{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0618556701030926,
  "eval_steps": 9,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.020618556701030927,
      "grad_norm": 3.2483479976654053,
      "learning_rate": 1e-05,
      "loss": 3.488,
      "step": 1
    },
    {
      "epoch": 0.020618556701030927,
      "eval_loss": 3.5999021530151367,
      "eval_runtime": 1.3665,
      "eval_samples_per_second": 30.003,
      "eval_steps_per_second": 4.391,
      "step": 1
    },
    {
      "epoch": 0.041237113402061855,
      "grad_norm": 3.2766311168670654,
      "learning_rate": 2e-05,
      "loss": 3.5399,
      "step": 2
    },
    {
      "epoch": 0.061855670103092786,
      "grad_norm": 3.8570547103881836,
      "learning_rate": 3e-05,
      "loss": 3.5997,
      "step": 3
    },
    {
      "epoch": 0.08247422680412371,
      "grad_norm": 3.21763014793396,
      "learning_rate": 4e-05,
      "loss": 3.4567,
      "step": 4
    },
    {
      "epoch": 0.10309278350515463,
      "grad_norm": 3.6865789890289307,
      "learning_rate": 5e-05,
      "loss": 3.5648,
      "step": 5
    },
    {
      "epoch": 0.12371134020618557,
      "grad_norm": 3.4130187034606934,
      "learning_rate": 6e-05,
      "loss": 3.4646,
      "step": 6
    },
    {
      "epoch": 0.14432989690721648,
      "grad_norm": 3.0942752361297607,
      "learning_rate": 7e-05,
      "loss": 3.6277,
      "step": 7
    },
    {
      "epoch": 0.16494845360824742,
      "grad_norm": 2.5932466983795166,
      "learning_rate": 8e-05,
      "loss": 3.3956,
      "step": 8
    },
    {
      "epoch": 0.18556701030927836,
      "grad_norm": 1.657394289970398,
      "learning_rate": 9e-05,
      "loss": 3.2032,
      "step": 9
    },
    {
      "epoch": 0.18556701030927836,
      "eval_loss": 3.2119903564453125,
      "eval_runtime": 0.4139,
      "eval_samples_per_second": 99.05,
      "eval_steps_per_second": 14.495,
      "step": 9
    },
    {
      "epoch": 0.20618556701030927,
      "grad_norm": 1.3544855117797852,
      "learning_rate": 0.0001,
      "loss": 3.1933,
      "step": 10
    },
    {
      "epoch": 0.2268041237113402,
      "grad_norm": 1.2699248790740967,
      "learning_rate": 9.99695413509548e-05,
      "loss": 3.2105,
      "step": 11
    },
    {
      "epoch": 0.24742268041237114,
      "grad_norm": 1.3860818147659302,
      "learning_rate": 9.987820251299122e-05,
      "loss": 3.0634,
      "step": 12
    },
    {
      "epoch": 0.26804123711340205,
      "grad_norm": 1.409780502319336,
      "learning_rate": 9.972609476841367e-05,
      "loss": 3.1134,
      "step": 13
    },
    {
      "epoch": 0.28865979381443296,
      "grad_norm": 1.366316318511963,
      "learning_rate": 9.951340343707852e-05,
      "loss": 3.061,
      "step": 14
    },
    {
      "epoch": 0.30927835051546393,
      "grad_norm": 1.3080459833145142,
      "learning_rate": 9.924038765061042e-05,
      "loss": 3.0381,
      "step": 15
    },
    {
      "epoch": 0.32989690721649484,
      "grad_norm": 1.0734084844589233,
      "learning_rate": 9.890738003669029e-05,
      "loss": 2.8557,
      "step": 16
    },
    {
      "epoch": 0.35051546391752575,
      "grad_norm": 1.0526894330978394,
      "learning_rate": 9.851478631379982e-05,
      "loss": 2.9582,
      "step": 17
    },
    {
      "epoch": 0.3711340206185567,
      "grad_norm": 1.0270817279815674,
      "learning_rate": 9.806308479691595e-05,
      "loss": 2.995,
      "step": 18
    },
    {
      "epoch": 0.3711340206185567,
      "eval_loss": 2.911524772644043,
      "eval_runtime": 0.4107,
      "eval_samples_per_second": 99.835,
      "eval_steps_per_second": 14.61,
      "step": 18
    },
    {
      "epoch": 0.3917525773195876,
      "grad_norm": 0.9986397624015808,
      "learning_rate": 9.755282581475769e-05,
      "loss": 2.9881,
      "step": 19
    },
    {
      "epoch": 0.41237113402061853,
      "grad_norm": 0.9932209849357605,
      "learning_rate": 9.698463103929542e-05,
      "loss": 2.8199,
      "step": 20
    },
    {
      "epoch": 0.4329896907216495,
      "grad_norm": 0.9946835041046143,
      "learning_rate": 9.635919272833938e-05,
      "loss": 2.8456,
      "step": 21
    },
    {
      "epoch": 0.4536082474226804,
      "grad_norm": 0.8482059240341187,
      "learning_rate": 9.567727288213005e-05,
      "loss": 2.7425,
      "step": 22
    },
    {
      "epoch": 0.4742268041237113,
      "grad_norm": 0.9843814969062805,
      "learning_rate": 9.493970231495835e-05,
      "loss": 2.727,
      "step": 23
    },
    {
      "epoch": 0.4948453608247423,
      "grad_norm": 0.7832030057907104,
      "learning_rate": 9.414737964294636e-05,
      "loss": 2.7126,
      "step": 24
    },
    {
      "epoch": 0.5154639175257731,
      "grad_norm": 0.813600480556488,
      "learning_rate": 9.330127018922194e-05,
      "loss": 2.7697,
      "step": 25
    },
    {
      "epoch": 0.5360824742268041,
      "grad_norm": 0.8165346384048462,
      "learning_rate": 9.24024048078213e-05,
      "loss": 2.6725,
      "step": 26
    },
    {
      "epoch": 0.5567010309278351,
      "grad_norm": 0.8682026267051697,
      "learning_rate": 9.145187862775209e-05,
      "loss": 2.7597,
      "step": 27
    },
    {
      "epoch": 0.5567010309278351,
      "eval_loss": 2.6847293376922607,
      "eval_runtime": 0.4158,
      "eval_samples_per_second": 98.602,
      "eval_steps_per_second": 14.43,
      "step": 27
    },
    {
      "epoch": 0.5773195876288659,
      "grad_norm": 0.9257245659828186,
      "learning_rate": 9.045084971874738e-05,
      "loss": 2.6435,
      "step": 28
    },
    {
      "epoch": 0.5979381443298969,
      "grad_norm": 0.8357184529304504,
      "learning_rate": 8.940053768033609e-05,
      "loss": 2.649,
      "step": 29
    },
    {
      "epoch": 0.6185567010309279,
      "grad_norm": 0.8404372334480286,
      "learning_rate": 8.83022221559489e-05,
      "loss": 2.607,
      "step": 30
    },
    {
      "epoch": 0.6391752577319587,
      "grad_norm": 0.7873509526252747,
      "learning_rate": 8.715724127386972e-05,
      "loss": 2.6266,
      "step": 31
    },
    {
      "epoch": 0.6597938144329897,
      "grad_norm": 0.8580664992332458,
      "learning_rate": 8.596699001693255e-05,
      "loss": 2.6335,
      "step": 32
    },
    {
      "epoch": 0.6804123711340206,
      "grad_norm": 0.7918985486030579,
      "learning_rate": 8.473291852294987e-05,
      "loss": 2.5781,
      "step": 33
    },
    {
      "epoch": 0.7010309278350515,
      "grad_norm": 0.9945449829101562,
      "learning_rate": 8.345653031794292e-05,
      "loss": 2.6204,
      "step": 34
    },
    {
      "epoch": 0.7216494845360825,
      "grad_norm": 0.972023069858551,
      "learning_rate": 8.213938048432697e-05,
      "loss": 2.4871,
      "step": 35
    },
    {
      "epoch": 0.7422680412371134,
      "grad_norm": 0.882148265838623,
      "learning_rate": 8.07830737662829e-05,
      "loss": 2.5851,
      "step": 36
    },
    {
      "epoch": 0.7422680412371134,
      "eval_loss": 2.549657106399536,
      "eval_runtime": 0.412,
      "eval_samples_per_second": 99.523,
      "eval_steps_per_second": 14.564,
      "step": 36
    },
    {
      "epoch": 0.7628865979381443,
      "grad_norm": 1.065983533859253,
      "learning_rate": 7.938926261462366e-05,
      "loss": 2.5385,
      "step": 37
    },
    {
      "epoch": 0.7835051546391752,
      "grad_norm": 0.8932860493659973,
      "learning_rate": 7.795964517353735e-05,
      "loss": 2.587,
      "step": 38
    },
    {
      "epoch": 0.8041237113402062,
      "grad_norm": 0.7516025304794312,
      "learning_rate": 7.649596321166024e-05,
      "loss": 2.4763,
      "step": 39
    },
    {
      "epoch": 0.8247422680412371,
      "grad_norm": 1.1583222150802612,
      "learning_rate": 7.500000000000001e-05,
      "loss": 2.4846,
      "step": 40
    },
    {
      "epoch": 0.845360824742268,
      "grad_norm": 0.797197699546814,
      "learning_rate": 7.347357813929454e-05,
      "loss": 2.5684,
      "step": 41
    },
    {
      "epoch": 0.865979381443299,
      "grad_norm": 0.8551321625709534,
      "learning_rate": 7.191855733945387e-05,
      "loss": 2.4838,
      "step": 42
    },
    {
      "epoch": 0.8865979381443299,
      "grad_norm": 0.8284348845481873,
      "learning_rate": 7.033683215379002e-05,
      "loss": 2.5329,
      "step": 43
    },
    {
      "epoch": 0.9072164948453608,
      "grad_norm": 0.9195374250411987,
      "learning_rate": 6.873032967079561e-05,
      "loss": 2.4583,
      "step": 44
    },
    {
      "epoch": 0.9278350515463918,
      "grad_norm": 0.8680117130279541,
      "learning_rate": 6.710100716628344e-05,
      "loss": 2.4121,
      "step": 45
    },
    {
      "epoch": 0.9278350515463918,
      "eval_loss": 2.4568657875061035,
      "eval_runtime": 0.4143,
      "eval_samples_per_second": 98.972,
      "eval_steps_per_second": 14.484,
      "step": 45
    },
    {
      "epoch": 0.9484536082474226,
      "grad_norm": 0.8823745250701904,
      "learning_rate": 6.545084971874738e-05,
      "loss": 2.4938,
      "step": 46
    },
    {
      "epoch": 0.9690721649484536,
      "grad_norm": 0.8819367289543152,
      "learning_rate": 6.378186779084995e-05,
      "loss": 2.5178,
      "step": 47
    },
    {
      "epoch": 0.9896907216494846,
      "grad_norm": 0.8066341280937195,
      "learning_rate": 6.209609477998338e-05,
      "loss": 2.4225,
      "step": 48
    },
    {
      "epoch": 1.0103092783505154,
      "grad_norm": 1.8289326429367065,
      "learning_rate": 6.0395584540887963e-05,
      "loss": 3.7317,
      "step": 49
    },
    {
      "epoch": 1.0309278350515463,
      "grad_norm": 0.7989461421966553,
      "learning_rate": 5.868240888334653e-05,
      "loss": 2.1872,
      "step": 50
    },
    {
      "epoch": 1.0515463917525774,
      "grad_norm": 0.8214808106422424,
      "learning_rate": 5.695865504800327e-05,
      "loss": 2.3618,
      "step": 51
    },
    {
      "epoch": 1.0721649484536082,
      "grad_norm": 0.8764827847480774,
      "learning_rate": 5.522642316338268e-05,
      "loss": 2.4086,
      "step": 52
    },
    {
      "epoch": 1.0927835051546393,
      "grad_norm": 0.8780012726783752,
      "learning_rate": 5.348782368720626e-05,
      "loss": 2.4445,
      "step": 53
    },
    {
      "epoch": 1.1134020618556701,
      "grad_norm": 0.8269762992858887,
      "learning_rate": 5.174497483512506e-05,
      "loss": 2.2305,
      "step": 54
    },
    {
      "epoch": 1.1134020618556701,
      "eval_loss": 2.3804149627685547,
      "eval_runtime": 0.457,
      "eval_samples_per_second": 89.72,
      "eval_steps_per_second": 13.13,
      "step": 54
    },
    {
      "epoch": 1.134020618556701,
      "grad_norm": 0.8428110480308533,
      "learning_rate": 5e-05,
      "loss": 2.4357,
      "step": 55
    },
    {
      "epoch": 1.1546391752577319,
      "grad_norm": 0.7851772904396057,
      "learning_rate": 4.825502516487497e-05,
      "loss": 2.121,
      "step": 56
    },
    {
      "epoch": 1.175257731958763,
      "grad_norm": 0.8829571604728699,
      "learning_rate": 4.6512176312793736e-05,
      "loss": 2.591,
      "step": 57
    },
    {
      "epoch": 1.1958762886597938,
      "grad_norm": 0.9812927842140198,
      "learning_rate": 4.477357683661734e-05,
      "loss": 2.4396,
      "step": 58
    },
    {
      "epoch": 1.2164948453608249,
      "grad_norm": 0.7928707003593445,
      "learning_rate": 4.3041344951996746e-05,
      "loss": 2.1293,
      "step": 59
    },
    {
      "epoch": 1.2371134020618557,
      "grad_norm": 0.7929445505142212,
      "learning_rate": 4.131759111665349e-05,
      "loss": 2.1948,
      "step": 60
    },
    {
      "epoch": 1.2577319587628866,
      "grad_norm": 0.8162508606910706,
      "learning_rate": 3.960441545911204e-05,
      "loss": 2.3651,
      "step": 61
    },
    {
      "epoch": 1.2783505154639174,
      "grad_norm": 0.8284922242164612,
      "learning_rate": 3.790390522001662e-05,
      "loss": 2.2187,
      "step": 62
    },
    {
      "epoch": 1.2989690721649485,
      "grad_norm": 0.852911114692688,
      "learning_rate": 3.6218132209150045e-05,
      "loss": 2.3589,
      "step": 63
    },
    {
      "epoch": 1.2989690721649485,
      "eval_loss": 2.3151369094848633,
      "eval_runtime": 0.4207,
      "eval_samples_per_second": 97.456,
      "eval_steps_per_second": 14.262,
      "step": 63
    },
    {
      "epoch": 1.3195876288659794,
      "grad_norm": 0.9148892760276794,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 2.4377,
      "step": 64
    },
    {
      "epoch": 1.3402061855670104,
      "grad_norm": 0.9453803300857544,
      "learning_rate": 3.289899283371657e-05,
      "loss": 2.3204,
      "step": 65
    },
    {
      "epoch": 1.3608247422680413,
      "grad_norm": 0.9316099286079407,
      "learning_rate": 3.12696703292044e-05,
      "loss": 2.3347,
      "step": 66
    },
    {
      "epoch": 1.3814432989690721,
      "grad_norm": 0.7478690147399902,
      "learning_rate": 2.9663167846209998e-05,
      "loss": 2.1146,
      "step": 67
    },
    {
      "epoch": 1.402061855670103,
      "grad_norm": 0.7751596570014954,
      "learning_rate": 2.8081442660546125e-05,
      "loss": 2.0994,
      "step": 68
    },
    {
      "epoch": 1.422680412371134,
      "grad_norm": 1.0089936256408691,
      "learning_rate": 2.6526421860705473e-05,
      "loss": 2.3435,
      "step": 69
    },
    {
      "epoch": 1.443298969072165,
      "grad_norm": 0.8275005221366882,
      "learning_rate": 2.500000000000001e-05,
      "loss": 2.3227,
      "step": 70
    },
    {
      "epoch": 1.463917525773196,
      "grad_norm": 0.9118816256523132,
      "learning_rate": 2.350403678833976e-05,
      "loss": 2.4716,
      "step": 71
    },
    {
      "epoch": 1.4845360824742269,
      "grad_norm": 0.8187054395675659,
      "learning_rate": 2.2040354826462668e-05,
      "loss": 2.1673,
      "step": 72
    },
    {
      "epoch": 1.4845360824742269,
      "eval_loss": 2.285522222518921,
      "eval_runtime": 0.4132,
      "eval_samples_per_second": 99.228,
      "eval_steps_per_second": 14.521,
      "step": 72
    },
    {
      "epoch": 1.5051546391752577,
      "grad_norm": 0.8783688545227051,
      "learning_rate": 2.061073738537635e-05,
      "loss": 2.5232,
      "step": 73
    },
    {
      "epoch": 1.5257731958762886,
      "grad_norm": 0.7773301005363464,
      "learning_rate": 1.9216926233717085e-05,
      "loss": 1.8729,
      "step": 74
    },
    {
      "epoch": 1.5463917525773194,
      "grad_norm": 0.9036346673965454,
      "learning_rate": 1.7860619515673033e-05,
      "loss": 2.3334,
      "step": 75
    },
    {
      "epoch": 1.5670103092783505,
      "grad_norm": 0.7970970273017883,
      "learning_rate": 1.6543469682057106e-05,
      "loss": 2.3437,
      "step": 76
    },
    {
      "epoch": 1.5876288659793816,
      "grad_norm": 0.7679882049560547,
      "learning_rate": 1.526708147705013e-05,
      "loss": 2.2173,
      "step": 77
    },
    {
      "epoch": 1.6082474226804124,
      "grad_norm": 0.8142104744911194,
      "learning_rate": 1.4033009983067452e-05,
      "loss": 2.2157,
      "step": 78
    },
    {
      "epoch": 1.6288659793814433,
      "grad_norm": 0.8778489828109741,
      "learning_rate": 1.2842758726130283e-05,
      "loss": 2.4768,
      "step": 79
    },
    {
      "epoch": 1.6494845360824741,
      "grad_norm": 0.8009780049324036,
      "learning_rate": 1.1697777844051105e-05,
      "loss": 2.128,
      "step": 80
    },
    {
      "epoch": 1.670103092783505,
      "grad_norm": 0.8642892241477966,
      "learning_rate": 1.0599462319663905e-05,
      "loss": 2.3527,
      "step": 81
    },
    {
      "epoch": 1.670103092783505,
      "eval_loss": 2.2726926803588867,
      "eval_runtime": 0.428,
      "eval_samples_per_second": 95.798,
      "eval_steps_per_second": 14.019,
      "step": 81
    },
    {
      "epoch": 1.690721649484536,
      "grad_norm": 0.7616624236106873,
      "learning_rate": 9.549150281252633e-06,
      "loss": 2.0641,
      "step": 82
    },
    {
      "epoch": 1.7113402061855671,
      "grad_norm": 0.8058216571807861,
      "learning_rate": 8.548121372247918e-06,
      "loss": 2.61,
      "step": 83
    },
    {
      "epoch": 1.731958762886598,
      "grad_norm": 0.7998701333999634,
      "learning_rate": 7.597595192178702e-06,
      "loss": 2.0878,
      "step": 84
    },
    {
      "epoch": 1.7525773195876289,
      "grad_norm": 0.8615913391113281,
      "learning_rate": 6.698729810778065e-06,
      "loss": 2.2409,
      "step": 85
    },
    {
      "epoch": 1.7731958762886597,
      "grad_norm": 0.8412482738494873,
      "learning_rate": 5.852620357053651e-06,
      "loss": 2.4182,
      "step": 86
    },
    {
      "epoch": 1.7938144329896906,
      "grad_norm": 0.8457291126251221,
      "learning_rate": 5.060297685041659e-06,
      "loss": 2.2028,
      "step": 87
    },
    {
      "epoch": 1.8144329896907216,
      "grad_norm": 0.895065426826477,
      "learning_rate": 4.322727117869951e-06,
      "loss": 2.3504,
      "step": 88
    },
    {
      "epoch": 1.8350515463917527,
      "grad_norm": 0.768960177898407,
      "learning_rate": 3.6408072716606346e-06,
      "loss": 2.0939,
      "step": 89
    },
    {
      "epoch": 1.8556701030927836,
      "grad_norm": 0.7762629389762878,
      "learning_rate": 3.0153689607045845e-06,
      "loss": 2.2403,
      "step": 90
    },
    {
      "epoch": 1.8556701030927836,
      "eval_loss": 2.270183563232422,
      "eval_runtime": 0.413,
      "eval_samples_per_second": 99.268,
      "eval_steps_per_second": 14.527,
      "step": 90
    },
    {
      "epoch": 1.8762886597938144,
      "grad_norm": 0.796205997467041,
      "learning_rate": 2.4471741852423237e-06,
      "loss": 2.3044,
      "step": 91
    },
    {
      "epoch": 1.8969072164948453,
      "grad_norm": 0.7898029088973999,
      "learning_rate": 1.9369152030840556e-06,
      "loss": 2.2555,
      "step": 92
    },
    {
      "epoch": 1.9175257731958761,
      "grad_norm": 0.786343514919281,
      "learning_rate": 1.4852136862001764e-06,
      "loss": 2.3107,
      "step": 93
    },
    {
      "epoch": 1.9381443298969072,
      "grad_norm": 0.7233936786651611,
      "learning_rate": 1.0926199633097157e-06,
      "loss": 2.1264,
      "step": 94
    },
    {
      "epoch": 1.9587628865979383,
      "grad_norm": 0.7573835849761963,
      "learning_rate": 7.596123493895991e-07,
      "loss": 2.1235,
      "step": 95
    },
    {
      "epoch": 1.9793814432989691,
      "grad_norm": 0.8158411383628845,
      "learning_rate": 4.865965629214819e-07,
      "loss": 2.3631,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.1228227615356445,
      "learning_rate": 2.7390523158633554e-07,
      "loss": 3.367,
      "step": 97
    },
    {
      "epoch": 2.020618556701031,
      "grad_norm": 0.8564669489860535,
      "learning_rate": 1.2179748700879012e-07,
      "loss": 2.226,
      "step": 98
    },
    {
      "epoch": 2.0412371134020617,
      "grad_norm": 0.7765387296676636,
      "learning_rate": 3.04586490452119e-08,
      "loss": 2.2284,
      "step": 99
    },
    {
      "epoch": 2.0412371134020617,
      "eval_loss": 2.268873929977417,
      "eval_runtime": 0.4154,
      "eval_samples_per_second": 98.694,
      "eval_steps_per_second": 14.443,
      "step": 99
    },
    {
      "epoch": 2.0618556701030926,
      "grad_norm": 0.7744645476341248,
      "learning_rate": 0.0,
      "loss": 2.1928,
      "step": 100
    }
  ],
  "logging_steps": 1,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 1379698511708160.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}