{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 10030,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9970089730807578,
      "grad_norm": 13791.8017578125,
      "learning_rate": 0.0005993999999999999,
      "loss": 1.2252,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.18732381009853702,
      "eval_loss": 4.833001136779785,
      "eval_runtime": 4.3793,
      "eval_samples_per_second": 51.835,
      "eval_steps_per_second": 1.827,
      "step": 1003
    },
    {
      "epoch": 1.9940179461615155,
      "grad_norm": 38028.28125,
      "learning_rate": 0.0005336212624584717,
      "loss": 1.104,
      "step": 2000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.19786718622033328,
      "eval_loss": 4.691581726074219,
      "eval_runtime": 3.8992,
      "eval_samples_per_second": 58.218,
      "eval_steps_per_second": 2.052,
      "step": 2006
    },
    {
      "epoch": 2.991026919242273,
      "grad_norm": 34522.31640625,
      "learning_rate": 0.0004671760797342192,
      "loss": 1.0865,
      "step": 3000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.20838469960430012,
      "eval_loss": 4.60223913192749,
      "eval_runtime": 3.82,
      "eval_samples_per_second": 59.424,
      "eval_steps_per_second": 2.094,
      "step": 3009
    },
    {
      "epoch": 3.988035892323031,
      "grad_norm": 9552.4384765625,
      "learning_rate": 0.00040073089700996676,
      "loss": 1.0446,
      "step": 4000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.24390285955671268,
      "eval_loss": 4.333094596862793,
      "eval_runtime": 3.8108,
      "eval_samples_per_second": 59.568,
      "eval_steps_per_second": 2.099,
      "step": 4012
    },
    {
      "epoch": 4.985044865403789,
      "grad_norm": 7808.9814453125,
      "learning_rate": 0.00033428571428571426,
      "loss": 0.9806,
      "step": 5000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.26604136313870186,
      "eval_loss": 4.115571022033691,
      "eval_runtime": 3.7314,
      "eval_samples_per_second": 60.836,
      "eval_steps_per_second": 2.144,
      "step": 5015
    },
    {
      "epoch": 5.982053838484546,
      "grad_norm": 7799.94580078125,
      "learning_rate": 0.00026784053156146177,
      "loss": 0.9375,
      "step": 6000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.2841539005319103,
      "eval_loss": 3.9862992763519287,
      "eval_runtime": 3.7749,
      "eval_samples_per_second": 60.134,
      "eval_steps_per_second": 2.119,
      "step": 6018
    },
    {
      "epoch": 6.979062811565304,
      "grad_norm": 7969.98388671875,
      "learning_rate": 0.00020139534883720927,
      "loss": 0.9074,
      "step": 7000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.2942231264601671,
      "eval_loss": 3.9006783962249756,
      "eval_runtime": 3.758,
      "eval_samples_per_second": 60.405,
      "eval_steps_per_second": 2.129,
      "step": 7021
    },
    {
      "epoch": 7.976071784646062,
      "grad_norm": 8524.130859375,
      "learning_rate": 0.0001349501661129568,
      "loss": 0.8839,
      "step": 8000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.3046802934558652,
      "eval_loss": 3.8308379650115967,
      "eval_runtime": 3.7591,
      "eval_samples_per_second": 60.388,
      "eval_steps_per_second": 2.128,
      "step": 8024
    },
    {
      "epoch": 8.97308075772682,
      "grad_norm": 8583.34375,
      "learning_rate": 6.850498338870431e-05,
      "loss": 0.867,
      "step": 9000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.31158564445632214,
      "eval_loss": 3.7815797328948975,
      "eval_runtime": 3.7451,
      "eval_samples_per_second": 60.613,
      "eval_steps_per_second": 2.136,
      "step": 9027
    },
    {
      "epoch": 9.970089730807578,
      "grad_norm": 8688.66796875,
      "learning_rate": 2.0598006644518273e-06,
      "loss": 0.8541,
      "step": 10000
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.3152495323154909,
      "eval_loss": 3.7565455436706543,
      "eval_runtime": 3.716,
      "eval_samples_per_second": 61.087,
      "eval_steps_per_second": 2.153,
      "step": 10030
    },
    {
      "epoch": 10.0,
      "step": 10030,
      "total_flos": 8.379374174208e+16,
      "train_loss": 0.9887108649713092,
      "train_runtime": 9029.3583,
      "train_samples_per_second": 35.516,
      "train_steps_per_second": 1.111
    }
  ],
  "logging_steps": 1000,
  "max_steps": 10030,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "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": 8.379374174208e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}