{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 0.9998523628652342, "eval_steps": 500, "global_step": 4656, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.010737246164789885, "grad_norm": 122.58504486083984, "learning_rate": 3.5714285714285718e-06, "loss": 25.5971, "step": 50 }, { "epoch": 0.02147449232957977, "grad_norm": 92.14105224609375, "learning_rate": 7.1428571428571436e-06, "loss": 15.8095, "step": 100 }, { "epoch": 0.03221173849436966, "grad_norm": 88.2498779296875, "learning_rate": 9.999879015387978e-06, "loss": 14.4751, "step": 150 }, { "epoch": 0.04294898465915954, "grad_norm": 79.90851593017578, "learning_rate": 9.995645168701038e-06, "loss": 10.6395, "step": 200 }, { "epoch": 0.05368623082394943, "grad_norm": 98.06950378417969, "learning_rate": 9.98536794504998e-06, "loss": 11.1393, "step": 250 }, { "epoch": 0.06442347698873932, "grad_norm": 95.09337615966797, "learning_rate": 9.969059777090564e-06, "loss": 11.2403, "step": 300 }, { "epoch": 0.0751607231535292, "grad_norm": 94.5640869140625, "learning_rate": 9.946740393286928e-06, "loss": 10.161, "step": 350 }, { "epoch": 0.08589796931831908, "grad_norm": 45.75733947753906, "learning_rate": 9.918436794045507e-06, "loss": 9.6319, "step": 400 }, { "epoch": 0.09663521548310897, "grad_norm": 44.17033767700195, "learning_rate": 9.884183219051837e-06, "loss": 10.4065, "step": 450 }, { "epoch": 0.10737246164789886, "grad_norm": 41.16874694824219, "learning_rate": 9.844021105849837e-06, "loss": 7.8502, "step": 500 }, { "epoch": 0.11810970781268874, "grad_norm": 84.32959747314453, "learning_rate": 9.797999039713586e-06, "loss": 9.3041, "step": 550 }, { "epoch": 0.12884695397747864, "grad_norm": 39.66738510131836, "learning_rate": 9.746172694872332e-06, "loss": 8.5279, "step": 600 }, { "epoch": 0.13958420014226852, "grad_norm": 39.50006866455078, "learning_rate": 9.688604767159736e-06, "loss": 9.5201, "step": 650 }, { "epoch": 0.1503214463070584, "grad_norm": 38.226261138916016, "learning_rate": 9.62536489816892e-06, "loss": 9.9717, "step": 700 }, { "epoch": 0.16105869247184829, "grad_norm": 35.42646789550781, "learning_rate": 9.556529591005001e-06, "loss": 9.0548, "step": 750 }, { "epoch": 0.17179593863663817, "grad_norm": 44.124263763427734, "learning_rate": 9.482182117737066e-06, "loss": 8.9763, "step": 800 }, { "epoch": 0.18253318480142805, "grad_norm": 43.76786422729492, "learning_rate": 9.402412418661541e-06, "loss": 9.8142, "step": 850 }, { "epoch": 0.19327043096621793, "grad_norm": 34.88469314575195, "learning_rate": 9.317316993498788e-06, "loss": 8.6092, "step": 900 }, { "epoch": 0.20400767713100781, "grad_norm": 38.54001998901367, "learning_rate": 9.226998784654606e-06, "loss": 7.4996, "step": 950 }, { "epoch": 0.21474492329579772, "grad_norm": 74.26197814941406, "learning_rate": 9.131567052687811e-06, "loss": 9.0709, "step": 1000 }, { "epoch": 0.2254821694605876, "grad_norm": 41.04670333862305, "learning_rate": 9.03113724413456e-06, "loss": 10.0284, "step": 1050 }, { "epoch": 0.2362194156253775, "grad_norm": 42.67012405395508, "learning_rate": 8.925830851849338e-06, "loss": 7.9923, "step": 1100 }, { "epoch": 0.24695666179016737, "grad_norm": 49.978736877441406, "learning_rate": 8.815775268031514e-06, "loss": 8.8409, "step": 1150 }, { "epoch": 0.2576939079549573, "grad_norm": 80.3055419921875, "learning_rate": 8.701103630115303e-06, "loss": 8.9708, "step": 1200 }, { "epoch": 0.26843115411974716, "grad_norm": 31.320579528808594, "learning_rate": 8.581954659709549e-06, "loss": 9.6101, "step": 1250 }, { "epoch": 0.27916840028453704, "grad_norm": 35.097434997558594, "learning_rate": 8.458472494782169e-06, "loss": 9.1278, "step": 1300 }, { "epoch": 0.2899056464493269, "grad_norm": 79.29938507080078, "learning_rate": 8.330806515292271e-06, "loss": 8.365, "step": 1350 }, { "epoch": 0.3006428926141168, "grad_norm": 35.84828186035156, "learning_rate": 8.199111162480871e-06, "loss": 9.4828, "step": 1400 }, { "epoch": 0.3113801387789067, "grad_norm": 28.13869857788086, "learning_rate": 8.063545752038854e-06, "loss": 7.6841, "step": 1450 }, { "epoch": 0.32211738494369657, "grad_norm": 48.257423400878906, "learning_rate": 7.924274281378153e-06, "loss": 7.3778, "step": 1500 }, { "epoch": 0.33285463110848645, "grad_norm": 39.27827835083008, "learning_rate": 7.781465231239318e-06, "loss": 10.1326, "step": 1550 }, { "epoch": 0.34359187727327634, "grad_norm": 29.741004943847656, "learning_rate": 7.635291361875474e-06, "loss": 8.61, "step": 1600 }, { "epoch": 0.3543291234380662, "grad_norm": 39.182769775390625, "learning_rate": 7.485929504059234e-06, "loss": 8.702, "step": 1650 }, { "epoch": 0.3650663696028561, "grad_norm": 32.254554748535156, "learning_rate": 7.333560345165371e-06, "loss": 7.5357, "step": 1700 }, { "epoch": 0.375803615767646, "grad_norm": 28.51645851135254, "learning_rate": 7.178368210588067e-06, "loss": 7.2457, "step": 1750 }, { "epoch": 0.38654086193243586, "grad_norm": 28.341947555541992, "learning_rate": 7.020540840757124e-06, "loss": 9.2959, "step": 1800 }, { "epoch": 0.39727810809722575, "grad_norm": 42.67914581298828, "learning_rate": 6.860269164022921e-06, "loss": 9.1062, "step": 1850 }, { "epoch": 0.40801535426201563, "grad_norm": 31.293323516845703, "learning_rate": 6.697747065684851e-06, "loss": 9.5109, "step": 1900 }, { "epoch": 0.4187526004268055, "grad_norm": 35.099021911621094, "learning_rate": 6.5331711534426326e-06, "loss": 8.1034, "step": 1950 }, { "epoch": 0.42948984659159545, "grad_norm": 35.72314453125, "learning_rate": 6.366740519554286e-06, "loss": 7.9167, "step": 2000 }, { "epoch": 0.44022709275638533, "grad_norm": 73.83097076416016, "learning_rate": 6.198656499988444e-06, "loss": 8.5164, "step": 2050 }, { "epoch": 0.4509643389211752, "grad_norm": 30.618099212646484, "learning_rate": 6.029122430862373e-06, "loss": 9.6928, "step": 2100 }, { "epoch": 0.4617015850859651, "grad_norm": 74.77922058105469, "learning_rate": 5.858343402460391e-06, "loss": 8.1368, "step": 2150 }, { "epoch": 0.472438831250755, "grad_norm": 26.513835906982422, "learning_rate": 5.68652601113019e-06, "loss": 7.5154, "step": 2200 }, { "epoch": 0.48317607741554486, "grad_norm": 70.47901916503906, "learning_rate": 5.513878109357228e-06, "loss": 8.6317, "step": 2250 }, { "epoch": 0.49391332358033474, "grad_norm": 33.31165313720703, "learning_rate": 5.3406085543195555e-06, "loss": 9.8199, "step": 2300 }, { "epoch": 0.5046505697451246, "grad_norm": 36.43808364868164, "learning_rate": 5.166926955227224e-06, "loss": 9.8981, "step": 2350 }, { "epoch": 0.5153878159099146, "grad_norm": 35.49153137207031, "learning_rate": 4.993043419751933e-06, "loss": 7.7103, "step": 2400 }, { "epoch": 0.5261250620747044, "grad_norm": 27.446149826049805, "learning_rate": 4.8191682998536905e-06, "loss": 7.0446, "step": 2450 }, { "epoch": 0.5368623082394943, "grad_norm": 35.19398498535156, "learning_rate": 4.645511937311934e-06, "loss": 8.806, "step": 2500 }, { "epoch": 0.5475995544042842, "grad_norm": 77.66211700439453, "learning_rate": 4.472284409268976e-06, "loss": 10.2757, "step": 2550 }, { "epoch": 0.5583368005690741, "grad_norm": 29.25933837890625, "learning_rate": 4.299695274093593e-06, "loss": 9.8498, "step": 2600 }, { "epoch": 0.569074046733864, "grad_norm": 32.54759979248047, "learning_rate": 4.1279533178721755e-06, "loss": 9.312, "step": 2650 }, { "epoch": 0.5798112928986539, "grad_norm": 77.79645538330078, "learning_rate": 3.957266301834145e-06, "loss": 8.5012, "step": 2700 }, { "epoch": 0.5905485390634437, "grad_norm": 30.680482864379883, "learning_rate": 3.7878407110171646e-06, "loss": 8.1592, "step": 2750 }, { "epoch": 0.6012857852282336, "grad_norm": 30.990217208862305, "learning_rate": 3.6198815044761847e-06, "loss": 9.4135, "step": 2800 }, { "epoch": 0.6120230313930235, "grad_norm": 34.311153411865234, "learning_rate": 3.4535918673385456e-06, "loss": 8.7601, "step": 2850 }, { "epoch": 0.6227602775578134, "grad_norm": 34.08210754394531, "learning_rate": 3.2891729650050096e-06, "loss": 8.4742, "step": 2900 }, { "epoch": 0.6334975237226033, "grad_norm": 30.370546340942383, "learning_rate": 3.1268236997941535e-06, "loss": 7.6472, "step": 2950 }, { "epoch": 0.6442347698873931, "grad_norm": 76.85076141357422, "learning_rate": 2.966740470324451e-06, "loss": 10.3334, "step": 3000 }, { "epoch": 0.654972016052183, "grad_norm": 33.31392288208008, "learning_rate": 2.8091169339251644e-06, "loss": 8.1425, "step": 3050 }, { "epoch": 0.6657092622169729, "grad_norm": 28.994285583496094, "learning_rate": 2.654143772363455e-06, "loss": 8.2618, "step": 3100 }, { "epoch": 0.6764465083817628, "grad_norm": 33.96207809448242, "learning_rate": 2.502008461171114e-06, "loss": 7.9787, "step": 3150 }, { "epoch": 0.6871837545465527, "grad_norm": 44.06400680541992, "learning_rate": 2.352895042849965e-06, "loss": 8.0353, "step": 3200 }, { "epoch": 0.6979210007113426, "grad_norm": 35.88554000854492, "learning_rate": 2.20698390423032e-06, "loss": 9.3106, "step": 3250 }, { "epoch": 0.7086582468761324, "grad_norm": 31.703388214111328, "learning_rate": 2.0644515582517803e-06, "loss": 8.5185, "step": 3300 }, { "epoch": 0.7193954930409223, "grad_norm": 34.17527770996094, "learning_rate": 1.9254704304304174e-06, "loss": 9.1151, "step": 3350 }, { "epoch": 0.7301327392057122, "grad_norm": 29.952329635620117, "learning_rate": 1.7902086502706256e-06, "loss": 7.4807, "step": 3400 }, { "epoch": 0.7408699853705021, "grad_norm": 29.618684768676758, "learning_rate": 1.658829847873965e-06, "loss": 8.402, "step": 3450 }, { "epoch": 0.751607231535292, "grad_norm": 35.65748596191406, "learning_rate": 1.5314929559910985e-06, "loss": 8.7392, "step": 3500 }, { "epoch": 0.7623444777000818, "grad_norm": 40.38413619995117, "learning_rate": 1.4083520177562154e-06, "loss": 6.4324, "step": 3550 }, { "epoch": 0.7730817238648717, "grad_norm": 38.58983612060547, "learning_rate": 1.2895560003365837e-06, "loss": 9.4396, "step": 3600 }, { "epoch": 0.7838189700296616, "grad_norm": 33.327510833740234, "learning_rate": 1.1752486147226505e-06, "loss": 7.8763, "step": 3650 }, { "epoch": 0.7945562161944515, "grad_norm": 31.827482223510742, "learning_rate": 1.0655681418766772e-06, "loss": 8.1159, "step": 3700 }, { "epoch": 0.8052934623592414, "grad_norm": 37.83740997314453, "learning_rate": 9.60647265450249e-07, "loss": 7.1248, "step": 3750 }, { "epoch": 0.8160307085240313, "grad_norm": 68.32665252685547, "learning_rate": 8.60612911273011e-07, "loss": 8.8583, "step": 3800 }, { "epoch": 0.8267679546888211, "grad_norm": 28.628252029418945, "learning_rate": 7.655860938068071e-07, "loss": 6.5486, "step": 3850 }, { "epoch": 0.837505200853611, "grad_norm": 35.551815032958984, "learning_rate": 6.756817697509755e-07, "loss": 8.1337, "step": 3900 }, { "epoch": 0.8482424470184009, "grad_norm": 29.21653175354004, "learning_rate": 5.910086989758862e-07, "loss": 7.0004, "step": 3950 }, { "epoch": 0.8589796931831909, "grad_norm": 28.43161392211914, "learning_rate": 5.11669312952977e-07, "loss": 8.2137, "step": 4000 }, { "epoch": 0.8697169393479808, "grad_norm": 27.239362716674805, "learning_rate": 4.377595908404225e-07, "loss": 7.9611, "step": 4050 }, { "epoch": 0.8804541855127707, "grad_norm": 27.20981788635254, "learning_rate": 3.693689433743658e-07, "loss": 8.2071, "step": 4100 }, { "epoch": 0.8911914316775605, "grad_norm": 30.86234474182129, "learning_rate": 3.065801047061517e-07, "loss": 6.6662, "step": 4150 }, { "epoch": 0.9019286778423504, "grad_norm": 72.03483581542969, "learning_rate": 2.4946903231642727e-07, "loss": 9.2256, "step": 4200 }, { "epoch": 0.9126659240071403, "grad_norm": 35.229248046875, "learning_rate": 1.9810481512716638e-07, "loss": 7.8211, "step": 4250 }, { "epoch": 0.9234031701719302, "grad_norm": 29.196678161621094, "learning_rate": 1.5254958992280022e-07, "loss": 7.9577, "step": 4300 }, { "epoch": 0.9341404163367201, "grad_norm": 31.289777755737305, "learning_rate": 1.128584661815435e-07, "loss": 7.572, "step": 4350 }, { "epoch": 0.94487766250151, "grad_norm": 29.68185043334961, "learning_rate": 7.907945940786033e-08, "loss": 7.5517, "step": 4400 }, { "epoch": 0.9556149086662998, "grad_norm": 28.52676773071289, "learning_rate": 5.125343304671459e-08, "loss": 10.374, "step": 4450 }, { "epoch": 0.9663521548310897, "grad_norm": 34.890655517578125, "learning_rate": 2.9414049049872883e-08, "loss": 7.1905, "step": 4500 }, { "epoch": 0.9770894009958796, "grad_norm": 40.517208099365234, "learning_rate": 1.35877271540652e-08, "loss": 8.6259, "step": 4550 }, { "epoch": 0.9878266471606695, "grad_norm": 29.639062881469727, "learning_rate": 3.7936129202648106e-09, "loss": 6.9152, "step": 4600 }, { "epoch": 0.9985638933254594, "grad_norm": 43.505428314208984, "learning_rate": 4.3554572743409463e-11, "loss": 10.0409, "step": 4650 }, { "epoch": 0.9998523628652342, "step": 4656, "total_flos": 3.2168639765195784e+19, "train_loss": 8.979798909315129, "train_runtime": 72260.7648, "train_samples_per_second": 4.124, "train_steps_per_second": 0.064 } ], "logging_steps": 50, "max_steps": 4656, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 1000, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 3.2168639765195784e+19, "train_batch_size": 1, "trial_name": null, "trial_params": null }