diff --git "a/checkpoint-2000/trainer_state.json" "b/checkpoint-2000/trainer_state.json" new file mode 100644--- /dev/null +++ "b/checkpoint-2000/trainer_state.json" @@ -0,0 +1,8146 @@ +{ + "best_global_step": 2000, + "best_metric": 5.208409309387207, + "best_model_checkpoint": "/gpfs/scratch/guoh/DNAFM/output/gencode_human_12.8k_12800/Gencode-BPE/checkpoint-2000", + "epoch": 0.5673758865248227, + "eval_steps": 125, + "global_step": 2000, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0005673758865248227, + "grad_norm": 1275.0146484375, + "loss": 281.4781, + "lr": 2e-06, + "step": 2, + "tokens_trained": 0.000192256 + }, + { + "epoch": 0.0011347517730496454, + "grad_norm": 1437.579833984375, + "loss": 267.2211, + "lr": 6e-06, + "step": 4, + "tokens_trained": 0.000382024 + }, + { + "epoch": 0.001702127659574468, + "grad_norm": 1719.271484375, + "loss": 219.3822, + "lr": 1e-05, + "step": 6, + "tokens_trained": 0.00057072 + }, + { + "epoch": 0.0022695035460992908, + "grad_norm": 1444.94970703125, + "loss": 133.8172, + "lr": 1.4e-05, + "step": 8, + "tokens_trained": 0.000761336 + }, + { + "epoch": 0.0028368794326241137, + "grad_norm": 238.9689178466797, + "loss": 90.8177, + "lr": 1.8e-05, + "step": 10, + "tokens_trained": 0.000953248 + }, + { + "epoch": 0.003404255319148936, + "grad_norm": 158.53497314453125, + "loss": 84.6922, + "lr": 2.2e-05, + "step": 12, + "tokens_trained": 0.00114424 + }, + { + "epoch": 0.003971631205673759, + "grad_norm": 146.10595703125, + "loss": 76.7055, + "lr": 2.6e-05, + "step": 14, + "tokens_trained": 0.001334104 + }, + { + "epoch": 0.0045390070921985815, + "grad_norm": 140.69964599609375, + "loss": 67.9952, + "lr": 3e-05, + "step": 16, + "tokens_trained": 0.00152392 + }, + { + "epoch": 0.005106382978723404, + "grad_norm": 108.80303192138672, + "loss": 57.8088, + "lr": 3.4000000000000007e-05, + "step": 18, + "tokens_trained": 0.001713872 + }, + { + "epoch": 0.005673758865248227, + "grad_norm": 106.82334899902344, + "loss": 48.6585, + "lr": 3.8e-05, + "step": 20, + "tokens_trained": 0.001903976 + }, + { + "epoch": 0.00624113475177305, + "grad_norm": 93.58769989013672, + "loss": 41.7984, + "lr": 4.2000000000000004e-05, + "step": 22, + "tokens_trained": 0.002094288 + }, + { + "epoch": 0.006808510638297872, + "grad_norm": 87.5854721069336, + "loss": 37.6201, + "lr": 4.6e-05, + "step": 24, + "tokens_trained": 0.002282496 + }, + { + "epoch": 0.007375886524822695, + "grad_norm": 84.12794494628906, + "loss": 35.0091, + "lr": 5e-05, + "step": 26, + "tokens_trained": 0.00247068 + }, + { + "epoch": 0.007943262411347518, + "grad_norm": 79.77535247802734, + "loss": 33.2253, + "lr": 5.4e-05, + "step": 28, + "tokens_trained": 0.002662888 + }, + { + "epoch": 0.00851063829787234, + "grad_norm": 66.42157745361328, + "loss": 32.0682, + "lr": 5.800000000000001e-05, + "step": 30, + "tokens_trained": 0.002851968 + }, + { + "epoch": 0.009078014184397163, + "grad_norm": 87.52485656738281, + "loss": 30.893, + "lr": 6.2e-05, + "step": 32, + "tokens_trained": 0.003041384 + }, + { + "epoch": 0.009645390070921986, + "grad_norm": 58.33614730834961, + "loss": 30.0513, + "lr": 6.6e-05, + "step": 34, + "tokens_trained": 0.003232872 + }, + { + "epoch": 0.010212765957446808, + "grad_norm": 54.629329681396484, + "loss": 29.0115, + "lr": 7.000000000000001e-05, + "step": 36, + "tokens_trained": 0.003423824 + }, + { + "epoch": 0.01078014184397163, + "grad_norm": 52.79097366333008, + "loss": 28.2084, + "lr": 7.4e-05, + "step": 38, + "tokens_trained": 0.003613232 + }, + { + "epoch": 0.011347517730496455, + "grad_norm": 54.481224060058594, + "loss": 27.4345, + "lr": 7.8e-05, + "step": 40, + "tokens_trained": 0.003800952 + }, + { + "epoch": 0.011914893617021277, + "grad_norm": 58.7069091796875, + "loss": 26.5936, + "lr": 8.2e-05, + "step": 42, + "tokens_trained": 0.003991512 + }, + { + "epoch": 0.0124822695035461, + "grad_norm": 49.30760955810547, + "loss": 26.0608, + "lr": 8.599999999999999e-05, + "step": 44, + "tokens_trained": 0.004180648 + }, + { + "epoch": 0.013049645390070922, + "grad_norm": 61.902587890625, + "loss": 25.5363, + "lr": 8.999999999999999e-05, + "step": 46, + "tokens_trained": 0.00437148 + }, + { + "epoch": 0.013617021276595745, + "grad_norm": 46.76111602783203, + "loss": 24.9599, + "lr": 9.400000000000001e-05, + "step": 48, + "tokens_trained": 0.004559344 + }, + { + "epoch": 0.014184397163120567, + "grad_norm": 57.06416702270508, + "loss": 24.4087, + "lr": 9.800000000000001e-05, + "step": 50, + "tokens_trained": 0.004749256 + }, + { + "epoch": 0.01475177304964539, + "grad_norm": 44.798736572265625, + "loss": 24.1444, + "lr": 0.000102, + "step": 52, + "tokens_trained": 0.004940192 + }, + { + "epoch": 0.015319148936170212, + "grad_norm": 40.29296875, + "loss": 23.6011, + "lr": 0.000106, + "step": 54, + "tokens_trained": 0.005130304 + }, + { + "epoch": 0.015886524822695036, + "grad_norm": 38.75099563598633, + "loss": 23.1781, + "lr": 0.00011, + "step": 56, + "tokens_trained": 0.005322864 + }, + { + "epoch": 0.016453900709219857, + "grad_norm": 37.470706939697266, + "loss": 22.9136, + "lr": 0.000114, + "step": 58, + "tokens_trained": 0.00551392 + }, + { + "epoch": 0.01702127659574468, + "grad_norm": 35.1894645690918, + "loss": 22.6336, + "lr": 0.000118, + "step": 60, + "tokens_trained": 0.005703096 + }, + { + "epoch": 0.017588652482269502, + "grad_norm": 35.136573791503906, + "loss": 22.2998, + "lr": 0.000122, + "step": 62, + "tokens_trained": 0.005892448 + }, + { + "epoch": 0.018156028368794326, + "grad_norm": 38.05111312866211, + "loss": 21.9401, + "lr": 0.000126, + "step": 64, + "tokens_trained": 0.006081656 + }, + { + "epoch": 0.01872340425531915, + "grad_norm": 35.63850021362305, + "loss": 21.7206, + "lr": 0.00013000000000000002, + "step": 66, + "tokens_trained": 0.006273032 + }, + { + "epoch": 0.01929078014184397, + "grad_norm": 34.327667236328125, + "loss": 21.4051, + "lr": 0.000134, + "step": 68, + "tokens_trained": 0.00646304 + }, + { + "epoch": 0.019858156028368795, + "grad_norm": 31.457059860229492, + "loss": 21.0774, + "lr": 0.00013800000000000002, + "step": 70, + "tokens_trained": 0.006652832 + }, + { + "epoch": 0.020425531914893616, + "grad_norm": 34.91672897338867, + "loss": 20.8718, + "lr": 0.00014199999999999998, + "step": 72, + "tokens_trained": 0.006843512 + }, + { + "epoch": 0.02099290780141844, + "grad_norm": 27.959579467773438, + "loss": 20.6932, + "lr": 0.000146, + "step": 74, + "tokens_trained": 0.007033584 + }, + { + "epoch": 0.02156028368794326, + "grad_norm": 26.569866180419922, + "loss": 20.4072, + "lr": 0.00015, + "step": 76, + "tokens_trained": 0.007224032 + }, + { + "epoch": 0.022127659574468085, + "grad_norm": 28.009904861450195, + "loss": 20.2229, + "lr": 0.000154, + "step": 78, + "tokens_trained": 0.00741368 + }, + { + "epoch": 0.02269503546099291, + "grad_norm": 28.892959594726562, + "loss": 20.0528, + "lr": 0.000158, + "step": 80, + "tokens_trained": 0.00760416 + }, + { + "epoch": 0.02326241134751773, + "grad_norm": 31.58131980895996, + "loss": 19.8016, + "lr": 0.000162, + "step": 82, + "tokens_trained": 0.007793952 + }, + { + "epoch": 0.023829787234042554, + "grad_norm": 31.01254653930664, + "loss": 19.634, + "lr": 0.00016600000000000002, + "step": 84, + "tokens_trained": 0.007980792 + }, + { + "epoch": 0.024397163120567375, + "grad_norm": 28.732515335083008, + "loss": 19.3777, + "lr": 0.00017, + "step": 86, + "tokens_trained": 0.008171968 + }, + { + "epoch": 0.0249645390070922, + "grad_norm": 24.31264877319336, + "loss": 19.1346, + "lr": 0.000174, + "step": 88, + "tokens_trained": 0.008361632 + }, + { + "epoch": 0.02553191489361702, + "grad_norm": 26.557010650634766, + "loss": 19.0014, + "lr": 0.000178, + "step": 90, + "tokens_trained": 0.008552328 + }, + { + "epoch": 0.026099290780141844, + "grad_norm": 21.156103134155273, + "loss": 18.7032, + "lr": 0.000182, + "step": 92, + "tokens_trained": 0.008743136 + }, + { + "epoch": 0.02666666666666667, + "grad_norm": 25.7484188079834, + "loss": 18.4836, + "lr": 0.000186, + "step": 94, + "tokens_trained": 0.008932056 + }, + { + "epoch": 0.02723404255319149, + "grad_norm": 22.27949333190918, + "loss": 18.2233, + "lr": 0.00019, + "step": 96, + "tokens_trained": 0.009121608 + }, + { + "epoch": 0.027801418439716313, + "grad_norm": 24.9247989654541, + "loss": 17.9867, + "lr": 0.000194, + "step": 98, + "tokens_trained": 0.009311008 + }, + { + "epoch": 0.028368794326241134, + "grad_norm": 24.302066802978516, + "loss": 17.8016, + "lr": 0.00019800000000000002, + "step": 100, + "tokens_trained": 0.009501456 + }, + { + "epoch": 0.02893617021276596, + "grad_norm": 23.458459854125977, + "loss": 17.6295, + "lr": 0.000202, + "step": 102, + "tokens_trained": 0.009693952 + }, + { + "epoch": 0.02950354609929078, + "grad_norm": 24.092350006103516, + "loss": 17.4593, + "lr": 0.000206, + "step": 104, + "tokens_trained": 0.009883328 + }, + { + "epoch": 0.030070921985815603, + "grad_norm": 22.54726219177246, + "loss": 17.2141, + "lr": 0.00021, + "step": 106, + "tokens_trained": 0.01007316 + }, + { + "epoch": 0.030638297872340424, + "grad_norm": 21.334760665893555, + "loss": 17.044, + "lr": 0.000214, + "step": 108, + "tokens_trained": 0.010266504 + }, + { + "epoch": 0.031205673758865248, + "grad_norm": 20.584287643432617, + "loss": 16.8919, + "lr": 0.000218, + "step": 110, + "tokens_trained": 0.010455736 + }, + { + "epoch": 0.03177304964539007, + "grad_norm": 23.51676368713379, + "loss": 16.751, + "lr": 0.000222, + "step": 112, + "tokens_trained": 0.010645208 + }, + { + "epoch": 0.03234042553191489, + "grad_norm": 23.278276443481445, + "loss": 16.5997, + "lr": 0.00022600000000000002, + "step": 114, + "tokens_trained": 0.010838928 + }, + { + "epoch": 0.032907801418439714, + "grad_norm": 25.4830265045166, + "loss": 16.3416, + "lr": 0.00023, + "step": 116, + "tokens_trained": 0.011027792 + }, + { + "epoch": 0.03347517730496454, + "grad_norm": 29.442413330078125, + "loss": 16.24, + "lr": 0.00023400000000000002, + "step": 118, + "tokens_trained": 0.011217456 + }, + { + "epoch": 0.03404255319148936, + "grad_norm": 21.77578353881836, + "loss": 16.1922, + "lr": 0.00023799999999999998, + "step": 120, + "tokens_trained": 0.01140804 + }, + { + "epoch": 0.03460992907801418, + "grad_norm": 27.040719985961914, + "loss": 15.9059, + "lr": 0.000242, + "step": 122, + "tokens_trained": 0.011597816 + }, + { + "epoch": 0.035177304964539004, + "grad_norm": 24.74480628967285, + "loss": 15.7818, + "lr": 0.000246, + "step": 124, + "tokens_trained": 0.011785624 + }, + { + "epoch": 0.03546099290780142, + "eval_loss": 15.553059577941895, + "eval_runtime": 23.5485, + "step": 125, + "tokens_trained": 0.011880832 + }, + { + "epoch": 0.03574468085106383, + "grad_norm": 23.13482666015625, + "loss": 15.5739, + "lr": 0.00025, + "step": 126, + "tokens_trained": 0.011975976 + }, + { + "epoch": 0.03631205673758865, + "grad_norm": 22.8618106842041, + "loss": 15.4302, + "lr": 0.000254, + "step": 128, + "tokens_trained": 0.012166744 + }, + { + "epoch": 0.03687943262411347, + "grad_norm": 26.804859161376953, + "loss": 15.3623, + "lr": 0.00025800000000000004, + "step": 130, + "tokens_trained": 0.01235436 + }, + { + "epoch": 0.0374468085106383, + "grad_norm": 21.826601028442383, + "loss": 15.1465, + "lr": 0.000262, + "step": 132, + "tokens_trained": 0.012544976 + }, + { + "epoch": 0.03801418439716312, + "grad_norm": 39.447086334228516, + "loss": 15.0137, + "lr": 0.000266, + "step": 134, + "tokens_trained": 0.012736352 + }, + { + "epoch": 0.03858156028368794, + "grad_norm": 23.44275665283203, + "loss": 14.9355, + "lr": 0.00027, + "step": 136, + "tokens_trained": 0.012925008 + }, + { + "epoch": 0.03914893617021276, + "grad_norm": 21.631427764892578, + "loss": 14.6825, + "lr": 0.00027400000000000005, + "step": 138, + "tokens_trained": 0.013114672 + }, + { + "epoch": 0.03971631205673759, + "grad_norm": 23.674650192260742, + "loss": 14.5194, + "lr": 0.00027800000000000004, + "step": 140, + "tokens_trained": 0.013304016 + }, + { + "epoch": 0.04028368794326241, + "grad_norm": 23.974796295166016, + "loss": 14.4829, + "lr": 0.00028199999999999997, + "step": 142, + "tokens_trained": 0.013496696 + }, + { + "epoch": 0.04085106382978723, + "grad_norm": 26.112201690673828, + "loss": 14.3027, + "lr": 0.00028599999999999996, + "step": 144, + "tokens_trained": 0.013684816 + }, + { + "epoch": 0.04141843971631206, + "grad_norm": 20.67386817932129, + "loss": 14.1499, + "lr": 0.00029, + "step": 146, + "tokens_trained": 0.013874832 + }, + { + "epoch": 0.04198581560283688, + "grad_norm": 24.253408432006836, + "loss": 13.9378, + "lr": 0.000294, + "step": 148, + "tokens_trained": 0.014065056 + }, + { + "epoch": 0.0425531914893617, + "grad_norm": 35.716087341308594, + "loss": 14.0562, + "lr": 0.000298, + "step": 150, + "tokens_trained": 0.014256784 + }, + { + "epoch": 0.04312056737588652, + "grad_norm": 29.414331436157227, + "loss": 14.0462, + "lr": 0.000302, + "step": 152, + "tokens_trained": 0.014446312 + }, + { + "epoch": 0.04368794326241135, + "grad_norm": 30.687482833862305, + "loss": 13.7603, + "lr": 0.000306, + "step": 154, + "tokens_trained": 0.014639872 + }, + { + "epoch": 0.04425531914893617, + "grad_norm": 29.806455612182617, + "loss": 13.708, + "lr": 0.00031, + "step": 156, + "tokens_trained": 0.014831112 + }, + { + "epoch": 0.04482269503546099, + "grad_norm": 24.900897979736328, + "loss": 13.548, + "lr": 0.000314, + "step": 158, + "tokens_trained": 0.015021288 + }, + { + "epoch": 0.04539007092198582, + "grad_norm": 24.29252815246582, + "loss": 13.3119, + "lr": 0.00031800000000000003, + "step": 160, + "tokens_trained": 0.01521228 + }, + { + "epoch": 0.04595744680851064, + "grad_norm": 20.68342399597168, + "loss": 13.1829, + "lr": 0.000322, + "step": 162, + "tokens_trained": 0.015403688 + }, + { + "epoch": 0.04652482269503546, + "grad_norm": 20.822795867919922, + "loss": 12.9044, + "lr": 0.000326, + "step": 164, + "tokens_trained": 0.015593416 + }, + { + "epoch": 0.04709219858156028, + "grad_norm": 21.689916610717773, + "loss": 12.6862, + "lr": 0.00033, + "step": 166, + "tokens_trained": 0.015784408 + }, + { + "epoch": 0.04765957446808511, + "grad_norm": 17.873889923095703, + "loss": 12.5502, + "lr": 0.00033400000000000004, + "step": 168, + "tokens_trained": 0.0159744 + }, + { + "epoch": 0.04822695035460993, + "grad_norm": 18.951616287231445, + "loss": 12.308, + "lr": 0.00033800000000000003, + "step": 170, + "tokens_trained": 0.016163736 + }, + { + "epoch": 0.04879432624113475, + "grad_norm": 15.146363258361816, + "loss": 12.1558, + "lr": 0.000342, + "step": 172, + "tokens_trained": 0.016353832 + }, + { + "epoch": 0.04936170212765958, + "grad_norm": 18.336984634399414, + "loss": 12.0386, + "lr": 0.000346, + "step": 174, + "tokens_trained": 0.016545088 + }, + { + "epoch": 0.0499290780141844, + "grad_norm": 17.221126556396484, + "loss": 11.8791, + "lr": 0.00035, + "step": 176, + "tokens_trained": 0.016735704 + }, + { + "epoch": 0.05049645390070922, + "grad_norm": 19.362564086914062, + "loss": 11.7224, + "lr": 0.000354, + "step": 178, + "tokens_trained": 0.016927944 + }, + { + "epoch": 0.05106382978723404, + "grad_norm": 15.564507484436035, + "loss": 11.6448, + "lr": 0.000358, + "step": 180, + "tokens_trained": 0.017116096 + }, + { + "epoch": 0.05163120567375887, + "grad_norm": 20.711383819580078, + "loss": 11.4398, + "lr": 0.000362, + "step": 182, + "tokens_trained": 0.01730564 + }, + { + "epoch": 0.05219858156028369, + "grad_norm": 18.627403259277344, + "loss": 11.3377, + "lr": 0.000366, + "step": 184, + "tokens_trained": 0.017495864 + }, + { + "epoch": 0.05276595744680851, + "grad_norm": 15.00942325592041, + "loss": 11.1416, + "lr": 0.00037, + "step": 186, + "tokens_trained": 0.017686464 + }, + { + "epoch": 0.05333333333333334, + "grad_norm": 17.070598602294922, + "loss": 11.0148, + "lr": 0.000374, + "step": 188, + "tokens_trained": 0.017879488 + }, + { + "epoch": 0.05390070921985816, + "grad_norm": 16.101457595825195, + "loss": 10.8874, + "lr": 0.000378, + "step": 190, + "tokens_trained": 0.018068312 + }, + { + "epoch": 0.05446808510638298, + "grad_norm": 15.613334655761719, + "loss": 10.7055, + "lr": 0.000382, + "step": 192, + "tokens_trained": 0.018255752 + }, + { + "epoch": 0.0550354609929078, + "grad_norm": 17.671857833862305, + "loss": 10.5706, + "lr": 0.000386, + "step": 194, + "tokens_trained": 0.018447096 + }, + { + "epoch": 0.05560283687943263, + "grad_norm": 16.080909729003906, + "loss": 10.4476, + "lr": 0.00039000000000000005, + "step": 196, + "tokens_trained": 0.018637264 + }, + { + "epoch": 0.05617021276595745, + "grad_norm": 15.02849292755127, + "loss": 10.2962, + "lr": 0.00039400000000000004, + "step": 198, + "tokens_trained": 0.018827552 + }, + { + "epoch": 0.05673758865248227, + "grad_norm": 14.990167617797852, + "loss": 10.1912, + "lr": 0.000398, + "step": 200, + "tokens_trained": 0.019018 + }, + { + "epoch": 0.05730496453900709, + "grad_norm": 15.390633583068848, + "loss": 10.0442, + "lr": 0.000402, + "step": 202, + "tokens_trained": 0.019209864 + }, + { + "epoch": 0.05787234042553192, + "grad_norm": 16.871570587158203, + "loss": 9.9685, + "lr": 0.00040600000000000006, + "step": 204, + "tokens_trained": 0.019400176 + }, + { + "epoch": 0.05843971631205674, + "grad_norm": 20.16544532775879, + "loss": 9.8531, + "lr": 0.00041, + "step": 206, + "tokens_trained": 0.019589424 + }, + { + "epoch": 0.05900709219858156, + "grad_norm": 16.825023651123047, + "loss": 9.7777, + "lr": 0.000414, + "step": 208, + "tokens_trained": 0.019779112 + }, + { + "epoch": 0.059574468085106386, + "grad_norm": 16.43510627746582, + "loss": 9.6122, + "lr": 0.00041799999999999997, + "step": 210, + "tokens_trained": 0.019970048 + }, + { + "epoch": 0.060141843971631206, + "grad_norm": 17.340473175048828, + "loss": 9.4859, + "lr": 0.000422, + "step": 212, + "tokens_trained": 0.020160968 + }, + { + "epoch": 0.06070921985815603, + "grad_norm": 15.019119262695312, + "loss": 9.3656, + "lr": 0.000426, + "step": 214, + "tokens_trained": 0.020349664 + }, + { + "epoch": 0.06127659574468085, + "grad_norm": 13.379194259643555, + "loss": 9.2348, + "lr": 0.00043, + "step": 216, + "tokens_trained": 0.020538192 + }, + { + "epoch": 0.061843971631205676, + "grad_norm": 16.71472930908203, + "loss": 9.2258, + "lr": 0.00043400000000000003, + "step": 218, + "tokens_trained": 0.020728936 + }, + { + "epoch": 0.062411347517730496, + "grad_norm": 12.743139266967773, + "loss": 9.0569, + "lr": 0.000438, + "step": 220, + "tokens_trained": 0.020917472 + }, + { + "epoch": 0.06297872340425532, + "grad_norm": 15.739934921264648, + "loss": 8.9623, + "lr": 0.000442, + "step": 222, + "tokens_trained": 0.02110928 + }, + { + "epoch": 0.06354609929078014, + "grad_norm": 14.23620891571045, + "loss": 8.8201, + "lr": 0.000446, + "step": 224, + "tokens_trained": 0.021300168 + }, + { + "epoch": 0.06411347517730497, + "grad_norm": 13.005538940429688, + "loss": 8.7235, + "lr": 0.00045000000000000004, + "step": 226, + "tokens_trained": 0.021490272 + }, + { + "epoch": 0.06468085106382979, + "grad_norm": 17.17629051208496, + "loss": 8.6907, + "lr": 0.00045400000000000003, + "step": 228, + "tokens_trained": 0.021681552 + }, + { + "epoch": 0.06524822695035461, + "grad_norm": 14.430739402770996, + "loss": 8.6196, + "lr": 0.000458, + "step": 230, + "tokens_trained": 0.02187236 + }, + { + "epoch": 0.06581560283687943, + "grad_norm": 14.575714111328125, + "loss": 8.4741, + "lr": 0.000462, + "step": 232, + "tokens_trained": 0.022061976 + }, + { + "epoch": 0.06638297872340425, + "grad_norm": 13.892754554748535, + "loss": 8.4118, + "lr": 0.00046600000000000005, + "step": 234, + "tokens_trained": 0.022252008 + }, + { + "epoch": 0.06695035460992908, + "grad_norm": 11.58240795135498, + "loss": 8.2781, + "lr": 0.00047, + "step": 236, + "tokens_trained": 0.02244284 + }, + { + "epoch": 0.0675177304964539, + "grad_norm": 13.022644996643066, + "loss": 8.2139, + "lr": 0.000474, + "step": 238, + "tokens_trained": 0.022631152 + }, + { + "epoch": 0.06808510638297872, + "grad_norm": 11.844677925109863, + "loss": 8.1134, + "lr": 0.00047799999999999996, + "step": 240, + "tokens_trained": 0.022821096 + }, + { + "epoch": 0.06865248226950355, + "grad_norm": 13.878067016601562, + "loss": 8.0221, + "lr": 0.000482, + "step": 242, + "tokens_trained": 0.023011656 + }, + { + "epoch": 0.06921985815602837, + "grad_norm": 12.34648323059082, + "loss": 7.9755, + "lr": 0.000486, + "step": 244, + "tokens_trained": 0.023201 + }, + { + "epoch": 0.06978723404255319, + "grad_norm": 14.238297462463379, + "loss": 7.8969, + "lr": 0.00049, + "step": 246, + "tokens_trained": 0.023391128 + }, + { + "epoch": 0.07035460992907801, + "grad_norm": 14.386019706726074, + "loss": 7.8627, + "lr": 0.000494, + "step": 248, + "tokens_trained": 0.023581768 + }, + { + "epoch": 0.07092198581560284, + "grad_norm": 13.623086929321289, + "loss": 7.7568, + "lr": 0.000498, + "step": 250, + "tokens_trained": 0.023771248 + }, + { + "epoch": 0.07092198581560284, + "eval_loss": 7.70297384262085, + "eval_runtime": 21.3853, + "step": 250, + "tokens_trained": 0.023771248 + }, + { + "epoch": 0.07148936170212766, + "grad_norm": 14.347646713256836, + "loss": 7.6842, + "lr": 0.0005020000000000001, + "step": 252, + "tokens_trained": 0.023961056 + }, + { + "epoch": 0.07205673758865248, + "grad_norm": 12.5592041015625, + "loss": 7.6516, + "lr": 0.000506, + "step": 254, + "tokens_trained": 0.024150968 + }, + { + "epoch": 0.0726241134751773, + "grad_norm": 13.219141960144043, + "loss": 7.5789, + "lr": 0.00051, + "step": 256, + "tokens_trained": 0.024340072 + }, + { + "epoch": 0.07319148936170213, + "grad_norm": 12.654081344604492, + "loss": 7.5369, + "lr": 0.000514, + "step": 258, + "tokens_trained": 0.024529296 + }, + { + "epoch": 0.07375886524822695, + "grad_norm": 13.136971473693848, + "loss": 7.4949, + "lr": 0.000518, + "step": 260, + "tokens_trained": 0.024719688 + }, + { + "epoch": 0.07432624113475177, + "grad_norm": 12.680288314819336, + "loss": 7.3904, + "lr": 0.000522, + "step": 262, + "tokens_trained": 0.024909632 + }, + { + "epoch": 0.0748936170212766, + "grad_norm": 12.754518508911133, + "loss": 7.3514, + "lr": 0.000526, + "step": 264, + "tokens_trained": 0.025098416 + }, + { + "epoch": 0.07546099290780142, + "grad_norm": 13.22311019897461, + "loss": 7.2951, + "lr": 0.0005300000000000001, + "step": 266, + "tokens_trained": 0.025287344 + }, + { + "epoch": 0.07602836879432624, + "grad_norm": 12.11903190612793, + "loss": 7.2229, + "lr": 0.0005340000000000001, + "step": 268, + "tokens_trained": 0.025477152 + }, + { + "epoch": 0.07659574468085106, + "grad_norm": 13.771833419799805, + "loss": 7.1815, + "lr": 0.0005380000000000001, + "step": 270, + "tokens_trained": 0.025668288 + }, + { + "epoch": 0.07716312056737588, + "grad_norm": 11.756864547729492, + "loss": 7.1669, + "lr": 0.0005420000000000001, + "step": 272, + "tokens_trained": 0.025858528 + }, + { + "epoch": 0.0777304964539007, + "grad_norm": 13.613094329833984, + "loss": 7.1079, + "lr": 0.000546, + "step": 274, + "tokens_trained": 0.026048616 + }, + { + "epoch": 0.07829787234042553, + "grad_norm": 10.001923561096191, + "loss": 7.0508, + "lr": 0.00055, + "step": 276, + "tokens_trained": 0.026236944 + }, + { + "epoch": 0.07886524822695036, + "grad_norm": 14.262083053588867, + "loss": 6.9955, + "lr": 0.000554, + "step": 278, + "tokens_trained": 0.026426848 + }, + { + "epoch": 0.07943262411347518, + "grad_norm": 12.381136894226074, + "loss": 6.9831, + "lr": 0.000558, + "step": 280, + "tokens_trained": 0.026616784 + }, + { + "epoch": 0.08, + "grad_norm": 9.815845489501953, + "loss": 6.917, + "lr": 0.0005620000000000001, + "step": 282, + "tokens_trained": 0.026805176 + }, + { + "epoch": 0.08056737588652482, + "grad_norm": 11.669997215270996, + "loss": 6.8999, + "lr": 0.000566, + "step": 284, + "tokens_trained": 0.02699488 + }, + { + "epoch": 0.08113475177304964, + "grad_norm": 12.770941734313965, + "loss": 6.8998, + "lr": 0.00057, + "step": 286, + "tokens_trained": 0.027185784 + }, + { + "epoch": 0.08170212765957446, + "grad_norm": 15.572457313537598, + "loss": 6.841, + "lr": 0.000574, + "step": 288, + "tokens_trained": 0.027375896 + }, + { + "epoch": 0.08226950354609928, + "grad_norm": 10.980833053588867, + "loss": 6.8545, + "lr": 0.000578, + "step": 290, + "tokens_trained": 0.02756588 + }, + { + "epoch": 0.08283687943262412, + "grad_norm": 11.678337097167969, + "loss": 6.7853, + "lr": 0.0005819999999999999, + "step": 292, + "tokens_trained": 0.02775456 + }, + { + "epoch": 0.08340425531914894, + "grad_norm": 9.77885913848877, + "loss": 6.7465, + "lr": 0.0005859999999999999, + "step": 294, + "tokens_trained": 0.027942856 + }, + { + "epoch": 0.08397163120567376, + "grad_norm": 13.62730884552002, + "loss": 6.7276, + "lr": 0.00059, + "step": 296, + "tokens_trained": 0.028133152 + }, + { + "epoch": 0.08453900709219858, + "grad_norm": 10.644404411315918, + "loss": 6.6802, + "lr": 0.000594, + "step": 298, + "tokens_trained": 0.028322192 + }, + { + "epoch": 0.0851063829787234, + "grad_norm": 11.130610466003418, + "loss": 6.6548, + "lr": 0.000598, + "step": 300, + "tokens_trained": 0.0285122 + }, + { + "epoch": 0.08567375886524822, + "grad_norm": 11.557455062866211, + "loss": 6.6155, + "lr": 0.000602, + "step": 302, + "tokens_trained": 0.028699792 + }, + { + "epoch": 0.08624113475177304, + "grad_norm": 9.276884078979492, + "loss": 6.5989, + "lr": 0.000606, + "step": 304, + "tokens_trained": 0.028889896 + }, + { + "epoch": 0.08680851063829788, + "grad_norm": 9.616179466247559, + "loss": 6.5773, + "lr": 0.00061, + "step": 306, + "tokens_trained": 0.029082272 + }, + { + "epoch": 0.0873758865248227, + "grad_norm": 10.575953483581543, + "loss": 6.5358, + "lr": 0.000614, + "step": 308, + "tokens_trained": 0.029273352 + }, + { + "epoch": 0.08794326241134752, + "grad_norm": 9.089850425720215, + "loss": 6.5088, + "lr": 0.0006180000000000001, + "step": 310, + "tokens_trained": 0.029463848 + }, + { + "epoch": 0.08851063829787234, + "grad_norm": 9.090002059936523, + "loss": 6.4849, + "lr": 0.000622, + "step": 312, + "tokens_trained": 0.029653272 + }, + { + "epoch": 0.08907801418439716, + "grad_norm": 12.038308143615723, + "loss": 6.4624, + "lr": 0.000626, + "step": 314, + "tokens_trained": 0.029841928 + }, + { + "epoch": 0.08964539007092198, + "grad_norm": 9.073866844177246, + "loss": 6.4515, + "lr": 0.00063, + "step": 316, + "tokens_trained": 0.030029808 + }, + { + "epoch": 0.0902127659574468, + "grad_norm": 8.727197647094727, + "loss": 6.43, + "lr": 0.000634, + "step": 318, + "tokens_trained": 0.030221288 + }, + { + "epoch": 0.09078014184397164, + "grad_norm": 14.558151245117188, + "loss": 6.4487, + "lr": 0.000638, + "step": 320, + "tokens_trained": 0.030410872 + }, + { + "epoch": 0.09134751773049646, + "grad_norm": 9.98914623260498, + "loss": 6.4279, + "lr": 0.000642, + "step": 322, + "tokens_trained": 0.030602376 + }, + { + "epoch": 0.09191489361702128, + "grad_norm": 10.395442962646484, + "loss": 6.4311, + "lr": 0.000646, + "step": 324, + "tokens_trained": 0.030792968 + }, + { + "epoch": 0.0924822695035461, + "grad_norm": 10.8250093460083, + "loss": 6.3726, + "lr": 0.0006500000000000001, + "step": 326, + "tokens_trained": 0.030982944 + }, + { + "epoch": 0.09304964539007092, + "grad_norm": 9.73416805267334, + "loss": 6.34, + "lr": 0.0006540000000000001, + "step": 328, + "tokens_trained": 0.031174928 + }, + { + "epoch": 0.09361702127659574, + "grad_norm": 8.596503257751465, + "loss": 6.3322, + "lr": 0.0006580000000000001, + "step": 330, + "tokens_trained": 0.031364288 + }, + { + "epoch": 0.09418439716312056, + "grad_norm": 8.49472427368164, + "loss": 6.3096, + "lr": 0.000662, + "step": 332, + "tokens_trained": 0.03155376 + }, + { + "epoch": 0.0947517730496454, + "grad_norm": 7.857503414154053, + "loss": 6.2368, + "lr": 0.000666, + "step": 334, + "tokens_trained": 0.031744368 + }, + { + "epoch": 0.09531914893617022, + "grad_norm": 9.007513999938965, + "loss": 6.198, + "lr": 0.00067, + "step": 336, + "tokens_trained": 0.031934136 + }, + { + "epoch": 0.09588652482269504, + "grad_norm": 8.185524940490723, + "loss": 6.2328, + "lr": 0.000674, + "step": 338, + "tokens_trained": 0.032124984 + }, + { + "epoch": 0.09645390070921986, + "grad_norm": 8.784396171569824, + "loss": 6.1945, + "lr": 0.0006780000000000001, + "step": 340, + "tokens_trained": 0.032316016 + }, + { + "epoch": 0.09702127659574468, + "grad_norm": 8.642311096191406, + "loss": 6.218, + "lr": 0.0006820000000000001, + "step": 342, + "tokens_trained": 0.032506224 + }, + { + "epoch": 0.0975886524822695, + "grad_norm": 8.493780136108398, + "loss": 6.194, + "lr": 0.0006860000000000001, + "step": 344, + "tokens_trained": 0.032696152 + }, + { + "epoch": 0.09815602836879432, + "grad_norm": 9.120508193969727, + "loss": 6.2241, + "lr": 0.00069, + "step": 346, + "tokens_trained": 0.032885688 + }, + { + "epoch": 0.09872340425531916, + "grad_norm": 9.34500503540039, + "loss": 6.1548, + "lr": 0.000694, + "step": 348, + "tokens_trained": 0.03307568 + }, + { + "epoch": 0.09929078014184398, + "grad_norm": 7.483356952667236, + "loss": 6.1282, + "lr": 0.0006979999999999999, + "step": 350, + "tokens_trained": 0.033267208 + }, + { + "epoch": 0.0998581560283688, + "grad_norm": 7.974069118499756, + "loss": 6.1032, + "lr": 0.0007019999999999999, + "step": 352, + "tokens_trained": 0.033458144 + }, + { + "epoch": 0.10042553191489362, + "grad_norm": 8.247384071350098, + "loss": 6.1698, + "lr": 0.0007059999999999999, + "step": 354, + "tokens_trained": 0.033650352 + }, + { + "epoch": 0.10099290780141844, + "grad_norm": 8.554885864257812, + "loss": 6.1429, + "lr": 0.00071, + "step": 356, + "tokens_trained": 0.033840232 + }, + { + "epoch": 0.10156028368794326, + "grad_norm": 7.209281921386719, + "loss": 6.0997, + "lr": 0.000714, + "step": 358, + "tokens_trained": 0.034030032 + }, + { + "epoch": 0.10212765957446808, + "grad_norm": 8.660383224487305, + "loss": 6.1497, + "lr": 0.000718, + "step": 360, + "tokens_trained": 0.034218592 + }, + { + "epoch": 0.10269503546099291, + "grad_norm": 9.382761001586914, + "loss": 6.0665, + "lr": 0.000722, + "step": 362, + "tokens_trained": 0.034408408 + }, + { + "epoch": 0.10326241134751774, + "grad_norm": 6.915714263916016, + "loss": 6.0636, + "lr": 0.000726, + "step": 364, + "tokens_trained": 0.034600016 + }, + { + "epoch": 0.10382978723404256, + "grad_norm": 7.8990631103515625, + "loss": 6.0975, + "lr": 0.00073, + "step": 366, + "tokens_trained": 0.034790792 + }, + { + "epoch": 0.10439716312056738, + "grad_norm": 8.859809875488281, + "loss": 6.0754, + "lr": 0.000734, + "step": 368, + "tokens_trained": 0.034981304 + }, + { + "epoch": 0.1049645390070922, + "grad_norm": 7.392801761627197, + "loss": 6.039, + "lr": 0.000738, + "step": 370, + "tokens_trained": 0.03516956 + }, + { + "epoch": 0.10553191489361702, + "grad_norm": 9.427324295043945, + "loss": 6.084, + "lr": 0.000742, + "step": 372, + "tokens_trained": 0.035358816 + }, + { + "epoch": 0.10609929078014184, + "grad_norm": 7.168910503387451, + "loss": 6.0498, + "lr": 0.000746, + "step": 374, + "tokens_trained": 0.035548016 + }, + { + "epoch": 0.10638297872340426, + "eval_loss": 6.038269996643066, + "eval_runtime": 21.3445, + "step": 375, + "tokens_trained": 0.035644104 + }, + { + "epoch": 0.10666666666666667, + "grad_norm": 7.899259567260742, + "loss": 6.0345, + "lr": 0.00075, + "step": 376, + "tokens_trained": 0.035739856 + }, + { + "epoch": 0.1072340425531915, + "grad_norm": 8.91533374786377, + "loss": 6.0386, + "lr": 0.000754, + "step": 378, + "tokens_trained": 0.035930264 + }, + { + "epoch": 0.10780141843971631, + "grad_norm": 6.998043060302734, + "loss": 6.0294, + "lr": 0.000758, + "step": 380, + "tokens_trained": 0.036119616 + }, + { + "epoch": 0.10836879432624114, + "grad_norm": 7.343894958496094, + "loss": 6.0116, + "lr": 0.000762, + "step": 382, + "tokens_trained": 0.036308416 + }, + { + "epoch": 0.10893617021276596, + "grad_norm": 8.182528495788574, + "loss": 5.9904, + "lr": 0.0007660000000000001, + "step": 384, + "tokens_trained": 0.036497264 + }, + { + "epoch": 0.10950354609929078, + "grad_norm": 7.927818775177002, + "loss": 6.0345, + "lr": 0.0007700000000000001, + "step": 386, + "tokens_trained": 0.036688192 + }, + { + "epoch": 0.1100709219858156, + "grad_norm": 8.07447338104248, + "loss": 5.9685, + "lr": 0.0007740000000000001, + "step": 388, + "tokens_trained": 0.036878256 + }, + { + "epoch": 0.11063829787234042, + "grad_norm": 7.281871318817139, + "loss": 6.0125, + "lr": 0.000778, + "step": 390, + "tokens_trained": 0.037068272 + }, + { + "epoch": 0.11120567375886525, + "grad_norm": 8.298929214477539, + "loss": 6.0071, + "lr": 0.000782, + "step": 392, + "tokens_trained": 0.037259464 + }, + { + "epoch": 0.11177304964539007, + "grad_norm": 7.546716690063477, + "loss": 5.9721, + "lr": 0.000786, + "step": 394, + "tokens_trained": 0.037449696 + }, + { + "epoch": 0.1123404255319149, + "grad_norm": 8.28548526763916, + "loss": 5.9819, + "lr": 0.00079, + "step": 396, + "tokens_trained": 0.037639672 + }, + { + "epoch": 0.11290780141843972, + "grad_norm": 7.064655303955078, + "loss": 5.9873, + "lr": 0.0007940000000000001, + "step": 398, + "tokens_trained": 0.03782712 + }, + { + "epoch": 0.11347517730496454, + "grad_norm": 7.743175506591797, + "loss": 5.9528, + "lr": 0.0007980000000000001, + "step": 400, + "tokens_trained": 0.03801792 + }, + { + "epoch": 0.11404255319148936, + "grad_norm": 7.00898551940918, + "loss": 5.9504, + "lr": 0.0008020000000000001, + "step": 402, + "tokens_trained": 0.038209176 + }, + { + "epoch": 0.11460992907801418, + "grad_norm": 7.9350409507751465, + "loss": 5.9555, + "lr": 0.0008060000000000001, + "step": 404, + "tokens_trained": 0.03839824 + }, + { + "epoch": 0.11517730496453901, + "grad_norm": 7.048569679260254, + "loss": 5.9787, + "lr": 0.0008100000000000001, + "step": 406, + "tokens_trained": 0.03858732 + }, + { + "epoch": 0.11574468085106383, + "grad_norm": 7.088194370269775, + "loss": 5.928, + "lr": 0.0008139999999999999, + "step": 408, + "tokens_trained": 0.038777712 + }, + { + "epoch": 0.11631205673758865, + "grad_norm": 8.230712890625, + "loss": 5.9716, + "lr": 0.0008179999999999999, + "step": 410, + "tokens_trained": 0.038969464 + }, + { + "epoch": 0.11687943262411347, + "grad_norm": 8.076972007751465, + "loss": 5.9624, + "lr": 0.0008219999999999999, + "step": 412, + "tokens_trained": 0.039162064 + }, + { + "epoch": 0.1174468085106383, + "grad_norm": 8.065289497375488, + "loss": 5.9937, + "lr": 0.000826, + "step": 414, + "tokens_trained": 0.039348688 + }, + { + "epoch": 0.11801418439716312, + "grad_norm": 6.393420696258545, + "loss": 5.9278, + "lr": 0.00083, + "step": 416, + "tokens_trained": 0.03953732 + }, + { + "epoch": 0.11858156028368794, + "grad_norm": 7.384702682495117, + "loss": 5.931, + "lr": 0.000834, + "step": 418, + "tokens_trained": 0.039729808 + }, + { + "epoch": 0.11914893617021277, + "grad_norm": 7.007425308227539, + "loss": 5.93, + "lr": 0.000838, + "step": 420, + "tokens_trained": 0.039921096 + }, + { + "epoch": 0.11971631205673759, + "grad_norm": 7.112692832946777, + "loss": 5.9625, + "lr": 0.000842, + "step": 422, + "tokens_trained": 0.040110856 + }, + { + "epoch": 0.12028368794326241, + "grad_norm": 8.484418869018555, + "loss": 5.9848, + "lr": 0.000846, + "step": 424, + "tokens_trained": 0.040300504 + }, + { + "epoch": 0.12085106382978723, + "grad_norm": 6.633459091186523, + "loss": 6.0226, + "lr": 0.00085, + "step": 426, + "tokens_trained": 0.04049056 + }, + { + "epoch": 0.12141843971631205, + "grad_norm": 7.796964168548584, + "loss": 5.9152, + "lr": 0.000854, + "step": 428, + "tokens_trained": 0.040680544 + }, + { + "epoch": 0.12198581560283688, + "grad_norm": 7.833578586578369, + "loss": 5.924, + "lr": 0.000858, + "step": 430, + "tokens_trained": 0.040873128 + }, + { + "epoch": 0.1225531914893617, + "grad_norm": 6.7470550537109375, + "loss": 5.9318, + "lr": 0.000862, + "step": 432, + "tokens_trained": 0.041063488 + }, + { + "epoch": 0.12312056737588653, + "grad_norm": 6.066318988800049, + "loss": 5.9569, + "lr": 0.000866, + "step": 434, + "tokens_trained": 0.041254368 + }, + { + "epoch": 0.12368794326241135, + "grad_norm": 6.753541469573975, + "loss": 5.8851, + "lr": 0.00087, + "step": 436, + "tokens_trained": 0.04144516 + }, + { + "epoch": 0.12425531914893617, + "grad_norm": 6.471331596374512, + "loss": 5.864, + "lr": 0.000874, + "step": 438, + "tokens_trained": 0.041636912 + }, + { + "epoch": 0.12482269503546099, + "grad_norm": 6.129056930541992, + "loss": 5.8965, + "lr": 0.000878, + "step": 440, + "tokens_trained": 0.041828104 + }, + { + "epoch": 0.1253900709219858, + "grad_norm": 6.478890895843506, + "loss": 5.8817, + "lr": 0.000882, + "step": 442, + "tokens_trained": 0.04201808 + }, + { + "epoch": 0.12595744680851065, + "grad_norm": 6.014713287353516, + "loss": 5.8268, + "lr": 0.0008860000000000001, + "step": 444, + "tokens_trained": 0.042207328 + }, + { + "epoch": 0.12652482269503545, + "grad_norm": 5.505755424499512, + "loss": 5.8684, + "lr": 0.0008900000000000001, + "step": 446, + "tokens_trained": 0.042398152 + }, + { + "epoch": 0.1270921985815603, + "grad_norm": 10.096606254577637, + "loss": 5.8608, + "lr": 0.000894, + "step": 448, + "tokens_trained": 0.042588984 + }, + { + "epoch": 0.1276595744680851, + "grad_norm": 6.388499736785889, + "loss": 5.8766, + "lr": 0.000898, + "step": 450, + "tokens_trained": 0.042778592 + }, + { + "epoch": 0.12822695035460993, + "grad_norm": 7.145125865936279, + "loss": 5.8571, + "lr": 0.000902, + "step": 452, + "tokens_trained": 0.042967176 + }, + { + "epoch": 0.12879432624113477, + "grad_norm": 6.826383113861084, + "loss": 5.8655, + "lr": 0.000906, + "step": 454, + "tokens_trained": 0.043158952 + }, + { + "epoch": 0.12936170212765957, + "grad_norm": 6.036892414093018, + "loss": 5.8775, + "lr": 0.00091, + "step": 456, + "tokens_trained": 0.043349288 + }, + { + "epoch": 0.1299290780141844, + "grad_norm": 6.36528205871582, + "loss": 5.8908, + "lr": 0.0009140000000000001, + "step": 458, + "tokens_trained": 0.043539888 + }, + { + "epoch": 0.13049645390070921, + "grad_norm": 6.317558288574219, + "loss": 5.8702, + "lr": 0.0009180000000000001, + "step": 460, + "tokens_trained": 0.04373232 + }, + { + "epoch": 0.13106382978723405, + "grad_norm": 6.427131175994873, + "loss": 5.8399, + "lr": 0.0009220000000000001, + "step": 462, + "tokens_trained": 0.043922744 + }, + { + "epoch": 0.13163120567375886, + "grad_norm": 5.666539669036865, + "loss": 5.7899, + "lr": 0.0009260000000000001, + "step": 464, + "tokens_trained": 0.044112888 + }, + { + "epoch": 0.1321985815602837, + "grad_norm": 5.241824150085449, + "loss": 5.8203, + "lr": 0.00093, + "step": 466, + "tokens_trained": 0.04430244 + }, + { + "epoch": 0.1327659574468085, + "grad_norm": 6.072646141052246, + "loss": 5.8367, + "lr": 0.000934, + "step": 468, + "tokens_trained": 0.044493528 + }, + { + "epoch": 0.13333333333333333, + "grad_norm": 6.414418697357178, + "loss": 5.8236, + "lr": 0.0009379999999999999, + "step": 470, + "tokens_trained": 0.044682328 + }, + { + "epoch": 0.13390070921985817, + "grad_norm": 6.958801746368408, + "loss": 5.8179, + "lr": 0.000942, + "step": 472, + "tokens_trained": 0.044874256 + }, + { + "epoch": 0.13446808510638297, + "grad_norm": 5.787843227386475, + "loss": 5.8478, + "lr": 0.000946, + "step": 474, + "tokens_trained": 0.045065616 + }, + { + "epoch": 0.1350354609929078, + "grad_norm": 5.5841240882873535, + "loss": 5.8307, + "lr": 0.00095, + "step": 476, + "tokens_trained": 0.045257024 + }, + { + "epoch": 0.13560283687943261, + "grad_norm": 6.607712745666504, + "loss": 5.8512, + "lr": 0.000954, + "step": 478, + "tokens_trained": 0.045446432 + }, + { + "epoch": 0.13617021276595745, + "grad_norm": 5.473597049713135, + "loss": 5.8174, + "lr": 0.000958, + "step": 480, + "tokens_trained": 0.045636392 + }, + { + "epoch": 0.13673758865248226, + "grad_norm": 5.435728549957275, + "loss": 5.8308, + "lr": 0.000962, + "step": 482, + "tokens_trained": 0.045823784 + }, + { + "epoch": 0.1373049645390071, + "grad_norm": 6.049300670623779, + "loss": 5.8293, + "lr": 0.000966, + "step": 484, + "tokens_trained": 0.046013408 + }, + { + "epoch": 0.13787234042553193, + "grad_norm": 6.311764717102051, + "loss": 5.8086, + "lr": 0.0009699999999999999, + "step": 486, + "tokens_trained": 0.046202528 + }, + { + "epoch": 0.13843971631205673, + "grad_norm": 5.886009216308594, + "loss": 5.7986, + "lr": 0.000974, + "step": 488, + "tokens_trained": 0.04639404 + }, + { + "epoch": 0.13900709219858157, + "grad_norm": 5.438202381134033, + "loss": 5.8473, + "lr": 0.000978, + "step": 490, + "tokens_trained": 0.046586512 + }, + { + "epoch": 0.13957446808510637, + "grad_norm": 5.08393669128418, + "loss": 5.7613, + "lr": 0.000982, + "step": 492, + "tokens_trained": 0.046777448 + }, + { + "epoch": 0.1401418439716312, + "grad_norm": 5.645389080047607, + "loss": 5.7723, + "lr": 0.0009860000000000001, + "step": 494, + "tokens_trained": 0.046966096 + }, + { + "epoch": 0.14070921985815601, + "grad_norm": 6.320916652679443, + "loss": 5.7772, + "lr": 0.00099, + "step": 496, + "tokens_trained": 0.047155152 + }, + { + "epoch": 0.14127659574468085, + "grad_norm": 5.573540210723877, + "loss": 5.7412, + "lr": 0.000994, + "step": 498, + "tokens_trained": 0.047345352 + }, + { + "epoch": 0.14184397163120568, + "grad_norm": 4.939594745635986, + "loss": 5.8208, + "lr": 0.000998, + "step": 500, + "tokens_trained": 0.047535016 + }, + { + "epoch": 0.14184397163120568, + "eval_loss": 5.799490928649902, + "eval_runtime": 20.8575, + "step": 500, + "tokens_trained": 0.047535016 + }, + { + "epoch": 0.1424113475177305, + "grad_norm": 5.805343151092529, + "loss": 5.7734, + "lr": 0.00099986013986014, + "step": 502, + "tokens_trained": 0.047724216 + }, + { + "epoch": 0.14297872340425533, + "grad_norm": 5.831176280975342, + "loss": 5.8044, + "lr": 0.0009995804195804196, + "step": 504, + "tokens_trained": 0.047914328 + }, + { + "epoch": 0.14354609929078013, + "grad_norm": 5.045091152191162, + "loss": 5.8133, + "lr": 0.0009993006993006994, + "step": 506, + "tokens_trained": 0.048105032 + }, + { + "epoch": 0.14411347517730497, + "grad_norm": 5.276819705963135, + "loss": 5.7555, + "lr": 0.000999020979020979, + "step": 508, + "tokens_trained": 0.048293104 + }, + { + "epoch": 0.14468085106382977, + "grad_norm": 5.710324287414551, + "loss": 5.7619, + "lr": 0.0009987412587412587, + "step": 510, + "tokens_trained": 0.048483888 + }, + { + "epoch": 0.1452482269503546, + "grad_norm": 4.9472527503967285, + "loss": 5.767, + "lr": 0.0009984615384615386, + "step": 512, + "tokens_trained": 0.04867336 + }, + { + "epoch": 0.14581560283687944, + "grad_norm": 5.410078525543213, + "loss": 5.7238, + "lr": 0.0009981818181818182, + "step": 514, + "tokens_trained": 0.048863104 + }, + { + "epoch": 0.14638297872340425, + "grad_norm": 6.025843143463135, + "loss": 5.7664, + "lr": 0.000997902097902098, + "step": 516, + "tokens_trained": 0.049053856 + }, + { + "epoch": 0.14695035460992908, + "grad_norm": 5.3211669921875, + "loss": 5.747, + "lr": 0.0009976223776223777, + "step": 518, + "tokens_trained": 0.049245104 + }, + { + "epoch": 0.1475177304964539, + "grad_norm": 6.059483051300049, + "loss": 5.7611, + "lr": 0.0009973426573426573, + "step": 520, + "tokens_trained": 0.049434368 + }, + { + "epoch": 0.14808510638297873, + "grad_norm": 5.362505912780762, + "loss": 5.7607, + "lr": 0.000997062937062937, + "step": 522, + "tokens_trained": 0.049622648 + }, + { + "epoch": 0.14865248226950353, + "grad_norm": 5.391371726989746, + "loss": 5.7857, + "lr": 0.0009967832167832168, + "step": 524, + "tokens_trained": 0.049812304 + }, + { + "epoch": 0.14921985815602837, + "grad_norm": 4.3839030265808105, + "loss": 5.7334, + "lr": 0.0009965034965034964, + "step": 526, + "tokens_trained": 0.05000356 + }, + { + "epoch": 0.1497872340425532, + "grad_norm": 5.008530616760254, + "loss": 5.7475, + "lr": 0.0009962237762237763, + "step": 528, + "tokens_trained": 0.050193304 + }, + { + "epoch": 0.150354609929078, + "grad_norm": 5.068671226501465, + "loss": 5.7866, + "lr": 0.000995944055944056, + "step": 530, + "tokens_trained": 0.050382856 + }, + { + "epoch": 0.15092198581560284, + "grad_norm": 5.399240493774414, + "loss": 5.6857, + "lr": 0.0009956643356643356, + "step": 532, + "tokens_trained": 0.050570864 + }, + { + "epoch": 0.15148936170212765, + "grad_norm": 5.689481735229492, + "loss": 5.7586, + "lr": 0.0009953846153846154, + "step": 534, + "tokens_trained": 0.050760384 + }, + { + "epoch": 0.15205673758865249, + "grad_norm": 4.652275562286377, + "loss": 5.7866, + "lr": 0.000995104895104895, + "step": 536, + "tokens_trained": 0.050952712 + }, + { + "epoch": 0.1526241134751773, + "grad_norm": 4.126920223236084, + "loss": 5.7261, + "lr": 0.000994825174825175, + "step": 538, + "tokens_trained": 0.051141656 + }, + { + "epoch": 0.15319148936170213, + "grad_norm": 4.233098030090332, + "loss": 5.6903, + "lr": 0.0009945454545454546, + "step": 540, + "tokens_trained": 0.051331256 + }, + { + "epoch": 0.15375886524822696, + "grad_norm": 4.271973133087158, + "loss": 5.7293, + "lr": 0.0009942657342657344, + "step": 542, + "tokens_trained": 0.051522072 + }, + { + "epoch": 0.15432624113475177, + "grad_norm": 4.653008937835693, + "loss": 5.7133, + "lr": 0.000993986013986014, + "step": 544, + "tokens_trained": 0.051711624 + }, + { + "epoch": 0.1548936170212766, + "grad_norm": 4.192624092102051, + "loss": 5.6876, + "lr": 0.0009937062937062937, + "step": 546, + "tokens_trained": 0.051901744 + }, + { + "epoch": 0.1554609929078014, + "grad_norm": 5.497848033905029, + "loss": 5.7378, + "lr": 0.0009934265734265735, + "step": 548, + "tokens_trained": 0.052092872 + }, + { + "epoch": 0.15602836879432624, + "grad_norm": 4.350259780883789, + "loss": 5.6533, + "lr": 0.0009931468531468532, + "step": 550, + "tokens_trained": 0.052281768 + }, + { + "epoch": 0.15659574468085105, + "grad_norm": 4.515641689300537, + "loss": 5.7492, + "lr": 0.000992867132867133, + "step": 552, + "tokens_trained": 0.052471848 + }, + { + "epoch": 0.15716312056737589, + "grad_norm": 4.628066539764404, + "loss": 5.7113, + "lr": 0.0009925874125874127, + "step": 554, + "tokens_trained": 0.052660168 + }, + { + "epoch": 0.15773049645390072, + "grad_norm": 4.8322930335998535, + "loss": 5.6696, + "lr": 0.0009923076923076923, + "step": 556, + "tokens_trained": 0.05284776 + }, + { + "epoch": 0.15829787234042553, + "grad_norm": 3.999706506729126, + "loss": 5.7296, + "lr": 0.000992027972027972, + "step": 558, + "tokens_trained": 0.053037344 + }, + { + "epoch": 0.15886524822695036, + "grad_norm": 4.332971572875977, + "loss": 5.7362, + "lr": 0.0009917482517482518, + "step": 560, + "tokens_trained": 0.053228168 + }, + { + "epoch": 0.15943262411347517, + "grad_norm": 4.500301361083984, + "loss": 5.6982, + "lr": 0.0009914685314685314, + "step": 562, + "tokens_trained": 0.05341856 + }, + { + "epoch": 0.16, + "grad_norm": 4.721808910369873, + "loss": 5.7166, + "lr": 0.0009911888111888113, + "step": 564, + "tokens_trained": 0.053608824 + }, + { + "epoch": 0.1605673758865248, + "grad_norm": 5.265316009521484, + "loss": 5.7069, + "lr": 0.000990909090909091, + "step": 566, + "tokens_trained": 0.053799728 + }, + { + "epoch": 0.16113475177304964, + "grad_norm": 5.024131774902344, + "loss": 5.7113, + "lr": 0.0009906293706293705, + "step": 568, + "tokens_trained": 0.05398944 + }, + { + "epoch": 0.16170212765957448, + "grad_norm": 4.063276767730713, + "loss": 5.6251, + "lr": 0.0009903496503496504, + "step": 570, + "tokens_trained": 0.054176512 + }, + { + "epoch": 0.1622695035460993, + "grad_norm": 4.15974760055542, + "loss": 5.6912, + "lr": 0.00099006993006993, + "step": 572, + "tokens_trained": 0.054367072 + }, + { + "epoch": 0.16283687943262412, + "grad_norm": 4.338894844055176, + "loss": 5.6807, + "lr": 0.0009897902097902099, + "step": 574, + "tokens_trained": 0.054559184 + }, + { + "epoch": 0.16340425531914893, + "grad_norm": 5.535487174987793, + "loss": 5.6765, + "lr": 0.0009895104895104895, + "step": 576, + "tokens_trained": 0.054748904 + }, + { + "epoch": 0.16397163120567376, + "grad_norm": 4.379040241241455, + "loss": 5.6884, + "lr": 0.0009892307692307694, + "step": 578, + "tokens_trained": 0.054936136 + }, + { + "epoch": 0.16453900709219857, + "grad_norm": 4.746179103851318, + "loss": 5.6885, + "lr": 0.000988951048951049, + "step": 580, + "tokens_trained": 0.055125584 + }, + { + "epoch": 0.1651063829787234, + "grad_norm": 4.949806213378906, + "loss": 5.7061, + "lr": 0.0009886713286713286, + "step": 582, + "tokens_trained": 0.055314608 + }, + { + "epoch": 0.16567375886524824, + "grad_norm": 4.507448196411133, + "loss": 5.6339, + "lr": 0.0009883916083916085, + "step": 584, + "tokens_trained": 0.055503992 + }, + { + "epoch": 0.16624113475177305, + "grad_norm": 4.131013870239258, + "loss": 5.7122, + "lr": 0.0009881118881118881, + "step": 586, + "tokens_trained": 0.055693376 + }, + { + "epoch": 0.16680851063829788, + "grad_norm": 5.32897424697876, + "loss": 5.7192, + "lr": 0.000987832167832168, + "step": 588, + "tokens_trained": 0.05588452 + }, + { + "epoch": 0.1673758865248227, + "grad_norm": 4.166877746582031, + "loss": 5.6666, + "lr": 0.0009875524475524476, + "step": 590, + "tokens_trained": 0.056073936 + }, + { + "epoch": 0.16794326241134752, + "grad_norm": 4.393389701843262, + "loss": 5.6113, + "lr": 0.0009872727272727273, + "step": 592, + "tokens_trained": 0.056262224 + }, + { + "epoch": 0.16851063829787233, + "grad_norm": 4.466696739196777, + "loss": 5.6466, + "lr": 0.000986993006993007, + "step": 594, + "tokens_trained": 0.056454008 + }, + { + "epoch": 0.16907801418439716, + "grad_norm": 3.9413373470306396, + "loss": 5.6838, + "lr": 0.0009867132867132867, + "step": 596, + "tokens_trained": 0.05664444 + }, + { + "epoch": 0.169645390070922, + "grad_norm": 3.594649314880371, + "loss": 5.6684, + "lr": 0.0009864335664335664, + "step": 598, + "tokens_trained": 0.056833864 + }, + { + "epoch": 0.1702127659574468, + "grad_norm": 3.5969483852386475, + "loss": 5.6619, + "lr": 0.0009861538461538462, + "step": 600, + "tokens_trained": 0.05702332 + }, + { + "epoch": 0.17078014184397164, + "grad_norm": 3.845414638519287, + "loss": 5.5855, + "lr": 0.0009858741258741259, + "step": 602, + "tokens_trained": 0.057212776 + }, + { + "epoch": 0.17134751773049645, + "grad_norm": 3.9198834896087646, + "loss": 5.6551, + "lr": 0.0009855944055944055, + "step": 604, + "tokens_trained": 0.05740152 + }, + { + "epoch": 0.17191489361702128, + "grad_norm": 3.6764986515045166, + "loss": 5.6228, + "lr": 0.0009853146853146854, + "step": 606, + "tokens_trained": 0.057595616 + }, + { + "epoch": 0.1724822695035461, + "grad_norm": 3.8210043907165527, + "loss": 5.6557, + "lr": 0.000985034965034965, + "step": 608, + "tokens_trained": 0.057783968 + }, + { + "epoch": 0.17304964539007092, + "grad_norm": 3.893644094467163, + "loss": 5.6675, + "lr": 0.0009847552447552449, + "step": 610, + "tokens_trained": 0.057974832 + }, + { + "epoch": 0.17361702127659576, + "grad_norm": 3.280839681625366, + "loss": 5.6442, + "lr": 0.0009844755244755245, + "step": 612, + "tokens_trained": 0.058166272 + }, + { + "epoch": 0.17418439716312056, + "grad_norm": 3.4350404739379883, + "loss": 5.6555, + "lr": 0.0009841958041958043, + "step": 614, + "tokens_trained": 0.058356008 + }, + { + "epoch": 0.1747517730496454, + "grad_norm": 3.7700448036193848, + "loss": 5.6138, + "lr": 0.000983916083916084, + "step": 616, + "tokens_trained": 0.058546792 + }, + { + "epoch": 0.1753191489361702, + "grad_norm": 3.8182730674743652, + "loss": 5.6931, + "lr": 0.0009836363636363636, + "step": 618, + "tokens_trained": 0.058736296 + }, + { + "epoch": 0.17588652482269504, + "grad_norm": 3.9105372428894043, + "loss": 5.6431, + "lr": 0.0009833566433566435, + "step": 620, + "tokens_trained": 0.058927576 + }, + { + "epoch": 0.17645390070921985, + "grad_norm": 3.8897712230682373, + "loss": 5.6203, + "lr": 0.000983076923076923, + "step": 622, + "tokens_trained": 0.059118416 + }, + { + "epoch": 0.17702127659574468, + "grad_norm": 3.512194871902466, + "loss": 5.6292, + "lr": 0.000982797202797203, + "step": 624, + "tokens_trained": 0.059308568 + }, + { + "epoch": 0.1773049645390071, + "eval_loss": 5.630118370056152, + "eval_runtime": 21.1591, + "step": 625, + "tokens_trained": 0.059404056 + }, + { + "epoch": 0.17758865248226952, + "grad_norm": 2.990100383758545, + "loss": 5.622, + "lr": 0.0009825174825174826, + "step": 626, + "tokens_trained": 0.059499776 + }, + { + "epoch": 0.17815602836879432, + "grad_norm": 3.0487334728240967, + "loss": 5.6629, + "lr": 0.0009822377622377622, + "step": 628, + "tokens_trained": 0.059690208 + }, + { + "epoch": 0.17872340425531916, + "grad_norm": 3.6905510425567627, + "loss": 5.6345, + "lr": 0.0009819580419580419, + "step": 630, + "tokens_trained": 0.059881352 + }, + { + "epoch": 0.17929078014184396, + "grad_norm": 3.302255630493164, + "loss": 5.6733, + "lr": 0.0009816783216783217, + "step": 632, + "tokens_trained": 0.060071896 + }, + { + "epoch": 0.1798581560283688, + "grad_norm": 3.6833834648132324, + "loss": 5.5868, + "lr": 0.0009813986013986014, + "step": 634, + "tokens_trained": 0.060260504 + }, + { + "epoch": 0.1804255319148936, + "grad_norm": 3.1528804302215576, + "loss": 5.6128, + "lr": 0.0009811188811188812, + "step": 636, + "tokens_trained": 0.060450584 + }, + { + "epoch": 0.18099290780141844, + "grad_norm": 3.788860559463501, + "loss": 5.6235, + "lr": 0.0009808391608391608, + "step": 638, + "tokens_trained": 0.060640872 + }, + { + "epoch": 0.18156028368794327, + "grad_norm": 3.192462682723999, + "loss": 5.545, + "lr": 0.0009805594405594405, + "step": 640, + "tokens_trained": 0.060832776 + }, + { + "epoch": 0.18212765957446808, + "grad_norm": 3.505732774734497, + "loss": 5.5801, + "lr": 0.0009802797202797203, + "step": 642, + "tokens_trained": 0.06102204 + }, + { + "epoch": 0.18269503546099292, + "grad_norm": 3.9589102268218994, + "loss": 5.6091, + "lr": 0.00098, + "step": 644, + "tokens_trained": 0.061209744 + }, + { + "epoch": 0.18326241134751772, + "grad_norm": 3.4410059452056885, + "loss": 5.6279, + "lr": 0.0009797202797202798, + "step": 646, + "tokens_trained": 0.061400392 + }, + { + "epoch": 0.18382978723404256, + "grad_norm": 3.7746005058288574, + "loss": 5.6124, + "lr": 0.0009794405594405595, + "step": 648, + "tokens_trained": 0.061592232 + }, + { + "epoch": 0.18439716312056736, + "grad_norm": 3.75022292137146, + "loss": 5.5826, + "lr": 0.000979160839160839, + "step": 650, + "tokens_trained": 0.061781824 + }, + { + "epoch": 0.1849645390070922, + "grad_norm": 3.7629313468933105, + "loss": 5.555, + "lr": 0.000978881118881119, + "step": 652, + "tokens_trained": 0.061972744 + }, + { + "epoch": 0.18553191489361703, + "grad_norm": 4.5046820640563965, + "loss": 5.5972, + "lr": 0.0009786013986013986, + "step": 654, + "tokens_trained": 0.062163456 + }, + { + "epoch": 0.18609929078014184, + "grad_norm": 3.443138599395752, + "loss": 5.6061, + "lr": 0.0009783216783216782, + "step": 656, + "tokens_trained": 0.06235208 + }, + { + "epoch": 0.18666666666666668, + "grad_norm": 3.2661828994750977, + "loss": 5.5479, + "lr": 0.000978041958041958, + "step": 658, + "tokens_trained": 0.062544416 + }, + { + "epoch": 0.18723404255319148, + "grad_norm": 3.9571003913879395, + "loss": 5.6069, + "lr": 0.000977762237762238, + "step": 660, + "tokens_trained": 0.062733992 + }, + { + "epoch": 0.18780141843971632, + "grad_norm": 3.705880641937256, + "loss": 5.5915, + "lr": 0.0009774825174825176, + "step": 662, + "tokens_trained": 0.062922536 + }, + { + "epoch": 0.18836879432624112, + "grad_norm": 4.066433429718018, + "loss": 5.6031, + "lr": 0.0009772027972027972, + "step": 664, + "tokens_trained": 0.063114224 + }, + { + "epoch": 0.18893617021276596, + "grad_norm": 3.356651782989502, + "loss": 5.6045, + "lr": 0.0009769230769230768, + "step": 666, + "tokens_trained": 0.063304616 + }, + { + "epoch": 0.1895035460992908, + "grad_norm": 3.8084938526153564, + "loss": 5.6138, + "lr": 0.0009766433566433567, + "step": 668, + "tokens_trained": 0.06349476 + }, + { + "epoch": 0.1900709219858156, + "grad_norm": 4.282619953155518, + "loss": 5.5704, + "lr": 0.0009763636363636363, + "step": 670, + "tokens_trained": 0.063684848 + }, + { + "epoch": 0.19063829787234043, + "grad_norm": 3.045057773590088, + "loss": 5.6427, + "lr": 0.0009760839160839161, + "step": 672, + "tokens_trained": 0.063875192 + }, + { + "epoch": 0.19120567375886524, + "grad_norm": 3.360164165496826, + "loss": 5.5778, + "lr": 0.0009758041958041958, + "step": 674, + "tokens_trained": 0.06406636 + }, + { + "epoch": 0.19177304964539008, + "grad_norm": 3.5778472423553467, + "loss": 5.5389, + "lr": 0.0009755244755244756, + "step": 676, + "tokens_trained": 0.064254376 + }, + { + "epoch": 0.19234042553191488, + "grad_norm": 3.34869384765625, + "loss": 5.5894, + "lr": 0.0009752447552447553, + "step": 678, + "tokens_trained": 0.0644448 + }, + { + "epoch": 0.19290780141843972, + "grad_norm": 3.083582878112793, + "loss": 5.5776, + "lr": 0.0009749650349650349, + "step": 680, + "tokens_trained": 0.064633712 + }, + { + "epoch": 0.19347517730496455, + "grad_norm": 3.345973491668701, + "loss": 5.5987, + "lr": 0.0009746853146853148, + "step": 682, + "tokens_trained": 0.064824808 + }, + { + "epoch": 0.19404255319148936, + "grad_norm": 3.9262702465057373, + "loss": 5.64, + "lr": 0.0009744055944055944, + "step": 684, + "tokens_trained": 0.065016224 + }, + { + "epoch": 0.1946099290780142, + "grad_norm": 3.298543930053711, + "loss": 5.587, + "lr": 0.0009741258741258742, + "step": 686, + "tokens_trained": 0.065204216 + }, + { + "epoch": 0.195177304964539, + "grad_norm": 3.118626832962036, + "loss": 5.5864, + "lr": 0.0009738461538461538, + "step": 688, + "tokens_trained": 0.065393256 + }, + { + "epoch": 0.19574468085106383, + "grad_norm": 2.983548402786255, + "loss": 5.5506, + "lr": 0.0009735664335664336, + "step": 690, + "tokens_trained": 0.06558324 + }, + { + "epoch": 0.19631205673758864, + "grad_norm": 3.5204527378082275, + "loss": 5.5336, + "lr": 0.0009732867132867133, + "step": 692, + "tokens_trained": 0.065775624 + }, + { + "epoch": 0.19687943262411348, + "grad_norm": 3.138550281524658, + "loss": 5.5677, + "lr": 0.000973006993006993, + "step": 694, + "tokens_trained": 0.0659666 + }, + { + "epoch": 0.1974468085106383, + "grad_norm": 3.0961053371429443, + "loss": 5.5714, + "lr": 0.0009727272727272728, + "step": 696, + "tokens_trained": 0.066155512 + }, + { + "epoch": 0.19801418439716312, + "grad_norm": 3.4929685592651367, + "loss": 5.5829, + "lr": 0.0009724475524475524, + "step": 698, + "tokens_trained": 0.06634576 + }, + { + "epoch": 0.19858156028368795, + "grad_norm": 3.1820616722106934, + "loss": 5.6108, + "lr": 0.0009721678321678323, + "step": 700, + "tokens_trained": 0.066537016 + }, + { + "epoch": 0.19914893617021276, + "grad_norm": 3.4244654178619385, + "loss": 5.6025, + "lr": 0.0009718881118881119, + "step": 702, + "tokens_trained": 0.066727856 + }, + { + "epoch": 0.1997163120567376, + "grad_norm": 3.258605480194092, + "loss": 5.5581, + "lr": 0.0009716083916083917, + "step": 704, + "tokens_trained": 0.066916672 + }, + { + "epoch": 0.2002836879432624, + "grad_norm": 2.7159688472747803, + "loss": 5.5478, + "lr": 0.0009713286713286713, + "step": 706, + "tokens_trained": 0.067107704 + }, + { + "epoch": 0.20085106382978724, + "grad_norm": 3.1941912174224854, + "loss": 5.6126, + "lr": 0.000971048951048951, + "step": 708, + "tokens_trained": 0.067297896 + }, + { + "epoch": 0.20141843971631207, + "grad_norm": 3.20470929145813, + "loss": 5.5628, + "lr": 0.0009707692307692308, + "step": 710, + "tokens_trained": 0.06748608 + }, + { + "epoch": 0.20198581560283688, + "grad_norm": 3.6400153636932373, + "loss": 5.5758, + "lr": 0.0009704895104895105, + "step": 712, + "tokens_trained": 0.0676766 + }, + { + "epoch": 0.2025531914893617, + "grad_norm": 2.881639003753662, + "loss": 5.5512, + "lr": 0.0009702097902097903, + "step": 714, + "tokens_trained": 0.067865848 + }, + { + "epoch": 0.20312056737588652, + "grad_norm": 3.1113905906677246, + "loss": 5.5396, + "lr": 0.0009699300699300699, + "step": 716, + "tokens_trained": 0.068055368 + }, + { + "epoch": 0.20368794326241135, + "grad_norm": 3.135014772415161, + "loss": 5.5763, + "lr": 0.0009696503496503498, + "step": 718, + "tokens_trained": 0.068248544 + }, + { + "epoch": 0.20425531914893616, + "grad_norm": 3.1870718002319336, + "loss": 5.5903, + "lr": 0.0009693706293706294, + "step": 720, + "tokens_trained": 0.068436944 + }, + { + "epoch": 0.204822695035461, + "grad_norm": 3.125596523284912, + "loss": 5.6033, + "lr": 0.0009690909090909091, + "step": 722, + "tokens_trained": 0.06862548 + }, + { + "epoch": 0.20539007092198583, + "grad_norm": 2.897671699523926, + "loss": 5.5946, + "lr": 0.0009688111888111888, + "step": 724, + "tokens_trained": 0.068815232 + }, + { + "epoch": 0.20595744680851064, + "grad_norm": 2.855313539505005, + "loss": 5.5731, + "lr": 0.0009685314685314685, + "step": 726, + "tokens_trained": 0.06900692 + }, + { + "epoch": 0.20652482269503547, + "grad_norm": 2.7760672569274902, + "loss": 5.4949, + "lr": 0.0009682517482517483, + "step": 728, + "tokens_trained": 0.069195376 + }, + { + "epoch": 0.20709219858156028, + "grad_norm": 2.9300007820129395, + "loss": 5.5491, + "lr": 0.000967972027972028, + "step": 730, + "tokens_trained": 0.069385512 + }, + { + "epoch": 0.2076595744680851, + "grad_norm": 3.299860954284668, + "loss": 5.5405, + "lr": 0.0009676923076923078, + "step": 732, + "tokens_trained": 0.069573304 + }, + { + "epoch": 0.20822695035460992, + "grad_norm": 3.300189256668091, + "loss": 5.5797, + "lr": 0.0009674125874125874, + "step": 734, + "tokens_trained": 0.069764248 + }, + { + "epoch": 0.20879432624113475, + "grad_norm": 2.932995557785034, + "loss": 5.5556, + "lr": 0.0009671328671328672, + "step": 736, + "tokens_trained": 0.06995496 + }, + { + "epoch": 0.2093617021276596, + "grad_norm": 2.6711719036102295, + "loss": 5.48, + "lr": 0.0009668531468531469, + "step": 738, + "tokens_trained": 0.070142776 + }, + { + "epoch": 0.2099290780141844, + "grad_norm": 2.833314895629883, + "loss": 5.542, + "lr": 0.0009665734265734266, + "step": 740, + "tokens_trained": 0.070332064 + }, + { + "epoch": 0.21049645390070923, + "grad_norm": 2.899843215942383, + "loss": 5.5649, + "lr": 0.0009662937062937063, + "step": 742, + "tokens_trained": 0.070523448 + }, + { + "epoch": 0.21106382978723404, + "grad_norm": 2.96528697013855, + "loss": 5.5277, + "lr": 0.000966013986013986, + "step": 744, + "tokens_trained": 0.070713768 + }, + { + "epoch": 0.21163120567375887, + "grad_norm": 2.921109437942505, + "loss": 5.5646, + "lr": 0.0009657342657342657, + "step": 746, + "tokens_trained": 0.070905704 + }, + { + "epoch": 0.21219858156028368, + "grad_norm": 3.2725329399108887, + "loss": 5.4786, + "lr": 0.0009654545454545455, + "step": 748, + "tokens_trained": 0.071096008 + }, + { + "epoch": 0.2127659574468085, + "grad_norm": 2.8296804428100586, + "loss": 5.573, + "lr": 0.0009651748251748252, + "step": 750, + "tokens_trained": 0.07128828 + }, + { + "epoch": 0.2127659574468085, + "eval_loss": 5.535472869873047, + "eval_runtime": 21.0109, + "step": 750, + "tokens_trained": 0.07128828 + }, + { + "epoch": 0.21333333333333335, + "grad_norm": 3.0509591102600098, + "loss": 5.6037, + "lr": 0.0009648951048951049, + "step": 752, + "tokens_trained": 0.071479496 + }, + { + "epoch": 0.21390070921985815, + "grad_norm": 2.6773571968078613, + "loss": 5.5266, + "lr": 0.0009646153846153846, + "step": 754, + "tokens_trained": 0.071668568 + }, + { + "epoch": 0.214468085106383, + "grad_norm": 2.9600210189819336, + "loss": 5.5362, + "lr": 0.0009643356643356644, + "step": 756, + "tokens_trained": 0.071860552 + }, + { + "epoch": 0.2150354609929078, + "grad_norm": 2.6674885749816895, + "loss": 5.5388, + "lr": 0.0009640559440559441, + "step": 758, + "tokens_trained": 0.07204912 + }, + { + "epoch": 0.21560283687943263, + "grad_norm": 2.50179386138916, + "loss": 5.5027, + "lr": 0.0009637762237762237, + "step": 760, + "tokens_trained": 0.072239952 + }, + { + "epoch": 0.21617021276595744, + "grad_norm": 2.843411684036255, + "loss": 5.5221, + "lr": 0.0009634965034965035, + "step": 762, + "tokens_trained": 0.07243076 + }, + { + "epoch": 0.21673758865248227, + "grad_norm": 2.8686277866363525, + "loss": 5.4896, + "lr": 0.0009632167832167832, + "step": 764, + "tokens_trained": 0.072623272 + }, + { + "epoch": 0.2173049645390071, + "grad_norm": 2.611424684524536, + "loss": 5.5557, + "lr": 0.000962937062937063, + "step": 766, + "tokens_trained": 0.07281408 + }, + { + "epoch": 0.2178723404255319, + "grad_norm": 3.013145685195923, + "loss": 5.4964, + "lr": 0.0009626573426573427, + "step": 768, + "tokens_trained": 0.073005016 + }, + { + "epoch": 0.21843971631205675, + "grad_norm": 2.8682022094726562, + "loss": 5.5232, + "lr": 0.0009623776223776224, + "step": 770, + "tokens_trained": 0.07319652 + }, + { + "epoch": 0.21900709219858155, + "grad_norm": 2.6478466987609863, + "loss": 5.5517, + "lr": 0.0009620979020979021, + "step": 772, + "tokens_trained": 0.073387048 + }, + { + "epoch": 0.2195744680851064, + "grad_norm": 2.7273097038269043, + "loss": 5.5572, + "lr": 0.0009618181818181818, + "step": 774, + "tokens_trained": 0.073577424 + }, + { + "epoch": 0.2201418439716312, + "grad_norm": 3.104907751083374, + "loss": 5.5081, + "lr": 0.0009615384615384616, + "step": 776, + "tokens_trained": 0.073766712 + }, + { + "epoch": 0.22070921985815603, + "grad_norm": 2.9616432189941406, + "loss": 5.5059, + "lr": 0.0009612587412587412, + "step": 778, + "tokens_trained": 0.073956272 + }, + { + "epoch": 0.22127659574468084, + "grad_norm": 3.330319881439209, + "loss": 5.4811, + "lr": 0.000960979020979021, + "step": 780, + "tokens_trained": 0.074144008 + }, + { + "epoch": 0.22184397163120567, + "grad_norm": 2.964371919631958, + "loss": 5.4763, + "lr": 0.0009606993006993007, + "step": 782, + "tokens_trained": 0.074333888 + }, + { + "epoch": 0.2224113475177305, + "grad_norm": 3.13899827003479, + "loss": 5.5262, + "lr": 0.0009604195804195805, + "step": 784, + "tokens_trained": 0.074523584 + }, + { + "epoch": 0.2229787234042553, + "grad_norm": 3.2576637268066406, + "loss": 5.4983, + "lr": 0.0009601398601398602, + "step": 786, + "tokens_trained": 0.074714128 + }, + { + "epoch": 0.22354609929078015, + "grad_norm": 2.916149616241455, + "loss": 5.504, + "lr": 0.0009598601398601398, + "step": 788, + "tokens_trained": 0.074905104 + }, + { + "epoch": 0.22411347517730495, + "grad_norm": 2.842733144760132, + "loss": 5.4997, + "lr": 0.0009595804195804196, + "step": 790, + "tokens_trained": 0.075096328 + }, + { + "epoch": 0.2246808510638298, + "grad_norm": 2.880695104598999, + "loss": 5.5131, + "lr": 0.0009593006993006993, + "step": 792, + "tokens_trained": 0.075286104 + }, + { + "epoch": 0.2252482269503546, + "grad_norm": 2.620516300201416, + "loss": 5.5291, + "lr": 0.0009590209790209791, + "step": 794, + "tokens_trained": 0.075477392 + }, + { + "epoch": 0.22581560283687943, + "grad_norm": 2.622455358505249, + "loss": 5.5433, + "lr": 0.0009587412587412587, + "step": 796, + "tokens_trained": 0.0756682 + }, + { + "epoch": 0.22638297872340427, + "grad_norm": 2.532047986984253, + "loss": 5.5169, + "lr": 0.0009584615384615385, + "step": 798, + "tokens_trained": 0.075856528 + }, + { + "epoch": 0.22695035460992907, + "grad_norm": 2.628110885620117, + "loss": 5.5369, + "lr": 0.0009581818181818182, + "step": 800, + "tokens_trained": 0.076046256 + }, + { + "epoch": 0.2275177304964539, + "grad_norm": 2.376600980758667, + "loss": 5.4888, + "lr": 0.000957902097902098, + "step": 802, + "tokens_trained": 0.076236016 + }, + { + "epoch": 0.22808510638297871, + "grad_norm": 2.433666706085205, + "loss": 5.5044, + "lr": 0.0009576223776223777, + "step": 804, + "tokens_trained": 0.07642324 + }, + { + "epoch": 0.22865248226950355, + "grad_norm": 2.3850929737091064, + "loss": 5.4941, + "lr": 0.0009573426573426573, + "step": 806, + "tokens_trained": 0.07661376 + }, + { + "epoch": 0.22921985815602836, + "grad_norm": 2.4664969444274902, + "loss": 5.5257, + "lr": 0.0009570629370629371, + "step": 808, + "tokens_trained": 0.076804952 + }, + { + "epoch": 0.2297872340425532, + "grad_norm": 2.8514602184295654, + "loss": 5.5335, + "lr": 0.0009567832167832168, + "step": 810, + "tokens_trained": 0.076995064 + }, + { + "epoch": 0.23035460992907802, + "grad_norm": 2.508887767791748, + "loss": 5.5093, + "lr": 0.0009565034965034966, + "step": 812, + "tokens_trained": 0.077185344 + }, + { + "epoch": 0.23092198581560283, + "grad_norm": 2.5842514038085938, + "loss": 5.5246, + "lr": 0.0009562237762237762, + "step": 814, + "tokens_trained": 0.077375232 + }, + { + "epoch": 0.23148936170212767, + "grad_norm": 2.621562957763672, + "loss": 5.4948, + "lr": 0.0009559440559440559, + "step": 816, + "tokens_trained": 0.07756528 + }, + { + "epoch": 0.23205673758865247, + "grad_norm": 2.3230698108673096, + "loss": 5.5367, + "lr": 0.0009556643356643357, + "step": 818, + "tokens_trained": 0.077754936 + }, + { + "epoch": 0.2326241134751773, + "grad_norm": 2.728039264678955, + "loss": 5.4548, + "lr": 0.0009553846153846154, + "step": 820, + "tokens_trained": 0.077944056 + }, + { + "epoch": 0.23319148936170211, + "grad_norm": 2.786271333694458, + "loss": 5.4701, + "lr": 0.0009551048951048952, + "step": 822, + "tokens_trained": 0.07813272 + }, + { + "epoch": 0.23375886524822695, + "grad_norm": 2.449995517730713, + "loss": 5.5505, + "lr": 0.0009548251748251748, + "step": 824, + "tokens_trained": 0.078321888 + }, + { + "epoch": 0.23432624113475178, + "grad_norm": 2.394447088241577, + "loss": 5.4709, + "lr": 0.0009545454545454546, + "step": 826, + "tokens_trained": 0.078510288 + }, + { + "epoch": 0.2348936170212766, + "grad_norm": 2.5857675075531006, + "loss": 5.4986, + "lr": 0.0009542657342657343, + "step": 828, + "tokens_trained": 0.078698032 + }, + { + "epoch": 0.23546099290780143, + "grad_norm": 2.728743314743042, + "loss": 5.4983, + "lr": 0.000953986013986014, + "step": 830, + "tokens_trained": 0.078890608 + }, + { + "epoch": 0.23602836879432623, + "grad_norm": 2.3619866371154785, + "loss": 5.4985, + "lr": 0.0009537062937062937, + "step": 832, + "tokens_trained": 0.079081968 + }, + { + "epoch": 0.23659574468085107, + "grad_norm": 2.6265158653259277, + "loss": 5.5088, + "lr": 0.0009534265734265734, + "step": 834, + "tokens_trained": 0.079270712 + }, + { + "epoch": 0.23716312056737587, + "grad_norm": 2.3731281757354736, + "loss": 5.4682, + "lr": 0.0009531468531468532, + "step": 836, + "tokens_trained": 0.079459912 + }, + { + "epoch": 0.2377304964539007, + "grad_norm": 2.375283718109131, + "loss": 5.4278, + "lr": 0.0009528671328671329, + "step": 838, + "tokens_trained": 0.079649408 + }, + { + "epoch": 0.23829787234042554, + "grad_norm": 2.6856729984283447, + "loss": 5.5277, + "lr": 0.0009525874125874127, + "step": 840, + "tokens_trained": 0.079839552 + }, + { + "epoch": 0.23886524822695035, + "grad_norm": 2.5037410259246826, + "loss": 5.5022, + "lr": 0.0009523076923076923, + "step": 842, + "tokens_trained": 0.08002732 + }, + { + "epoch": 0.23943262411347518, + "grad_norm": 2.25175404548645, + "loss": 5.4918, + "lr": 0.000952027972027972, + "step": 844, + "tokens_trained": 0.080216416 + }, + { + "epoch": 0.24, + "grad_norm": 2.3555264472961426, + "loss": 5.5134, + "lr": 0.0009517482517482518, + "step": 846, + "tokens_trained": 0.080406928 + }, + { + "epoch": 0.24056737588652483, + "grad_norm": 2.390998601913452, + "loss": 5.4721, + "lr": 0.0009514685314685315, + "step": 848, + "tokens_trained": 0.080596232 + }, + { + "epoch": 0.24113475177304963, + "grad_norm": 2.1585986614227295, + "loss": 5.4511, + "lr": 0.0009511888111888112, + "step": 850, + "tokens_trained": 0.080786848 + }, + { + "epoch": 0.24170212765957447, + "grad_norm": 2.7733986377716064, + "loss": 5.5269, + "lr": 0.0009509090909090909, + "step": 852, + "tokens_trained": 0.080978144 + }, + { + "epoch": 0.2422695035460993, + "grad_norm": 2.8021209239959717, + "loss": 5.4751, + "lr": 0.0009506293706293707, + "step": 854, + "tokens_trained": 0.081167712 + }, + { + "epoch": 0.2428368794326241, + "grad_norm": 2.5434224605560303, + "loss": 5.5154, + "lr": 0.0009503496503496504, + "step": 856, + "tokens_trained": 0.081357584 + }, + { + "epoch": 0.24340425531914894, + "grad_norm": 2.456421136856079, + "loss": 5.5459, + "lr": 0.0009500699300699301, + "step": 858, + "tokens_trained": 0.081545992 + }, + { + "epoch": 0.24397163120567375, + "grad_norm": 2.317312002182007, + "loss": 5.4644, + "lr": 0.0009497902097902098, + "step": 860, + "tokens_trained": 0.081735392 + }, + { + "epoch": 0.24453900709219858, + "grad_norm": 2.3580780029296875, + "loss": 5.4359, + "lr": 0.0009495104895104895, + "step": 862, + "tokens_trained": 0.081925608 + }, + { + "epoch": 0.2451063829787234, + "grad_norm": 2.6440224647521973, + "loss": 5.4757, + "lr": 0.0009492307692307693, + "step": 864, + "tokens_trained": 0.08211328 + }, + { + "epoch": 0.24567375886524823, + "grad_norm": 2.5468132495880127, + "loss": 5.4115, + "lr": 0.000948951048951049, + "step": 866, + "tokens_trained": 0.082303736 + }, + { + "epoch": 0.24624113475177306, + "grad_norm": 2.431992530822754, + "loss": 5.4655, + "lr": 0.0009486713286713286, + "step": 868, + "tokens_trained": 0.082492896 + }, + { + "epoch": 0.24680851063829787, + "grad_norm": 2.443335771560669, + "loss": 5.4684, + "lr": 0.0009483916083916084, + "step": 870, + "tokens_trained": 0.082684024 + }, + { + "epoch": 0.2473758865248227, + "grad_norm": 2.6467180252075195, + "loss": 5.5017, + "lr": 0.0009481118881118881, + "step": 872, + "tokens_trained": 0.08287444 + }, + { + "epoch": 0.2479432624113475, + "grad_norm": 2.6044974327087402, + "loss": 5.4637, + "lr": 0.0009478321678321679, + "step": 874, + "tokens_trained": 0.08306436 + }, + { + "epoch": 0.24822695035460993, + "eval_loss": 5.4816508293151855, + "eval_runtime": 20.9467, + "step": 875, + "tokens_trained": 0.083158888 + }, + { + "epoch": 0.24851063829787234, + "grad_norm": 2.6221189498901367, + "loss": 5.4785, + "lr": 0.0009475524475524476, + "step": 876, + "tokens_trained": 0.083253472 + }, + { + "epoch": 0.24907801418439715, + "grad_norm": 2.409327983856201, + "loss": 5.42, + "lr": 0.0009472727272727273, + "step": 878, + "tokens_trained": 0.08344528 + }, + { + "epoch": 0.24964539007092199, + "grad_norm": 2.2504723072052, + "loss": 5.399, + "lr": 0.000946993006993007, + "step": 880, + "tokens_trained": 0.083635752 + }, + { + "epoch": 0.2502127659574468, + "grad_norm": 2.3018665313720703, + "loss": 5.4512, + "lr": 0.0009467132867132868, + "step": 882, + "tokens_trained": 0.08382576 + }, + { + "epoch": 0.2507801418439716, + "grad_norm": 2.5774636268615723, + "loss": 5.4592, + "lr": 0.0009464335664335665, + "step": 884, + "tokens_trained": 0.084016232 + }, + { + "epoch": 0.25134751773049646, + "grad_norm": 2.614935874938965, + "loss": 5.4772, + "lr": 0.0009461538461538461, + "step": 886, + "tokens_trained": 0.084206992 + }, + { + "epoch": 0.2519148936170213, + "grad_norm": 2.4281506538391113, + "loss": 5.4972, + "lr": 0.0009458741258741259, + "step": 888, + "tokens_trained": 0.084395848 + }, + { + "epoch": 0.2524822695035461, + "grad_norm": 2.3668100833892822, + "loss": 5.4505, + "lr": 0.0009455944055944056, + "step": 890, + "tokens_trained": 0.084583704 + }, + { + "epoch": 0.2530496453900709, + "grad_norm": 2.1937146186828613, + "loss": 5.4981, + "lr": 0.0009453146853146854, + "step": 892, + "tokens_trained": 0.08477096 + }, + { + "epoch": 0.25361702127659574, + "grad_norm": 2.2917556762695312, + "loss": 5.4224, + "lr": 0.000945034965034965, + "step": 894, + "tokens_trained": 0.084961048 + }, + { + "epoch": 0.2541843971631206, + "grad_norm": 2.1254703998565674, + "loss": 5.4409, + "lr": 0.0009447552447552447, + "step": 896, + "tokens_trained": 0.085153256 + }, + { + "epoch": 0.2547517730496454, + "grad_norm": 2.267159938812256, + "loss": 5.4527, + "lr": 0.0009444755244755245, + "step": 898, + "tokens_trained": 0.085343128 + }, + { + "epoch": 0.2553191489361702, + "grad_norm": 2.1975555419921875, + "loss": 5.516, + "lr": 0.0009441958041958042, + "step": 900, + "tokens_trained": 0.085534024 + }, + { + "epoch": 0.255886524822695, + "grad_norm": 2.3459436893463135, + "loss": 5.4592, + "lr": 0.000943916083916084, + "step": 902, + "tokens_trained": 0.085725136 + }, + { + "epoch": 0.25645390070921986, + "grad_norm": 2.4788501262664795, + "loss": 5.3937, + "lr": 0.0009436363636363636, + "step": 904, + "tokens_trained": 0.08591548 + }, + { + "epoch": 0.2570212765957447, + "grad_norm": 2.415065288543701, + "loss": 5.3991, + "lr": 0.0009433566433566434, + "step": 906, + "tokens_trained": 0.086105008 + }, + { + "epoch": 0.25758865248226953, + "grad_norm": 2.1260058879852295, + "loss": 5.4122, + "lr": 0.0009430769230769231, + "step": 908, + "tokens_trained": 0.08629424 + }, + { + "epoch": 0.2581560283687943, + "grad_norm": 2.1759092807769775, + "loss": 5.4663, + "lr": 0.0009427972027972029, + "step": 910, + "tokens_trained": 0.086485784 + }, + { + "epoch": 0.25872340425531914, + "grad_norm": 2.3481245040893555, + "loss": 5.4398, + "lr": 0.0009425174825174825, + "step": 912, + "tokens_trained": 0.086676744 + }, + { + "epoch": 0.259290780141844, + "grad_norm": 2.312612533569336, + "loss": 5.4615, + "lr": 0.0009422377622377622, + "step": 914, + "tokens_trained": 0.086866424 + }, + { + "epoch": 0.2598581560283688, + "grad_norm": 2.4709548950195312, + "loss": 5.4062, + "lr": 0.000941958041958042, + "step": 916, + "tokens_trained": 0.087055824 + }, + { + "epoch": 0.2604255319148936, + "grad_norm": 2.3664543628692627, + "loss": 5.4696, + "lr": 0.0009416783216783217, + "step": 918, + "tokens_trained": 0.087244136 + }, + { + "epoch": 0.26099290780141843, + "grad_norm": 2.423687696456909, + "loss": 5.4762, + "lr": 0.0009413986013986015, + "step": 920, + "tokens_trained": 0.087432584 + }, + { + "epoch": 0.26156028368794326, + "grad_norm": 2.4002890586853027, + "loss": 5.4743, + "lr": 0.0009411188811188811, + "step": 922, + "tokens_trained": 0.087622248 + }, + { + "epoch": 0.2621276595744681, + "grad_norm": 2.107527494430542, + "loss": 5.4013, + "lr": 0.0009408391608391608, + "step": 924, + "tokens_trained": 0.087809888 + }, + { + "epoch": 0.26269503546099293, + "grad_norm": 2.05177640914917, + "loss": 5.4601, + "lr": 0.0009405594405594406, + "step": 926, + "tokens_trained": 0.088002704 + }, + { + "epoch": 0.2632624113475177, + "grad_norm": 2.303874969482422, + "loss": 5.456, + "lr": 0.0009402797202797203, + "step": 928, + "tokens_trained": 0.088191344 + }, + { + "epoch": 0.26382978723404255, + "grad_norm": 2.4369659423828125, + "loss": 5.4162, + "lr": 0.00094, + "step": 930, + "tokens_trained": 0.088380832 + }, + { + "epoch": 0.2643971631205674, + "grad_norm": 2.4750819206237793, + "loss": 5.455, + "lr": 0.0009397202797202797, + "step": 932, + "tokens_trained": 0.088569936 + }, + { + "epoch": 0.2649645390070922, + "grad_norm": 2.09557843208313, + "loss": 5.4273, + "lr": 0.0009394405594405595, + "step": 934, + "tokens_trained": 0.08876116 + }, + { + "epoch": 0.265531914893617, + "grad_norm": 2.0984373092651367, + "loss": 5.4342, + "lr": 0.0009391608391608392, + "step": 936, + "tokens_trained": 0.088951032 + }, + { + "epoch": 0.26609929078014183, + "grad_norm": 2.1150097846984863, + "loss": 5.4344, + "lr": 0.000938881118881119, + "step": 938, + "tokens_trained": 0.08914124 + }, + { + "epoch": 0.26666666666666666, + "grad_norm": 2.1577563285827637, + "loss": 5.455, + "lr": 0.0009386013986013986, + "step": 940, + "tokens_trained": 0.089330952 + }, + { + "epoch": 0.2672340425531915, + "grad_norm": 2.0483016967773438, + "loss": 5.413, + "lr": 0.0009383216783216783, + "step": 942, + "tokens_trained": 0.08952116 + }, + { + "epoch": 0.26780141843971633, + "grad_norm": 2.3116559982299805, + "loss": 5.455, + "lr": 0.0009380419580419581, + "step": 944, + "tokens_trained": 0.089712888 + }, + { + "epoch": 0.2683687943262411, + "grad_norm": 2.2459256649017334, + "loss": 5.3971, + "lr": 0.0009377622377622378, + "step": 946, + "tokens_trained": 0.089903936 + }, + { + "epoch": 0.26893617021276595, + "grad_norm": 2.3048787117004395, + "loss": 5.4454, + "lr": 0.0009374825174825175, + "step": 948, + "tokens_trained": 0.090095888 + }, + { + "epoch": 0.2695035460992908, + "grad_norm": 2.196735143661499, + "loss": 5.4101, + "lr": 0.0009372027972027972, + "step": 950, + "tokens_trained": 0.090287472 + }, + { + "epoch": 0.2700709219858156, + "grad_norm": 2.3908562660217285, + "loss": 5.4731, + "lr": 0.0009369230769230769, + "step": 952, + "tokens_trained": 0.090476568 + }, + { + "epoch": 0.27063829787234045, + "grad_norm": 2.154932975769043, + "loss": 5.4104, + "lr": 0.0009366433566433567, + "step": 954, + "tokens_trained": 0.090665592 + }, + { + "epoch": 0.27120567375886523, + "grad_norm": 2.340907096862793, + "loss": 5.3707, + "lr": 0.0009363636363636364, + "step": 956, + "tokens_trained": 0.090853232 + }, + { + "epoch": 0.27177304964539006, + "grad_norm": 2.1736438274383545, + "loss": 5.4484, + "lr": 0.0009360839160839161, + "step": 958, + "tokens_trained": 0.091043808 + }, + { + "epoch": 0.2723404255319149, + "grad_norm": 2.3518154621124268, + "loss": 5.4919, + "lr": 0.0009358041958041958, + "step": 960, + "tokens_trained": 0.09123384 + }, + { + "epoch": 0.27290780141843973, + "grad_norm": 2.6673426628112793, + "loss": 5.4008, + "lr": 0.0009355244755244755, + "step": 962, + "tokens_trained": 0.091422544 + }, + { + "epoch": 0.2734751773049645, + "grad_norm": 2.4755311012268066, + "loss": 5.4533, + "lr": 0.0009352447552447553, + "step": 964, + "tokens_trained": 0.09161544 + }, + { + "epoch": 0.27404255319148935, + "grad_norm": 2.338452100753784, + "loss": 5.4953, + "lr": 0.0009349650349650349, + "step": 966, + "tokens_trained": 0.091806344 + }, + { + "epoch": 0.2746099290780142, + "grad_norm": 2.170426845550537, + "loss": 5.4588, + "lr": 0.0009346853146853147, + "step": 968, + "tokens_trained": 0.091996648 + }, + { + "epoch": 0.275177304964539, + "grad_norm": 2.2587599754333496, + "loss": 5.4547, + "lr": 0.0009344055944055944, + "step": 970, + "tokens_trained": 0.09218848 + }, + { + "epoch": 0.27574468085106385, + "grad_norm": 2.0009043216705322, + "loss": 5.4116, + "lr": 0.0009341258741258742, + "step": 972, + "tokens_trained": 0.092377984 + }, + { + "epoch": 0.27631205673758863, + "grad_norm": 2.0617294311523438, + "loss": 5.4541, + "lr": 0.0009338461538461539, + "step": 974, + "tokens_trained": 0.092569472 + }, + { + "epoch": 0.27687943262411346, + "grad_norm": 2.059300661087036, + "loss": 5.4414, + "lr": 0.0009335664335664336, + "step": 976, + "tokens_trained": 0.092758496 + }, + { + "epoch": 0.2774468085106383, + "grad_norm": 2.2815263271331787, + "loss": 5.4435, + "lr": 0.0009332867132867133, + "step": 978, + "tokens_trained": 0.092950368 + }, + { + "epoch": 0.27801418439716313, + "grad_norm": 2.1770365238189697, + "loss": 5.4481, + "lr": 0.0009330069930069929, + "step": 980, + "tokens_trained": 0.093140552 + }, + { + "epoch": 0.27858156028368797, + "grad_norm": 2.0089797973632812, + "loss": 5.4117, + "lr": 0.0009327272727272728, + "step": 982, + "tokens_trained": 0.093332312 + }, + { + "epoch": 0.27914893617021275, + "grad_norm": 2.2188286781311035, + "loss": 5.4594, + "lr": 0.0009324475524475524, + "step": 984, + "tokens_trained": 0.093520792 + }, + { + "epoch": 0.2797163120567376, + "grad_norm": 2.310481548309326, + "loss": 5.393, + "lr": 0.0009321678321678322, + "step": 986, + "tokens_trained": 0.093710608 + }, + { + "epoch": 0.2802836879432624, + "grad_norm": 2.3832972049713135, + "loss": 5.4277, + "lr": 0.0009318881118881119, + "step": 988, + "tokens_trained": 0.093900952 + }, + { + "epoch": 0.28085106382978725, + "grad_norm": 2.011126756668091, + "loss": 5.4097, + "lr": 0.0009316083916083917, + "step": 990, + "tokens_trained": 0.094091 + }, + { + "epoch": 0.28141843971631203, + "grad_norm": 2.2632968425750732, + "loss": 5.4388, + "lr": 0.0009313286713286714, + "step": 992, + "tokens_trained": 0.094281216 + }, + { + "epoch": 0.28198581560283686, + "grad_norm": 2.3477587699890137, + "loss": 5.3728, + "lr": 0.000931048951048951, + "step": 994, + "tokens_trained": 0.094470264 + }, + { + "epoch": 0.2825531914893617, + "grad_norm": 2.486196756362915, + "loss": 5.414, + "lr": 0.0009307692307692308, + "step": 996, + "tokens_trained": 0.094662816 + }, + { + "epoch": 0.28312056737588653, + "grad_norm": 2.5286316871643066, + "loss": 5.4063, + "lr": 0.0009304895104895104, + "step": 998, + "tokens_trained": 0.094852896 + }, + { + "epoch": 0.28368794326241137, + "grad_norm": 2.712406635284424, + "loss": 5.4274, + "lr": 0.0009302097902097903, + "step": 1000, + "tokens_trained": 0.09504336 + }, + { + "epoch": 0.28368794326241137, + "eval_loss": 5.434450626373291, + "eval_runtime": 21.0388, + "step": 1000, + "tokens_trained": 0.09504336 + }, + { + "epoch": 0.28425531914893615, + "grad_norm": 2.245316505432129, + "loss": 5.3551, + "lr": 0.0009299300699300699, + "step": 1002, + "tokens_trained": 0.095233944 + }, + { + "epoch": 0.284822695035461, + "grad_norm": 2.335533618927002, + "loss": 5.4608, + "lr": 0.0009296503496503497, + "step": 1004, + "tokens_trained": 0.095423184 + }, + { + "epoch": 0.2853900709219858, + "grad_norm": 2.232128858566284, + "loss": 5.4374, + "lr": 0.0009293706293706294, + "step": 1006, + "tokens_trained": 0.095612672 + }, + { + "epoch": 0.28595744680851065, + "grad_norm": 2.148329257965088, + "loss": 5.463, + "lr": 0.0009290909090909091, + "step": 1008, + "tokens_trained": 0.095802784 + }, + { + "epoch": 0.2865248226950355, + "grad_norm": 1.9580810070037842, + "loss": 5.291, + "lr": 0.0009288111888111889, + "step": 1010, + "tokens_trained": 0.095990776 + }, + { + "epoch": 0.28709219858156027, + "grad_norm": 1.9873988628387451, + "loss": 5.4103, + "lr": 0.0009285314685314685, + "step": 1012, + "tokens_trained": 0.096180648 + }, + { + "epoch": 0.2876595744680851, + "grad_norm": 2.0297746658325195, + "loss": 5.4078, + "lr": 0.0009282517482517483, + "step": 1014, + "tokens_trained": 0.09637224 + }, + { + "epoch": 0.28822695035460993, + "grad_norm": 1.928497076034546, + "loss": 5.3162, + "lr": 0.0009279720279720279, + "step": 1016, + "tokens_trained": 0.096561744 + }, + { + "epoch": 0.28879432624113477, + "grad_norm": 2.1219675540924072, + "loss": 5.4358, + "lr": 0.0009276923076923078, + "step": 1018, + "tokens_trained": 0.096752296 + }, + { + "epoch": 0.28936170212765955, + "grad_norm": 2.0021066665649414, + "loss": 5.4232, + "lr": 0.0009274125874125874, + "step": 1020, + "tokens_trained": 0.096943856 + }, + { + "epoch": 0.2899290780141844, + "grad_norm": 1.9920068979263306, + "loss": 5.407, + "lr": 0.0009271328671328671, + "step": 1022, + "tokens_trained": 0.097133632 + }, + { + "epoch": 0.2904964539007092, + "grad_norm": 1.8810361623764038, + "loss": 5.4293, + "lr": 0.0009268531468531469, + "step": 1024, + "tokens_trained": 0.097325976 + }, + { + "epoch": 0.29106382978723405, + "grad_norm": 1.8560134172439575, + "loss": 5.4236, + "lr": 0.0009265734265734266, + "step": 1026, + "tokens_trained": 0.0975142 + }, + { + "epoch": 0.2916312056737589, + "grad_norm": 2.1735010147094727, + "loss": 5.4252, + "lr": 0.0009262937062937064, + "step": 1028, + "tokens_trained": 0.0977042 + }, + { + "epoch": 0.29219858156028367, + "grad_norm": 2.2467288970947266, + "loss": 5.3756, + "lr": 0.000926013986013986, + "step": 1030, + "tokens_trained": 0.097893376 + }, + { + "epoch": 0.2927659574468085, + "grad_norm": 1.9609313011169434, + "loss": 5.4091, + "lr": 0.0009257342657342658, + "step": 1032, + "tokens_trained": 0.0980824 + }, + { + "epoch": 0.29333333333333333, + "grad_norm": 2.116384267807007, + "loss": 5.4001, + "lr": 0.0009254545454545454, + "step": 1034, + "tokens_trained": 0.098271304 + }, + { + "epoch": 0.29390070921985817, + "grad_norm": 2.1869800090789795, + "loss": 5.4102, + "lr": 0.0009251748251748252, + "step": 1036, + "tokens_trained": 0.098461528 + }, + { + "epoch": 0.294468085106383, + "grad_norm": 2.2882192134857178, + "loss": 5.4723, + "lr": 0.0009248951048951049, + "step": 1038, + "tokens_trained": 0.09865268 + }, + { + "epoch": 0.2950354609929078, + "grad_norm": 2.1590888500213623, + "loss": 5.3523, + "lr": 0.0009246153846153846, + "step": 1040, + "tokens_trained": 0.098842688 + }, + { + "epoch": 0.2956028368794326, + "grad_norm": 2.284207582473755, + "loss": 5.4647, + "lr": 0.0009243356643356644, + "step": 1042, + "tokens_trained": 0.099031544 + }, + { + "epoch": 0.29617021276595745, + "grad_norm": 2.333207845687866, + "loss": 5.4655, + "lr": 0.0009240559440559441, + "step": 1044, + "tokens_trained": 0.09922264 + }, + { + "epoch": 0.2967375886524823, + "grad_norm": 2.357572555541992, + "loss": 5.3909, + "lr": 0.0009237762237762239, + "step": 1046, + "tokens_trained": 0.099411416 + }, + { + "epoch": 0.29730496453900707, + "grad_norm": 1.88053297996521, + "loss": 5.4119, + "lr": 0.0009234965034965035, + "step": 1048, + "tokens_trained": 0.099602112 + }, + { + "epoch": 0.2978723404255319, + "grad_norm": 1.8860585689544678, + "loss": 5.3867, + "lr": 0.0009232167832167832, + "step": 1050, + "tokens_trained": 0.099792552 + }, + { + "epoch": 0.29843971631205674, + "grad_norm": 2.000173568725586, + "loss": 5.3773, + "lr": 0.0009229370629370629, + "step": 1052, + "tokens_trained": 0.099981752 + }, + { + "epoch": 0.29900709219858157, + "grad_norm": 2.015394926071167, + "loss": 5.3936, + "lr": 0.0009226573426573427, + "step": 1054, + "tokens_trained": 0.10017428 + }, + { + "epoch": 0.2995744680851064, + "grad_norm": 2.0050301551818848, + "loss": 5.3653, + "lr": 0.0009223776223776224, + "step": 1056, + "tokens_trained": 0.100364544 + }, + { + "epoch": 0.3001418439716312, + "grad_norm": 1.7397475242614746, + "loss": 5.3224, + "lr": 0.0009220979020979021, + "step": 1058, + "tokens_trained": 0.100555648 + }, + { + "epoch": 0.300709219858156, + "grad_norm": 1.9808533191680908, + "loss": 5.3822, + "lr": 0.0009218181818181819, + "step": 1060, + "tokens_trained": 0.100744968 + }, + { + "epoch": 0.30127659574468085, + "grad_norm": 2.034329652786255, + "loss": 5.3961, + "lr": 0.0009215384615384616, + "step": 1062, + "tokens_trained": 0.100934176 + }, + { + "epoch": 0.3018439716312057, + "grad_norm": 2.1286778450012207, + "loss": 5.4061, + "lr": 0.0009212587412587413, + "step": 1064, + "tokens_trained": 0.101125216 + }, + { + "epoch": 0.3024113475177305, + "grad_norm": 2.131822347640991, + "loss": 5.3675, + "lr": 0.000920979020979021, + "step": 1066, + "tokens_trained": 0.101314504 + }, + { + "epoch": 0.3029787234042553, + "grad_norm": 2.162069320678711, + "loss": 5.4552, + "lr": 0.0009206993006993007, + "step": 1068, + "tokens_trained": 0.101503352 + }, + { + "epoch": 0.30354609929078014, + "grad_norm": 2.5730931758880615, + "loss": 5.3978, + "lr": 0.0009204195804195804, + "step": 1070, + "tokens_trained": 0.101691504 + }, + { + "epoch": 0.30411347517730497, + "grad_norm": 2.2053022384643555, + "loss": 5.3604, + "lr": 0.0009201398601398602, + "step": 1072, + "tokens_trained": 0.101883072 + }, + { + "epoch": 0.3046808510638298, + "grad_norm": 2.1578407287597656, + "loss": 5.4236, + "lr": 0.0009198601398601398, + "step": 1074, + "tokens_trained": 0.102075832 + }, + { + "epoch": 0.3052482269503546, + "grad_norm": 2.0061423778533936, + "loss": 5.3882, + "lr": 0.0009195804195804196, + "step": 1076, + "tokens_trained": 0.102266768 + }, + { + "epoch": 0.3058156028368794, + "grad_norm": 1.8915576934814453, + "loss": 5.3539, + "lr": 0.0009193006993006993, + "step": 1078, + "tokens_trained": 0.102457096 + }, + { + "epoch": 0.30638297872340425, + "grad_norm": 2.15053129196167, + "loss": 5.4222, + "lr": 0.0009190209790209791, + "step": 1080, + "tokens_trained": 0.102647544 + }, + { + "epoch": 0.3069503546099291, + "grad_norm": 2.241217851638794, + "loss": 5.4275, + "lr": 0.0009187412587412588, + "step": 1082, + "tokens_trained": 0.10283904 + }, + { + "epoch": 0.3075177304964539, + "grad_norm": 2.37854266166687, + "loss": 5.419, + "lr": 0.0009184615384615385, + "step": 1084, + "tokens_trained": 0.103028464 + }, + { + "epoch": 0.3080851063829787, + "grad_norm": 2.00118350982666, + "loss": 5.4225, + "lr": 0.0009181818181818182, + "step": 1086, + "tokens_trained": 0.10321896 + }, + { + "epoch": 0.30865248226950354, + "grad_norm": 2.2643723487854004, + "loss": 5.4487, + "lr": 0.0009179020979020978, + "step": 1088, + "tokens_trained": 0.103409256 + }, + { + "epoch": 0.30921985815602837, + "grad_norm": 2.4618585109710693, + "loss": 5.4211, + "lr": 0.0009176223776223777, + "step": 1090, + "tokens_trained": 0.103597272 + }, + { + "epoch": 0.3097872340425532, + "grad_norm": 2.141491174697876, + "loss": 5.3758, + "lr": 0.0009173426573426573, + "step": 1092, + "tokens_trained": 0.103786128 + }, + { + "epoch": 0.31035460992907804, + "grad_norm": 1.9777475595474243, + "loss": 5.4129, + "lr": 0.0009170629370629371, + "step": 1094, + "tokens_trained": 0.103974864 + }, + { + "epoch": 0.3109219858156028, + "grad_norm": 1.9153270721435547, + "loss": 5.3912, + "lr": 0.0009167832167832168, + "step": 1096, + "tokens_trained": 0.104163864 + }, + { + "epoch": 0.31148936170212765, + "grad_norm": 2.172558546066284, + "loss": 5.3425, + "lr": 0.0009165034965034966, + "step": 1098, + "tokens_trained": 0.104353136 + }, + { + "epoch": 0.3120567375886525, + "grad_norm": 2.049896717071533, + "loss": 5.3732, + "lr": 0.0009162237762237763, + "step": 1100, + "tokens_trained": 0.10454476 + }, + { + "epoch": 0.3126241134751773, + "grad_norm": 1.9415545463562012, + "loss": 5.3873, + "lr": 0.0009159440559440559, + "step": 1102, + "tokens_trained": 0.104734296 + }, + { + "epoch": 0.3131914893617021, + "grad_norm": 1.7280856370925903, + "loss": 5.3857, + "lr": 0.0009156643356643357, + "step": 1104, + "tokens_trained": 0.104925648 + }, + { + "epoch": 0.31375886524822694, + "grad_norm": 1.9120069742202759, + "loss": 5.3216, + "lr": 0.0009153846153846153, + "step": 1106, + "tokens_trained": 0.105115776 + }, + { + "epoch": 0.31432624113475177, + "grad_norm": 2.007101058959961, + "loss": 5.4074, + "lr": 0.0009151048951048952, + "step": 1108, + "tokens_trained": 0.105305656 + }, + { + "epoch": 0.3148936170212766, + "grad_norm": 1.9159268140792847, + "loss": 5.3625, + "lr": 0.0009148251748251748, + "step": 1110, + "tokens_trained": 0.105494632 + }, + { + "epoch": 0.31546099290780144, + "grad_norm": 1.9235239028930664, + "loss": 5.3362, + "lr": 0.0009145454545454546, + "step": 1112, + "tokens_trained": 0.105683536 + }, + { + "epoch": 0.3160283687943262, + "grad_norm": 1.8954299688339233, + "loss": 5.3531, + "lr": 0.0009142657342657343, + "step": 1114, + "tokens_trained": 0.105873176 + }, + { + "epoch": 0.31659574468085105, + "grad_norm": 2.026578426361084, + "loss": 5.408, + "lr": 0.000913986013986014, + "step": 1116, + "tokens_trained": 0.10606276 + }, + { + "epoch": 0.3171631205673759, + "grad_norm": 1.9014806747436523, + "loss": 5.363, + "lr": 0.0009137062937062938, + "step": 1118, + "tokens_trained": 0.106254616 + }, + { + "epoch": 0.3177304964539007, + "grad_norm": 1.849649429321289, + "loss": 5.3811, + "lr": 0.0009134265734265734, + "step": 1120, + "tokens_trained": 0.106445376 + }, + { + "epoch": 0.31829787234042556, + "grad_norm": 1.7405186891555786, + "loss": 5.3504, + "lr": 0.0009131468531468532, + "step": 1122, + "tokens_trained": 0.106636072 + }, + { + "epoch": 0.31886524822695034, + "grad_norm": 1.867285966873169, + "loss": 5.3675, + "lr": 0.0009128671328671328, + "step": 1124, + "tokens_trained": 0.106827896 + }, + { + "epoch": 0.3191489361702128, + "eval_loss": 5.3796281814575195, + "eval_runtime": 20.7444, + "step": 1125, + "tokens_trained": 0.106922416 + }, + { + "epoch": 0.31943262411347517, + "grad_norm": 1.8044356107711792, + "loss": 5.3717, + "lr": 0.0009125874125874127, + "step": 1126, + "tokens_trained": 0.107016056 + }, + { + "epoch": 0.32, + "grad_norm": 1.6348600387573242, + "loss": 5.4676, + "lr": 0.0009123076923076923, + "step": 1128, + "tokens_trained": 0.107203992 + }, + { + "epoch": 0.32056737588652484, + "grad_norm": 1.7802475690841675, + "loss": 5.3304, + "lr": 0.000912027972027972, + "step": 1130, + "tokens_trained": 0.107394224 + }, + { + "epoch": 0.3211347517730496, + "grad_norm": 1.7695430517196655, + "loss": 5.3611, + "lr": 0.0009117482517482518, + "step": 1132, + "tokens_trained": 0.107583464 + }, + { + "epoch": 0.32170212765957445, + "grad_norm": 2.026853322982788, + "loss": 5.363, + "lr": 0.0009114685314685315, + "step": 1134, + "tokens_trained": 0.107776088 + }, + { + "epoch": 0.3222695035460993, + "grad_norm": 1.803208589553833, + "loss": 5.3801, + "lr": 0.0009111888111888113, + "step": 1136, + "tokens_trained": 0.107964416 + }, + { + "epoch": 0.3228368794326241, + "grad_norm": 1.812386155128479, + "loss": 5.3721, + "lr": 0.0009109090909090909, + "step": 1138, + "tokens_trained": 0.108153104 + }, + { + "epoch": 0.32340425531914896, + "grad_norm": 1.605839490890503, + "loss": 5.3339, + "lr": 0.0009106293706293707, + "step": 1140, + "tokens_trained": 0.108341408 + }, + { + "epoch": 0.32397163120567374, + "grad_norm": 1.7169313430786133, + "loss": 5.4051, + "lr": 0.0009103496503496503, + "step": 1142, + "tokens_trained": 0.108532312 + }, + { + "epoch": 0.3245390070921986, + "grad_norm": 2.0499444007873535, + "loss": 5.2992, + "lr": 0.0009100699300699301, + "step": 1144, + "tokens_trained": 0.108721864 + }, + { + "epoch": 0.3251063829787234, + "grad_norm": 1.988674521446228, + "loss": 5.3862, + "lr": 0.0009097902097902098, + "step": 1146, + "tokens_trained": 0.108912352 + }, + { + "epoch": 0.32567375886524824, + "grad_norm": 1.8733936548233032, + "loss": 5.3627, + "lr": 0.0009095104895104895, + "step": 1148, + "tokens_trained": 0.109101952 + }, + { + "epoch": 0.3262411347517731, + "grad_norm": 1.978704810142517, + "loss": 5.3668, + "lr": 0.0009092307692307692, + "step": 1150, + "tokens_trained": 0.109292712 + }, + { + "epoch": 0.32680851063829786, + "grad_norm": 1.9723341464996338, + "loss": 5.3545, + "lr": 0.000908951048951049, + "step": 1152, + "tokens_trained": 0.109484992 + }, + { + "epoch": 0.3273758865248227, + "grad_norm": 2.165728807449341, + "loss": 5.3731, + "lr": 0.0009086713286713288, + "step": 1154, + "tokens_trained": 0.109674336 + }, + { + "epoch": 0.3279432624113475, + "grad_norm": 1.9241019487380981, + "loss": 5.3456, + "lr": 0.0009083916083916084, + "step": 1156, + "tokens_trained": 0.109863368 + }, + { + "epoch": 0.32851063829787236, + "grad_norm": 1.9442275762557983, + "loss": 5.4065, + "lr": 0.0009081118881118881, + "step": 1158, + "tokens_trained": 0.110051744 + }, + { + "epoch": 0.32907801418439714, + "grad_norm": 1.7714000940322876, + "loss": 5.3888, + "lr": 0.0009078321678321678, + "step": 1160, + "tokens_trained": 0.11024344 + }, + { + "epoch": 0.329645390070922, + "grad_norm": 2.043646812438965, + "loss": 5.3835, + "lr": 0.0009075524475524476, + "step": 1162, + "tokens_trained": 0.11043488 + }, + { + "epoch": 0.3302127659574468, + "grad_norm": 1.837196946144104, + "loss": 5.3554, + "lr": 0.0009072727272727273, + "step": 1164, + "tokens_trained": 0.110626104 + }, + { + "epoch": 0.33078014184397164, + "grad_norm": 1.874135971069336, + "loss": 5.3457, + "lr": 0.000906993006993007, + "step": 1166, + "tokens_trained": 0.110814768 + }, + { + "epoch": 0.3313475177304965, + "grad_norm": 1.6493511199951172, + "loss": 5.3118, + "lr": 0.0009067132867132866, + "step": 1168, + "tokens_trained": 0.111004104 + }, + { + "epoch": 0.33191489361702126, + "grad_norm": 1.8386362791061401, + "loss": 5.3422, + "lr": 0.0009064335664335665, + "step": 1170, + "tokens_trained": 0.11119544 + }, + { + "epoch": 0.3324822695035461, + "grad_norm": 2.020859718322754, + "loss": 5.3565, + "lr": 0.0009061538461538462, + "step": 1172, + "tokens_trained": 0.111384384 + }, + { + "epoch": 0.3330496453900709, + "grad_norm": 2.049401044845581, + "loss": 5.3358, + "lr": 0.0009058741258741259, + "step": 1174, + "tokens_trained": 0.111573944 + }, + { + "epoch": 0.33361702127659576, + "grad_norm": 1.965345025062561, + "loss": 5.3431, + "lr": 0.0009055944055944056, + "step": 1176, + "tokens_trained": 0.111763504 + }, + { + "epoch": 0.3341843971631206, + "grad_norm": 1.9792066812515259, + "loss": 5.3579, + "lr": 0.0009053146853146853, + "step": 1178, + "tokens_trained": 0.111953664 + }, + { + "epoch": 0.3347517730496454, + "grad_norm": 1.7790883779525757, + "loss": 5.3499, + "lr": 0.0009050349650349651, + "step": 1180, + "tokens_trained": 0.11214324 + }, + { + "epoch": 0.3353191489361702, + "grad_norm": 1.6504682302474976, + "loss": 5.3415, + "lr": 0.0009047552447552448, + "step": 1182, + "tokens_trained": 0.112331256 + }, + { + "epoch": 0.33588652482269504, + "grad_norm": 1.9687312841415405, + "loss": 5.3565, + "lr": 0.0009044755244755245, + "step": 1184, + "tokens_trained": 0.11252208 + }, + { + "epoch": 0.3364539007092199, + "grad_norm": 1.7077507972717285, + "loss": 5.3568, + "lr": 0.0009041958041958041, + "step": 1186, + "tokens_trained": 0.112714272 + }, + { + "epoch": 0.33702127659574466, + "grad_norm": 1.6311697959899902, + "loss": 5.3345, + "lr": 0.000903916083916084, + "step": 1188, + "tokens_trained": 0.11290428 + }, + { + "epoch": 0.3375886524822695, + "grad_norm": 1.975233793258667, + "loss": 5.4161, + "lr": 0.0009036363636363637, + "step": 1190, + "tokens_trained": 0.113093984 + }, + { + "epoch": 0.3381560283687943, + "grad_norm": 1.7567362785339355, + "loss": 5.3481, + "lr": 0.0009033566433566434, + "step": 1192, + "tokens_trained": 0.113284904 + }, + { + "epoch": 0.33872340425531916, + "grad_norm": 2.121367931365967, + "loss": 5.3729, + "lr": 0.0009030769230769231, + "step": 1194, + "tokens_trained": 0.113477952 + }, + { + "epoch": 0.339290780141844, + "grad_norm": 2.143253803253174, + "loss": 5.3866, + "lr": 0.0009027972027972027, + "step": 1196, + "tokens_trained": 0.11366872 + }, + { + "epoch": 0.3398581560283688, + "grad_norm": 2.1118557453155518, + "loss": 5.3501, + "lr": 0.0009025174825174826, + "step": 1198, + "tokens_trained": 0.113861552 + }, + { + "epoch": 0.3404255319148936, + "grad_norm": 1.8132637739181519, + "loss": 5.3325, + "lr": 0.0009022377622377622, + "step": 1200, + "tokens_trained": 0.114051176 + }, + { + "epoch": 0.34099290780141844, + "grad_norm": 1.761227011680603, + "loss": 5.3629, + "lr": 0.000901958041958042, + "step": 1202, + "tokens_trained": 0.114240808 + }, + { + "epoch": 0.3415602836879433, + "grad_norm": 1.8358371257781982, + "loss": 5.3644, + "lr": 0.0009016783216783216, + "step": 1204, + "tokens_trained": 0.114430968 + }, + { + "epoch": 0.3421276595744681, + "grad_norm": 2.0768542289733887, + "loss": 5.3705, + "lr": 0.0009013986013986014, + "step": 1206, + "tokens_trained": 0.114620544 + }, + { + "epoch": 0.3426950354609929, + "grad_norm": 1.6928143501281738, + "loss": 5.2534, + "lr": 0.0009011188811188812, + "step": 1208, + "tokens_trained": 0.114811928 + }, + { + "epoch": 0.3432624113475177, + "grad_norm": 1.8634029626846313, + "loss": 5.3679, + "lr": 0.0009008391608391609, + "step": 1210, + "tokens_trained": 0.115002912 + }, + { + "epoch": 0.34382978723404256, + "grad_norm": 1.8048954010009766, + "loss": 5.3049, + "lr": 0.0009005594405594406, + "step": 1212, + "tokens_trained": 0.115192544 + }, + { + "epoch": 0.3443971631205674, + "grad_norm": 1.9170348644256592, + "loss": 5.2457, + "lr": 0.0009002797202797202, + "step": 1214, + "tokens_trained": 0.115383248 + }, + { + "epoch": 0.3449645390070922, + "grad_norm": 1.788751482963562, + "loss": 5.3678, + "lr": 0.0009000000000000001, + "step": 1216, + "tokens_trained": 0.115574304 + }, + { + "epoch": 0.345531914893617, + "grad_norm": 1.9751293659210205, + "loss": 5.3352, + "lr": 0.0008997202797202797, + "step": 1218, + "tokens_trained": 0.115766008 + }, + { + "epoch": 0.34609929078014184, + "grad_norm": 1.8202649354934692, + "loss": 5.37, + "lr": 0.0008994405594405595, + "step": 1220, + "tokens_trained": 0.11595804 + }, + { + "epoch": 0.3466666666666667, + "grad_norm": 1.656063199043274, + "loss": 5.3664, + "lr": 0.0008991608391608391, + "step": 1222, + "tokens_trained": 0.116146344 + }, + { + "epoch": 0.3472340425531915, + "grad_norm": 1.7509667873382568, + "loss": 5.3636, + "lr": 0.0008988811188811188, + "step": 1224, + "tokens_trained": 0.116334568 + }, + { + "epoch": 0.3478014184397163, + "grad_norm": 1.8556638956069946, + "loss": 5.3577, + "lr": 0.0008986013986013987, + "step": 1226, + "tokens_trained": 0.116525704 + }, + { + "epoch": 0.3483687943262411, + "grad_norm": 2.026033639907837, + "loss": 5.3657, + "lr": 0.0008983216783216783, + "step": 1228, + "tokens_trained": 0.116716032 + }, + { + "epoch": 0.34893617021276596, + "grad_norm": 1.6965924501419067, + "loss": 5.304, + "lr": 0.0008980419580419581, + "step": 1230, + "tokens_trained": 0.116904832 + }, + { + "epoch": 0.3495035460992908, + "grad_norm": 1.8144174814224243, + "loss": 5.3759, + "lr": 0.0008977622377622377, + "step": 1232, + "tokens_trained": 0.117095536 + }, + { + "epoch": 0.3500709219858156, + "grad_norm": 1.7229580879211426, + "loss": 5.3244, + "lr": 0.0008974825174825176, + "step": 1234, + "tokens_trained": 0.117285952 + }, + { + "epoch": 0.3506382978723404, + "grad_norm": 1.722578525543213, + "loss": 5.3442, + "lr": 0.0008972027972027972, + "step": 1236, + "tokens_trained": 0.117477488 + }, + { + "epoch": 0.35120567375886524, + "grad_norm": 1.8006796836853027, + "loss": 5.3624, + "lr": 0.000896923076923077, + "step": 1238, + "tokens_trained": 0.117667352 + }, + { + "epoch": 0.3517730496453901, + "grad_norm": 1.7172250747680664, + "loss": 5.3002, + "lr": 0.0008966433566433566, + "step": 1240, + "tokens_trained": 0.117856504 + }, + { + "epoch": 0.3523404255319149, + "grad_norm": 1.8281760215759277, + "loss": 5.3311, + "lr": 0.0008963636363636363, + "step": 1242, + "tokens_trained": 0.11804676 + }, + { + "epoch": 0.3529078014184397, + "grad_norm": 1.7666652202606201, + "loss": 5.3847, + "lr": 0.0008960839160839162, + "step": 1244, + "tokens_trained": 0.118235688 + }, + { + "epoch": 0.3534751773049645, + "grad_norm": 1.7723621129989624, + "loss": 5.3506, + "lr": 0.0008958041958041958, + "step": 1246, + "tokens_trained": 0.11842632 + }, + { + "epoch": 0.35404255319148936, + "grad_norm": 1.7779643535614014, + "loss": 5.3066, + "lr": 0.0008955244755244756, + "step": 1248, + "tokens_trained": 0.118616536 + }, + { + "epoch": 0.3546099290780142, + "grad_norm": 1.746245265007019, + "loss": 5.2993, + "lr": 0.0008952447552447552, + "step": 1250, + "tokens_trained": 0.118807672 + }, + { + "epoch": 0.3546099290780142, + "eval_loss": 5.34489107131958, + "eval_runtime": 21.0838, + "step": 1250, + "tokens_trained": 0.118807672 + }, + { + "epoch": 0.35517730496453903, + "grad_norm": 1.8439521789550781, + "loss": 5.3796, + "lr": 0.000894965034965035, + "step": 1252, + "tokens_trained": 0.118996672 + }, + { + "epoch": 0.3557446808510638, + "grad_norm": 1.7830157279968262, + "loss": 5.3435, + "lr": 0.0008946853146853147, + "step": 1254, + "tokens_trained": 0.119189544 + }, + { + "epoch": 0.35631205673758864, + "grad_norm": 1.6022379398345947, + "loss": 5.3772, + "lr": 0.0008944055944055944, + "step": 1256, + "tokens_trained": 0.119379312 + }, + { + "epoch": 0.3568794326241135, + "grad_norm": 1.6100343465805054, + "loss": 5.3411, + "lr": 0.0008941258741258741, + "step": 1258, + "tokens_trained": 0.119572072 + }, + { + "epoch": 0.3574468085106383, + "grad_norm": 1.7826210260391235, + "loss": 5.317, + "lr": 0.0008938461538461538, + "step": 1260, + "tokens_trained": 0.119761304 + }, + { + "epoch": 0.3580141843971631, + "grad_norm": 1.510432243347168, + "loss": 5.4018, + "lr": 0.0008935664335664337, + "step": 1262, + "tokens_trained": 0.11994984 + }, + { + "epoch": 0.35858156028368793, + "grad_norm": 1.7209227085113525, + "loss": 5.3651, + "lr": 0.0008932867132867133, + "step": 1264, + "tokens_trained": 0.120139368 + }, + { + "epoch": 0.35914893617021276, + "grad_norm": 1.7528654336929321, + "loss": 5.3329, + "lr": 0.000893006993006993, + "step": 1266, + "tokens_trained": 0.1203308 + }, + { + "epoch": 0.3597163120567376, + "grad_norm": 1.8427083492279053, + "loss": 5.3897, + "lr": 0.0008927272727272727, + "step": 1268, + "tokens_trained": 0.12052112 + }, + { + "epoch": 0.36028368794326243, + "grad_norm": 1.530527114868164, + "loss": 5.3407, + "lr": 0.0008924475524475525, + "step": 1270, + "tokens_trained": 0.120709456 + }, + { + "epoch": 0.3608510638297872, + "grad_norm": 1.5996145009994507, + "loss": 5.3697, + "lr": 0.0008921678321678322, + "step": 1272, + "tokens_trained": 0.12089976 + }, + { + "epoch": 0.36141843971631205, + "grad_norm": 1.5235425233840942, + "loss": 5.335, + "lr": 0.0008918881118881119, + "step": 1274, + "tokens_trained": 0.121089184 + }, + { + "epoch": 0.3619858156028369, + "grad_norm": 1.757206678390503, + "loss": 5.2983, + "lr": 0.0008916083916083916, + "step": 1276, + "tokens_trained": 0.1212798 + }, + { + "epoch": 0.3625531914893617, + "grad_norm": 1.5952467918395996, + "loss": 5.3593, + "lr": 0.0008913286713286713, + "step": 1278, + "tokens_trained": 0.121472816 + }, + { + "epoch": 0.36312056737588655, + "grad_norm": 1.6975666284561157, + "loss": 5.3867, + "lr": 0.0008910489510489512, + "step": 1280, + "tokens_trained": 0.121659944 + }, + { + "epoch": 0.36368794326241133, + "grad_norm": 1.8659151792526245, + "loss": 5.3032, + "lr": 0.0008907692307692308, + "step": 1282, + "tokens_trained": 0.121848552 + }, + { + "epoch": 0.36425531914893616, + "grad_norm": 1.8692409992218018, + "loss": 5.3643, + "lr": 0.0008904895104895105, + "step": 1284, + "tokens_trained": 0.12203916 + }, + { + "epoch": 0.364822695035461, + "grad_norm": 1.786490559577942, + "loss": 5.4001, + "lr": 0.0008902097902097902, + "step": 1286, + "tokens_trained": 0.122228464 + }, + { + "epoch": 0.36539007092198583, + "grad_norm": 1.6635786294937134, + "loss": 5.3158, + "lr": 0.00088993006993007, + "step": 1288, + "tokens_trained": 0.122419768 + }, + { + "epoch": 0.3659574468085106, + "grad_norm": 1.8413279056549072, + "loss": 5.315, + "lr": 0.0008896503496503497, + "step": 1290, + "tokens_trained": 0.122608512 + }, + { + "epoch": 0.36652482269503545, + "grad_norm": 1.802370548248291, + "loss": 5.3203, + "lr": 0.0008893706293706294, + "step": 1292, + "tokens_trained": 0.122795944 + }, + { + "epoch": 0.3670921985815603, + "grad_norm": 1.5968035459518433, + "loss": 5.3833, + "lr": 0.000889090909090909, + "step": 1294, + "tokens_trained": 0.1229842 + }, + { + "epoch": 0.3676595744680851, + "grad_norm": 1.8354761600494385, + "loss": 5.3365, + "lr": 0.0008888111888111888, + "step": 1296, + "tokens_trained": 0.123175336 + }, + { + "epoch": 0.36822695035460995, + "grad_norm": 1.925227403640747, + "loss": 5.3687, + "lr": 0.0008885314685314686, + "step": 1298, + "tokens_trained": 0.123366848 + }, + { + "epoch": 0.36879432624113473, + "grad_norm": 1.7477060556411743, + "loss": 5.4033, + "lr": 0.0008882517482517483, + "step": 1300, + "tokens_trained": 0.123556656 + }, + { + "epoch": 0.36936170212765956, + "grad_norm": 1.8925527334213257, + "loss": 5.2854, + "lr": 0.000887972027972028, + "step": 1302, + "tokens_trained": 0.12374612 + }, + { + "epoch": 0.3699290780141844, + "grad_norm": 1.8904681205749512, + "loss": 5.2903, + "lr": 0.0008876923076923077, + "step": 1304, + "tokens_trained": 0.123936192 + }, + { + "epoch": 0.37049645390070923, + "grad_norm": 1.9903556108474731, + "loss": 5.2994, + "lr": 0.0008874125874125875, + "step": 1306, + "tokens_trained": 0.124126112 + }, + { + "epoch": 0.37106382978723407, + "grad_norm": 2.014011859893799, + "loss": 5.353, + "lr": 0.0008871328671328671, + "step": 1308, + "tokens_trained": 0.124314592 + }, + { + "epoch": 0.37163120567375885, + "grad_norm": 1.9086287021636963, + "loss": 5.3924, + "lr": 0.0008868531468531469, + "step": 1310, + "tokens_trained": 0.124503496 + }, + { + "epoch": 0.3721985815602837, + "grad_norm": 1.8927134275436401, + "loss": 5.3098, + "lr": 0.0008865734265734265, + "step": 1312, + "tokens_trained": 0.124693296 + }, + { + "epoch": 0.3727659574468085, + "grad_norm": 1.850883960723877, + "loss": 5.356, + "lr": 0.0008862937062937063, + "step": 1314, + "tokens_trained": 0.124883528 + }, + { + "epoch": 0.37333333333333335, + "grad_norm": 1.813315510749817, + "loss": 5.3564, + "lr": 0.0008860139860139861, + "step": 1316, + "tokens_trained": 0.125072328 + }, + { + "epoch": 0.37390070921985813, + "grad_norm": 1.6776509284973145, + "loss": 5.3348, + "lr": 0.0008857342657342658, + "step": 1318, + "tokens_trained": 0.125263128 + }, + { + "epoch": 0.37446808510638296, + "grad_norm": 1.7775620222091675, + "loss": 5.298, + "lr": 0.0008854545454545455, + "step": 1320, + "tokens_trained": 0.125453944 + }, + { + "epoch": 0.3750354609929078, + "grad_norm": 1.6916086673736572, + "loss": 5.332, + "lr": 0.0008851748251748251, + "step": 1322, + "tokens_trained": 0.125644264 + }, + { + "epoch": 0.37560283687943263, + "grad_norm": 1.7182034254074097, + "loss": 5.3405, + "lr": 0.000884895104895105, + "step": 1324, + "tokens_trained": 0.125835256 + }, + { + "epoch": 0.37617021276595747, + "grad_norm": 1.690463662147522, + "loss": 5.355, + "lr": 0.0008846153846153846, + "step": 1326, + "tokens_trained": 0.126025952 + }, + { + "epoch": 0.37673758865248225, + "grad_norm": 1.7073352336883545, + "loss": 5.3304, + "lr": 0.0008843356643356644, + "step": 1328, + "tokens_trained": 0.126217456 + }, + { + "epoch": 0.3773049645390071, + "grad_norm": 1.6633049249649048, + "loss": 5.2724, + "lr": 0.000884055944055944, + "step": 1330, + "tokens_trained": 0.126407688 + }, + { + "epoch": 0.3778723404255319, + "grad_norm": 1.618843913078308, + "loss": 5.2952, + "lr": 0.0008837762237762238, + "step": 1332, + "tokens_trained": 0.126599504 + }, + { + "epoch": 0.37843971631205675, + "grad_norm": 1.7496757507324219, + "loss": 5.2846, + "lr": 0.0008834965034965036, + "step": 1334, + "tokens_trained": 0.126787648 + }, + { + "epoch": 0.3790070921985816, + "grad_norm": 1.7284750938415527, + "loss": 5.3229, + "lr": 0.0008832167832167832, + "step": 1336, + "tokens_trained": 0.126977568 + }, + { + "epoch": 0.37957446808510636, + "grad_norm": 1.55423903465271, + "loss": 5.3112, + "lr": 0.000882937062937063, + "step": 1338, + "tokens_trained": 0.12716944 + }, + { + "epoch": 0.3801418439716312, + "grad_norm": 1.5783073902130127, + "loss": 5.3002, + "lr": 0.0008826573426573426, + "step": 1340, + "tokens_trained": 0.127357296 + }, + { + "epoch": 0.38070921985815603, + "grad_norm": 1.6970964670181274, + "loss": 5.3003, + "lr": 0.0008823776223776225, + "step": 1342, + "tokens_trained": 0.127547112 + }, + { + "epoch": 0.38127659574468087, + "grad_norm": 1.8086830377578735, + "loss": 5.3018, + "lr": 0.0008820979020979021, + "step": 1344, + "tokens_trained": 0.12773616 + }, + { + "epoch": 0.38184397163120565, + "grad_norm": 1.6589199304580688, + "loss": 5.2903, + "lr": 0.0008818181818181819, + "step": 1346, + "tokens_trained": 0.127924704 + }, + { + "epoch": 0.3824113475177305, + "grad_norm": 1.6546344757080078, + "loss": 5.2639, + "lr": 0.0008815384615384615, + "step": 1348, + "tokens_trained": 0.128114848 + }, + { + "epoch": 0.3829787234042553, + "grad_norm": 1.6867282390594482, + "loss": 5.2713, + "lr": 0.0008812587412587412, + "step": 1350, + "tokens_trained": 0.12830572 + }, + { + "epoch": 0.38354609929078015, + "grad_norm": 1.6336158514022827, + "loss": 5.2688, + "lr": 0.0008809790209790211, + "step": 1352, + "tokens_trained": 0.128497336 + }, + { + "epoch": 0.384113475177305, + "grad_norm": 1.591659665107727, + "loss": 5.3073, + "lr": 0.0008806993006993007, + "step": 1354, + "tokens_trained": 0.128689088 + }, + { + "epoch": 0.38468085106382977, + "grad_norm": 1.6427522897720337, + "loss": 5.2649, + "lr": 0.0008804195804195805, + "step": 1356, + "tokens_trained": 0.128879208 + }, + { + "epoch": 0.3852482269503546, + "grad_norm": 1.693124771118164, + "loss": 5.334, + "lr": 0.0008801398601398601, + "step": 1358, + "tokens_trained": 0.129069376 + }, + { + "epoch": 0.38581560283687943, + "grad_norm": 1.6677430868148804, + "loss": 5.3091, + "lr": 0.00087986013986014, + "step": 1360, + "tokens_trained": 0.12925972 + }, + { + "epoch": 0.38638297872340427, + "grad_norm": 1.6829359531402588, + "loss": 5.3529, + "lr": 0.0008795804195804196, + "step": 1362, + "tokens_trained": 0.129449816 + }, + { + "epoch": 0.3869503546099291, + "grad_norm": 1.6984829902648926, + "loss": 5.2832, + "lr": 0.0008793006993006993, + "step": 1364, + "tokens_trained": 0.129638736 + }, + { + "epoch": 0.3875177304964539, + "grad_norm": 1.6351298093795776, + "loss": 5.3654, + "lr": 0.000879020979020979, + "step": 1366, + "tokens_trained": 0.129831512 + }, + { + "epoch": 0.3880851063829787, + "grad_norm": 1.588394045829773, + "loss": 5.3203, + "lr": 0.0008787412587412587, + "step": 1368, + "tokens_trained": 0.130021424 + }, + { + "epoch": 0.38865248226950355, + "grad_norm": 1.7608240842819214, + "loss": 5.3387, + "lr": 0.0008784615384615386, + "step": 1370, + "tokens_trained": 0.130211848 + }, + { + "epoch": 0.3892198581560284, + "grad_norm": 1.7742120027542114, + "loss": 5.3054, + "lr": 0.0008781818181818182, + "step": 1372, + "tokens_trained": 0.130400256 + }, + { + "epoch": 0.38978723404255317, + "grad_norm": 1.8751057386398315, + "loss": 5.3569, + "lr": 0.000877902097902098, + "step": 1374, + "tokens_trained": 0.130591616 + }, + { + "epoch": 0.3900709219858156, + "eval_loss": 5.315512180328369, + "eval_runtime": 20.9232, + "step": 1375, + "tokens_trained": 0.130685128 + }, + { + "epoch": 0.390354609929078, + "grad_norm": 1.8666746616363525, + "loss": 5.3088, + "lr": 0.0008776223776223776, + "step": 1376, + "tokens_trained": 0.130781056 + }, + { + "epoch": 0.39092198581560283, + "grad_norm": 1.8694190979003906, + "loss": 5.2391, + "lr": 0.0008773426573426574, + "step": 1378, + "tokens_trained": 0.130971152 + }, + { + "epoch": 0.39148936170212767, + "grad_norm": 2.0663323402404785, + "loss": 5.3497, + "lr": 0.0008770629370629371, + "step": 1380, + "tokens_trained": 0.131163224 + }, + { + "epoch": 0.3920567375886525, + "grad_norm": 1.956207275390625, + "loss": 5.3227, + "lr": 0.0008767832167832168, + "step": 1382, + "tokens_trained": 0.131353832 + }, + { + "epoch": 0.3926241134751773, + "grad_norm": 1.6816498041152954, + "loss": 5.2626, + "lr": 0.0008765034965034965, + "step": 1384, + "tokens_trained": 0.13154472 + }, + { + "epoch": 0.3931914893617021, + "grad_norm": 1.655116319656372, + "loss": 5.3334, + "lr": 0.0008762237762237762, + "step": 1386, + "tokens_trained": 0.131732128 + }, + { + "epoch": 0.39375886524822695, + "grad_norm": 1.6439241170883179, + "loss": 5.3038, + "lr": 0.0008759440559440561, + "step": 1388, + "tokens_trained": 0.131920728 + }, + { + "epoch": 0.3943262411347518, + "grad_norm": 1.5000464916229248, + "loss": 5.2798, + "lr": 0.0008756643356643357, + "step": 1390, + "tokens_trained": 0.1321094 + }, + { + "epoch": 0.3948936170212766, + "grad_norm": 1.7129333019256592, + "loss": 5.2918, + "lr": 0.0008753846153846154, + "step": 1392, + "tokens_trained": 0.132299832 + }, + { + "epoch": 0.3954609929078014, + "grad_norm": 1.7489241361618042, + "loss": 5.3551, + "lr": 0.0008751048951048951, + "step": 1394, + "tokens_trained": 0.13249016 + }, + { + "epoch": 0.39602836879432624, + "grad_norm": 1.6597840785980225, + "loss": 5.3718, + "lr": 0.0008748251748251749, + "step": 1396, + "tokens_trained": 0.132680568 + }, + { + "epoch": 0.39659574468085107, + "grad_norm": 1.8800175189971924, + "loss": 5.3578, + "lr": 0.0008745454545454546, + "step": 1398, + "tokens_trained": 0.132871296 + }, + { + "epoch": 0.3971631205673759, + "grad_norm": 1.8190884590148926, + "loss": 5.2714, + "lr": 0.0008742657342657343, + "step": 1400, + "tokens_trained": 0.133062288 + }, + { + "epoch": 0.3977304964539007, + "grad_norm": 1.602634310722351, + "loss": 5.2914, + "lr": 0.0008739860139860139, + "step": 1402, + "tokens_trained": 0.133252584 + }, + { + "epoch": 0.3982978723404255, + "grad_norm": 1.7363992929458618, + "loss": 5.3154, + "lr": 0.0008737062937062937, + "step": 1404, + "tokens_trained": 0.133444784 + }, + { + "epoch": 0.39886524822695035, + "grad_norm": 1.7578014135360718, + "loss": 5.3735, + "lr": 0.0008734265734265734, + "step": 1406, + "tokens_trained": 0.133636288 + }, + { + "epoch": 0.3994326241134752, + "grad_norm": 1.8847187757492065, + "loss": 5.3118, + "lr": 0.0008731468531468532, + "step": 1408, + "tokens_trained": 0.133825824 + }, + { + "epoch": 0.4, + "grad_norm": 1.750780701637268, + "loss": 5.3101, + "lr": 0.0008728671328671329, + "step": 1410, + "tokens_trained": 0.134016688 + }, + { + "epoch": 0.4005673758865248, + "grad_norm": 1.6785613298416138, + "loss": 5.2823, + "lr": 0.0008725874125874126, + "step": 1412, + "tokens_trained": 0.134208992 + }, + { + "epoch": 0.40113475177304964, + "grad_norm": 1.7419382333755493, + "loss": 5.2388, + "lr": 0.0008723076923076924, + "step": 1414, + "tokens_trained": 0.134398376 + }, + { + "epoch": 0.40170212765957447, + "grad_norm": 1.6936920881271362, + "loss": 5.2824, + "lr": 0.000872027972027972, + "step": 1416, + "tokens_trained": 0.134589264 + }, + { + "epoch": 0.4022695035460993, + "grad_norm": 1.7408183813095093, + "loss": 5.2692, + "lr": 0.0008717482517482518, + "step": 1418, + "tokens_trained": 0.134776568 + }, + { + "epoch": 0.40283687943262414, + "grad_norm": 1.7089916467666626, + "loss": 5.2309, + "lr": 0.0008714685314685314, + "step": 1420, + "tokens_trained": 0.134967616 + }, + { + "epoch": 0.4034042553191489, + "grad_norm": 1.6850922107696533, + "loss": 5.3656, + "lr": 0.0008711888111888112, + "step": 1422, + "tokens_trained": 0.135158272 + }, + { + "epoch": 0.40397163120567375, + "grad_norm": 1.546431064605713, + "loss": 5.3455, + "lr": 0.0008709090909090909, + "step": 1424, + "tokens_trained": 0.135349512 + }, + { + "epoch": 0.4045390070921986, + "grad_norm": 1.3656421899795532, + "loss": 5.2842, + "lr": 0.0008706293706293707, + "step": 1426, + "tokens_trained": 0.135538512 + }, + { + "epoch": 0.4051063829787234, + "grad_norm": 1.5918062925338745, + "loss": 5.3243, + "lr": 0.0008703496503496504, + "step": 1428, + "tokens_trained": 0.13572968 + }, + { + "epoch": 0.4056737588652482, + "grad_norm": 1.563009262084961, + "loss": 5.2539, + "lr": 0.00087006993006993, + "step": 1430, + "tokens_trained": 0.135919568 + }, + { + "epoch": 0.40624113475177304, + "grad_norm": 1.6144121885299683, + "loss": 5.2844, + "lr": 0.0008697902097902099, + "step": 1432, + "tokens_trained": 0.136109304 + }, + { + "epoch": 0.40680851063829787, + "grad_norm": 1.5911130905151367, + "loss": 5.3205, + "lr": 0.0008695104895104895, + "step": 1434, + "tokens_trained": 0.136296696 + }, + { + "epoch": 0.4073758865248227, + "grad_norm": 1.60932457447052, + "loss": 5.3783, + "lr": 0.0008692307692307693, + "step": 1436, + "tokens_trained": 0.136484912 + }, + { + "epoch": 0.40794326241134754, + "grad_norm": 1.559644341468811, + "loss": 5.2785, + "lr": 0.0008689510489510489, + "step": 1438, + "tokens_trained": 0.136675736 + }, + { + "epoch": 0.4085106382978723, + "grad_norm": 1.5167043209075928, + "loss": 5.3224, + "lr": 0.0008686713286713287, + "step": 1440, + "tokens_trained": 0.136864928 + }, + { + "epoch": 0.40907801418439715, + "grad_norm": 1.5843397378921509, + "loss": 5.3075, + "lr": 0.0008683916083916084, + "step": 1442, + "tokens_trained": 0.137056688 + }, + { + "epoch": 0.409645390070922, + "grad_norm": 1.581120491027832, + "loss": 5.2863, + "lr": 0.0008681118881118881, + "step": 1444, + "tokens_trained": 0.137244664 + }, + { + "epoch": 0.4102127659574468, + "grad_norm": 1.6355490684509277, + "loss": 5.348, + "lr": 0.0008678321678321679, + "step": 1446, + "tokens_trained": 0.13743372 + }, + { + "epoch": 0.41078014184397166, + "grad_norm": 1.5543185472488403, + "loss": 5.3268, + "lr": 0.0008675524475524475, + "step": 1448, + "tokens_trained": 0.13762696 + }, + { + "epoch": 0.41134751773049644, + "grad_norm": 1.5313750505447388, + "loss": 5.2784, + "lr": 0.0008672727272727273, + "step": 1450, + "tokens_trained": 0.137817376 + }, + { + "epoch": 0.41191489361702127, + "grad_norm": 1.7918111085891724, + "loss": 5.3063, + "lr": 0.000866993006993007, + "step": 1452, + "tokens_trained": 0.138007944 + }, + { + "epoch": 0.4124822695035461, + "grad_norm": 1.5105966329574585, + "loss": 5.2432, + "lr": 0.0008667132867132868, + "step": 1454, + "tokens_trained": 0.138199776 + }, + { + "epoch": 0.41304964539007094, + "grad_norm": 1.4441865682601929, + "loss": 5.269, + "lr": 0.0008664335664335664, + "step": 1456, + "tokens_trained": 0.13839124 + }, + { + "epoch": 0.4136170212765957, + "grad_norm": 1.473544955253601, + "loss": 5.2377, + "lr": 0.0008661538461538461, + "step": 1458, + "tokens_trained": 0.138580704 + }, + { + "epoch": 0.41418439716312055, + "grad_norm": 1.6085572242736816, + "loss": 5.245, + "lr": 0.0008658741258741259, + "step": 1460, + "tokens_trained": 0.138770176 + }, + { + "epoch": 0.4147517730496454, + "grad_norm": 1.609894871711731, + "loss": 5.3124, + "lr": 0.0008655944055944056, + "step": 1462, + "tokens_trained": 0.138961656 + }, + { + "epoch": 0.4153191489361702, + "grad_norm": 1.6923688650131226, + "loss": 5.3099, + "lr": 0.0008653146853146854, + "step": 1464, + "tokens_trained": 0.139151128 + }, + { + "epoch": 0.41588652482269506, + "grad_norm": 1.7480796575546265, + "loss": 5.2608, + "lr": 0.000865034965034965, + "step": 1466, + "tokens_trained": 0.139341168 + }, + { + "epoch": 0.41645390070921984, + "grad_norm": 1.725832223892212, + "loss": 5.2863, + "lr": 0.0008647552447552448, + "step": 1468, + "tokens_trained": 0.139530448 + }, + { + "epoch": 0.41702127659574467, + "grad_norm": 1.7886406183242798, + "loss": 5.231, + "lr": 0.0008644755244755245, + "step": 1470, + "tokens_trained": 0.13972244 + }, + { + "epoch": 0.4175886524822695, + "grad_norm": 1.803231954574585, + "loss": 5.2428, + "lr": 0.0008641958041958042, + "step": 1472, + "tokens_trained": 0.139913136 + }, + { + "epoch": 0.41815602836879434, + "grad_norm": 1.5347254276275635, + "loss": 5.2215, + "lr": 0.0008639160839160839, + "step": 1474, + "tokens_trained": 0.140104072 + }, + { + "epoch": 0.4187234042553192, + "grad_norm": 1.4485915899276733, + "loss": 5.2364, + "lr": 0.0008636363636363636, + "step": 1476, + "tokens_trained": 0.140294312 + }, + { + "epoch": 0.41929078014184396, + "grad_norm": 1.6130446195602417, + "loss": 5.3088, + "lr": 0.0008633566433566434, + "step": 1478, + "tokens_trained": 0.140482968 + }, + { + "epoch": 0.4198581560283688, + "grad_norm": 1.5839030742645264, + "loss": 5.3215, + "lr": 0.0008630769230769231, + "step": 1480, + "tokens_trained": 0.140674208 + }, + { + "epoch": 0.4204255319148936, + "grad_norm": 1.7519373893737793, + "loss": 5.3331, + "lr": 0.0008627972027972029, + "step": 1482, + "tokens_trained": 0.140864408 + }, + { + "epoch": 0.42099290780141846, + "grad_norm": 1.6718385219573975, + "loss": 5.231, + "lr": 0.0008625174825174825, + "step": 1484, + "tokens_trained": 0.141054696 + }, + { + "epoch": 0.42156028368794324, + "grad_norm": 1.5733797550201416, + "loss": 5.2621, + "lr": 0.0008622377622377622, + "step": 1486, + "tokens_trained": 0.141245712 + }, + { + "epoch": 0.4221276595744681, + "grad_norm": 1.549985647201538, + "loss": 5.2574, + "lr": 0.000861958041958042, + "step": 1488, + "tokens_trained": 0.141434232 + }, + { + "epoch": 0.4226950354609929, + "grad_norm": 1.651908278465271, + "loss": 5.2953, + "lr": 0.0008616783216783217, + "step": 1490, + "tokens_trained": 0.141623936 + }, + { + "epoch": 0.42326241134751774, + "grad_norm": 1.5680350065231323, + "loss": 5.288, + "lr": 0.0008613986013986014, + "step": 1492, + "tokens_trained": 0.141813904 + }, + { + "epoch": 0.4238297872340426, + "grad_norm": 1.5155646800994873, + "loss": 5.2529, + "lr": 0.0008611188811188811, + "step": 1494, + "tokens_trained": 0.14200372 + }, + { + "epoch": 0.42439716312056736, + "grad_norm": 1.5949562788009644, + "loss": 5.3064, + "lr": 0.0008608391608391609, + "step": 1496, + "tokens_trained": 0.142194496 + }, + { + "epoch": 0.4249645390070922, + "grad_norm": 1.6359357833862305, + "loss": 5.3452, + "lr": 0.0008605594405594406, + "step": 1498, + "tokens_trained": 0.142384592 + }, + { + "epoch": 0.425531914893617, + "grad_norm": 1.648120403289795, + "loss": 5.3427, + "lr": 0.0008602797202797203, + "step": 1500, + "tokens_trained": 0.142573368 + }, + { + "epoch": 0.425531914893617, + "eval_loss": 5.282389163970947, + "eval_runtime": 20.5657, + "step": 1500, + "tokens_trained": 0.142573368 + }, + { + "epoch": 0.42609929078014186, + "grad_norm": 1.6313989162445068, + "loss": 5.2442, + "lr": 0.00086, + "step": 1502, + "tokens_trained": 0.142764584 + }, + { + "epoch": 0.4266666666666667, + "grad_norm": 1.447824239730835, + "loss": 5.2979, + "lr": 0.0008597202797202797, + "step": 1504, + "tokens_trained": 0.142953912 + }, + { + "epoch": 0.4272340425531915, + "grad_norm": 1.4285600185394287, + "loss": 5.317, + "lr": 0.0008594405594405595, + "step": 1506, + "tokens_trained": 0.143145944 + }, + { + "epoch": 0.4278014184397163, + "grad_norm": 1.4464077949523926, + "loss": 5.2746, + "lr": 0.0008591608391608392, + "step": 1508, + "tokens_trained": 0.1433374 + }, + { + "epoch": 0.42836879432624114, + "grad_norm": 1.3554625511169434, + "loss": 5.276, + "lr": 0.0008588811188811188, + "step": 1510, + "tokens_trained": 0.143529088 + }, + { + "epoch": 0.428936170212766, + "grad_norm": 1.4690148830413818, + "loss": 5.2976, + "lr": 0.0008586013986013986, + "step": 1512, + "tokens_trained": 0.1437192 + }, + { + "epoch": 0.42950354609929076, + "grad_norm": 1.4911222457885742, + "loss": 5.2727, + "lr": 0.0008583216783216783, + "step": 1514, + "tokens_trained": 0.143907728 + }, + { + "epoch": 0.4300709219858156, + "grad_norm": 1.5823880434036255, + "loss": 5.2481, + "lr": 0.0008580419580419581, + "step": 1516, + "tokens_trained": 0.144097048 + }, + { + "epoch": 0.4306382978723404, + "grad_norm": 1.486588716506958, + "loss": 5.2561, + "lr": 0.0008577622377622378, + "step": 1518, + "tokens_trained": 0.14428652 + }, + { + "epoch": 0.43120567375886526, + "grad_norm": 1.5762882232666016, + "loss": 5.267, + "lr": 0.0008574825174825175, + "step": 1520, + "tokens_trained": 0.144476848 + }, + { + "epoch": 0.4317730496453901, + "grad_norm": 1.6832828521728516, + "loss": 5.3329, + "lr": 0.0008572027972027972, + "step": 1522, + "tokens_trained": 0.144667568 + }, + { + "epoch": 0.4323404255319149, + "grad_norm": 1.7036137580871582, + "loss": 5.2326, + "lr": 0.000856923076923077, + "step": 1524, + "tokens_trained": 0.144860328 + }, + { + "epoch": 0.4329078014184397, + "grad_norm": 1.8102291822433472, + "loss": 5.251, + "lr": 0.0008566433566433567, + "step": 1526, + "tokens_trained": 0.1450528 + }, + { + "epoch": 0.43347517730496454, + "grad_norm": 1.667229413986206, + "loss": 5.2841, + "lr": 0.0008563636363636363, + "step": 1528, + "tokens_trained": 0.145240952 + }, + { + "epoch": 0.4340425531914894, + "grad_norm": 1.6709800958633423, + "loss": 5.2387, + "lr": 0.0008560839160839161, + "step": 1530, + "tokens_trained": 0.145431376 + }, + { + "epoch": 0.4346099290780142, + "grad_norm": 1.600885272026062, + "loss": 5.2179, + "lr": 0.0008558041958041958, + "step": 1532, + "tokens_trained": 0.145620184 + }, + { + "epoch": 0.435177304964539, + "grad_norm": 1.5783873796463013, + "loss": 5.2432, + "lr": 0.0008555244755244756, + "step": 1534, + "tokens_trained": 0.145810616 + }, + { + "epoch": 0.4357446808510638, + "grad_norm": 1.5059685707092285, + "loss": 5.2604, + "lr": 0.0008552447552447553, + "step": 1536, + "tokens_trained": 0.14600232 + }, + { + "epoch": 0.43631205673758866, + "grad_norm": 1.5880341529846191, + "loss": 5.249, + "lr": 0.000854965034965035, + "step": 1538, + "tokens_trained": 0.146192504 + }, + { + "epoch": 0.4368794326241135, + "grad_norm": 1.430004596710205, + "loss": 5.2668, + "lr": 0.0008546853146853147, + "step": 1540, + "tokens_trained": 0.146382264 + }, + { + "epoch": 0.4374468085106383, + "grad_norm": 1.4099256992340088, + "loss": 5.2839, + "lr": 0.0008544055944055944, + "step": 1542, + "tokens_trained": 0.146570432 + }, + { + "epoch": 0.4380141843971631, + "grad_norm": 1.3938827514648438, + "loss": 5.2534, + "lr": 0.0008541258741258742, + "step": 1544, + "tokens_trained": 0.146763736 + }, + { + "epoch": 0.43858156028368794, + "grad_norm": 1.4359923601150513, + "loss": 5.2202, + "lr": 0.0008538461538461538, + "step": 1546, + "tokens_trained": 0.146953944 + }, + { + "epoch": 0.4391489361702128, + "grad_norm": 1.5405043363571167, + "loss": 5.2613, + "lr": 0.0008535664335664336, + "step": 1548, + "tokens_trained": 0.147144664 + }, + { + "epoch": 0.4397163120567376, + "grad_norm": 1.6448051929473877, + "loss": 5.299, + "lr": 0.0008532867132867133, + "step": 1550, + "tokens_trained": 0.147335064 + }, + { + "epoch": 0.4402836879432624, + "grad_norm": 1.6528949737548828, + "loss": 5.3004, + "lr": 0.000853006993006993, + "step": 1552, + "tokens_trained": 0.147524088 + }, + { + "epoch": 0.4408510638297872, + "grad_norm": 1.637702226638794, + "loss": 5.2298, + "lr": 0.0008527272727272728, + "step": 1554, + "tokens_trained": 0.147716296 + }, + { + "epoch": 0.44141843971631206, + "grad_norm": 1.7230212688446045, + "loss": 5.2806, + "lr": 0.0008524475524475524, + "step": 1556, + "tokens_trained": 0.147905216 + }, + { + "epoch": 0.4419858156028369, + "grad_norm": 1.6216089725494385, + "loss": 5.3062, + "lr": 0.0008521678321678322, + "step": 1558, + "tokens_trained": 0.148092312 + }, + { + "epoch": 0.4425531914893617, + "grad_norm": 1.5734955072402954, + "loss": 5.2607, + "lr": 0.0008518881118881119, + "step": 1560, + "tokens_trained": 0.148282712 + }, + { + "epoch": 0.4431205673758865, + "grad_norm": 1.6687103509902954, + "loss": 5.2737, + "lr": 0.0008516083916083917, + "step": 1562, + "tokens_trained": 0.148474672 + }, + { + "epoch": 0.44368794326241134, + "grad_norm": 1.547277569770813, + "loss": 5.3183, + "lr": 0.0008513286713286713, + "step": 1564, + "tokens_trained": 0.148667824 + }, + { + "epoch": 0.4442553191489362, + "grad_norm": 1.3782074451446533, + "loss": 5.266, + "lr": 0.000851048951048951, + "step": 1566, + "tokens_trained": 0.14885704 + }, + { + "epoch": 0.444822695035461, + "grad_norm": 1.5648273229599, + "loss": 5.2954, + "lr": 0.0008507692307692308, + "step": 1568, + "tokens_trained": 0.14904804 + }, + { + "epoch": 0.4453900709219858, + "grad_norm": 1.5675908327102661, + "loss": 5.2897, + "lr": 0.0008504895104895105, + "step": 1570, + "tokens_trained": 0.149237048 + }, + { + "epoch": 0.4459574468085106, + "grad_norm": 1.5399287939071655, + "loss": 5.2993, + "lr": 0.0008502097902097903, + "step": 1572, + "tokens_trained": 0.149427328 + }, + { + "epoch": 0.44652482269503546, + "grad_norm": 1.7170253992080688, + "loss": 5.2756, + "lr": 0.0008499300699300699, + "step": 1574, + "tokens_trained": 0.149618448 + }, + { + "epoch": 0.4470921985815603, + "grad_norm": 1.5694142580032349, + "loss": 5.2294, + "lr": 0.0008496503496503497, + "step": 1576, + "tokens_trained": 0.149809416 + }, + { + "epoch": 0.44765957446808513, + "grad_norm": 1.5410487651824951, + "loss": 5.2392, + "lr": 0.0008493706293706294, + "step": 1578, + "tokens_trained": 0.149999608 + }, + { + "epoch": 0.4482269503546099, + "grad_norm": 1.5991896390914917, + "loss": 5.2569, + "lr": 0.0008490909090909091, + "step": 1580, + "tokens_trained": 0.150190224 + }, + { + "epoch": 0.44879432624113474, + "grad_norm": 1.5861775875091553, + "loss": 5.3151, + "lr": 0.0008488111888111888, + "step": 1582, + "tokens_trained": 0.150380592 + }, + { + "epoch": 0.4493617021276596, + "grad_norm": 1.530462622642517, + "loss": 5.3242, + "lr": 0.0008485314685314685, + "step": 1584, + "tokens_trained": 0.15056992 + }, + { + "epoch": 0.4499290780141844, + "grad_norm": 1.5658655166625977, + "loss": 5.2933, + "lr": 0.0008482517482517483, + "step": 1586, + "tokens_trained": 0.150760336 + }, + { + "epoch": 0.4504964539007092, + "grad_norm": 1.4187430143356323, + "loss": 5.2235, + "lr": 0.000847972027972028, + "step": 1588, + "tokens_trained": 0.150949088 + }, + { + "epoch": 0.451063829787234, + "grad_norm": 1.6921541690826416, + "loss": 5.2496, + "lr": 0.0008476923076923078, + "step": 1590, + "tokens_trained": 0.151140016 + }, + { + "epoch": 0.45163120567375886, + "grad_norm": 1.6049220561981201, + "loss": 5.2767, + "lr": 0.0008474125874125874, + "step": 1592, + "tokens_trained": 0.151330944 + }, + { + "epoch": 0.4521985815602837, + "grad_norm": 1.513168454170227, + "loss": 5.2904, + "lr": 0.0008471328671328671, + "step": 1594, + "tokens_trained": 0.151520152 + }, + { + "epoch": 0.45276595744680853, + "grad_norm": 1.5247087478637695, + "loss": 5.2391, + "lr": 0.0008468531468531469, + "step": 1596, + "tokens_trained": 0.151711592 + }, + { + "epoch": 0.4533333333333333, + "grad_norm": 1.5005898475646973, + "loss": 5.3025, + "lr": 0.0008465734265734266, + "step": 1598, + "tokens_trained": 0.151902736 + }, + { + "epoch": 0.45390070921985815, + "grad_norm": 1.3196156024932861, + "loss": 5.3025, + "lr": 0.0008462937062937063, + "step": 1600, + "tokens_trained": 0.152094032 + }, + { + "epoch": 0.454468085106383, + "grad_norm": 1.5037102699279785, + "loss": 5.2348, + "lr": 0.000846013986013986, + "step": 1602, + "tokens_trained": 0.15228336 + }, + { + "epoch": 0.4550354609929078, + "grad_norm": 1.404539942741394, + "loss": 5.2551, + "lr": 0.0008457342657342658, + "step": 1604, + "tokens_trained": 0.152474776 + }, + { + "epoch": 0.45560283687943265, + "grad_norm": 1.4784883260726929, + "loss": 5.2927, + "lr": 0.0008454545454545455, + "step": 1606, + "tokens_trained": 0.152663392 + }, + { + "epoch": 0.45617021276595743, + "grad_norm": 1.3743332624435425, + "loss": 5.2542, + "lr": 0.0008451748251748252, + "step": 1608, + "tokens_trained": 0.152852512 + }, + { + "epoch": 0.45673758865248226, + "grad_norm": 1.4161995649337769, + "loss": 5.2518, + "lr": 0.0008448951048951049, + "step": 1610, + "tokens_trained": 0.15304428 + }, + { + "epoch": 0.4573049645390071, + "grad_norm": 1.5045989751815796, + "loss": 5.2735, + "lr": 0.0008446153846153846, + "step": 1612, + "tokens_trained": 0.153234632 + }, + { + "epoch": 0.45787234042553193, + "grad_norm": 1.3695783615112305, + "loss": 5.2294, + "lr": 0.0008443356643356644, + "step": 1614, + "tokens_trained": 0.1534248 + }, + { + "epoch": 0.4584397163120567, + "grad_norm": 1.4551646709442139, + "loss": 5.2639, + "lr": 0.0008440559440559441, + "step": 1616, + "tokens_trained": 0.153614944 + }, + { + "epoch": 0.45900709219858155, + "grad_norm": 1.5018376111984253, + "loss": 5.2989, + "lr": 0.0008437762237762238, + "step": 1618, + "tokens_trained": 0.153803784 + }, + { + "epoch": 0.4595744680851064, + "grad_norm": 1.5295960903167725, + "loss": 5.33, + "lr": 0.0008434965034965035, + "step": 1620, + "tokens_trained": 0.153993752 + }, + { + "epoch": 0.4601418439716312, + "grad_norm": 1.417626142501831, + "loss": 5.2134, + "lr": 0.0008432167832167832, + "step": 1622, + "tokens_trained": 0.154184448 + }, + { + "epoch": 0.46070921985815605, + "grad_norm": 1.5715348720550537, + "loss": 5.2782, + "lr": 0.000842937062937063, + "step": 1624, + "tokens_trained": 0.154373632 + }, + { + "epoch": 0.46099290780141844, + "eval_loss": 5.266384601593018, + "eval_runtime": 21.0916, + "step": 1625, + "tokens_trained": 0.154468808 + }, + { + "epoch": 0.46127659574468083, + "grad_norm": 1.5504534244537354, + "loss": 5.2307, + "lr": 0.0008426573426573427, + "step": 1626, + "tokens_trained": 0.154564864 + }, + { + "epoch": 0.46184397163120566, + "grad_norm": 1.483108401298523, + "loss": 5.2578, + "lr": 0.0008423776223776224, + "step": 1628, + "tokens_trained": 0.154755312 + }, + { + "epoch": 0.4624113475177305, + "grad_norm": 1.5631264448165894, + "loss": 5.3291, + "lr": 0.0008420979020979021, + "step": 1630, + "tokens_trained": 0.154943736 + }, + { + "epoch": 0.46297872340425533, + "grad_norm": 1.4680705070495605, + "loss": 5.2256, + "lr": 0.0008418181818181819, + "step": 1632, + "tokens_trained": 0.15513452 + }, + { + "epoch": 0.46354609929078017, + "grad_norm": 1.468338966369629, + "loss": 5.2712, + "lr": 0.0008415384615384616, + "step": 1634, + "tokens_trained": 0.155325288 + }, + { + "epoch": 0.46411347517730495, + "grad_norm": 1.4557780027389526, + "loss": 5.2808, + "lr": 0.0008412587412587412, + "step": 1636, + "tokens_trained": 0.155515328 + }, + { + "epoch": 0.4646808510638298, + "grad_norm": 1.4534999132156372, + "loss": 5.2707, + "lr": 0.000840979020979021, + "step": 1638, + "tokens_trained": 0.155706752 + }, + { + "epoch": 0.4652482269503546, + "grad_norm": 1.4011393785476685, + "loss": 5.3028, + "lr": 0.0008406993006993006, + "step": 1640, + "tokens_trained": 0.155895336 + }, + { + "epoch": 0.46581560283687945, + "grad_norm": 1.307922601699829, + "loss": 5.2188, + "lr": 0.0008404195804195805, + "step": 1642, + "tokens_trained": 0.156085936 + }, + { + "epoch": 0.46638297872340423, + "grad_norm": 1.359922170639038, + "loss": 5.2863, + "lr": 0.0008401398601398602, + "step": 1644, + "tokens_trained": 0.15627636 + }, + { + "epoch": 0.46695035460992906, + "grad_norm": 1.6204577684402466, + "loss": 5.2877, + "lr": 0.0008398601398601399, + "step": 1646, + "tokens_trained": 0.156465192 + }, + { + "epoch": 0.4675177304964539, + "grad_norm": 1.7367322444915771, + "loss": 5.2501, + "lr": 0.0008395804195804196, + "step": 1648, + "tokens_trained": 0.15665336 + }, + { + "epoch": 0.46808510638297873, + "grad_norm": 1.7013088464736938, + "loss": 5.2522, + "lr": 0.0008393006993006993, + "step": 1650, + "tokens_trained": 0.156843128 + }, + { + "epoch": 0.46865248226950357, + "grad_norm": 1.6429578065872192, + "loss": 5.2978, + "lr": 0.0008390209790209791, + "step": 1652, + "tokens_trained": 0.157034328 + }, + { + "epoch": 0.46921985815602835, + "grad_norm": 1.527243733406067, + "loss": 5.2384, + "lr": 0.0008387412587412587, + "step": 1654, + "tokens_trained": 0.157222784 + }, + { + "epoch": 0.4697872340425532, + "grad_norm": 1.4792861938476562, + "loss": 5.2149, + "lr": 0.0008384615384615385, + "step": 1656, + "tokens_trained": 0.15741308 + }, + { + "epoch": 0.470354609929078, + "grad_norm": 1.4050098657608032, + "loss": 5.229, + "lr": 0.0008381818181818181, + "step": 1658, + "tokens_trained": 0.157603872 + }, + { + "epoch": 0.47092198581560285, + "grad_norm": 1.4799182415008545, + "loss": 5.2235, + "lr": 0.000837902097902098, + "step": 1660, + "tokens_trained": 0.157793352 + }, + { + "epoch": 0.4714893617021277, + "grad_norm": 1.4031378030776978, + "loss": 5.23, + "lr": 0.0008376223776223776, + "step": 1662, + "tokens_trained": 0.157984416 + }, + { + "epoch": 0.47205673758865246, + "grad_norm": 1.5775604248046875, + "loss": 5.2811, + "lr": 0.0008373426573426573, + "step": 1664, + "tokens_trained": 0.158176048 + }, + { + "epoch": 0.4726241134751773, + "grad_norm": 1.4855432510375977, + "loss": 5.2363, + "lr": 0.0008370629370629371, + "step": 1666, + "tokens_trained": 0.158368152 + }, + { + "epoch": 0.47319148936170213, + "grad_norm": 1.5609453916549683, + "loss": 5.2984, + "lr": 0.0008367832167832168, + "step": 1668, + "tokens_trained": 0.15855684 + }, + { + "epoch": 0.47375886524822697, + "grad_norm": 1.5052629709243774, + "loss": 5.213, + "lr": 0.0008365034965034966, + "step": 1670, + "tokens_trained": 0.15874712 + }, + { + "epoch": 0.47432624113475175, + "grad_norm": 1.5655242204666138, + "loss": 5.2551, + "lr": 0.0008362237762237762, + "step": 1672, + "tokens_trained": 0.158937104 + }, + { + "epoch": 0.4748936170212766, + "grad_norm": 1.301142930984497, + "loss": 5.1564, + "lr": 0.000835944055944056, + "step": 1674, + "tokens_trained": 0.159128096 + }, + { + "epoch": 0.4754609929078014, + "grad_norm": 1.5447527170181274, + "loss": 5.2547, + "lr": 0.0008356643356643356, + "step": 1676, + "tokens_trained": 0.159318968 + }, + { + "epoch": 0.47602836879432625, + "grad_norm": 1.638100266456604, + "loss": 5.2301, + "lr": 0.0008353846153846154, + "step": 1678, + "tokens_trained": 0.159508648 + }, + { + "epoch": 0.4765957446808511, + "grad_norm": 1.6203068494796753, + "loss": 5.2644, + "lr": 0.0008351048951048951, + "step": 1680, + "tokens_trained": 0.159698648 + }, + { + "epoch": 0.47716312056737586, + "grad_norm": 1.4097110033035278, + "loss": 5.2047, + "lr": 0.0008348251748251748, + "step": 1682, + "tokens_trained": 0.159887392 + }, + { + "epoch": 0.4777304964539007, + "grad_norm": 1.3377385139465332, + "loss": 5.2685, + "lr": 0.0008345454545454546, + "step": 1684, + "tokens_trained": 0.160076904 + }, + { + "epoch": 0.47829787234042553, + "grad_norm": 1.4079371690750122, + "loss": 5.2842, + "lr": 0.0008342657342657343, + "step": 1686, + "tokens_trained": 0.160266712 + }, + { + "epoch": 0.47886524822695037, + "grad_norm": 1.6039987802505493, + "loss": 5.2248, + "lr": 0.0008339860139860141, + "step": 1688, + "tokens_trained": 0.160455464 + }, + { + "epoch": 0.4794326241134752, + "grad_norm": 1.639218807220459, + "loss": 5.2007, + "lr": 0.0008337062937062937, + "step": 1690, + "tokens_trained": 0.16064472 + }, + { + "epoch": 0.48, + "grad_norm": 1.8226710557937622, + "loss": 5.2427, + "lr": 0.0008334265734265734, + "step": 1692, + "tokens_trained": 0.160835192 + }, + { + "epoch": 0.4805673758865248, + "grad_norm": 1.6480419635772705, + "loss": 5.1944, + "lr": 0.0008331468531468531, + "step": 1694, + "tokens_trained": 0.161025272 + }, + { + "epoch": 0.48113475177304965, + "grad_norm": 1.666717290878296, + "loss": 5.2879, + "lr": 0.0008328671328671329, + "step": 1696, + "tokens_trained": 0.161214016 + }, + { + "epoch": 0.4817021276595745, + "grad_norm": 1.5092660188674927, + "loss": 5.2612, + "lr": 0.0008325874125874126, + "step": 1698, + "tokens_trained": 0.161405448 + }, + { + "epoch": 0.48226950354609927, + "grad_norm": 1.4042121171951294, + "loss": 5.2373, + "lr": 0.0008323076923076923, + "step": 1700, + "tokens_trained": 0.161595896 + }, + { + "epoch": 0.4828368794326241, + "grad_norm": 1.4937382936477661, + "loss": 5.2172, + "lr": 0.000832027972027972, + "step": 1702, + "tokens_trained": 0.161783904 + }, + { + "epoch": 0.48340425531914893, + "grad_norm": 1.4652959108352661, + "loss": 5.2704, + "lr": 0.0008317482517482518, + "step": 1704, + "tokens_trained": 0.161975888 + }, + { + "epoch": 0.48397163120567377, + "grad_norm": 1.3021745681762695, + "loss": 5.2672, + "lr": 0.0008314685314685315, + "step": 1706, + "tokens_trained": 0.162165808 + }, + { + "epoch": 0.4845390070921986, + "grad_norm": 1.3580701351165771, + "loss": 5.2467, + "lr": 0.0008311888111888112, + "step": 1708, + "tokens_trained": 0.162355152 + }, + { + "epoch": 0.4851063829787234, + "grad_norm": 1.480072259902954, + "loss": 5.2797, + "lr": 0.0008309090909090909, + "step": 1710, + "tokens_trained": 0.162544744 + }, + { + "epoch": 0.4856737588652482, + "grad_norm": 1.3532829284667969, + "loss": 5.2556, + "lr": 0.0008306293706293706, + "step": 1712, + "tokens_trained": 0.162734976 + }, + { + "epoch": 0.48624113475177305, + "grad_norm": 1.240332007408142, + "loss": 5.2153, + "lr": 0.0008303496503496504, + "step": 1714, + "tokens_trained": 0.162924992 + }, + { + "epoch": 0.4868085106382979, + "grad_norm": 1.4141086339950562, + "loss": 5.2056, + "lr": 0.00083006993006993, + "step": 1716, + "tokens_trained": 0.163114008 + }, + { + "epoch": 0.4873758865248227, + "grad_norm": 1.321721076965332, + "loss": 5.2223, + "lr": 0.0008297902097902098, + "step": 1718, + "tokens_trained": 0.163304416 + }, + { + "epoch": 0.4879432624113475, + "grad_norm": 1.5437248945236206, + "loss": 5.2727, + "lr": 0.0008295104895104895, + "step": 1720, + "tokens_trained": 0.163493816 + }, + { + "epoch": 0.48851063829787233, + "grad_norm": 1.7218859195709229, + "loss": 5.2323, + "lr": 0.0008292307692307693, + "step": 1722, + "tokens_trained": 0.163683984 + }, + { + "epoch": 0.48907801418439717, + "grad_norm": 1.5534045696258545, + "loss": 5.1983, + "lr": 0.000828951048951049, + "step": 1724, + "tokens_trained": 0.163874968 + }, + { + "epoch": 0.489645390070922, + "grad_norm": 1.3675404787063599, + "loss": 5.2086, + "lr": 0.0008286713286713287, + "step": 1726, + "tokens_trained": 0.164065152 + }, + { + "epoch": 0.4902127659574468, + "grad_norm": 1.5178970098495483, + "loss": 5.2529, + "lr": 0.0008283916083916084, + "step": 1728, + "tokens_trained": 0.164255952 + }, + { + "epoch": 0.4907801418439716, + "grad_norm": 1.4910545349121094, + "loss": 5.2931, + "lr": 0.000828111888111888, + "step": 1730, + "tokens_trained": 0.164447112 + }, + { + "epoch": 0.49134751773049645, + "grad_norm": 1.5647637844085693, + "loss": 5.2603, + "lr": 0.0008278321678321679, + "step": 1732, + "tokens_trained": 0.16463704 + }, + { + "epoch": 0.4919148936170213, + "grad_norm": 1.4607906341552734, + "loss": 5.2702, + "lr": 0.0008275524475524475, + "step": 1734, + "tokens_trained": 0.164827312 + }, + { + "epoch": 0.4924822695035461, + "grad_norm": 1.5806026458740234, + "loss": 5.2356, + "lr": 0.0008272727272727273, + "step": 1736, + "tokens_trained": 0.165015224 + }, + { + "epoch": 0.4930496453900709, + "grad_norm": 1.5417263507843018, + "loss": 5.262, + "lr": 0.000826993006993007, + "step": 1738, + "tokens_trained": 0.16520484 + }, + { + "epoch": 0.49361702127659574, + "grad_norm": 1.511680245399475, + "loss": 5.2634, + "lr": 0.0008267132867132868, + "step": 1740, + "tokens_trained": 0.165393064 + }, + { + "epoch": 0.49418439716312057, + "grad_norm": 1.4468717575073242, + "loss": 5.2452, + "lr": 0.0008264335664335665, + "step": 1742, + "tokens_trained": 0.165584472 + }, + { + "epoch": 0.4947517730496454, + "grad_norm": 1.423187017440796, + "loss": 5.2533, + "lr": 0.0008261538461538461, + "step": 1744, + "tokens_trained": 0.165773768 + }, + { + "epoch": 0.49531914893617024, + "grad_norm": 1.512462854385376, + "loss": 5.2152, + "lr": 0.0008258741258741259, + "step": 1746, + "tokens_trained": 0.165963456 + }, + { + "epoch": 0.495886524822695, + "grad_norm": 1.4620780944824219, + "loss": 5.2511, + "lr": 0.0008255944055944055, + "step": 1748, + "tokens_trained": 0.166152136 + }, + { + "epoch": 0.49645390070921985, + "grad_norm": 1.4943009614944458, + "loss": 5.2829, + "lr": 0.0008253146853146854, + "step": 1750, + "tokens_trained": 0.16634248 + }, + { + "epoch": 0.49645390070921985, + "eval_loss": 5.23966646194458, + "eval_runtime": 20.5954, + "step": 1750, + "tokens_trained": 0.16634248 + }, + { + "epoch": 0.4970212765957447, + "grad_norm": 1.6739267110824585, + "loss": 5.2306, + "lr": 0.000825034965034965, + "step": 1752, + "tokens_trained": 0.166532864 + }, + { + "epoch": 0.4975886524822695, + "grad_norm": 1.6125763654708862, + "loss": 5.2845, + "lr": 0.0008247552447552448, + "step": 1754, + "tokens_trained": 0.166722944 + }, + { + "epoch": 0.4981560283687943, + "grad_norm": 1.5872310400009155, + "loss": 5.2075, + "lr": 0.0008244755244755245, + "step": 1756, + "tokens_trained": 0.16691184 + }, + { + "epoch": 0.49872340425531914, + "grad_norm": 1.4396610260009766, + "loss": 5.2532, + "lr": 0.0008241958041958042, + "step": 1758, + "tokens_trained": 0.167101896 + }, + { + "epoch": 0.49929078014184397, + "grad_norm": 1.363879680633545, + "loss": 5.2252, + "lr": 0.000823916083916084, + "step": 1760, + "tokens_trained": 0.167289384 + }, + { + "epoch": 0.4998581560283688, + "grad_norm": 1.395561695098877, + "loss": 5.2097, + "lr": 0.0008236363636363636, + "step": 1762, + "tokens_trained": 0.167479424 + }, + { + "epoch": 0.5004255319148936, + "grad_norm": 1.413736343383789, + "loss": 5.2283, + "lr": 0.0008233566433566434, + "step": 1764, + "tokens_trained": 0.167668256 + }, + { + "epoch": 0.5009929078014185, + "grad_norm": 1.4240859746932983, + "loss": 5.2574, + "lr": 0.000823076923076923, + "step": 1766, + "tokens_trained": 0.167858616 + }, + { + "epoch": 0.5015602836879433, + "grad_norm": 1.437165379524231, + "loss": 5.2511, + "lr": 0.0008227972027972029, + "step": 1768, + "tokens_trained": 0.168048272 + }, + { + "epoch": 0.502127659574468, + "grad_norm": 1.458575963973999, + "loss": 5.2183, + "lr": 0.0008225174825174825, + "step": 1770, + "tokens_trained": 0.168240184 + }, + { + "epoch": 0.5026950354609929, + "grad_norm": 1.5224673748016357, + "loss": 5.259, + "lr": 0.0008222377622377622, + "step": 1772, + "tokens_trained": 0.168429536 + }, + { + "epoch": 0.5032624113475177, + "grad_norm": 1.578438401222229, + "loss": 5.2108, + "lr": 0.000821958041958042, + "step": 1774, + "tokens_trained": 0.168619312 + }, + { + "epoch": 0.5038297872340426, + "grad_norm": 1.4880632162094116, + "loss": 5.229, + "lr": 0.0008216783216783217, + "step": 1776, + "tokens_trained": 0.168808344 + }, + { + "epoch": 0.5043971631205674, + "grad_norm": 1.3741049766540527, + "loss": 5.2873, + "lr": 0.0008213986013986015, + "step": 1778, + "tokens_trained": 0.168999112 + }, + { + "epoch": 0.5049645390070922, + "grad_norm": 1.4396610260009766, + "loss": 5.3237, + "lr": 0.0008211188811188811, + "step": 1780, + "tokens_trained": 0.169189288 + }, + { + "epoch": 0.505531914893617, + "grad_norm": 1.4296880960464478, + "loss": 5.2228, + "lr": 0.0008208391608391609, + "step": 1782, + "tokens_trained": 0.16937864 + }, + { + "epoch": 0.5060992907801418, + "grad_norm": 1.5704258680343628, + "loss": 5.2569, + "lr": 0.0008205594405594405, + "step": 1784, + "tokens_trained": 0.169569024 + }, + { + "epoch": 0.5066666666666667, + "grad_norm": 1.458261489868164, + "loss": 5.1818, + "lr": 0.0008202797202797203, + "step": 1786, + "tokens_trained": 0.16975932 + }, + { + "epoch": 0.5072340425531915, + "grad_norm": 1.5307244062423706, + "loss": 5.2684, + "lr": 0.00082, + "step": 1788, + "tokens_trained": 0.169949064 + }, + { + "epoch": 0.5078014184397163, + "grad_norm": 1.3966363668441772, + "loss": 5.2125, + "lr": 0.0008197202797202797, + "step": 1790, + "tokens_trained": 0.170139352 + }, + { + "epoch": 0.5083687943262412, + "grad_norm": 1.4094839096069336, + "loss": 5.2518, + "lr": 0.0008194405594405595, + "step": 1792, + "tokens_trained": 0.170330336 + }, + { + "epoch": 0.5089361702127659, + "grad_norm": 1.266122817993164, + "loss": 5.2409, + "lr": 0.0008191608391608392, + "step": 1794, + "tokens_trained": 0.170521848 + }, + { + "epoch": 0.5095035460992908, + "grad_norm": 1.3079488277435303, + "loss": 5.182, + "lr": 0.000818881118881119, + "step": 1796, + "tokens_trained": 0.170710664 + }, + { + "epoch": 0.5100709219858156, + "grad_norm": 1.2961090803146362, + "loss": 5.2456, + "lr": 0.0008186013986013986, + "step": 1798, + "tokens_trained": 0.170900016 + }, + { + "epoch": 0.5106382978723404, + "grad_norm": 1.3402773141860962, + "loss": 5.1888, + "lr": 0.0008183216783216783, + "step": 1800, + "tokens_trained": 0.171089824 + }, + { + "epoch": 0.5112056737588653, + "grad_norm": 1.386769413948059, + "loss": 5.1715, + "lr": 0.000818041958041958, + "step": 1802, + "tokens_trained": 0.171279448 + }, + { + "epoch": 0.51177304964539, + "grad_norm": 1.4280421733856201, + "loss": 5.2131, + "lr": 0.0008177622377622378, + "step": 1804, + "tokens_trained": 0.17147048 + }, + { + "epoch": 0.512340425531915, + "grad_norm": 1.4805412292480469, + "loss": 5.2379, + "lr": 0.0008174825174825175, + "step": 1806, + "tokens_trained": 0.171662264 + }, + { + "epoch": 0.5129078014184397, + "grad_norm": 1.4608936309814453, + "loss": 5.2412, + "lr": 0.0008172027972027972, + "step": 1808, + "tokens_trained": 0.171853176 + }, + { + "epoch": 0.5134751773049645, + "grad_norm": 1.550136923789978, + "loss": 5.1828, + "lr": 0.000816923076923077, + "step": 1810, + "tokens_trained": 0.172043344 + }, + { + "epoch": 0.5140425531914894, + "grad_norm": 1.4756869077682495, + "loss": 5.199, + "lr": 0.0008166433566433567, + "step": 1812, + "tokens_trained": 0.172231952 + }, + { + "epoch": 0.5146099290780142, + "grad_norm": 1.4199044704437256, + "loss": 5.2074, + "lr": 0.0008163636363636364, + "step": 1814, + "tokens_trained": 0.172420376 + }, + { + "epoch": 0.5151773049645391, + "grad_norm": 1.3477959632873535, + "loss": 5.1672, + "lr": 0.0008160839160839161, + "step": 1816, + "tokens_trained": 0.172610248 + }, + { + "epoch": 0.5157446808510638, + "grad_norm": 1.3331218957901, + "loss": 5.2267, + "lr": 0.0008158041958041958, + "step": 1818, + "tokens_trained": 0.172799168 + }, + { + "epoch": 0.5163120567375886, + "grad_norm": 1.2391384840011597, + "loss": 5.2088, + "lr": 0.0008155244755244755, + "step": 1820, + "tokens_trained": 0.172989328 + }, + { + "epoch": 0.5168794326241135, + "grad_norm": 1.3377013206481934, + "loss": 5.2279, + "lr": 0.0008152447552447553, + "step": 1822, + "tokens_trained": 0.173179376 + }, + { + "epoch": 0.5174468085106383, + "grad_norm": 1.285628318786621, + "loss": 5.3006, + "lr": 0.000814965034965035, + "step": 1824, + "tokens_trained": 0.173370408 + }, + { + "epoch": 0.5180141843971631, + "grad_norm": 1.2010120153427124, + "loss": 5.2264, + "lr": 0.0008146853146853147, + "step": 1826, + "tokens_trained": 0.173561144 + }, + { + "epoch": 0.518581560283688, + "grad_norm": 1.2953096628189087, + "loss": 5.1879, + "lr": 0.0008144055944055944, + "step": 1828, + "tokens_trained": 0.173753592 + }, + { + "epoch": 0.5191489361702127, + "grad_norm": 1.256910800933838, + "loss": 5.2402, + "lr": 0.0008141258741258742, + "step": 1830, + "tokens_trained": 0.173943752 + }, + { + "epoch": 0.5197163120567376, + "grad_norm": 1.338755488395691, + "loss": 5.2556, + "lr": 0.0008138461538461539, + "step": 1832, + "tokens_trained": 0.174130504 + }, + { + "epoch": 0.5202836879432624, + "grad_norm": 1.380715012550354, + "loss": 5.2047, + "lr": 0.0008135664335664336, + "step": 1834, + "tokens_trained": 0.174322088 + }, + { + "epoch": 0.5208510638297872, + "grad_norm": 1.4989492893218994, + "loss": 5.1873, + "lr": 0.0008132867132867133, + "step": 1836, + "tokens_trained": 0.17451164 + }, + { + "epoch": 0.5214184397163121, + "grad_norm": 1.3239110708236694, + "loss": 5.202, + "lr": 0.000813006993006993, + "step": 1838, + "tokens_trained": 0.174701896 + }, + { + "epoch": 0.5219858156028369, + "grad_norm": 1.397745132446289, + "loss": 5.2259, + "lr": 0.0008127272727272728, + "step": 1840, + "tokens_trained": 0.174892336 + }, + { + "epoch": 0.5225531914893617, + "grad_norm": 1.3992305994033813, + "loss": 5.1771, + "lr": 0.0008124475524475524, + "step": 1842, + "tokens_trained": 0.17508276 + }, + { + "epoch": 0.5231205673758865, + "grad_norm": 1.38923180103302, + "loss": 5.1981, + "lr": 0.0008121678321678322, + "step": 1844, + "tokens_trained": 0.175273272 + }, + { + "epoch": 0.5236879432624113, + "grad_norm": 1.478642225265503, + "loss": 5.2533, + "lr": 0.0008118881118881119, + "step": 1846, + "tokens_trained": 0.175462352 + }, + { + "epoch": 0.5242553191489362, + "grad_norm": 1.332709789276123, + "loss": 5.2205, + "lr": 0.0008116083916083917, + "step": 1848, + "tokens_trained": 0.175648128 + }, + { + "epoch": 0.524822695035461, + "grad_norm": 1.4612590074539185, + "loss": 5.2207, + "lr": 0.0008113286713286714, + "step": 1850, + "tokens_trained": 0.175837712 + }, + { + "epoch": 0.5253900709219859, + "grad_norm": 1.4682700634002686, + "loss": 5.2576, + "lr": 0.000811048951048951, + "step": 1852, + "tokens_trained": 0.176029512 + }, + { + "epoch": 0.5259574468085106, + "grad_norm": 1.3380264043807983, + "loss": 5.2435, + "lr": 0.0008107692307692308, + "step": 1854, + "tokens_trained": 0.176220432 + }, + { + "epoch": 0.5265248226950354, + "grad_norm": 1.2452281713485718, + "loss": 5.2973, + "lr": 0.0008104895104895104, + "step": 1856, + "tokens_trained": 0.176412144 + }, + { + "epoch": 0.5270921985815603, + "grad_norm": 1.392592191696167, + "loss": 5.2028, + "lr": 0.0008102097902097903, + "step": 1858, + "tokens_trained": 0.17660144 + }, + { + "epoch": 0.5276595744680851, + "grad_norm": 1.4258657693862915, + "loss": 5.2342, + "lr": 0.0008099300699300699, + "step": 1860, + "tokens_trained": 0.176790424 + }, + { + "epoch": 0.52822695035461, + "grad_norm": 1.4627033472061157, + "loss": 5.1732, + "lr": 0.0008096503496503497, + "step": 1862, + "tokens_trained": 0.176983296 + }, + { + "epoch": 0.5287943262411348, + "grad_norm": 1.4448645114898682, + "loss": 5.2001, + "lr": 0.0008093706293706294, + "step": 1864, + "tokens_trained": 0.177174544 + }, + { + "epoch": 0.5293617021276595, + "grad_norm": 1.3879749774932861, + "loss": 5.1642, + "lr": 0.0008090909090909092, + "step": 1866, + "tokens_trained": 0.17736428 + }, + { + "epoch": 0.5299290780141844, + "grad_norm": 1.2791417837142944, + "loss": 5.1975, + "lr": 0.0008088111888111889, + "step": 1868, + "tokens_trained": 0.177553752 + }, + { + "epoch": 0.5304964539007092, + "grad_norm": 1.3620632886886597, + "loss": 5.1742, + "lr": 0.0008085314685314685, + "step": 1870, + "tokens_trained": 0.177746448 + }, + { + "epoch": 0.531063829787234, + "grad_norm": 1.2759565114974976, + "loss": 5.2076, + "lr": 0.0008082517482517483, + "step": 1872, + "tokens_trained": 0.177937888 + }, + { + "epoch": 0.5316312056737589, + "grad_norm": 1.3390915393829346, + "loss": 5.2387, + "lr": 0.0008079720279720279, + "step": 1874, + "tokens_trained": 0.178127776 + }, + { + "epoch": 0.5319148936170213, + "eval_loss": 5.228371620178223, + "eval_runtime": 20.9372, + "step": 1875, + "tokens_trained": 0.17822376 + }, + { + "epoch": 0.5321985815602837, + "grad_norm": 1.3872885704040527, + "loss": 5.2053, + "lr": 0.0008076923076923078, + "step": 1876, + "tokens_trained": 0.178318616 + }, + { + "epoch": 0.5327659574468085, + "grad_norm": 1.4238568544387817, + "loss": 5.2091, + "lr": 0.0008074125874125874, + "step": 1878, + "tokens_trained": 0.178509272 + }, + { + "epoch": 0.5333333333333333, + "grad_norm": 1.3352588415145874, + "loss": 5.2471, + "lr": 0.0008071328671328671, + "step": 1880, + "tokens_trained": 0.178698016 + }, + { + "epoch": 0.5339007092198581, + "grad_norm": 1.2931993007659912, + "loss": 5.2315, + "lr": 0.0008068531468531469, + "step": 1882, + "tokens_trained": 0.17888628 + }, + { + "epoch": 0.534468085106383, + "grad_norm": 1.3475919961929321, + "loss": 5.2337, + "lr": 0.0008065734265734265, + "step": 1884, + "tokens_trained": 0.179076944 + }, + { + "epoch": 0.5350354609929078, + "grad_norm": 1.3263812065124512, + "loss": 5.2017, + "lr": 0.0008062937062937064, + "step": 1886, + "tokens_trained": 0.179266128 + }, + { + "epoch": 0.5356028368794327, + "grad_norm": 1.3956594467163086, + "loss": 5.1907, + "lr": 0.000806013986013986, + "step": 1888, + "tokens_trained": 0.179454848 + }, + { + "epoch": 0.5361702127659574, + "grad_norm": 1.4399393796920776, + "loss": 5.216, + "lr": 0.0008057342657342658, + "step": 1890, + "tokens_trained": 0.179643992 + }, + { + "epoch": 0.5367375886524822, + "grad_norm": 1.278714656829834, + "loss": 5.1689, + "lr": 0.0008054545454545454, + "step": 1892, + "tokens_trained": 0.179831416 + }, + { + "epoch": 0.5373049645390071, + "grad_norm": 1.3517796993255615, + "loss": 5.1319, + "lr": 0.0008051748251748253, + "step": 1894, + "tokens_trained": 0.180022528 + }, + { + "epoch": 0.5378723404255319, + "grad_norm": 1.2710460424423218, + "loss": 5.1619, + "lr": 0.0008048951048951049, + "step": 1896, + "tokens_trained": 0.180212936 + }, + { + "epoch": 0.5384397163120568, + "grad_norm": 1.3603075742721558, + "loss": 5.1615, + "lr": 0.0008046153846153846, + "step": 1898, + "tokens_trained": 0.180404648 + }, + { + "epoch": 0.5390070921985816, + "grad_norm": 1.422122836112976, + "loss": 5.1801, + "lr": 0.0008043356643356644, + "step": 1900, + "tokens_trained": 0.18059388 + }, + { + "epoch": 0.5395744680851063, + "grad_norm": 1.4242218732833862, + "loss": 5.2367, + "lr": 0.000804055944055944, + "step": 1902, + "tokens_trained": 0.180783248 + }, + { + "epoch": 0.5401418439716312, + "grad_norm": 1.4476134777069092, + "loss": 5.252, + "lr": 0.0008037762237762239, + "step": 1904, + "tokens_trained": 0.180971152 + }, + { + "epoch": 0.540709219858156, + "grad_norm": 1.4724863767623901, + "loss": 5.2042, + "lr": 0.0008034965034965035, + "step": 1906, + "tokens_trained": 0.181159992 + }, + { + "epoch": 0.5412765957446809, + "grad_norm": 1.4014806747436523, + "loss": 5.2514, + "lr": 0.0008032167832167832, + "step": 1908, + "tokens_trained": 0.18135032 + }, + { + "epoch": 0.5418439716312057, + "grad_norm": 1.3511682748794556, + "loss": 5.2036, + "lr": 0.0008029370629370629, + "step": 1910, + "tokens_trained": 0.181540312 + }, + { + "epoch": 0.5424113475177305, + "grad_norm": 1.3011739253997803, + "loss": 5.24, + "lr": 0.0008026573426573427, + "step": 1912, + "tokens_trained": 0.181731104 + }, + { + "epoch": 0.5429787234042553, + "grad_norm": 1.2753015756607056, + "loss": 5.25, + "lr": 0.0008023776223776224, + "step": 1914, + "tokens_trained": 0.18192008 + }, + { + "epoch": 0.5435460992907801, + "grad_norm": 1.4685192108154297, + "loss": 5.1619, + "lr": 0.0008020979020979021, + "step": 1916, + "tokens_trained": 0.182110072 + }, + { + "epoch": 0.544113475177305, + "grad_norm": 1.4695900678634644, + "loss": 5.2626, + "lr": 0.0008018181818181818, + "step": 1918, + "tokens_trained": 0.182300224 + }, + { + "epoch": 0.5446808510638298, + "grad_norm": 1.4895613193511963, + "loss": 5.1766, + "lr": 0.0008015384615384615, + "step": 1920, + "tokens_trained": 0.182490712 + }, + { + "epoch": 0.5452482269503546, + "grad_norm": 1.3073184490203857, + "loss": 5.2281, + "lr": 0.0008012587412587414, + "step": 1922, + "tokens_trained": 0.182681168 + }, + { + "epoch": 0.5458156028368795, + "grad_norm": 1.2414125204086304, + "loss": 5.2099, + "lr": 0.000800979020979021, + "step": 1924, + "tokens_trained": 0.182870504 + }, + { + "epoch": 0.5463829787234042, + "grad_norm": 1.2407176494598389, + "loss": 5.1116, + "lr": 0.0008006993006993007, + "step": 1926, + "tokens_trained": 0.1830618 + }, + { + "epoch": 0.546950354609929, + "grad_norm": 1.4507744312286377, + "loss": 5.1658, + "lr": 0.0008004195804195804, + "step": 1928, + "tokens_trained": 0.183250072 + }, + { + "epoch": 0.5475177304964539, + "grad_norm": 1.348907232284546, + "loss": 5.231, + "lr": 0.0008001398601398602, + "step": 1930, + "tokens_trained": 0.18344004 + }, + { + "epoch": 0.5480851063829787, + "grad_norm": 1.4393324851989746, + "loss": 5.2393, + "lr": 0.0007998601398601399, + "step": 1932, + "tokens_trained": 0.183630032 + }, + { + "epoch": 0.5486524822695036, + "grad_norm": 1.3569602966308594, + "loss": 5.2068, + "lr": 0.0007995804195804196, + "step": 1934, + "tokens_trained": 0.183820816 + }, + { + "epoch": 0.5492198581560284, + "grad_norm": 1.362021803855896, + "loss": 5.1641, + "lr": 0.0007993006993006992, + "step": 1936, + "tokens_trained": 0.184009824 + }, + { + "epoch": 0.5497872340425531, + "grad_norm": 1.2926445007324219, + "loss": 5.1983, + "lr": 0.000799020979020979, + "step": 1938, + "tokens_trained": 0.184199544 + }, + { + "epoch": 0.550354609929078, + "grad_norm": 1.3065440654754639, + "loss": 5.3009, + "lr": 0.0007987412587412588, + "step": 1940, + "tokens_trained": 0.1843906 + }, + { + "epoch": 0.5509219858156028, + "grad_norm": 1.3288060426712036, + "loss": 5.2347, + "lr": 0.0007984615384615385, + "step": 1942, + "tokens_trained": 0.184580304 + }, + { + "epoch": 0.5514893617021277, + "grad_norm": 1.4742496013641357, + "loss": 5.1497, + "lr": 0.0007981818181818182, + "step": 1944, + "tokens_trained": 0.184771832 + }, + { + "epoch": 0.5520567375886525, + "grad_norm": 1.3907397985458374, + "loss": 5.2001, + "lr": 0.0007979020979020979, + "step": 1946, + "tokens_trained": 0.184963744 + }, + { + "epoch": 0.5526241134751773, + "grad_norm": 1.3324332237243652, + "loss": 5.2056, + "lr": 0.0007976223776223777, + "step": 1948, + "tokens_trained": 0.185152248 + }, + { + "epoch": 0.5531914893617021, + "grad_norm": 1.258155345916748, + "loss": 5.1999, + "lr": 0.0007973426573426573, + "step": 1950, + "tokens_trained": 0.18534196 + }, + { + "epoch": 0.5537588652482269, + "grad_norm": 1.3515956401824951, + "loss": 5.1988, + "lr": 0.0007970629370629371, + "step": 1952, + "tokens_trained": 0.18553156 + }, + { + "epoch": 0.5543262411347518, + "grad_norm": 1.535507321357727, + "loss": 5.2198, + "lr": 0.0007967832167832167, + "step": 1954, + "tokens_trained": 0.185719792 + }, + { + "epoch": 0.5548936170212766, + "grad_norm": 1.3124226331710815, + "loss": 5.1468, + "lr": 0.0007965034965034965, + "step": 1956, + "tokens_trained": 0.18591288 + }, + { + "epoch": 0.5554609929078014, + "grad_norm": 1.2720654010772705, + "loss": 5.1939, + "lr": 0.0007962237762237763, + "step": 1958, + "tokens_trained": 0.186102344 + }, + { + "epoch": 0.5560283687943263, + "grad_norm": 1.2731753587722778, + "loss": 5.2063, + "lr": 0.000795944055944056, + "step": 1960, + "tokens_trained": 0.186291976 + }, + { + "epoch": 0.556595744680851, + "grad_norm": 1.3020576238632202, + "loss": 5.266, + "lr": 0.0007956643356643357, + "step": 1962, + "tokens_trained": 0.186483504 + }, + { + "epoch": 0.5571631205673759, + "grad_norm": 1.300626277923584, + "loss": 5.2159, + "lr": 0.0007953846153846153, + "step": 1964, + "tokens_trained": 0.18667372 + }, + { + "epoch": 0.5577304964539007, + "grad_norm": 1.3075426816940308, + "loss": 5.2136, + "lr": 0.0007951048951048952, + "step": 1966, + "tokens_trained": 0.186864808 + }, + { + "epoch": 0.5582978723404255, + "grad_norm": 1.4623394012451172, + "loss": 5.2081, + "lr": 0.0007948251748251748, + "step": 1968, + "tokens_trained": 0.187056272 + }, + { + "epoch": 0.5588652482269504, + "grad_norm": 1.4950625896453857, + "loss": 5.1885, + "lr": 0.0007945454545454546, + "step": 1970, + "tokens_trained": 0.187244464 + }, + { + "epoch": 0.5594326241134752, + "grad_norm": 1.517152190208435, + "loss": 5.2558, + "lr": 0.0007942657342657342, + "step": 1972, + "tokens_trained": 0.187433216 + }, + { + "epoch": 0.56, + "grad_norm": 1.4226372241973877, + "loss": 5.236, + "lr": 0.000793986013986014, + "step": 1974, + "tokens_trained": 0.187622632 + }, + { + "epoch": 0.5605673758865248, + "grad_norm": 1.3692735433578491, + "loss": 5.2089, + "lr": 0.0007937062937062938, + "step": 1976, + "tokens_trained": 0.18781324 + }, + { + "epoch": 0.5611347517730496, + "grad_norm": 1.3344841003417969, + "loss": 5.2052, + "lr": 0.0007934265734265734, + "step": 1978, + "tokens_trained": 0.188002488 + }, + { + "epoch": 0.5617021276595745, + "grad_norm": 1.3929632902145386, + "loss": 5.2353, + "lr": 0.0007931468531468532, + "step": 1980, + "tokens_trained": 0.188194712 + }, + { + "epoch": 0.5622695035460993, + "grad_norm": 1.3147000074386597, + "loss": 5.2071, + "lr": 0.0007928671328671328, + "step": 1982, + "tokens_trained": 0.188387056 + }, + { + "epoch": 0.5628368794326241, + "grad_norm": 1.351483702659607, + "loss": 5.2196, + "lr": 0.0007925874125874127, + "step": 1984, + "tokens_trained": 0.188579048 + }, + { + "epoch": 0.563404255319149, + "grad_norm": 1.3840581178665161, + "loss": 5.1889, + "lr": 0.0007923076923076923, + "step": 1986, + "tokens_trained": 0.18876896 + }, + { + "epoch": 0.5639716312056737, + "grad_norm": 1.3427214622497559, + "loss": 5.192, + "lr": 0.000792027972027972, + "step": 1988, + "tokens_trained": 0.18895832 + }, + { + "epoch": 0.5645390070921986, + "grad_norm": 1.2931344509124756, + "loss": 5.1942, + "lr": 0.0007917482517482517, + "step": 1990, + "tokens_trained": 0.18915036 + }, + { + "epoch": 0.5651063829787234, + "grad_norm": 1.2408664226531982, + "loss": 5.2014, + "lr": 0.0007914685314685314, + "step": 1992, + "tokens_trained": 0.189339784 + }, + { + "epoch": 0.5656737588652482, + "grad_norm": 1.342760682106018, + "loss": 5.2056, + "lr": 0.0007911888111888113, + "step": 1994, + "tokens_trained": 0.189530776 + }, + { + "epoch": 0.5662411347517731, + "grad_norm": 1.2647815942764282, + "loss": 5.2338, + "lr": 0.0007909090909090909, + "step": 1996, + "tokens_trained": 0.189720312 + }, + { + "epoch": 0.5668085106382978, + "grad_norm": 1.1956689357757568, + "loss": 5.1464, + "lr": 0.0007906293706293707, + "step": 1998, + "tokens_trained": 0.189909592 + }, + { + "epoch": 0.5673758865248227, + "grad_norm": 1.287185549736023, + "loss": 5.1919, + "lr": 0.0007903496503496503, + "step": 2000, + "tokens_trained": 0.190100544 + }, + { + "epoch": 0.5673758865248227, + "eval_loss": 5.208409309387207, + "eval_runtime": 21.1643, + "step": 2000, + "tokens_trained": 0.190100544 + } + ], + "logging_steps": 2, + "max_steps": 7650, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 125, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 0.0, + "train_batch_size": 32, + "trial_name": null, + "trial_params": null +}