{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.076923076923077,
  "eval_steps": 5,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 0.9935800099826351,
      "learning_rate": 0.0001,
      "loss": 1.9258,
      "step": 1
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 1.2486035475371888,
      "learning_rate": 0.00018660254037844388,
      "loss": 2.0231,
      "step": 5
    },
    {
      "epoch": 0.7692307692307693,
      "eval_loss": 1.8074404001235962,
      "eval_runtime": 2.6326,
      "eval_samples_per_second": 7.597,
      "eval_steps_per_second": 1.14,
      "step": 5
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.6291642388150667,
      "learning_rate": 0.00011736481776669306,
      "loss": 1.2461,
      "step": 10
    },
    {
      "epoch": 1.5384615384615383,
      "eval_loss": 1.3001079559326172,
      "eval_runtime": 1.3446,
      "eval_samples_per_second": 14.874,
      "eval_steps_per_second": 2.231,
      "step": 10
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 0.44475405058884243,
      "learning_rate": 3.5721239031346066e-05,
      "loss": 1.0347,
      "step": 15
    },
    {
      "epoch": 2.3076923076923075,
      "eval_loss": 1.1189016103744507,
      "eval_runtime": 1.34,
      "eval_samples_per_second": 14.926,
      "eval_steps_per_second": 2.239,
      "step": 15
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 0.5502394281465781,
      "learning_rate": 0.0,
      "loss": 1.0003,
      "step": 20
    },
    {
      "epoch": 3.076923076923077,
      "eval_loss": 1.0752954483032227,
      "eval_runtime": 1.3421,
      "eval_samples_per_second": 14.902,
      "eval_steps_per_second": 2.235,
      "step": 20
    },
    {
      "epoch": 3.076923076923077,
      "step": 20,
      "total_flos": 460885032960.0,
      "train_loss": 1.3211758494377137,
      "train_runtime": 84.6621,
      "train_samples_per_second": 3.78,
      "train_steps_per_second": 0.236
    }
  ],
  "logging_steps": 5,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 460885032960.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}