{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.8674136321195145,
  "eval_steps": 50,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "eval_loss": 0.1086791530251503,
      "eval_runtime": 3.1139,
      "eval_samples_per_second": 32.114,
      "eval_steps_per_second": 4.175,
      "step": 50
    },
    {
      "epoch": 0.09,
      "eval_loss": 0.10340642929077148,
      "eval_runtime": 3.1088,
      "eval_samples_per_second": 32.167,
      "eval_steps_per_second": 4.182,
      "step": 100
    },
    {
      "epoch": 0.14,
      "eval_loss": 0.09758836030960083,
      "eval_runtime": 3.1297,
      "eval_samples_per_second": 31.952,
      "eval_steps_per_second": 4.154,
      "step": 150
    },
    {
      "epoch": 0.19,
      "eval_loss": 0.09741945564746857,
      "eval_runtime": 3.1428,
      "eval_samples_per_second": 31.818,
      "eval_steps_per_second": 4.136,
      "step": 200
    },
    {
      "epoch": 0.23,
      "eval_loss": 0.09413551539182663,
      "eval_runtime": 3.1346,
      "eval_samples_per_second": 31.902,
      "eval_steps_per_second": 4.147,
      "step": 250
    },
    {
      "epoch": 0.28,
      "eval_loss": 0.09605029970407486,
      "eval_runtime": 3.1389,
      "eval_samples_per_second": 31.858,
      "eval_steps_per_second": 4.142,
      "step": 300
    },
    {
      "epoch": 0.33,
      "eval_loss": 0.0930955782532692,
      "eval_runtime": 3.1561,
      "eval_samples_per_second": 31.685,
      "eval_steps_per_second": 4.119,
      "step": 350
    },
    {
      "epoch": 0.37,
      "eval_loss": 0.0926249772310257,
      "eval_runtime": 3.1213,
      "eval_samples_per_second": 32.038,
      "eval_steps_per_second": 4.165,
      "step": 400
    },
    {
      "epoch": 0.42,
      "eval_loss": 0.0899224504828453,
      "eval_runtime": 3.1292,
      "eval_samples_per_second": 31.957,
      "eval_steps_per_second": 4.154,
      "step": 450
    },
    {
      "epoch": 0.47,
      "learning_rate": 8.75e-05,
      "loss": 0.1297,
      "step": 500
    },
    {
      "epoch": 0.47,
      "eval_loss": 0.0851517766714096,
      "eval_runtime": 3.1369,
      "eval_samples_per_second": 31.879,
      "eval_steps_per_second": 4.144,
      "step": 500
    },
    {
      "epoch": 0.51,
      "eval_loss": 0.08426132053136826,
      "eval_runtime": 3.1311,
      "eval_samples_per_second": 31.938,
      "eval_steps_per_second": 4.152,
      "step": 550
    },
    {
      "epoch": 0.56,
      "eval_loss": 0.0841633602976799,
      "eval_runtime": 3.1188,
      "eval_samples_per_second": 32.064,
      "eval_steps_per_second": 4.168,
      "step": 600
    },
    {
      "epoch": 0.61,
      "eval_loss": 0.08286924660205841,
      "eval_runtime": 3.1352,
      "eval_samples_per_second": 31.896,
      "eval_steps_per_second": 4.147,
      "step": 650
    },
    {
      "epoch": 0.65,
      "eval_loss": 0.08303254097700119,
      "eval_runtime": 3.1217,
      "eval_samples_per_second": 32.034,
      "eval_steps_per_second": 4.164,
      "step": 700
    },
    {
      "epoch": 0.7,
      "eval_loss": 0.08298823982477188,
      "eval_runtime": 3.152,
      "eval_samples_per_second": 31.726,
      "eval_steps_per_second": 4.124,
      "step": 750
    },
    {
      "epoch": 0.75,
      "eval_loss": 0.08352884650230408,
      "eval_runtime": 3.1051,
      "eval_samples_per_second": 32.205,
      "eval_steps_per_second": 4.187,
      "step": 800
    },
    {
      "epoch": 0.79,
      "eval_loss": 0.0866646021604538,
      "eval_runtime": 3.123,
      "eval_samples_per_second": 32.02,
      "eval_steps_per_second": 4.163,
      "step": 850
    },
    {
      "epoch": 0.84,
      "eval_loss": 0.08790910989046097,
      "eval_runtime": 3.1472,
      "eval_samples_per_second": 31.774,
      "eval_steps_per_second": 4.131,
      "step": 900
    },
    {
      "epoch": 0.89,
      "eval_loss": 0.0884072408080101,
      "eval_runtime": 3.1322,
      "eval_samples_per_second": 31.926,
      "eval_steps_per_second": 4.15,
      "step": 950
    },
    {
      "epoch": 0.93,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.1137,
      "step": 1000
    },
    {
      "epoch": 0.93,
      "eval_loss": 0.08399125933647156,
      "eval_runtime": 3.1237,
      "eval_samples_per_second": 32.013,
      "eval_steps_per_second": 4.162,
      "step": 1000
    },
    {
      "epoch": 0.98,
      "eval_loss": 0.08606167882680893,
      "eval_runtime": 3.1401,
      "eval_samples_per_second": 31.846,
      "eval_steps_per_second": 4.14,
      "step": 1050
    },
    {
      "epoch": 1.03,
      "eval_loss": 0.08386250585317612,
      "eval_runtime": 3.0992,
      "eval_samples_per_second": 32.267,
      "eval_steps_per_second": 4.195,
      "step": 1100
    },
    {
      "epoch": 1.07,
      "eval_loss": 0.08104114234447479,
      "eval_runtime": 3.1205,
      "eval_samples_per_second": 32.046,
      "eval_steps_per_second": 4.166,
      "step": 1150
    },
    {
      "epoch": 1.12,
      "eval_loss": 0.08137733489274979,
      "eval_runtime": 3.1344,
      "eval_samples_per_second": 31.904,
      "eval_steps_per_second": 4.148,
      "step": 1200
    },
    {
      "epoch": 1.17,
      "eval_loss": 0.08145874738693237,
      "eval_runtime": 3.1106,
      "eval_samples_per_second": 32.148,
      "eval_steps_per_second": 4.179,
      "step": 1250
    },
    {
      "epoch": 1.21,
      "eval_loss": 0.0812147930264473,
      "eval_runtime": 3.1172,
      "eval_samples_per_second": 32.08,
      "eval_steps_per_second": 4.17,
      "step": 1300
    },
    {
      "epoch": 1.26,
      "eval_loss": 0.0826956108212471,
      "eval_runtime": 3.1201,
      "eval_samples_per_second": 32.05,
      "eval_steps_per_second": 4.167,
      "step": 1350
    },
    {
      "epoch": 1.31,
      "eval_loss": 0.08307519555091858,
      "eval_runtime": 3.1328,
      "eval_samples_per_second": 31.92,
      "eval_steps_per_second": 4.15,
      "step": 1400
    },
    {
      "epoch": 1.35,
      "eval_loss": 0.07891510426998138,
      "eval_runtime": 3.1304,
      "eval_samples_per_second": 31.945,
      "eval_steps_per_second": 4.153,
      "step": 1450
    },
    {
      "epoch": 1.4,
      "learning_rate": 6.25e-05,
      "loss": 0.0976,
      "step": 1500
    },
    {
      "epoch": 1.4,
      "eval_loss": 0.07503539323806763,
      "eval_runtime": 3.1408,
      "eval_samples_per_second": 31.839,
      "eval_steps_per_second": 4.139,
      "step": 1500
    },
    {
      "epoch": 1.45,
      "eval_loss": 0.07038568705320358,
      "eval_runtime": 3.1598,
      "eval_samples_per_second": 31.647,
      "eval_steps_per_second": 4.114,
      "step": 1550
    },
    {
      "epoch": 1.49,
      "eval_loss": 0.07097797840833664,
      "eval_runtime": 3.1403,
      "eval_samples_per_second": 31.844,
      "eval_steps_per_second": 4.14,
      "step": 1600
    },
    {
      "epoch": 1.54,
      "eval_loss": 0.07105908542871475,
      "eval_runtime": 3.1203,
      "eval_samples_per_second": 32.048,
      "eval_steps_per_second": 4.166,
      "step": 1650
    },
    {
      "epoch": 1.59,
      "eval_loss": 0.06974442303180695,
      "eval_runtime": 3.1279,
      "eval_samples_per_second": 31.97,
      "eval_steps_per_second": 4.156,
      "step": 1700
    },
    {
      "epoch": 1.63,
      "eval_loss": 0.07010236382484436,
      "eval_runtime": 3.1368,
      "eval_samples_per_second": 31.88,
      "eval_steps_per_second": 4.144,
      "step": 1750
    },
    {
      "epoch": 1.68,
      "eval_loss": 0.06847620010375977,
      "eval_runtime": 3.1578,
      "eval_samples_per_second": 31.667,
      "eval_steps_per_second": 4.117,
      "step": 1800
    },
    {
      "epoch": 1.73,
      "eval_loss": 0.06789014488458633,
      "eval_runtime": 3.1274,
      "eval_samples_per_second": 31.975,
      "eval_steps_per_second": 4.157,
      "step": 1850
    },
    {
      "epoch": 1.77,
      "eval_loss": 0.06896524131298065,
      "eval_runtime": 3.1488,
      "eval_samples_per_second": 31.758,
      "eval_steps_per_second": 4.129,
      "step": 1900
    },
    {
      "epoch": 1.82,
      "eval_loss": 0.06796181946992874,
      "eval_runtime": 3.1466,
      "eval_samples_per_second": 31.78,
      "eval_steps_per_second": 4.131,
      "step": 1950
    },
    {
      "epoch": 1.87,
      "learning_rate": 5e-05,
      "loss": 0.0908,
      "step": 2000
    },
    {
      "epoch": 1.87,
      "eval_loss": 0.06683148443698883,
      "eval_runtime": 3.129,
      "eval_samples_per_second": 31.959,
      "eval_steps_per_second": 4.155,
      "step": 2000
    }
  ],
  "logging_steps": 500,
  "max_steps": 4000,
  "num_train_epochs": 4,
  "save_steps": 1000,
  "total_flos": 2.738687031115776e+16,
  "trial_name": null,
  "trial_params": null
}