{
  "best_metric": 0.33868587017059326,
  "best_model_checkpoint": "data/Llama-31-8B_task-3_180-samples_config-2_auto/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": 2.021467447280884,
      "learning_rate": 2.5e-06,
      "loss": 2.0799,
      "step": 1
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 3.11173939704895,
      "learning_rate": 5e-06,
      "loss": 2.6403,
      "step": 2
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 2.315504550933838,
      "learning_rate": 1e-05,
      "loss": 2.2947,
      "step": 4
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 1.801814079284668,
      "learning_rate": 1.5e-05,
      "loss": 2.1182,
      "step": 6
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 2.036524534225464,
      "learning_rate": 2e-05,
      "loss": 2.1694,
      "step": 8
    },
    {
      "epoch": 0.9411764705882353,
      "eval_loss": 2.0003082752227783,
      "eval_runtime": 32.0224,
      "eval_samples_per_second": 1.124,
      "eval_steps_per_second": 1.124,
      "step": 8
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 2.5105128288269043,
      "learning_rate": 2.5e-05,
      "loss": 1.8779,
      "step": 10
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 2.7568750381469727,
      "learning_rate": 3e-05,
      "loss": 1.7988,
      "step": 12
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 2.2622737884521484,
      "learning_rate": 3.5e-05,
      "loss": 1.5424,
      "step": 14
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 2.128488063812256,
      "learning_rate": 4e-05,
      "loss": 1.1443,
      "step": 16
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.8652411103248596,
      "eval_runtime": 32.1727,
      "eval_samples_per_second": 1.119,
      "eval_steps_per_second": 1.119,
      "step": 17
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 1.4408377408981323,
      "learning_rate": 4.5e-05,
      "loss": 0.922,
      "step": 18
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 0.9458075165748596,
      "learning_rate": 5e-05,
      "loss": 0.7236,
      "step": 20
    },
    {
      "epoch": 2.588235294117647,
      "grad_norm": 0.6567085385322571,
      "learning_rate": 5.500000000000001e-05,
      "loss": 0.477,
      "step": 22
    },
    {
      "epoch": 2.8235294117647056,
      "grad_norm": 0.9464641809463501,
      "learning_rate": 6e-05,
      "loss": 0.5201,
      "step": 24
    },
    {
      "epoch": 2.9411764705882355,
      "eval_loss": 0.4725404977798462,
      "eval_runtime": 32.3745,
      "eval_samples_per_second": 1.112,
      "eval_steps_per_second": 1.112,
      "step": 25
    },
    {
      "epoch": 3.0588235294117645,
      "grad_norm": 1.6472339630126953,
      "learning_rate": 6.500000000000001e-05,
      "loss": 0.3977,
      "step": 26
    },
    {
      "epoch": 3.2941176470588234,
      "grad_norm": 0.6393092274665833,
      "learning_rate": 7e-05,
      "loss": 0.4804,
      "step": 28
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 0.28454142808914185,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.347,
      "step": 30
    },
    {
      "epoch": 3.764705882352941,
      "grad_norm": 1.2879621982574463,
      "learning_rate": 8e-05,
      "loss": 0.4046,
      "step": 32
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.7320904731750488,
      "learning_rate": 8.5e-05,
      "loss": 0.2979,
      "step": 34
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.3814656436443329,
      "eval_runtime": 32.0174,
      "eval_samples_per_second": 1.124,
      "eval_steps_per_second": 1.124,
      "step": 34
    },
    {
      "epoch": 4.235294117647059,
      "grad_norm": 0.3547268807888031,
      "learning_rate": 9e-05,
      "loss": 0.3104,
      "step": 36
    },
    {
      "epoch": 4.470588235294118,
      "grad_norm": 0.2421128749847412,
      "learning_rate": 9.5e-05,
      "loss": 0.3709,
      "step": 38
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 0.30042824149131775,
      "learning_rate": 0.0001,
      "loss": 0.33,
      "step": 40
    },
    {
      "epoch": 4.9411764705882355,
      "grad_norm": 0.2670217454433441,
      "learning_rate": 9.999238475781957e-05,
      "loss": 0.3128,
      "step": 42
    },
    {
      "epoch": 4.9411764705882355,
      "eval_loss": 0.37381914258003235,
      "eval_runtime": 32.0812,
      "eval_samples_per_second": 1.122,
      "eval_steps_per_second": 1.122,
      "step": 42
    },
    {
      "epoch": 5.176470588235294,
      "grad_norm": 0.3223280608654022,
      "learning_rate": 9.99695413509548e-05,
      "loss": 0.3502,
      "step": 44
    },
    {
      "epoch": 5.411764705882353,
      "grad_norm": 0.09799011051654816,
      "learning_rate": 9.99314767377287e-05,
      "loss": 0.1791,
      "step": 46
    },
    {
      "epoch": 5.647058823529412,
      "grad_norm": 0.28713056445121765,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.2813,
      "step": 48
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 0.2577309310436249,
      "learning_rate": 9.980973490458728e-05,
      "loss": 0.3107,
      "step": 50
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.3409940004348755,
      "eval_runtime": 32.0722,
      "eval_samples_per_second": 1.122,
      "eval_steps_per_second": 1.122,
      "step": 51
    },
    {
      "epoch": 6.117647058823529,
      "grad_norm": 0.2809261679649353,
      "learning_rate": 9.972609476841367e-05,
      "loss": 0.3084,
      "step": 52
    },
    {
      "epoch": 6.352941176470588,
      "grad_norm": 0.22198711335659027,
      "learning_rate": 9.962730758206611e-05,
      "loss": 0.296,
      "step": 54
    },
    {
      "epoch": 6.588235294117647,
      "grad_norm": 0.21486568450927734,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.2075,
      "step": 56
    },
    {
      "epoch": 6.823529411764706,
      "grad_norm": 0.30816760659217834,
      "learning_rate": 9.938441702975689e-05,
      "loss": 0.2488,
      "step": 58
    },
    {
      "epoch": 6.9411764705882355,
      "eval_loss": 0.33882033824920654,
      "eval_runtime": 32.0738,
      "eval_samples_per_second": 1.122,
      "eval_steps_per_second": 1.122,
      "step": 59
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 1.1969307661056519,
      "learning_rate": 9.924038765061042e-05,
      "loss": 0.3702,
      "step": 60
    },
    {
      "epoch": 7.294117647058823,
      "grad_norm": 0.2791982591152191,
      "learning_rate": 9.908135917238321e-05,
      "loss": 0.2952,
      "step": 62
    },
    {
      "epoch": 7.529411764705882,
      "grad_norm": 0.5182144045829773,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.2397,
      "step": 64
    },
    {
      "epoch": 7.764705882352941,
      "grad_norm": 0.2453116476535797,
      "learning_rate": 9.871850323926177e-05,
      "loss": 0.1793,
      "step": 66
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.2609338164329529,
      "learning_rate": 9.851478631379982e-05,
      "loss": 0.252,
      "step": 68
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.33868587017059326,
      "eval_runtime": 32.0677,
      "eval_samples_per_second": 1.123,
      "eval_steps_per_second": 1.123,
      "step": 68
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 0.5208426713943481,
      "learning_rate": 9.829629131445342e-05,
      "loss": 0.2136,
      "step": 70
    },
    {
      "epoch": 8.470588235294118,
      "grad_norm": 0.26494300365448,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.2568,
      "step": 72
    },
    {
      "epoch": 8.705882352941176,
      "grad_norm": 0.25580376386642456,
      "learning_rate": 9.781523779815179e-05,
      "loss": 0.1604,
      "step": 74
    },
    {
      "epoch": 8.941176470588236,
      "grad_norm": 0.42419931292533875,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.2324,
      "step": 76
    },
    {
      "epoch": 8.941176470588236,
      "eval_loss": 0.3474370539188385,
      "eval_runtime": 32.0689,
      "eval_samples_per_second": 1.123,
      "eval_steps_per_second": 1.123,
      "step": 76
    },
    {
      "epoch": 9.176470588235293,
      "grad_norm": 0.37841469049453735,
      "learning_rate": 9.727592877996585e-05,
      "loss": 0.2023,
      "step": 78
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 0.3214286267757416,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.1913,
      "step": 80
    },
    {
      "epoch": 9.647058823529411,
      "grad_norm": 0.3668440878391266,
      "learning_rate": 9.667902132486009e-05,
      "loss": 0.1572,
      "step": 82
    },
    {
      "epoch": 9.882352941176471,
      "grad_norm": 0.4008190929889679,
      "learning_rate": 9.635919272833938e-05,
      "loss": 0.1655,
      "step": 84
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.39318332076072693,
      "eval_runtime": 32.0708,
      "eval_samples_per_second": 1.123,
      "eval_steps_per_second": 1.123,
      "step": 85
    },
    {
      "epoch": 10.117647058823529,
      "grad_norm": 0.29186657071113586,
      "learning_rate": 9.602524267262203e-05,
      "loss": 0.1415,
      "step": 86
    },
    {
      "epoch": 10.352941176470589,
      "grad_norm": 0.38893839716911316,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.0942,
      "step": 88
    },
    {
      "epoch": 10.588235294117647,
      "grad_norm": 0.4997704327106476,
      "learning_rate": 9.53153893518325e-05,
      "loss": 0.1556,
      "step": 90
    },
    {
      "epoch": 10.823529411764707,
      "grad_norm": 0.5584585666656494,
      "learning_rate": 9.493970231495835e-05,
      "loss": 0.1782,
      "step": 92
    },
    {
      "epoch": 10.941176470588236,
      "eval_loss": 0.4127330482006073,
      "eval_runtime": 32.0702,
      "eval_samples_per_second": 1.123,
      "eval_steps_per_second": 1.123,
      "step": 93
    },
    {
      "epoch": 11.058823529411764,
      "grad_norm": 4.571016788482666,
      "learning_rate": 9.45503262094184e-05,
      "loss": 0.1421,
      "step": 94
    },
    {
      "epoch": 11.294117647058824,
      "grad_norm": 0.4855145514011383,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.1204,
      "step": 96
    },
    {
      "epoch": 11.529411764705882,
      "grad_norm": 0.5573723912239075,
      "learning_rate": 9.373098535696979e-05,
      "loss": 0.1159,
      "step": 98
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 0.7781118750572205,
      "learning_rate": 9.330127018922194e-05,
      "loss": 0.0704,
      "step": 100
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.5146734714508057,
      "learning_rate": 9.285836503510562e-05,
      "loss": 0.0536,
      "step": 102
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.44917285442352295,
      "eval_runtime": 32.0608,
      "eval_samples_per_second": 1.123,
      "eval_steps_per_second": 1.123,
      "step": 102
    },
    {
      "epoch": 12.235294117647058,
      "grad_norm": 0.7036318182945251,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.0835,
      "step": 104
    },
    {
      "epoch": 12.470588235294118,
      "grad_norm": 0.44796207547187805,
      "learning_rate": 9.193352839727121e-05,
      "loss": 0.0482,
      "step": 106
    },
    {
      "epoch": 12.705882352941176,
      "grad_norm": 0.5520998239517212,
      "learning_rate": 9.145187862775209e-05,
      "loss": 0.0383,
      "step": 108
    },
    {
      "epoch": 12.941176470588236,
      "grad_norm": 0.6558694243431091,
      "learning_rate": 9.09576022144496e-05,
      "loss": 0.0425,
      "step": 110
    },
    {
      "epoch": 12.941176470588236,
      "eval_loss": 0.5432764887809753,
      "eval_runtime": 32.0739,
      "eval_samples_per_second": 1.122,
      "eval_steps_per_second": 1.122,
      "step": 110
    },
    {
      "epoch": 13.176470588235293,
      "grad_norm": 0.3389807641506195,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0481,
      "step": 112
    },
    {
      "epoch": 13.411764705882353,
      "grad_norm": 0.3057907521724701,
      "learning_rate": 8.993177550236464e-05,
      "loss": 0.0239,
      "step": 114
    },
    {
      "epoch": 13.647058823529411,
      "grad_norm": 0.5624842643737793,
      "learning_rate": 8.940053768033609e-05,
      "loss": 0.0407,
      "step": 116
    },
    {
      "epoch": 13.882352941176471,
      "grad_norm": 0.5928642153739929,
      "learning_rate": 8.885729807284856e-05,
      "loss": 0.0347,
      "step": 118
    },
    {
      "epoch": 14.0,
      "eval_loss": 0.609021782875061,
      "eval_runtime": 32.0809,
      "eval_samples_per_second": 1.122,
      "eval_steps_per_second": 1.122,
      "step": 119
    },
    {
      "epoch": 14.117647058823529,
      "grad_norm": 0.3645310699939728,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.0333,
      "step": 120
    },
    {
      "epoch": 14.352941176470589,
      "grad_norm": 0.3996390998363495,
      "learning_rate": 8.773547901113862e-05,
      "loss": 0.0217,
      "step": 122
    },
    {
      "epoch": 14.588235294117647,
      "grad_norm": 0.6839818358421326,
      "learning_rate": 8.715724127386972e-05,
      "loss": 0.0347,
      "step": 124
    },
    {
      "epoch": 14.823529411764707,
      "grad_norm": 1.0604524612426758,
      "learning_rate": 8.656768508095853e-05,
      "loss": 0.0231,
      "step": 126
    },
    {
      "epoch": 14.941176470588236,
      "eval_loss": 0.5973975658416748,
      "eval_runtime": 32.0187,
      "eval_samples_per_second": 1.124,
      "eval_steps_per_second": 1.124,
      "step": 127
    },
    {
      "epoch": 14.941176470588236,
      "step": 127,
      "total_flos": 1.6116622709856666e+17,
      "train_loss": 0.43878336019051356,
      "train_runtime": 5909.9825,
      "train_samples_per_second": 1.151,
      "train_steps_per_second": 0.068
    }
  ],
  "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": 1.6116622709856666e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}