{
  "best_metric": 1.9783430099487305,
  "best_model_checkpoint": "./outputs/checkpoint-4100",
  "epoch": 2.987249544626594,
  "eval_steps": 100,
  "global_step": 4100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 2.7403,
      "step": 100
    },
    {
      "epoch": 0.07,
      "eval_loss": 2.642388343811035,
      "eval_runtime": 205.2517,
      "eval_samples_per_second": 30.567,
      "eval_steps_per_second": 3.825,
      "step": 100
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 2.606,
      "step": 200
    },
    {
      "epoch": 0.15,
      "eval_loss": 2.5923426151275635,
      "eval_runtime": 208.3105,
      "eval_samples_per_second": 30.119,
      "eval_steps_per_second": 3.768,
      "step": 200
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 2.5632,
      "step": 300
    },
    {
      "epoch": 0.22,
      "eval_loss": 2.5577170848846436,
      "eval_runtime": 205.6257,
      "eval_samples_per_second": 30.512,
      "eval_steps_per_second": 3.818,
      "step": 300
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 2.5369,
      "step": 400
    },
    {
      "epoch": 0.29,
      "eval_loss": 2.5247888565063477,
      "eval_runtime": 205.7237,
      "eval_samples_per_second": 30.497,
      "eval_steps_per_second": 3.816,
      "step": 400
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 2.4954,
      "step": 500
    },
    {
      "epoch": 0.36,
      "eval_loss": 2.498934745788574,
      "eval_runtime": 207.1055,
      "eval_samples_per_second": 30.294,
      "eval_steps_per_second": 3.79,
      "step": 500
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 2.469,
      "step": 600
    },
    {
      "epoch": 0.44,
      "eval_loss": 2.470606565475464,
      "eval_runtime": 205.8606,
      "eval_samples_per_second": 30.477,
      "eval_steps_per_second": 3.813,
      "step": 600
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 2.4509,
      "step": 700
    },
    {
      "epoch": 0.51,
      "eval_loss": 2.446812391281128,
      "eval_runtime": 205.2079,
      "eval_samples_per_second": 30.574,
      "eval_steps_per_second": 3.825,
      "step": 700
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 2.4284,
      "step": 800
    },
    {
      "epoch": 0.58,
      "eval_loss": 2.42557954788208,
      "eval_runtime": 205.3478,
      "eval_samples_per_second": 30.553,
      "eval_steps_per_second": 3.823,
      "step": 800
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002,
      "loss": 2.3916,
      "step": 900
    },
    {
      "epoch": 0.66,
      "eval_loss": 2.403062343597412,
      "eval_runtime": 204.8842,
      "eval_samples_per_second": 30.622,
      "eval_steps_per_second": 3.831,
      "step": 900
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 2.3883,
      "step": 1000
    },
    {
      "epoch": 0.73,
      "eval_loss": 2.384012222290039,
      "eval_runtime": 204.4165,
      "eval_samples_per_second": 30.692,
      "eval_steps_per_second": 3.84,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 2.382,
      "step": 1100
    },
    {
      "epoch": 0.8,
      "eval_loss": 2.361379384994507,
      "eval_runtime": 204.6076,
      "eval_samples_per_second": 30.664,
      "eval_steps_per_second": 3.837,
      "step": 1100
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 2.3372,
      "step": 1200
    },
    {
      "epoch": 0.87,
      "eval_loss": 2.3411779403686523,
      "eval_runtime": 204.8024,
      "eval_samples_per_second": 30.634,
      "eval_steps_per_second": 3.833,
      "step": 1200
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002,
      "loss": 2.3496,
      "step": 1300
    },
    {
      "epoch": 0.95,
      "eval_loss": 2.3248472213745117,
      "eval_runtime": 204.1277,
      "eval_samples_per_second": 30.736,
      "eval_steps_per_second": 3.846,
      "step": 1300
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 2.3001,
      "step": 1400
    },
    {
      "epoch": 1.02,
      "eval_loss": 2.301698684692383,
      "eval_runtime": 204.3213,
      "eval_samples_per_second": 30.707,
      "eval_steps_per_second": 3.842,
      "step": 1400
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002,
      "loss": 2.2598,
      "step": 1500
    },
    {
      "epoch": 1.09,
      "eval_loss": 2.284607172012329,
      "eval_runtime": 204.5306,
      "eval_samples_per_second": 30.675,
      "eval_steps_per_second": 3.838,
      "step": 1500
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 2.2518,
      "step": 1600
    },
    {
      "epoch": 1.17,
      "eval_loss": 2.2680861949920654,
      "eval_runtime": 204.3902,
      "eval_samples_per_second": 30.696,
      "eval_steps_per_second": 3.841,
      "step": 1600
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002,
      "loss": 2.2518,
      "step": 1700
    },
    {
      "epoch": 1.24,
      "eval_loss": 2.2509853839874268,
      "eval_runtime": 204.5985,
      "eval_samples_per_second": 30.665,
      "eval_steps_per_second": 3.837,
      "step": 1700
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002,
      "loss": 2.2139,
      "step": 1800
    },
    {
      "epoch": 1.31,
      "eval_loss": 2.2367281913757324,
      "eval_runtime": 206.3249,
      "eval_samples_per_second": 30.408,
      "eval_steps_per_second": 3.805,
      "step": 1800
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 2.1986,
      "step": 1900
    },
    {
      "epoch": 1.38,
      "eval_loss": 2.222033977508545,
      "eval_runtime": 205.0156,
      "eval_samples_per_second": 30.603,
      "eval_steps_per_second": 3.829,
      "step": 1900
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0002,
      "loss": 2.1927,
      "step": 2000
    },
    {
      "epoch": 1.46,
      "eval_loss": 2.2064921855926514,
      "eval_runtime": 205.4048,
      "eval_samples_per_second": 30.545,
      "eval_steps_per_second": 3.822,
      "step": 2000
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 2.1762,
      "step": 2100
    },
    {
      "epoch": 1.53,
      "eval_loss": 2.1937856674194336,
      "eval_runtime": 204.8721,
      "eval_samples_per_second": 30.624,
      "eval_steps_per_second": 3.832,
      "step": 2100
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 2.1584,
      "step": 2200
    },
    {
      "epoch": 1.6,
      "eval_loss": 2.1781294345855713,
      "eval_runtime": 204.973,
      "eval_samples_per_second": 30.609,
      "eval_steps_per_second": 3.83,
      "step": 2200
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002,
      "loss": 2.1481,
      "step": 2300
    },
    {
      "epoch": 1.68,
      "eval_loss": 2.1652047634124756,
      "eval_runtime": 204.8186,
      "eval_samples_per_second": 30.632,
      "eval_steps_per_second": 3.833,
      "step": 2300
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0002,
      "loss": 2.1553,
      "step": 2400
    },
    {
      "epoch": 1.75,
      "eval_loss": 2.1539299488067627,
      "eval_runtime": 204.9017,
      "eval_samples_per_second": 30.62,
      "eval_steps_per_second": 3.831,
      "step": 2400
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 2.1242,
      "step": 2500
    },
    {
      "epoch": 1.82,
      "eval_loss": 2.140552520751953,
      "eval_runtime": 204.6745,
      "eval_samples_per_second": 30.654,
      "eval_steps_per_second": 3.835,
      "step": 2500
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0002,
      "loss": 2.1241,
      "step": 2600
    },
    {
      "epoch": 1.89,
      "eval_loss": 2.129112482070923,
      "eval_runtime": 204.7891,
      "eval_samples_per_second": 30.636,
      "eval_steps_per_second": 3.833,
      "step": 2600
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0002,
      "loss": 2.108,
      "step": 2700
    },
    {
      "epoch": 1.97,
      "eval_loss": 2.1130168437957764,
      "eval_runtime": 204.8827,
      "eval_samples_per_second": 30.622,
      "eval_steps_per_second": 3.831,
      "step": 2700
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0002,
      "loss": 2.0624,
      "step": 2800
    },
    {
      "epoch": 2.04,
      "eval_loss": 2.1079933643341064,
      "eval_runtime": 205.1262,
      "eval_samples_per_second": 30.586,
      "eval_steps_per_second": 3.827,
      "step": 2800
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0002,
      "loss": 2.0368,
      "step": 2900
    },
    {
      "epoch": 2.11,
      "eval_loss": 2.0953664779663086,
      "eval_runtime": 204.7517,
      "eval_samples_per_second": 30.642,
      "eval_steps_per_second": 3.834,
      "step": 2900
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0002,
      "loss": 2.0563,
      "step": 3000
    },
    {
      "epoch": 2.19,
      "eval_loss": 2.0847907066345215,
      "eval_runtime": 205.0373,
      "eval_samples_per_second": 30.599,
      "eval_steps_per_second": 3.829,
      "step": 3000
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0002,
      "loss": 2.0254,
      "step": 3100
    },
    {
      "epoch": 2.26,
      "eval_loss": 2.0736167430877686,
      "eval_runtime": 205.2299,
      "eval_samples_per_second": 30.571,
      "eval_steps_per_second": 3.825,
      "step": 3100
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0002,
      "loss": 2.0397,
      "step": 3200
    },
    {
      "epoch": 2.33,
      "eval_loss": 2.0629310607910156,
      "eval_runtime": 205.2704,
      "eval_samples_per_second": 30.565,
      "eval_steps_per_second": 3.824,
      "step": 3200
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0002,
      "loss": 2.0083,
      "step": 3300
    },
    {
      "epoch": 2.4,
      "eval_loss": 2.0515356063842773,
      "eval_runtime": 205.1563,
      "eval_samples_per_second": 30.582,
      "eval_steps_per_second": 3.826,
      "step": 3300
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0002,
      "loss": 2.0018,
      "step": 3400
    },
    {
      "epoch": 2.48,
      "eval_loss": 2.0455892086029053,
      "eval_runtime": 280.8694,
      "eval_samples_per_second": 22.338,
      "eval_steps_per_second": 2.795,
      "step": 3400
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0002,
      "loss": 2.0001,
      "step": 3500
    },
    {
      "epoch": 2.55,
      "eval_loss": 2.0334508419036865,
      "eval_runtime": 220.5599,
      "eval_samples_per_second": 28.446,
      "eval_steps_per_second": 3.559,
      "step": 3500
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0002,
      "loss": 1.9962,
      "step": 3600
    },
    {
      "epoch": 2.62,
      "eval_loss": 2.0254855155944824,
      "eval_runtime": 204.9938,
      "eval_samples_per_second": 30.606,
      "eval_steps_per_second": 3.829,
      "step": 3600
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0002,
      "loss": 1.9767,
      "step": 3700
    },
    {
      "epoch": 2.7,
      "eval_loss": 2.015181303024292,
      "eval_runtime": 205.3577,
      "eval_samples_per_second": 30.552,
      "eval_steps_per_second": 3.823,
      "step": 3700
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0002,
      "loss": 1.9679,
      "step": 3800
    },
    {
      "epoch": 2.77,
      "eval_loss": 2.00793194770813,
      "eval_runtime": 205.2214,
      "eval_samples_per_second": 30.572,
      "eval_steps_per_second": 3.825,
      "step": 3800
    },
    {
      "epoch": 2.84,
      "learning_rate": 0.0002,
      "loss": 1.9555,
      "step": 3900
    },
    {
      "epoch": 2.84,
      "eval_loss": 1.9973456859588623,
      "eval_runtime": 205.4846,
      "eval_samples_per_second": 30.533,
      "eval_steps_per_second": 3.82,
      "step": 3900
    },
    {
      "epoch": 2.91,
      "learning_rate": 0.0002,
      "loss": 1.9549,
      "step": 4000
    },
    {
      "epoch": 2.91,
      "eval_loss": 1.990099549293518,
      "eval_runtime": 205.2933,
      "eval_samples_per_second": 30.561,
      "eval_steps_per_second": 3.824,
      "step": 4000
    },
    {
      "epoch": 2.99,
      "learning_rate": 0.0002,
      "loss": 1.9559,
      "step": 4100
    },
    {
      "epoch": 2.99,
      "eval_loss": 1.9783430099487305,
      "eval_runtime": 205.0037,
      "eval_samples_per_second": 30.604,
      "eval_steps_per_second": 3.829,
      "step": 4100
    }
  ],
  "logging_steps": 100,
  "max_steps": 4116,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 1.1959366338920448e+17,
  "trial_name": null,
  "trial_params": null
}