| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.0, |
| "eval_steps": 12, |
| "global_step": 63, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.047619047619047616, |
| "grad_norm": 55.872169494628906, |
| "learning_rate": 0.0, |
| "loss": 11.3984375, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.09523809523809523, |
| "grad_norm": 39.94508361816406, |
| "learning_rate": 2.5e-05, |
| "loss": 8.734375, |
| "step": 2 |
| }, |
| { |
| "epoch": 0.14285714285714285, |
| "grad_norm": 30.2918701171875, |
| "learning_rate": 5e-05, |
| "loss": 6.33984375, |
| "step": 3 |
| }, |
| { |
| "epoch": 0.19047619047619047, |
| "grad_norm": 11.614492416381836, |
| "learning_rate": 4.9970167022408685e-05, |
| "loss": 4.24609375, |
| "step": 4 |
| }, |
| { |
| "epoch": 0.23809523809523808, |
| "grad_norm": 10.849442481994629, |
| "learning_rate": 4.988074720132825e-05, |
| "loss": 2.6611328125, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.2857142857142857, |
| "grad_norm": 6.369261264801025, |
| "learning_rate": 4.9731977662049233e-05, |
| "loss": 1.529296875, |
| "step": 6 |
| }, |
| { |
| "epoch": 0.3333333333333333, |
| "grad_norm": 1.0310866832733154, |
| "learning_rate": 4.9524252914645555e-05, |
| "loss": 2.18267822265625, |
| "step": 7 |
| }, |
| { |
| "epoch": 0.38095238095238093, |
| "grad_norm": 2.9220826625823975, |
| "learning_rate": 4.9258123807804715e-05, |
| "loss": 0.80126953125, |
| "step": 8 |
| }, |
| { |
| "epoch": 0.42857142857142855, |
| "grad_norm": 2.5777978897094727, |
| "learning_rate": 4.8934296068076105e-05, |
| "loss": 0.8701171875, |
| "step": 9 |
| }, |
| { |
| "epoch": 0.47619047619047616, |
| "grad_norm": 2.27763295173645, |
| "learning_rate": 4.855362842841111e-05, |
| "loss": 0.7646484375, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.5238095238095238, |
| "grad_norm": 1.504328727722168, |
| "learning_rate": 4.811713035095761e-05, |
| "loss": 0.79931640625, |
| "step": 11 |
| }, |
| { |
| "epoch": 0.5714285714285714, |
| "grad_norm": 1.8779164552688599, |
| "learning_rate": 4.7625959350147905e-05, |
| "loss": 0.83056640625, |
| "step": 12 |
| }, |
| { |
| "epoch": 0.6190476190476191, |
| "grad_norm": 0.6685611605644226, |
| "learning_rate": 4.70814179231785e-05, |
| "loss": 1.894866943359375, |
| "step": 13 |
| }, |
| { |
| "epoch": 0.6666666666666666, |
| "grad_norm": 1.276977300643921, |
| "learning_rate": 4.648495009602168e-05, |
| "loss": 0.64892578125, |
| "step": 14 |
| }, |
| { |
| "epoch": 0.7142857142857143, |
| "grad_norm": 1.3761838674545288, |
| "learning_rate": 4.5838137594128254e-05, |
| "loss": 0.689208984375, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.7142857142857143, |
| "eval_loss": 1.626708984375, |
| "eval_runtime": 5.6958, |
| "eval_samples_per_second": 0.351, |
| "eval_steps_per_second": 0.176, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.7619047619047619, |
| "grad_norm": 1.0244767665863037, |
| "learning_rate": 4.5142695647975993e-05, |
| "loss": 0.554443359375, |
| "step": 16 |
| }, |
| { |
| "epoch": 0.8095238095238095, |
| "grad_norm": 0.4870399832725525, |
| "learning_rate": 4.44004684445867e-05, |
| "loss": 1.8179779052734375, |
| "step": 17 |
| }, |
| { |
| "epoch": 0.8571428571428571, |
| "grad_norm": 0.820194661617279, |
| "learning_rate": 4.361342423707385e-05, |
| "loss": 1.12042236328125, |
| "step": 18 |
| }, |
| { |
| "epoch": 0.9047619047619048, |
| "grad_norm": 0.8197793364524841, |
| "learning_rate": 4.2783650125189096e-05, |
| "loss": 0.616943359375, |
| "step": 19 |
| }, |
| { |
| "epoch": 0.9523809523809523, |
| "grad_norm": 0.5198944211006165, |
| "learning_rate": 4.191334652070895e-05, |
| "loss": 1.839569091796875, |
| "step": 20 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 0.34501922130584717, |
| "learning_rate": 4.1004821312338285e-05, |
| "loss": 1.6414794921875, |
| "step": 21 |
| }, |
| { |
| "epoch": 1.0476190476190477, |
| "grad_norm": 0.873728334903717, |
| "learning_rate": 4.006048374560445e-05, |
| "loss": 0.578125, |
| "step": 22 |
| }, |
| { |
| "epoch": 1.0952380952380953, |
| "grad_norm": 0.9984555840492249, |
| "learning_rate": 3.90828380339712e-05, |
| "loss": 0.4749755859375, |
| "step": 23 |
| }, |
| { |
| "epoch": 1.1428571428571428, |
| "grad_norm": 0.9322866797447205, |
| "learning_rate": 3.8074476718114706e-05, |
| "loss": 0.59375, |
| "step": 24 |
| }, |
| { |
| "epoch": 1.1904761904761905, |
| "grad_norm": 1.126771330833435, |
| "learning_rate": 3.7038073790971875e-05, |
| "loss": 0.451416015625, |
| "step": 25 |
| }, |
| { |
| "epoch": 1.2380952380952381, |
| "grad_norm": 0.921955406665802, |
| "learning_rate": 3.597637760679167e-05, |
| "loss": 0.4898681640625, |
| "step": 26 |
| }, |
| { |
| "epoch": 1.2857142857142856, |
| "grad_norm": 1.160629153251648, |
| "learning_rate": 3.4892203592993786e-05, |
| "loss": 0.4130859375, |
| "step": 27 |
| }, |
| { |
| "epoch": 1.2857142857142856, |
| "eval_loss": 1.51025390625, |
| "eval_runtime": 5.4865, |
| "eval_samples_per_second": 0.365, |
| "eval_steps_per_second": 0.182, |
| "step": 27 |
| }, |
| { |
| "epoch": 1.3333333333333333, |
| "grad_norm": 0.5455287098884583, |
| "learning_rate": 3.3788426784161216e-05, |
| "loss": 1.712249755859375, |
| "step": 28 |
| }, |
| { |
| "epoch": 1.380952380952381, |
| "grad_norm": 1.361806869506836, |
| "learning_rate": 3.2667974197965405e-05, |
| "loss": 0.3602294921875, |
| "step": 29 |
| }, |
| { |
| "epoch": 1.4285714285714286, |
| "grad_norm": 1.2242106199264526, |
| "learning_rate": 3.1533817073241556e-05, |
| "loss": 0.5125732421875, |
| "step": 30 |
| }, |
| { |
| "epoch": 1.4761904761904763, |
| "grad_norm": 1.058226466178894, |
| "learning_rate": 3.03889629907974e-05, |
| "loss": 0.3873291015625, |
| "step": 31 |
| }, |
| { |
| "epoch": 1.5238095238095237, |
| "grad_norm": 1.107188105583191, |
| "learning_rate": 2.923644789784955e-05, |
| "loss": 0.5450439453125, |
| "step": 32 |
| }, |
| { |
| "epoch": 1.5714285714285714, |
| "grad_norm": 1.150083303451538, |
| "learning_rate": 2.807932805723725e-05, |
| "loss": 0.51953125, |
| "step": 33 |
| }, |
| { |
| "epoch": 1.619047619047619, |
| "grad_norm": 0.534706175327301, |
| "learning_rate": 2.692067194276276e-05, |
| "loss": 1.4922409057617188, |
| "step": 34 |
| }, |
| { |
| "epoch": 1.6666666666666665, |
| "grad_norm": 1.0580393075942993, |
| "learning_rate": 2.5763552102150456e-05, |
| "loss": 0.4501953125, |
| "step": 35 |
| }, |
| { |
| "epoch": 1.7142857142857144, |
| "grad_norm": 1.0441676378250122, |
| "learning_rate": 2.4611037009202603e-05, |
| "loss": 0.478515625, |
| "step": 36 |
| }, |
| { |
| "epoch": 1.7619047619047619, |
| "grad_norm": 0.8805537819862366, |
| "learning_rate": 2.3466182926758456e-05, |
| "loss": 0.3509521484375, |
| "step": 37 |
| }, |
| { |
| "epoch": 1.8095238095238095, |
| "grad_norm": 0.4713503420352936, |
| "learning_rate": 2.2332025802034607e-05, |
| "loss": 1.53851318359375, |
| "step": 38 |
| }, |
| { |
| "epoch": 1.8571428571428572, |
| "grad_norm": 0.6947212219238281, |
| "learning_rate": 2.1211573215838792e-05, |
| "loss": 0.891815185546875, |
| "step": 39 |
| }, |
| { |
| "epoch": 1.8571428571428572, |
| "eval_loss": 1.4468994140625, |
| "eval_runtime": 5.1599, |
| "eval_samples_per_second": 0.388, |
| "eval_steps_per_second": 0.194, |
| "step": 39 |
| }, |
| { |
| "epoch": 1.9047619047619047, |
| "grad_norm": 0.8006393909454346, |
| "learning_rate": 2.010779640700622e-05, |
| "loss": 0.4466552734375, |
| "step": 40 |
| }, |
| { |
| "epoch": 1.9523809523809523, |
| "grad_norm": 0.47487154603004456, |
| "learning_rate": 1.9023622393208336e-05, |
| "loss": 1.5706024169921875, |
| "step": 41 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 0.3503785729408264, |
| "learning_rate": 1.796192620902814e-05, |
| "loss": 1.513275146484375, |
| "step": 42 |
| }, |
| { |
| "epoch": 2.0476190476190474, |
| "grad_norm": 0.8355963230133057, |
| "learning_rate": 1.692552328188531e-05, |
| "loss": 0.43603515625, |
| "step": 43 |
| }, |
| { |
| "epoch": 2.0952380952380953, |
| "grad_norm": 0.7735680937767029, |
| "learning_rate": 1.5917161966028815e-05, |
| "loss": 0.34686279296875, |
| "step": 44 |
| }, |
| { |
| "epoch": 2.142857142857143, |
| "grad_norm": 0.7461949586868286, |
| "learning_rate": 1.4939516254395546e-05, |
| "loss": 0.43896484375, |
| "step": 45 |
| }, |
| { |
| "epoch": 2.1904761904761907, |
| "grad_norm": 1.0518755912780762, |
| "learning_rate": 1.399517868766172e-05, |
| "loss": 0.325439453125, |
| "step": 46 |
| }, |
| { |
| "epoch": 2.238095238095238, |
| "grad_norm": 0.8924214243888855, |
| "learning_rate": 1.3086653479291062e-05, |
| "loss": 0.3905029296875, |
| "step": 47 |
| }, |
| { |
| "epoch": 2.2857142857142856, |
| "grad_norm": 0.8009278774261475, |
| "learning_rate": 1.2216349874810906e-05, |
| "loss": 0.30126953125, |
| "step": 48 |
| }, |
| { |
| "epoch": 2.3333333333333335, |
| "grad_norm": 0.5341566801071167, |
| "learning_rate": 1.1386575762926155e-05, |
| "loss": 1.5106887817382812, |
| "step": 49 |
| }, |
| { |
| "epoch": 2.380952380952381, |
| "grad_norm": 0.8630108833312988, |
| "learning_rate": 1.0599531555413309e-05, |
| "loss": 0.266845703125, |
| "step": 50 |
| }, |
| { |
| "epoch": 2.4285714285714284, |
| "grad_norm": 0.8821431994438171, |
| "learning_rate": 9.857304352024019e-06, |
| "loss": 0.3577880859375, |
| "step": 51 |
| }, |
| { |
| "epoch": 2.4285714285714284, |
| "eval_loss": 1.4066162109375, |
| "eval_runtime": 5.4612, |
| "eval_samples_per_second": 0.366, |
| "eval_steps_per_second": 0.183, |
| "step": 51 |
| }, |
| { |
| "epoch": 2.4761904761904763, |
| "grad_norm": 0.6822161674499512, |
| "learning_rate": 9.161862405871748e-06, |
| "loss": 0.29779052734375, |
| "step": 52 |
| }, |
| { |
| "epoch": 2.5238095238095237, |
| "grad_norm": 0.9326254725456238, |
| "learning_rate": 8.515049903978325e-06, |
| "loss": 0.45281982421875, |
| "step": 53 |
| }, |
| { |
| "epoch": 2.571428571428571, |
| "grad_norm": 1.0303138494491577, |
| "learning_rate": 7.918582076821507e-06, |
| "loss": 0.39422607421875, |
| "step": 54 |
| }, |
| { |
| "epoch": 2.619047619047619, |
| "grad_norm": 0.5410661101341248, |
| "learning_rate": 7.374040649852105e-06, |
| "loss": 1.334136962890625, |
| "step": 55 |
| }, |
| { |
| "epoch": 2.6666666666666665, |
| "grad_norm": 0.9201714396476746, |
| "learning_rate": 6.882869649042397e-06, |
| "loss": 0.35498046875, |
| "step": 56 |
| }, |
| { |
| "epoch": 2.7142857142857144, |
| "grad_norm": 0.8627307415008545, |
| "learning_rate": 6.446371571588896e-06, |
| "loss": 0.39990234375, |
| "step": 57 |
| }, |
| { |
| "epoch": 2.761904761904762, |
| "grad_norm": 0.6887866258621216, |
| "learning_rate": 6.065703931923894e-06, |
| "loss": 0.2904052734375, |
| "step": 58 |
| }, |
| { |
| "epoch": 2.8095238095238093, |
| "grad_norm": 0.4894810616970062, |
| "learning_rate": 5.741876192195292e-06, |
| "loss": 1.4284439086914062, |
| "step": 59 |
| }, |
| { |
| "epoch": 2.857142857142857, |
| "grad_norm": 0.7044135332107544, |
| "learning_rate": 5.47574708535445e-06, |
| "loss": 0.8047714233398438, |
| "step": 60 |
| }, |
| { |
| "epoch": 2.9047619047619047, |
| "grad_norm": 0.9605301022529602, |
| "learning_rate": 5.268022337950767e-06, |
| "loss": 0.38262939453125, |
| "step": 61 |
| }, |
| { |
| "epoch": 2.9523809523809526, |
| "grad_norm": 0.49748826026916504, |
| "learning_rate": 5.119252798671747e-06, |
| "loss": 1.4863128662109375, |
| "step": 62 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 0.3692787289619446, |
| "learning_rate": 5.029832977591314e-06, |
| "loss": 1.4801177978515625, |
| "step": 63 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 1.3985595703125, |
| "eval_runtime": 5.1407, |
| "eval_samples_per_second": 0.389, |
| "eval_steps_per_second": 0.195, |
| "step": 63 |
| } |
| ], |
| "logging_steps": 1.0, |
| "max_steps": 63, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 0, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.3702596321542144e+16, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
