{
  "best_global_step": 1000,
  "best_metric": 0.3800264000892639,
  "best_model_checkpoint": "check-point/checkpoint-1000",
  "epoch": 7.763975155279503,
  "eval_steps": 500,
  "global_step": 2500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.5527950310559007,
      "eval_entropy": 0.36121490036231885,
      "eval_loss": 0.3807576298713684,
      "eval_mean_token_accuracy": 0.8985943699228591,
      "eval_num_tokens": 8180731.0,
      "eval_runtime": 8.7103,
      "eval_samples_per_second": 126.632,
      "eval_steps_per_second": 7.922,
      "step": 500
    },
    {
      "entropy": 0.45297335986024845,
      "epoch": 2.0,
      "grad_norm": 0.74609375,
      "learning_rate": 0.00016519230769230768,
      "loss": 0.4535,
      "mean_token_accuracy": 0.8832034343703193,
      "num_tokens": 10531260.0,
      "step": 644
    },
    {
      "epoch": 3.1055900621118013,
      "eval_entropy": 0.24365942028985507,
      "eval_loss": 0.3800264000892639,
      "eval_mean_token_accuracy": 0.9076558178749637,
      "eval_num_tokens": 16353946.0,
      "eval_runtime": 8.7106,
      "eval_samples_per_second": 126.627,
      "eval_steps_per_second": 7.921,
      "step": 1000
    },
    {
      "entropy": 0.2585039790372671,
      "epoch": 4.0,
      "grad_norm": 0.65234375,
      "learning_rate": 0.00012391025641025641,
      "loss": 0.2582,
      "mean_token_accuracy": 0.9224253959537293,
      "num_tokens": 21062520.0,
      "step": 1288
    },
    {
      "epoch": 4.658385093167702,
      "eval_entropy": 0.19808650362318841,
      "eval_loss": 0.3817984163761139,
      "eval_mean_token_accuracy": 0.9157982235369475,
      "eval_num_tokens": 24533627.0,
      "eval_runtime": 8.7626,
      "eval_samples_per_second": 125.876,
      "eval_steps_per_second": 7.874,
      "step": 1500
    },
    {
      "entropy": 0.13021414620535715,
      "epoch": 6.0,
      "grad_norm": 0.81640625,
      "learning_rate": 8.262820512820512e-05,
      "loss": 0.125,
      "mean_token_accuracy": 0.9599104345955464,
      "num_tokens": 31593780.0,
      "step": 1932
    },
    {
      "epoch": 6.211180124223603,
      "eval_entropy": 0.11409505208333333,
      "eval_loss": 0.4657194912433624,
      "eval_mean_token_accuracy": 0.9217576470927916,
      "eval_num_tokens": 32707230.0,
      "eval_runtime": 12.0522,
      "eval_samples_per_second": 91.519,
      "eval_steps_per_second": 5.725,
      "step": 2000
    },
    {
      "epoch": 7.763975155279503,
      "eval_entropy": 0.09595434216485507,
      "eval_loss": 0.5112115144729614,
      "eval_mean_token_accuracy": 0.9232899140620577,
      "eval_num_tokens": 40887516.0,
      "eval_runtime": 12.0489,
      "eval_samples_per_second": 91.544,
      "eval_steps_per_second": 5.727,
      "step": 2500
    }
  ],
  "logging_steps": 644,
  "max_steps": 3220,
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8.784869136228557e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}