{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0038639876352395673,
  "eval_steps": 500,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.7173828125,
      "epoch": 0.0007727975270479134,
      "learning_rate": 0.00011999999999999999,
      "loss": 3.8298,
      "mean_token_accuracy": 0.5266443401575088,
      "num_input_tokens_seen": 8532,
      "num_tokens": 6819.0,
      "step": 5,
      "train_runtime": 5.2614,
      "train_tokens_per_second": 1621.61
    },
    {
      "entropy": 0.8421875,
      "epoch": 0.0015455950540958269,
      "learning_rate": 0.00027,
      "loss": 2.5375,
      "mean_token_accuracy": 0.5753881394863128,
      "num_input_tokens_seen": 19448,
      "num_tokens": 14529.0,
      "step": 10,
      "train_runtime": 9.0474,
      "train_tokens_per_second": 2149.573
    },
    {
      "entropy": 1.1810546875,
      "epoch": 0.00231839258114374,
      "learning_rate": 0.0002985402103112355,
      "loss": 1.5425,
      "mean_token_accuracy": 0.6471151679754257,
      "num_input_tokens_seen": 29460,
      "num_tokens": 21905.0,
      "step": 15,
      "train_runtime": 12.415,
      "train_tokens_per_second": 2372.931
    },
    {
      "entropy": 1.571484375,
      "epoch": 0.0030911901081916537,
      "learning_rate": 0.00029265847744427303,
      "loss": 1.3481,
      "mean_token_accuracy": 0.6753059208393097,
      "num_input_tokens_seen": 39712,
      "num_tokens": 29725.0,
      "step": 20,
      "train_runtime": 15.9429,
      "train_tokens_per_second": 2490.896
    },
    {
      "entropy": 1.392578125,
      "epoch": 0.0038639876352395673,
      "learning_rate": 0.000282442138928839,
      "loss": 1.2909,
      "mean_token_accuracy": 0.6818175077438354,
      "num_input_tokens_seen": 51300,
      "num_tokens": 38298.0,
      "step": 25,
      "train_runtime": 19.3266,
      "train_tokens_per_second": 2654.371
    }
  ],
  "logging_steps": 5,
  "max_steps": 100,
  "num_input_tokens_seen": 51300,
  "num_train_epochs": 1,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 435228451430400.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}