{
  "best_metric": 0.6152952313423157,
  "best_model_checkpoint": "data/Gemma-2-2B_task-2_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": 0.7513045072555542,
      "learning_rate": 1.1764705882352942e-06,
      "loss": 0.96,
      "step": 1
    },
    {
      "epoch": 0.11764705882352941,
      "grad_norm": 0.7305058240890503,
      "learning_rate": 2.3529411764705885e-06,
      "loss": 1.0322,
      "step": 2
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 0.8322392702102661,
      "learning_rate": 4.705882352941177e-06,
      "loss": 1.0413,
      "step": 4
    },
    {
      "epoch": 0.35294117647058826,
      "grad_norm": 0.8096673488616943,
      "learning_rate": 7.058823529411765e-06,
      "loss": 1.0079,
      "step": 6
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 0.7323368787765503,
      "learning_rate": 9.411764705882354e-06,
      "loss": 1.01,
      "step": 8
    },
    {
      "epoch": 0.5882352941176471,
      "grad_norm": 0.7627260684967041,
      "learning_rate": 1.1764705882352942e-05,
      "loss": 1.0219,
      "step": 10
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 0.7248629331588745,
      "learning_rate": 1.411764705882353e-05,
      "loss": 0.9561,
      "step": 12
    },
    {
      "epoch": 0.8235294117647058,
      "grad_norm": 0.673678994178772,
      "learning_rate": 1.647058823529412e-05,
      "loss": 0.9349,
      "step": 14
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 0.7298202514648438,
      "learning_rate": 1.8823529411764708e-05,
      "loss": 0.962,
      "step": 16
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.9374876022338867,
      "eval_runtime": 14.325,
      "eval_samples_per_second": 2.513,
      "eval_steps_per_second": 2.513,
      "step": 17
    },
    {
      "epoch": 1.0588235294117647,
      "grad_norm": 0.6586418747901917,
      "learning_rate": 2.1176470588235296e-05,
      "loss": 1.1488,
      "step": 18
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 0.561080276966095,
      "learning_rate": 2.3529411764705884e-05,
      "loss": 0.8605,
      "step": 20
    },
    {
      "epoch": 1.2941176470588236,
      "grad_norm": 0.6036514639854431,
      "learning_rate": 2.5882352941176475e-05,
      "loss": 0.9143,
      "step": 22
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 0.5451833605766296,
      "learning_rate": 2.823529411764706e-05,
      "loss": 0.8186,
      "step": 24
    },
    {
      "epoch": 1.5294117647058822,
      "grad_norm": 0.5691556334495544,
      "learning_rate": 3.058823529411765e-05,
      "loss": 0.8046,
      "step": 26
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 0.49707120656967163,
      "learning_rate": 3.294117647058824e-05,
      "loss": 0.939,
      "step": 28
    },
    {
      "epoch": 1.7647058823529411,
      "grad_norm": 0.5690445899963379,
      "learning_rate": 3.529411764705883e-05,
      "loss": 0.7478,
      "step": 30
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 0.6026608347892761,
      "learning_rate": 3.7647058823529415e-05,
      "loss": 0.7134,
      "step": 32
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.5153243541717529,
      "learning_rate": 4e-05,
      "loss": 0.7695,
      "step": 34
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.7491788864135742,
      "eval_runtime": 14.3189,
      "eval_samples_per_second": 2.514,
      "eval_steps_per_second": 2.514,
      "step": 34
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 0.54787677526474,
      "learning_rate": 4.235294117647059e-05,
      "loss": 0.711,
      "step": 36
    },
    {
      "epoch": 2.235294117647059,
      "grad_norm": 0.450118750333786,
      "learning_rate": 4.470588235294118e-05,
      "loss": 0.6345,
      "step": 38
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 0.49056726694107056,
      "learning_rate": 4.705882352941177e-05,
      "loss": 0.697,
      "step": 40
    },
    {
      "epoch": 2.4705882352941178,
      "grad_norm": 0.4761519432067871,
      "learning_rate": 4.9411764705882355e-05,
      "loss": 0.8166,
      "step": 42
    },
    {
      "epoch": 2.588235294117647,
      "grad_norm": 0.5589882731437683,
      "learning_rate": 5.176470588235295e-05,
      "loss": 0.6736,
      "step": 44
    },
    {
      "epoch": 2.7058823529411766,
      "grad_norm": 0.561333954334259,
      "learning_rate": 5.411764705882353e-05,
      "loss": 0.6738,
      "step": 46
    },
    {
      "epoch": 2.8235294117647056,
      "grad_norm": 0.4906809329986572,
      "learning_rate": 5.647058823529412e-05,
      "loss": 0.6545,
      "step": 48
    },
    {
      "epoch": 2.9411764705882355,
      "grad_norm": 0.5915781855583191,
      "learning_rate": 5.882352941176471e-05,
      "loss": 0.6479,
      "step": 50
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.6654016971588135,
      "eval_runtime": 14.3138,
      "eval_samples_per_second": 2.515,
      "eval_steps_per_second": 2.515,
      "step": 51
    },
    {
      "epoch": 3.0588235294117645,
      "grad_norm": 1.159887671470642,
      "learning_rate": 6.11764705882353e-05,
      "loss": 0.5645,
      "step": 52
    },
    {
      "epoch": 3.176470588235294,
      "grad_norm": 0.42328324913978577,
      "learning_rate": 6.352941176470588e-05,
      "loss": 0.595,
      "step": 54
    },
    {
      "epoch": 3.2941176470588234,
      "grad_norm": 0.9672607183456421,
      "learning_rate": 6.588235294117648e-05,
      "loss": 0.6181,
      "step": 56
    },
    {
      "epoch": 3.411764705882353,
      "grad_norm": 0.6226010322570801,
      "learning_rate": 6.823529411764707e-05,
      "loss": 0.791,
      "step": 58
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 0.48087257146835327,
      "learning_rate": 7.058823529411765e-05,
      "loss": 0.5587,
      "step": 60
    },
    {
      "epoch": 3.6470588235294117,
      "grad_norm": 0.49178069829940796,
      "learning_rate": 7.294117647058823e-05,
      "loss": 0.5947,
      "step": 62
    },
    {
      "epoch": 3.764705882352941,
      "grad_norm": 0.746466338634491,
      "learning_rate": 7.529411764705883e-05,
      "loss": 0.5642,
      "step": 64
    },
    {
      "epoch": 3.8823529411764706,
      "grad_norm": 0.5014656782150269,
      "learning_rate": 7.764705882352942e-05,
      "loss": 0.5431,
      "step": 66
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.488460510969162,
      "learning_rate": 8e-05,
      "loss": 0.5369,
      "step": 68
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.6236481666564941,
      "eval_runtime": 14.3164,
      "eval_samples_per_second": 2.515,
      "eval_steps_per_second": 2.515,
      "step": 68
    },
    {
      "epoch": 4.117647058823529,
      "grad_norm": 0.6756840348243713,
      "learning_rate": 8.23529411764706e-05,
      "loss": 0.6918,
      "step": 70
    },
    {
      "epoch": 4.235294117647059,
      "grad_norm": 0.5347814559936523,
      "learning_rate": 8.470588235294118e-05,
      "loss": 0.4818,
      "step": 72
    },
    {
      "epoch": 4.352941176470588,
      "grad_norm": 0.531356692314148,
      "learning_rate": 8.705882352941177e-05,
      "loss": 0.5312,
      "step": 74
    },
    {
      "epoch": 4.470588235294118,
      "grad_norm": 0.5759537816047668,
      "learning_rate": 8.941176470588236e-05,
      "loss": 0.5072,
      "step": 76
    },
    {
      "epoch": 4.588235294117647,
      "grad_norm": 0.6224028468132019,
      "learning_rate": 9.176470588235295e-05,
      "loss": 0.508,
      "step": 78
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 0.6810643672943115,
      "learning_rate": 9.411764705882353e-05,
      "loss": 0.4936,
      "step": 80
    },
    {
      "epoch": 4.823529411764706,
      "grad_norm": 0.7130584716796875,
      "learning_rate": 9.647058823529412e-05,
      "loss": 0.516,
      "step": 82
    },
    {
      "epoch": 4.9411764705882355,
      "grad_norm": 0.6670802235603333,
      "learning_rate": 9.882352941176471e-05,
      "loss": 0.4817,
      "step": 84
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.6152952313423157,
      "eval_runtime": 14.328,
      "eval_samples_per_second": 2.513,
      "eval_steps_per_second": 2.513,
      "step": 85
    },
    {
      "epoch": 5.0588235294117645,
      "grad_norm": 0.7109234929084778,
      "learning_rate": 9.99995783847866e-05,
      "loss": 0.4423,
      "step": 86
    },
    {
      "epoch": 5.176470588235294,
      "grad_norm": 0.6109732985496521,
      "learning_rate": 9.999620550574153e-05,
      "loss": 0.4178,
      "step": 88
    },
    {
      "epoch": 5.294117647058823,
      "grad_norm": 0.7292227745056152,
      "learning_rate": 9.998945997517956e-05,
      "loss": 0.4594,
      "step": 90
    },
    {
      "epoch": 5.411764705882353,
      "grad_norm": 0.8408915400505066,
      "learning_rate": 9.997934224814173e-05,
      "loss": 0.4302,
      "step": 92
    },
    {
      "epoch": 5.529411764705882,
      "grad_norm": 0.925163984298706,
      "learning_rate": 9.996585300715116e-05,
      "loss": 0.4013,
      "step": 94
    },
    {
      "epoch": 5.647058823529412,
      "grad_norm": 0.8225246667861938,
      "learning_rate": 9.994899316216708e-05,
      "loss": 0.3892,
      "step": 96
    },
    {
      "epoch": 5.764705882352941,
      "grad_norm": 1.006257176399231,
      "learning_rate": 9.992876385052345e-05,
      "loss": 0.4298,
      "step": 98
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 0.9953014850616455,
      "learning_rate": 9.990516643685222e-05,
      "loss": 0.509,
      "step": 100
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.8502301573753357,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.4256,
      "step": 102
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.6260381937026978,
      "eval_runtime": 14.3197,
      "eval_samples_per_second": 2.514,
      "eval_steps_per_second": 2.514,
      "step": 102
    },
    {
      "epoch": 6.117647058823529,
      "grad_norm": 0.7284648418426514,
      "learning_rate": 9.984787389787688e-05,
      "loss": 0.3137,
      "step": 104
    },
    {
      "epoch": 6.235294117647059,
      "grad_norm": 1.1452425718307495,
      "learning_rate": 9.981418263742148e-05,
      "loss": 0.403,
      "step": 106
    },
    {
      "epoch": 6.352941176470588,
      "grad_norm": 0.8531830906867981,
      "learning_rate": 9.977713100437509e-05,
      "loss": 0.3199,
      "step": 108
    },
    {
      "epoch": 6.470588235294118,
      "grad_norm": 1.5461158752441406,
      "learning_rate": 9.973672149817232e-05,
      "loss": 0.3179,
      "step": 110
    },
    {
      "epoch": 6.588235294117647,
      "grad_norm": 1.0349928140640259,
      "learning_rate": 9.96929568447637e-05,
      "loss": 0.2666,
      "step": 112
    },
    {
      "epoch": 6.705882352941177,
      "grad_norm": 1.1756819486618042,
      "learning_rate": 9.964583999643174e-05,
      "loss": 0.3221,
      "step": 114
    },
    {
      "epoch": 6.823529411764706,
      "grad_norm": 1.0901398658752441,
      "learning_rate": 9.95953741315919e-05,
      "loss": 0.3165,
      "step": 116
    },
    {
      "epoch": 6.9411764705882355,
      "grad_norm": 1.0658103227615356,
      "learning_rate": 9.954156265457801e-05,
      "loss": 0.3157,
      "step": 118
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.7120265960693359,
      "eval_runtime": 14.3174,
      "eval_samples_per_second": 2.514,
      "eval_steps_per_second": 2.514,
      "step": 119
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 0.9966153502464294,
      "learning_rate": 9.948440919541278e-05,
      "loss": 0.3223,
      "step": 120
    },
    {
      "epoch": 7.176470588235294,
      "grad_norm": 0.9575842022895813,
      "learning_rate": 9.942391760956277e-05,
      "loss": 0.21,
      "step": 122
    },
    {
      "epoch": 7.294117647058823,
      "grad_norm": 1.0436619520187378,
      "learning_rate": 9.936009197767845e-05,
      "loss": 0.2066,
      "step": 124
    },
    {
      "epoch": 7.411764705882353,
      "grad_norm": 1.516401767730713,
      "learning_rate": 9.929293660531888e-05,
      "loss": 0.2019,
      "step": 126
    },
    {
      "epoch": 7.529411764705882,
      "grad_norm": 1.3807342052459717,
      "learning_rate": 9.922245602266118e-05,
      "loss": 0.2161,
      "step": 128
    },
    {
      "epoch": 7.647058823529412,
      "grad_norm": 1.4491366147994995,
      "learning_rate": 9.91486549841951e-05,
      "loss": 0.2008,
      "step": 130
    },
    {
      "epoch": 7.764705882352941,
      "grad_norm": 1.322877287864685,
      "learning_rate": 9.90715384684021e-05,
      "loss": 0.1657,
      "step": 132
    },
    {
      "epoch": 7.882352941176471,
      "grad_norm": 1.2288844585418701,
      "learning_rate": 9.899111167741966e-05,
      "loss": 0.2384,
      "step": 134
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.2698296308517456,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.2073,
      "step": 136
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.8672888278961182,
      "eval_runtime": 14.3211,
      "eval_samples_per_second": 2.514,
      "eval_steps_per_second": 2.514,
      "step": 136
    },
    {
      "epoch": 8.117647058823529,
      "grad_norm": 1.0830202102661133,
      "learning_rate": 9.882034919459555e-05,
      "loss": 0.1166,
      "step": 138
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 1.026341199874878,
      "learning_rate": 9.873002502207503e-05,
      "loss": 0.1275,
      "step": 140
    },
    {
      "epoch": 8.352941176470589,
      "grad_norm": 1.3224797248840332,
      "learning_rate": 9.863641361223024e-05,
      "loss": 0.1206,
      "step": 142
    },
    {
      "epoch": 8.470588235294118,
      "grad_norm": 1.3426852226257324,
      "learning_rate": 9.853952127991372e-05,
      "loss": 0.1422,
      "step": 144
    },
    {
      "epoch": 8.588235294117647,
      "grad_norm": 1.8169254064559937,
      "learning_rate": 9.843935456130295e-05,
      "loss": 0.1676,
      "step": 146
    },
    {
      "epoch": 8.705882352941176,
      "grad_norm": 1.5044218301773071,
      "learning_rate": 9.833592021345937e-05,
      "loss": 0.1163,
      "step": 148
    },
    {
      "epoch": 8.823529411764707,
      "grad_norm": 1.3765360116958618,
      "learning_rate": 9.822922521387276e-05,
      "loss": 0.1118,
      "step": 150
    },
    {
      "epoch": 8.941176470588236,
      "grad_norm": 1.3853477239608765,
      "learning_rate": 9.811927675999036e-05,
      "loss": 0.1196,
      "step": 152
    },
    {
      "epoch": 9.0,
      "eval_loss": 1.061182975769043,
      "eval_runtime": 14.3282,
      "eval_samples_per_second": 2.513,
      "eval_steps_per_second": 2.513,
      "step": 153
    },
    {
      "epoch": 9.058823529411764,
      "grad_norm": 0.9441946744918823,
      "learning_rate": 9.800608226873142e-05,
      "loss": 0.0887,
      "step": 154
    },
    {
      "epoch": 9.176470588235293,
      "grad_norm": 1.1457839012145996,
      "learning_rate": 9.788964937598689e-05,
      "loss": 0.0752,
      "step": 156
    },
    {
      "epoch": 9.294117647058824,
      "grad_norm": 1.1113783121109009,
      "learning_rate": 9.776998593610428e-05,
      "loss": 0.0594,
      "step": 158
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 1.4485852718353271,
      "learning_rate": 9.764710002135784e-05,
      "loss": 0.1021,
      "step": 160
    },
    {
      "epoch": 9.529411764705882,
      "grad_norm": 0.9894552826881409,
      "learning_rate": 9.752099992140399e-05,
      "loss": 0.0555,
      "step": 162
    },
    {
      "epoch": 9.647058823529411,
      "grad_norm": 1.9262152910232544,
      "learning_rate": 9.739169414272217e-05,
      "loss": 0.0778,
      "step": 164
    },
    {
      "epoch": 9.764705882352942,
      "grad_norm": 1.211846113204956,
      "learning_rate": 9.725919140804099e-05,
      "loss": 0.0877,
      "step": 166
    },
    {
      "epoch": 9.882352941176471,
      "grad_norm": 1.2289079427719116,
      "learning_rate": 9.71235006557497e-05,
      "loss": 0.0637,
      "step": 168
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.492928147315979,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.0925,
      "step": 170
    },
    {
      "epoch": 10.0,
      "eval_loss": 1.1790716648101807,
      "eval_runtime": 14.3168,
      "eval_samples_per_second": 2.515,
      "eval_steps_per_second": 2.515,
      "step": 170
    },
    {
      "epoch": 10.117647058823529,
      "grad_norm": 0.8547347187995911,
      "learning_rate": 9.684259192656553e-05,
      "loss": 0.0327,
      "step": 172
    },
    {
      "epoch": 10.235294117647058,
      "grad_norm": 1.168129801750183,
      "learning_rate": 9.669739289925577e-05,
      "loss": 0.0406,
      "step": 174
    },
    {
      "epoch": 10.352941176470589,
      "grad_norm": 0.9869183301925659,
      "learning_rate": 9.654904375222385e-05,
      "loss": 0.0458,
      "step": 176
    },
    {
      "epoch": 10.470588235294118,
      "grad_norm": 1.345210313796997,
      "learning_rate": 9.639755449282875e-05,
      "loss": 0.0487,
      "step": 178
    },
    {
      "epoch": 10.588235294117647,
      "grad_norm": 1.5439075231552124,
      "learning_rate": 9.62429353402556e-05,
      "loss": 0.0692,
      "step": 180
    },
    {
      "epoch": 10.705882352941176,
      "grad_norm": 1.3598881959915161,
      "learning_rate": 9.608519672482636e-05,
      "loss": 0.0506,
      "step": 182
    },
    {
      "epoch": 10.823529411764707,
      "grad_norm": 1.1148260831832886,
      "learning_rate": 9.592434928729616e-05,
      "loss": 0.0507,
      "step": 184
    },
    {
      "epoch": 10.941176470588236,
      "grad_norm": 1.2796578407287598,
      "learning_rate": 9.576040387813552e-05,
      "loss": 0.0527,
      "step": 186
    },
    {
      "epoch": 11.0,
      "eval_loss": 1.314186453819275,
      "eval_runtime": 14.3955,
      "eval_samples_per_second": 2.501,
      "eval_steps_per_second": 2.501,
      "step": 187
    },
    {
      "epoch": 11.058823529411764,
      "grad_norm": 1.0173205137252808,
      "learning_rate": 9.559337155679842e-05,
      "loss": 0.0427,
      "step": 188
    },
    {
      "epoch": 11.176470588235293,
      "grad_norm": 1.0243877172470093,
      "learning_rate": 9.542326359097619e-05,
      "loss": 0.0334,
      "step": 190
    },
    {
      "epoch": 11.294117647058824,
      "grad_norm": 0.6837695240974426,
      "learning_rate": 9.525009145583745e-05,
      "loss": 0.028,
      "step": 192
    },
    {
      "epoch": 11.411764705882353,
      "grad_norm": 0.9165364503860474,
      "learning_rate": 9.507386683325404e-05,
      "loss": 0.0251,
      "step": 194
    },
    {
      "epoch": 11.529411764705882,
      "grad_norm": 1.333146572113037,
      "learning_rate": 9.489460161101291e-05,
      "loss": 0.0419,
      "step": 196
    },
    {
      "epoch": 11.647058823529411,
      "grad_norm": 1.0797488689422607,
      "learning_rate": 9.471230788201429e-05,
      "loss": 0.0292,
      "step": 198
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 1.510589599609375,
      "learning_rate": 9.452699794345581e-05,
      "loss": 0.0579,
      "step": 200
    },
    {
      "epoch": 11.882352941176471,
      "grad_norm": 1.140371561050415,
      "learning_rate": 9.43386842960031e-05,
      "loss": 0.0329,
      "step": 202
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.9838421940803528,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.031,
      "step": 204
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.3305846452713013,
      "eval_runtime": 14.3149,
      "eval_samples_per_second": 2.515,
      "eval_steps_per_second": 2.515,
      "step": 204
    },
    {
      "epoch": 12.0,
      "step": 204,
      "total_flos": 5.499045460888781e+16,
      "train_loss": 0.4080328904274924,
      "train_runtime": 2337.636,
      "train_samples_per_second": 2.909,
      "train_steps_per_second": 0.364
    }
  ],
  "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": 5.499045460888781e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}