| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.4676125848241828, | |
| "eval_steps": 500, | |
| "global_step": 4000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.030845157310302282, | |
| "grad_norm": 2.8206074237823486, | |
| "learning_rate": 1.0277492291880782e-05, | |
| "loss": 1.8082, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.061690314620604564, | |
| "grad_norm": 3.4183013439178467, | |
| "learning_rate": 2.0554984583761563e-05, | |
| "loss": 0.6538, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.09253547193090685, | |
| "grad_norm": 2.170591354370117, | |
| "learning_rate": 3.083247687564235e-05, | |
| "loss": 0.4563, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.12338062924120913, | |
| "grad_norm": 1.4687080383300781, | |
| "learning_rate": 4.110996916752313e-05, | |
| "loss": 0.4263, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15422578655151142, | |
| "grad_norm": 1.836676836013794, | |
| "learning_rate": 5.1387461459403907e-05, | |
| "loss": 0.3994, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.1850709438618137, | |
| "grad_norm": 1.2718663215637207, | |
| "learning_rate": 6.16649537512847e-05, | |
| "loss": 0.3665, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.215916101172116, | |
| "grad_norm": 1.6945191621780396, | |
| "learning_rate": 7.194244604316547e-05, | |
| "loss": 0.3577, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.24676125848241826, | |
| "grad_norm": 1.2829898595809937, | |
| "learning_rate": 8.221993833504625e-05, | |
| "loss": 0.347, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.27760641579272055, | |
| "grad_norm": 1.01521635055542, | |
| "learning_rate": 9.249743062692704e-05, | |
| "loss": 0.3288, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.30845157310302285, | |
| "grad_norm": 1.522111415863037, | |
| "learning_rate": 0.00010277492291880781, | |
| "loss": 0.3267, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.3392967304133251, | |
| "grad_norm": 0.9678927659988403, | |
| "learning_rate": 0.00011305241521068859, | |
| "loss": 0.3198, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.3701418877236274, | |
| "grad_norm": 1.2144405841827393, | |
| "learning_rate": 0.0001233299075025694, | |
| "loss": 0.3099, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.4009870450339297, | |
| "grad_norm": 1.3122639656066895, | |
| "learning_rate": 0.00013360739979445017, | |
| "loss": 0.2929, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.431832202344232, | |
| "grad_norm": 1.0934101343154907, | |
| "learning_rate": 0.00014388489208633093, | |
| "loss": 0.3003, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.4626773596545342, | |
| "grad_norm": 0.7938969731330872, | |
| "learning_rate": 0.00015416238437821172, | |
| "loss": 0.2956, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.4935225169648365, | |
| "grad_norm": 0.6571168303489685, | |
| "learning_rate": 0.0001644398766700925, | |
| "loss": 0.2736, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.5243676742751388, | |
| "grad_norm": 1.0073938369750977, | |
| "learning_rate": 0.0001747173689619733, | |
| "loss": 0.2892, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.5552128315854411, | |
| "grad_norm": 0.9874083399772644, | |
| "learning_rate": 0.00018499486125385408, | |
| "loss": 0.2723, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.5860579888957433, | |
| "grad_norm": 1.1770968437194824, | |
| "learning_rate": 0.00019527235354573487, | |
| "loss": 0.2855, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.6169031462060457, | |
| "grad_norm": 1.00326669216156, | |
| "learning_rate": 0.00019997622717095418, | |
| "loss": 0.2587, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6477483035163479, | |
| "grad_norm": 1.0380828380584717, | |
| "learning_rate": 0.0001998067088192682, | |
| "loss": 0.2764, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.6785934608266502, | |
| "grad_norm": 1.430301547050476, | |
| "learning_rate": 0.00019947447034120033, | |
| "loss": 0.2565, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.7094386181369525, | |
| "grad_norm": 1.0970648527145386, | |
| "learning_rate": 0.00019898005340261433, | |
| "loss": 0.2685, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.7402837754472548, | |
| "grad_norm": 0.9110261797904968, | |
| "learning_rate": 0.0001983242640774473, | |
| "loss": 0.2489, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.7711289327575571, | |
| "grad_norm": 1.0265332460403442, | |
| "learning_rate": 0.00019750817153352506, | |
| "loss": 0.2425, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.8019740900678594, | |
| "grad_norm": 0.8820884823799133, | |
| "learning_rate": 0.00019653310628944164, | |
| "loss": 0.2402, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.8328192473781616, | |
| "grad_norm": 0.933083713054657, | |
| "learning_rate": 0.00019540065804534467, | |
| "loss": 0.2444, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.863664404688464, | |
| "grad_norm": 0.8229042887687683, | |
| "learning_rate": 0.00019411267309116375, | |
| "loss": 0.2172, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.8945095619987662, | |
| "grad_norm": 0.9641085267066956, | |
| "learning_rate": 0.00019267125129650688, | |
| "loss": 0.2501, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.9253547193090684, | |
| "grad_norm": 0.6993410587310791, | |
| "learning_rate": 0.00019107874268713254, | |
| "loss": 0.2246, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.9561998766193708, | |
| "grad_norm": 1.0574674606323242, | |
| "learning_rate": 0.00018933774361357917, | |
| "loss": 0.2265, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.987045033929673, | |
| "grad_norm": 1.0701500177383423, | |
| "learning_rate": 0.0001874510925181983, | |
| "loss": 0.2116, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.0178901912399754, | |
| "grad_norm": 0.44411325454711914, | |
| "learning_rate": 0.0001854218653074927, | |
| "loss": 0.1978, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 1.0487353485502775, | |
| "grad_norm": 0.8408161401748657, | |
| "learning_rate": 0.0001832533703373043, | |
| "loss": 0.1701, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.0795805058605799, | |
| "grad_norm": 0.7417210936546326, | |
| "learning_rate": 0.0001809491430190276, | |
| "loss": 0.1873, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.1104256631708822, | |
| "grad_norm": 0.859341025352478, | |
| "learning_rate": 0.00017851294005564254, | |
| "loss": 0.1717, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.1412708204811843, | |
| "grad_norm": 0.6794934272766113, | |
| "learning_rate": 0.0001759487333169642, | |
| "loss": 0.1732, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 1.1721159777914867, | |
| "grad_norm": 0.5129622220993042, | |
| "learning_rate": 0.00017326070336409427, | |
| "loss": 0.172, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.202961135101789, | |
| "grad_norm": 0.5867941379547119, | |
| "learning_rate": 0.00017045323263363272, | |
| "loss": 0.1724, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 1.2338062924120914, | |
| "grad_norm": 0.887195348739624, | |
| "learning_rate": 0.0001675308982927608, | |
| "loss": 0.1655, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.2646514497223935, | |
| "grad_norm": 0.4721340835094452, | |
| "learning_rate": 0.0001644984647768447, | |
| "loss": 0.1642, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 1.2954966070326959, | |
| "grad_norm": 0.551078200340271, | |
| "learning_rate": 0.00016136087602172582, | |
| "loss": 0.1678, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.3263417643429982, | |
| "grad_norm": 0.97613525390625, | |
| "learning_rate": 0.00015812324740336248, | |
| "loss": 0.159, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 1.3571869216533003, | |
| "grad_norm": 0.7279055714607239, | |
| "learning_rate": 0.00015479085739796328, | |
| "loss": 0.1612, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.3880320789636027, | |
| "grad_norm": 0.9861264228820801, | |
| "learning_rate": 0.0001513691389762097, | |
| "loss": 0.1578, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 1.418877236273905, | |
| "grad_norm": 0.7998865246772766, | |
| "learning_rate": 0.00014786367074559828, | |
| "loss": 0.1569, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.4497223935842074, | |
| "grad_norm": 0.851041853427887, | |
| "learning_rate": 0.0001442801678553436, | |
| "loss": 0.1486, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 1.4805675508945095, | |
| "grad_norm": 0.42780816555023193, | |
| "learning_rate": 0.00014062447267866986, | |
| "loss": 0.1486, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.5114127082048119, | |
| "grad_norm": 0.7399270534515381, | |
| "learning_rate": 0.00013690254528768225, | |
| "loss": 0.1364, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 1.542257865515114, | |
| "grad_norm": 1.0372874736785889, | |
| "learning_rate": 0.0001331204537363485, | |
| "loss": 0.1393, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.5731030228254164, | |
| "grad_norm": 1.0774208307266235, | |
| "learning_rate": 0.00012928436416743098, | |
| "loss": 0.1468, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 1.6039481801357187, | |
| "grad_norm": 0.849371075630188, | |
| "learning_rate": 0.00012540053075949987, | |
| "loss": 0.1357, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.634793337446021, | |
| "grad_norm": 0.9047374725341797, | |
| "learning_rate": 0.00012147528553041718, | |
| "loss": 0.1292, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 1.6656384947563234, | |
| "grad_norm": 0.46791982650756836, | |
| "learning_rate": 0.00011751502801391479, | |
| "loss": 0.1308, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.6964836520666255, | |
| "grad_norm": 0.8372901082038879, | |
| "learning_rate": 0.00011352621482609807, | |
| "loss": 0.1401, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 1.7273288093769277, | |
| "grad_norm": 0.7648535966873169, | |
| "learning_rate": 0.00010951534913888515, | |
| "loss": 0.1245, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.75817396668723, | |
| "grad_norm": 0.4626932442188263, | |
| "learning_rate": 0.00010548897007754374, | |
| "loss": 0.1288, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 1.7890191239975324, | |
| "grad_norm": 0.5342707633972168, | |
| "learning_rate": 0.00010145364205961125, | |
| "loss": 0.1237, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.8198642813078347, | |
| "grad_norm": 0.7241202592849731, | |
| "learning_rate": 9.74159440925796e-05, | |
| "loss": 0.1168, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 1.850709438618137, | |
| "grad_norm": 0.48138442635536194, | |
| "learning_rate": 9.338245904779345e-05, | |
| "loss": 0.1149, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.8815545959284392, | |
| "grad_norm": 0.5214439630508423, | |
| "learning_rate": 8.93597629280487e-05, | |
| "loss": 0.1175, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 1.9123997532387416, | |
| "grad_norm": 0.5231944918632507, | |
| "learning_rate": 8.535441414638937e-05, | |
| "loss": 0.1168, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 1.9432449105490437, | |
| "grad_norm": 0.7800565958023071, | |
| "learning_rate": 8.13729428335819e-05, | |
| "loss": 0.1081, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 1.974090067859346, | |
| "grad_norm": 0.592022716999054, | |
| "learning_rate": 7.742184019169945e-05, | |
| "loss": 0.115, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.0049352251696484, | |
| "grad_norm": 0.4546678960323334, | |
| "learning_rate": 7.350754791117384e-05, | |
| "loss": 0.1002, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 2.0357803824799507, | |
| "grad_norm": 0.4683123826980591, | |
| "learning_rate": 6.963644766856894e-05, | |
| "loss": 0.0696, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.066625539790253, | |
| "grad_norm": 0.4901474416255951, | |
| "learning_rate": 6.581485072219755e-05, | |
| "loss": 0.0696, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 2.097470697100555, | |
| "grad_norm": 0.7581807971000671, | |
| "learning_rate": 6.204898762254524e-05, | |
| "loss": 0.0705, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.1283158544108574, | |
| "grad_norm": 0.5819096565246582, | |
| "learning_rate": 5.8344998054276115e-05, | |
| "loss": 0.0695, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 2.1591610117211597, | |
| "grad_norm": 0.7029954791069031, | |
| "learning_rate": 5.4708920826382035e-05, | |
| "loss": 0.0683, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.190006169031462, | |
| "grad_norm": 0.6168348789215088, | |
| "learning_rate": 5.114668402679472e-05, | |
| "loss": 0.0675, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 2.2208513263417644, | |
| "grad_norm": 0.5696262717247009, | |
| "learning_rate": 4.766409535751225e-05, | |
| "loss": 0.0678, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.2516964836520668, | |
| "grad_norm": 0.7154285907745361, | |
| "learning_rate": 4.426683266599702e-05, | |
| "loss": 0.0655, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 2.2825416409623687, | |
| "grad_norm": 0.6194272041320801, | |
| "learning_rate": 4.0960434688282515e-05, | |
| "loss": 0.0623, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.313386798272671, | |
| "grad_norm": 0.4507332444190979, | |
| "learning_rate": 3.775029201888051e-05, | |
| "loss": 0.0645, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 2.3442319555829734, | |
| "grad_norm": 0.42285481095314026, | |
| "learning_rate": 3.4641638322211456e-05, | |
| "loss": 0.0623, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.3750771128932757, | |
| "grad_norm": 0.4432896077632904, | |
| "learning_rate": 3.1639541799886083e-05, | |
| "loss": 0.0625, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 2.405922270203578, | |
| "grad_norm": 0.4708622097969055, | |
| "learning_rate": 2.874889692774978e-05, | |
| "loss": 0.063, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.4367674275138804, | |
| "grad_norm": 0.30538269877433777, | |
| "learning_rate": 2.5974416476161167e-05, | |
| "loss": 0.06, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 2.4676125848241828, | |
| "grad_norm": 0.7072311639785767, | |
| "learning_rate": 2.3320623826514897e-05, | |
| "loss": 0.0623, | |
| "step": 4000 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 4863, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.8423096729862144e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |