{
  "best_metric": 2.62109375,
  "best_model_checkpoint": "/home/zhengxinyong/outputs/bloom-7b1_de_continual-pretrain_100000samples_-1vocab_original/checkpoint-15000",
  "epoch": 1.6485328058028355,
  "global_step": 15000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.27,
      "learning_rate": 0.0001,
      "loss": 3.0311,
      "step": 2500
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001,
      "loss": 2.7334,
      "step": 5000
    },
    {
      "epoch": 0.55,
      "eval_loss": 2.775390625,
      "eval_runtime": 12.1862,
      "eval_samples_per_second": 65.976,
      "eval_steps_per_second": 8.288,
      "step": 5000
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0001,
      "loss": 2.6307,
      "step": 7500
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0001,
      "loss": 2.4576,
      "step": 10000
    },
    {
      "epoch": 1.1,
      "eval_loss": 2.68359375,
      "eval_runtime": 12.1913,
      "eval_samples_per_second": 65.949,
      "eval_steps_per_second": 8.285,
      "step": 10000
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0001,
      "loss": 2.2705,
      "step": 12500
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0001,
      "loss": 2.2921,
      "step": 15000
    },
    {
      "epoch": 1.65,
      "eval_loss": 2.62109375,
      "eval_runtime": 12.1817,
      "eval_samples_per_second": 66.0,
      "eval_steps_per_second": 8.291,
      "step": 15000
    }
  ],
  "max_steps": 25000,
  "num_train_epochs": 3,
  "total_flos": 4.4542120389519606e+18,
  "trial_name": null,
  "trial_params": null
}