{
  "best_metric": 0.5744918584823608,
  "best_model_checkpoint": "data/Llama-31-8B_task-2_180-samples_config-2_auto/checkpoint-51",
  "epoch": 12.941176470588236,
  "eval_steps": 500,
  "global_step": 110,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11764705882352941,
      "grad_norm": 0.3844596743583679,
      "learning_rate": 2.5e-06,
      "loss": 0.9123,
      "step": 1
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 0.41780412197113037,
      "learning_rate": 5e-06,
      "loss": 0.9403,
      "step": 2
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 0.376588374376297,
      "learning_rate": 1e-05,
      "loss": 0.8811,
      "step": 4
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 0.3761625289916992,
      "learning_rate": 1.5e-05,
      "loss": 0.8925,
      "step": 6
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 0.37995854020118713,
      "learning_rate": 2e-05,
      "loss": 0.8788,
      "step": 8
    },
    {
      "epoch": 0.9411764705882353,
      "eval_loss": 0.877348005771637,
      "eval_runtime": 38.0414,
      "eval_samples_per_second": 0.946,
      "eval_steps_per_second": 0.946,
      "step": 8
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 0.3556970953941345,
      "learning_rate": 2.5e-05,
      "loss": 0.851,
      "step": 10
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 0.3419400155544281,
      "learning_rate": 3e-05,
      "loss": 0.8325,
      "step": 12
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 0.3205713629722595,
      "learning_rate": 3.5e-05,
      "loss": 0.8229,
      "step": 14
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 0.2835601568222046,
      "learning_rate": 4e-05,
      "loss": 0.7758,
      "step": 16
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.7601401805877686,
      "eval_runtime": 38.0471,
      "eval_samples_per_second": 0.946,
      "eval_steps_per_second": 0.946,
      "step": 17
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 0.26591286063194275,
      "learning_rate": 4.5e-05,
      "loss": 0.7652,
      "step": 18
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 0.3065170347690582,
      "learning_rate": 5e-05,
      "loss": 0.6993,
      "step": 20
    },
    {
      "epoch": 2.588235294117647,
      "grad_norm": 0.3026050627231598,
      "learning_rate": 5.500000000000001e-05,
      "loss": 0.7078,
      "step": 22
    },
    {
      "epoch": 2.8235294117647056,
      "grad_norm": 0.3148728907108307,
      "learning_rate": 6e-05,
      "loss": 0.7042,
      "step": 24
    },
    {
      "epoch": 2.9411764705882355,
      "eval_loss": 0.6668568253517151,
      "eval_runtime": 38.0439,
      "eval_samples_per_second": 0.946,
      "eval_steps_per_second": 0.946,
      "step": 25
    },
    {
      "epoch": 3.0588235294117645,
      "grad_norm": 0.25821393728256226,
      "learning_rate": 6.500000000000001e-05,
      "loss": 0.6441,
      "step": 26
    },
    {
      "epoch": 3.2941176470588234,
      "grad_norm": 0.2481173425912857,
      "learning_rate": 7e-05,
      "loss": 0.6323,
      "step": 28
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 0.3203742504119873,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.6177,
      "step": 30
    },
    {
      "epoch": 3.764705882352941,
      "grad_norm": 0.30813756585121155,
      "learning_rate": 8e-05,
      "loss": 0.6197,
      "step": 32
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.26221516728401184,
      "learning_rate": 8.5e-05,
      "loss": 0.5735,
      "step": 34
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.6089106202125549,
      "eval_runtime": 38.0493,
      "eval_samples_per_second": 0.946,
      "eval_steps_per_second": 0.946,
      "step": 34
    },
    {
      "epoch": 4.235294117647059,
      "grad_norm": 0.2948972284793854,
      "learning_rate": 9e-05,
      "loss": 0.5694,
      "step": 36
    },
    {
      "epoch": 4.470588235294118,
      "grad_norm": 0.27812206745147705,
      "learning_rate": 9.5e-05,
      "loss": 0.5668,
      "step": 38
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 0.2674737870693207,
      "learning_rate": 0.0001,
      "loss": 0.5461,
      "step": 40
    },
    {
      "epoch": 4.9411764705882355,
      "grad_norm": 0.2772054970264435,
      "learning_rate": 9.999238475781957e-05,
      "loss": 0.5334,
      "step": 42
    },
    {
      "epoch": 4.9411764705882355,
      "eval_loss": 0.5867595672607422,
      "eval_runtime": 38.041,
      "eval_samples_per_second": 0.946,
      "eval_steps_per_second": 0.946,
      "step": 42
    },
    {
      "epoch": 5.176470588235294,
      "grad_norm": 0.2802392840385437,
      "learning_rate": 9.99695413509548e-05,
      "loss": 0.4954,
      "step": 44
    },
    {
      "epoch": 5.411764705882353,
      "grad_norm": 0.2730778157711029,
      "learning_rate": 9.99314767377287e-05,
      "loss": 0.4932,
      "step": 46
    },
    {
      "epoch": 5.647058823529412,
      "grad_norm": 0.30770865082740784,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.4719,
      "step": 48
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 0.32184281945228577,
      "learning_rate": 9.980973490458728e-05,
      "loss": 0.5073,
      "step": 50
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.5744918584823608,
      "eval_runtime": 38.0671,
      "eval_samples_per_second": 0.946,
      "eval_steps_per_second": 0.946,
      "step": 51
    },
    {
      "epoch": 6.117647058823529,
      "grad_norm": 0.2975810170173645,
      "learning_rate": 9.972609476841367e-05,
      "loss": 0.4778,
      "step": 52
    },
    {
      "epoch": 6.352941176470588,
      "grad_norm": 0.31714895367622375,
      "learning_rate": 9.962730758206611e-05,
      "loss": 0.4401,
      "step": 54
    },
    {
      "epoch": 6.588235294117647,
      "grad_norm": 0.3397834897041321,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.4288,
      "step": 56
    },
    {
      "epoch": 6.823529411764706,
      "grad_norm": 0.4086342751979828,
      "learning_rate": 9.938441702975689e-05,
      "loss": 0.42,
      "step": 58
    },
    {
      "epoch": 6.9411764705882355,
      "eval_loss": 0.5879939198493958,
      "eval_runtime": 38.0523,
      "eval_samples_per_second": 0.946,
      "eval_steps_per_second": 0.946,
      "step": 59
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 0.37034332752227783,
      "learning_rate": 9.924038765061042e-05,
      "loss": 0.4178,
      "step": 60
    },
    {
      "epoch": 7.294117647058823,
      "grad_norm": 0.3547601103782654,
      "learning_rate": 9.908135917238321e-05,
      "loss": 0.3715,
      "step": 62
    },
    {
      "epoch": 7.529411764705882,
      "grad_norm": 0.3913765549659729,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.3461,
      "step": 64
    },
    {
      "epoch": 7.764705882352941,
      "grad_norm": 0.4355185925960541,
      "learning_rate": 9.871850323926177e-05,
      "loss": 0.331,
      "step": 66
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.4763782322406769,
      "learning_rate": 9.851478631379982e-05,
      "loss": 0.3519,
      "step": 68
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.622241735458374,
      "eval_runtime": 38.0418,
      "eval_samples_per_second": 0.946,
      "eval_steps_per_second": 0.946,
      "step": 68
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 0.43703243136405945,
      "learning_rate": 9.829629131445342e-05,
      "loss": 0.2681,
      "step": 70
    },
    {
      "epoch": 8.470588235294118,
      "grad_norm": 0.5235260725021362,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.2863,
      "step": 72
    },
    {
      "epoch": 8.705882352941176,
      "grad_norm": 0.5810825228691101,
      "learning_rate": 9.781523779815179e-05,
      "loss": 0.2418,
      "step": 74
    },
    {
      "epoch": 8.941176470588236,
      "grad_norm": 0.6036997437477112,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.2325,
      "step": 76
    },
    {
      "epoch": 8.941176470588236,
      "eval_loss": 0.7260810136795044,
      "eval_runtime": 38.0425,
      "eval_samples_per_second": 0.946,
      "eval_steps_per_second": 0.946,
      "step": 76
    },
    {
      "epoch": 9.176470588235293,
      "grad_norm": 0.542436420917511,
      "learning_rate": 9.727592877996585e-05,
      "loss": 0.1959,
      "step": 78
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 0.6471271514892578,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.1574,
      "step": 80
    },
    {
      "epoch": 9.647058823529411,
      "grad_norm": 0.6145007610321045,
      "learning_rate": 9.667902132486009e-05,
      "loss": 0.1524,
      "step": 82
    },
    {
      "epoch": 9.882352941176471,
      "grad_norm": 0.6503257155418396,
      "learning_rate": 9.635919272833938e-05,
      "loss": 0.1557,
      "step": 84
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.7672773599624634,
      "eval_runtime": 38.0449,
      "eval_samples_per_second": 0.946,
      "eval_steps_per_second": 0.946,
      "step": 85
    },
    {
      "epoch": 10.117647058823529,
      "grad_norm": 0.4677157700061798,
      "learning_rate": 9.602524267262203e-05,
      "loss": 0.1066,
      "step": 86
    },
    {
      "epoch": 10.352941176470589,
      "grad_norm": 0.6145541667938232,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.084,
      "step": 88
    },
    {
      "epoch": 10.588235294117647,
      "grad_norm": 0.7448076009750366,
      "learning_rate": 9.53153893518325e-05,
      "loss": 0.0945,
      "step": 90
    },
    {
      "epoch": 10.823529411764707,
      "grad_norm": 0.6103171110153198,
      "learning_rate": 9.493970231495835e-05,
      "loss": 0.0979,
      "step": 92
    },
    {
      "epoch": 10.941176470588236,
      "eval_loss": 0.8393203020095825,
      "eval_runtime": 38.0419,
      "eval_samples_per_second": 0.946,
      "eval_steps_per_second": 0.946,
      "step": 93
    },
    {
      "epoch": 11.058823529411764,
      "grad_norm": 0.5315003395080566,
      "learning_rate": 9.45503262094184e-05,
      "loss": 0.0927,
      "step": 94
    },
    {
      "epoch": 11.294117647058824,
      "grad_norm": 0.5819242000579834,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.0711,
      "step": 96
    },
    {
      "epoch": 11.529411764705882,
      "grad_norm": 0.7706305980682373,
      "learning_rate": 9.373098535696979e-05,
      "loss": 0.0557,
      "step": 98
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 0.7562894225120544,
      "learning_rate": 9.330127018922194e-05,
      "loss": 0.0471,
      "step": 100
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.5847508311271667,
      "learning_rate": 9.285836503510562e-05,
      "loss": 0.0516,
      "step": 102
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.0522946119308472,
      "eval_runtime": 38.0432,
      "eval_samples_per_second": 0.946,
      "eval_steps_per_second": 0.946,
      "step": 102
    },
    {
      "epoch": 12.235294117647058,
      "grad_norm": 0.5992071032524109,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.0411,
      "step": 104
    },
    {
      "epoch": 12.470588235294118,
      "grad_norm": 0.47745710611343384,
      "learning_rate": 9.193352839727121e-05,
      "loss": 0.0384,
      "step": 106
    },
    {
      "epoch": 12.705882352941176,
      "grad_norm": 0.6165940761566162,
      "learning_rate": 9.145187862775209e-05,
      "loss": 0.036,
      "step": 108
    },
    {
      "epoch": 12.941176470588236,
      "grad_norm": 0.6436859369277954,
      "learning_rate": 9.09576022144496e-05,
      "loss": 0.0399,
      "step": 110
    },
    {
      "epoch": 12.941176470588236,
      "eval_loss": 1.1250815391540527,
      "eval_runtime": 38.0469,
      "eval_samples_per_second": 0.946,
      "eval_steps_per_second": 0.946,
      "step": 110
    },
    {
      "epoch": 12.941176470588236,
      "step": 110,
      "total_flos": 1.755282932634747e+17,
      "train_loss": 0.42093315938995646,
      "train_runtime": 6240.1976,
      "train_samples_per_second": 1.09,
      "train_steps_per_second": 0.064
    }
  ],
  "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": 1.755282932634747e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}