{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9846153846153847,
  "eval_steps": 9,
  "global_step": 97,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03076923076923077,
      "grad_norm": 0.19627095758914948,
      "learning_rate": 1e-05,
      "loss": 10.3769,
      "step": 1
    },
    {
      "epoch": 0.03076923076923077,
      "eval_loss": 10.386632919311523,
      "eval_runtime": 0.0801,
      "eval_samples_per_second": 1361.483,
      "eval_steps_per_second": 49.963,
      "step": 1
    },
    {
      "epoch": 0.06153846153846154,
      "grad_norm": 0.20535489916801453,
      "learning_rate": 2e-05,
      "loss": 10.3764,
      "step": 2
    },
    {
      "epoch": 0.09230769230769231,
      "grad_norm": 0.1900486946105957,
      "learning_rate": 3e-05,
      "loss": 10.382,
      "step": 3
    },
    {
      "epoch": 0.12307692307692308,
      "grad_norm": 0.2189124971628189,
      "learning_rate": 4e-05,
      "loss": 10.3854,
      "step": 4
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 0.19613224267959595,
      "learning_rate": 5e-05,
      "loss": 10.3846,
      "step": 5
    },
    {
      "epoch": 0.18461538461538463,
      "grad_norm": 0.21051953732967377,
      "learning_rate": 6e-05,
      "loss": 10.3951,
      "step": 6
    },
    {
      "epoch": 0.2153846153846154,
      "grad_norm": 0.193317711353302,
      "learning_rate": 7e-05,
      "loss": 10.4099,
      "step": 7
    },
    {
      "epoch": 0.24615384615384617,
      "grad_norm": 0.22925445437431335,
      "learning_rate": 8e-05,
      "loss": 10.4001,
      "step": 8
    },
    {
      "epoch": 0.27692307692307694,
      "grad_norm": 0.2118426263332367,
      "learning_rate": 9e-05,
      "loss": 10.4226,
      "step": 9
    },
    {
      "epoch": 0.27692307692307694,
      "eval_loss": 10.384289741516113,
      "eval_runtime": 0.075,
      "eval_samples_per_second": 1452.898,
      "eval_steps_per_second": 53.317,
      "step": 9
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 0.22410708665847778,
      "learning_rate": 0.0001,
      "loss": 10.3974,
      "step": 10
    },
    {
      "epoch": 0.3384615384615385,
      "grad_norm": 0.28219085931777954,
      "learning_rate": 9.996740476948385e-05,
      "loss": 10.3593,
      "step": 11
    },
    {
      "epoch": 0.36923076923076925,
      "grad_norm": 0.2661738991737366,
      "learning_rate": 9.98696615758975e-05,
      "loss": 10.3826,
      "step": 12
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.26590314507484436,
      "learning_rate": 9.970689785771798e-05,
      "loss": 10.3853,
      "step": 13
    },
    {
      "epoch": 0.4307692307692308,
      "grad_norm": 0.23882359266281128,
      "learning_rate": 9.947932582778188e-05,
      "loss": 10.3944,
      "step": 14
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 0.2391405999660492,
      "learning_rate": 9.918724219660013e-05,
      "loss": 10.357,
      "step": 15
    },
    {
      "epoch": 0.49230769230769234,
      "grad_norm": 0.2403474599123001,
      "learning_rate": 9.883102778550434e-05,
      "loss": 10.3803,
      "step": 16
    },
    {
      "epoch": 0.5230769230769231,
      "grad_norm": 0.22196514904499054,
      "learning_rate": 9.841114703012817e-05,
      "loss": 10.3643,
      "step": 17
    },
    {
      "epoch": 0.5538461538461539,
      "grad_norm": 0.245796337723732,
      "learning_rate": 9.792814737487207e-05,
      "loss": 10.4181,
      "step": 18
    },
    {
      "epoch": 0.5538461538461539,
      "eval_loss": 10.378185272216797,
      "eval_runtime": 0.0762,
      "eval_samples_per_second": 1429.641,
      "eval_steps_per_second": 52.464,
      "step": 18
    },
    {
      "epoch": 0.5846153846153846,
      "grad_norm": 0.2934595048427582,
      "learning_rate": 9.738265855914013e-05,
      "loss": 10.3524,
      "step": 19
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 0.29529669880867004,
      "learning_rate": 9.677539179628005e-05,
      "loss": 10.3852,
      "step": 20
    },
    {
      "epoch": 0.6461538461538462,
      "grad_norm": 0.2764834761619568,
      "learning_rate": 9.610713884629666e-05,
      "loss": 10.3627,
      "step": 21
    },
    {
      "epoch": 0.676923076923077,
      "grad_norm": 0.270579993724823,
      "learning_rate": 9.537877098354786e-05,
      "loss": 10.3802,
      "step": 22
    },
    {
      "epoch": 0.7076923076923077,
      "grad_norm": 0.28736695647239685,
      "learning_rate": 9.459123786076912e-05,
      "loss": 10.3475,
      "step": 23
    },
    {
      "epoch": 0.7384615384615385,
      "grad_norm": 0.27069252729415894,
      "learning_rate": 9.374556627090749e-05,
      "loss": 10.3726,
      "step": 24
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 0.2778293192386627,
      "learning_rate": 9.284285880837946e-05,
      "loss": 10.3683,
      "step": 25
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.3016285002231598,
      "learning_rate": 9.188429243149824e-05,
      "loss": 10.3652,
      "step": 26
    },
    {
      "epoch": 0.8307692307692308,
      "grad_norm": 0.3102306127548218,
      "learning_rate": 9.087111692794459e-05,
      "loss": 10.3612,
      "step": 27
    },
    {
      "epoch": 0.8307692307692308,
      "eval_loss": 10.371162414550781,
      "eval_runtime": 0.0804,
      "eval_samples_per_second": 1355.392,
      "eval_steps_per_second": 49.739,
      "step": 27
    },
    {
      "epoch": 0.8615384615384616,
      "grad_norm": 0.2973518967628479,
      "learning_rate": 8.980465328528219e-05,
      "loss": 10.3562,
      "step": 28
    },
    {
      "epoch": 0.8923076923076924,
      "grad_norm": 0.27339646220207214,
      "learning_rate": 8.868629196864182e-05,
      "loss": 10.3745,
      "step": 29
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 0.30518829822540283,
      "learning_rate": 8.751749110782012e-05,
      "loss": 10.3788,
      "step": 30
    },
    {
      "epoch": 0.9538461538461539,
      "grad_norm": 0.30676740407943726,
      "learning_rate": 8.629977459615655e-05,
      "loss": 10.3631,
      "step": 31
    },
    {
      "epoch": 0.9846153846153847,
      "grad_norm": 0.3426137864589691,
      "learning_rate": 8.503473010366713e-05,
      "loss": 10.3782,
      "step": 32
    },
    {
      "epoch": 1.0153846153846153,
      "grad_norm": 0.43814149498939514,
      "learning_rate": 8.37240070070257e-05,
      "loss": 14.7399,
      "step": 33
    },
    {
      "epoch": 1.0461538461538462,
      "grad_norm": 0.39517074823379517,
      "learning_rate": 8.236931423909138e-05,
      "loss": 11.6581,
      "step": 34
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 0.286582887172699,
      "learning_rate": 8.097241806078615e-05,
      "loss": 9.8718,
      "step": 35
    },
    {
      "epoch": 1.1076923076923078,
      "grad_norm": 0.3339255154132843,
      "learning_rate": 7.953513975822755e-05,
      "loss": 9.8388,
      "step": 36
    },
    {
      "epoch": 1.1076923076923078,
      "eval_loss": 10.362679481506348,
      "eval_runtime": 0.0735,
      "eval_samples_per_second": 1482.077,
      "eval_steps_per_second": 54.388,
      "step": 36
    },
    {
      "epoch": 1.1384615384615384,
      "grad_norm": 0.38969117403030396,
      "learning_rate": 7.805935326811912e-05,
      "loss": 9.6292,
      "step": 37
    },
    {
      "epoch": 1.1692307692307693,
      "grad_norm": 0.4563569724559784,
      "learning_rate": 7.654698273449435e-05,
      "loss": 11.4989,
      "step": 38
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.5030809044837952,
      "learning_rate": 7.500000000000001e-05,
      "loss": 11.8672,
      "step": 39
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 0.32794782519340515,
      "learning_rate": 7.342042203498951e-05,
      "loss": 9.4949,
      "step": 40
    },
    {
      "epoch": 1.2615384615384615,
      "grad_norm": 0.3771244287490845,
      "learning_rate": 7.181030830777837e-05,
      "loss": 8.6339,
      "step": 41
    },
    {
      "epoch": 1.2923076923076924,
      "grad_norm": 0.37634986639022827,
      "learning_rate": 7.017175809949044e-05,
      "loss": 9.5719,
      "step": 42
    },
    {
      "epoch": 1.323076923076923,
      "grad_norm": 0.5357686877250671,
      "learning_rate": 6.850690776699573e-05,
      "loss": 13.4886,
      "step": 43
    },
    {
      "epoch": 1.353846153846154,
      "grad_norm": 0.41375601291656494,
      "learning_rate": 6.681792795750875e-05,
      "loss": 10.2368,
      "step": 44
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.3804188370704651,
      "learning_rate": 6.510702077847863e-05,
      "loss": 8.42,
      "step": 45
    },
    {
      "epoch": 1.3846153846153846,
      "eval_loss": 10.35329532623291,
      "eval_runtime": 0.0829,
      "eval_samples_per_second": 1314.084,
      "eval_steps_per_second": 48.223,
      "step": 45
    },
    {
      "epoch": 1.4153846153846155,
      "grad_norm": 0.5498846173286438,
      "learning_rate": 6.337641692646106e-05,
      "loss": 10.7262,
      "step": 46
    },
    {
      "epoch": 1.4461538461538461,
      "grad_norm": 0.4699338972568512,
      "learning_rate": 6.162837277871553e-05,
      "loss": 10.9246,
      "step": 47
    },
    {
      "epoch": 1.476923076923077,
      "grad_norm": 0.47309648990631104,
      "learning_rate": 5.9865167451320005e-05,
      "loss": 10.4618,
      "step": 48
    },
    {
      "epoch": 1.5076923076923077,
      "grad_norm": 0.44090601801872253,
      "learning_rate": 5.808909982763825e-05,
      "loss": 9.5822,
      "step": 49
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.6003273129463196,
      "learning_rate": 5.6302485561014475e-05,
      "loss": 12.5208,
      "step": 50
    },
    {
      "epoch": 1.5692307692307692,
      "grad_norm": 0.43022680282592773,
      "learning_rate": 5.4507654055603275e-05,
      "loss": 9.2575,
      "step": 51
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.38099291920661926,
      "learning_rate": 5.270694542927088e-05,
      "loss": 9.0721,
      "step": 52
    },
    {
      "epoch": 1.6307692307692307,
      "grad_norm": 0.5643202662467957,
      "learning_rate": 5.090270746252802e-05,
      "loss": 11.2072,
      "step": 53
    },
    {
      "epoch": 1.6615384615384614,
      "grad_norm": 0.4488671123981476,
      "learning_rate": 4.909729253747197e-05,
      "loss": 9.4193,
      "step": 54
    },
    {
      "epoch": 1.6615384615384614,
      "eval_loss": 10.343835830688477,
      "eval_runtime": 0.071,
      "eval_samples_per_second": 1535.673,
      "eval_steps_per_second": 56.355,
      "step": 54
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 0.5811948776245117,
      "learning_rate": 4.729305457072913e-05,
      "loss": 11.7481,
      "step": 55
    },
    {
      "epoch": 1.7230769230769232,
      "grad_norm": 0.41047021746635437,
      "learning_rate": 4.549234594439674e-05,
      "loss": 9.4589,
      "step": 56
    },
    {
      "epoch": 1.7538461538461538,
      "grad_norm": 0.522146999835968,
      "learning_rate": 4.3697514438985536e-05,
      "loss": 11.107,
      "step": 57
    },
    {
      "epoch": 1.7846153846153845,
      "grad_norm": 0.5175060629844666,
      "learning_rate": 4.1910900172361764e-05,
      "loss": 10.9767,
      "step": 58
    },
    {
      "epoch": 1.8153846153846154,
      "grad_norm": 0.4755529463291168,
      "learning_rate": 4.0134832548680006e-05,
      "loss": 9.3121,
      "step": 59
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 0.4807398021221161,
      "learning_rate": 3.8371627221284495e-05,
      "loss": 9.6356,
      "step": 60
    },
    {
      "epoch": 1.876923076923077,
      "grad_norm": 0.5616713166236877,
      "learning_rate": 3.6623583073538966e-05,
      "loss": 10.963,
      "step": 61
    },
    {
      "epoch": 1.9076923076923076,
      "grad_norm": 0.5250815153121948,
      "learning_rate": 3.489297922152136e-05,
      "loss": 10.1303,
      "step": 62
    },
    {
      "epoch": 1.9384615384615385,
      "grad_norm": 0.6839239001274109,
      "learning_rate": 3.3182072042491244e-05,
      "loss": 11.6986,
      "step": 63
    },
    {
      "epoch": 1.9384615384615385,
      "eval_loss": 10.335771560668945,
      "eval_runtime": 0.0748,
      "eval_samples_per_second": 1457.284,
      "eval_steps_per_second": 53.478,
      "step": 63
    },
    {
      "epoch": 1.9692307692307693,
      "grad_norm": 0.5417065620422363,
      "learning_rate": 3.149309223300428e-05,
      "loss": 9.5666,
      "step": 64
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.8907036185264587,
      "learning_rate": 2.982824190050958e-05,
      "loss": 15.1453,
      "step": 65
    },
    {
      "epoch": 2.0307692307692307,
      "grad_norm": 0.5486352443695068,
      "learning_rate": 2.8189691692221627e-05,
      "loss": 10.3319,
      "step": 66
    },
    {
      "epoch": 2.0615384615384613,
      "grad_norm": 0.5944435000419617,
      "learning_rate": 2.65795779650105e-05,
      "loss": 10.3267,
      "step": 67
    },
    {
      "epoch": 2.0923076923076924,
      "grad_norm": 0.5956578254699707,
      "learning_rate": 2.500000000000001e-05,
      "loss": 10.313,
      "step": 68
    },
    {
      "epoch": 2.123076923076923,
      "grad_norm": 0.592414379119873,
      "learning_rate": 2.3453017265505673e-05,
      "loss": 10.3379,
      "step": 69
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.6250660419464111,
      "learning_rate": 2.194064673188089e-05,
      "loss": 10.3427,
      "step": 70
    },
    {
      "epoch": 2.184615384615385,
      "grad_norm": 0.6292226910591125,
      "learning_rate": 2.0464860241772455e-05,
      "loss": 10.3298,
      "step": 71
    },
    {
      "epoch": 2.2153846153846155,
      "grad_norm": 0.5584103465080261,
      "learning_rate": 1.902758193921385e-05,
      "loss": 10.3342,
      "step": 72
    },
    {
      "epoch": 2.2153846153846155,
      "eval_loss": 10.32968807220459,
      "eval_runtime": 0.0724,
      "eval_samples_per_second": 1505.761,
      "eval_steps_per_second": 55.257,
      "step": 72
    },
    {
      "epoch": 2.246153846153846,
      "grad_norm": 0.5744684934616089,
      "learning_rate": 1.7630685760908622e-05,
      "loss": 10.3275,
      "step": 73
    },
    {
      "epoch": 2.276923076923077,
      "grad_norm": 0.6510607600212097,
      "learning_rate": 1.6275992992974308e-05,
      "loss": 10.3399,
      "step": 74
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 0.6014554500579834,
      "learning_rate": 1.4965269896332885e-05,
      "loss": 10.3441,
      "step": 75
    },
    {
      "epoch": 2.3384615384615386,
      "grad_norm": 0.6675054430961609,
      "learning_rate": 1.3700225403843469e-05,
      "loss": 10.3299,
      "step": 76
    },
    {
      "epoch": 2.3692307692307693,
      "grad_norm": 0.5850571990013123,
      "learning_rate": 1.2482508892179884e-05,
      "loss": 10.3144,
      "step": 77
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.6561471223831177,
      "learning_rate": 1.1313708031358183e-05,
      "loss": 10.3181,
      "step": 78
    },
    {
      "epoch": 2.430769230769231,
      "grad_norm": 0.6304017305374146,
      "learning_rate": 1.0195346714717813e-05,
      "loss": 10.3316,
      "step": 79
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.627305269241333,
      "learning_rate": 9.12888307205541e-06,
      "loss": 10.321,
      "step": 80
    },
    {
      "epoch": 2.4923076923076923,
      "grad_norm": 0.6237972974777222,
      "learning_rate": 8.115707568501768e-06,
      "loss": 10.3571,
      "step": 81
    },
    {
      "epoch": 2.4923076923076923,
      "eval_loss": 10.326565742492676,
      "eval_runtime": 0.0741,
      "eval_samples_per_second": 1471.008,
      "eval_steps_per_second": 53.982,
      "step": 81
    },
    {
      "epoch": 2.523076923076923,
      "grad_norm": 0.627188503742218,
      "learning_rate": 7.157141191620548e-06,
      "loss": 10.3486,
      "step": 82
    },
    {
      "epoch": 2.5538461538461537,
      "grad_norm": 0.63338303565979,
      "learning_rate": 6.2544337290925185e-06,
      "loss": 10.3175,
      "step": 83
    },
    {
      "epoch": 2.5846153846153848,
      "grad_norm": 0.5604279041290283,
      "learning_rate": 5.408762139230888e-06,
      "loss": 10.3171,
      "step": 84
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 0.6100577712059021,
      "learning_rate": 4.621229016452156e-06,
      "loss": 10.3251,
      "step": 85
    },
    {
      "epoch": 2.646153846153846,
      "grad_norm": 0.6354370713233948,
      "learning_rate": 3.892861153703342e-06,
      "loss": 10.3134,
      "step": 86
    },
    {
      "epoch": 2.676923076923077,
      "grad_norm": 0.6003406643867493,
      "learning_rate": 3.2246082037199532e-06,
      "loss": 10.3045,
      "step": 87
    },
    {
      "epoch": 2.707692307692308,
      "grad_norm": 0.5505728125572205,
      "learning_rate": 2.6173414408598827e-06,
      "loss": 10.3232,
      "step": 88
    },
    {
      "epoch": 2.7384615384615385,
      "grad_norm": 0.6088215708732605,
      "learning_rate": 2.0718526251279346e-06,
      "loss": 10.3132,
      "step": 89
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 0.6168414950370789,
      "learning_rate": 1.5888529698718346e-06,
      "loss": 10.3199,
      "step": 90
    },
    {
      "epoch": 2.769230769230769,
      "eval_loss": 10.325533866882324,
      "eval_runtime": 0.0753,
      "eval_samples_per_second": 1447.667,
      "eval_steps_per_second": 53.125,
      "step": 90
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.6565277576446533,
      "learning_rate": 1.1689722144956671e-06,
      "loss": 10.3263,
      "step": 91
    },
    {
      "epoch": 2.830769230769231,
      "grad_norm": 0.7084165811538696,
      "learning_rate": 8.127578033998662e-07,
      "loss": 10.3452,
      "step": 92
    },
    {
      "epoch": 2.8615384615384616,
      "grad_norm": 0.5756314396858215,
      "learning_rate": 5.206741722181386e-07,
      "loss": 10.319,
      "step": 93
    },
    {
      "epoch": 2.8923076923076922,
      "grad_norm": 0.5791900753974915,
      "learning_rate": 2.9310214228202013e-07,
      "loss": 10.308,
      "step": 94
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 0.6099293828010559,
      "learning_rate": 1.3033842410251075e-07,
      "loss": 10.315,
      "step": 95
    },
    {
      "epoch": 2.953846153846154,
      "grad_norm": 0.5944267511367798,
      "learning_rate": 3.259523051615254e-08,
      "loss": 10.3374,
      "step": 96
    },
    {
      "epoch": 2.9846153846153847,
      "grad_norm": 0.618131160736084,
      "learning_rate": 0.0,
      "loss": 10.3129,
      "step": 97
    }
  ],
  "logging_steps": 1,
  "max_steps": 97,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 20293349277696.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}