{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9845207382417147,
  "eval_steps": 200,
  "global_step": 10000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "eval_loss": 0.24929340183734894,
      "eval_runtime": 152.0524,
      "eval_samples_per_second": 942.675,
      "eval_steps_per_second": 3.683,
      "step": 200
    },
    {
      "epoch": 0.08,
      "eval_loss": 0.397050142288208,
      "eval_runtime": 151.7752,
      "eval_samples_per_second": 944.397,
      "eval_steps_per_second": 3.69,
      "step": 400
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.751934907719786e-05,
      "loss": 0.4919,
      "step": 500
    },
    {
      "epoch": 0.12,
      "eval_loss": 0.619749903678894,
      "eval_runtime": 153.0348,
      "eval_samples_per_second": 936.624,
      "eval_steps_per_second": 3.659,
      "step": 600
    },
    {
      "epoch": 0.16,
      "eval_loss": 0.5482326149940491,
      "eval_runtime": 151.3209,
      "eval_samples_per_second": 947.232,
      "eval_steps_per_second": 3.701,
      "step": 800
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.5038698154395716e-05,
      "loss": 0.9307,
      "step": 1000
    },
    {
      "epoch": 0.2,
      "eval_loss": 0.861940324306488,
      "eval_runtime": 150.9686,
      "eval_samples_per_second": 949.442,
      "eval_steps_per_second": 3.709,
      "step": 1000
    },
    {
      "epoch": 0.24,
      "eval_loss": 0.5618771910667419,
      "eval_runtime": 151.1536,
      "eval_samples_per_second": 948.28,
      "eval_steps_per_second": 3.705,
      "step": 1200
    },
    {
      "epoch": 0.28,
      "eval_loss": 0.7757473587989807,
      "eval_runtime": 151.2249,
      "eval_samples_per_second": 947.834,
      "eval_steps_per_second": 3.703,
      "step": 1400
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.255804723159357e-05,
      "loss": 1.6552,
      "step": 1500
    },
    {
      "epoch": 0.32,
      "eval_loss": 0.5049824714660645,
      "eval_runtime": 151.8547,
      "eval_samples_per_second": 943.902,
      "eval_steps_per_second": 3.688,
      "step": 1600
    },
    {
      "epoch": 0.36,
      "eval_loss": 1.1517618894577026,
      "eval_runtime": 151.1455,
      "eval_samples_per_second": 948.331,
      "eval_steps_per_second": 3.705,
      "step": 1800
    },
    {
      "epoch": 0.4,
      "learning_rate": 4.0077396308791423e-05,
      "loss": 1.1387,
      "step": 2000
    },
    {
      "epoch": 0.4,
      "eval_loss": 1.0938903093338013,
      "eval_runtime": 151.2372,
      "eval_samples_per_second": 947.756,
      "eval_steps_per_second": 3.703,
      "step": 2000
    },
    {
      "epoch": 0.44,
      "eval_loss": 9.282928466796875,
      "eval_runtime": 150.978,
      "eval_samples_per_second": 949.383,
      "eval_steps_per_second": 3.709,
      "step": 2200
    },
    {
      "epoch": 0.48,
      "eval_loss": 0.2713712155818939,
      "eval_runtime": 151.4204,
      "eval_samples_per_second": 946.61,
      "eval_steps_per_second": 3.698,
      "step": 2400
    },
    {
      "epoch": 0.5,
      "learning_rate": 3.7596745385989284e-05,
      "loss": 8.5966,
      "step": 2500
    },
    {
      "epoch": 0.52,
      "eval_loss": 0.12629659473896027,
      "eval_runtime": 151.6925,
      "eval_samples_per_second": 944.911,
      "eval_steps_per_second": 3.692,
      "step": 2600
    },
    {
      "epoch": 0.56,
      "eval_loss": 0.11906945705413818,
      "eval_runtime": 151.1292,
      "eval_samples_per_second": 948.434,
      "eval_steps_per_second": 3.705,
      "step": 2800
    },
    {
      "epoch": 0.6,
      "learning_rate": 3.5116094463187144e-05,
      "loss": 0.1233,
      "step": 3000
    },
    {
      "epoch": 0.6,
      "eval_loss": 0.11606267094612122,
      "eval_runtime": 151.1195,
      "eval_samples_per_second": 948.494,
      "eval_steps_per_second": 3.706,
      "step": 3000
    },
    {
      "epoch": 0.64,
      "eval_loss": 0.11504378169775009,
      "eval_runtime": 151.2469,
      "eval_samples_per_second": 947.695,
      "eval_steps_per_second": 3.703,
      "step": 3200
    },
    {
      "epoch": 0.67,
      "eval_loss": 0.11453282833099365,
      "eval_runtime": 151.278,
      "eval_samples_per_second": 947.501,
      "eval_steps_per_second": 3.702,
      "step": 3400
    },
    {
      "epoch": 0.69,
      "learning_rate": 3.2635443540385e-05,
      "loss": 0.1166,
      "step": 3500
    },
    {
      "epoch": 0.71,
      "eval_loss": 0.1138230413198471,
      "eval_runtime": 151.2516,
      "eval_samples_per_second": 947.666,
      "eval_steps_per_second": 3.702,
      "step": 3600
    },
    {
      "epoch": 0.75,
      "eval_loss": 0.11347956955432892,
      "eval_runtime": 151.7267,
      "eval_samples_per_second": 944.699,
      "eval_steps_per_second": 3.691,
      "step": 3800
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.0154792617582855e-05,
      "loss": 0.1151,
      "step": 4000
    },
    {
      "epoch": 0.79,
      "eval_loss": 0.11324501782655716,
      "eval_runtime": 151.6688,
      "eval_samples_per_second": 945.059,
      "eval_steps_per_second": 3.692,
      "step": 4000
    },
    {
      "epoch": 0.83,
      "eval_loss": 0.11296597123146057,
      "eval_runtime": 151.2257,
      "eval_samples_per_second": 947.829,
      "eval_steps_per_second": 3.703,
      "step": 4200
    },
    {
      "epoch": 0.87,
      "eval_loss": 0.11248484253883362,
      "eval_runtime": 151.3062,
      "eval_samples_per_second": 947.324,
      "eval_steps_per_second": 3.701,
      "step": 4400
    },
    {
      "epoch": 0.89,
      "learning_rate": 2.7674141694780715e-05,
      "loss": 0.1131,
      "step": 4500
    },
    {
      "epoch": 0.91,
      "eval_loss": 0.11222843080759048,
      "eval_runtime": 151.1995,
      "eval_samples_per_second": 947.992,
      "eval_steps_per_second": 3.704,
      "step": 4600
    },
    {
      "epoch": 0.95,
      "eval_loss": 0.1118897795677185,
      "eval_runtime": 151.3101,
      "eval_samples_per_second": 947.3,
      "eval_steps_per_second": 3.701,
      "step": 4800
    },
    {
      "epoch": 0.99,
      "learning_rate": 2.519349077197857e-05,
      "loss": 0.1132,
      "step": 5000
    },
    {
      "epoch": 0.99,
      "eval_loss": 0.11164118349552155,
      "eval_runtime": 151.2544,
      "eval_samples_per_second": 947.648,
      "eval_steps_per_second": 3.702,
      "step": 5000
    },
    {
      "epoch": 1.03,
      "eval_loss": 0.11147473752498627,
      "eval_runtime": 151.2525,
      "eval_samples_per_second": 947.661,
      "eval_steps_per_second": 3.702,
      "step": 5200
    },
    {
      "epoch": 1.07,
      "eval_loss": 0.1114504486322403,
      "eval_runtime": 151.3584,
      "eval_samples_per_second": 946.997,
      "eval_steps_per_second": 3.7,
      "step": 5400
    },
    {
      "epoch": 1.09,
      "learning_rate": 2.2712839849176426e-05,
      "loss": 0.1123,
      "step": 5500
    },
    {
      "epoch": 1.11,
      "eval_loss": 0.1111702173948288,
      "eval_runtime": 151.4231,
      "eval_samples_per_second": 946.592,
      "eval_steps_per_second": 3.698,
      "step": 5600
    },
    {
      "epoch": 1.15,
      "eval_loss": 0.11109592020511627,
      "eval_runtime": 151.3306,
      "eval_samples_per_second": 947.171,
      "eval_steps_per_second": 3.701,
      "step": 5800
    },
    {
      "epoch": 1.19,
      "learning_rate": 2.0232188926374283e-05,
      "loss": 0.1116,
      "step": 6000
    },
    {
      "epoch": 1.19,
      "eval_loss": 0.11104920506477356,
      "eval_runtime": 151.4541,
      "eval_samples_per_second": 946.399,
      "eval_steps_per_second": 3.697,
      "step": 6000
    },
    {
      "epoch": 1.23,
      "eval_loss": 0.11098004877567291,
      "eval_runtime": 151.5152,
      "eval_samples_per_second": 946.017,
      "eval_steps_per_second": 3.696,
      "step": 6200
    },
    {
      "epoch": 1.27,
      "eval_loss": 0.11082496494054794,
      "eval_runtime": 151.4327,
      "eval_samples_per_second": 946.533,
      "eval_steps_per_second": 3.698,
      "step": 6400
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.775153800357214e-05,
      "loss": 0.1132,
      "step": 6500
    },
    {
      "epoch": 1.31,
      "eval_loss": 0.11074619740247726,
      "eval_runtime": 151.2307,
      "eval_samples_per_second": 947.797,
      "eval_steps_per_second": 3.703,
      "step": 6600
    },
    {
      "epoch": 1.35,
      "eval_loss": 0.11220254749059677,
      "eval_runtime": 151.2796,
      "eval_samples_per_second": 947.491,
      "eval_steps_per_second": 3.702,
      "step": 6800
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.5270887080769993e-05,
      "loss": 0.2039,
      "step": 7000
    },
    {
      "epoch": 1.39,
      "eval_loss": 0.11101505160331726,
      "eval_runtime": 151.22,
      "eval_samples_per_second": 947.864,
      "eval_steps_per_second": 3.703,
      "step": 7000
    },
    {
      "epoch": 1.43,
      "eval_loss": 0.11076509952545166,
      "eval_runtime": 151.1874,
      "eval_samples_per_second": 948.068,
      "eval_steps_per_second": 3.704,
      "step": 7200
    },
    {
      "epoch": 1.47,
      "eval_loss": 0.11064371466636658,
      "eval_runtime": 151.2298,
      "eval_samples_per_second": 947.803,
      "eval_steps_per_second": 3.703,
      "step": 7400
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.2790236157967852e-05,
      "loss": 0.1107,
      "step": 7500
    },
    {
      "epoch": 1.51,
      "eval_loss": 0.1105843037366867,
      "eval_runtime": 151.246,
      "eval_samples_per_second": 947.701,
      "eval_steps_per_second": 3.703,
      "step": 7600
    },
    {
      "epoch": 1.55,
      "eval_loss": 0.11054880172014236,
      "eval_runtime": 151.2949,
      "eval_samples_per_second": 947.395,
      "eval_steps_per_second": 3.701,
      "step": 7800
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.0309585235165709e-05,
      "loss": 0.1115,
      "step": 8000
    },
    {
      "epoch": 1.59,
      "eval_loss": 0.11044388264417648,
      "eval_runtime": 151.2161,
      "eval_samples_per_second": 947.888,
      "eval_steps_per_second": 3.703,
      "step": 8000
    },
    {
      "epoch": 1.63,
      "eval_loss": 0.11042939871549606,
      "eval_runtime": 151.2469,
      "eval_samples_per_second": 947.695,
      "eval_steps_per_second": 3.703,
      "step": 8200
    },
    {
      "epoch": 1.67,
      "eval_loss": 0.11039307713508606,
      "eval_runtime": 151.2801,
      "eval_samples_per_second": 947.487,
      "eval_steps_per_second": 3.702,
      "step": 8400
    },
    {
      "epoch": 1.69,
      "learning_rate": 7.828934312363564e-06,
      "loss": 0.1106,
      "step": 8500
    },
    {
      "epoch": 1.71,
      "eval_loss": 0.1104336529970169,
      "eval_runtime": 151.2744,
      "eval_samples_per_second": 947.523,
      "eval_steps_per_second": 3.702,
      "step": 8600
    },
    {
      "epoch": 1.75,
      "eval_loss": 0.11029984056949615,
      "eval_runtime": 151.1978,
      "eval_samples_per_second": 948.003,
      "eval_steps_per_second": 3.704,
      "step": 8800
    },
    {
      "epoch": 1.79,
      "learning_rate": 5.348283389561421e-06,
      "loss": 0.1092,
      "step": 9000
    },
    {
      "epoch": 1.79,
      "eval_loss": 0.11028754711151123,
      "eval_runtime": 151.3426,
      "eval_samples_per_second": 947.096,
      "eval_steps_per_second": 3.7,
      "step": 9000
    },
    {
      "epoch": 1.83,
      "eval_loss": 0.11026974767446518,
      "eval_runtime": 151.187,
      "eval_samples_per_second": 948.071,
      "eval_steps_per_second": 3.704,
      "step": 9200
    },
    {
      "epoch": 1.87,
      "eval_loss": 0.11021895706653595,
      "eval_runtime": 151.1765,
      "eval_samples_per_second": 948.137,
      "eval_steps_per_second": 3.704,
      "step": 9400
    },
    {
      "epoch": 1.89,
      "learning_rate": 2.867632466759278e-06,
      "loss": 0.111,
      "step": 9500
    },
    {
      "epoch": 1.91,
      "eval_loss": 0.11018586158752441,
      "eval_runtime": 151.3213,
      "eval_samples_per_second": 947.229,
      "eval_steps_per_second": 3.701,
      "step": 9600
    },
    {
      "epoch": 1.94,
      "eval_loss": 0.11020087450742722,
      "eval_runtime": 151.3699,
      "eval_samples_per_second": 946.925,
      "eval_steps_per_second": 3.7,
      "step": 9800
    },
    {
      "epoch": 1.98,
      "learning_rate": 3.869815439571344e-07,
      "loss": 0.1109,
      "step": 10000
    },
    {
      "epoch": 1.98,
      "eval_loss": 0.11018609255552292,
      "eval_runtime": 151.328,
      "eval_samples_per_second": 947.187,
      "eval_steps_per_second": 3.701,
      "step": 10000
    }
  ],
  "logging_steps": 500,
  "max_steps": 10078,
  "num_train_epochs": 2,
  "save_steps": 1000,
  "total_flos": 1.6722690048e+17,
  "trial_name": null,
  "trial_params": null
}