06, + "loss": 4.1248, + "step": 2927 + }, + { + "epoch": 0.9364168782106378, + "grad_norm": 0.6699708104133606, + "learning_rate": 6.354765510544346e-06, + "loss": 4.2041, + "step": 2928 + }, + { + "epoch": 0.9367366927182235, + "grad_norm": 1.4264156818389893, + "learning_rate": 6.291282598903091e-06, + "loss": 4.1338, + "step": 2929 + }, + { + "epoch": 0.937056507225809, + "grad_norm": 0.8606477379798889, + "learning_rate": 6.228115012076729e-06, + "loss": 4.2471, + "step": 2930 + }, + { + "epoch": 0.9373763217333946, + "grad_norm": 0.6733378171920776, + "learning_rate": 6.165262817881678e-06, + "loss": 4.1329, + "step": 2931 + }, + { + "epoch": 0.9376961362409802, + "grad_norm": 0.8740752935409546, + "learning_rate": 6.102726083795961e-06, + "loss": 4.1531, + "step": 2932 + }, + { + "epoch": 0.9380159507485658, + "grad_norm": 0.6417012810707092, + "learning_rate": 6.040504876958741e-06, + "loss": 4.1521, + "step": 2933 + }, + { + "epoch": 0.9383357652561515, + "grad_norm": 1.6858338117599487, + "learning_rate": 5.978599264170614e-06, + "loss": 4.1414, + "step": 2934 + }, + { + "epoch": 0.938655579763737, + "grad_norm": 0.9434038400650024, + "learning_rate": 5.917009311893217e-06, + "loss": 4.1067, + "step": 2935 + }, + { + "epoch": 0.9389753942713226, + "grad_norm": 1.1570518016815186, + "learning_rate": 5.855735086249358e-06, + "loss": 4.1728, + "step": 2936 + }, + { + "epoch": 0.9392952087789083, + "grad_norm": 0.9964786767959595, + "learning_rate": 5.794776653022881e-06, + "loss": 4.125, + "step": 2937 + }, + { + "epoch": 0.9396150232864938, + "grad_norm": 0.8901183009147644, + "learning_rate": 5.7341340776585035e-06, + "loss": 4.1379, + "step": 2938 + }, + { + "epoch": 0.9399348377940794, + "grad_norm": 1.3609243631362915, + "learning_rate": 5.673807425262045e-06, + "loss": 4.1647, + "step": 2939 + }, + { + "epoch": 0.940254652301665, + "grad_norm": 3.80275559425354, + "learning_rate": 5.613796760599898e-06, + "loss": 4.0865, + "step": 2940 + }, + { + "epoch": 0.9405744668092506, + "grad_norm": 0.909650981426239, + "learning_rate": 5.554102148099393e-06, + "loss": 4.2426, + "step": 2941 + }, + { + "epoch": 0.9408942813168363, + "grad_norm": 1.1343904733657837, + "learning_rate": 5.494723651848532e-06, + "loss": 4.2275, + "step": 2942 + }, + { + "epoch": 0.9412140958244218, + "grad_norm": 3.4002182483673096, + "learning_rate": 5.435661335595753e-06, + "loss": 4.1426, + "step": 2943 + }, + { + "epoch": 0.9415339103320074, + "grad_norm": 0.8503502607345581, + "learning_rate": 5.376915262750369e-06, + "loss": 4.2002, + "step": 2944 + }, + { + "epoch": 0.9418537248395931, + "grad_norm": 0.8439334034919739, + "learning_rate": 5.3184854963818305e-06, + "loss": 4.0841, + "step": 2945 + }, + { + "epoch": 0.9421735393471786, + "grad_norm": 0.6317617893218994, + "learning_rate": 5.260372099220289e-06, + "loss": 4.1769, + "step": 2946 + }, + { + "epoch": 0.9424933538547642, + "grad_norm": 0.7237291932106018, + "learning_rate": 5.202575133656039e-06, + "loss": 4.2168, + "step": 2947 + }, + { + "epoch": 0.9428131683623499, + "grad_norm": 0.9313304424285889, + "learning_rate": 5.145094661739746e-06, + "loss": 4.147, + "step": 2948 + }, + { + "epoch": 0.9431329828699354, + "grad_norm": 20.669403076171875, + "learning_rate": 5.087930745182278e-06, + "loss": 4.1734, + "step": 2949 + }, + { + "epoch": 0.9434527973775211, + "grad_norm": 0.8501777648925781, + "learning_rate": 5.031083445354644e-06, + "loss": 4.1611, + "step": 2950 + }, + { + "epoch": 0.9437726118851066, + "grad_norm": 0.6755287051200867, + "learning_rate": 4.9745528232879915e-06, + "loss": 4.1229, + "step": 2951 + }, + { + "epoch": 0.9440924263926922, + "grad_norm": 1.415075659751892, + "learning_rate": 4.918338939673372e-06, + "loss": 4.2392, + "step": 2952 + }, + { + "epoch": 0.9444122409002779, + "grad_norm": 3.2531232833862305, + "learning_rate": 4.862441854861809e-06, + "loss": 4.1843, + "step": 2953 + }, + { + "epoch": 0.9447320554078634, + "grad_norm": 0.8109667897224426, + "learning_rate": 4.806861628864333e-06, + "loss": 4.2085, + "step": 2954 + }, + { + "epoch": 0.9450518699154491, + "grad_norm": 0.5986691117286682, + "learning_rate": 4.751598321351679e-06, + "loss": 4.1806, + "step": 2955 + }, + { + "epoch": 0.9453716844230347, + "grad_norm": 0.8207346796989441, + "learning_rate": 4.6966519916543875e-06, + "loss": 4.1153, + "step": 2956 + }, + { + "epoch": 0.9456914989306202, + "grad_norm": 0.7765918970108032, + "learning_rate": 4.642022698762638e-06, + "loss": 4.0931, + "step": 2957 + }, + { + "epoch": 0.9460113134382059, + "grad_norm": 0.9279441237449646, + "learning_rate": 4.5877105013262805e-06, + "loss": 4.0834, + "step": 2958 + }, + { + "epoch": 0.9463311279457914, + "grad_norm": 7.6889119148254395, + "learning_rate": 4.533715457654741e-06, + "loss": 4.1973, + "step": 2959 + }, + { + "epoch": 0.946650942453377, + "grad_norm": 5.713785171508789, + "learning_rate": 4.480037625716981e-06, + "loss": 4.1071, + "step": 2960 + }, + { + "epoch": 0.9469707569609627, + "grad_norm": 3.001783847808838, + "learning_rate": 4.4266770631413374e-06, + "loss": 4.2032, + "step": 2961 + }, + { + "epoch": 0.9472905714685482, + "grad_norm": 1.2258671522140503, + "learning_rate": 4.373633827215517e-06, + "loss": 4.2993, + "step": 2962 + }, + { + "epoch": 0.9476103859761339, + "grad_norm": 1.2415916919708252, + "learning_rate": 4.3209079748866e-06, + "loss": 4.0942, + "step": 2963 + }, + { + "epoch": 0.9479302004837195, + "grad_norm": 0.7851530909538269, + "learning_rate": 4.268499562760907e-06, + "loss": 4.1672, + "step": 2964 + }, + { + "epoch": 0.948250014991305, + "grad_norm": 3.1025571823120117, + "learning_rate": 4.216408647103997e-06, + "loss": 4.1698, + "step": 2965 + }, + { + "epoch": 0.9485698294988907, + "grad_norm": 1.134128212928772, + "learning_rate": 4.164635283840468e-06, + "loss": 4.1732, + "step": 2966 + }, + { + "epoch": 0.9488896440064762, + "grad_norm": 0.7273879051208496, + "learning_rate": 4.113179528554089e-06, + "loss": 4.1805, + "step": 2967 + }, + { + "epoch": 0.9492094585140618, + "grad_norm": 2.2934703826904297, + "learning_rate": 4.062041436487573e-06, + "loss": 4.1697, + "step": 2968 + }, + { + "epoch": 0.9495292730216475, + "grad_norm": 0.7853653430938721, + "learning_rate": 4.011221062542636e-06, + "loss": 4.0711, + "step": 2969 + }, + { + "epoch": 0.949849087529233, + "grad_norm": 0.7500988841056824, + "learning_rate": 3.9607184612799325e-06, + "loss": 4.178, + "step": 2970 + }, + { + "epoch": 0.9501689020368187, + "grad_norm": 0.7080156207084656, + "learning_rate": 3.910533686918826e-06, + "loss": 4.1784, + "step": 2971 + }, + { + "epoch": 0.9504887165444043, + "grad_norm": 1.8000987768173218, + "learning_rate": 3.860666793337585e-06, + "loss": 4.1274, + "step": 2972 + }, + { + "epoch": 0.9508085310519898, + "grad_norm": 0.6100180149078369, + "learning_rate": 3.811117834073152e-06, + "loss": 4.1278, + "step": 2973 + }, + { + "epoch": 0.9511283455595755, + "grad_norm": 0.8149864673614502, + "learning_rate": 3.761886862321173e-06, + "loss": 4.1442, + "step": 2974 + }, + { + "epoch": 0.951448160067161, + "grad_norm": 1.2226074934005737, + "learning_rate": 3.7129739309358362e-06, + "loss": 4.1006, + "step": 2975 + }, + { + "epoch": 0.9517679745747466, + "grad_norm": 0.8216033577919006, + "learning_rate": 3.664379092429903e-06, + "loss": 4.1184, + "step": 2976 + }, + { + "epoch": 0.9520877890823323, + "grad_norm": 1.7109731435775757, + "learning_rate": 3.6161023989747075e-06, + "loss": 4.2134, + "step": 2977 + }, + { + "epoch": 0.9524076035899178, + "grad_norm": 1.8109666109085083, + "learning_rate": 3.5681439023999224e-06, + "loss": 4.1576, + "step": 2978 + }, + { + "epoch": 0.9527274180975035, + "grad_norm": 0.8563116192817688, + "learning_rate": 3.5205036541936626e-06, + "loss": 4.1564, + "step": 2979 + }, + { + "epoch": 0.9530472326050891, + "grad_norm": 2.034409284591675, + "learning_rate": 3.4731817055023812e-06, + "loss": 4.1303, + "step": 2980 + }, + { + "epoch": 0.9533670471126746, + "grad_norm": 0.8260316848754883, + "learning_rate": 3.4261781071307393e-06, + "loss": 4.1298, + "step": 2981 + }, + { + "epoch": 0.9536868616202603, + "grad_norm": 0.7619715332984924, + "learning_rate": 3.3794929095417034e-06, + "loss": 4.1778, + "step": 2982 + }, + { + "epoch": 0.9540066761278458, + "grad_norm": 2.5158939361572266, + "learning_rate": 3.3331261628563145e-06, + "loss": 4.082, + "step": 2983 + }, + { + "epoch": 0.9543264906354314, + "grad_norm": 0.9901076555252075, + "learning_rate": 3.2870779168538196e-06, + "loss": 4.1302, + "step": 2984 + }, + { + "epoch": 0.9546463051430171, + "grad_norm": 3.47987699508667, + "learning_rate": 3.2413482209714737e-06, + "loss": 4.0875, + "step": 2985 + }, + { + "epoch": 0.9549661196506026, + "grad_norm": 1.6151037216186523, + "learning_rate": 3.195937124304504e-06, + "loss": 4.212, + "step": 2986 + }, + { + "epoch": 0.9552859341581883, + "grad_norm": 1.4174071550369263, + "learning_rate": 3.150844675606212e-06, + "loss": 4.0668, + "step": 2987 + }, + { + "epoch": 0.9556057486657739, + "grad_norm": 1.9378107786178589, + "learning_rate": 3.10607092328764e-06, + "loss": 4.1585, + "step": 2988 + }, + { + "epoch": 0.9559255631733594, + "grad_norm": 1.7095003128051758, + "learning_rate": 3.0616159154177366e-06, + "loss": 4.134, + "step": 2989 + }, + { + "epoch": 0.9562453776809451, + "grad_norm": 8.621370315551758, + "learning_rate": 3.0174796997233908e-06, + "loss": 4.1905, + "step": 2990 + }, + { + "epoch": 0.9565651921885306, + "grad_norm": 0.6244714260101318, + "learning_rate": 2.973662323588999e-06, + "loss": 4.1201, + "step": 2991 + }, + { + "epoch": 0.9568850066961162, + "grad_norm": 0.9261051416397095, + "learning_rate": 2.930163834056831e-06, + "loss": 4.2435, + "step": 2992 + }, + { + "epoch": 0.9572048212037019, + "grad_norm": 1.0151216983795166, + "learning_rate": 2.8869842778266983e-06, + "loss": 4.1934, + "step": 2993 + }, + { + "epoch": 0.9575246357112874, + "grad_norm": 1.3376774787902832, + "learning_rate": 2.844123701256051e-06, + "loss": 4.2021, + "step": 2994 + }, + { + "epoch": 0.9578444502188731, + "grad_norm": 0.9204723238945007, + "learning_rate": 2.801582150359882e-06, + "loss": 4.0851, + "step": 2995 + }, + { + "epoch": 0.9581642647264587, + "grad_norm": 1.2936477661132812, + "learning_rate": 2.7593596708106904e-06, + "loss": 4.0481, + "step": 2996 + }, + { + "epoch": 0.9584840792340442, + "grad_norm": 0.796598494052887, + "learning_rate": 2.717456307938415e-06, + "loss": 4.1471, + "step": 2997 + }, + { + "epoch": 0.9588038937416299, + "grad_norm": 0.8803315758705139, + "learning_rate": 2.6758721067303367e-06, + "loss": 4.2105, + "step": 2998 + }, + { + "epoch": 0.9591237082492154, + "grad_norm": 2.119453191757202, + "learning_rate": 2.634607111831177e-06, + "loss": 4.1448, + "step": 2999 + }, + { + "epoch": 0.959443522756801, + "grad_norm": 0.9366162419319153, + "learning_rate": 2.5936613675428985e-06, + "loss": 4.0965, + "step": 3000 + }, + { + "epoch": 0.959443522756801, + "eval_runtime": 22.1777, + "eval_samples_per_second": 85.536, + "eval_steps_per_second": 21.418, + "step": 3000 + }, + { + "epoch": 0.9597633372643867, + "grad_norm": 0.9046416282653809, + "learning_rate": 2.5530349178247033e-06, + "loss": 4.1669, + "step": 3001 + }, + { + "epoch": 0.9600831517719722, + "grad_norm": 0.6622817516326904, + "learning_rate": 2.512727806293069e-06, + "loss": 4.1817, + "step": 3002 + }, + { + "epoch": 0.9604029662795579, + "grad_norm": 1.492269515991211, + "learning_rate": 2.4727400762215798e-06, + "loss": 4.0929, + "step": 3003 + }, + { + "epoch": 0.9607227807871435, + "grad_norm": 0.9400051236152649, + "learning_rate": 2.4330717705409287e-06, + "loss": 4.0636, + "step": 3004 + }, + { + "epoch": 0.961042595294729, + "grad_norm": 0.8403382897377014, + "learning_rate": 2.393722931838882e-06, + "loss": 4.1766, + "step": 3005 + }, + { + "epoch": 0.9613624098023147, + "grad_norm": 0.7138628959655762, + "learning_rate": 2.3546936023603134e-06, + "loss": 4.2131, + "step": 3006 + }, + { + "epoch": 0.9616822243099002, + "grad_norm": 5.644319534301758, + "learning_rate": 2.315983824006906e-06, + "loss": 4.195, + "step": 3007 + }, + { + "epoch": 0.9620020388174858, + "grad_norm": 0.7298849821090698, + "learning_rate": 2.277593638337416e-06, + "loss": 4.1334, + "step": 3008 + }, + { + "epoch": 0.9623218533250715, + "grad_norm": 3.01814603805542, + "learning_rate": 2.2395230865674075e-06, + "loss": 4.1278, + "step": 3009 + }, + { + "epoch": 0.962641667832657, + "grad_norm": 1.0043747425079346, + "learning_rate": 2.201772209569319e-06, + "loss": 4.1642, + "step": 3010 + }, + { + "epoch": 0.9629614823402427, + "grad_norm": 2.418339490890503, + "learning_rate": 2.164341047872398e-06, + "loss": 4.2048, + "step": 3011 + }, + { + "epoch": 0.9632812968478283, + "grad_norm": 2.0579919815063477, + "learning_rate": 2.127229641662598e-06, + "loss": 4.132, + "step": 3012 + }, + { + "epoch": 0.9636011113554138, + "grad_norm": 0.6686593890190125, + "learning_rate": 2.0904380307826483e-06, + "loss": 4.1404, + "step": 3013 + }, + { + "epoch": 0.9639209258629995, + "grad_norm": 1.4202622175216675, + "learning_rate": 2.053966254731887e-06, + "loss": 4.1232, + "step": 3014 + }, + { + "epoch": 0.964240740370585, + "grad_norm": 0.9033358693122864, + "learning_rate": 2.0178143526663248e-06, + "loss": 4.2485, + "step": 3015 + }, + { + "epoch": 0.9645605548781706, + "grad_norm": 0.8088997006416321, + "learning_rate": 1.981982363398549e-06, + "loss": 4.1986, + "step": 3016 + }, + { + "epoch": 0.9648803693857563, + "grad_norm": 1.034389853477478, + "learning_rate": 1.9464703253976533e-06, + "loss": 4.0827, + "step": 3017 + }, + { + "epoch": 0.9652001838933418, + "grad_norm": 0.7637101411819458, + "learning_rate": 1.911278276789241e-06, + "loss": 4.1197, + "step": 3018 + }, + { + "epoch": 0.9655199984009275, + "grad_norm": 2.229572057723999, + "learning_rate": 1.8764062553554227e-06, + "loss": 4.0562, + "step": 3019 + }, + { + "epoch": 0.9658398129085131, + "grad_norm": 0.6626889109611511, + "learning_rate": 1.8418542985347174e-06, + "loss": 4.2373, + "step": 3020 + }, + { + "epoch": 0.9661596274160986, + "grad_norm": 0.6541303992271423, + "learning_rate": 1.8076224434219523e-06, + "loss": 4.1088, + "step": 3021 + }, + { + "epoch": 0.9664794419236843, + "grad_norm": 1.221030354499817, + "learning_rate": 1.773710726768396e-06, + "loss": 4.1433, + "step": 3022 + }, + { + "epoch": 0.9667992564312698, + "grad_norm": 1.8171106576919556, + "learning_rate": 1.7401191849815255e-06, + "loss": 4.1747, + "step": 3023 + }, + { + "epoch": 0.9671190709388555, + "grad_norm": 2.3712589740753174, + "learning_rate": 1.7068478541251263e-06, + "loss": 4.1589, + "step": 3024 + }, + { + "epoch": 0.9674388854464411, + "grad_norm": 1.5026735067367554, + "learning_rate": 1.673896769919192e-06, + "loss": 4.1045, + "step": 3025 + }, + { + "epoch": 0.9677586999540266, + "grad_norm": 0.5768782496452332, + "learning_rate": 1.6412659677399908e-06, + "loss": 4.1551, + "step": 3026 + }, + { + "epoch": 0.9680785144616123, + "grad_norm": 0.8237771987915039, + "learning_rate": 1.608955482619767e-06, + "loss": 4.1686, + "step": 3027 + }, + { + "epoch": 0.9683983289691979, + "grad_norm": 3.6280362606048584, + "learning_rate": 1.5769653492470057e-06, + "loss": 4.1417, + "step": 3028 + }, + { + "epoch": 0.9687181434767834, + "grad_norm": 1.3089876174926758, + "learning_rate": 1.5452956019661678e-06, + "loss": 4.1256, + "step": 3029 + }, + { + "epoch": 0.9690379579843691, + "grad_norm": 1.0755126476287842, + "learning_rate": 1.5139462747778885e-06, + "loss": 4.1386, + "step": 3030 + }, + { + "epoch": 0.9693577724919547, + "grad_norm": 0.6938768029212952, + "learning_rate": 1.4829174013386126e-06, + "loss": 4.1207, + "step": 3031 + }, + { + "epoch": 0.9696775869995403, + "grad_norm": 0.9844991564750671, + "learning_rate": 1.4522090149609256e-06, + "loss": 4.1421, + "step": 3032 + }, + { + "epoch": 0.9699974015071259, + "grad_norm": 0.711318850517273, + "learning_rate": 1.4218211486132558e-06, + "loss": 4.1968, + "step": 3033 + }, + { + "epoch": 0.9703172160147114, + "grad_norm": 1.010939598083496, + "learning_rate": 1.3917538349198731e-06, + "loss": 4.1652, + "step": 3034 + }, + { + "epoch": 0.9706370305222971, + "grad_norm": 0.9667810201644897, + "learning_rate": 1.3620071061609894e-06, + "loss": 4.1123, + "step": 3035 + }, + { + "epoch": 0.9709568450298827, + "grad_norm": 2.0549793243408203, + "learning_rate": 1.332580994272625e-06, + "loss": 4.0604, + "step": 3036 + }, + { + "epoch": 0.9712766595374682, + "grad_norm": 1.7428183555603027, + "learning_rate": 1.3034755308465428e-06, + "loss": 4.2364, + "step": 3037 + }, + { + "epoch": 0.9715964740450539, + "grad_norm": 0.8890702128410339, + "learning_rate": 1.2746907471302803e-06, + "loss": 4.1222, + "step": 3038 + }, + { + "epoch": 0.9719162885526395, + "grad_norm": 1.2009294033050537, + "learning_rate": 1.2462266740270843e-06, + "loss": 4.1332, + "step": 3039 + }, + { + "epoch": 0.9722361030602251, + "grad_norm": 0.9344691038131714, + "learning_rate": 1.2180833420959436e-06, + "loss": 4.2131, + "step": 3040 + }, + { + "epoch": 0.9725559175678107, + "grad_norm": 8.247675895690918, + "learning_rate": 1.190260781551422e-06, + "loss": 4.1067, + "step": 3041 + }, + { + "epoch": 0.9728757320753962, + "grad_norm": 1.7461212873458862, + "learning_rate": 1.1627590222637594e-06, + "loss": 4.1469, + "step": 3042 + }, + { + "epoch": 0.9731955465829819, + "grad_norm": 0.775938093662262, + "learning_rate": 1.1355780937587378e-06, + "loss": 4.1465, + "step": 3043 + }, + { + "epoch": 0.9735153610905675, + "grad_norm": 0.7440497279167175, + "learning_rate": 1.1087180252177475e-06, + "loss": 4.1029, + "step": 3044 + }, + { + "epoch": 0.973835175598153, + "grad_norm": 0.9629155397415161, + "learning_rate": 1.0821788454776548e-06, + "loss": 4.2221, + "step": 3045 + }, + { + "epoch": 0.9741549901057387, + "grad_norm": 1.415415644645691, + "learning_rate": 1.0559605830308682e-06, + "loss": 4.1579, + "step": 3046 + }, + { + "epoch": 0.9744748046133243, + "grad_norm": 0.8382353782653809, + "learning_rate": 1.030063266025205e-06, + "loss": 4.1315, + "step": 3047 + }, + { + "epoch": 0.9747946191209099, + "grad_norm": 0.6998452544212341, + "learning_rate": 1.0044869222639917e-06, + "loss": 4.2049, + "step": 3048 + }, + { + "epoch": 0.9751144336284955, + "grad_norm": 1.6525304317474365, + "learning_rate": 9.79231579205897e-07, + "loss": 4.1292, + "step": 3049 + }, + { + "epoch": 0.975434248136081, + "grad_norm": 0.940997838973999, + "learning_rate": 9.54297263964965e-07, + "loss": 4.0344, + "step": 3050 + }, + { + "epoch": 0.9757540626436667, + "grad_norm": 1.1762572526931763, + "learning_rate": 9.29684003310649e-07, + "loss": 4.2244, + "step": 3051 + }, + { + "epoch": 0.9760738771512523, + "grad_norm": 0.7123349905014038, + "learning_rate": 9.053918236676116e-07, + "loss": 4.2288, + "step": 3052 + }, + { + "epoch": 0.9763936916588378, + "grad_norm": 1.1401041746139526, + "learning_rate": 8.814207511159243e-07, + "loss": 4.2165, + "step": 3053 + }, + { + "epoch": 0.9767135061664235, + "grad_norm": 0.6380232572555542, + "learning_rate": 8.577708113908011e-07, + "loss": 4.247, + "step": 3054 + }, + { + "epoch": 0.977033320674009, + "grad_norm": 0.9629085659980774, + "learning_rate": 8.344420298827981e-07, + "loss": 4.1811, + "step": 3055 + }, + { + "epoch": 0.9773531351815947, + "grad_norm": 0.8233169913291931, + "learning_rate": 8.114344316376143e-07, + "loss": 4.2324, + "step": 3056 + }, + { + "epoch": 0.9776729496891803, + "grad_norm": 1.0156099796295166, + "learning_rate": 7.887480413561243e-07, + "loss": 4.181, + "step": 3057 + }, + { + "epoch": 0.9779927641967658, + "grad_norm": 1.231768250465393, + "learning_rate": 7.663828833943786e-07, + "loss": 4.1446, + "step": 3058 + }, + { + "epoch": 0.9783125787043515, + "grad_norm": 0.7119415402412415, + "learning_rate": 7.443389817635371e-07, + "loss": 4.1661, + "step": 3059 + }, + { + "epoch": 0.9786323932119371, + "grad_norm": 0.828500509262085, + "learning_rate": 7.226163601298685e-07, + "loss": 4.1525, + "step": 3060 + }, + { + "epoch": 0.9789522077195226, + "grad_norm": 1.2262293100357056, + "learning_rate": 7.01215041814751e-07, + "loss": 4.2199, + "step": 3061 + }, + { + "epoch": 0.9792720222271083, + "grad_norm": 0.9760488867759705, + "learning_rate": 6.801350497945391e-07, + "loss": 4.1662, + "step": 3062 + }, + { + "epoch": 0.9795918367346939, + "grad_norm": 2.3235695362091064, + "learning_rate": 6.593764067006624e-07, + "loss": 4.0375, + "step": 3063 + }, + { + "epoch": 0.9799116512422795, + "grad_norm": 1.4631023406982422, + "learning_rate": 6.389391348195272e-07, + "loss": 4.1815, + "step": 3064 + }, + { + "epoch": 0.9802314657498651, + "grad_norm": 0.776385486125946, + "learning_rate": 6.188232560925155e-07, + "loss": 4.0884, + "step": 3065 + }, + { + "epoch": 0.9805512802574506, + "grad_norm": 1.8445664644241333, + "learning_rate": 5.990287921160186e-07, + "loss": 4.1904, + "step": 3066 + }, + { + "epoch": 0.9808710947650363, + "grad_norm": 1.4369531869888306, + "learning_rate": 5.79555764141304e-07, + "loss": 4.1334, + "step": 3067 + }, + { + "epoch": 0.9811909092726219, + "grad_norm": 1.0996965169906616, + "learning_rate": 5.604041930745485e-07, + "loss": 4.1985, + "step": 3068 + }, + { + "epoch": 0.9815107237802074, + "grad_norm": 0.6361686587333679, + "learning_rate": 5.415740994768048e-07, + "loss": 4.2496, + "step": 3069 + }, + { + "epoch": 0.9818305382877931, + "grad_norm": 1.0933880805969238, + "learning_rate": 5.230655035640352e-07, + "loss": 4.1126, + "step": 3070 + }, + { + "epoch": 0.9821503527953787, + "grad_norm": 1.8671218156814575, + "learning_rate": 5.048784252069782e-07, + "loss": 4.1701, + "step": 3071 + }, + { + "epoch": 0.9824701673029643, + "grad_norm": 0.5897703766822815, + "learning_rate": 4.870128839312815e-07, + "loss": 4.1879, + "step": 3072 + }, + { + "epoch": 0.9827899818105499, + "grad_norm": 1.1557186841964722, + "learning_rate": 4.6946889891726903e-07, + "loss": 4.118, + "step": 3073 + }, + { + "epoch": 0.9831097963181354, + "grad_norm": 1.0101734399795532, + "learning_rate": 4.5224648900017424e-07, + "loss": 4.1539, + "step": 3074 + }, + { + "epoch": 0.9834296108257211, + "grad_norm": 1.1650283336639404, + "learning_rate": 4.353456726699067e-07, + "loss": 4.1303, + "step": 3075 + }, + { + "epoch": 0.9837494253333067, + "grad_norm": 0.716679036617279, + "learning_rate": 4.1876646807111893e-07, + "loss": 4.0996, + "step": 3076 + }, + { + "epoch": 0.9840692398408922, + "grad_norm": 1.0437771081924438, + "learning_rate": 4.025088930031728e-07, + "loss": 4.0889, + "step": 3077 + }, + { + "epoch": 0.9843890543484779, + "grad_norm": 0.9004671573638916, + "learning_rate": 3.8657296492023984e-07, + "loss": 4.0952, + "step": 3078 + }, + { + "epoch": 0.9847088688560635, + "grad_norm": 2.723461151123047, + "learning_rate": 3.709587009309678e-07, + "loss": 4.0832, + "step": 3079 + }, + { + "epoch": 0.9850286833636491, + "grad_norm": 31.49422264099121, + "learning_rate": 3.5566611779888066e-07, + "loss": 4.0763, + "step": 3080 + }, + { + "epoch": 0.9853484978712347, + "grad_norm": 0.9104720950126648, + "learning_rate": 3.406952319420453e-07, + "loss": 3.9452, + "step": 3081 + }, + { + "epoch": 0.9856683123788202, + "grad_norm": 0.9721198678016663, + "learning_rate": 3.260460594330716e-07, + "loss": 4.1809, + "step": 3082 + }, + { + "epoch": 0.9859881268864059, + "grad_norm": 1.181180715560913, + "learning_rate": 3.1171861599937896e-07, + "loss": 4.2267, + "step": 3083 + }, + { + "epoch": 0.9863079413939915, + "grad_norm": 1.084585428237915, + "learning_rate": 2.9771291702279655e-07, + "loss": 4.2163, + "step": 3084 + }, + { + "epoch": 0.986627755901577, + "grad_norm": 1.203477382659912, + "learning_rate": 2.840289775398297e-07, + "loss": 4.1947, + "step": 3085 + }, + { + "epoch": 0.9869475704091627, + "grad_norm": 2.1086409091949463, + "learning_rate": 2.7066681224149344e-07, + "loss": 4.1699, + "step": 3086 + }, + { + "epoch": 0.9872673849167483, + "grad_norm": 0.9736827611923218, + "learning_rate": 2.5762643547337924e-07, + "loss": 4.1535, + "step": 3087 + }, + { + "epoch": 0.9875871994243339, + "grad_norm": 0.9415672421455383, + "learning_rate": 2.4490786123562144e-07, + "loss": 4.1539, + "step": 3088 + }, + { + "epoch": 0.9879070139319195, + "grad_norm": 7.882657051086426, + "learning_rate": 2.3251110318283083e-07, + "loss": 4.0907, + "step": 3089 + }, + { + "epoch": 0.988226828439505, + "grad_norm": 2.238258123397827, + "learning_rate": 2.204361746241279e-07, + "loss": 4.1646, + "step": 3090 + }, + { + "epoch": 0.9885466429470907, + "grad_norm": 0.7429603934288025, + "learning_rate": 2.0868308852310943e-07, + "loss": 4.1225, + "step": 3091 + }, + { + "epoch": 0.9888664574546763, + "grad_norm": 1.7196696996688843, + "learning_rate": 1.9725185749784879e-07, + "loss": 4.1234, + "step": 3092 + }, + { + "epoch": 0.9891862719622619, + "grad_norm": 1.3329929113388062, + "learning_rate": 1.861424938208955e-07, + "loss": 4.0764, + "step": 3093 + }, + { + "epoch": 0.9895060864698475, + "grad_norm": 4.000940322875977, + "learning_rate": 1.753550094192424e-07, + "loss": 4.163, + "step": 3094 + }, + { + "epoch": 0.9898259009774331, + "grad_norm": 1.3527119159698486, + "learning_rate": 1.6488941587429193e-07, + "loss": 4.1331, + "step": 3095 + }, + { + "epoch": 0.9901457154850187, + "grad_norm": 1.9178422689437866, + "learning_rate": 1.547457244218564e-07, + "loss": 4.1183, + "step": 3096 + }, + { + "epoch": 0.9904655299926043, + "grad_norm": 0.7177301049232483, + "learning_rate": 1.4492394595219115e-07, + "loss": 4.0857, + "step": 3097 + }, + { + "epoch": 0.9907853445001898, + "grad_norm": 0.7928923964500427, + "learning_rate": 1.3542409100992802e-07, + "loss": 4.1367, + "step": 3098 + }, + { + "epoch": 0.9911051590077755, + "grad_norm": 2.0621325969696045, + "learning_rate": 1.2624616979407532e-07, + "loss": 4.1868, + "step": 3099 + }, + { + "epoch": 0.9914249735153611, + "grad_norm": 3.547945261001587, + "learning_rate": 1.1739019215801781e-07, + "loss": 4.2042, + "step": 3100 + }, + { + "epoch": 0.9914249735153611, + "eval_runtime": 21.1669, + "eval_samples_per_second": 89.621, + "eval_steps_per_second": 22.441, + "step": 3100 + }, + { + "epoch": 0.9917447880229467, + "grad_norm": 2.798473358154297, + "learning_rate": 1.0885616760951676e-07, + "loss": 4.087, + "step": 3101 + }, + { + "epoch": 0.9920646025305323, + "grad_norm": 2.1363327503204346, + "learning_rate": 1.0064410531067657e-07, + "loss": 4.1217, + "step": 3102 + }, + { + "epoch": 0.9923844170381179, + "grad_norm": 0.9180671572685242, + "learning_rate": 9.27540140779448e-08, + "loss": 4.2427, + "step": 3103 + }, + { + "epoch": 0.9927042315457035, + "grad_norm": 0.9050948619842529, + "learning_rate": 8.51859023821122e-08, + "loss": 4.1748, + "step": 3104 + }, + { + "epoch": 0.9930240460532891, + "grad_norm": 43.1258544921875, + "learning_rate": 7.793977834824605e-08, + "loss": 4.1679, + "step": 3105 + }, + { + "epoch": 0.9933438605608746, + "grad_norm": 0.9948246479034424, + "learning_rate": 7.101564975579011e-08, + "loss": 4.1565, + "step": 3106 + }, + { + "epoch": 0.9936636750684603, + "grad_norm": 1.709020733833313, + "learning_rate": 6.441352403849798e-08, + "loss": 4.1669, + "step": 3107 + }, + { + "epoch": 0.9939834895760459, + "grad_norm": 0.569662868976593, + "learning_rate": 5.813340828429991e-08, + "loss": 4.1436, + "step": 3108 + }, + { + "epoch": 0.9943033040836315, + "grad_norm": 5.566103458404541, + "learning_rate": 5.217530923560254e-08, + "loss": 4.1018, + "step": 3109 + }, + { + "epoch": 0.9946231185912171, + "grad_norm": 1.3223363161087036, + "learning_rate": 4.6539233288955816e-08, + "loss": 4.1737, + "step": 3110 + }, + { + "epoch": 0.9949429330988027, + "grad_norm": 0.9689614176750183, + "learning_rate": 4.122518649525286e-08, + "loss": 4.1415, + "step": 3111 + }, + { + "epoch": 0.9952627476063883, + "grad_norm": 1.1901214122772217, + "learning_rate": 3.623317455959673e-08, + "loss": 4.1675, + "step": 3112 + }, + { + "epoch": 0.9955825621139739, + "grad_norm": 0.898078203201294, + "learning_rate": 3.156320284146696e-08, + "loss": 4.0298, + "step": 3113 + }, + { + "epoch": 0.9959023766215594, + "grad_norm": 0.7803918719291687, + "learning_rate": 2.7215276354486393e-08, + "loss": 4.1385, + "step": 3114 + }, + { + "epoch": 0.9962221911291451, + "grad_norm": 3.619258165359497, + "learning_rate": 2.3189399766587735e-08, + "loss": 4.1092, + "step": 3115 + }, + { + "epoch": 0.9965420056367307, + "grad_norm": 0.6324862241744995, + "learning_rate": 1.948557739994694e-08, + "loss": 4.2265, + "step": 3116 + }, + { + "epoch": 0.9968618201443163, + "grad_norm": 1.4645570516586304, + "learning_rate": 1.6103813230949892e-08, + "loss": 4.1126, + "step": 3117 + }, + { + "epoch": 0.9971816346519019, + "grad_norm": 1.1848136186599731, + "learning_rate": 1.3044110890292336e-08, + "loss": 4.1932, + "step": 3118 + }, + { + "epoch": 0.9975014491594875, + "grad_norm": 1.3195313215255737, + "learning_rate": 1.0306473662813341e-08, + "loss": 4.1381, + "step": 3119 + }, + { + "epoch": 0.9978212636670731, + "grad_norm": 0.6921530961990356, + "learning_rate": 7.89090448766183e-09, + "loss": 4.1312, + "step": 3120 + }, + { + "epoch": 0.9981410781746587, + "grad_norm": 5.658547401428223, + "learning_rate": 5.7974059581633595e-09, + "loss": 4.1864, + "step": 3121 + }, + { + "epoch": 0.9984608926822442, + "grad_norm": 1.4388262033462524, + "learning_rate": 4.02598032192003e-09, + "loss": 4.1867, + "step": 3122 + }, + { + "epoch": 0.9987807071898299, + "grad_norm": 2.132417917251587, + "learning_rate": 2.5766294807438858e-09, + "loss": 4.076, + "step": 3123 + }, + { + "epoch": 0.9991005216974155, + "grad_norm": 1.1271979808807373, + "learning_rate": 1.4493549905902902e-09, + "loss": 4.1576, + "step": 3124 + }, + { + "epoch": 0.9994203362050011, + "grad_norm": 1.7157942056655884, + "learning_rate": 6.441580617577713e-10, + "loss": 4.1391, + "step": 3125 + }, + { + "epoch": 0.9997401507125867, + "grad_norm": 0.7376135587692261, + "learning_rate": 1.6103955865487407e-10, + "loss": 4.1225, + "step": 3126 + }, + { + "epoch": 0.9997401507125867, + "step": 3126, + "total_flos": 2.6228062174681498e+17, + "train_loss": 4.706350558397484, + "train_runtime": 27646.8629, + "train_samples_per_second": 7.238, + "train_steps_per_second": 0.113 + }, + { + "epoch": 0.9997401507125867, + "eval_runtime": 22.6361, + "eval_samples_per_second": 83.804, + "eval_steps_per_second": 20.984, + "step": 3126 + } + ], + "logging_steps": 1, + "max_steps": 3126, + "num_input_tokens_seen": 0, + "num_train_epochs": 1, + "save_steps": 2000, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 2.6228062174681498e+17, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +}