{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 4385,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.28506271379703535,
      "grad_norm": 0.8346652984619141,
      "learning_rate": 0.0001885974914481186,
      "loss": 1.1468,
      "step": 250
    },
    {
      "epoch": 0.5701254275940707,
      "grad_norm": 0.7933174967765808,
      "learning_rate": 0.00017719498289623718,
      "loss": 0.9799,
      "step": 500
    },
    {
      "epoch": 0.855188141391106,
      "grad_norm": 0.7169309258460999,
      "learning_rate": 0.00016579247434435576,
      "loss": 0.943,
      "step": 750
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.8826740384101868,
      "eval_runtime": 87.0511,
      "eval_samples_per_second": 17.909,
      "eval_steps_per_second": 1.126,
      "step": 877
    },
    {
      "epoch": 1.1402508551881414,
      "grad_norm": 0.7942357659339905,
      "learning_rate": 0.00015438996579247434,
      "loss": 0.9029,
      "step": 1000
    },
    {
      "epoch": 1.4253135689851768,
      "grad_norm": 0.833415150642395,
      "learning_rate": 0.00014298745724059295,
      "loss": 0.8786,
      "step": 1250
    },
    {
      "epoch": 1.710376282782212,
      "grad_norm": 0.8092217445373535,
      "learning_rate": 0.0001315849486887115,
      "loss": 0.8515,
      "step": 1500
    },
    {
      "epoch": 1.9954389965792474,
      "grad_norm": 0.7863802909851074,
      "learning_rate": 0.00012018244013683011,
      "loss": 0.8452,
      "step": 1750
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.8214807510375977,
      "eval_runtime": 87.0295,
      "eval_samples_per_second": 17.913,
      "eval_steps_per_second": 1.126,
      "step": 1754
    },
    {
      "epoch": 2.280501710376283,
      "grad_norm": 1.0299004316329956,
      "learning_rate": 0.0001087799315849487,
      "loss": 0.8139,
      "step": 2000
    },
    {
      "epoch": 2.565564424173318,
      "grad_norm": 0.8788676857948303,
      "learning_rate": 9.737742303306729e-05,
      "loss": 0.8172,
      "step": 2250
    },
    {
      "epoch": 2.8506271379703536,
      "grad_norm": 0.9441242218017578,
      "learning_rate": 8.597491448118586e-05,
      "loss": 0.7919,
      "step": 2500
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.7865139842033386,
      "eval_runtime": 87.0332,
      "eval_samples_per_second": 17.913,
      "eval_steps_per_second": 1.126,
      "step": 2631
    },
    {
      "epoch": 3.135689851767389,
      "grad_norm": 1.2991793155670166,
      "learning_rate": 7.457240592930444e-05,
      "loss": 0.7961,
      "step": 2750
    },
    {
      "epoch": 3.420752565564424,
      "grad_norm": 0.9629215598106384,
      "learning_rate": 6.316989737742304e-05,
      "loss": 0.7761,
      "step": 3000
    },
    {
      "epoch": 3.7058152793614596,
      "grad_norm": 1.0487836599349976,
      "learning_rate": 5.1767388825541616e-05,
      "loss": 0.7673,
      "step": 3250
    },
    {
      "epoch": 3.990877993158495,
      "grad_norm": 0.9421872496604919,
      "learning_rate": 4.0364880273660207e-05,
      "loss": 0.7717,
      "step": 3500
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.76755291223526,
      "eval_runtime": 87.0332,
      "eval_samples_per_second": 17.913,
      "eval_steps_per_second": 1.126,
      "step": 3508
    },
    {
      "epoch": 4.27594070695553,
      "grad_norm": 0.9843165874481201,
      "learning_rate": 2.896237172177879e-05,
      "loss": 0.7556,
      "step": 3750
    },
    {
      "epoch": 4.561003420752566,
      "grad_norm": 0.908370852470398,
      "learning_rate": 1.7559863169897377e-05,
      "loss": 0.7452,
      "step": 4000
    },
    {
      "epoch": 4.846066134549601,
      "grad_norm": 1.0073881149291992,
      "learning_rate": 6.2029646522234895e-06,
      "loss": 0.7658,
      "step": 4250
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.7607811093330383,
      "eval_runtime": 87.0224,
      "eval_samples_per_second": 17.915,
      "eval_steps_per_second": 1.126,
      "step": 4385
    }
  ],
  "logging_steps": 250,
  "max_steps": 4385,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 2.2306963458097152e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}