{
  "best_metric": 1.116302251815796,
  "best_model_checkpoint": "./outputs/checkpoint-3800",
  "epoch": 2.769398907103825,
  "eval_steps": 100,
  "global_step": 3800,
  "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": 1.765,
      "step": 100
    },
    {
      "epoch": 0.07,
      "eval_loss": 1.631608486175537,
      "eval_runtime": 430.1354,
      "eval_samples_per_second": 14.586,
      "eval_steps_per_second": 1.825,
      "step": 100
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 1.6077,
      "step": 200
    },
    {
      "epoch": 0.15,
      "eval_loss": 1.5870035886764526,
      "eval_runtime": 417.8578,
      "eval_samples_per_second": 15.015,
      "eval_steps_per_second": 1.879,
      "step": 200
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 1.5755,
      "step": 300
    },
    {
      "epoch": 0.22,
      "eval_loss": 1.5563029050827026,
      "eval_runtime": 417.8787,
      "eval_samples_per_second": 15.014,
      "eval_steps_per_second": 1.879,
      "step": 300
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 1.5445,
      "step": 400
    },
    {
      "epoch": 0.29,
      "eval_loss": 1.5300686359405518,
      "eval_runtime": 417.9846,
      "eval_samples_per_second": 15.01,
      "eval_steps_per_second": 1.878,
      "step": 400
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 1.5114,
      "step": 500
    },
    {
      "epoch": 0.36,
      "eval_loss": 1.5074832439422607,
      "eval_runtime": 417.962,
      "eval_samples_per_second": 15.011,
      "eval_steps_per_second": 1.878,
      "step": 500
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 1.4946,
      "step": 600
    },
    {
      "epoch": 0.44,
      "eval_loss": 1.4870833158493042,
      "eval_runtime": 417.6568,
      "eval_samples_per_second": 15.022,
      "eval_steps_per_second": 1.88,
      "step": 600
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 1.472,
      "step": 700
    },
    {
      "epoch": 0.51,
      "eval_loss": 1.4683642387390137,
      "eval_runtime": 417.7114,
      "eval_samples_per_second": 15.02,
      "eval_steps_per_second": 1.879,
      "step": 700
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 1.4622,
      "step": 800
    },
    {
      "epoch": 0.58,
      "eval_loss": 1.4501450061798096,
      "eval_runtime": 417.7046,
      "eval_samples_per_second": 15.02,
      "eval_steps_per_second": 1.879,
      "step": 800
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002,
      "loss": 1.433,
      "step": 900
    },
    {
      "epoch": 0.66,
      "eval_loss": 1.4323909282684326,
      "eval_runtime": 417.8696,
      "eval_samples_per_second": 15.014,
      "eval_steps_per_second": 1.879,
      "step": 900
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 1.4268,
      "step": 1000
    },
    {
      "epoch": 0.73,
      "eval_loss": 1.4162870645523071,
      "eval_runtime": 417.7667,
      "eval_samples_per_second": 15.018,
      "eval_steps_per_second": 1.879,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 1.4125,
      "step": 1100
    },
    {
      "epoch": 0.8,
      "eval_loss": 1.4018020629882812,
      "eval_runtime": 417.5926,
      "eval_samples_per_second": 15.024,
      "eval_steps_per_second": 1.88,
      "step": 1100
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 1.3846,
      "step": 1200
    },
    {
      "epoch": 0.87,
      "eval_loss": 1.3857518434524536,
      "eval_runtime": 417.6267,
      "eval_samples_per_second": 15.023,
      "eval_steps_per_second": 1.88,
      "step": 1200
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002,
      "loss": 1.3933,
      "step": 1300
    },
    {
      "epoch": 0.95,
      "eval_loss": 1.3718016147613525,
      "eval_runtime": 417.2876,
      "eval_samples_per_second": 15.035,
      "eval_steps_per_second": 1.881,
      "step": 1300
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 1.3556,
      "step": 1400
    },
    {
      "epoch": 1.02,
      "eval_loss": 1.3577570915222168,
      "eval_runtime": 417.5078,
      "eval_samples_per_second": 15.027,
      "eval_steps_per_second": 1.88,
      "step": 1400
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002,
      "loss": 1.3316,
      "step": 1500
    },
    {
      "epoch": 1.09,
      "eval_loss": 1.3446884155273438,
      "eval_runtime": 417.3899,
      "eval_samples_per_second": 15.032,
      "eval_steps_per_second": 1.881,
      "step": 1500
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 1.3207,
      "step": 1600
    },
    {
      "epoch": 1.17,
      "eval_loss": 1.3314917087554932,
      "eval_runtime": 417.7717,
      "eval_samples_per_second": 15.018,
      "eval_steps_per_second": 1.879,
      "step": 1600
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002,
      "loss": 1.3212,
      "step": 1700
    },
    {
      "epoch": 1.24,
      "eval_loss": 1.318053960800171,
      "eval_runtime": 417.56,
      "eval_samples_per_second": 15.025,
      "eval_steps_per_second": 1.88,
      "step": 1700
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002,
      "loss": 1.2967,
      "step": 1800
    },
    {
      "epoch": 1.31,
      "eval_loss": 1.3063520193099976,
      "eval_runtime": 417.5913,
      "eval_samples_per_second": 15.024,
      "eval_steps_per_second": 1.88,
      "step": 1800
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 1.2883,
      "step": 1900
    },
    {
      "epoch": 1.38,
      "eval_loss": 1.2934125661849976,
      "eval_runtime": 417.7511,
      "eval_samples_per_second": 15.019,
      "eval_steps_per_second": 1.879,
      "step": 1900
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0002,
      "loss": 1.276,
      "step": 2000
    },
    {
      "epoch": 1.46,
      "eval_loss": 1.2826957702636719,
      "eval_runtime": 448.0598,
      "eval_samples_per_second": 14.003,
      "eval_steps_per_second": 1.752,
      "step": 2000
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 1.2659,
      "step": 2100
    },
    {
      "epoch": 1.53,
      "eval_loss": 1.2710933685302734,
      "eval_runtime": 417.5822,
      "eval_samples_per_second": 15.025,
      "eval_steps_per_second": 1.88,
      "step": 2100
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 1.2451,
      "step": 2200
    },
    {
      "epoch": 1.6,
      "eval_loss": 1.259511113166809,
      "eval_runtime": 417.6409,
      "eval_samples_per_second": 15.022,
      "eval_steps_per_second": 1.88,
      "step": 2200
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002,
      "loss": 1.2436,
      "step": 2300
    },
    {
      "epoch": 1.68,
      "eval_loss": 1.2497906684875488,
      "eval_runtime": 417.7302,
      "eval_samples_per_second": 15.019,
      "eval_steps_per_second": 1.879,
      "step": 2300
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0002,
      "loss": 1.2412,
      "step": 2400
    },
    {
      "epoch": 1.75,
      "eval_loss": 1.2375760078430176,
      "eval_runtime": 417.7184,
      "eval_samples_per_second": 15.02,
      "eval_steps_per_second": 1.879,
      "step": 2400
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 1.2223,
      "step": 2500
    },
    {
      "epoch": 1.82,
      "eval_loss": 1.2266868352890015,
      "eval_runtime": 417.9231,
      "eval_samples_per_second": 15.012,
      "eval_steps_per_second": 1.878,
      "step": 2500
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0002,
      "loss": 1.2106,
      "step": 2600
    },
    {
      "epoch": 1.89,
      "eval_loss": 1.216168999671936,
      "eval_runtime": 417.7189,
      "eval_samples_per_second": 15.02,
      "eval_steps_per_second": 1.879,
      "step": 2600
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0002,
      "loss": 1.2034,
      "step": 2700
    },
    {
      "epoch": 1.97,
      "eval_loss": 1.2053289413452148,
      "eval_runtime": 417.5961,
      "eval_samples_per_second": 15.024,
      "eval_steps_per_second": 1.88,
      "step": 2700
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0002,
      "loss": 1.1787,
      "step": 2800
    },
    {
      "epoch": 2.04,
      "eval_loss": 1.1963659524917603,
      "eval_runtime": 417.7494,
      "eval_samples_per_second": 15.019,
      "eval_steps_per_second": 1.879,
      "step": 2800
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0002,
      "loss": 1.1541,
      "step": 2900
    },
    {
      "epoch": 2.11,
      "eval_loss": 1.1863473653793335,
      "eval_runtime": 417.6096,
      "eval_samples_per_second": 15.024,
      "eval_steps_per_second": 1.88,
      "step": 2900
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0002,
      "loss": 1.1567,
      "step": 3000
    },
    {
      "epoch": 2.19,
      "eval_loss": 1.17583167552948,
      "eval_runtime": 417.6171,
      "eval_samples_per_second": 15.023,
      "eval_steps_per_second": 1.88,
      "step": 3000
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0002,
      "loss": 1.1395,
      "step": 3100
    },
    {
      "epoch": 2.26,
      "eval_loss": 1.1677733659744263,
      "eval_runtime": 417.4863,
      "eval_samples_per_second": 15.028,
      "eval_steps_per_second": 1.88,
      "step": 3100
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0002,
      "loss": 1.1528,
      "step": 3200
    },
    {
      "epoch": 2.33,
      "eval_loss": 1.1819865703582764,
      "eval_runtime": 339.2738,
      "eval_samples_per_second": 18.492,
      "eval_steps_per_second": 2.314,
      "step": 3200
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0002,
      "loss": 1.142,
      "step": 3300
    },
    {
      "epoch": 2.41,
      "eval_loss": 1.168481469154358,
      "eval_runtime": 345.6314,
      "eval_samples_per_second": 18.152,
      "eval_steps_per_second": 2.271,
      "step": 3300
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0002,
      "loss": 1.1252,
      "step": 3400
    },
    {
      "epoch": 2.48,
      "eval_loss": 1.1566288471221924,
      "eval_runtime": 339.7257,
      "eval_samples_per_second": 18.468,
      "eval_steps_per_second": 2.311,
      "step": 3400
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0002,
      "loss": 1.1227,
      "step": 3500
    },
    {
      "epoch": 2.55,
      "eval_loss": 1.1449061632156372,
      "eval_runtime": 339.6535,
      "eval_samples_per_second": 18.472,
      "eval_steps_per_second": 2.311,
      "step": 3500
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0002,
      "loss": 1.116,
      "step": 3600
    },
    {
      "epoch": 2.62,
      "eval_loss": 1.1353813409805298,
      "eval_runtime": 340.0732,
      "eval_samples_per_second": 18.449,
      "eval_steps_per_second": 2.308,
      "step": 3600
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0002,
      "loss": 1.1106,
      "step": 3700
    },
    {
      "epoch": 2.7,
      "eval_loss": 1.1255244016647339,
      "eval_runtime": 341.3582,
      "eval_samples_per_second": 18.38,
      "eval_steps_per_second": 2.3,
      "step": 3700
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0002,
      "loss": 1.0945,
      "step": 3800
    },
    {
      "epoch": 2.77,
      "eval_loss": 1.116302251815796,
      "eval_runtime": 339.9589,
      "eval_samples_per_second": 18.455,
      "eval_steps_per_second": 2.309,
      "step": 3800
    }
  ],
  "logging_steps": 100,
  "max_steps": 4116,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 1.2320741436517632e+18,
  "trial_name": null,
  "trial_params": null
}