{
  "best_metric": 0.6108916997909546,
  "best_model_checkpoint": "data/Gemma-2-2B_task-2_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": 0.6985878944396973,
      "learning_rate": 2.5e-06,
      "loss": 0.9961,
      "step": 1
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 0.7494945526123047,
      "learning_rate": 5e-06,
      "loss": 1.0417,
      "step": 2
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 0.684769868850708,
      "learning_rate": 1e-05,
      "loss": 1.0099,
      "step": 4
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 0.6675004363059998,
      "learning_rate": 1.5e-05,
      "loss": 0.9974,
      "step": 6
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 0.6743130683898926,
      "learning_rate": 2e-05,
      "loss": 0.9786,
      "step": 8
    },
    {
      "epoch": 0.9411764705882353,
      "eval_loss": 1.0014729499816895,
      "eval_runtime": 14.3258,
      "eval_samples_per_second": 2.513,
      "eval_steps_per_second": 2.513,
      "step": 8
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 0.6840775609016418,
      "learning_rate": 2.5e-05,
      "loss": 1.0714,
      "step": 10
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 0.5774727463722229,
      "learning_rate": 3e-05,
      "loss": 0.9531,
      "step": 12
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 0.6025673747062683,
      "learning_rate": 3.5e-05,
      "loss": 0.9684,
      "step": 14
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 0.5042005777359009,
      "learning_rate": 4e-05,
      "loss": 0.8258,
      "step": 16
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.8337339162826538,
      "eval_runtime": 14.3136,
      "eval_samples_per_second": 2.515,
      "eval_steps_per_second": 2.515,
      "step": 17
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 0.47736313939094543,
      "learning_rate": 4.5e-05,
      "loss": 0.836,
      "step": 18
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 0.4720657765865326,
      "learning_rate": 5e-05,
      "loss": 0.7616,
      "step": 20
    },
    {
      "epoch": 2.588235294117647,
      "grad_norm": 0.41114121675491333,
      "learning_rate": 5.500000000000001e-05,
      "loss": 0.8426,
      "step": 22
    },
    {
      "epoch": 2.8235294117647056,
      "grad_norm": 0.43080058693885803,
      "learning_rate": 6e-05,
      "loss": 0.7427,
      "step": 24
    },
    {
      "epoch": 2.9411764705882355,
      "eval_loss": 0.7352527976036072,
      "eval_runtime": 14.3146,
      "eval_samples_per_second": 2.515,
      "eval_steps_per_second": 2.515,
      "step": 25
    },
    {
      "epoch": 3.0588235294117645,
      "grad_norm": 0.42164093255996704,
      "learning_rate": 6.500000000000001e-05,
      "loss": 0.6823,
      "step": 26
    },
    {
      "epoch": 3.2941176470588234,
      "grad_norm": 0.3572733998298645,
      "learning_rate": 7e-05,
      "loss": 0.6814,
      "step": 28
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 0.4064764380455017,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.7735,
      "step": 30
    },
    {
      "epoch": 3.764705882352941,
      "grad_norm": 0.38751062750816345,
      "learning_rate": 8e-05,
      "loss": 0.652,
      "step": 32
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.3583742678165436,
      "learning_rate": 8.5e-05,
      "loss": 0.6073,
      "step": 34
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.6722074151039124,
      "eval_runtime": 14.3189,
      "eval_samples_per_second": 2.514,
      "eval_steps_per_second": 2.514,
      "step": 34
    },
    {
      "epoch": 4.235294117647059,
      "grad_norm": 0.3805805742740631,
      "learning_rate": 9e-05,
      "loss": 0.6967,
      "step": 36
    },
    {
      "epoch": 4.470588235294118,
      "grad_norm": 0.3615116775035858,
      "learning_rate": 9.5e-05,
      "loss": 0.6047,
      "step": 38
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 0.3218657374382019,
      "learning_rate": 0.0001,
      "loss": 0.5826,
      "step": 40
    },
    {
      "epoch": 4.9411764705882355,
      "grad_norm": 0.32369837164878845,
      "learning_rate": 9.999238475781957e-05,
      "loss": 0.5792,
      "step": 42
    },
    {
      "epoch": 4.9411764705882355,
      "eval_loss": 0.6399049162864685,
      "eval_runtime": 14.3245,
      "eval_samples_per_second": 2.513,
      "eval_steps_per_second": 2.513,
      "step": 42
    },
    {
      "epoch": 5.176470588235294,
      "grad_norm": 0.3278553783893585,
      "learning_rate": 9.99695413509548e-05,
      "loss": 0.5413,
      "step": 44
    },
    {
      "epoch": 5.411764705882353,
      "grad_norm": 0.3335334360599518,
      "learning_rate": 9.99314767377287e-05,
      "loss": 0.5675,
      "step": 46
    },
    {
      "epoch": 5.647058823529412,
      "grad_norm": 0.32360556721687317,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.5056,
      "step": 48
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 0.39473217725753784,
      "learning_rate": 9.980973490458728e-05,
      "loss": 0.5866,
      "step": 50
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.6175941228866577,
      "eval_runtime": 14.3347,
      "eval_samples_per_second": 2.511,
      "eval_steps_per_second": 2.511,
      "step": 51
    },
    {
      "epoch": 6.117647058823529,
      "grad_norm": 0.35811060667037964,
      "learning_rate": 9.972609476841367e-05,
      "loss": 0.5178,
      "step": 52
    },
    {
      "epoch": 6.352941176470588,
      "grad_norm": 0.34254446625709534,
      "learning_rate": 9.962730758206611e-05,
      "loss": 0.5349,
      "step": 54
    },
    {
      "epoch": 6.588235294117647,
      "grad_norm": 0.39761653542518616,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.4716,
      "step": 56
    },
    {
      "epoch": 6.823529411764706,
      "grad_norm": 0.39172688126564026,
      "learning_rate": 9.938441702975689e-05,
      "loss": 0.4814,
      "step": 58
    },
    {
      "epoch": 6.9411764705882355,
      "eval_loss": 0.6108916997909546,
      "eval_runtime": 14.3165,
      "eval_samples_per_second": 2.515,
      "eval_steps_per_second": 2.515,
      "step": 59
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 0.46634534001350403,
      "learning_rate": 9.924038765061042e-05,
      "loss": 0.5187,
      "step": 60
    },
    {
      "epoch": 7.294117647058823,
      "grad_norm": 0.3935166895389557,
      "learning_rate": 9.908135917238321e-05,
      "loss": 0.433,
      "step": 62
    },
    {
      "epoch": 7.529411764705882,
      "grad_norm": 0.5637087821960449,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.4196,
      "step": 64
    },
    {
      "epoch": 7.764705882352941,
      "grad_norm": 0.4940302073955536,
      "learning_rate": 9.871850323926177e-05,
      "loss": 0.399,
      "step": 66
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.49748435616493225,
      "learning_rate": 9.851478631379982e-05,
      "loss": 0.4326,
      "step": 68
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.6347079277038574,
      "eval_runtime": 14.3222,
      "eval_samples_per_second": 2.514,
      "eval_steps_per_second": 2.514,
      "step": 68
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 0.5274530053138733,
      "learning_rate": 9.829629131445342e-05,
      "loss": 0.3546,
      "step": 70
    },
    {
      "epoch": 8.470588235294118,
      "grad_norm": 0.5347787737846375,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.3737,
      "step": 72
    },
    {
      "epoch": 8.705882352941176,
      "grad_norm": 0.5760989785194397,
      "learning_rate": 9.781523779815179e-05,
      "loss": 0.3905,
      "step": 74
    },
    {
      "epoch": 8.941176470588236,
      "grad_norm": 0.7981486320495605,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.3304,
      "step": 76
    },
    {
      "epoch": 8.941176470588236,
      "eval_loss": 0.6800416111946106,
      "eval_runtime": 14.3137,
      "eval_samples_per_second": 2.515,
      "eval_steps_per_second": 2.515,
      "step": 76
    },
    {
      "epoch": 9.176470588235293,
      "grad_norm": 0.6213347911834717,
      "learning_rate": 9.727592877996585e-05,
      "loss": 0.3038,
      "step": 78
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 0.8229406476020813,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.2932,
      "step": 80
    },
    {
      "epoch": 9.647058823529411,
      "grad_norm": 0.8098807334899902,
      "learning_rate": 9.667902132486009e-05,
      "loss": 0.2583,
      "step": 82
    },
    {
      "epoch": 9.882352941176471,
      "grad_norm": 0.8053330183029175,
      "learning_rate": 9.635919272833938e-05,
      "loss": 0.271,
      "step": 84
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.7600449323654175,
      "eval_runtime": 14.3124,
      "eval_samples_per_second": 2.515,
      "eval_steps_per_second": 2.515,
      "step": 85
    },
    {
      "epoch": 10.117647058823529,
      "grad_norm": 0.7272791862487793,
      "learning_rate": 9.602524267262203e-05,
      "loss": 0.2196,
      "step": 86
    },
    {
      "epoch": 10.352941176470589,
      "grad_norm": 0.8117285370826721,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.1931,
      "step": 88
    },
    {
      "epoch": 10.588235294117647,
      "grad_norm": 0.8827134966850281,
      "learning_rate": 9.53153893518325e-05,
      "loss": 0.2183,
      "step": 90
    },
    {
      "epoch": 10.823529411764707,
      "grad_norm": 1.0259592533111572,
      "learning_rate": 9.493970231495835e-05,
      "loss": 0.198,
      "step": 92
    },
    {
      "epoch": 10.941176470588236,
      "eval_loss": 0.9147105813026428,
      "eval_runtime": 14.3087,
      "eval_samples_per_second": 2.516,
      "eval_steps_per_second": 2.516,
      "step": 93
    },
    {
      "epoch": 11.058823529411764,
      "grad_norm": 0.8768501877784729,
      "learning_rate": 9.45503262094184e-05,
      "loss": 0.1804,
      "step": 94
    },
    {
      "epoch": 11.294117647058824,
      "grad_norm": 0.8787427544593811,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.1389,
      "step": 96
    },
    {
      "epoch": 11.529411764705882,
      "grad_norm": 0.9567432403564453,
      "learning_rate": 9.373098535696979e-05,
      "loss": 0.1155,
      "step": 98
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 1.0750855207443237,
      "learning_rate": 9.330127018922194e-05,
      "loss": 0.1217,
      "step": 100
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.9014711976051331,
      "learning_rate": 9.285836503510562e-05,
      "loss": 0.0956,
      "step": 102
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.082717776298523,
      "eval_runtime": 14.3089,
      "eval_samples_per_second": 2.516,
      "eval_steps_per_second": 2.516,
      "step": 102
    },
    {
      "epoch": 12.235294117647058,
      "grad_norm": 0.9323349595069885,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.0915,
      "step": 104
    },
    {
      "epoch": 12.470588235294118,
      "grad_norm": 1.0960149765014648,
      "learning_rate": 9.193352839727121e-05,
      "loss": 0.0713,
      "step": 106
    },
    {
      "epoch": 12.705882352941176,
      "grad_norm": 0.9801832437515259,
      "learning_rate": 9.145187862775209e-05,
      "loss": 0.059,
      "step": 108
    },
    {
      "epoch": 12.941176470588236,
      "grad_norm": 1.0168565511703491,
      "learning_rate": 9.09576022144496e-05,
      "loss": 0.0691,
      "step": 110
    },
    {
      "epoch": 12.941176470588236,
      "eval_loss": 1.213892936706543,
      "eval_runtime": 14.3174,
      "eval_samples_per_second": 2.514,
      "eval_steps_per_second": 2.514,
      "step": 110
    },
    {
      "epoch": 13.176470588235293,
      "grad_norm": 0.7937895655632019,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0614,
      "step": 112
    },
    {
      "epoch": 13.411764705882353,
      "grad_norm": 0.8961052894592285,
      "learning_rate": 8.993177550236464e-05,
      "loss": 0.0418,
      "step": 114
    },
    {
      "epoch": 13.647058823529411,
      "grad_norm": 0.8488242626190186,
      "learning_rate": 8.940053768033609e-05,
      "loss": 0.0447,
      "step": 116
    },
    {
      "epoch": 13.882352941176471,
      "grad_norm": 0.7885847687721252,
      "learning_rate": 8.885729807284856e-05,
      "loss": 0.0451,
      "step": 118
    },
    {
      "epoch": 14.0,
      "eval_loss": 1.3164575099945068,
      "eval_runtime": 14.3198,
      "eval_samples_per_second": 2.514,
      "eval_steps_per_second": 2.514,
      "step": 119
    },
    {
      "epoch": 14.0,
      "step": 119,
      "total_flos": 6.415553046931046e+16,
      "train_loss": 0.4680662748067319,
      "train_runtime": 2723.8485,
      "train_samples_per_second": 2.496,
      "train_steps_per_second": 0.147
    }
  ],
  "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": 6.415553046931046e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}