{
  "best_metric": 1.0226036310195923,
  "best_model_checkpoint": "data/Gemma-2-2B_task-1_180-samples_config-2_full/checkpoint-68",
  "epoch": 14.941176470588236,
  "eval_steps": 500,
  "global_step": 127,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11764705882352941,
      "grad_norm": 1.2928869724273682,
      "learning_rate": 2.5e-06,
      "loss": 2.5121,
      "step": 1
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 1.3557415008544922,
      "learning_rate": 5e-06,
      "loss": 2.4289,
      "step": 2
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 1.4175913333892822,
      "learning_rate": 1e-05,
      "loss": 2.4884,
      "step": 4
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 1.468453288078308,
      "learning_rate": 1.5e-05,
      "loss": 2.5267,
      "step": 6
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 1.396510362625122,
      "learning_rate": 2e-05,
      "loss": 2.5025,
      "step": 8
    },
    {
      "epoch": 0.9411764705882353,
      "eval_loss": 2.4361391067504883,
      "eval_runtime": 3.2171,
      "eval_samples_per_second": 11.19,
      "eval_steps_per_second": 11.19,
      "step": 8
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 1.2768471240997314,
      "learning_rate": 2.5e-05,
      "loss": 2.3866,
      "step": 10
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 1.2308349609375,
      "learning_rate": 3e-05,
      "loss": 2.3513,
      "step": 12
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 1.2498621940612793,
      "learning_rate": 3.5e-05,
      "loss": 2.271,
      "step": 14
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 1.0906994342803955,
      "learning_rate": 4e-05,
      "loss": 2.0775,
      "step": 16
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.908151626586914,
      "eval_runtime": 3.2514,
      "eval_samples_per_second": 11.072,
      "eval_steps_per_second": 11.072,
      "step": 17
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 0.9101805090904236,
      "learning_rate": 4.5e-05,
      "loss": 1.9483,
      "step": 18
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 0.8433398008346558,
      "learning_rate": 5e-05,
      "loss": 1.7828,
      "step": 20
    },
    {
      "epoch": 2.588235294117647,
      "grad_norm": 0.8993494510650635,
      "learning_rate": 5.500000000000001e-05,
      "loss": 1.709,
      "step": 22
    },
    {
      "epoch": 2.8235294117647056,
      "grad_norm": 0.8818881511688232,
      "learning_rate": 6e-05,
      "loss": 1.63,
      "step": 24
    },
    {
      "epoch": 2.9411764705882355,
      "eval_loss": 1.471886157989502,
      "eval_runtime": 3.2084,
      "eval_samples_per_second": 11.221,
      "eval_steps_per_second": 11.221,
      "step": 25
    },
    {
      "epoch": 3.0588235294117645,
      "grad_norm": 0.917806088924408,
      "learning_rate": 6.500000000000001e-05,
      "loss": 1.4438,
      "step": 26
    },
    {
      "epoch": 3.2941176470588234,
      "grad_norm": 0.7994047403335571,
      "learning_rate": 7e-05,
      "loss": 1.3963,
      "step": 28
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 0.7407734990119934,
      "learning_rate": 7.500000000000001e-05,
      "loss": 1.289,
      "step": 30
    },
    {
      "epoch": 3.764705882352941,
      "grad_norm": 0.5310801267623901,
      "learning_rate": 8e-05,
      "loss": 1.2196,
      "step": 32
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.5224806070327759,
      "learning_rate": 8.5e-05,
      "loss": 1.1623,
      "step": 34
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.181559681892395,
      "eval_runtime": 3.2359,
      "eval_samples_per_second": 11.125,
      "eval_steps_per_second": 11.125,
      "step": 34
    },
    {
      "epoch": 4.235294117647059,
      "grad_norm": 0.45314764976501465,
      "learning_rate": 9e-05,
      "loss": 1.1274,
      "step": 36
    },
    {
      "epoch": 4.470588235294118,
      "grad_norm": 0.4329112768173218,
      "learning_rate": 9.5e-05,
      "loss": 1.1191,
      "step": 38
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 0.4611508548259735,
      "learning_rate": 0.0001,
      "loss": 1.0931,
      "step": 40
    },
    {
      "epoch": 4.9411764705882355,
      "grad_norm": 0.4493888020515442,
      "learning_rate": 9.999238475781957e-05,
      "loss": 1.0588,
      "step": 42
    },
    {
      "epoch": 4.9411764705882355,
      "eval_loss": 1.0970077514648438,
      "eval_runtime": 3.249,
      "eval_samples_per_second": 11.08,
      "eval_steps_per_second": 11.08,
      "step": 42
    },
    {
      "epoch": 5.176470588235294,
      "grad_norm": 0.4374525249004364,
      "learning_rate": 9.99695413509548e-05,
      "loss": 1.0416,
      "step": 44
    },
    {
      "epoch": 5.411764705882353,
      "grad_norm": 0.40576106309890747,
      "learning_rate": 9.99314767377287e-05,
      "loss": 1.0197,
      "step": 46
    },
    {
      "epoch": 5.647058823529412,
      "grad_norm": 0.511160671710968,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.9458,
      "step": 48
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 0.43824997544288635,
      "learning_rate": 9.980973490458728e-05,
      "loss": 1.0217,
      "step": 50
    },
    {
      "epoch": 6.0,
      "eval_loss": 1.0481396913528442,
      "eval_runtime": 3.2147,
      "eval_samples_per_second": 11.199,
      "eval_steps_per_second": 11.199,
      "step": 51
    },
    {
      "epoch": 6.117647058823529,
      "grad_norm": 0.3990187346935272,
      "learning_rate": 9.972609476841367e-05,
      "loss": 0.9254,
      "step": 52
    },
    {
      "epoch": 6.352941176470588,
      "grad_norm": 0.4180718660354614,
      "learning_rate": 9.962730758206611e-05,
      "loss": 0.9386,
      "step": 54
    },
    {
      "epoch": 6.588235294117647,
      "grad_norm": 0.41820135712623596,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.9483,
      "step": 56
    },
    {
      "epoch": 6.823529411764706,
      "grad_norm": 0.47258704900741577,
      "learning_rate": 9.938441702975689e-05,
      "loss": 0.914,
      "step": 58
    },
    {
      "epoch": 6.9411764705882355,
      "eval_loss": 1.0287858247756958,
      "eval_runtime": 3.2255,
      "eval_samples_per_second": 11.161,
      "eval_steps_per_second": 11.161,
      "step": 59
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 0.4506057798862457,
      "learning_rate": 9.924038765061042e-05,
      "loss": 0.9126,
      "step": 60
    },
    {
      "epoch": 7.294117647058823,
      "grad_norm": 0.45049503445625305,
      "learning_rate": 9.908135917238321e-05,
      "loss": 0.8682,
      "step": 62
    },
    {
      "epoch": 7.529411764705882,
      "grad_norm": 0.4842223823070526,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.824,
      "step": 64
    },
    {
      "epoch": 7.764705882352941,
      "grad_norm": 0.5061436295509338,
      "learning_rate": 9.871850323926177e-05,
      "loss": 0.8147,
      "step": 66
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.5219563841819763,
      "learning_rate": 9.851478631379982e-05,
      "loss": 0.8449,
      "step": 68
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.0226036310195923,
      "eval_runtime": 3.1968,
      "eval_samples_per_second": 11.261,
      "eval_steps_per_second": 11.261,
      "step": 68
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 0.5255036354064941,
      "learning_rate": 9.829629131445342e-05,
      "loss": 0.7815,
      "step": 70
    },
    {
      "epoch": 8.470588235294118,
      "grad_norm": 0.5476519465446472,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.7882,
      "step": 72
    },
    {
      "epoch": 8.705882352941176,
      "grad_norm": 0.6667208075523376,
      "learning_rate": 9.781523779815179e-05,
      "loss": 0.7192,
      "step": 74
    },
    {
      "epoch": 8.941176470588236,
      "grad_norm": 0.762292206287384,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.7558,
      "step": 76
    },
    {
      "epoch": 8.941176470588236,
      "eval_loss": 1.0446863174438477,
      "eval_runtime": 3.2563,
      "eval_samples_per_second": 11.056,
      "eval_steps_per_second": 11.056,
      "step": 76
    },
    {
      "epoch": 9.176470588235293,
      "grad_norm": 0.6393022537231445,
      "learning_rate": 9.727592877996585e-05,
      "loss": 0.6956,
      "step": 78
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 0.664159893989563,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.6336,
      "step": 80
    },
    {
      "epoch": 9.647058823529411,
      "grad_norm": 0.7166351675987244,
      "learning_rate": 9.667902132486009e-05,
      "loss": 0.6638,
      "step": 82
    },
    {
      "epoch": 9.882352941176471,
      "grad_norm": 0.7807050943374634,
      "learning_rate": 9.635919272833938e-05,
      "loss": 0.6646,
      "step": 84
    },
    {
      "epoch": 10.0,
      "eval_loss": 1.0844277143478394,
      "eval_runtime": 3.1399,
      "eval_samples_per_second": 11.465,
      "eval_steps_per_second": 11.465,
      "step": 85
    },
    {
      "epoch": 10.117647058823529,
      "grad_norm": 0.7433176636695862,
      "learning_rate": 9.602524267262203e-05,
      "loss": 0.5804,
      "step": 86
    },
    {
      "epoch": 10.352941176470589,
      "grad_norm": 0.8067631125450134,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.5205,
      "step": 88
    },
    {
      "epoch": 10.588235294117647,
      "grad_norm": 1.005212426185608,
      "learning_rate": 9.53153893518325e-05,
      "loss": 0.5561,
      "step": 90
    },
    {
      "epoch": 10.823529411764707,
      "grad_norm": 1.0650252103805542,
      "learning_rate": 9.493970231495835e-05,
      "loss": 0.5742,
      "step": 92
    },
    {
      "epoch": 10.941176470588236,
      "eval_loss": 1.1509406566619873,
      "eval_runtime": 3.1742,
      "eval_samples_per_second": 11.342,
      "eval_steps_per_second": 11.342,
      "step": 93
    },
    {
      "epoch": 11.058823529411764,
      "grad_norm": 0.8970515131950378,
      "learning_rate": 9.45503262094184e-05,
      "loss": 0.5529,
      "step": 94
    },
    {
      "epoch": 11.294117647058824,
      "grad_norm": 1.0291845798492432,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.4454,
      "step": 96
    },
    {
      "epoch": 11.529411764705882,
      "grad_norm": 1.4001314640045166,
      "learning_rate": 9.373098535696979e-05,
      "loss": 0.4481,
      "step": 98
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 1.4160633087158203,
      "learning_rate": 9.330127018922194e-05,
      "loss": 0.4078,
      "step": 100
    },
    {
      "epoch": 12.0,
      "grad_norm": 1.1785461902618408,
      "learning_rate": 9.285836503510562e-05,
      "loss": 0.3766,
      "step": 102
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.3009971380233765,
      "eval_runtime": 3.2192,
      "eval_samples_per_second": 11.183,
      "eval_steps_per_second": 11.183,
      "step": 102
    },
    {
      "epoch": 12.235294117647058,
      "grad_norm": 1.2713053226470947,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.3695,
      "step": 104
    },
    {
      "epoch": 12.470588235294118,
      "grad_norm": 1.2278716564178467,
      "learning_rate": 9.193352839727121e-05,
      "loss": 0.3281,
      "step": 106
    },
    {
      "epoch": 12.705882352941176,
      "grad_norm": 1.6842726469039917,
      "learning_rate": 9.145187862775209e-05,
      "loss": 0.2966,
      "step": 108
    },
    {
      "epoch": 12.941176470588236,
      "grad_norm": 1.4139273166656494,
      "learning_rate": 9.09576022144496e-05,
      "loss": 0.3088,
      "step": 110
    },
    {
      "epoch": 12.941176470588236,
      "eval_loss": 1.5293265581130981,
      "eval_runtime": 3.1577,
      "eval_samples_per_second": 11.401,
      "eval_steps_per_second": 11.401,
      "step": 110
    },
    {
      "epoch": 13.176470588235293,
      "grad_norm": 1.2564139366149902,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.2694,
      "step": 112
    },
    {
      "epoch": 13.411764705882353,
      "grad_norm": 1.33437180519104,
      "learning_rate": 8.993177550236464e-05,
      "loss": 0.2205,
      "step": 114
    },
    {
      "epoch": 13.647058823529411,
      "grad_norm": 1.3570257425308228,
      "learning_rate": 8.940053768033609e-05,
      "loss": 0.2163,
      "step": 116
    },
    {
      "epoch": 13.882352941176471,
      "grad_norm": 1.5274231433868408,
      "learning_rate": 8.885729807284856e-05,
      "loss": 0.2417,
      "step": 118
    },
    {
      "epoch": 14.0,
      "eval_loss": 1.7321001291275024,
      "eval_runtime": 3.1501,
      "eval_samples_per_second": 11.428,
      "eval_steps_per_second": 11.428,
      "step": 119
    },
    {
      "epoch": 14.117647058823529,
      "grad_norm": 1.0535608530044556,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.2081,
      "step": 120
    },
    {
      "epoch": 14.352941176470589,
      "grad_norm": 1.19734787940979,
      "learning_rate": 8.773547901113862e-05,
      "loss": 0.1673,
      "step": 122
    },
    {
      "epoch": 14.588235294117647,
      "grad_norm": 1.3768539428710938,
      "learning_rate": 8.715724127386972e-05,
      "loss": 0.1801,
      "step": 124
    },
    {
      "epoch": 14.823529411764707,
      "grad_norm": 1.5754146575927734,
      "learning_rate": 8.656768508095853e-05,
      "loss": 0.1632,
      "step": 126
    },
    {
      "epoch": 14.941176470588236,
      "eval_loss": 1.9125914573669434,
      "eval_runtime": 3.1775,
      "eval_samples_per_second": 11.33,
      "eval_steps_per_second": 11.33,
      "step": 127
    },
    {
      "epoch": 14.941176470588236,
      "step": 127,
      "total_flos": 8474841308987392.0,
      "train_loss": 0.9817079439407258,
      "train_runtime": 869.6221,
      "train_samples_per_second": 7.819,
      "train_steps_per_second": 0.46
    }
  ],
  "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": 8474841308987392.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}