{
  "best_metric": 0.6501245498657227,
  "best_model_checkpoint": "data/Gemma-2-2B_task-2_180-samples_config-2/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": 0.8071649670600891,
      "learning_rate": 2.5e-06,
      "loss": 1.0778,
      "step": 1
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 0.9397239685058594,
      "learning_rate": 5e-06,
      "loss": 1.1159,
      "step": 2
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 0.8109443187713623,
      "learning_rate": 1e-05,
      "loss": 1.0942,
      "step": 4
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 0.7966590523719788,
      "learning_rate": 1.5e-05,
      "loss": 1.031,
      "step": 6
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 0.7742748260498047,
      "learning_rate": 2e-05,
      "loss": 1.0192,
      "step": 8
    },
    {
      "epoch": 0.9411764705882353,
      "eval_loss": 1.054410457611084,
      "eval_runtime": 13.9398,
      "eval_samples_per_second": 2.583,
      "eval_steps_per_second": 2.583,
      "step": 8
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 0.8211076259613037,
      "learning_rate": 2.5e-05,
      "loss": 1.145,
      "step": 10
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 0.6804890632629395,
      "learning_rate": 3e-05,
      "loss": 1.0073,
      "step": 12
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 0.6264609098434448,
      "learning_rate": 3.5e-05,
      "loss": 0.9536,
      "step": 14
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 0.5299414992332458,
      "learning_rate": 4e-05,
      "loss": 0.8716,
      "step": 16
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.8600956797599792,
      "eval_runtime": 13.929,
      "eval_samples_per_second": 2.585,
      "eval_steps_per_second": 2.585,
      "step": 17
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 0.4857208728790283,
      "learning_rate": 4.5e-05,
      "loss": 0.8599,
      "step": 18
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 0.4451720714569092,
      "learning_rate": 5e-05,
      "loss": 0.7818,
      "step": 20
    },
    {
      "epoch": 2.588235294117647,
      "grad_norm": 0.4310362935066223,
      "learning_rate": 5.500000000000001e-05,
      "loss": 0.8738,
      "step": 22
    },
    {
      "epoch": 2.8235294117647056,
      "grad_norm": 0.39287930727005005,
      "learning_rate": 6e-05,
      "loss": 0.7904,
      "step": 24
    },
    {
      "epoch": 2.9411764705882355,
      "eval_loss": 0.7689363956451416,
      "eval_runtime": 13.9401,
      "eval_samples_per_second": 2.582,
      "eval_steps_per_second": 2.582,
      "step": 25
    },
    {
      "epoch": 3.0588235294117645,
      "grad_norm": 0.38156506419181824,
      "learning_rate": 6.500000000000001e-05,
      "loss": 0.7176,
      "step": 26
    },
    {
      "epoch": 3.2941176470588234,
      "grad_norm": 0.31532859802246094,
      "learning_rate": 7e-05,
      "loss": 0.7099,
      "step": 28
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 0.3182987570762634,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.8206,
      "step": 30
    },
    {
      "epoch": 3.764705882352941,
      "grad_norm": 0.330995112657547,
      "learning_rate": 8e-05,
      "loss": 0.7029,
      "step": 32
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.31390658020973206,
      "learning_rate": 8.5e-05,
      "loss": 0.6518,
      "step": 34
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.712177574634552,
      "eval_runtime": 13.9348,
      "eval_samples_per_second": 2.583,
      "eval_steps_per_second": 2.583,
      "step": 34
    },
    {
      "epoch": 4.235294117647059,
      "grad_norm": 0.28448107838630676,
      "learning_rate": 9e-05,
      "loss": 0.7639,
      "step": 36
    },
    {
      "epoch": 4.470588235294118,
      "grad_norm": 0.3131851553916931,
      "learning_rate": 9.5e-05,
      "loss": 0.6385,
      "step": 38
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 0.312388151884079,
      "learning_rate": 0.0001,
      "loss": 0.6184,
      "step": 40
    },
    {
      "epoch": 4.9411764705882355,
      "grad_norm": 0.2886146605014801,
      "learning_rate": 9.999238475781957e-05,
      "loss": 0.6153,
      "step": 42
    },
    {
      "epoch": 4.9411764705882355,
      "eval_loss": 0.6815809011459351,
      "eval_runtime": 13.9344,
      "eval_samples_per_second": 2.584,
      "eval_steps_per_second": 2.584,
      "step": 42
    },
    {
      "epoch": 5.176470588235294,
      "grad_norm": 0.29206588864326477,
      "learning_rate": 9.99695413509548e-05,
      "loss": 0.6103,
      "step": 44
    },
    {
      "epoch": 5.411764705882353,
      "grad_norm": 0.29846033453941345,
      "learning_rate": 9.99314767377287e-05,
      "loss": 0.6346,
      "step": 46
    },
    {
      "epoch": 5.647058823529412,
      "grad_norm": 0.31835803389549255,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.545,
      "step": 48
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 0.40522563457489014,
      "learning_rate": 9.980973490458728e-05,
      "loss": 0.6309,
      "step": 50
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.6577179431915283,
      "eval_runtime": 13.9343,
      "eval_samples_per_second": 2.584,
      "eval_steps_per_second": 2.584,
      "step": 51
    },
    {
      "epoch": 6.117647058823529,
      "grad_norm": 0.29965972900390625,
      "learning_rate": 9.972609476841367e-05,
      "loss": 0.5626,
      "step": 52
    },
    {
      "epoch": 6.352941176470588,
      "grad_norm": 0.3087371587753296,
      "learning_rate": 9.962730758206611e-05,
      "loss": 0.5514,
      "step": 54
    },
    {
      "epoch": 6.588235294117647,
      "grad_norm": 0.3645492494106293,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.5493,
      "step": 56
    },
    {
      "epoch": 6.823529411764706,
      "grad_norm": 0.3288743197917938,
      "learning_rate": 9.938441702975689e-05,
      "loss": 0.5496,
      "step": 58
    },
    {
      "epoch": 6.9411764705882355,
      "eval_loss": 0.6501245498657227,
      "eval_runtime": 13.9329,
      "eval_samples_per_second": 2.584,
      "eval_steps_per_second": 2.584,
      "step": 59
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 0.5126956105232239,
      "learning_rate": 9.924038765061042e-05,
      "loss": 0.5752,
      "step": 60
    },
    {
      "epoch": 7.294117647058823,
      "grad_norm": 0.3698246479034424,
      "learning_rate": 9.908135917238321e-05,
      "loss": 0.4895,
      "step": 62
    },
    {
      "epoch": 7.529411764705882,
      "grad_norm": 0.37893158197402954,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.4573,
      "step": 64
    },
    {
      "epoch": 7.764705882352941,
      "grad_norm": 0.41835057735443115,
      "learning_rate": 9.871850323926177e-05,
      "loss": 0.4518,
      "step": 66
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.43825429677963257,
      "learning_rate": 9.851478631379982e-05,
      "loss": 0.5015,
      "step": 68
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.654900312423706,
      "eval_runtime": 13.9273,
      "eval_samples_per_second": 2.585,
      "eval_steps_per_second": 2.585,
      "step": 68
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 0.5615823864936829,
      "learning_rate": 9.829629131445342e-05,
      "loss": 0.4371,
      "step": 70
    },
    {
      "epoch": 8.470588235294118,
      "grad_norm": 0.4718203842639923,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.4436,
      "step": 72
    },
    {
      "epoch": 8.705882352941176,
      "grad_norm": 0.5241195559501648,
      "learning_rate": 9.781523779815179e-05,
      "loss": 0.4396,
      "step": 74
    },
    {
      "epoch": 8.941176470588236,
      "grad_norm": 0.5938544273376465,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.3755,
      "step": 76
    },
    {
      "epoch": 8.941176470588236,
      "eval_loss": 0.7021616697311401,
      "eval_runtime": 13.9224,
      "eval_samples_per_second": 2.586,
      "eval_steps_per_second": 2.586,
      "step": 76
    },
    {
      "epoch": 9.176470588235293,
      "grad_norm": 0.6253803968429565,
      "learning_rate": 9.727592877996585e-05,
      "loss": 0.3811,
      "step": 78
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 0.7442609071731567,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.3416,
      "step": 80
    },
    {
      "epoch": 9.647058823529411,
      "grad_norm": 1.2256662845611572,
      "learning_rate": 9.667902132486009e-05,
      "loss": 0.3305,
      "step": 82
    },
    {
      "epoch": 9.882352941176471,
      "grad_norm": 0.739233136177063,
      "learning_rate": 9.635919272833938e-05,
      "loss": 0.341,
      "step": 84
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.7519544363021851,
      "eval_runtime": 13.9236,
      "eval_samples_per_second": 2.586,
      "eval_steps_per_second": 2.586,
      "step": 85
    },
    {
      "epoch": 10.117647058823529,
      "grad_norm": 0.7215401530265808,
      "learning_rate": 9.602524267262203e-05,
      "loss": 0.2774,
      "step": 86
    },
    {
      "epoch": 10.352941176470589,
      "grad_norm": 0.7284573912620544,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.2448,
      "step": 88
    },
    {
      "epoch": 10.588235294117647,
      "grad_norm": 1.1683156490325928,
      "learning_rate": 9.53153893518325e-05,
      "loss": 0.2814,
      "step": 90
    },
    {
      "epoch": 10.823529411764707,
      "grad_norm": 0.8636814951896667,
      "learning_rate": 9.493970231495835e-05,
      "loss": 0.2562,
      "step": 92
    },
    {
      "epoch": 10.941176470588236,
      "eval_loss": 0.8381435871124268,
      "eval_runtime": 13.9377,
      "eval_samples_per_second": 2.583,
      "eval_steps_per_second": 2.583,
      "step": 93
    },
    {
      "epoch": 11.058823529411764,
      "grad_norm": 0.8777179718017578,
      "learning_rate": 9.45503262094184e-05,
      "loss": 0.2658,
      "step": 94
    },
    {
      "epoch": 11.294117647058824,
      "grad_norm": 1.0160168409347534,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.1938,
      "step": 96
    },
    {
      "epoch": 11.529411764705882,
      "grad_norm": 1.087310552597046,
      "learning_rate": 9.373098535696979e-05,
      "loss": 0.1638,
      "step": 98
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 1.4101260900497437,
      "learning_rate": 9.330127018922194e-05,
      "loss": 0.1702,
      "step": 100
    },
    {
      "epoch": 12.0,
      "grad_norm": 1.0575681924819946,
      "learning_rate": 9.285836503510562e-05,
      "loss": 0.152,
      "step": 102
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.0347379446029663,
      "eval_runtime": 13.915,
      "eval_samples_per_second": 2.587,
      "eval_steps_per_second": 2.587,
      "step": 102
    },
    {
      "epoch": 12.235294117647058,
      "grad_norm": 0.9104964137077332,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.1142,
      "step": 104
    },
    {
      "epoch": 12.470588235294118,
      "grad_norm": 1.0433629751205444,
      "learning_rate": 9.193352839727121e-05,
      "loss": 0.0918,
      "step": 106
    },
    {
      "epoch": 12.705882352941176,
      "grad_norm": 1.135378360748291,
      "learning_rate": 9.145187862775209e-05,
      "loss": 0.1074,
      "step": 108
    },
    {
      "epoch": 12.941176470588236,
      "grad_norm": 1.3610491752624512,
      "learning_rate": 9.09576022144496e-05,
      "loss": 0.1132,
      "step": 110
    },
    {
      "epoch": 12.941176470588236,
      "eval_loss": 1.187706470489502,
      "eval_runtime": 13.9286,
      "eval_samples_per_second": 2.585,
      "eval_steps_per_second": 2.585,
      "step": 110
    },
    {
      "epoch": 13.176470588235293,
      "grad_norm": 0.8360520005226135,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0778,
      "step": 112
    },
    {
      "epoch": 13.411764705882353,
      "grad_norm": 0.8965087532997131,
      "learning_rate": 8.993177550236464e-05,
      "loss": 0.0628,
      "step": 114
    },
    {
      "epoch": 13.647058823529411,
      "grad_norm": 0.7961492538452148,
      "learning_rate": 8.940053768033609e-05,
      "loss": 0.0552,
      "step": 116
    },
    {
      "epoch": 13.882352941176471,
      "grad_norm": 1.0190213918685913,
      "learning_rate": 8.885729807284856e-05,
      "loss": 0.0618,
      "step": 118
    },
    {
      "epoch": 14.0,
      "eval_loss": 1.3359065055847168,
      "eval_runtime": 13.9157,
      "eval_samples_per_second": 2.587,
      "eval_steps_per_second": 2.587,
      "step": 119
    },
    {
      "epoch": 14.0,
      "step": 119,
      "total_flos": 6.237753178482278e+16,
      "train_loss": 0.515900345910497,
      "train_runtime": 2643.9097,
      "train_samples_per_second": 2.572,
      "train_steps_per_second": 0.151
    }
  ],
  "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": 6.237753178482278e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}