{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9420289855072463,
  "eval_steps": 1,
  "global_step": 68,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.028985507246376812,
      "eval_loss": 3.890916585922241,
      "eval_runtime": 2.366,
      "eval_samples_per_second": 253.597,
      "eval_steps_per_second": 31.7,
      "step": 1
    },
    {
      "epoch": 0.057971014492753624,
      "grad_norm": 251.42497029296223,
      "learning_rate": 6.666666666666667e-06,
      "loss": 3.8462,
      "step": 2
    },
    {
      "epoch": 0.057971014492753624,
      "eval_loss": 3.1606125831604004,
      "eval_runtime": 2.3735,
      "eval_samples_per_second": 252.789,
      "eval_steps_per_second": 31.599,
      "step": 2
    },
    {
      "epoch": 0.08695652173913043,
      "eval_loss": 1.4003069400787354,
      "eval_runtime": 2.3791,
      "eval_samples_per_second": 252.195,
      "eval_steps_per_second": 31.524,
      "step": 3
    },
    {
      "epoch": 0.11594202898550725,
      "grad_norm": 174.88885660985272,
      "learning_rate": 9.994161134161635e-06,
      "loss": 2.3026,
      "step": 4
    },
    {
      "epoch": 0.11594202898550725,
      "eval_loss": 0.5247076749801636,
      "eval_runtime": 2.3571,
      "eval_samples_per_second": 254.551,
      "eval_steps_per_second": 31.819,
      "step": 4
    },
    {
      "epoch": 0.14492753623188406,
      "eval_loss": 0.25349560379981995,
      "eval_runtime": 2.3667,
      "eval_samples_per_second": 253.52,
      "eval_steps_per_second": 31.69,
      "step": 5
    },
    {
      "epoch": 0.17391304347826086,
      "grad_norm": 9.541839408808285,
      "learning_rate": 9.947531997255256e-06,
      "loss": 0.3725,
      "step": 6
    },
    {
      "epoch": 0.17391304347826086,
      "eval_loss": 0.12238868325948715,
      "eval_runtime": 2.3709,
      "eval_samples_per_second": 253.068,
      "eval_steps_per_second": 31.633,
      "step": 6
    },
    {
      "epoch": 0.2028985507246377,
      "eval_loss": 0.07106433808803558,
      "eval_runtime": 2.3595,
      "eval_samples_per_second": 254.287,
      "eval_steps_per_second": 31.786,
      "step": 7
    },
    {
      "epoch": 0.2318840579710145,
      "grad_norm": 8.373395170519098,
      "learning_rate": 9.854709087130261e-06,
      "loss": 0.1704,
      "step": 8
    },
    {
      "epoch": 0.2318840579710145,
      "eval_loss": 0.07050631195306778,
      "eval_runtime": 2.374,
      "eval_samples_per_second": 252.739,
      "eval_steps_per_second": 31.592,
      "step": 8
    },
    {
      "epoch": 0.2608695652173913,
      "eval_loss": 0.0841919556260109,
      "eval_runtime": 2.3733,
      "eval_samples_per_second": 252.81,
      "eval_steps_per_second": 31.601,
      "step": 9
    },
    {
      "epoch": 0.2898550724637681,
      "grad_norm": 9.328057178580242,
      "learning_rate": 9.716559066288716e-06,
      "loss": 0.0719,
      "step": 10
    },
    {
      "epoch": 0.2898550724637681,
      "eval_loss": 0.06837386637926102,
      "eval_runtime": 2.402,
      "eval_samples_per_second": 249.788,
      "eval_steps_per_second": 31.223,
      "step": 10
    },
    {
      "epoch": 0.3188405797101449,
      "eval_loss": 0.08372741937637329,
      "eval_runtime": 2.3771,
      "eval_samples_per_second": 252.413,
      "eval_steps_per_second": 31.552,
      "step": 11
    },
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 8.195685627940097,
      "learning_rate": 9.534371804252727e-06,
      "loss": 0.0719,
      "step": 12
    },
    {
      "epoch": 0.34782608695652173,
      "eval_loss": 0.07937659323215485,
      "eval_runtime": 2.3703,
      "eval_samples_per_second": 253.131,
      "eval_steps_per_second": 31.641,
      "step": 12
    },
    {
      "epoch": 0.37681159420289856,
      "eval_loss": 0.06787987053394318,
      "eval_runtime": 2.3654,
      "eval_samples_per_second": 253.659,
      "eval_steps_per_second": 31.707,
      "step": 13
    },
    {
      "epoch": 0.4057971014492754,
      "grad_norm": 3.0846120042199954,
      "learning_rate": 9.309848334400247e-06,
      "loss": 0.0729,
      "step": 14
    },
    {
      "epoch": 0.4057971014492754,
      "eval_loss": 0.060705069452524185,
      "eval_runtime": 2.3698,
      "eval_samples_per_second": 253.186,
      "eval_steps_per_second": 31.648,
      "step": 14
    },
    {
      "epoch": 0.43478260869565216,
      "eval_loss": 0.06819155067205429,
      "eval_runtime": 2.3712,
      "eval_samples_per_second": 253.037,
      "eval_steps_per_second": 31.63,
      "step": 15
    },
    {
      "epoch": 0.463768115942029,
      "grad_norm": 3.7022895578403414,
      "learning_rate": 9.045084971874738e-06,
      "loss": 0.0639,
      "step": 16
    },
    {
      "epoch": 0.463768115942029,
      "eval_loss": 0.06595086306333542,
      "eval_runtime": 2.3702,
      "eval_samples_per_second": 253.148,
      "eval_steps_per_second": 31.643,
      "step": 16
    },
    {
      "epoch": 0.4927536231884058,
      "eval_loss": 0.06074570491909981,
      "eval_runtime": 2.3929,
      "eval_samples_per_second": 250.74,
      "eval_steps_per_second": 31.342,
      "step": 17
    },
    {
      "epoch": 0.5217391304347826,
      "grad_norm": 2.6201997383559235,
      "learning_rate": 8.742553740855507e-06,
      "loss": 0.0659,
      "step": 18
    },
    {
      "epoch": 0.5217391304347826,
      "eval_loss": 0.060938794165849686,
      "eval_runtime": 2.3734,
      "eval_samples_per_second": 252.797,
      "eval_steps_per_second": 31.6,
      "step": 18
    },
    {
      "epoch": 0.5507246376811594,
      "eval_loss": 0.05989724025130272,
      "eval_runtime": 2.386,
      "eval_samples_per_second": 251.47,
      "eval_steps_per_second": 31.434,
      "step": 19
    },
    {
      "epoch": 0.5797101449275363,
      "grad_norm": 1.5759739495214995,
      "learning_rate": 8.405079293933986e-06,
      "loss": 0.0584,
      "step": 20
    },
    {
      "epoch": 0.5797101449275363,
      "eval_loss": 0.05950001999735832,
      "eval_runtime": 2.3751,
      "eval_samples_per_second": 252.625,
      "eval_steps_per_second": 31.578,
      "step": 20
    },
    {
      "epoch": 0.6086956521739131,
      "eval_loss": 0.057929884642362595,
      "eval_runtime": 2.3951,
      "eval_samples_per_second": 250.515,
      "eval_steps_per_second": 31.314,
      "step": 21
    },
    {
      "epoch": 0.6376811594202898,
      "grad_norm": 0.9083257875769617,
      "learning_rate": 8.035812539093557e-06,
      "loss": 0.059,
      "step": 22
    },
    {
      "epoch": 0.6376811594202898,
      "eval_loss": 0.05716191604733467,
      "eval_runtime": 2.3793,
      "eval_samples_per_second": 252.176,
      "eval_steps_per_second": 31.522,
      "step": 22
    },
    {
      "epoch": 0.6666666666666666,
      "eval_loss": 0.05785393714904785,
      "eval_runtime": 2.3743,
      "eval_samples_per_second": 252.704,
      "eval_steps_per_second": 31.588,
      "step": 23
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 9.258583060973042,
      "learning_rate": 7.638201220530664e-06,
      "loss": 0.1069,
      "step": 24
    },
    {
      "epoch": 0.6956521739130435,
      "eval_loss": 0.06170507147908211,
      "eval_runtime": 2.3968,
      "eval_samples_per_second": 250.337,
      "eval_steps_per_second": 31.292,
      "step": 24
    },
    {
      "epoch": 0.7246376811594203,
      "eval_loss": 0.06007671728730202,
      "eval_runtime": 2.375,
      "eval_samples_per_second": 252.631,
      "eval_steps_per_second": 31.579,
      "step": 25
    },
    {
      "epoch": 0.7536231884057971,
      "grad_norm": 2.788879674143748,
      "learning_rate": 7.215957727996208e-06,
      "loss": 0.0585,
      "step": 26
    },
    {
      "epoch": 0.7536231884057971,
      "eval_loss": 0.05631522089242935,
      "eval_runtime": 2.4038,
      "eval_samples_per_second": 249.609,
      "eval_steps_per_second": 31.201,
      "step": 26
    },
    {
      "epoch": 0.782608695652174,
      "eval_loss": 0.05981193110346794,
      "eval_runtime": 2.3841,
      "eval_samples_per_second": 251.665,
      "eval_steps_per_second": 31.458,
      "step": 27
    },
    {
      "epoch": 0.8115942028985508,
      "grad_norm": 3.982184927790719,
      "learning_rate": 6.773024435212678e-06,
      "loss": 0.097,
      "step": 28
    },
    {
      "epoch": 0.8115942028985508,
      "eval_loss": 0.05898861214518547,
      "eval_runtime": 2.3921,
      "eval_samples_per_second": 250.824,
      "eval_steps_per_second": 31.353,
      "step": 28
    },
    {
      "epoch": 0.8405797101449275,
      "eval_loss": 0.05481765791773796,
      "eval_runtime": 2.3767,
      "eval_samples_per_second": 252.451,
      "eval_steps_per_second": 31.556,
      "step": 29
    },
    {
      "epoch": 0.8695652173913043,
      "grad_norm": 0.18833058180333875,
      "learning_rate": 6.313536890992935e-06,
      "loss": 0.059,
      "step": 30
    },
    {
      "epoch": 0.8695652173913043,
      "eval_loss": 0.05593809857964516,
      "eval_runtime": 2.3764,
      "eval_samples_per_second": 252.478,
      "eval_steps_per_second": 31.56,
      "step": 30
    },
    {
      "epoch": 0.8985507246376812,
      "eval_loss": 0.05695917829871178,
      "eval_runtime": 2.39,
      "eval_samples_per_second": 251.049,
      "eval_steps_per_second": 31.381,
      "step": 31
    },
    {
      "epoch": 0.927536231884058,
      "grad_norm": 3.4944330077548207,
      "learning_rate": 5.841785206735192e-06,
      "loss": 0.0695,
      "step": 32
    },
    {
      "epoch": 0.927536231884058,
      "eval_loss": 0.05482754111289978,
      "eval_runtime": 2.3734,
      "eval_samples_per_second": 252.799,
      "eval_steps_per_second": 31.6,
      "step": 32
    },
    {
      "epoch": 0.9565217391304348,
      "eval_loss": 0.055433232337236404,
      "eval_runtime": 2.3729,
      "eval_samples_per_second": 252.86,
      "eval_steps_per_second": 31.607,
      "step": 33
    },
    {
      "epoch": 0.9855072463768116,
      "grad_norm": 2.742927364863374,
      "learning_rate": 5.362174000808813e-06,
      "loss": 0.0533,
      "step": 34
    },
    {
      "epoch": 0.9855072463768116,
      "eval_loss": 0.05639192834496498,
      "eval_runtime": 2.3727,
      "eval_samples_per_second": 252.873,
      "eval_steps_per_second": 31.609,
      "step": 34
    },
    {
      "epoch": 1.0144927536231885,
      "eval_loss": 0.054112281650304794,
      "eval_runtime": 2.37,
      "eval_samples_per_second": 253.168,
      "eval_steps_per_second": 31.646,
      "step": 35
    },
    {
      "epoch": 1.0144927536231885,
      "grad_norm": 1.5488276127209792,
      "learning_rate": 4.87918127381934e-06,
      "loss": 0.0544,
      "step": 36
    },
    {
      "epoch": 1.0144927536231885,
      "eval_loss": 0.0547836609184742,
      "eval_runtime": 2.372,
      "eval_samples_per_second": 252.956,
      "eval_steps_per_second": 31.619,
      "step": 36
    },
    {
      "epoch": 1.0434782608695652,
      "eval_loss": 0.05551725998520851,
      "eval_runtime": 2.376,
      "eval_samples_per_second": 252.528,
      "eval_steps_per_second": 31.566,
      "step": 37
    },
    {
      "epoch": 1.0724637681159421,
      "grad_norm": 2.1899578742270838,
      "learning_rate": 4.397316598723385e-06,
      "loss": 0.0555,
      "step": 38
    },
    {
      "epoch": 1.0724637681159421,
      "eval_loss": 0.05312129110097885,
      "eval_runtime": 2.3838,
      "eval_samples_per_second": 251.695,
      "eval_steps_per_second": 31.462,
      "step": 38
    },
    {
      "epoch": 1.1014492753623188,
      "eval_loss": 0.053158555179834366,
      "eval_runtime": 2.3919,
      "eval_samples_per_second": 250.844,
      "eval_steps_per_second": 31.355,
      "step": 39
    },
    {
      "epoch": 1.1304347826086956,
      "grad_norm": 2.1733146824122724,
      "learning_rate": 3.92107901616097e-06,
      "loss": 0.0524,
      "step": 40
    },
    {
      "epoch": 1.1304347826086956,
      "eval_loss": 0.05355316773056984,
      "eval_runtime": 2.3826,
      "eval_samples_per_second": 251.826,
      "eval_steps_per_second": 31.478,
      "step": 40
    },
    {
      "epoch": 1.1594202898550725,
      "eval_loss": 0.05187664180994034,
      "eval_runtime": 2.3848,
      "eval_samples_per_second": 251.594,
      "eval_steps_per_second": 31.449,
      "step": 41
    },
    {
      "epoch": 1.1884057971014492,
      "grad_norm": 2.315639907284586,
      "learning_rate": 3.4549150281252635e-06,
      "loss": 0.0641,
      "step": 42
    },
    {
      "epoch": 1.1884057971014492,
      "eval_loss": 0.05204891413450241,
      "eval_runtime": 2.3997,
      "eval_samples_per_second": 250.028,
      "eval_steps_per_second": 31.254,
      "step": 42
    },
    {
      "epoch": 1.2173913043478262,
      "eval_loss": 0.052227165549993515,
      "eval_runtime": 2.3864,
      "eval_samples_per_second": 251.429,
      "eval_steps_per_second": 31.429,
      "step": 43
    },
    {
      "epoch": 1.2463768115942029,
      "grad_norm": 1.5853052624042796,
      "learning_rate": 3.0031770821715233e-06,
      "loss": 0.0494,
      "step": 44
    },
    {
      "epoch": 1.2463768115942029,
      "eval_loss": 0.05136393383145332,
      "eval_runtime": 2.3856,
      "eval_samples_per_second": 251.513,
      "eval_steps_per_second": 31.439,
      "step": 44
    },
    {
      "epoch": 1.2753623188405796,
      "eval_loss": 0.051076941192150116,
      "eval_runtime": 2.3923,
      "eval_samples_per_second": 250.8,
      "eval_steps_per_second": 31.35,
      "step": 45
    },
    {
      "epoch": 1.3043478260869565,
      "grad_norm": 0.2342975899295018,
      "learning_rate": 2.57008293378697e-06,
      "loss": 0.0502,
      "step": 46
    },
    {
      "epoch": 1.3043478260869565,
      "eval_loss": 0.05139908567070961,
      "eval_runtime": 2.3812,
      "eval_samples_per_second": 251.974,
      "eval_steps_per_second": 31.497,
      "step": 46
    },
    {
      "epoch": 1.3333333333333333,
      "eval_loss": 0.05105065554380417,
      "eval_runtime": 2.4018,
      "eval_samples_per_second": 249.811,
      "eval_steps_per_second": 31.226,
      "step": 47
    },
    {
      "epoch": 1.3623188405797102,
      "grad_norm": 1.150174671283694,
      "learning_rate": 2.159676266344222e-06,
      "loss": 0.0482,
      "step": 48
    },
    {
      "epoch": 1.3623188405797102,
      "eval_loss": 0.050515007227659225,
      "eval_runtime": 2.3849,
      "eval_samples_per_second": 251.588,
      "eval_steps_per_second": 31.449,
      "step": 48
    },
    {
      "epoch": 1.391304347826087,
      "eval_loss": 0.05112989619374275,
      "eval_runtime": 2.3861,
      "eval_samples_per_second": 251.452,
      "eval_steps_per_second": 31.432,
      "step": 49
    },
    {
      "epoch": 1.4202898550724639,
      "grad_norm": 0.9809479332869758,
      "learning_rate": 1.7757889363191484e-06,
      "loss": 0.0472,
      "step": 50
    },
    {
      "epoch": 1.4202898550724639,
      "eval_loss": 0.050852496176958084,
      "eval_runtime": 2.3888,
      "eval_samples_per_second": 251.167,
      "eval_steps_per_second": 31.396,
      "step": 50
    },
    {
      "epoch": 1.4492753623188406,
      "eval_loss": 0.04979565367102623,
      "eval_runtime": 2.3955,
      "eval_samples_per_second": 250.469,
      "eval_steps_per_second": 31.309,
      "step": 51
    },
    {
      "epoch": 1.4782608695652173,
      "grad_norm": 0.7181974427552101,
      "learning_rate": 1.4220051962793952e-06,
      "loss": 0.0478,
      "step": 52
    },
    {
      "epoch": 1.4782608695652173,
      "eval_loss": 0.04979529604315758,
      "eval_runtime": 2.3784,
      "eval_samples_per_second": 252.271,
      "eval_steps_per_second": 31.534,
      "step": 52
    },
    {
      "epoch": 1.5072463768115942,
      "eval_loss": 0.050219178199768066,
      "eval_runtime": 2.3756,
      "eval_samples_per_second": 252.57,
      "eval_steps_per_second": 31.571,
      "step": 53
    },
    {
      "epoch": 1.5362318840579712,
      "grad_norm": 1.7774492629444423,
      "learning_rate": 1.1016282296838887e-06,
      "loss": 0.055,
      "step": 54
    },
    {
      "epoch": 1.5362318840579712,
      "eval_loss": 0.04986535757780075,
      "eval_runtime": 2.3832,
      "eval_samples_per_second": 251.765,
      "eval_steps_per_second": 31.471,
      "step": 54
    },
    {
      "epoch": 1.5652173913043477,
      "eval_loss": 0.04925783351063728,
      "eval_runtime": 2.3863,
      "eval_samples_per_second": 251.431,
      "eval_steps_per_second": 31.429,
      "step": 55
    },
    {
      "epoch": 1.5942028985507246,
      "grad_norm": 0.7119057580240911,
      "learning_rate": 8.176493099488664e-07,
      "loss": 0.0459,
      "step": 56
    },
    {
      "epoch": 1.5942028985507246,
      "eval_loss": 0.049321793019771576,
      "eval_runtime": 2.3816,
      "eval_samples_per_second": 251.931,
      "eval_steps_per_second": 31.491,
      "step": 56
    },
    {
      "epoch": 1.6231884057971016,
      "eval_loss": 0.049656953662633896,
      "eval_runtime": 2.3857,
      "eval_samples_per_second": 251.499,
      "eval_steps_per_second": 31.437,
      "step": 57
    },
    {
      "epoch": 1.6521739130434783,
      "grad_norm": 1.6066866518550498,
      "learning_rate": 5.727198717339511e-07,
      "loss": 0.0492,
      "step": 58
    },
    {
      "epoch": 1.6521739130434783,
      "eval_loss": 0.04972606897354126,
      "eval_runtime": 2.3792,
      "eval_samples_per_second": 252.181,
      "eval_steps_per_second": 31.523,
      "step": 58
    },
    {
      "epoch": 1.681159420289855,
      "eval_loss": 0.04940681904554367,
      "eval_runtime": 2.3921,
      "eval_samples_per_second": 250.828,
      "eval_steps_per_second": 31.354,
      "step": 59
    },
    {
      "epoch": 1.710144927536232,
      "grad_norm": 1.5684134897039972,
      "learning_rate": 3.691267552111183e-07,
      "loss": 0.0504,
      "step": 60
    },
    {
      "epoch": 1.710144927536232,
      "eval_loss": 0.04902585968375206,
      "eval_runtime": 2.3816,
      "eval_samples_per_second": 251.932,
      "eval_steps_per_second": 31.491,
      "step": 60
    },
    {
      "epoch": 1.7391304347826086,
      "eval_loss": 0.04878399893641472,
      "eval_runtime": 2.3923,
      "eval_samples_per_second": 250.809,
      "eval_steps_per_second": 31.351,
      "step": 61
    },
    {
      "epoch": 1.7681159420289854,
      "grad_norm": 3.0252941822207946,
      "learning_rate": 2.0877085445416889e-07,
      "loss": 0.0564,
      "step": 62
    },
    {
      "epoch": 1.7681159420289854,
      "eval_loss": 0.04881977662444115,
      "eval_runtime": 2.3726,
      "eval_samples_per_second": 252.892,
      "eval_steps_per_second": 31.611,
      "step": 62
    },
    {
      "epoch": 1.7971014492753623,
      "eval_loss": 0.04876958206295967,
      "eval_runtime": 2.383,
      "eval_samples_per_second": 251.783,
      "eval_steps_per_second": 31.473,
      "step": 63
    },
    {
      "epoch": 1.8260869565217392,
      "grad_norm": 0.5275348821410851,
      "learning_rate": 9.314936930293283e-08,
      "loss": 0.0503,
      "step": 64
    },
    {
      "epoch": 1.8260869565217392,
      "eval_loss": 0.048753608018159866,
      "eval_runtime": 2.3935,
      "eval_samples_per_second": 250.679,
      "eval_steps_per_second": 31.335,
      "step": 64
    },
    {
      "epoch": 1.855072463768116,
      "eval_loss": 0.04874425381422043,
      "eval_runtime": 2.3788,
      "eval_samples_per_second": 252.224,
      "eval_steps_per_second": 31.528,
      "step": 65
    },
    {
      "epoch": 1.8840579710144927,
      "grad_norm": 0.7635061728851181,
      "learning_rate": 2.3341826411756863e-08,
      "loss": 0.0495,
      "step": 66
    },
    {
      "epoch": 1.8840579710144927,
      "eval_loss": 0.048726994544267654,
      "eval_runtime": 2.3833,
      "eval_samples_per_second": 251.751,
      "eval_steps_per_second": 31.469,
      "step": 66
    },
    {
      "epoch": 1.9130434782608696,
      "eval_loss": 0.048674505203962326,
      "eval_runtime": 2.3749,
      "eval_samples_per_second": 252.645,
      "eval_steps_per_second": 31.581,
      "step": 67
    },
    {
      "epoch": 1.9420289855072463,
      "grad_norm": 0.4446688344675618,
      "learning_rate": 0.0,
      "loss": 0.0446,
      "step": 68
    },
    {
      "epoch": 1.9420289855072463,
      "eval_loss": 0.048731766641139984,
      "eval_runtime": 2.3934,
      "eval_samples_per_second": 250.692,
      "eval_steps_per_second": 31.336,
      "step": 68
    }
  ],
  "logging_steps": 2,
  "max_steps": 68,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 35518238687232.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}