{
  "best_metric": 0.6522939801216125,
  "best_model_checkpoint": "data/Gemma-2-2B_task-2_180-samples_config-1/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": 0.8471795320510864,
      "learning_rate": 1.1764705882352942e-06,
      "loss": 1.0521,
      "step": 1
    },
    {
      "epoch": 0.11764705882352941,
      "grad_norm": 0.8471609950065613,
      "learning_rate": 2.3529411764705885e-06,
      "loss": 1.1036,
      "step": 2
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 1.1077567338943481,
      "learning_rate": 4.705882352941177e-06,
      "loss": 1.1169,
      "step": 4
    },
    {
      "epoch": 0.35294117647058826,
      "grad_norm": 1.3564423322677612,
      "learning_rate": 7.058823529411765e-06,
      "loss": 1.1274,
      "step": 6
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 0.8793351650238037,
      "learning_rate": 9.411764705882354e-06,
      "loss": 1.0579,
      "step": 8
    },
    {
      "epoch": 0.5882352941176471,
      "grad_norm": 0.8737766146659851,
      "learning_rate": 1.1764705882352942e-05,
      "loss": 1.0669,
      "step": 10
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 0.7959626317024231,
      "learning_rate": 1.411764705882353e-05,
      "loss": 0.9733,
      "step": 12
    },
    {
      "epoch": 0.8235294117647058,
      "grad_norm": 0.794722855091095,
      "learning_rate": 1.647058823529412e-05,
      "loss": 1.0265,
      "step": 14
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 0.8228878974914551,
      "learning_rate": 1.8823529411764708e-05,
      "loss": 0.9382,
      "step": 16
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.9762291312217712,
      "eval_runtime": 13.9309,
      "eval_samples_per_second": 2.584,
      "eval_steps_per_second": 2.584,
      "step": 17
    },
    {
      "epoch": 1.0588235294117647,
      "grad_norm": 0.748081386089325,
      "learning_rate": 2.1176470588235296e-05,
      "loss": 1.1948,
      "step": 18
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 0.6090679168701172,
      "learning_rate": 2.3529411764705884e-05,
      "loss": 0.9314,
      "step": 20
    },
    {
      "epoch": 1.2941176470588236,
      "grad_norm": 0.5461698174476624,
      "learning_rate": 2.5882352941176475e-05,
      "loss": 0.9145,
      "step": 22
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 0.5632569193840027,
      "learning_rate": 2.823529411764706e-05,
      "loss": 0.8891,
      "step": 24
    },
    {
      "epoch": 1.5294117647058822,
      "grad_norm": 0.5248648524284363,
      "learning_rate": 3.058823529411765e-05,
      "loss": 0.7771,
      "step": 26
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 0.4509378671646118,
      "learning_rate": 3.294117647058824e-05,
      "loss": 0.9258,
      "step": 28
    },
    {
      "epoch": 1.7647058823529411,
      "grad_norm": 0.6190409660339355,
      "learning_rate": 3.529411764705883e-05,
      "loss": 0.8323,
      "step": 30
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 0.4419402480125427,
      "learning_rate": 3.7647058823529415e-05,
      "loss": 0.7181,
      "step": 32
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.5190498232841492,
      "learning_rate": 4e-05,
      "loss": 0.7998,
      "step": 34
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.7799161076545715,
      "eval_runtime": 13.9288,
      "eval_samples_per_second": 2.585,
      "eval_steps_per_second": 2.585,
      "step": 34
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 0.4938206076622009,
      "learning_rate": 4.235294117647059e-05,
      "loss": 0.7448,
      "step": 36
    },
    {
      "epoch": 2.235294117647059,
      "grad_norm": 0.39435631036758423,
      "learning_rate": 4.470588235294118e-05,
      "loss": 0.7043,
      "step": 38
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 0.4283750653266907,
      "learning_rate": 4.705882352941177e-05,
      "loss": 0.6884,
      "step": 40
    },
    {
      "epoch": 2.4705882352941178,
      "grad_norm": 0.4009155333042145,
      "learning_rate": 4.9411764705882355e-05,
      "loss": 0.8442,
      "step": 42
    },
    {
      "epoch": 2.588235294117647,
      "grad_norm": 0.5209583640098572,
      "learning_rate": 5.176470588235295e-05,
      "loss": 0.7254,
      "step": 44
    },
    {
      "epoch": 2.7058823529411766,
      "grad_norm": 0.5622742176055908,
      "learning_rate": 5.411764705882353e-05,
      "loss": 0.7274,
      "step": 46
    },
    {
      "epoch": 2.8235294117647056,
      "grad_norm": 0.5544571280479431,
      "learning_rate": 5.647058823529412e-05,
      "loss": 0.7078,
      "step": 48
    },
    {
      "epoch": 2.9411764705882355,
      "grad_norm": 0.41800931096076965,
      "learning_rate": 5.882352941176471e-05,
      "loss": 0.6681,
      "step": 50
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.7045078873634338,
      "eval_runtime": 13.9411,
      "eval_samples_per_second": 2.582,
      "eval_steps_per_second": 2.582,
      "step": 51
    },
    {
      "epoch": 3.0588235294117645,
      "grad_norm": 0.3819270133972168,
      "learning_rate": 6.11764705882353e-05,
      "loss": 0.6347,
      "step": 52
    },
    {
      "epoch": 3.176470588235294,
      "grad_norm": 0.4059789478778839,
      "learning_rate": 6.352941176470588e-05,
      "loss": 0.6442,
      "step": 54
    },
    {
      "epoch": 3.2941176470588234,
      "grad_norm": 0.36979779601097107,
      "learning_rate": 6.588235294117648e-05,
      "loss": 0.6311,
      "step": 56
    },
    {
      "epoch": 3.411764705882353,
      "grad_norm": 0.5177358984947205,
      "learning_rate": 6.823529411764707e-05,
      "loss": 0.8655,
      "step": 58
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 0.5120185017585754,
      "learning_rate": 7.058823529411765e-05,
      "loss": 0.5978,
      "step": 60
    },
    {
      "epoch": 3.6470588235294117,
      "grad_norm": 0.4440748989582062,
      "learning_rate": 7.294117647058823e-05,
      "loss": 0.6496,
      "step": 62
    },
    {
      "epoch": 3.764705882352941,
      "grad_norm": 0.5182046890258789,
      "learning_rate": 7.529411764705883e-05,
      "loss": 0.6069,
      "step": 64
    },
    {
      "epoch": 3.8823529411764706,
      "grad_norm": 0.44228434562683105,
      "learning_rate": 7.764705882352942e-05,
      "loss": 0.622,
      "step": 66
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.43491390347480774,
      "learning_rate": 8e-05,
      "loss": 0.539,
      "step": 68
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.6650347709655762,
      "eval_runtime": 13.9399,
      "eval_samples_per_second": 2.583,
      "eval_steps_per_second": 2.583,
      "step": 68
    },
    {
      "epoch": 4.117647058823529,
      "grad_norm": 0.8099572658538818,
      "learning_rate": 8.23529411764706e-05,
      "loss": 0.7707,
      "step": 70
    },
    {
      "epoch": 4.235294117647059,
      "grad_norm": 0.4558533728122711,
      "learning_rate": 8.470588235294118e-05,
      "loss": 0.5399,
      "step": 72
    },
    {
      "epoch": 4.352941176470588,
      "grad_norm": 0.4668755829334259,
      "learning_rate": 8.705882352941177e-05,
      "loss": 0.5596,
      "step": 74
    },
    {
      "epoch": 4.470588235294118,
      "grad_norm": 0.48760464787483215,
      "learning_rate": 8.941176470588236e-05,
      "loss": 0.5395,
      "step": 76
    },
    {
      "epoch": 4.588235294117647,
      "grad_norm": 0.5810846090316772,
      "learning_rate": 9.176470588235295e-05,
      "loss": 0.5389,
      "step": 78
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 0.5398671627044678,
      "learning_rate": 9.411764705882353e-05,
      "loss": 0.5122,
      "step": 80
    },
    {
      "epoch": 4.823529411764706,
      "grad_norm": 0.70992511510849,
      "learning_rate": 9.647058823529412e-05,
      "loss": 0.5351,
      "step": 82
    },
    {
      "epoch": 4.9411764705882355,
      "grad_norm": 0.6651508808135986,
      "learning_rate": 9.882352941176471e-05,
      "loss": 0.532,
      "step": 84
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.6522939801216125,
      "eval_runtime": 13.9323,
      "eval_samples_per_second": 2.584,
      "eval_steps_per_second": 2.584,
      "step": 85
    },
    {
      "epoch": 5.0588235294117645,
      "grad_norm": 0.6531202793121338,
      "learning_rate": 9.99995783847866e-05,
      "loss": 0.517,
      "step": 86
    },
    {
      "epoch": 5.176470588235294,
      "grad_norm": 0.6586031913757324,
      "learning_rate": 9.999620550574153e-05,
      "loss": 0.4782,
      "step": 88
    },
    {
      "epoch": 5.294117647058823,
      "grad_norm": 0.7373224496841431,
      "learning_rate": 9.998945997517956e-05,
      "loss": 0.5178,
      "step": 90
    },
    {
      "epoch": 5.411764705882353,
      "grad_norm": 0.6238186359405518,
      "learning_rate": 9.997934224814173e-05,
      "loss": 0.5067,
      "step": 92
    },
    {
      "epoch": 5.529411764705882,
      "grad_norm": 0.6471043229103088,
      "learning_rate": 9.996585300715116e-05,
      "loss": 0.4334,
      "step": 94
    },
    {
      "epoch": 5.647058823529412,
      "grad_norm": 0.6911196112632751,
      "learning_rate": 9.994899316216708e-05,
      "loss": 0.435,
      "step": 96
    },
    {
      "epoch": 5.764705882352941,
      "grad_norm": 0.7732041478157043,
      "learning_rate": 9.992876385052345e-05,
      "loss": 0.4769,
      "step": 98
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 0.9492689967155457,
      "learning_rate": 9.990516643685222e-05,
      "loss": 0.5408,
      "step": 100
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.7098620533943176,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.4525,
      "step": 102
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.6651446223258972,
      "eval_runtime": 13.9461,
      "eval_samples_per_second": 2.581,
      "eval_steps_per_second": 2.581,
      "step": 102
    },
    {
      "epoch": 6.117647058823529,
      "grad_norm": 0.6771420836448669,
      "learning_rate": 9.984787389787688e-05,
      "loss": 0.3573,
      "step": 104
    },
    {
      "epoch": 6.235294117647059,
      "grad_norm": 1.0061311721801758,
      "learning_rate": 9.981418263742148e-05,
      "loss": 0.4105,
      "step": 106
    },
    {
      "epoch": 6.352941176470588,
      "grad_norm": 1.0207984447479248,
      "learning_rate": 9.977713100437509e-05,
      "loss": 0.336,
      "step": 108
    },
    {
      "epoch": 6.470588235294118,
      "grad_norm": 0.95076584815979,
      "learning_rate": 9.973672149817232e-05,
      "loss": 0.3713,
      "step": 110
    },
    {
      "epoch": 6.588235294117647,
      "grad_norm": 0.9174726605415344,
      "learning_rate": 9.96929568447637e-05,
      "loss": 0.361,
      "step": 112
    },
    {
      "epoch": 6.705882352941177,
      "grad_norm": 0.9004970788955688,
      "learning_rate": 9.964583999643174e-05,
      "loss": 0.4135,
      "step": 114
    },
    {
      "epoch": 6.823529411764706,
      "grad_norm": 0.9365142583847046,
      "learning_rate": 9.95953741315919e-05,
      "loss": 0.381,
      "step": 116
    },
    {
      "epoch": 6.9411764705882355,
      "grad_norm": 0.8976054191589355,
      "learning_rate": 9.954156265457801e-05,
      "loss": 0.372,
      "step": 118
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.7216023802757263,
      "eval_runtime": 13.9301,
      "eval_samples_per_second": 2.584,
      "eval_steps_per_second": 2.584,
      "step": 119
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 1.4741501808166504,
      "learning_rate": 9.948440919541278e-05,
      "loss": 0.3764,
      "step": 120
    },
    {
      "epoch": 7.176470588235294,
      "grad_norm": 0.9651046395301819,
      "learning_rate": 9.942391760956277e-05,
      "loss": 0.2564,
      "step": 122
    },
    {
      "epoch": 7.294117647058823,
      "grad_norm": 1.0543420314788818,
      "learning_rate": 9.936009197767845e-05,
      "loss": 0.2659,
      "step": 124
    },
    {
      "epoch": 7.411764705882353,
      "grad_norm": 1.4210118055343628,
      "learning_rate": 9.929293660531888e-05,
      "loss": 0.231,
      "step": 126
    },
    {
      "epoch": 7.529411764705882,
      "grad_norm": 1.2298481464385986,
      "learning_rate": 9.922245602266118e-05,
      "loss": 0.2299,
      "step": 128
    },
    {
      "epoch": 7.647058823529412,
      "grad_norm": 1.2452986240386963,
      "learning_rate": 9.91486549841951e-05,
      "loss": 0.2399,
      "step": 130
    },
    {
      "epoch": 7.764705882352941,
      "grad_norm": 1.1392860412597656,
      "learning_rate": 9.90715384684021e-05,
      "loss": 0.2242,
      "step": 132
    },
    {
      "epoch": 7.882352941176471,
      "grad_norm": 1.2908728122711182,
      "learning_rate": 9.899111167741966e-05,
      "loss": 0.2832,
      "step": 134
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.4094887971878052,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.2505,
      "step": 136
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.8553277254104614,
      "eval_runtime": 13.9361,
      "eval_samples_per_second": 2.583,
      "eval_steps_per_second": 2.583,
      "step": 136
    },
    {
      "epoch": 8.117647058823529,
      "grad_norm": 1.0582916736602783,
      "learning_rate": 9.882034919459555e-05,
      "loss": 0.1599,
      "step": 138
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 1.3344570398330688,
      "learning_rate": 9.873002502207503e-05,
      "loss": 0.1608,
      "step": 140
    },
    {
      "epoch": 8.352941176470589,
      "grad_norm": 1.665604829788208,
      "learning_rate": 9.863641361223024e-05,
      "loss": 0.1675,
      "step": 142
    },
    {
      "epoch": 8.470588235294118,
      "grad_norm": 1.5989713668823242,
      "learning_rate": 9.853952127991372e-05,
      "loss": 0.1836,
      "step": 144
    },
    {
      "epoch": 8.588235294117647,
      "grad_norm": 1.955446720123291,
      "learning_rate": 9.843935456130295e-05,
      "loss": 0.2194,
      "step": 146
    },
    {
      "epoch": 8.705882352941176,
      "grad_norm": 1.3413044214248657,
      "learning_rate": 9.833592021345937e-05,
      "loss": 0.1342,
      "step": 148
    },
    {
      "epoch": 8.823529411764707,
      "grad_norm": 1.3233766555786133,
      "learning_rate": 9.822922521387276e-05,
      "loss": 0.1287,
      "step": 150
    },
    {
      "epoch": 8.941176470588236,
      "grad_norm": 1.5602322816848755,
      "learning_rate": 9.811927675999036e-05,
      "loss": 0.1207,
      "step": 152
    },
    {
      "epoch": 9.0,
      "eval_loss": 1.0733627080917358,
      "eval_runtime": 13.9558,
      "eval_samples_per_second": 2.58,
      "eval_steps_per_second": 2.58,
      "step": 153
    },
    {
      "epoch": 9.058823529411764,
      "grad_norm": 1.2735803127288818,
      "learning_rate": 9.800608226873142e-05,
      "loss": 0.1248,
      "step": 154
    },
    {
      "epoch": 9.176470588235293,
      "grad_norm": 1.2522315979003906,
      "learning_rate": 9.788964937598689e-05,
      "loss": 0.1028,
      "step": 156
    },
    {
      "epoch": 9.294117647058824,
      "grad_norm": 1.4622876644134521,
      "learning_rate": 9.776998593610428e-05,
      "loss": 0.0673,
      "step": 158
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 1.8606466054916382,
      "learning_rate": 9.764710002135784e-05,
      "loss": 0.1134,
      "step": 160
    },
    {
      "epoch": 9.529411764705882,
      "grad_norm": 1.7573667764663696,
      "learning_rate": 9.752099992140399e-05,
      "loss": 0.0848,
      "step": 162
    },
    {
      "epoch": 9.647058823529411,
      "grad_norm": 1.3051804304122925,
      "learning_rate": 9.739169414272217e-05,
      "loss": 0.0811,
      "step": 164
    },
    {
      "epoch": 9.764705882352942,
      "grad_norm": 1.233198642730713,
      "learning_rate": 9.725919140804099e-05,
      "loss": 0.092,
      "step": 166
    },
    {
      "epoch": 9.882352941176471,
      "grad_norm": 1.4986467361450195,
      "learning_rate": 9.71235006557497e-05,
      "loss": 0.0783,
      "step": 168
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.5419331789016724,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.085,
      "step": 170
    },
    {
      "epoch": 10.0,
      "eval_loss": 1.1967495679855347,
      "eval_runtime": 13.9383,
      "eval_samples_per_second": 2.583,
      "eval_steps_per_second": 2.583,
      "step": 170
    },
    {
      "epoch": 10.117647058823529,
      "grad_norm": 0.8988786339759827,
      "learning_rate": 9.684259192656553e-05,
      "loss": 0.0368,
      "step": 172
    },
    {
      "epoch": 10.235294117647058,
      "grad_norm": 1.1210275888442993,
      "learning_rate": 9.669739289925577e-05,
      "loss": 0.0411,
      "step": 174
    },
    {
      "epoch": 10.352941176470589,
      "grad_norm": 1.1623586416244507,
      "learning_rate": 9.654904375222385e-05,
      "loss": 0.043,
      "step": 176
    },
    {
      "epoch": 10.470588235294118,
      "grad_norm": 1.0533829927444458,
      "learning_rate": 9.639755449282875e-05,
      "loss": 0.0472,
      "step": 178
    },
    {
      "epoch": 10.588235294117647,
      "grad_norm": 1.7356975078582764,
      "learning_rate": 9.62429353402556e-05,
      "loss": 0.0923,
      "step": 180
    },
    {
      "epoch": 10.705882352941176,
      "grad_norm": 1.7056307792663574,
      "learning_rate": 9.608519672482636e-05,
      "loss": 0.0463,
      "step": 182
    },
    {
      "epoch": 10.823529411764707,
      "grad_norm": 1.4538213014602661,
      "learning_rate": 9.592434928729616e-05,
      "loss": 0.0604,
      "step": 184
    },
    {
      "epoch": 10.941176470588236,
      "grad_norm": 0.9947254657745361,
      "learning_rate": 9.576040387813552e-05,
      "loss": 0.0601,
      "step": 186
    },
    {
      "epoch": 11.0,
      "eval_loss": 1.312574863433838,
      "eval_runtime": 13.9333,
      "eval_samples_per_second": 2.584,
      "eval_steps_per_second": 2.584,
      "step": 187
    },
    {
      "epoch": 11.058823529411764,
      "grad_norm": 0.938515841960907,
      "learning_rate": 9.559337155679842e-05,
      "loss": 0.0436,
      "step": 188
    },
    {
      "epoch": 11.176470588235293,
      "grad_norm": 1.0058748722076416,
      "learning_rate": 9.542326359097619e-05,
      "loss": 0.0307,
      "step": 190
    },
    {
      "epoch": 11.294117647058824,
      "grad_norm": 1.376985788345337,
      "learning_rate": 9.525009145583745e-05,
      "loss": 0.0357,
      "step": 192
    },
    {
      "epoch": 11.411764705882353,
      "grad_norm": 0.9462356567382812,
      "learning_rate": 9.507386683325404e-05,
      "loss": 0.0178,
      "step": 194
    },
    {
      "epoch": 11.529411764705882,
      "grad_norm": 1.5170210599899292,
      "learning_rate": 9.489460161101291e-05,
      "loss": 0.0342,
      "step": 196
    },
    {
      "epoch": 11.647058823529411,
      "grad_norm": 1.0396932363510132,
      "learning_rate": 9.471230788201429e-05,
      "loss": 0.0295,
      "step": 198
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 1.3947901725769043,
      "learning_rate": 9.452699794345581e-05,
      "loss": 0.0501,
      "step": 200
    },
    {
      "epoch": 11.882352941176471,
      "grad_norm": 1.241542100906372,
      "learning_rate": 9.43386842960031e-05,
      "loss": 0.0337,
      "step": 202
    },
    {
      "epoch": 12.0,
      "grad_norm": 1.05393385887146,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.0263,
      "step": 204
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.3857388496398926,
      "eval_runtime": 13.9348,
      "eval_samples_per_second": 2.583,
      "eval_steps_per_second": 2.583,
      "step": 204
    },
    {
      "epoch": 12.0,
      "step": 204,
      "total_flos": 5.346645589465498e+16,
      "train_loss": 0.44165483409283207,
      "train_runtime": 2279.3203,
      "train_samples_per_second": 2.983,
      "train_steps_per_second": 0.373
    }
  ],
  "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": 5.346645589465498e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}