{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 11350,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.8810572687224669,
      "grad_norm": 18084.95703125,
      "learning_rate": 0.0005993999999999999,
      "loss": 1.1021,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2970001674393294,
      "eval_loss": 4.178283214569092,
      "eval_runtime": 615.0579,
      "eval_samples_per_second": 60.807,
      "eval_steps_per_second": 1.901,
      "step": 1135
    },
    {
      "epoch": 1.7621145374449338,
      "grad_norm": 26064.150390625,
      "learning_rate": 0.0005420869565217391,
      "loss": 0.9898,
      "step": 2000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3064508094645081,
      "eval_loss": 4.065672397613525,
      "eval_runtime": 614.7114,
      "eval_samples_per_second": 60.842,
      "eval_steps_per_second": 1.902,
      "step": 2270
    },
    {
      "epoch": 2.643171806167401,
      "grad_norm": 14842.248046875,
      "learning_rate": 0.00048411594202898544,
      "loss": 0.9617,
      "step": 3000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3191253911278085,
      "eval_loss": 3.9150826930999756,
      "eval_runtime": 613.7049,
      "eval_samples_per_second": 60.941,
      "eval_steps_per_second": 1.905,
      "step": 3405
    },
    {
      "epoch": 3.5242290748898677,
      "grad_norm": 9546.3798828125,
      "learning_rate": 0.00042614492753623184,
      "loss": 0.9291,
      "step": 4000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.33467485375221073,
      "eval_loss": 3.7638590335845947,
      "eval_runtime": 614.245,
      "eval_samples_per_second": 60.888,
      "eval_steps_per_second": 1.903,
      "step": 4540
    },
    {
      "epoch": 4.405286343612334,
      "grad_norm": 6838.88623046875,
      "learning_rate": 0.00036817391304347824,
      "loss": 0.8903,
      "step": 5000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.34916521029333275,
      "eval_loss": 3.6331233978271484,
      "eval_runtime": 614.1793,
      "eval_samples_per_second": 60.894,
      "eval_steps_per_second": 1.903,
      "step": 5675
    },
    {
      "epoch": 5.286343612334802,
      "grad_norm": 6897.5400390625,
      "learning_rate": 0.00031020289855072464,
      "loss": 0.8546,
      "step": 6000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.3598596125872516,
      "eval_loss": 3.546229124069214,
      "eval_runtime": 613.9133,
      "eval_samples_per_second": 60.921,
      "eval_steps_per_second": 1.904,
      "step": 6810
    },
    {
      "epoch": 6.167400881057269,
      "grad_norm": 7466.97265625,
      "learning_rate": 0.000252231884057971,
      "loss": 0.8321,
      "step": 7000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.36795211235179004,
      "eval_loss": 3.4829158782958984,
      "eval_runtime": 613.499,
      "eval_samples_per_second": 60.962,
      "eval_steps_per_second": 1.905,
      "step": 7945
    },
    {
      "epoch": 7.048458149779735,
      "grad_norm": 7247.333984375,
      "learning_rate": 0.00019426086956521736,
      "loss": 0.8128,
      "step": 8000
    },
    {
      "epoch": 7.929515418502203,
      "grad_norm": 7852.26904296875,
      "learning_rate": 0.00013628985507246376,
      "loss": 0.798,
      "step": 9000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.3742137676988604,
      "eval_loss": 3.4337821006774902,
      "eval_runtime": 612.7115,
      "eval_samples_per_second": 61.04,
      "eval_steps_per_second": 1.908,
      "step": 9080
    },
    {
      "epoch": 8.810572687224669,
      "grad_norm": 7869.26025390625,
      "learning_rate": 7.831884057971013e-05,
      "loss": 0.7836,
      "step": 10000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.3786116140104859,
      "eval_loss": 3.4001269340515137,
      "eval_runtime": 613.2076,
      "eval_samples_per_second": 60.991,
      "eval_steps_per_second": 1.906,
      "step": 10215
    },
    {
      "epoch": 9.691629955947137,
      "grad_norm": 8053.40234375,
      "learning_rate": 2.034782608695652e-05,
      "loss": 0.7747,
      "step": 11000
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.3821338572789016,
      "eval_loss": 3.3799285888671875,
      "eval_runtime": 613.3109,
      "eval_samples_per_second": 60.98,
      "eval_steps_per_second": 1.906,
      "step": 11350
    },
    {
      "epoch": 10.0,
      "step": 11350,
      "total_flos": 9.489081434112e+16,
      "train_loss": 0.8809447758208288,
      "train_runtime": 16232.7867,
      "train_samples_per_second": 22.372,
      "train_steps_per_second": 0.699
    }
  ],
  "logging_steps": 1000,
  "max_steps": 11350,
  "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": 9.489081434112e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}