{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 3000,
  "global_step": 88686,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.382721060821325e-05,
      "grad_norm": 146.0,
      "learning_rate": 0.00029999661727893914,
      "loss": 28.0,
      "step": 1
    },
    {
      "epoch": 0.10148163182463973,
      "grad_norm": 1.0625,
      "learning_rate": 0.000289851836817536,
      "loss": 4.8045,
      "step": 3000
    },
    {
      "epoch": 0.10148163182463973,
      "eval_loss": 3.3920648097991943,
      "eval_runtime": 84.2066,
      "eval_samples_per_second": 1121.111,
      "eval_steps_per_second": 8.764,
      "step": 3000
    },
    {
      "epoch": 0.20296326364927947,
      "grad_norm": 1.1875,
      "learning_rate": 0.000279703673635072,
      "loss": 4.1932,
      "step": 6000
    },
    {
      "epoch": 0.20296326364927947,
      "eval_loss": 3.254700183868408,
      "eval_runtime": 83.4018,
      "eval_samples_per_second": 1131.93,
      "eval_steps_per_second": 8.849,
      "step": 6000
    },
    {
      "epoch": 0.30444489547391923,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026955551045260807,
      "loss": 4.1098,
      "step": 9000
    },
    {
      "epoch": 0.30444489547391923,
      "eval_loss": 3.204374074935913,
      "eval_runtime": 83.3884,
      "eval_samples_per_second": 1132.112,
      "eval_steps_per_second": 8.85,
      "step": 9000
    },
    {
      "epoch": 0.40592652729855894,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002594073472701441,
      "loss": 4.0752,
      "step": 12000
    },
    {
      "epoch": 0.40592652729855894,
      "eval_loss": 3.173970937728882,
      "eval_runtime": 83.3988,
      "eval_samples_per_second": 1131.971,
      "eval_steps_per_second": 8.849,
      "step": 12000
    },
    {
      "epoch": 0.5074081591231987,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002492591840876801,
      "loss": 4.0515,
      "step": 15000
    },
    {
      "epoch": 0.5074081591231987,
      "eval_loss": 3.157351016998291,
      "eval_runtime": 83.395,
      "eval_samples_per_second": 1132.022,
      "eval_steps_per_second": 8.849,
      "step": 15000
    },
    {
      "epoch": 0.6088897909478385,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002391110209052161,
      "loss": 4.0446,
      "step": 18000
    },
    {
      "epoch": 0.6088897909478385,
      "eval_loss": 3.1490302085876465,
      "eval_runtime": 83.4067,
      "eval_samples_per_second": 1131.863,
      "eval_steps_per_second": 8.848,
      "step": 18000
    },
    {
      "epoch": 0.7103714227724782,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022896285772275215,
      "loss": 4.036,
      "step": 21000
    },
    {
      "epoch": 0.7103714227724782,
      "eval_loss": 3.1390371322631836,
      "eval_runtime": 83.3672,
      "eval_samples_per_second": 1132.399,
      "eval_steps_per_second": 8.852,
      "step": 21000
    },
    {
      "epoch": 0.8118530545971179,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002188146945402882,
      "loss": 4.0298,
      "step": 24000
    },
    {
      "epoch": 0.8118530545971179,
      "eval_loss": 3.1364541053771973,
      "eval_runtime": 83.5197,
      "eval_samples_per_second": 1130.332,
      "eval_steps_per_second": 8.836,
      "step": 24000
    },
    {
      "epoch": 0.9133346864217576,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020866653135782423,
      "loss": 4.0273,
      "step": 27000
    },
    {
      "epoch": 0.9133346864217576,
      "eval_loss": 3.1308434009552,
      "eval_runtime": 83.5474,
      "eval_samples_per_second": 1129.957,
      "eval_steps_per_second": 8.833,
      "step": 27000
    },
    {
      "epoch": 1.0148163182463974,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019851836817536025,
      "loss": 4.023,
      "step": 30000
    },
    {
      "epoch": 1.0148163182463974,
      "eval_loss": 3.1285781860351562,
      "eval_runtime": 83.5362,
      "eval_samples_per_second": 1130.109,
      "eval_steps_per_second": 8.834,
      "step": 30000
    },
    {
      "epoch": 1.116297950071037,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001883702049928963,
      "loss": 4.0229,
      "step": 33000
    },
    {
      "epoch": 1.116297950071037,
      "eval_loss": 3.1244707107543945,
      "eval_runtime": 83.5446,
      "eval_samples_per_second": 1129.996,
      "eval_steps_per_second": 8.834,
      "step": 33000
    },
    {
      "epoch": 1.217779581895677,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001782220418104323,
      "loss": 4.0158,
      "step": 36000
    },
    {
      "epoch": 1.217779581895677,
      "eval_loss": 3.1234333515167236,
      "eval_runtime": 83.5361,
      "eval_samples_per_second": 1130.11,
      "eval_steps_per_second": 8.835,
      "step": 36000
    },
    {
      "epoch": 1.3192612137203166,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016807387862796832,
      "loss": 4.0163,
      "step": 39000
    },
    {
      "epoch": 1.3192612137203166,
      "eval_loss": 3.121718406677246,
      "eval_runtime": 83.5263,
      "eval_samples_per_second": 1130.242,
      "eval_steps_per_second": 8.836,
      "step": 39000
    },
    {
      "epoch": 1.4207428455449564,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015792571544550436,
      "loss": 4.0156,
      "step": 42000
    },
    {
      "epoch": 1.4207428455449564,
      "eval_loss": 3.1207656860351562,
      "eval_runtime": 83.531,
      "eval_samples_per_second": 1130.179,
      "eval_steps_per_second": 8.835,
      "step": 42000
    },
    {
      "epoch": 1.522224477369596,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014777755226304037,
      "loss": 4.0147,
      "step": 45000
    },
    {
      "epoch": 1.522224477369596,
      "eval_loss": 3.119389295578003,
      "eval_runtime": 83.5355,
      "eval_samples_per_second": 1130.119,
      "eval_steps_per_second": 8.835,
      "step": 45000
    },
    {
      "epoch": 1.6237061091942357,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001376293890805764,
      "loss": 4.0114,
      "step": 48000
    },
    {
      "epoch": 1.6237061091942357,
      "eval_loss": 3.11716628074646,
      "eval_runtime": 83.398,
      "eval_samples_per_second": 1131.981,
      "eval_steps_per_second": 8.849,
      "step": 48000
    },
    {
      "epoch": 1.7251877410188756,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012748122589811243,
      "loss": 4.0113,
      "step": 51000
    },
    {
      "epoch": 1.7251877410188756,
      "eval_loss": 3.1180977821350098,
      "eval_runtime": 83.3949,
      "eval_samples_per_second": 1132.024,
      "eval_steps_per_second": 8.849,
      "step": 51000
    },
    {
      "epoch": 1.8266693728435153,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011733306271564845,
      "loss": 4.0097,
      "step": 54000
    },
    {
      "epoch": 1.8266693728435153,
      "eval_loss": 3.116975784301758,
      "eval_runtime": 83.3757,
      "eval_samples_per_second": 1132.284,
      "eval_steps_per_second": 8.851,
      "step": 54000
    },
    {
      "epoch": 1.928151004668155,
      "grad_norm": 1.140625,
      "learning_rate": 0.00010718489953318448,
      "loss": 4.0105,
      "step": 57000
    },
    {
      "epoch": 1.928151004668155,
      "eval_loss": 3.116128921508789,
      "eval_runtime": 83.3931,
      "eval_samples_per_second": 1132.049,
      "eval_steps_per_second": 8.85,
      "step": 57000
    },
    {
      "epoch": 2.029632636492795,
      "grad_norm": 1.1953125,
      "learning_rate": 9.703673635072052e-05,
      "loss": 4.0115,
      "step": 60000
    },
    {
      "epoch": 2.029632636492795,
      "eval_loss": 3.1166157722473145,
      "eval_runtime": 83.3926,
      "eval_samples_per_second": 1132.055,
      "eval_steps_per_second": 8.85,
      "step": 60000
    },
    {
      "epoch": 2.1311142683174347,
      "grad_norm": 1.4609375,
      "learning_rate": 8.688857316825655e-05,
      "loss": 4.0117,
      "step": 63000
    },
    {
      "epoch": 2.1311142683174347,
      "eval_loss": 3.1167428493499756,
      "eval_runtime": 83.4107,
      "eval_samples_per_second": 1131.809,
      "eval_steps_per_second": 8.848,
      "step": 63000
    },
    {
      "epoch": 2.232595900142074,
      "grad_norm": 1.2109375,
      "learning_rate": 7.674040998579256e-05,
      "loss": 4.0082,
      "step": 66000
    },
    {
      "epoch": 2.232595900142074,
      "eval_loss": 3.116044282913208,
      "eval_runtime": 83.4033,
      "eval_samples_per_second": 1131.909,
      "eval_steps_per_second": 8.849,
      "step": 66000
    },
    {
      "epoch": 2.334077531966714,
      "grad_norm": 1.265625,
      "learning_rate": 6.659224680332859e-05,
      "loss": 4.0089,
      "step": 69000
    },
    {
      "epoch": 2.334077531966714,
      "eval_loss": 3.1157476902008057,
      "eval_runtime": 83.3805,
      "eval_samples_per_second": 1132.219,
      "eval_steps_per_second": 8.851,
      "step": 69000
    },
    {
      "epoch": 2.435559163791354,
      "grad_norm": 1.2421875,
      "learning_rate": 5.644408362086462e-05,
      "loss": 4.0087,
      "step": 72000
    },
    {
      "epoch": 2.435559163791354,
      "eval_loss": 3.1155996322631836,
      "eval_runtime": 83.4329,
      "eval_samples_per_second": 1131.508,
      "eval_steps_per_second": 8.845,
      "step": 72000
    },
    {
      "epoch": 2.5370407956159937,
      "grad_norm": 1.34375,
      "learning_rate": 4.629592043840065e-05,
      "loss": 4.0104,
      "step": 75000
    },
    {
      "epoch": 2.5370407956159937,
      "eval_loss": 3.115832567214966,
      "eval_runtime": 83.4175,
      "eval_samples_per_second": 1131.716,
      "eval_steps_per_second": 8.847,
      "step": 75000
    },
    {
      "epoch": 2.638522427440633,
      "grad_norm": 1.3203125,
      "learning_rate": 3.614775725593667e-05,
      "loss": 4.0088,
      "step": 78000
    },
    {
      "epoch": 2.638522427440633,
      "eval_loss": 3.115938425064087,
      "eval_runtime": 83.3984,
      "eval_samples_per_second": 1131.977,
      "eval_steps_per_second": 8.849,
      "step": 78000
    },
    {
      "epoch": 2.740004059265273,
      "grad_norm": 1.234375,
      "learning_rate": 2.59995940734727e-05,
      "loss": 4.007,
      "step": 81000
    },
    {
      "epoch": 2.740004059265273,
      "eval_loss": 3.115938425064087,
      "eval_runtime": 83.4084,
      "eval_samples_per_second": 1131.841,
      "eval_steps_per_second": 8.848,
      "step": 81000
    },
    {
      "epoch": 2.841485691089913,
      "grad_norm": 1.3984375,
      "learning_rate": 1.5851430891008727e-05,
      "loss": 4.0116,
      "step": 84000
    },
    {
      "epoch": 2.841485691089913,
      "eval_loss": 3.116044282913208,
      "eval_runtime": 83.4021,
      "eval_samples_per_second": 1131.925,
      "eval_steps_per_second": 8.849,
      "step": 84000
    },
    {
      "epoch": 2.9429673229145523,
      "grad_norm": 1.2109375,
      "learning_rate": 5.703267708544753e-06,
      "loss": 4.0099,
      "step": 87000
    },
    {
      "epoch": 2.9429673229145523,
      "eval_loss": 3.115874767303467,
      "eval_runtime": 83.5591,
      "eval_samples_per_second": 1129.799,
      "eval_steps_per_second": 8.832,
      "step": 87000
    },
    {
      "epoch": 3.0,
      "step": 88686,
      "total_flos": 1.19291602746581e+18,
      "train_loss": 4.055220102383691,
      "train_runtime": 32766.006,
      "train_samples_per_second": 346.44,
      "train_steps_per_second": 2.707
    }
  ],
  "logging_steps": 3000,
  "max_steps": 88686,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.19291602746581e+18,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": null
}