{
  "best_metric": 0.3731086850166321,
  "best_model_checkpoint": "data/Gemma-2-2B_task-3_180-samples_config-2_auto/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": 5.491757392883301,
      "learning_rate": 2.5e-06,
      "loss": 2.9251,
      "step": 1
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 13.161396980285645,
      "learning_rate": 5e-06,
      "loss": 3.7577,
      "step": 2
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 6.202048301696777,
      "learning_rate": 1e-05,
      "loss": 3.2505,
      "step": 4
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 4.814983367919922,
      "learning_rate": 1.5e-05,
      "loss": 2.9929,
      "step": 6
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 5.386263847351074,
      "learning_rate": 2e-05,
      "loss": 3.1214,
      "step": 8
    },
    {
      "epoch": 0.9411764705882353,
      "eval_loss": 2.8617448806762695,
      "eval_runtime": 11.3523,
      "eval_samples_per_second": 3.171,
      "eval_steps_per_second": 3.171,
      "step": 8
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 6.3948564529418945,
      "learning_rate": 2.5e-05,
      "loss": 2.7687,
      "step": 10
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 7.420201778411865,
      "learning_rate": 3e-05,
      "loss": 2.5214,
      "step": 12
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 4.606271266937256,
      "learning_rate": 3.5e-05,
      "loss": 2.1349,
      "step": 14
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 4.027822494506836,
      "learning_rate": 4e-05,
      "loss": 1.4487,
      "step": 16
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.1770029067993164,
      "eval_runtime": 11.3486,
      "eval_samples_per_second": 3.172,
      "eval_steps_per_second": 3.172,
      "step": 17
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 2.65433406829834,
      "learning_rate": 4.5e-05,
      "loss": 1.2553,
      "step": 18
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 1.9925974607467651,
      "learning_rate": 5e-05,
      "loss": 0.9851,
      "step": 20
    },
    {
      "epoch": 2.588235294117647,
      "grad_norm": 1.5486540794372559,
      "learning_rate": 5.500000000000001e-05,
      "loss": 0.8038,
      "step": 22
    },
    {
      "epoch": 2.8235294117647056,
      "grad_norm": 0.8094419836997986,
      "learning_rate": 6e-05,
      "loss": 0.6083,
      "step": 24
    },
    {
      "epoch": 2.9411764705882355,
      "eval_loss": 0.5479112267494202,
      "eval_runtime": 11.3381,
      "eval_samples_per_second": 3.175,
      "eval_steps_per_second": 3.175,
      "step": 25
    },
    {
      "epoch": 3.0588235294117645,
      "grad_norm": 0.6637625694274902,
      "learning_rate": 6.500000000000001e-05,
      "loss": 0.4663,
      "step": 26
    },
    {
      "epoch": 3.2941176470588234,
      "grad_norm": 0.606617271900177,
      "learning_rate": 7e-05,
      "loss": 0.5634,
      "step": 28
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 0.7731356024742126,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.5085,
      "step": 30
    },
    {
      "epoch": 3.764705882352941,
      "grad_norm": 0.6013955473899841,
      "learning_rate": 8e-05,
      "loss": 0.4741,
      "step": 32
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.5140759944915771,
      "learning_rate": 8.5e-05,
      "loss": 0.3481,
      "step": 34
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.43571236729621887,
      "eval_runtime": 11.346,
      "eval_samples_per_second": 3.173,
      "eval_steps_per_second": 3.173,
      "step": 34
    },
    {
      "epoch": 4.235294117647059,
      "grad_norm": 0.4210691750049591,
      "learning_rate": 9e-05,
      "loss": 0.4212,
      "step": 36
    },
    {
      "epoch": 4.470588235294118,
      "grad_norm": 0.49330905079841614,
      "learning_rate": 9.5e-05,
      "loss": 0.4154,
      "step": 38
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 0.41598206758499146,
      "learning_rate": 0.0001,
      "loss": 0.3712,
      "step": 40
    },
    {
      "epoch": 4.9411764705882355,
      "grad_norm": 0.6862059831619263,
      "learning_rate": 9.999238475781957e-05,
      "loss": 0.3365,
      "step": 42
    },
    {
      "epoch": 4.9411764705882355,
      "eval_loss": 0.39443454146385193,
      "eval_runtime": 11.3548,
      "eval_samples_per_second": 3.17,
      "eval_steps_per_second": 3.17,
      "step": 42
    },
    {
      "epoch": 5.176470588235294,
      "grad_norm": 0.42712822556495667,
      "learning_rate": 9.99695413509548e-05,
      "loss": 0.359,
      "step": 44
    },
    {
      "epoch": 5.411764705882353,
      "grad_norm": 0.7315046787261963,
      "learning_rate": 9.99314767377287e-05,
      "loss": 0.2014,
      "step": 46
    },
    {
      "epoch": 5.647058823529412,
      "grad_norm": 0.3802564740180969,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.2993,
      "step": 48
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 0.4529002904891968,
      "learning_rate": 9.980973490458728e-05,
      "loss": 0.3762,
      "step": 50
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.37851595878601074,
      "eval_runtime": 11.352,
      "eval_samples_per_second": 3.171,
      "eval_steps_per_second": 3.171,
      "step": 51
    },
    {
      "epoch": 6.117647058823529,
      "grad_norm": 0.28589263558387756,
      "learning_rate": 9.972609476841367e-05,
      "loss": 0.3172,
      "step": 52
    },
    {
      "epoch": 6.352941176470588,
      "grad_norm": 0.34178394079208374,
      "learning_rate": 9.962730758206611e-05,
      "loss": 0.3431,
      "step": 54
    },
    {
      "epoch": 6.588235294117647,
      "grad_norm": 0.310181200504303,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.2081,
      "step": 56
    },
    {
      "epoch": 6.823529411764706,
      "grad_norm": 0.4242079257965088,
      "learning_rate": 9.938441702975689e-05,
      "loss": 0.2475,
      "step": 58
    },
    {
      "epoch": 6.9411764705882355,
      "eval_loss": 0.3731086850166321,
      "eval_runtime": 11.3401,
      "eval_samples_per_second": 3.175,
      "eval_steps_per_second": 3.175,
      "step": 59
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 0.510144829750061,
      "learning_rate": 9.924038765061042e-05,
      "loss": 0.4045,
      "step": 60
    },
    {
      "epoch": 7.294117647058823,
      "grad_norm": 0.35856202244758606,
      "learning_rate": 9.908135917238321e-05,
      "loss": 0.2731,
      "step": 62
    },
    {
      "epoch": 7.529411764705882,
      "grad_norm": 0.3888154625892639,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.2129,
      "step": 64
    },
    {
      "epoch": 7.764705882352941,
      "grad_norm": 0.3968857228755951,
      "learning_rate": 9.871850323926177e-05,
      "loss": 0.1634,
      "step": 66
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.41991421580314636,
      "learning_rate": 9.851478631379982e-05,
      "loss": 0.2295,
      "step": 68
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.3856033682823181,
      "eval_runtime": 11.3495,
      "eval_samples_per_second": 3.172,
      "eval_steps_per_second": 3.172,
      "step": 68
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 0.4918777048587799,
      "learning_rate": 9.829629131445342e-05,
      "loss": 0.1766,
      "step": 70
    },
    {
      "epoch": 8.470588235294118,
      "grad_norm": 0.3988642692565918,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.2184,
      "step": 72
    },
    {
      "epoch": 8.705882352941176,
      "grad_norm": 0.4547469913959503,
      "learning_rate": 9.781523779815179e-05,
      "loss": 0.1559,
      "step": 74
    },
    {
      "epoch": 8.941176470588236,
      "grad_norm": 2.7850542068481445,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.1897,
      "step": 76
    },
    {
      "epoch": 8.941176470588236,
      "eval_loss": 0.43631118535995483,
      "eval_runtime": 11.353,
      "eval_samples_per_second": 3.171,
      "eval_steps_per_second": 3.171,
      "step": 76
    },
    {
      "epoch": 9.176470588235293,
      "grad_norm": 0.46989014744758606,
      "learning_rate": 9.727592877996585e-05,
      "loss": 0.1458,
      "step": 78
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 0.6614084243774414,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.1595,
      "step": 80
    },
    {
      "epoch": 9.647058823529411,
      "grad_norm": 0.6373603940010071,
      "learning_rate": 9.667902132486009e-05,
      "loss": 0.119,
      "step": 82
    },
    {
      "epoch": 9.882352941176471,
      "grad_norm": 0.6148824691772461,
      "learning_rate": 9.635919272833938e-05,
      "loss": 0.1273,
      "step": 84
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.5126776695251465,
      "eval_runtime": 11.3437,
      "eval_samples_per_second": 3.174,
      "eval_steps_per_second": 3.174,
      "step": 85
    },
    {
      "epoch": 10.117647058823529,
      "grad_norm": 0.3797300159931183,
      "learning_rate": 9.602524267262203e-05,
      "loss": 0.0989,
      "step": 86
    },
    {
      "epoch": 10.352941176470589,
      "grad_norm": 0.4829336404800415,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.0709,
      "step": 88
    },
    {
      "epoch": 10.588235294117647,
      "grad_norm": 0.5972853899002075,
      "learning_rate": 9.53153893518325e-05,
      "loss": 0.1048,
      "step": 90
    },
    {
      "epoch": 10.823529411764707,
      "grad_norm": 0.7292377352714539,
      "learning_rate": 9.493970231495835e-05,
      "loss": 0.1212,
      "step": 92
    },
    {
      "epoch": 10.941176470588236,
      "eval_loss": 0.5697892308235168,
      "eval_runtime": 11.3388,
      "eval_samples_per_second": 3.175,
      "eval_steps_per_second": 3.175,
      "step": 93
    },
    {
      "epoch": 11.058823529411764,
      "grad_norm": 0.3437809646129608,
      "learning_rate": 9.45503262094184e-05,
      "loss": 0.0805,
      "step": 94
    },
    {
      "epoch": 11.294117647058824,
      "grad_norm": 0.7272744178771973,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.0757,
      "step": 96
    },
    {
      "epoch": 11.529411764705882,
      "grad_norm": 0.8306965231895447,
      "learning_rate": 9.373098535696979e-05,
      "loss": 0.076,
      "step": 98
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 0.6787627339363098,
      "learning_rate": 9.330127018922194e-05,
      "loss": 0.0411,
      "step": 100
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.4867553412914276,
      "learning_rate": 9.285836503510562e-05,
      "loss": 0.031,
      "step": 102
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.7056739330291748,
      "eval_runtime": 11.3483,
      "eval_samples_per_second": 3.172,
      "eval_steps_per_second": 3.172,
      "step": 102
    },
    {
      "epoch": 12.235294117647058,
      "grad_norm": 1.0010762214660645,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.0518,
      "step": 104
    },
    {
      "epoch": 12.470588235294118,
      "grad_norm": 0.9890713095664978,
      "learning_rate": 9.193352839727121e-05,
      "loss": 0.0314,
      "step": 106
    },
    {
      "epoch": 12.705882352941176,
      "grad_norm": 0.43616411089897156,
      "learning_rate": 9.145187862775209e-05,
      "loss": 0.0253,
      "step": 108
    },
    {
      "epoch": 12.941176470588236,
      "grad_norm": 0.6954380869865417,
      "learning_rate": 9.09576022144496e-05,
      "loss": 0.0283,
      "step": 110
    },
    {
      "epoch": 12.941176470588236,
      "eval_loss": 0.7807791829109192,
      "eval_runtime": 11.3378,
      "eval_samples_per_second": 3.175,
      "eval_steps_per_second": 3.175,
      "step": 110
    },
    {
      "epoch": 13.176470588235293,
      "grad_norm": 1.154528260231018,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0331,
      "step": 112
    },
    {
      "epoch": 13.411764705882353,
      "grad_norm": 0.37544330954551697,
      "learning_rate": 8.993177550236464e-05,
      "loss": 0.0174,
      "step": 114
    },
    {
      "epoch": 13.647058823529411,
      "grad_norm": 0.800521969795227,
      "learning_rate": 8.940053768033609e-05,
      "loss": 0.0341,
      "step": 116
    },
    {
      "epoch": 13.882352941176471,
      "grad_norm": 0.6290589570999146,
      "learning_rate": 8.885729807284856e-05,
      "loss": 0.0227,
      "step": 118
    },
    {
      "epoch": 14.0,
      "eval_loss": 0.8175501823425293,
      "eval_runtime": 11.3375,
      "eval_samples_per_second": 3.175,
      "eval_steps_per_second": 3.175,
      "step": 119
    },
    {
      "epoch": 14.0,
      "step": 119,
      "total_flos": 5.163085364566426e+16,
      "train_loss": 0.58860215244173,
      "train_runtime": 2168.2081,
      "train_samples_per_second": 3.136,
      "train_steps_per_second": 0.184
    }
  ],
  "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": 5.163085364566426e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}