{
  "best_metric": 1.2130030393600464,
  "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": 2.1907,
      "step": 100
    },
    {
      "epoch": 0.07,
      "eval_loss": 2.0242879390716553,
      "eval_runtime": 144.9414,
      "eval_samples_per_second": 43.286,
      "eval_steps_per_second": 5.416,
      "step": 100
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 1.9738,
      "step": 200
    },
    {
      "epoch": 0.15,
      "eval_loss": 1.9313020706176758,
      "eval_runtime": 143.9274,
      "eval_samples_per_second": 43.591,
      "eval_steps_per_second": 5.454,
      "step": 200
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 1.9003,
      "step": 300
    },
    {
      "epoch": 0.22,
      "eval_loss": 1.8709847927093506,
      "eval_runtime": 144.0274,
      "eval_samples_per_second": 43.561,
      "eval_steps_per_second": 5.45,
      "step": 300
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 1.849,
      "step": 400
    },
    {
      "epoch": 0.29,
      "eval_loss": 1.8261810541152954,
      "eval_runtime": 144.0251,
      "eval_samples_per_second": 43.562,
      "eval_steps_per_second": 5.45,
      "step": 400
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 1.7978,
      "step": 500
    },
    {
      "epoch": 0.36,
      "eval_loss": 1.783812403678894,
      "eval_runtime": 143.9209,
      "eval_samples_per_second": 43.593,
      "eval_steps_per_second": 5.454,
      "step": 500
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 1.7696,
      "step": 600
    },
    {
      "epoch": 0.44,
      "eval_loss": 1.7559951543807983,
      "eval_runtime": 143.9146,
      "eval_samples_per_second": 43.595,
      "eval_steps_per_second": 5.455,
      "step": 600
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 1.7308,
      "step": 700
    },
    {
      "epoch": 0.51,
      "eval_loss": 1.7238949537277222,
      "eval_runtime": 143.9925,
      "eval_samples_per_second": 43.572,
      "eval_steps_per_second": 5.452,
      "step": 700
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 1.7094,
      "step": 800
    },
    {
      "epoch": 0.58,
      "eval_loss": 1.6951390504837036,
      "eval_runtime": 143.9787,
      "eval_samples_per_second": 43.576,
      "eval_steps_per_second": 5.452,
      "step": 800
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002,
      "loss": 1.674,
      "step": 900
    },
    {
      "epoch": 0.66,
      "eval_loss": 1.6701356172561646,
      "eval_runtime": 144.0222,
      "eval_samples_per_second": 43.563,
      "eval_steps_per_second": 5.451,
      "step": 900
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 1.6589,
      "step": 1000
    },
    {
      "epoch": 0.73,
      "eval_loss": 1.6467818021774292,
      "eval_runtime": 143.9955,
      "eval_samples_per_second": 43.571,
      "eval_steps_per_second": 5.452,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 1.6385,
      "step": 1100
    },
    {
      "epoch": 0.8,
      "eval_loss": 1.6242790222167969,
      "eval_runtime": 144.0793,
      "eval_samples_per_second": 43.545,
      "eval_steps_per_second": 5.448,
      "step": 1100
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 1.6033,
      "step": 1200
    },
    {
      "epoch": 0.87,
      "eval_loss": 1.6063916683197021,
      "eval_runtime": 144.0465,
      "eval_samples_per_second": 43.555,
      "eval_steps_per_second": 5.45,
      "step": 1200
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002,
      "loss": 1.6104,
      "step": 1300
    },
    {
      "epoch": 0.95,
      "eval_loss": 1.582713007926941,
      "eval_runtime": 144.0533,
      "eval_samples_per_second": 43.553,
      "eval_steps_per_second": 5.449,
      "step": 1300
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 1.5638,
      "step": 1400
    },
    {
      "epoch": 1.02,
      "eval_loss": 1.5626935958862305,
      "eval_runtime": 143.9538,
      "eval_samples_per_second": 43.583,
      "eval_steps_per_second": 5.453,
      "step": 1400
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002,
      "loss": 1.5362,
      "step": 1500
    },
    {
      "epoch": 1.09,
      "eval_loss": 1.5464439392089844,
      "eval_runtime": 143.8789,
      "eval_samples_per_second": 43.606,
      "eval_steps_per_second": 5.456,
      "step": 1500
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 1.5169,
      "step": 1600
    },
    {
      "epoch": 1.17,
      "eval_loss": 1.529025673866272,
      "eval_runtime": 143.9243,
      "eval_samples_per_second": 43.592,
      "eval_steps_per_second": 5.454,
      "step": 1600
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002,
      "loss": 1.5148,
      "step": 1700
    },
    {
      "epoch": 1.24,
      "eval_loss": 1.511838436126709,
      "eval_runtime": 143.9329,
      "eval_samples_per_second": 43.59,
      "eval_steps_per_second": 5.454,
      "step": 1700
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002,
      "loss": 1.4882,
      "step": 1800
    },
    {
      "epoch": 1.31,
      "eval_loss": 1.4979889392852783,
      "eval_runtime": 143.9162,
      "eval_samples_per_second": 43.595,
      "eval_steps_per_second": 5.455,
      "step": 1800
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 1.4777,
      "step": 1900
    },
    {
      "epoch": 1.38,
      "eval_loss": 1.4818388223648071,
      "eval_runtime": 143.9405,
      "eval_samples_per_second": 43.587,
      "eval_steps_per_second": 5.454,
      "step": 1900
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0002,
      "loss": 1.4608,
      "step": 2000
    },
    {
      "epoch": 1.46,
      "eval_loss": 1.4687143564224243,
      "eval_runtime": 143.8741,
      "eval_samples_per_second": 43.608,
      "eval_steps_per_second": 5.456,
      "step": 2000
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 1.4461,
      "step": 2100
    },
    {
      "epoch": 1.53,
      "eval_loss": 1.452826738357544,
      "eval_runtime": 143.8579,
      "eval_samples_per_second": 43.612,
      "eval_steps_per_second": 5.457,
      "step": 2100
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 1.421,
      "step": 2200
    },
    {
      "epoch": 1.6,
      "eval_loss": 1.4381742477416992,
      "eval_runtime": 143.8526,
      "eval_samples_per_second": 43.614,
      "eval_steps_per_second": 5.457,
      "step": 2200
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002,
      "loss": 1.42,
      "step": 2300
    },
    {
      "epoch": 1.68,
      "eval_loss": 1.4257140159606934,
      "eval_runtime": 143.9412,
      "eval_samples_per_second": 43.587,
      "eval_steps_per_second": 5.454,
      "step": 2300
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0002,
      "loss": 1.4195,
      "step": 2400
    },
    {
      "epoch": 1.75,
      "eval_loss": 1.4125311374664307,
      "eval_runtime": 143.8958,
      "eval_samples_per_second": 43.601,
      "eval_steps_per_second": 5.455,
      "step": 2400
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 1.3915,
      "step": 2500
    },
    {
      "epoch": 1.82,
      "eval_loss": 1.3995251655578613,
      "eval_runtime": 143.9086,
      "eval_samples_per_second": 43.597,
      "eval_steps_per_second": 5.455,
      "step": 2500
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0002,
      "loss": 1.3804,
      "step": 2600
    },
    {
      "epoch": 1.89,
      "eval_loss": 1.38627290725708,
      "eval_runtime": 143.9593,
      "eval_samples_per_second": 43.582,
      "eval_steps_per_second": 5.453,
      "step": 2600
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0002,
      "loss": 1.3689,
      "step": 2700
    },
    {
      "epoch": 1.97,
      "eval_loss": 1.3713726997375488,
      "eval_runtime": 143.9492,
      "eval_samples_per_second": 43.585,
      "eval_steps_per_second": 5.453,
      "step": 2700
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0002,
      "loss": 1.3402,
      "step": 2800
    },
    {
      "epoch": 2.04,
      "eval_loss": 1.3606998920440674,
      "eval_runtime": 144.0067,
      "eval_samples_per_second": 43.567,
      "eval_steps_per_second": 5.451,
      "step": 2800
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0002,
      "loss": 1.3149,
      "step": 2900
    },
    {
      "epoch": 2.11,
      "eval_loss": 1.346933126449585,
      "eval_runtime": 143.8935,
      "eval_samples_per_second": 43.602,
      "eval_steps_per_second": 5.455,
      "step": 2900
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0002,
      "loss": 1.315,
      "step": 3000
    },
    {
      "epoch": 2.19,
      "eval_loss": 1.3364778757095337,
      "eval_runtime": 144.0179,
      "eval_samples_per_second": 43.564,
      "eval_steps_per_second": 5.451,
      "step": 3000
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0002,
      "loss": 1.2945,
      "step": 3100
    },
    {
      "epoch": 2.26,
      "eval_loss": 1.3254159688949585,
      "eval_runtime": 143.9738,
      "eval_samples_per_second": 43.577,
      "eval_steps_per_second": 5.452,
      "step": 3100
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0002,
      "loss": 1.2991,
      "step": 3200
    },
    {
      "epoch": 2.33,
      "eval_loss": 1.3148905038833618,
      "eval_runtime": 147.7174,
      "eval_samples_per_second": 42.473,
      "eval_steps_per_second": 5.314,
      "step": 3200
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0002,
      "loss": 1.28,
      "step": 3300
    },
    {
      "epoch": 2.4,
      "eval_loss": 1.3021882772445679,
      "eval_runtime": 143.959,
      "eval_samples_per_second": 43.582,
      "eval_steps_per_second": 5.453,
      "step": 3300
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0002,
      "loss": 1.2619,
      "step": 3400
    },
    {
      "epoch": 2.48,
      "eval_loss": 1.2918113470077515,
      "eval_runtime": 143.9692,
      "eval_samples_per_second": 43.579,
      "eval_steps_per_second": 5.453,
      "step": 3400
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0002,
      "loss": 1.2629,
      "step": 3500
    },
    {
      "epoch": 2.55,
      "eval_loss": 1.2808809280395508,
      "eval_runtime": 143.992,
      "eval_samples_per_second": 43.572,
      "eval_steps_per_second": 5.452,
      "step": 3500
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0002,
      "loss": 1.2515,
      "step": 3600
    },
    {
      "epoch": 2.62,
      "eval_loss": 1.2679202556610107,
      "eval_runtime": 144.0312,
      "eval_samples_per_second": 43.56,
      "eval_steps_per_second": 5.45,
      "step": 3600
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0002,
      "loss": 1.2346,
      "step": 3700
    },
    {
      "epoch": 2.7,
      "eval_loss": 1.255784034729004,
      "eval_runtime": 144.0435,
      "eval_samples_per_second": 43.556,
      "eval_steps_per_second": 5.45,
      "step": 3700
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0002,
      "loss": 1.2112,
      "step": 3800
    },
    {
      "epoch": 2.77,
      "eval_loss": 1.2465852499008179,
      "eval_runtime": 144.0134,
      "eval_samples_per_second": 43.565,
      "eval_steps_per_second": 5.451,
      "step": 3800
    },
    {
      "epoch": 2.84,
      "learning_rate": 0.0002,
      "loss": 1.221,
      "step": 3900
    },
    {
      "epoch": 2.84,
      "eval_loss": 1.2340775728225708,
      "eval_runtime": 144.91,
      "eval_samples_per_second": 43.296,
      "eval_steps_per_second": 5.417,
      "step": 3900
    },
    {
      "epoch": 2.91,
      "learning_rate": 0.0002,
      "loss": 1.2027,
      "step": 4000
    },
    {
      "epoch": 2.91,
      "eval_loss": 1.2261732816696167,
      "eval_runtime": 143.9516,
      "eval_samples_per_second": 43.584,
      "eval_steps_per_second": 5.453,
      "step": 4000
    },
    {
      "epoch": 2.99,
      "learning_rate": 0.0002,
      "loss": 1.2013,
      "step": 4100
    },
    {
      "epoch": 2.99,
      "eval_loss": 1.2130030393600464,
      "eval_runtime": 143.9131,
      "eval_samples_per_second": 43.596,
      "eval_steps_per_second": 5.455,
      "step": 4100
    }
  ],
  "logging_steps": 100,
  "max_steps": 4116,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 2.4438234279579648e+17,
  "trial_name": null,
  "trial_params": null
}