{
  "best_metric": 0.7009265422821045,
  "best_model_checkpoint": "data/Llama-31-8B_task-2_180-samples_config-2/checkpoint-59",
  "epoch": 14.0,
  "eval_steps": 500,
  "global_step": 119,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11764705882352941,
      "grad_norm": 0.6937819123268127,
      "learning_rate": 2.5e-06,
      "loss": 1.1573,
      "step": 1
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 0.7594375610351562,
      "learning_rate": 5e-06,
      "loss": 1.1671,
      "step": 2
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 0.6988368630409241,
      "learning_rate": 1e-05,
      "loss": 1.1422,
      "step": 4
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 0.652793824672699,
      "learning_rate": 1.5e-05,
      "loss": 1.0974,
      "step": 6
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 0.6293764114379883,
      "learning_rate": 2e-05,
      "loss": 1.0575,
      "step": 8
    },
    {
      "epoch": 0.9411764705882353,
      "eval_loss": 1.0785504579544067,
      "eval_runtime": 37.0113,
      "eval_samples_per_second": 0.973,
      "eval_steps_per_second": 0.973,
      "step": 8
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 0.6449664831161499,
      "learning_rate": 2.5e-05,
      "loss": 1.0769,
      "step": 10
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 0.558076798915863,
      "learning_rate": 3e-05,
      "loss": 1.0193,
      "step": 12
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 0.4432999789714813,
      "learning_rate": 3.5e-05,
      "loss": 0.9205,
      "step": 14
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 0.3556315004825592,
      "learning_rate": 4e-05,
      "loss": 0.9283,
      "step": 16
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.898560106754303,
      "eval_runtime": 37.0041,
      "eval_samples_per_second": 0.973,
      "eval_steps_per_second": 0.973,
      "step": 17
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 0.29573854804039,
      "learning_rate": 4.5e-05,
      "loss": 0.9063,
      "step": 18
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 0.32392242550849915,
      "learning_rate": 5e-05,
      "loss": 0.8399,
      "step": 20
    },
    {
      "epoch": 2.588235294117647,
      "grad_norm": 0.321981281042099,
      "learning_rate": 5.500000000000001e-05,
      "loss": 0.8435,
      "step": 22
    },
    {
      "epoch": 2.8235294117647056,
      "grad_norm": 0.33737948536872864,
      "learning_rate": 6e-05,
      "loss": 0.8658,
      "step": 24
    },
    {
      "epoch": 2.9411764705882355,
      "eval_loss": 0.8191602230072021,
      "eval_runtime": 37.0132,
      "eval_samples_per_second": 0.973,
      "eval_steps_per_second": 0.973,
      "step": 25
    },
    {
      "epoch": 3.0588235294117645,
      "grad_norm": 0.2735823094844818,
      "learning_rate": 6.500000000000001e-05,
      "loss": 0.7951,
      "step": 26
    },
    {
      "epoch": 3.2941176470588234,
      "grad_norm": 0.2502347528934479,
      "learning_rate": 7e-05,
      "loss": 0.7837,
      "step": 28
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 0.22765275835990906,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.7809,
      "step": 30
    },
    {
      "epoch": 3.764705882352941,
      "grad_norm": 0.24914021790027618,
      "learning_rate": 8e-05,
      "loss": 0.7784,
      "step": 32
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.22613126039505005,
      "learning_rate": 8.5e-05,
      "loss": 0.7202,
      "step": 34
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.7615711092948914,
      "eval_runtime": 37.0211,
      "eval_samples_per_second": 0.972,
      "eval_steps_per_second": 0.972,
      "step": 34
    },
    {
      "epoch": 4.235294117647059,
      "grad_norm": 0.23252207040786743,
      "learning_rate": 9e-05,
      "loss": 0.744,
      "step": 36
    },
    {
      "epoch": 4.470588235294118,
      "grad_norm": 0.2108556032180786,
      "learning_rate": 9.5e-05,
      "loss": 0.7133,
      "step": 38
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 0.2563748359680176,
      "learning_rate": 0.0001,
      "loss": 0.6765,
      "step": 40
    },
    {
      "epoch": 4.9411764705882355,
      "grad_norm": 0.23268263041973114,
      "learning_rate": 9.999238475781957e-05,
      "loss": 0.6781,
      "step": 42
    },
    {
      "epoch": 4.9411764705882355,
      "eval_loss": 0.7277030348777771,
      "eval_runtime": 37.0149,
      "eval_samples_per_second": 0.973,
      "eval_steps_per_second": 0.973,
      "step": 42
    },
    {
      "epoch": 5.176470588235294,
      "grad_norm": 0.29484882950782776,
      "learning_rate": 9.99695413509548e-05,
      "loss": 0.6667,
      "step": 44
    },
    {
      "epoch": 5.411764705882353,
      "grad_norm": 0.25725889205932617,
      "learning_rate": 9.99314767377287e-05,
      "loss": 0.6563,
      "step": 46
    },
    {
      "epoch": 5.647058823529412,
      "grad_norm": 0.2961147725582123,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.6092,
      "step": 48
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 0.2545129954814911,
      "learning_rate": 9.980973490458728e-05,
      "loss": 0.648,
      "step": 50
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.7057892680168152,
      "eval_runtime": 37.0131,
      "eval_samples_per_second": 0.973,
      "eval_steps_per_second": 0.973,
      "step": 51
    },
    {
      "epoch": 6.117647058823529,
      "grad_norm": 0.26211220026016235,
      "learning_rate": 9.972609476841367e-05,
      "loss": 0.619,
      "step": 52
    },
    {
      "epoch": 6.352941176470588,
      "grad_norm": 0.2682921588420868,
      "learning_rate": 9.962730758206611e-05,
      "loss": 0.5516,
      "step": 54
    },
    {
      "epoch": 6.588235294117647,
      "grad_norm": 0.31296029686927795,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.5951,
      "step": 56
    },
    {
      "epoch": 6.823529411764706,
      "grad_norm": 0.3109372854232788,
      "learning_rate": 9.938441702975689e-05,
      "loss": 0.5956,
      "step": 58
    },
    {
      "epoch": 6.9411764705882355,
      "eval_loss": 0.7009265422821045,
      "eval_runtime": 37.0181,
      "eval_samples_per_second": 0.972,
      "eval_steps_per_second": 0.972,
      "step": 59
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 0.35290929675102234,
      "learning_rate": 9.924038765061042e-05,
      "loss": 0.572,
      "step": 60
    },
    {
      "epoch": 7.294117647058823,
      "grad_norm": 0.32779383659362793,
      "learning_rate": 9.908135917238321e-05,
      "loss": 0.523,
      "step": 62
    },
    {
      "epoch": 7.529411764705882,
      "grad_norm": 0.3676365613937378,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.4772,
      "step": 64
    },
    {
      "epoch": 7.764705882352941,
      "grad_norm": 0.44564926624298096,
      "learning_rate": 9.871850323926177e-05,
      "loss": 0.4732,
      "step": 66
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.44333919882774353,
      "learning_rate": 9.851478631379982e-05,
      "loss": 0.5123,
      "step": 68
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.7150763869285583,
      "eval_runtime": 37.0145,
      "eval_samples_per_second": 0.973,
      "eval_steps_per_second": 0.973,
      "step": 68
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 0.4024699926376343,
      "learning_rate": 9.829629131445342e-05,
      "loss": 0.4392,
      "step": 70
    },
    {
      "epoch": 8.470588235294118,
      "grad_norm": 0.4763796627521515,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.4338,
      "step": 72
    },
    {
      "epoch": 8.705882352941176,
      "grad_norm": 0.5328041911125183,
      "learning_rate": 9.781523779815179e-05,
      "loss": 0.3874,
      "step": 74
    },
    {
      "epoch": 8.941176470588236,
      "grad_norm": 0.5904791951179504,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.3458,
      "step": 76
    },
    {
      "epoch": 8.941176470588236,
      "eval_loss": 0.8067886829376221,
      "eval_runtime": 37.0152,
      "eval_samples_per_second": 0.973,
      "eval_steps_per_second": 0.973,
      "step": 76
    },
    {
      "epoch": 9.176470588235293,
      "grad_norm": 0.6169093251228333,
      "learning_rate": 9.727592877996585e-05,
      "loss": 0.3306,
      "step": 78
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 0.7063341736793518,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.271,
      "step": 80
    },
    {
      "epoch": 9.647058823529411,
      "grad_norm": 0.718392014503479,
      "learning_rate": 9.667902132486009e-05,
      "loss": 0.2661,
      "step": 82
    },
    {
      "epoch": 9.882352941176471,
      "grad_norm": 0.8483970761299133,
      "learning_rate": 9.635919272833938e-05,
      "loss": 0.2817,
      "step": 84
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.8642737865447998,
      "eval_runtime": 37.0195,
      "eval_samples_per_second": 0.972,
      "eval_steps_per_second": 0.972,
      "step": 85
    },
    {
      "epoch": 10.117647058823529,
      "grad_norm": 0.8544296026229858,
      "learning_rate": 9.602524267262203e-05,
      "loss": 0.214,
      "step": 86
    },
    {
      "epoch": 10.352941176470589,
      "grad_norm": 1.7004799842834473,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.1626,
      "step": 88
    },
    {
      "epoch": 10.588235294117647,
      "grad_norm": 0.7825019955635071,
      "learning_rate": 9.53153893518325e-05,
      "loss": 0.1836,
      "step": 90
    },
    {
      "epoch": 10.823529411764707,
      "grad_norm": 0.8101257085800171,
      "learning_rate": 9.493970231495835e-05,
      "loss": 0.1755,
      "step": 92
    },
    {
      "epoch": 10.941176470588236,
      "eval_loss": 1.069422721862793,
      "eval_runtime": 37.0062,
      "eval_samples_per_second": 0.973,
      "eval_steps_per_second": 0.973,
      "step": 93
    },
    {
      "epoch": 11.058823529411764,
      "grad_norm": 1.0563342571258545,
      "learning_rate": 9.45503262094184e-05,
      "loss": 0.1677,
      "step": 94
    },
    {
      "epoch": 11.294117647058824,
      "grad_norm": 0.8128485083580017,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.0921,
      "step": 96
    },
    {
      "epoch": 11.529411764705882,
      "grad_norm": 0.7511378526687622,
      "learning_rate": 9.373098535696979e-05,
      "loss": 0.0805,
      "step": 98
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 0.7049291133880615,
      "learning_rate": 9.330127018922194e-05,
      "loss": 0.0825,
      "step": 100
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.6958498954772949,
      "learning_rate": 9.285836503510562e-05,
      "loss": 0.0778,
      "step": 102
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.1663720607757568,
      "eval_runtime": 37.0211,
      "eval_samples_per_second": 0.972,
      "eval_steps_per_second": 0.972,
      "step": 102
    },
    {
      "epoch": 12.235294117647058,
      "grad_norm": 0.6885355710983276,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.0467,
      "step": 104
    },
    {
      "epoch": 12.470588235294118,
      "grad_norm": 0.6764589548110962,
      "learning_rate": 9.193352839727121e-05,
      "loss": 0.0366,
      "step": 106
    },
    {
      "epoch": 12.705882352941176,
      "grad_norm": 0.9820438027381897,
      "learning_rate": 9.145187862775209e-05,
      "loss": 0.0443,
      "step": 108
    },
    {
      "epoch": 12.941176470588236,
      "grad_norm": 0.829129695892334,
      "learning_rate": 9.09576022144496e-05,
      "loss": 0.0487,
      "step": 110
    },
    {
      "epoch": 12.941176470588236,
      "eval_loss": 1.3311554193496704,
      "eval_runtime": 37.0171,
      "eval_samples_per_second": 0.973,
      "eval_steps_per_second": 0.973,
      "step": 110
    },
    {
      "epoch": 13.176470588235293,
      "grad_norm": 0.5137343406677246,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0307,
      "step": 112
    },
    {
      "epoch": 13.411764705882353,
      "grad_norm": 0.6914269924163818,
      "learning_rate": 8.993177550236464e-05,
      "loss": 0.0281,
      "step": 114
    },
    {
      "epoch": 13.647058823529411,
      "grad_norm": 0.7916271686553955,
      "learning_rate": 8.940053768033609e-05,
      "loss": 0.0249,
      "step": 116
    },
    {
      "epoch": 13.882352941176471,
      "grad_norm": 0.741131603717804,
      "learning_rate": 8.885729807284856e-05,
      "loss": 0.028,
      "step": 118
    },
    {
      "epoch": 14.0,
      "eval_loss": 1.5136417150497437,
      "eval_runtime": 37.0097,
      "eval_samples_per_second": 0.973,
      "eval_steps_per_second": 0.973,
      "step": 119
    },
    {
      "epoch": 14.0,
      "step": 119,
      "total_flos": 1.8224671303611187e+17,
      "train_loss": 0.5096657525028,
      "train_runtime": 6505.133,
      "train_samples_per_second": 1.045,
      "train_steps_per_second": 0.061
    }
  ],
  "logging_steps": 2,
  "max_steps": 400,
  "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": 1.8224671303611187e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}