{
  "best_metric": 0.3313491940498352,
  "best_model_checkpoint": "data/Llama-31-8B_task-3_180-samples_config-1_auto/checkpoint-85",
  "epoch": 12.0,
  "eval_steps": 500,
  "global_step": 204,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.058823529411764705,
      "grad_norm": 2.790146827697754,
      "learning_rate": 1.1764705882352942e-06,
      "loss": 2.4521,
      "step": 1
    },
    {
      "epoch": 0.11764705882352941,
      "grad_norm": 1.3707395792007446,
      "learning_rate": 2.3529411764705885e-06,
      "loss": 1.7077,
      "step": 2
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 2.8815231323242188,
      "learning_rate": 4.705882352941177e-06,
      "loss": 2.6331,
      "step": 4
    },
    {
      "epoch": 0.35294117647058826,
      "grad_norm": 1.3376513719558716,
      "learning_rate": 7.058823529411765e-06,
      "loss": 2.3661,
      "step": 6
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 2.828890323638916,
      "learning_rate": 9.411764705882354e-06,
      "loss": 2.1508,
      "step": 8
    },
    {
      "epoch": 0.5882352941176471,
      "grad_norm": 2.8286173343658447,
      "learning_rate": 1.1764705882352942e-05,
      "loss": 2.2319,
      "step": 10
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 2.2810916900634766,
      "learning_rate": 1.411764705882353e-05,
      "loss": 1.808,
      "step": 12
    },
    {
      "epoch": 0.8235294117647058,
      "grad_norm": 3.351749897003174,
      "learning_rate": 1.647058823529412e-05,
      "loss": 2.1354,
      "step": 14
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 1.3132500648498535,
      "learning_rate": 1.8823529411764708e-05,
      "loss": 1.7081,
      "step": 16
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.5541911125183105,
      "eval_runtime": 32.0115,
      "eval_samples_per_second": 1.125,
      "eval_steps_per_second": 1.125,
      "step": 17
    },
    {
      "epoch": 1.0588235294117647,
      "grad_norm": 0.9471787810325623,
      "learning_rate": 2.1176470588235296e-05,
      "loss": 1.4844,
      "step": 18
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 2.1010115146636963,
      "learning_rate": 2.3529411764705884e-05,
      "loss": 1.4457,
      "step": 20
    },
    {
      "epoch": 1.2941176470588236,
      "grad_norm": 1.4327553510665894,
      "learning_rate": 2.5882352941176475e-05,
      "loss": 1.1948,
      "step": 22
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 1.9643908739089966,
      "learning_rate": 2.823529411764706e-05,
      "loss": 0.9873,
      "step": 24
    },
    {
      "epoch": 1.5294117647058822,
      "grad_norm": 1.6198688745498657,
      "learning_rate": 3.058823529411765e-05,
      "loss": 0.7604,
      "step": 26
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 1.122072458267212,
      "learning_rate": 3.294117647058824e-05,
      "loss": 0.6013,
      "step": 28
    },
    {
      "epoch": 1.7647058823529411,
      "grad_norm": 1.1838219165802002,
      "learning_rate": 3.529411764705883e-05,
      "loss": 0.7327,
      "step": 30
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 1.0292763710021973,
      "learning_rate": 3.7647058823529415e-05,
      "loss": 0.4018,
      "step": 32
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.8941810131072998,
      "learning_rate": 4e-05,
      "loss": 0.509,
      "step": 34
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.47733280062675476,
      "eval_runtime": 32.0078,
      "eval_samples_per_second": 1.125,
      "eval_steps_per_second": 1.125,
      "step": 34
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 0.6447968482971191,
      "learning_rate": 4.235294117647059e-05,
      "loss": 0.4526,
      "step": 36
    },
    {
      "epoch": 2.235294117647059,
      "grad_norm": 0.7567365169525146,
      "learning_rate": 4.470588235294118e-05,
      "loss": 0.4536,
      "step": 38
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 0.8407445549964905,
      "learning_rate": 4.705882352941177e-05,
      "loss": 0.4453,
      "step": 40
    },
    {
      "epoch": 2.4705882352941178,
      "grad_norm": 0.3435295522212982,
      "learning_rate": 4.9411764705882355e-05,
      "loss": 0.325,
      "step": 42
    },
    {
      "epoch": 2.588235294117647,
      "grad_norm": 0.3069871664047241,
      "learning_rate": 5.176470588235295e-05,
      "loss": 0.3309,
      "step": 44
    },
    {
      "epoch": 2.7058823529411766,
      "grad_norm": 0.37769752740859985,
      "learning_rate": 5.411764705882353e-05,
      "loss": 0.3642,
      "step": 46
    },
    {
      "epoch": 2.8235294117647056,
      "grad_norm": 0.37647154927253723,
      "learning_rate": 5.647058823529412e-05,
      "loss": 0.3983,
      "step": 48
    },
    {
      "epoch": 2.9411764705882355,
      "grad_norm": 0.4306583106517792,
      "learning_rate": 5.882352941176471e-05,
      "loss": 0.3144,
      "step": 50
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.3655799627304077,
      "eval_runtime": 32.007,
      "eval_samples_per_second": 1.125,
      "eval_steps_per_second": 1.125,
      "step": 51
    },
    {
      "epoch": 3.0588235294117645,
      "grad_norm": 0.3212827742099762,
      "learning_rate": 6.11764705882353e-05,
      "loss": 0.2755,
      "step": 52
    },
    {
      "epoch": 3.176470588235294,
      "grad_norm": 1.307051658630371,
      "learning_rate": 6.352941176470588e-05,
      "loss": 0.448,
      "step": 54
    },
    {
      "epoch": 3.2941176470588234,
      "grad_norm": 0.37125203013420105,
      "learning_rate": 6.588235294117648e-05,
      "loss": 0.312,
      "step": 56
    },
    {
      "epoch": 3.411764705882353,
      "grad_norm": 0.23212695121765137,
      "learning_rate": 6.823529411764707e-05,
      "loss": 0.4181,
      "step": 58
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 5.962123870849609,
      "learning_rate": 7.058823529411765e-05,
      "loss": 0.258,
      "step": 60
    },
    {
      "epoch": 3.6470588235294117,
      "grad_norm": 0.6083526611328125,
      "learning_rate": 7.294117647058823e-05,
      "loss": 0.2953,
      "step": 62
    },
    {
      "epoch": 3.764705882352941,
      "grad_norm": 0.3373955190181732,
      "learning_rate": 7.529411764705883e-05,
      "loss": 0.4026,
      "step": 64
    },
    {
      "epoch": 3.8823529411764706,
      "grad_norm": 0.45089852809906006,
      "learning_rate": 7.764705882352942e-05,
      "loss": 0.2389,
      "step": 66
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.4510911703109741,
      "learning_rate": 8e-05,
      "loss": 0.2899,
      "step": 68
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.34543025493621826,
      "eval_runtime": 32.0083,
      "eval_samples_per_second": 1.125,
      "eval_steps_per_second": 1.125,
      "step": 68
    },
    {
      "epoch": 4.117647058823529,
      "grad_norm": 0.642993688583374,
      "learning_rate": 8.23529411764706e-05,
      "loss": 0.265,
      "step": 70
    },
    {
      "epoch": 4.235294117647059,
      "grad_norm": 0.37745243310928345,
      "learning_rate": 8.470588235294118e-05,
      "loss": 0.2576,
      "step": 72
    },
    {
      "epoch": 4.352941176470588,
      "grad_norm": 0.4048821032047272,
      "learning_rate": 8.705882352941177e-05,
      "loss": 0.3338,
      "step": 74
    },
    {
      "epoch": 4.470588235294118,
      "grad_norm": 0.40954768657684326,
      "learning_rate": 8.941176470588236e-05,
      "loss": 0.3065,
      "step": 76
    },
    {
      "epoch": 4.588235294117647,
      "grad_norm": 0.30904123187065125,
      "learning_rate": 9.176470588235295e-05,
      "loss": 0.3121,
      "step": 78
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 1.0630900859832764,
      "learning_rate": 9.411764705882353e-05,
      "loss": 0.2534,
      "step": 80
    },
    {
      "epoch": 4.823529411764706,
      "grad_norm": 0.4571320712566376,
      "learning_rate": 9.647058823529412e-05,
      "loss": 0.3327,
      "step": 82
    },
    {
      "epoch": 4.9411764705882355,
      "grad_norm": 0.42991459369659424,
      "learning_rate": 9.882352941176471e-05,
      "loss": 0.2162,
      "step": 84
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.3313491940498352,
      "eval_runtime": 32.0012,
      "eval_samples_per_second": 1.125,
      "eval_steps_per_second": 1.125,
      "step": 85
    },
    {
      "epoch": 5.0588235294117645,
      "grad_norm": 0.4001903831958771,
      "learning_rate": 9.99995783847866e-05,
      "loss": 0.2581,
      "step": 86
    },
    {
      "epoch": 5.176470588235294,
      "grad_norm": 0.4010772705078125,
      "learning_rate": 9.999620550574153e-05,
      "loss": 0.2951,
      "step": 88
    },
    {
      "epoch": 5.294117647058823,
      "grad_norm": 1.2183829545974731,
      "learning_rate": 9.998945997517956e-05,
      "loss": 0.2299,
      "step": 90
    },
    {
      "epoch": 5.411764705882353,
      "grad_norm": 0.0022620977833867073,
      "learning_rate": 9.997934224814173e-05,
      "loss": 0.0719,
      "step": 92
    },
    {
      "epoch": 5.529411764705882,
      "grad_norm": 0.49284520745277405,
      "learning_rate": 9.996585300715116e-05,
      "loss": 0.2196,
      "step": 94
    },
    {
      "epoch": 5.647058823529412,
      "grad_norm": 0.47940441966056824,
      "learning_rate": 9.994899316216708e-05,
      "loss": 0.2111,
      "step": 96
    },
    {
      "epoch": 5.764705882352941,
      "grad_norm": 0.5573323965072632,
      "learning_rate": 9.992876385052345e-05,
      "loss": 0.2654,
      "step": 98
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 0.4234023094177246,
      "learning_rate": 9.990516643685222e-05,
      "loss": 0.2256,
      "step": 100
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.678525984287262,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.3465,
      "step": 102
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.33739909529685974,
      "eval_runtime": 32.0036,
      "eval_samples_per_second": 1.125,
      "eval_steps_per_second": 1.125,
      "step": 102
    },
    {
      "epoch": 6.117647058823529,
      "grad_norm": 0.4869462251663208,
      "learning_rate": 9.984787389787688e-05,
      "loss": 0.1337,
      "step": 104
    },
    {
      "epoch": 6.235294117647059,
      "grad_norm": 0.45321589708328247,
      "learning_rate": 9.981418263742148e-05,
      "loss": 0.2013,
      "step": 106
    },
    {
      "epoch": 6.352941176470588,
      "grad_norm": 0.3684188425540924,
      "learning_rate": 9.977713100437509e-05,
      "loss": 0.2076,
      "step": 108
    },
    {
      "epoch": 6.470588235294118,
      "grad_norm": 0.41488441824913025,
      "learning_rate": 9.973672149817232e-05,
      "loss": 0.1307,
      "step": 110
    },
    {
      "epoch": 6.588235294117647,
      "grad_norm": 0.5205825567245483,
      "learning_rate": 9.96929568447637e-05,
      "loss": 0.1358,
      "step": 112
    },
    {
      "epoch": 6.705882352941177,
      "grad_norm": 0.5574342608451843,
      "learning_rate": 9.964583999643174e-05,
      "loss": 0.1133,
      "step": 114
    },
    {
      "epoch": 6.823529411764706,
      "grad_norm": 0.8607386350631714,
      "learning_rate": 9.95953741315919e-05,
      "loss": 0.2207,
      "step": 116
    },
    {
      "epoch": 6.9411764705882355,
      "grad_norm": 0.7382214069366455,
      "learning_rate": 9.954156265457801e-05,
      "loss": 0.2498,
      "step": 118
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.35563692450523376,
      "eval_runtime": 31.9979,
      "eval_samples_per_second": 1.125,
      "eval_steps_per_second": 1.125,
      "step": 119
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 0.5253345966339111,
      "learning_rate": 9.948440919541278e-05,
      "loss": 0.2012,
      "step": 120
    },
    {
      "epoch": 7.176470588235294,
      "grad_norm": 0.616706371307373,
      "learning_rate": 9.942391760956277e-05,
      "loss": 0.1718,
      "step": 122
    },
    {
      "epoch": 7.294117647058823,
      "grad_norm": 0.5785579681396484,
      "learning_rate": 9.936009197767845e-05,
      "loss": 0.1278,
      "step": 124
    },
    {
      "epoch": 7.411764705882353,
      "grad_norm": 0.8303074240684509,
      "learning_rate": 9.929293660531888e-05,
      "loss": 0.1065,
      "step": 126
    },
    {
      "epoch": 7.529411764705882,
      "grad_norm": 0.8201299905776978,
      "learning_rate": 9.922245602266118e-05,
      "loss": 0.1027,
      "step": 128
    },
    {
      "epoch": 7.647058823529412,
      "grad_norm": 0.5928418040275574,
      "learning_rate": 9.91486549841951e-05,
      "loss": 0.091,
      "step": 130
    },
    {
      "epoch": 7.764705882352941,
      "grad_norm": 3.87727952003479,
      "learning_rate": 9.90715384684021e-05,
      "loss": 0.1228,
      "step": 132
    },
    {
      "epoch": 7.882352941176471,
      "grad_norm": 0.5786693692207336,
      "learning_rate": 9.899111167741966e-05,
      "loss": 0.1583,
      "step": 134
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.5984454154968262,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.1023,
      "step": 136
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.41042014956474304,
      "eval_runtime": 32.0113,
      "eval_samples_per_second": 1.125,
      "eval_steps_per_second": 1.125,
      "step": 136
    },
    {
      "epoch": 8.117647058823529,
      "grad_norm": 0.5814339518547058,
      "learning_rate": 9.882034919459555e-05,
      "loss": 0.0595,
      "step": 138
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 0.7130110263824463,
      "learning_rate": 9.873002502207503e-05,
      "loss": 0.0864,
      "step": 140
    },
    {
      "epoch": 8.352941176470589,
      "grad_norm": 0.7068061232566833,
      "learning_rate": 9.863641361223024e-05,
      "loss": 0.0951,
      "step": 142
    },
    {
      "epoch": 8.470588235294118,
      "grad_norm": 0.0006599414627999067,
      "learning_rate": 9.853952127991372e-05,
      "loss": 0.0686,
      "step": 144
    },
    {
      "epoch": 8.588235294117647,
      "grad_norm": 0.8574017882347107,
      "learning_rate": 9.843935456130295e-05,
      "loss": 0.0731,
      "step": 146
    },
    {
      "epoch": 8.705882352941176,
      "grad_norm": 0.38401010632514954,
      "learning_rate": 9.833592021345937e-05,
      "loss": 0.0227,
      "step": 148
    },
    {
      "epoch": 8.823529411764707,
      "grad_norm": 0.6860135793685913,
      "learning_rate": 9.822922521387276e-05,
      "loss": 0.0732,
      "step": 150
    },
    {
      "epoch": 8.941176470588236,
      "grad_norm": 1.105839490890503,
      "learning_rate": 9.811927675999036e-05,
      "loss": 0.0862,
      "step": 152
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.4675668179988861,
      "eval_runtime": 32.0127,
      "eval_samples_per_second": 1.125,
      "eval_steps_per_second": 1.125,
      "step": 153
    },
    {
      "epoch": 9.058823529411764,
      "grad_norm": 0.5514603853225708,
      "learning_rate": 9.800608226873142e-05,
      "loss": 0.0442,
      "step": 154
    },
    {
      "epoch": 9.176470588235293,
      "grad_norm": 0.3878594636917114,
      "learning_rate": 9.788964937598689e-05,
      "loss": 0.0347,
      "step": 156
    },
    {
      "epoch": 9.294117647058824,
      "grad_norm": 0.5091602802276611,
      "learning_rate": 9.776998593610428e-05,
      "loss": 0.0354,
      "step": 158
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 0.5891442894935608,
      "learning_rate": 9.764710002135784e-05,
      "loss": 0.0444,
      "step": 160
    },
    {
      "epoch": 9.529411764705882,
      "grad_norm": 0.6681481003761292,
      "learning_rate": 9.752099992140399e-05,
      "loss": 0.0281,
      "step": 162
    },
    {
      "epoch": 9.647058823529411,
      "grad_norm": 0.5979675650596619,
      "learning_rate": 9.739169414272217e-05,
      "loss": 0.0385,
      "step": 164
    },
    {
      "epoch": 9.764705882352942,
      "grad_norm": 0.8215705156326294,
      "learning_rate": 9.725919140804099e-05,
      "loss": 0.0464,
      "step": 166
    },
    {
      "epoch": 9.882352941176471,
      "grad_norm": 0.4677208364009857,
      "learning_rate": 9.71235006557497e-05,
      "loss": 0.0284,
      "step": 168
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.5602480173110962,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.0549,
      "step": 170
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.49783653020858765,
      "eval_runtime": 32.0118,
      "eval_samples_per_second": 1.125,
      "eval_steps_per_second": 1.125,
      "step": 170
    },
    {
      "epoch": 10.117647058823529,
      "grad_norm": 0.38908106088638306,
      "learning_rate": 9.684259192656553e-05,
      "loss": 0.0192,
      "step": 172
    },
    {
      "epoch": 10.235294117647058,
      "grad_norm": 0.6568523049354553,
      "learning_rate": 9.669739289925577e-05,
      "loss": 0.0176,
      "step": 174
    },
    {
      "epoch": 10.352941176470589,
      "grad_norm": 0.6142391562461853,
      "learning_rate": 9.654904375222385e-05,
      "loss": 0.0137,
      "step": 176
    },
    {
      "epoch": 10.470588235294118,
      "grad_norm": 0.6784031391143799,
      "learning_rate": 9.639755449282875e-05,
      "loss": 0.0305,
      "step": 178
    },
    {
      "epoch": 10.588235294117647,
      "grad_norm": 0.5321065783500671,
      "learning_rate": 9.62429353402556e-05,
      "loss": 0.0226,
      "step": 180
    },
    {
      "epoch": 10.705882352941176,
      "grad_norm": 0.5628299117088318,
      "learning_rate": 9.608519672482636e-05,
      "loss": 0.0551,
      "step": 182
    },
    {
      "epoch": 10.823529411764707,
      "grad_norm": 0.4615156054496765,
      "learning_rate": 9.592434928729616e-05,
      "loss": 0.0304,
      "step": 184
    },
    {
      "epoch": 10.941176470588236,
      "grad_norm": 0.7079306244850159,
      "learning_rate": 9.576040387813552e-05,
      "loss": 0.032,
      "step": 186
    },
    {
      "epoch": 11.0,
      "eval_loss": 0.5855510234832764,
      "eval_runtime": 32.0018,
      "eval_samples_per_second": 1.125,
      "eval_steps_per_second": 1.125,
      "step": 187
    },
    {
      "epoch": 11.058823529411764,
      "grad_norm": 0.2959321439266205,
      "learning_rate": 9.559337155679842e-05,
      "loss": 0.0096,
      "step": 188
    },
    {
      "epoch": 11.176470588235293,
      "grad_norm": 0.6475893259048462,
      "learning_rate": 9.542326359097619e-05,
      "loss": 0.0203,
      "step": 190
    },
    {
      "epoch": 11.294117647058824,
      "grad_norm": 1.12827730178833,
      "learning_rate": 9.525009145583745e-05,
      "loss": 0.0221,
      "step": 192
    },
    {
      "epoch": 11.411764705882353,
      "grad_norm": 0.5682121515274048,
      "learning_rate": 9.507386683325404e-05,
      "loss": 0.0179,
      "step": 194
    },
    {
      "epoch": 11.529411764705882,
      "grad_norm": 0.4497404396533966,
      "learning_rate": 9.489460161101291e-05,
      "loss": 0.0236,
      "step": 196
    },
    {
      "epoch": 11.647058823529411,
      "grad_norm": 0.08139002323150635,
      "learning_rate": 9.471230788201429e-05,
      "loss": 0.009,
      "step": 198
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 0.5133498311042786,
      "learning_rate": 9.452699794345581e-05,
      "loss": 0.0204,
      "step": 200
    },
    {
      "epoch": 11.882352941176471,
      "grad_norm": 0.38405758142471313,
      "learning_rate": 9.43386842960031e-05,
      "loss": 0.011,
      "step": 202
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.24333085119724274,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.0072,
      "step": 204
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.5622949600219727,
      "eval_runtime": 32.0008,
      "eval_samples_per_second": 1.125,
      "eval_steps_per_second": 1.125,
      "step": 204
    },
    {
      "epoch": 12.0,
      "step": 204,
      "total_flos": 1.2893298226941133e+17,
      "train_loss": 0.38738249295719845,
      "train_runtime": 4719.6721,
      "train_samples_per_second": 1.441,
      "train_steps_per_second": 0.18
    }
  ],
  "logging_steps": 2,
  "max_steps": 850,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 25,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 7,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.2893298226941133e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}