{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 18717,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08014104824491104,
      "grad_norm": 0.4942402243614197,
      "learning_rate": 4.866431586258482e-05,
      "loss": 0.772,
      "step": 500
    },
    {
      "epoch": 0.16028209648982208,
      "grad_norm": 0.36161601543426514,
      "learning_rate": 4.732863172516964e-05,
      "loss": 0.2015,
      "step": 1000
    },
    {
      "epoch": 0.24042314473473314,
      "grad_norm": 0.2504396438598633,
      "learning_rate": 4.599294758775445e-05,
      "loss": 0.1906,
      "step": 1500
    },
    {
      "epoch": 0.32056419297964417,
      "grad_norm": 0.21929016709327698,
      "learning_rate": 4.4657263450339265e-05,
      "loss": 0.1902,
      "step": 2000
    },
    {
      "epoch": 0.4007052412245552,
      "grad_norm": 0.37331002950668335,
      "learning_rate": 4.332157931292408e-05,
      "loss": 0.1844,
      "step": 2500
    },
    {
      "epoch": 0.4808462894694663,
      "grad_norm": 0.2946544587612152,
      "learning_rate": 4.19858951755089e-05,
      "loss": 0.1811,
      "step": 3000
    },
    {
      "epoch": 0.5609873377143773,
      "grad_norm": 0.326863557100296,
      "learning_rate": 4.065021103809372e-05,
      "loss": 0.1779,
      "step": 3500
    },
    {
      "epoch": 0.6411283859592883,
      "grad_norm": 0.3597277104854584,
      "learning_rate": 3.931452690067853e-05,
      "loss": 0.1756,
      "step": 4000
    },
    {
      "epoch": 0.7212694342041994,
      "grad_norm": 0.3043977916240692,
      "learning_rate": 3.7978842763263345e-05,
      "loss": 0.1767,
      "step": 4500
    },
    {
      "epoch": 0.8014104824491104,
      "grad_norm": 0.21228274703025818,
      "learning_rate": 3.664315862584816e-05,
      "loss": 0.1748,
      "step": 5000
    },
    {
      "epoch": 0.8815515306940215,
      "grad_norm": 0.32453158497810364,
      "learning_rate": 3.530747448843298e-05,
      "loss": 0.1744,
      "step": 5500
    },
    {
      "epoch": 0.9616925789389326,
      "grad_norm": 0.19325225055217743,
      "learning_rate": 3.397179035101779e-05,
      "loss": 0.174,
      "step": 6000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.1885562390089035,
      "eval_runtime": 108.5754,
      "eval_samples_per_second": 188.671,
      "eval_steps_per_second": 11.798,
      "step": 6239
    },
    {
      "epoch": 1.0418336271838435,
      "grad_norm": 0.27379509806632996,
      "learning_rate": 3.263610621360261e-05,
      "loss": 0.172,
      "step": 6500
    },
    {
      "epoch": 1.1219746754287545,
      "grad_norm": 0.35208427906036377,
      "learning_rate": 3.1300422076187425e-05,
      "loss": 0.1703,
      "step": 7000
    },
    {
      "epoch": 1.2021157236736657,
      "grad_norm": 0.2537246644496918,
      "learning_rate": 2.9964737938772242e-05,
      "loss": 0.1698,
      "step": 7500
    },
    {
      "epoch": 1.2822567719185767,
      "grad_norm": 0.32513222098350525,
      "learning_rate": 2.862905380135706e-05,
      "loss": 0.1695,
      "step": 8000
    },
    {
      "epoch": 1.3623978201634879,
      "grad_norm": 0.3062923550605774,
      "learning_rate": 2.729336966394187e-05,
      "loss": 0.1723,
      "step": 8500
    },
    {
      "epoch": 1.4425388684083988,
      "grad_norm": 0.3160272240638733,
      "learning_rate": 2.5957685526526688e-05,
      "loss": 0.1697,
      "step": 9000
    },
    {
      "epoch": 1.5226799166533098,
      "grad_norm": 0.21418999135494232,
      "learning_rate": 2.4622001389111505e-05,
      "loss": 0.1645,
      "step": 9500
    },
    {
      "epoch": 1.6028209648982208,
      "grad_norm": 0.1781550943851471,
      "learning_rate": 2.3286317251696322e-05,
      "loss": 0.1679,
      "step": 10000
    },
    {
      "epoch": 1.6829620131431318,
      "grad_norm": 0.24800395965576172,
      "learning_rate": 2.1950633114281136e-05,
      "loss": 0.1671,
      "step": 10500
    },
    {
      "epoch": 1.763103061388043,
      "grad_norm": 0.28134429454803467,
      "learning_rate": 2.0614948976865954e-05,
      "loss": 0.1645,
      "step": 11000
    },
    {
      "epoch": 1.8432441096329542,
      "grad_norm": 0.3860167860984802,
      "learning_rate": 1.9279264839450767e-05,
      "loss": 0.1637,
      "step": 11500
    },
    {
      "epoch": 1.9233851578778651,
      "grad_norm": 0.2310725599527359,
      "learning_rate": 1.7943580702035585e-05,
      "loss": 0.1644,
      "step": 12000
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.18438942730426788,
      "eval_runtime": 108.2951,
      "eval_samples_per_second": 189.159,
      "eval_steps_per_second": 11.829,
      "step": 12478
    },
    {
      "epoch": 2.003526206122776,
      "grad_norm": 0.3371495306491852,
      "learning_rate": 1.66078965646204e-05,
      "loss": 0.1658,
      "step": 12500
    },
    {
      "epoch": 2.083667254367687,
      "grad_norm": 0.27058491110801697,
      "learning_rate": 1.5272212427205216e-05,
      "loss": 0.1605,
      "step": 13000
    },
    {
      "epoch": 2.163808302612598,
      "grad_norm": 0.2933259904384613,
      "learning_rate": 1.3936528289790032e-05,
      "loss": 0.1641,
      "step": 13500
    },
    {
      "epoch": 2.243949350857509,
      "grad_norm": 0.31803345680236816,
      "learning_rate": 1.2600844152374847e-05,
      "loss": 0.1618,
      "step": 14000
    },
    {
      "epoch": 2.3240903991024204,
      "grad_norm": 0.29237860441207886,
      "learning_rate": 1.1265160014959663e-05,
      "loss": 0.1629,
      "step": 14500
    },
    {
      "epoch": 2.4042314473473314,
      "grad_norm": 0.7349903583526611,
      "learning_rate": 9.929475877544479e-06,
      "loss": 0.1632,
      "step": 15000
    },
    {
      "epoch": 2.4843724955922424,
      "grad_norm": 0.32003411650657654,
      "learning_rate": 8.593791740129294e-06,
      "loss": 0.1658,
      "step": 15500
    },
    {
      "epoch": 2.5645135438371534,
      "grad_norm": 0.29341480135917664,
      "learning_rate": 7.258107602714111e-06,
      "loss": 0.1631,
      "step": 16000
    },
    {
      "epoch": 2.6446545920820643,
      "grad_norm": 0.3577577769756317,
      "learning_rate": 5.9224234652989265e-06,
      "loss": 0.1667,
      "step": 16500
    },
    {
      "epoch": 2.7247956403269757,
      "grad_norm": 0.20905520021915436,
      "learning_rate": 4.586739327883742e-06,
      "loss": 0.1629,
      "step": 17000
    },
    {
      "epoch": 2.8049366885718863,
      "grad_norm": 0.21252945065498352,
      "learning_rate": 3.2510551904685583e-06,
      "loss": 0.1624,
      "step": 17500
    },
    {
      "epoch": 2.8850777368167977,
      "grad_norm": 0.23013170063495636,
      "learning_rate": 1.915371053053374e-06,
      "loss": 0.1606,
      "step": 18000
    },
    {
      "epoch": 2.9652187850617087,
      "grad_norm": 0.2925206422805786,
      "learning_rate": 5.796869156381899e-07,
      "loss": 0.1652,
      "step": 18500
    }
  ],
  "logging_steps": 500,
  "max_steps": 18717,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.0132668486254592e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}