{
  "best_metric": 0.4326106309890747,
  "best_model_checkpoint": "data/Llama-31-8B_task-3_180-samples_config-2/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": 2.2055137157440186,
      "learning_rate": 2.5e-06,
      "loss": 2.3266,
      "step": 1
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 3.225144386291504,
      "learning_rate": 5e-06,
      "loss": 2.8101,
      "step": 2
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 2.424504280090332,
      "learning_rate": 1e-05,
      "loss": 2.475,
      "step": 4
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 2.089503765106201,
      "learning_rate": 1.5e-05,
      "loss": 2.3515,
      "step": 6
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 2.2423853874206543,
      "learning_rate": 2e-05,
      "loss": 2.3768,
      "step": 8
    },
    {
      "epoch": 0.9411764705882353,
      "eval_loss": 2.232229709625244,
      "eval_runtime": 31.8754,
      "eval_samples_per_second": 1.129,
      "eval_steps_per_second": 1.129,
      "step": 8
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 2.619719982147217,
      "learning_rate": 2.5e-05,
      "loss": 2.0995,
      "step": 10
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 2.8407723903656006,
      "learning_rate": 3e-05,
      "loss": 1.9656,
      "step": 12
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 2.3627195358276367,
      "learning_rate": 3.5e-05,
      "loss": 1.6764,
      "step": 14
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 2.162893295288086,
      "learning_rate": 4e-05,
      "loss": 1.3409,
      "step": 16
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.0478061437606812,
      "eval_runtime": 31.8614,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 17
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 1.5112247467041016,
      "learning_rate": 4.5e-05,
      "loss": 1.08,
      "step": 18
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 1.0380314588546753,
      "learning_rate": 5e-05,
      "loss": 0.9124,
      "step": 20
    },
    {
      "epoch": 2.588235294117647,
      "grad_norm": 0.7666953802108765,
      "learning_rate": 5.500000000000001e-05,
      "loss": 0.602,
      "step": 22
    },
    {
      "epoch": 2.8235294117647056,
      "grad_norm": 0.9076225757598877,
      "learning_rate": 6e-05,
      "loss": 0.6367,
      "step": 24
    },
    {
      "epoch": 2.9411764705882355,
      "eval_loss": 0.6136518716812134,
      "eval_runtime": 31.8662,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 25
    },
    {
      "epoch": 3.0588235294117645,
      "grad_norm": 0.8183653950691223,
      "learning_rate": 6.500000000000001e-05,
      "loss": 0.4824,
      "step": 26
    },
    {
      "epoch": 3.2941176470588234,
      "grad_norm": 0.4505358636379242,
      "learning_rate": 7e-05,
      "loss": 0.6645,
      "step": 28
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 0.7439283132553101,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.4431,
      "step": 30
    },
    {
      "epoch": 3.764705882352941,
      "grad_norm": 1.1238969564437866,
      "learning_rate": 8e-05,
      "loss": 0.5535,
      "step": 32
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.838411808013916,
      "learning_rate": 8.5e-05,
      "loss": 0.4263,
      "step": 34
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.5191609263420105,
      "eval_runtime": 31.8706,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 34
    },
    {
      "epoch": 4.235294117647059,
      "grad_norm": 0.3909635841846466,
      "learning_rate": 9e-05,
      "loss": 0.4403,
      "step": 36
    },
    {
      "epoch": 4.470588235294118,
      "grad_norm": 0.42084378004074097,
      "learning_rate": 9.5e-05,
      "loss": 0.5502,
      "step": 38
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 0.3670160472393036,
      "learning_rate": 0.0001,
      "loss": 0.4249,
      "step": 40
    },
    {
      "epoch": 4.9411764705882355,
      "grad_norm": 0.23142799735069275,
      "learning_rate": 9.999238475781957e-05,
      "loss": 0.3991,
      "step": 42
    },
    {
      "epoch": 4.9411764705882355,
      "eval_loss": 0.49313750863075256,
      "eval_runtime": 31.8705,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 42
    },
    {
      "epoch": 5.176470588235294,
      "grad_norm": 0.33376431465148926,
      "learning_rate": 9.99695413509548e-05,
      "loss": 0.4703,
      "step": 44
    },
    {
      "epoch": 5.411764705882353,
      "grad_norm": 0.10726317018270493,
      "learning_rate": 9.99314767377287e-05,
      "loss": 0.2559,
      "step": 46
    },
    {
      "epoch": 5.647058823529412,
      "grad_norm": 0.249160498380661,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.3673,
      "step": 48
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 0.7038401961326599,
      "learning_rate": 9.980973490458728e-05,
      "loss": 0.4279,
      "step": 50
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.4563770294189453,
      "eval_runtime": 31.8648,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 51
    },
    {
      "epoch": 6.117647058823529,
      "grad_norm": 0.25906163454055786,
      "learning_rate": 9.972609476841367e-05,
      "loss": 0.3876,
      "step": 52
    },
    {
      "epoch": 6.352941176470588,
      "grad_norm": 0.2575547397136688,
      "learning_rate": 9.962730758206611e-05,
      "loss": 0.3968,
      "step": 54
    },
    {
      "epoch": 6.588235294117647,
      "grad_norm": 0.3523285388946533,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.3124,
      "step": 56
    },
    {
      "epoch": 6.823529411764706,
      "grad_norm": 0.34885844588279724,
      "learning_rate": 9.938441702975689e-05,
      "loss": 0.3042,
      "step": 58
    },
    {
      "epoch": 6.9411764705882355,
      "eval_loss": 0.44394323229789734,
      "eval_runtime": 31.8697,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 59
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 0.37693843245506287,
      "learning_rate": 9.924038765061042e-05,
      "loss": 0.4905,
      "step": 60
    },
    {
      "epoch": 7.294117647058823,
      "grad_norm": 0.32455551624298096,
      "learning_rate": 9.908135917238321e-05,
      "loss": 0.3874,
      "step": 62
    },
    {
      "epoch": 7.529411764705882,
      "grad_norm": 0.7495986223220825,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.3324,
      "step": 64
    },
    {
      "epoch": 7.764705882352941,
      "grad_norm": 0.28021636605262756,
      "learning_rate": 9.871850323926177e-05,
      "loss": 0.2254,
      "step": 66
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.30057859420776367,
      "learning_rate": 9.851478631379982e-05,
      "loss": 0.2975,
      "step": 68
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.4326106309890747,
      "eval_runtime": 31.8689,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 68
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 0.4154025614261627,
      "learning_rate": 9.829629131445342e-05,
      "loss": 0.2872,
      "step": 70
    },
    {
      "epoch": 8.470588235294118,
      "grad_norm": 0.32561734318733215,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.3324,
      "step": 72
    },
    {
      "epoch": 8.705882352941176,
      "grad_norm": 0.2880627512931824,
      "learning_rate": 9.781523779815179e-05,
      "loss": 0.2095,
      "step": 74
    },
    {
      "epoch": 8.941176470588236,
      "grad_norm": 0.5983346700668335,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.2668,
      "step": 76
    },
    {
      "epoch": 8.941176470588236,
      "eval_loss": 0.45965951681137085,
      "eval_runtime": 31.8716,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 76
    },
    {
      "epoch": 9.176470588235293,
      "grad_norm": 0.4076274037361145,
      "learning_rate": 9.727592877996585e-05,
      "loss": 0.2338,
      "step": 78
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 0.4654916226863861,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.218,
      "step": 80
    },
    {
      "epoch": 9.647058823529411,
      "grad_norm": 0.5463346242904663,
      "learning_rate": 9.667902132486009e-05,
      "loss": 0.1795,
      "step": 82
    },
    {
      "epoch": 9.882352941176471,
      "grad_norm": 0.46247029304504395,
      "learning_rate": 9.635919272833938e-05,
      "loss": 0.179,
      "step": 84
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.48587766289711,
      "eval_runtime": 31.8715,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 85
    },
    {
      "epoch": 10.117647058823529,
      "grad_norm": 0.4759255051612854,
      "learning_rate": 9.602524267262203e-05,
      "loss": 0.1765,
      "step": 86
    },
    {
      "epoch": 10.352941176470589,
      "grad_norm": 0.5803537964820862,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.081,
      "step": 88
    },
    {
      "epoch": 10.588235294117647,
      "grad_norm": 1.32803475856781,
      "learning_rate": 9.53153893518325e-05,
      "loss": 0.1678,
      "step": 90
    },
    {
      "epoch": 10.823529411764707,
      "grad_norm": 0.6951686143875122,
      "learning_rate": 9.493970231495835e-05,
      "loss": 0.1451,
      "step": 92
    },
    {
      "epoch": 10.941176470588236,
      "eval_loss": 0.5927227735519409,
      "eval_runtime": 31.8672,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 93
    },
    {
      "epoch": 11.058823529411764,
      "grad_norm": 0.6281619668006897,
      "learning_rate": 9.45503262094184e-05,
      "loss": 0.1226,
      "step": 94
    },
    {
      "epoch": 11.294117647058824,
      "grad_norm": 1.5004541873931885,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.095,
      "step": 96
    },
    {
      "epoch": 11.529411764705882,
      "grad_norm": 0.788500189781189,
      "learning_rate": 9.373098535696979e-05,
      "loss": 0.074,
      "step": 98
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 0.7135135531425476,
      "learning_rate": 9.330127018922194e-05,
      "loss": 0.0389,
      "step": 100
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.8762282133102417,
      "learning_rate": 9.285836503510562e-05,
      "loss": 0.053,
      "step": 102
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.7126303911209106,
      "eval_runtime": 31.8778,
      "eval_samples_per_second": 1.129,
      "eval_steps_per_second": 1.129,
      "step": 102
    },
    {
      "epoch": 12.235294117647058,
      "grad_norm": 0.6949164867401123,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.0412,
      "step": 104
    },
    {
      "epoch": 12.470588235294118,
      "grad_norm": 0.8660315871238708,
      "learning_rate": 9.193352839727121e-05,
      "loss": 0.0321,
      "step": 106
    },
    {
      "epoch": 12.705882352941176,
      "grad_norm": 0.8823907375335693,
      "learning_rate": 9.145187862775209e-05,
      "loss": 0.0229,
      "step": 108
    },
    {
      "epoch": 12.941176470588236,
      "grad_norm": 0.7816784381866455,
      "learning_rate": 9.09576022144496e-05,
      "loss": 0.0246,
      "step": 110
    },
    {
      "epoch": 12.941176470588236,
      "eval_loss": 0.8255257606506348,
      "eval_runtime": 31.8789,
      "eval_samples_per_second": 1.129,
      "eval_steps_per_second": 1.129,
      "step": 110
    },
    {
      "epoch": 13.176470588235293,
      "grad_norm": 0.32839006185531616,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0185,
      "step": 112
    },
    {
      "epoch": 13.411764705882353,
      "grad_norm": 0.25856518745422363,
      "learning_rate": 8.993177550236464e-05,
      "loss": 0.0097,
      "step": 114
    },
    {
      "epoch": 13.647058823529411,
      "grad_norm": 0.8626673221588135,
      "learning_rate": 8.940053768033609e-05,
      "loss": 0.0171,
      "step": 116
    },
    {
      "epoch": 13.882352941176471,
      "grad_norm": 0.8288297057151794,
      "learning_rate": 8.885729807284856e-05,
      "loss": 0.0229,
      "step": 118
    },
    {
      "epoch": 14.0,
      "eval_loss": 0.8519785404205322,
      "eval_runtime": 31.8704,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 119
    },
    {
      "epoch": 14.117647058823529,
      "grad_norm": 0.48234981298446655,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.0188,
      "step": 120
    },
    {
      "epoch": 14.352941176470589,
      "grad_norm": 0.28182467818260193,
      "learning_rate": 8.773547901113862e-05,
      "loss": 0.006,
      "step": 122
    },
    {
      "epoch": 14.588235294117647,
      "grad_norm": 0.7883679866790771,
      "learning_rate": 8.715724127386972e-05,
      "loss": 0.011,
      "step": 124
    },
    {
      "epoch": 14.823529411764707,
      "grad_norm": 1.031003475189209,
      "learning_rate": 8.656768508095853e-05,
      "loss": 0.0233,
      "step": 126
    },
    {
      "epoch": 14.941176470588236,
      "eval_loss": 0.841795802116394,
      "eval_runtime": 31.8708,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 127
    },
    {
      "epoch": 14.941176470588236,
      "step": 127,
      "total_flos": 1.5999389691753267e+17,
      "train_loss": 0.5077414073886013,
      "train_runtime": 5870.3299,
      "train_samples_per_second": 1.158,
      "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": 1.5999389691753267e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}