| { | |
| "best_metric": 1.47908163, | |
| "best_model_checkpoint": "D:\\_____NEW_NN\\LLM\\MiniCPM-V\\finetune\\output\\phi3-vision-128k-instruct\\v9-20240710-235159\\checkpoint-500", | |
| "epoch": 2.8828828828828827, | |
| "eval_steps": 50, | |
| "global_step": 800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "acc": 0.4856407, | |
| "epoch": 0.0036036036036036037, | |
| "grad_norm": 0.734375, | |
| "learning_rate": 2.4107142857142856e-06, | |
| "loss": 2.42667556, | |
| "memory(GiB)": 18.11, | |
| "step": 1, | |
| "train_speed(iter/s)": 0.072451 | |
| }, | |
| { | |
| "acc": 0.50815099, | |
| "epoch": 0.018018018018018018, | |
| "grad_norm": 0.671875, | |
| "learning_rate": 1.2053571428571429e-05, | |
| "loss": 2.28746271, | |
| "memory(GiB)": 19.3, | |
| "step": 5, | |
| "train_speed(iter/s)": 0.081978 | |
| }, | |
| { | |
| "acc": 0.50680609, | |
| "epoch": 0.036036036036036036, | |
| "grad_norm": 0.76953125, | |
| "learning_rate": 2.4107142857142858e-05, | |
| "loss": 2.29894772, | |
| "memory(GiB)": 19.3, | |
| "step": 10, | |
| "train_speed(iter/s)": 0.084125 | |
| }, | |
| { | |
| "acc": 0.51412601, | |
| "epoch": 0.05405405405405406, | |
| "grad_norm": 0.76171875, | |
| "learning_rate": 3.616071428571428e-05, | |
| "loss": 2.34161263, | |
| "memory(GiB)": 19.7, | |
| "step": 15, | |
| "train_speed(iter/s)": 0.08456 | |
| }, | |
| { | |
| "acc": 0.52338777, | |
| "epoch": 0.07207207207207207, | |
| "grad_norm": 0.6015625, | |
| "learning_rate": 4.8214285714285716e-05, | |
| "loss": 2.23036633, | |
| "memory(GiB)": 19.88, | |
| "step": 20, | |
| "train_speed(iter/s)": 0.084117 | |
| }, | |
| { | |
| "acc": 0.55944238, | |
| "epoch": 0.09009009009009009, | |
| "grad_norm": 0.66796875, | |
| "learning_rate": 6.026785714285715e-05, | |
| "loss": 2.01084595, | |
| "memory(GiB)": 19.93, | |
| "step": 25, | |
| "train_speed(iter/s)": 0.084444 | |
| }, | |
| { | |
| "acc": 0.57758675, | |
| "epoch": 0.10810810810810811, | |
| "grad_norm": 0.765625, | |
| "learning_rate": 7.232142857142856e-05, | |
| "loss": 1.94100876, | |
| "memory(GiB)": 20.21, | |
| "step": 30, | |
| "train_speed(iter/s)": 0.085158 | |
| }, | |
| { | |
| "acc": 0.5666451, | |
| "epoch": 0.12612612612612611, | |
| "grad_norm": 0.796875, | |
| "learning_rate": 8.4375e-05, | |
| "loss": 1.96992569, | |
| "memory(GiB)": 19.42, | |
| "step": 35, | |
| "train_speed(iter/s)": 0.085562 | |
| }, | |
| { | |
| "acc": 0.55766659, | |
| "epoch": 0.14414414414414414, | |
| "grad_norm": 0.828125, | |
| "learning_rate": 9.642857142857143e-05, | |
| "loss": 2.01305885, | |
| "memory(GiB)": 19.71, | |
| "step": 40, | |
| "train_speed(iter/s)": 0.0857 | |
| }, | |
| { | |
| "acc": 0.56964116, | |
| "epoch": 0.16216216216216217, | |
| "grad_norm": 0.83203125, | |
| "learning_rate": 0.00010848214285714286, | |
| "loss": 1.925914, | |
| "memory(GiB)": 19.68, | |
| "step": 45, | |
| "train_speed(iter/s)": 0.08577 | |
| }, | |
| { | |
| "acc": 0.56270452, | |
| "epoch": 0.18018018018018017, | |
| "grad_norm": 0.9375, | |
| "learning_rate": 0.0001205357142857143, | |
| "loss": 1.94923038, | |
| "memory(GiB)": 19.65, | |
| "step": 50, | |
| "train_speed(iter/s)": 0.085942 | |
| }, | |
| { | |
| "epoch": 0.18018018018018017, | |
| "eval_acc": 0.5890983000739098, | |
| "eval_loss": 1.795773983001709, | |
| "eval_runtime": 136.6505, | |
| "eval_samples_per_second": 1.105, | |
| "eval_steps_per_second": 0.556, | |
| "step": 50 | |
| }, | |
| { | |
| "acc": 0.57772484, | |
| "epoch": 0.1981981981981982, | |
| "grad_norm": 0.7265625, | |
| "learning_rate": 0.00013258928571428571, | |
| "loss": 1.86195869, | |
| "memory(GiB)": 23.11, | |
| "step": 55, | |
| "train_speed(iter/s)": 0.070857 | |
| }, | |
| { | |
| "acc": 0.59196444, | |
| "epoch": 0.21621621621621623, | |
| "grad_norm": 0.8125, | |
| "learning_rate": 0.00013499518432841625, | |
| "loss": 1.74724998, | |
| "memory(GiB)": 19.42, | |
| "step": 60, | |
| "train_speed(iter/s)": 0.071911 | |
| }, | |
| { | |
| "acc": 0.57253065, | |
| "epoch": 0.23423423423423423, | |
| "grad_norm": 0.69921875, | |
| "learning_rate": 0.00013497562184025362, | |
| "loss": 1.87580814, | |
| "memory(GiB)": 19.61, | |
| "step": 65, | |
| "train_speed(iter/s)": 0.072807 | |
| }, | |
| { | |
| "acc": 0.59546819, | |
| "epoch": 0.25225225225225223, | |
| "grad_norm": 0.73046875, | |
| "learning_rate": 0.00013494101591406666, | |
| "loss": 1.73464546, | |
| "memory(GiB)": 19.58, | |
| "step": 70, | |
| "train_speed(iter/s)": 0.073652 | |
| }, | |
| { | |
| "acc": 0.59667702, | |
| "epoch": 0.2702702702702703, | |
| "grad_norm": 0.8203125, | |
| "learning_rate": 0.00013489137426511745, | |
| "loss": 1.69518318, | |
| "memory(GiB)": 18.19, | |
| "step": 75, | |
| "train_speed(iter/s)": 0.074445 | |
| }, | |
| { | |
| "acc": 0.61824327, | |
| "epoch": 0.2882882882882883, | |
| "grad_norm": 0.828125, | |
| "learning_rate": 0.00013482670796082633, | |
| "loss": 1.64374161, | |
| "memory(GiB)": 19.52, | |
| "step": 80, | |
| "train_speed(iter/s)": 0.075071 | |
| }, | |
| { | |
| "acc": 0.60798159, | |
| "epoch": 0.3063063063063063, | |
| "grad_norm": 0.7734375, | |
| "learning_rate": 0.00013474703141830443, | |
| "loss": 1.68669338, | |
| "memory(GiB)": 19.57, | |
| "step": 85, | |
| "train_speed(iter/s)": 0.07562 | |
| }, | |
| { | |
| "acc": 0.5981144, | |
| "epoch": 0.32432432432432434, | |
| "grad_norm": 0.80078125, | |
| "learning_rate": 0.00013465236240113953, | |
| "loss": 1.701264, | |
| "memory(GiB)": 20.19, | |
| "step": 90, | |
| "train_speed(iter/s)": 0.076188 | |
| }, | |
| { | |
| "acc": 0.59871612, | |
| "epoch": 0.34234234234234234, | |
| "grad_norm": 1.0234375, | |
| "learning_rate": 0.00013454272201543564, | |
| "loss": 1.76608849, | |
| "memory(GiB)": 19.35, | |
| "step": 95, | |
| "train_speed(iter/s)": 0.076637 | |
| }, | |
| { | |
| "acc": 0.61396523, | |
| "epoch": 0.36036036036036034, | |
| "grad_norm": 0.7109375, | |
| "learning_rate": 0.00013441813470510747, | |
| "loss": 1.61449242, | |
| "memory(GiB)": 19.69, | |
| "step": 100, | |
| "train_speed(iter/s)": 0.077075 | |
| }, | |
| { | |
| "epoch": 0.36036036036036034, | |
| "eval_acc": 0.6091648189209165, | |
| "eval_loss": 1.6449466943740845, | |
| "eval_runtime": 134.5726, | |
| "eval_samples_per_second": 1.122, | |
| "eval_steps_per_second": 0.565, | |
| "step": 100 | |
| }, | |
| { | |
| "acc": 0.61147785, | |
| "epoch": 0.3783783783783784, | |
| "grad_norm": 0.69921875, | |
| "learning_rate": 0.00013427862824643083, | |
| "loss": 1.60589867, | |
| "memory(GiB)": 21.03, | |
| "step": 105, | |
| "train_speed(iter/s)": 0.070426 | |
| }, | |
| { | |
| "acc": 0.6038115, | |
| "epoch": 0.3963963963963964, | |
| "grad_norm": 0.88671875, | |
| "learning_rate": 0.00013412423374184996, | |
| "loss": 1.69055023, | |
| "memory(GiB)": 19.44, | |
| "step": 110, | |
| "train_speed(iter/s)": 0.07105 | |
| }, | |
| { | |
| "acc": 0.62303677, | |
| "epoch": 0.4144144144144144, | |
| "grad_norm": 0.84375, | |
| "learning_rate": 0.00013395498561304334, | |
| "loss": 1.5716897, | |
| "memory(GiB)": 19.27, | |
| "step": 115, | |
| "train_speed(iter/s)": 0.071618 | |
| }, | |
| { | |
| "acc": 0.6214046, | |
| "epoch": 0.43243243243243246, | |
| "grad_norm": 0.640625, | |
| "learning_rate": 0.00013377092159324956, | |
| "loss": 1.57531881, | |
| "memory(GiB)": 19.36, | |
| "step": 120, | |
| "train_speed(iter/s)": 0.07209 | |
| }, | |
| { | |
| "acc": 0.58676672, | |
| "epoch": 0.45045045045045046, | |
| "grad_norm": 0.68359375, | |
| "learning_rate": 0.00013357208271885473, | |
| "loss": 1.74933128, | |
| "memory(GiB)": 19.32, | |
| "step": 125, | |
| "train_speed(iter/s)": 0.072581 | |
| }, | |
| { | |
| "acc": 0.59380612, | |
| "epoch": 0.46846846846846846, | |
| "grad_norm": 0.7890625, | |
| "learning_rate": 0.00013335851332024374, | |
| "loss": 1.69583378, | |
| "memory(GiB)": 20.18, | |
| "step": 130, | |
| "train_speed(iter/s)": 0.073016 | |
| }, | |
| { | |
| "acc": 0.62007999, | |
| "epoch": 0.4864864864864865, | |
| "grad_norm": 0.73828125, | |
| "learning_rate": 0.0001331302610119168, | |
| "loss": 1.60020466, | |
| "memory(GiB)": 19.52, | |
| "step": 135, | |
| "train_speed(iter/s)": 0.073417 | |
| }, | |
| { | |
| "acc": 0.6116991, | |
| "epoch": 0.5045045045045045, | |
| "grad_norm": 1.1015625, | |
| "learning_rate": 0.00013288737668187408, | |
| "loss": 1.62470894, | |
| "memory(GiB)": 19.47, | |
| "step": 140, | |
| "train_speed(iter/s)": 0.073817 | |
| }, | |
| { | |
| "acc": 0.60051751, | |
| "epoch": 0.5225225225225225, | |
| "grad_norm": 0.87109375, | |
| "learning_rate": 0.00013262991448027034, | |
| "loss": 1.6651041, | |
| "memory(GiB)": 19.42, | |
| "step": 145, | |
| "train_speed(iter/s)": 0.074194 | |
| }, | |
| { | |
| "acc": 0.60736594, | |
| "epoch": 0.5405405405405406, | |
| "grad_norm": 0.76953125, | |
| "learning_rate": 0.00013235793180734238, | |
| "loss": 1.64281559, | |
| "memory(GiB)": 19.53, | |
| "step": 150, | |
| "train_speed(iter/s)": 0.074547 | |
| }, | |
| { | |
| "epoch": 0.5405405405405406, | |
| "eval_acc": 0.6190317812269032, | |
| "eval_loss": 1.5917434692382812, | |
| "eval_runtime": 135.0141, | |
| "eval_samples_per_second": 1.118, | |
| "eval_steps_per_second": 0.563, | |
| "step": 150 | |
| }, | |
| { | |
| "acc": 0.61663914, | |
| "epoch": 0.5585585585585585, | |
| "grad_norm": 1.0625, | |
| "learning_rate": 0.00013207148930061195, | |
| "loss": 1.60914173, | |
| "memory(GiB)": 23.05, | |
| "step": 155, | |
| "train_speed(iter/s)": 0.070306 | |
| }, | |
| { | |
| "acc": 0.60967774, | |
| "epoch": 0.5765765765765766, | |
| "grad_norm": 0.76953125, | |
| "learning_rate": 0.00013177065082136668, | |
| "loss": 1.59582939, | |
| "memory(GiB)": 19.47, | |
| "step": 160, | |
| "train_speed(iter/s)": 0.070712 | |
| }, | |
| { | |
| "acc": 0.63630972, | |
| "epoch": 0.5945945945945946, | |
| "grad_norm": 0.70703125, | |
| "learning_rate": 0.00013145548344042262, | |
| "loss": 1.50356016, | |
| "memory(GiB)": 19.62, | |
| "step": 165, | |
| "train_speed(iter/s)": 0.071104 | |
| }, | |
| { | |
| "acc": 0.60439692, | |
| "epoch": 0.6126126126126126, | |
| "grad_norm": 0.73046875, | |
| "learning_rate": 0.00013112605742317095, | |
| "loss": 1.67050171, | |
| "memory(GiB)": 19.41, | |
| "step": 170, | |
| "train_speed(iter/s)": 0.071478 | |
| }, | |
| { | |
| "acc": 0.62380457, | |
| "epoch": 0.6306306306306306, | |
| "grad_norm": 0.76171875, | |
| "learning_rate": 0.0001307824462139125, | |
| "loss": 1.53042831, | |
| "memory(GiB)": 19.5, | |
| "step": 175, | |
| "train_speed(iter/s)": 0.071843 | |
| }, | |
| { | |
| "acc": 0.61549187, | |
| "epoch": 0.6486486486486487, | |
| "grad_norm": 0.7578125, | |
| "learning_rate": 0.00013042472641948386, | |
| "loss": 1.59476538, | |
| "memory(GiB)": 19.53, | |
| "step": 180, | |
| "train_speed(iter/s)": 0.072168 | |
| }, | |
| { | |
| "acc": 0.64418182, | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 1.1796875, | |
| "learning_rate": 0.0001300529777921779, | |
| "loss": 1.47999802, | |
| "memory(GiB)": 19.32, | |
| "step": 185, | |
| "train_speed(iter/s)": 0.072501 | |
| }, | |
| { | |
| "acc": 0.62201657, | |
| "epoch": 0.6846846846846847, | |
| "grad_norm": 0.6484375, | |
| "learning_rate": 0.00012966728321196346, | |
| "loss": 1.5685544, | |
| "memory(GiB)": 19.47, | |
| "step": 190, | |
| "train_speed(iter/s)": 0.072821 | |
| }, | |
| { | |
| "acc": 0.61418505, | |
| "epoch": 0.7027027027027027, | |
| "grad_norm": 0.8984375, | |
| "learning_rate": 0.00012926772866800757, | |
| "loss": 1.6284462, | |
| "memory(GiB)": 19.45, | |
| "step": 195, | |
| "train_speed(iter/s)": 0.073127 | |
| }, | |
| { | |
| "acc": 0.62820964, | |
| "epoch": 0.7207207207207207, | |
| "grad_norm": 0.8515625, | |
| "learning_rate": 0.00012885440323950434, | |
| "loss": 1.54364405, | |
| "memory(GiB)": 19.53, | |
| "step": 200, | |
| "train_speed(iter/s)": 0.073413 | |
| }, | |
| { | |
| "epoch": 0.7207207207207207, | |
| "eval_acc": 0.6269770879526977, | |
| "eval_loss": 1.5466336011886597, | |
| "eval_runtime": 134.7868, | |
| "eval_samples_per_second": 1.12, | |
| "eval_steps_per_second": 0.564, | |
| "step": 200 | |
| }, | |
| { | |
| "acc": 0.6605804, | |
| "epoch": 0.7387387387387387, | |
| "grad_norm": 0.7578125, | |
| "learning_rate": 0.00012842739907581525, | |
| "loss": 1.42957153, | |
| "memory(GiB)": 23.0, | |
| "step": 205, | |
| "train_speed(iter/s)": 0.070232 | |
| }, | |
| { | |
| "acc": 0.61267309, | |
| "epoch": 0.7567567567567568, | |
| "grad_norm": 0.90234375, | |
| "learning_rate": 0.00012798681137592477, | |
| "loss": 1.62853241, | |
| "memory(GiB)": 17.96, | |
| "step": 210, | |
| "train_speed(iter/s)": 0.070571 | |
| }, | |
| { | |
| "acc": 0.63069816, | |
| "epoch": 0.7747747747747747, | |
| "grad_norm": 0.89453125, | |
| "learning_rate": 0.00012753273836721597, | |
| "loss": 1.56295233, | |
| "memory(GiB)": 19.4, | |
| "step": 215, | |
| "train_speed(iter/s)": 0.070892 | |
| }, | |
| { | |
| "acc": 0.60362072, | |
| "epoch": 0.7927927927927928, | |
| "grad_norm": 1.0703125, | |
| "learning_rate": 0.00012706528128357127, | |
| "loss": 1.63038826, | |
| "memory(GiB)": 19.37, | |
| "step": 220, | |
| "train_speed(iter/s)": 0.071181 | |
| }, | |
| { | |
| "acc": 0.62272639, | |
| "epoch": 0.8108108108108109, | |
| "grad_norm": 0.8828125, | |
| "learning_rate": 0.00012658454434280253, | |
| "loss": 1.5756237, | |
| "memory(GiB)": 19.62, | |
| "step": 225, | |
| "train_speed(iter/s)": 0.071466 | |
| }, | |
| { | |
| "acc": 0.59926658, | |
| "epoch": 0.8288288288288288, | |
| "grad_norm": 0.75390625, | |
| "learning_rate": 0.00012609063472341633, | |
| "loss": 1.60503426, | |
| "memory(GiB)": 19.63, | |
| "step": 230, | |
| "train_speed(iter/s)": 0.071751 | |
| }, | |
| { | |
| "acc": 0.60133944, | |
| "epoch": 0.8468468468468469, | |
| "grad_norm": 1.3515625, | |
| "learning_rate": 0.0001255836625407187, | |
| "loss": 1.64450779, | |
| "memory(GiB)": 19.31, | |
| "step": 235, | |
| "train_speed(iter/s)": 0.072034 | |
| }, | |
| { | |
| "acc": 0.64020758, | |
| "epoch": 0.8648648648648649, | |
| "grad_norm": 0.9375, | |
| "learning_rate": 0.00012506374082226534, | |
| "loss": 1.47053967, | |
| "memory(GiB)": 18.85, | |
| "step": 240, | |
| "train_speed(iter/s)": 0.072286 | |
| }, | |
| { | |
| "acc": 0.62713485, | |
| "epoch": 0.8828828828828829, | |
| "grad_norm": 0.82421875, | |
| "learning_rate": 0.00012453098548266276, | |
| "loss": 1.51464148, | |
| "memory(GiB)": 19.35, | |
| "step": 245, | |
| "train_speed(iter/s)": 0.07254 | |
| }, | |
| { | |
| "acc": 0.6202302, | |
| "epoch": 0.9009009009009009, | |
| "grad_norm": 0.625, | |
| "learning_rate": 0.0001239855152977253, | |
| "loss": 1.54778471, | |
| "memory(GiB)": 19.53, | |
| "step": 250, | |
| "train_speed(iter/s)": 0.072758 | |
| }, | |
| { | |
| "epoch": 0.9009009009009009, | |
| "eval_acc": 0.6308573540280857, | |
| "eval_loss": 1.510523796081543, | |
| "eval_runtime": 134.5445, | |
| "eval_samples_per_second": 1.122, | |
| "eval_steps_per_second": 0.565, | |
| "step": 250 | |
| }, | |
| { | |
| "acc": 0.63671951, | |
| "epoch": 0.918918918918919, | |
| "grad_norm": 1.7109375, | |
| "learning_rate": 0.00012342745187799459, | |
| "loss": 1.48321924, | |
| "memory(GiB)": 19.53, | |
| "step": 255, | |
| "train_speed(iter/s)": 0.070273 | |
| }, | |
| { | |
| "acc": 0.63577223, | |
| "epoch": 0.9369369369369369, | |
| "grad_norm": 0.7890625, | |
| "learning_rate": 0.000122856919641627, | |
| "loss": 1.50699987, | |
| "memory(GiB)": 19.94, | |
| "step": 260, | |
| "train_speed(iter/s)": 0.070553 | |
| }, | |
| { | |
| "acc": 0.64953299, | |
| "epoch": 0.954954954954955, | |
| "grad_norm": 0.85546875, | |
| "learning_rate": 0.000122274045786655, | |
| "loss": 1.46005678, | |
| "memory(GiB)": 20.1, | |
| "step": 265, | |
| "train_speed(iter/s)": 0.070802 | |
| }, | |
| { | |
| "acc": 0.62153759, | |
| "epoch": 0.972972972972973, | |
| "grad_norm": 1.0625, | |
| "learning_rate": 0.00012167896026262893, | |
| "loss": 1.55834417, | |
| "memory(GiB)": 19.86, | |
| "step": 270, | |
| "train_speed(iter/s)": 0.071052 | |
| }, | |
| { | |
| "acc": 0.64055209, | |
| "epoch": 0.990990990990991, | |
| "grad_norm": 1.125, | |
| "learning_rate": 0.00012107179574164504, | |
| "loss": 1.54932261, | |
| "memory(GiB)": 20.06, | |
| "step": 275, | |
| "train_speed(iter/s)": 0.071274 | |
| }, | |
| { | |
| "acc": 0.62708969, | |
| "epoch": 1.009009009009009, | |
| "grad_norm": 0.671875, | |
| "learning_rate": 0.00012045268758876699, | |
| "loss": 1.49731979, | |
| "memory(GiB)": 19.82, | |
| "step": 280, | |
| "train_speed(iter/s)": 0.07152 | |
| }, | |
| { | |
| "acc": 0.6689836, | |
| "epoch": 1.027027027027027, | |
| "grad_norm": 0.859375, | |
| "learning_rate": 0.00011982177383184648, | |
| "loss": 1.2817215, | |
| "memory(GiB)": 19.85, | |
| "step": 285, | |
| "train_speed(iter/s)": 0.07175 | |
| }, | |
| { | |
| "acc": 0.67519293, | |
| "epoch": 1.045045045045045, | |
| "grad_norm": 1.046875, | |
| "learning_rate": 0.00011917919513075066, | |
| "loss": 1.28632126, | |
| "memory(GiB)": 19.98, | |
| "step": 290, | |
| "train_speed(iter/s)": 0.071951 | |
| }, | |
| { | |
| "acc": 0.67276659, | |
| "epoch": 1.063063063063063, | |
| "grad_norm": 0.8984375, | |
| "learning_rate": 0.00011852509474600237, | |
| "loss": 1.27065611, | |
| "memory(GiB)": 20.03, | |
| "step": 295, | |
| "train_speed(iter/s)": 0.072155 | |
| }, | |
| { | |
| "acc": 0.64641519, | |
| "epoch": 1.0810810810810811, | |
| "grad_norm": 0.98046875, | |
| "learning_rate": 0.00011785961850684083, | |
| "loss": 1.38271847, | |
| "memory(GiB)": 19.09, | |
| "step": 300, | |
| "train_speed(iter/s)": 0.072371 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "eval_acc": 0.6305617147080562, | |
| "eval_loss": 1.523685097694397, | |
| "eval_runtime": 134.8234, | |
| "eval_samples_per_second": 1.12, | |
| "eval_steps_per_second": 0.564, | |
| "step": 300 | |
| }, | |
| { | |
| "acc": 0.67837138, | |
| "epoch": 1.0990990990990992, | |
| "grad_norm": 0.953125, | |
| "learning_rate": 0.00011718291477870959, | |
| "loss": 1.29290819, | |
| "memory(GiB)": 22.8, | |
| "step": 305, | |
| "train_speed(iter/s)": 0.070277 | |
| }, | |
| { | |
| "acc": 0.67195911, | |
| "epoch": 1.117117117117117, | |
| "grad_norm": 1.796875, | |
| "learning_rate": 0.00011649513443017889, | |
| "loss": 1.24073734, | |
| "memory(GiB)": 19.39, | |
| "step": 310, | |
| "train_speed(iter/s)": 0.070516 | |
| }, | |
| { | |
| "acc": 0.69478951, | |
| "epoch": 1.135135135135135, | |
| "grad_norm": 1.203125, | |
| "learning_rate": 0.00011579643079931018, | |
| "loss": 1.20378675, | |
| "memory(GiB)": 19.38, | |
| "step": 315, | |
| "train_speed(iter/s)": 0.070713 | |
| }, | |
| { | |
| "acc": 0.68726826, | |
| "epoch": 1.1531531531531531, | |
| "grad_norm": 0.98828125, | |
| "learning_rate": 0.00011508695965946992, | |
| "loss": 1.23284683, | |
| "memory(GiB)": 19.98, | |
| "step": 320, | |
| "train_speed(iter/s)": 0.070919 | |
| }, | |
| { | |
| "acc": 0.65419765, | |
| "epoch": 1.1711711711711712, | |
| "grad_norm": 0.93359375, | |
| "learning_rate": 0.00011436687918460052, | |
| "loss": 1.37520065, | |
| "memory(GiB)": 20.02, | |
| "step": 325, | |
| "train_speed(iter/s)": 0.071117 | |
| }, | |
| { | |
| "acc": 0.66610641, | |
| "epoch": 1.1891891891891893, | |
| "grad_norm": 0.8671875, | |
| "learning_rate": 0.000113636349913956, | |
| "loss": 1.30743008, | |
| "memory(GiB)": 19.35, | |
| "step": 330, | |
| "train_speed(iter/s)": 0.071322 | |
| }, | |
| { | |
| "acc": 0.67390976, | |
| "epoch": 1.2072072072072073, | |
| "grad_norm": 1.6640625, | |
| "learning_rate": 0.00011289553471631045, | |
| "loss": 1.28322783, | |
| "memory(GiB)": 19.49, | |
| "step": 335, | |
| "train_speed(iter/s)": 0.071518 | |
| }, | |
| { | |
| "acc": 0.68137512, | |
| "epoch": 1.2252252252252251, | |
| "grad_norm": 0.6953125, | |
| "learning_rate": 0.00011214459875364693, | |
| "loss": 1.23027716, | |
| "memory(GiB)": 19.38, | |
| "step": 340, | |
| "train_speed(iter/s)": 0.071692 | |
| }, | |
| { | |
| "acc": 0.67859125, | |
| "epoch": 1.2432432432432432, | |
| "grad_norm": 0.78515625, | |
| "learning_rate": 0.00011138370944433531, | |
| "loss": 1.22896252, | |
| "memory(GiB)": 20.06, | |
| "step": 345, | |
| "train_speed(iter/s)": 0.071876 | |
| }, | |
| { | |
| "acc": 0.66445112, | |
| "epoch": 1.2612612612612613, | |
| "grad_norm": 0.90234375, | |
| "learning_rate": 0.00011061303642580694, | |
| "loss": 1.30674038, | |
| "memory(GiB)": 19.49, | |
| "step": 350, | |
| "train_speed(iter/s)": 0.072045 | |
| }, | |
| { | |
| "epoch": 1.2612612612612613, | |
| "eval_acc": 0.6356245380635624, | |
| "eval_loss": 1.5072119235992432, | |
| "eval_runtime": 134.5232, | |
| "eval_samples_per_second": 1.122, | |
| "eval_steps_per_second": 0.565, | |
| "step": 350 | |
| }, | |
| { | |
| "acc": 0.67729836, | |
| "epoch": 1.2792792792792793, | |
| "grad_norm": 0.90625, | |
| "learning_rate": 0.00010983275151673467, | |
| "loss": 1.24173574, | |
| "memory(GiB)": 18.93, | |
| "step": 355, | |
| "train_speed(iter/s)": 0.07029 | |
| }, | |
| { | |
| "acc": 0.7040791, | |
| "epoch": 1.2972972972972974, | |
| "grad_norm": 0.84765625, | |
| "learning_rate": 0.00010904302867872639, | |
| "loss": 1.17582674, | |
| "memory(GiB)": 19.29, | |
| "step": 360, | |
| "train_speed(iter/s)": 0.070479 | |
| }, | |
| { | |
| "acc": 0.66356058, | |
| "epoch": 1.3153153153153152, | |
| "grad_norm": 0.82421875, | |
| "learning_rate": 0.00010824404397754104, | |
| "loss": 1.26798725, | |
| "memory(GiB)": 19.36, | |
| "step": 365, | |
| "train_speed(iter/s)": 0.070661 | |
| }, | |
| { | |
| "acc": 0.69379635, | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 0.98828125, | |
| "learning_rate": 0.0001074359755438354, | |
| "loss": 1.24331112, | |
| "memory(GiB)": 20.16, | |
| "step": 370, | |
| "train_speed(iter/s)": 0.070843 | |
| }, | |
| { | |
| "acc": 0.68220735, | |
| "epoch": 1.3513513513513513, | |
| "grad_norm": 0.94140625, | |
| "learning_rate": 0.00010661900353345051, | |
| "loss": 1.20891714, | |
| "memory(GiB)": 19.61, | |
| "step": 375, | |
| "train_speed(iter/s)": 0.071015 | |
| }, | |
| { | |
| "acc": 0.67620883, | |
| "epoch": 1.3693693693693694, | |
| "grad_norm": 1.0625, | |
| "learning_rate": 0.0001057933100872466, | |
| "loss": 1.23957863, | |
| "memory(GiB)": 20.17, | |
| "step": 380, | |
| "train_speed(iter/s)": 0.071181 | |
| }, | |
| { | |
| "acc": 0.63655629, | |
| "epoch": 1.3873873873873874, | |
| "grad_norm": 0.78515625, | |
| "learning_rate": 0.00010495907929049546, | |
| "loss": 1.44390507, | |
| "memory(GiB)": 19.25, | |
| "step": 385, | |
| "train_speed(iter/s)": 0.071356 | |
| }, | |
| { | |
| "acc": 0.67883902, | |
| "epoch": 1.4054054054054055, | |
| "grad_norm": 0.8828125, | |
| "learning_rate": 0.00010411649713183925, | |
| "loss": 1.29691544, | |
| "memory(GiB)": 18.78, | |
| "step": 390, | |
| "train_speed(iter/s)": 0.071515 | |
| }, | |
| { | |
| "acc": 0.67202511, | |
| "epoch": 1.4234234234234235, | |
| "grad_norm": 0.953125, | |
| "learning_rate": 0.00010326575146182521, | |
| "loss": 1.31318274, | |
| "memory(GiB)": 19.88, | |
| "step": 395, | |
| "train_speed(iter/s)": 0.071677 | |
| }, | |
| { | |
| "acc": 0.69274058, | |
| "epoch": 1.4414414414414414, | |
| "grad_norm": 0.82421875, | |
| "learning_rate": 0.00010240703195102489, | |
| "loss": 1.15976305, | |
| "memory(GiB)": 19.46, | |
| "step": 400, | |
| "train_speed(iter/s)": 0.071832 | |
| }, | |
| { | |
| "epoch": 1.4414414414414414, | |
| "eval_acc": 0.6368440502586844, | |
| "eval_loss": 1.4986343383789062, | |
| "eval_runtime": 134.3425, | |
| "eval_samples_per_second": 1.124, | |
| "eval_steps_per_second": 0.566, | |
| "step": 400 | |
| }, | |
| { | |
| "acc": 0.71039405, | |
| "epoch": 1.4594594594594594, | |
| "grad_norm": 0.77734375, | |
| "learning_rate": 0.0001015405300477479, | |
| "loss": 1.12253609, | |
| "memory(GiB)": 19.92, | |
| "step": 405, | |
| "train_speed(iter/s)": 0.070298 | |
| }, | |
| { | |
| "acc": 0.71356583, | |
| "epoch": 1.4774774774774775, | |
| "grad_norm": 0.84375, | |
| "learning_rate": 0.0001006664389353592, | |
| "loss": 1.13753939, | |
| "memory(GiB)": 19.31, | |
| "step": 410, | |
| "train_speed(iter/s)": 0.070457 | |
| }, | |
| { | |
| "acc": 0.675458, | |
| "epoch": 1.4954954954954955, | |
| "grad_norm": 1.1328125, | |
| "learning_rate": 9.978495348920958e-05, | |
| "loss": 1.29233532, | |
| "memory(GiB)": 19.06, | |
| "step": 415, | |
| "train_speed(iter/s)": 0.070616 | |
| }, | |
| { | |
| "acc": 0.67761598, | |
| "epoch": 1.5135135135135136, | |
| "grad_norm": 0.6875, | |
| "learning_rate": 9.889627023318897e-05, | |
| "loss": 1.22440186, | |
| "memory(GiB)": 19.16, | |
| "step": 420, | |
| "train_speed(iter/s)": 0.070773 | |
| }, | |
| { | |
| "acc": 0.67492404, | |
| "epoch": 1.5315315315315314, | |
| "grad_norm": 0.81640625, | |
| "learning_rate": 9.800058729591212e-05, | |
| "loss": 1.22408361, | |
| "memory(GiB)": 19.97, | |
| "step": 425, | |
| "train_speed(iter/s)": 0.070935 | |
| }, | |
| { | |
| "acc": 0.68050842, | |
| "epoch": 1.5495495495495497, | |
| "grad_norm": 0.84765625, | |
| "learning_rate": 9.70981043665466e-05, | |
| "loss": 1.2078824, | |
| "memory(GiB)": 19.92, | |
| "step": 430, | |
| "train_speed(iter/s)": 0.07109 | |
| }, | |
| { | |
| "acc": 0.6750885, | |
| "epoch": 1.5675675675675675, | |
| "grad_norm": 0.66796875, | |
| "learning_rate": 9.618902265029284e-05, | |
| "loss": 1.28742075, | |
| "memory(GiB)": 19.27, | |
| "step": 435, | |
| "train_speed(iter/s)": 0.071229 | |
| }, | |
| { | |
| "acc": 0.64411507, | |
| "epoch": 1.5855855855855856, | |
| "grad_norm": 0.95703125, | |
| "learning_rate": 9.527354482352616e-05, | |
| "loss": 1.37240067, | |
| "memory(GiB)": 20.21, | |
| "step": 440, | |
| "train_speed(iter/s)": 0.071374 | |
| }, | |
| { | |
| "acc": 0.67574663, | |
| "epoch": 1.6036036036036037, | |
| "grad_norm": 0.83984375, | |
| "learning_rate": 9.435187498861085e-05, | |
| "loss": 1.27780771, | |
| "memory(GiB)": 19.95, | |
| "step": 445, | |
| "train_speed(iter/s)": 0.071519 | |
| }, | |
| { | |
| "acc": 0.67897987, | |
| "epoch": 1.6216216216216215, | |
| "grad_norm": 1.2265625, | |
| "learning_rate": 9.342421862839632e-05, | |
| "loss": 1.26616125, | |
| "memory(GiB)": 19.32, | |
| "step": 450, | |
| "train_speed(iter/s)": 0.071661 | |
| }, | |
| { | |
| "epoch": 1.6216216216216215, | |
| "eval_acc": 0.6424611973392461, | |
| "eval_loss": 1.4772522449493408, | |
| "eval_runtime": 134.5995, | |
| "eval_samples_per_second": 1.122, | |
| "eval_steps_per_second": 0.565, | |
| "step": 450 | |
| }, | |
| { | |
| "acc": 0.66755495, | |
| "epoch": 1.6396396396396398, | |
| "grad_norm": 1.0390625, | |
| "learning_rate": 9.249078256040541e-05, | |
| "loss": 1.30118093, | |
| "memory(GiB)": 22.82, | |
| "step": 455, | |
| "train_speed(iter/s)": 0.070312 | |
| }, | |
| { | |
| "acc": 0.66560607, | |
| "epoch": 1.6576576576576576, | |
| "grad_norm": 1.0546875, | |
| "learning_rate": 9.155177489072527e-05, | |
| "loss": 1.31042576, | |
| "memory(GiB)": 19.56, | |
| "step": 460, | |
| "train_speed(iter/s)": 0.070454 | |
| }, | |
| { | |
| "acc": 0.67957892, | |
| "epoch": 1.6756756756756757, | |
| "grad_norm": 1.3828125, | |
| "learning_rate": 9.060740496761082e-05, | |
| "loss": 1.31165123, | |
| "memory(GiB)": 19.38, | |
| "step": 465, | |
| "train_speed(iter/s)": 0.070592 | |
| }, | |
| { | |
| "acc": 0.6744031, | |
| "epoch": 1.6936936936936937, | |
| "grad_norm": 1.4140625, | |
| "learning_rate": 8.965788333481144e-05, | |
| "loss": 1.26758223, | |
| "memory(GiB)": 19.42, | |
| "step": 470, | |
| "train_speed(iter/s)": 0.070726 | |
| }, | |
| { | |
| "acc": 0.66551232, | |
| "epoch": 1.7117117117117115, | |
| "grad_norm": 0.98046875, | |
| "learning_rate": 8.870342168463085e-05, | |
| "loss": 1.27216129, | |
| "memory(GiB)": 19.27, | |
| "step": 475, | |
| "train_speed(iter/s)": 0.070864 | |
| }, | |
| { | |
| "acc": 0.65833273, | |
| "epoch": 1.7297297297297298, | |
| "grad_norm": 0.9140625, | |
| "learning_rate": 8.77442328107313e-05, | |
| "loss": 1.32684155, | |
| "memory(GiB)": 19.48, | |
| "step": 480, | |
| "train_speed(iter/s)": 0.070997 | |
| }, | |
| { | |
| "acc": 0.68646383, | |
| "epoch": 1.7477477477477477, | |
| "grad_norm": 1.3671875, | |
| "learning_rate": 8.678053056069184e-05, | |
| "loss": 1.2200016, | |
| "memory(GiB)": 19.24, | |
| "step": 485, | |
| "train_speed(iter/s)": 0.071136 | |
| }, | |
| { | |
| "acc": 0.69040904, | |
| "epoch": 1.7657657657657657, | |
| "grad_norm": 1.6171875, | |
| "learning_rate": 8.581252978833194e-05, | |
| "loss": 1.18706884, | |
| "memory(GiB)": 19.53, | |
| "step": 490, | |
| "train_speed(iter/s)": 0.07127 | |
| }, | |
| { | |
| "acc": 0.66571455, | |
| "epoch": 1.7837837837837838, | |
| "grad_norm": 0.8515625, | |
| "learning_rate": 8.484044630581057e-05, | |
| "loss": 1.29456005, | |
| "memory(GiB)": 20.09, | |
| "step": 495, | |
| "train_speed(iter/s)": 0.071401 | |
| }, | |
| { | |
| "acc": 0.67682033, | |
| "epoch": 1.8018018018018018, | |
| "grad_norm": 1.0, | |
| "learning_rate": 8.386449683551164e-05, | |
| "loss": 1.20547714, | |
| "memory(GiB)": 19.95, | |
| "step": 500, | |
| "train_speed(iter/s)": 0.071533 | |
| }, | |
| { | |
| "epoch": 1.8018018018018018, | |
| "eval_acc": 0.6413155949741316, | |
| "eval_loss": 1.479081630706787, | |
| "eval_runtime": 134.2299, | |
| "eval_samples_per_second": 1.125, | |
| "eval_steps_per_second": 0.566, | |
| "step": 500 | |
| }, | |
| { | |
| "acc": 0.67326751, | |
| "epoch": 1.8198198198198199, | |
| "grad_norm": 1.0546875, | |
| "learning_rate": 8.288489896172669e-05, | |
| "loss": 1.25247726, | |
| "memory(GiB)": 20.29, | |
| "step": 505, | |
| "train_speed(iter/s)": 0.070304 | |
| }, | |
| { | |
| "acc": 0.66375732, | |
| "epoch": 1.8378378378378377, | |
| "grad_norm": 0.9296875, | |
| "learning_rate": 8.190187108214514e-05, | |
| "loss": 1.28065901, | |
| "memory(GiB)": 20.04, | |
| "step": 510, | |
| "train_speed(iter/s)": 0.070438 | |
| }, | |
| { | |
| "acc": 0.69006267, | |
| "epoch": 1.855855855855856, | |
| "grad_norm": 1.0234375, | |
| "learning_rate": 8.091563235916343e-05, | |
| "loss": 1.13905525, | |
| "memory(GiB)": 20.03, | |
| "step": 515, | |
| "train_speed(iter/s)": 0.070569 | |
| }, | |
| { | |
| "acc": 0.69745221, | |
| "epoch": 1.8738738738738738, | |
| "grad_norm": 0.96484375, | |
| "learning_rate": 7.992640267102351e-05, | |
| "loss": 1.14712362, | |
| "memory(GiB)": 18.5, | |
| "step": 520, | |
| "train_speed(iter/s)": 0.070709 | |
| }, | |
| { | |
| "acc": 0.6707756, | |
| "epoch": 1.8918918918918919, | |
| "grad_norm": 1.328125, | |
| "learning_rate": 7.893440256279186e-05, | |
| "loss": 1.30717278, | |
| "memory(GiB)": 20.66, | |
| "step": 525, | |
| "train_speed(iter/s)": 0.07083 | |
| }, | |
| { | |
| "acc": 0.66872559, | |
| "epoch": 1.90990990990991, | |
| "grad_norm": 0.9765625, | |
| "learning_rate": 7.793985319718982e-05, | |
| "loss": 1.28408003, | |
| "memory(GiB)": 19.48, | |
| "step": 530, | |
| "train_speed(iter/s)": 0.070948 | |
| }, | |
| { | |
| "acc": 0.68111048, | |
| "epoch": 1.9279279279279278, | |
| "grad_norm": 0.76171875, | |
| "learning_rate": 7.694297630528612e-05, | |
| "loss": 1.21391411, | |
| "memory(GiB)": 19.88, | |
| "step": 535, | |
| "train_speed(iter/s)": 0.071071 | |
| }, | |
| { | |
| "acc": 0.65094652, | |
| "epoch": 1.945945945945946, | |
| "grad_norm": 0.83203125, | |
| "learning_rate": 7.594399413706277e-05, | |
| "loss": 1.34138126, | |
| "memory(GiB)": 19.9, | |
| "step": 540, | |
| "train_speed(iter/s)": 0.071193 | |
| }, | |
| { | |
| "acc": 0.67896776, | |
| "epoch": 1.9639639639639639, | |
| "grad_norm": 0.796875, | |
| "learning_rate": 7.494312941186529e-05, | |
| "loss": 1.22575331, | |
| "memory(GiB)": 19.43, | |
| "step": 545, | |
| "train_speed(iter/s)": 0.071302 | |
| }, | |
| { | |
| "acc": 0.6839644, | |
| "epoch": 1.981981981981982, | |
| "grad_norm": 0.78515625, | |
| "learning_rate": 7.394060526874825e-05, | |
| "loss": 1.25017443, | |
| "memory(GiB)": 19.25, | |
| "step": 550, | |
| "train_speed(iter/s)": 0.07142 | |
| }, | |
| { | |
| "epoch": 1.981981981981982, | |
| "eval_acc": 0.645269770879527, | |
| "eval_loss": 1.4606801271438599, | |
| "eval_runtime": 134.7756, | |
| "eval_samples_per_second": 1.12, | |
| "eval_steps_per_second": 0.564, | |
| "step": 550 | |
| }, | |
| { | |
| "acc": 0.68771811, | |
| "epoch": 2.0, | |
| "grad_norm": 0.81640625, | |
| "learning_rate": 7.293664521672729e-05, | |
| "loss": 1.22415581, | |
| "memory(GiB)": 22.67, | |
| "step": 555, | |
| "train_speed(iter/s)": 0.070304 | |
| }, | |
| { | |
| "acc": 0.741537, | |
| "epoch": 2.018018018018018, | |
| "grad_norm": 0.6171875, | |
| "learning_rate": 7.193147308494851e-05, | |
| "loss": 0.95370378, | |
| "memory(GiB)": 19.64, | |
| "step": 560, | |
| "train_speed(iter/s)": 0.070425 | |
| }, | |
| { | |
| "acc": 0.75044699, | |
| "epoch": 2.036036036036036, | |
| "grad_norm": 1.09375, | |
| "learning_rate": 7.09253129727867e-05, | |
| "loss": 0.95568914, | |
| "memory(GiB)": 19.4, | |
| "step": 565, | |
| "train_speed(iter/s)": 0.070541 | |
| }, | |
| { | |
| "acc": 0.75126195, | |
| "epoch": 2.054054054054054, | |
| "grad_norm": 1.3671875, | |
| "learning_rate": 6.991838919988322e-05, | |
| "loss": 0.92719631, | |
| "memory(GiB)": 19.54, | |
| "step": 570, | |
| "train_speed(iter/s)": 0.070658 | |
| }, | |
| { | |
| "acc": 0.74883032, | |
| "epoch": 2.0720720720720722, | |
| "grad_norm": 1.0078125, | |
| "learning_rate": 6.891092625613469e-05, | |
| "loss": 0.92080975, | |
| "memory(GiB)": 20.17, | |
| "step": 575, | |
| "train_speed(iter/s)": 0.07077 | |
| }, | |
| { | |
| "acc": 0.76222944, | |
| "epoch": 2.09009009009009, | |
| "grad_norm": 0.99609375, | |
| "learning_rate": 6.790314875164393e-05, | |
| "loss": 0.88407106, | |
| "memory(GiB)": 19.57, | |
| "step": 580, | |
| "train_speed(iter/s)": 0.070882 | |
| }, | |
| { | |
| "acc": 0.76224823, | |
| "epoch": 2.108108108108108, | |
| "grad_norm": 1.0859375, | |
| "learning_rate": 6.689528136664377e-05, | |
| "loss": 0.85150976, | |
| "memory(GiB)": 19.54, | |
| "step": 585, | |
| "train_speed(iter/s)": 0.070995 | |
| }, | |
| { | |
| "acc": 0.73958569, | |
| "epoch": 2.126126126126126, | |
| "grad_norm": 1.3828125, | |
| "learning_rate": 6.588754880140573e-05, | |
| "loss": 0.92128286, | |
| "memory(GiB)": 19.58, | |
| "step": 590, | |
| "train_speed(iter/s)": 0.071101 | |
| }, | |
| { | |
| "acc": 0.74549003, | |
| "epoch": 2.144144144144144, | |
| "grad_norm": 1.359375, | |
| "learning_rate": 6.488017572614363e-05, | |
| "loss": 0.90851021, | |
| "memory(GiB)": 18.59, | |
| "step": 595, | |
| "train_speed(iter/s)": 0.071211 | |
| }, | |
| { | |
| "acc": 0.73912826, | |
| "epoch": 2.1621621621621623, | |
| "grad_norm": 1.3125, | |
| "learning_rate": 6.387338673092443e-05, | |
| "loss": 0.92900734, | |
| "memory(GiB)": 19.54, | |
| "step": 600, | |
| "train_speed(iter/s)": 0.071321 | |
| }, | |
| { | |
| "epoch": 2.1621621621621623, | |
| "eval_acc": 0.6320768662232077, | |
| "eval_loss": 1.5818341970443726, | |
| "eval_runtime": 134.4691, | |
| "eval_samples_per_second": 1.123, | |
| "eval_steps_per_second": 0.565, | |
| "step": 600 | |
| }, | |
| { | |
| "acc": 0.75979438, | |
| "epoch": 2.18018018018018, | |
| "grad_norm": 1.09375, | |
| "learning_rate": 6.286740627559656e-05, | |
| "loss": 0.89129753, | |
| "memory(GiB)": 22.37, | |
| "step": 605, | |
| "train_speed(iter/s)": 0.070301 | |
| }, | |
| { | |
| "acc": 0.72820721, | |
| "epoch": 2.1981981981981984, | |
| "grad_norm": 2.15625, | |
| "learning_rate": 6.186245863974757e-05, | |
| "loss": 0.96495447, | |
| "memory(GiB)": 19.6, | |
| "step": 610, | |
| "train_speed(iter/s)": 0.070413 | |
| }, | |
| { | |
| "acc": 0.75764585, | |
| "epoch": 2.2162162162162162, | |
| "grad_norm": 1.0078125, | |
| "learning_rate": 6.0858767872701715e-05, | |
| "loss": 0.89218092, | |
| "memory(GiB)": 20.15, | |
| "step": 615, | |
| "train_speed(iter/s)": 0.070515 | |
| }, | |
| { | |
| "acc": 0.75772595, | |
| "epoch": 2.234234234234234, | |
| "grad_norm": 1.6328125, | |
| "learning_rate": 5.985655774356901e-05, | |
| "loss": 0.89191771, | |
| "memory(GiB)": 19.46, | |
| "step": 620, | |
| "train_speed(iter/s)": 0.070627 | |
| }, | |
| { | |
| "acc": 0.7377079, | |
| "epoch": 2.2522522522522523, | |
| "grad_norm": 1.1875, | |
| "learning_rate": 5.8856051691356884e-05, | |
| "loss": 0.94241228, | |
| "memory(GiB)": 19.35, | |
| "step": 625, | |
| "train_speed(iter/s)": 0.070733 | |
| }, | |
| { | |
| "acc": 0.77948771, | |
| "epoch": 2.27027027027027, | |
| "grad_norm": 1.2890625, | |
| "learning_rate": 5.785747277515506e-05, | |
| "loss": 0.79317036, | |
| "memory(GiB)": 20.48, | |
| "step": 630, | |
| "train_speed(iter/s)": 0.070844 | |
| }, | |
| { | |
| "acc": 0.76766949, | |
| "epoch": 2.2882882882882885, | |
| "grad_norm": 0.97265625, | |
| "learning_rate": 5.686104362440552e-05, | |
| "loss": 0.82855272, | |
| "memory(GiB)": 20.12, | |
| "step": 635, | |
| "train_speed(iter/s)": 0.070945 | |
| }, | |
| { | |
| "acc": 0.74998231, | |
| "epoch": 2.3063063063063063, | |
| "grad_norm": 2.9375, | |
| "learning_rate": 5.586698638926811e-05, | |
| "loss": 0.93049393, | |
| "memory(GiB)": 20.06, | |
| "step": 640, | |
| "train_speed(iter/s)": 0.071044 | |
| }, | |
| { | |
| "acc": 0.75094385, | |
| "epoch": 2.3243243243243246, | |
| "grad_norm": 1.1875, | |
| "learning_rate": 5.487552269109287e-05, | |
| "loss": 0.86875353, | |
| "memory(GiB)": 19.33, | |
| "step": 645, | |
| "train_speed(iter/s)": 0.071146 | |
| }, | |
| { | |
| "acc": 0.74836354, | |
| "epoch": 2.3423423423423424, | |
| "grad_norm": 1.1328125, | |
| "learning_rate": 5.388687357301051e-05, | |
| "loss": 0.88861446, | |
| "memory(GiB)": 20.11, | |
| "step": 650, | |
| "train_speed(iter/s)": 0.071249 | |
| }, | |
| { | |
| "epoch": 2.3423423423423424, | |
| "eval_acc": 0.630709534368071, | |
| "eval_loss": 1.5767972469329834, | |
| "eval_runtime": 134.3063, | |
| "eval_samples_per_second": 1.124, | |
| "eval_steps_per_second": 0.566, | |
| "step": 650 | |
| }, | |
| { | |
| "acc": 0.76697993, | |
| "epoch": 2.3603603603603602, | |
| "grad_norm": 1.2734375, | |
| "learning_rate": 5.290125945065162e-05, | |
| "loss": 0.85701361, | |
| "memory(GiB)": 22.96, | |
| "step": 655, | |
| "train_speed(iter/s)": 0.070324 | |
| }, | |
| { | |
| "acc": 0.76252317, | |
| "epoch": 2.3783783783783785, | |
| "grad_norm": 1.0390625, | |
| "learning_rate": 5.191890006300573e-05, | |
| "loss": 0.85787058, | |
| "memory(GiB)": 20.13, | |
| "step": 660, | |
| "train_speed(iter/s)": 0.070422 | |
| }, | |
| { | |
| "acc": 0.7651772, | |
| "epoch": 2.3963963963963963, | |
| "grad_norm": 1.1875, | |
| "learning_rate": 5.094001442343155e-05, | |
| "loss": 0.8521904, | |
| "memory(GiB)": 19.86, | |
| "step": 665, | |
| "train_speed(iter/s)": 0.070523 | |
| }, | |
| { | |
| "acc": 0.73847542, | |
| "epoch": 2.4144144144144146, | |
| "grad_norm": 1.2734375, | |
| "learning_rate": 4.996482077082849e-05, | |
| "loss": 0.95858736, | |
| "memory(GiB)": 19.29, | |
| "step": 670, | |
| "train_speed(iter/s)": 0.070628 | |
| }, | |
| { | |
| "acc": 0.74675932, | |
| "epoch": 2.4324324324324325, | |
| "grad_norm": 1.2734375, | |
| "learning_rate": 4.899353652098139e-05, | |
| "loss": 0.86487961, | |
| "memory(GiB)": 18.64, | |
| "step": 675, | |
| "train_speed(iter/s)": 0.070727 | |
| }, | |
| { | |
| "acc": 0.73309464, | |
| "epoch": 2.4504504504504503, | |
| "grad_norm": 1.8671875, | |
| "learning_rate": 4.802637821808819e-05, | |
| "loss": 0.93775883, | |
| "memory(GiB)": 19.78, | |
| "step": 680, | |
| "train_speed(iter/s)": 0.070825 | |
| }, | |
| { | |
| "acc": 0.76575212, | |
| "epoch": 2.4684684684684686, | |
| "grad_norm": 1.03125, | |
| "learning_rate": 4.706356148648246e-05, | |
| "loss": 0.8259285, | |
| "memory(GiB)": 19.9, | |
| "step": 685, | |
| "train_speed(iter/s)": 0.07092 | |
| }, | |
| { | |
| "acc": 0.76865396, | |
| "epoch": 2.4864864864864864, | |
| "grad_norm": 1.3125, | |
| "learning_rate": 4.6105300982560625e-05, | |
| "loss": 0.84868517, | |
| "memory(GiB)": 19.19, | |
| "step": 690, | |
| "train_speed(iter/s)": 0.071014 | |
| }, | |
| { | |
| "acc": 0.75694928, | |
| "epoch": 2.5045045045045047, | |
| "grad_norm": 1.03125, | |
| "learning_rate": 4.515181034692515e-05, | |
| "loss": 0.87043924, | |
| "memory(GiB)": 19.95, | |
| "step": 695, | |
| "train_speed(iter/s)": 0.071105 | |
| }, | |
| { | |
| "acc": 0.75771561, | |
| "epoch": 2.5225225225225225, | |
| "grad_norm": 1.3515625, | |
| "learning_rate": 4.420330215675415e-05, | |
| "loss": 0.86245804, | |
| "memory(GiB)": 19.18, | |
| "step": 700, | |
| "train_speed(iter/s)": 0.071194 | |
| }, | |
| { | |
| "epoch": 2.5225225225225225, | |
| "eval_acc": 0.6335181079083518, | |
| "eval_loss": 1.5894646644592285, | |
| "eval_runtime": 134.225, | |
| "eval_samples_per_second": 1.125, | |
| "eval_steps_per_second": 0.566, | |
| "step": 700 | |
| }, | |
| { | |
| "acc": 0.76191721, | |
| "epoch": 2.5405405405405403, | |
| "grad_norm": 1.71875, | |
| "learning_rate": 4.325998787840818e-05, | |
| "loss": 0.85848246, | |
| "memory(GiB)": 19.14, | |
| "step": 705, | |
| "train_speed(iter/s)": 0.070324 | |
| }, | |
| { | |
| "acc": 0.76571012, | |
| "epoch": 2.5585585585585586, | |
| "grad_norm": 1.15625, | |
| "learning_rate": 4.2322077820284477e-05, | |
| "loss": 0.85979414, | |
| "memory(GiB)": 20.01, | |
| "step": 710, | |
| "train_speed(iter/s)": 0.070422 | |
| }, | |
| { | |
| "acc": 0.73852654, | |
| "epoch": 2.5765765765765765, | |
| "grad_norm": 1.6484375, | |
| "learning_rate": 4.138978108592962e-05, | |
| "loss": 0.90148897, | |
| "memory(GiB)": 19.05, | |
| "step": 715, | |
| "train_speed(iter/s)": 0.070518 | |
| }, | |
| { | |
| "acc": 0.76960816, | |
| "epoch": 2.5945945945945947, | |
| "grad_norm": 3.71875, | |
| "learning_rate": 4.046330552742053e-05, | |
| "loss": 0.88053255, | |
| "memory(GiB)": 19.25, | |
| "step": 720, | |
| "train_speed(iter/s)": 0.070616 | |
| }, | |
| { | |
| "acc": 0.77552128, | |
| "epoch": 2.6126126126126126, | |
| "grad_norm": 0.96484375, | |
| "learning_rate": 3.954285769902474e-05, | |
| "loss": 0.83608866, | |
| "memory(GiB)": 19.96, | |
| "step": 725, | |
| "train_speed(iter/s)": 0.070707 | |
| }, | |
| { | |
| "acc": 0.76034231, | |
| "epoch": 2.6306306306306304, | |
| "grad_norm": 1.078125, | |
| "learning_rate": 3.8628642811149894e-05, | |
| "loss": 0.84258709, | |
| "memory(GiB)": 19.75, | |
| "step": 730, | |
| "train_speed(iter/s)": 0.070796 | |
| }, | |
| { | |
| "acc": 0.73506665, | |
| "epoch": 2.6486486486486487, | |
| "grad_norm": 2.125, | |
| "learning_rate": 3.772086468459271e-05, | |
| "loss": 0.96418314, | |
| "memory(GiB)": 19.94, | |
| "step": 735, | |
| "train_speed(iter/s)": 0.070887 | |
| }, | |
| { | |
| "acc": 0.74339218, | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 1.3359375, | |
| "learning_rate": 3.6819725705098094e-05, | |
| "loss": 0.94632616, | |
| "memory(GiB)": 19.98, | |
| "step": 740, | |
| "train_speed(iter/s)": 0.070978 | |
| }, | |
| { | |
| "acc": 0.75258017, | |
| "epoch": 2.684684684684685, | |
| "grad_norm": 1.328125, | |
| "learning_rate": 3.592542677823787e-05, | |
| "loss": 0.89630384, | |
| "memory(GiB)": 19.9, | |
| "step": 745, | |
| "train_speed(iter/s)": 0.071065 | |
| }, | |
| { | |
| "acc": 0.7422905, | |
| "epoch": 2.7027027027027026, | |
| "grad_norm": 1.46875, | |
| "learning_rate": 3.503816728461963e-05, | |
| "loss": 0.92554636, | |
| "memory(GiB)": 19.94, | |
| "step": 750, | |
| "train_speed(iter/s)": 0.071152 | |
| }, | |
| { | |
| "epoch": 2.7027027027027026, | |
| "eval_acc": 0.6360679970436068, | |
| "eval_loss": 1.577430248260498, | |
| "eval_runtime": 134.0595, | |
| "eval_samples_per_second": 1.126, | |
| "eval_steps_per_second": 0.567, | |
| "step": 750 | |
| }, | |
| { | |
| "acc": 0.76009235, | |
| "epoch": 2.7207207207207205, | |
| "grad_norm": 1.7265625, | |
| "learning_rate": 3.415814503543563e-05, | |
| "loss": 0.89433851, | |
| "memory(GiB)": 19.38, | |
| "step": 755, | |
| "train_speed(iter/s)": 0.070345 | |
| }, | |
| { | |
| "acc": 0.75049233, | |
| "epoch": 2.7387387387387387, | |
| "grad_norm": 1.453125, | |
| "learning_rate": 3.3285556228361483e-05, | |
| "loss": 0.90194426, | |
| "memory(GiB)": 19.78, | |
| "step": 760, | |
| "train_speed(iter/s)": 0.070432 | |
| }, | |
| { | |
| "acc": 0.73652792, | |
| "epoch": 2.756756756756757, | |
| "grad_norm": 1.375, | |
| "learning_rate": 3.2420595403814615e-05, | |
| "loss": 0.94170513, | |
| "memory(GiB)": 19.18, | |
| "step": 765, | |
| "train_speed(iter/s)": 0.070517 | |
| }, | |
| { | |
| "acc": 0.74097948, | |
| "epoch": 2.774774774774775, | |
| "grad_norm": 1.171875, | |
| "learning_rate": 3.156345540158226e-05, | |
| "loss": 0.92526283, | |
| "memory(GiB)": 19.96, | |
| "step": 770, | |
| "train_speed(iter/s)": 0.070603 | |
| }, | |
| { | |
| "acc": 0.77357135, | |
| "epoch": 2.7927927927927927, | |
| "grad_norm": 1.21875, | |
| "learning_rate": 3.0714327317828445e-05, | |
| "loss": 0.84344234, | |
| "memory(GiB)": 19.42, | |
| "step": 775, | |
| "train_speed(iter/s)": 0.070681 | |
| }, | |
| { | |
| "acc": 0.76570077, | |
| "epoch": 2.810810810810811, | |
| "grad_norm": 1.4765625, | |
| "learning_rate": 2.9873400462489982e-05, | |
| "loss": 0.85261898, | |
| "memory(GiB)": 19.91, | |
| "step": 780, | |
| "train_speed(iter/s)": 0.070768 | |
| }, | |
| { | |
| "acc": 0.73979292, | |
| "epoch": 2.828828828828829, | |
| "grad_norm": 1.375, | |
| "learning_rate": 2.904086231707032e-05, | |
| "loss": 0.94777365, | |
| "memory(GiB)": 19.72, | |
| "step": 785, | |
| "train_speed(iter/s)": 0.07085 | |
| }, | |
| { | |
| "acc": 0.75035534, | |
| "epoch": 2.846846846846847, | |
| "grad_norm": 1.1484375, | |
| "learning_rate": 2.8216898492841355e-05, | |
| "loss": 0.88380022, | |
| "memory(GiB)": 19.09, | |
| "step": 790, | |
| "train_speed(iter/s)": 0.070936 | |
| }, | |
| { | |
| "acc": 0.76033754, | |
| "epoch": 2.864864864864865, | |
| "grad_norm": 1.078125, | |
| "learning_rate": 2.7401692689462153e-05, | |
| "loss": 0.84767551, | |
| "memory(GiB)": 20.02, | |
| "step": 795, | |
| "train_speed(iter/s)": 0.071016 | |
| }, | |
| { | |
| "acc": 0.74806399, | |
| "epoch": 2.8828828828828827, | |
| "grad_norm": 1.53125, | |
| "learning_rate": 2.6595426654023643e-05, | |
| "loss": 0.92544088, | |
| "memory(GiB)": 19.88, | |
| "step": 800, | |
| "train_speed(iter/s)": 0.0711 | |
| }, | |
| { | |
| "epoch": 2.8828828828828827, | |
| "eval_acc": 0.635920177383592, | |
| "eval_loss": 1.5869847536087036, | |
| "eval_runtime": 134.517, | |
| "eval_samples_per_second": 1.123, | |
| "eval_steps_per_second": 0.565, | |
| "step": 800 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 1108, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.3166381763355443e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |