trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "global_step": 155,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.32,
      "learning_rate": 9.35483870967742e-06,
      "loss": 0.2297,
      "max_memory_allocated (GB)": 83.39,
      "memory_allocated (GB)": 19.37,
      "step": 10,
      "total_memory_available (GB)": 93.74
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.70967741935484e-06,
      "loss": 0.0863,
      "max_memory_allocated (GB)": 83.39,
      "memory_allocated (GB)": 19.37,
      "step": 20,
      "total_memory_available (GB)": 93.74
    },
    {
      "epoch": 0.97,
      "learning_rate": 8.064516129032258e-06,
      "loss": 0.0828,
      "max_memory_allocated (GB)": 83.39,
      "memory_allocated (GB)": 19.37,
      "step": 30,
      "total_memory_available (GB)": 93.74
    },
    {
      "epoch": 1.29,
      "learning_rate": 7.4193548387096784e-06,
      "loss": 0.0625,
      "max_memory_allocated (GB)": 83.39,
      "memory_allocated (GB)": 19.37,
      "step": 40,
      "total_memory_available (GB)": 93.74
    },
    {
      "epoch": 1.61,
      "learning_rate": 6.774193548387097e-06,
      "loss": 0.0641,
      "max_memory_allocated (GB)": 83.39,
      "memory_allocated (GB)": 19.37,
      "step": 50,
      "total_memory_available (GB)": 93.74
    },
    {
      "epoch": 1.94,
      "learning_rate": 6.129032258064517e-06,
      "loss": 0.0613,
      "max_memory_allocated (GB)": 83.39,
      "memory_allocated (GB)": 19.37,
      "step": 60,
      "total_memory_available (GB)": 93.74
    },
    {
      "epoch": 2.26,
      "learning_rate": 5.483870967741935e-06,
      "loss": 0.0559,
      "max_memory_allocated (GB)": 83.39,
      "memory_allocated (GB)": 19.37,
      "step": 70,
      "total_memory_available (GB)": 93.74
    },
    {
      "epoch": 2.58,
      "learning_rate": 4.838709677419355e-06,
      "loss": 0.0602,
      "max_memory_allocated (GB)": 83.39,
      "memory_allocated (GB)": 19.37,
      "step": 80,
      "total_memory_available (GB)": 93.74
    },
    {
      "epoch": 2.9,
      "learning_rate": 4.193548387096774e-06,
      "loss": 0.066,
      "max_memory_allocated (GB)": 83.39,
      "memory_allocated (GB)": 19.37,
      "step": 90,
      "total_memory_available (GB)": 93.74
    },
    {
      "epoch": 3.23,
      "learning_rate": 3.548387096774194e-06,
      "loss": 0.0539,
      "max_memory_allocated (GB)": 83.39,
      "memory_allocated (GB)": 19.37,
      "step": 100,
      "total_memory_available (GB)": 93.74
    },
    {
      "epoch": 3.55,
      "learning_rate": 2.903225806451613e-06,
      "loss": 0.0523,
      "max_memory_allocated (GB)": 83.39,
      "memory_allocated (GB)": 19.37,
      "step": 110,
      "total_memory_available (GB)": 93.74
    },
    {
      "epoch": 3.87,
      "learning_rate": 2.2580645161290324e-06,
      "loss": 0.0555,
      "max_memory_allocated (GB)": 83.39,
      "memory_allocated (GB)": 19.37,
      "step": 120,
      "total_memory_available (GB)": 93.74
    },
    {
      "epoch": 4.19,
      "learning_rate": 1.6129032258064516e-06,
      "loss": 0.0586,
      "max_memory_allocated (GB)": 83.39,
      "memory_allocated (GB)": 19.37,
      "step": 130,
      "total_memory_available (GB)": 93.74
    },
    {
      "epoch": 4.52,
      "learning_rate": 9.67741935483871e-07,
      "loss": 0.0555,
      "max_memory_allocated (GB)": 83.39,
      "memory_allocated (GB)": 19.37,
      "step": 140,
      "total_memory_available (GB)": 93.74
    },
    {
      "epoch": 4.84,
      "learning_rate": 3.2258064516129035e-07,
      "loss": 0.0527,
      "max_memory_allocated (GB)": 83.39,
      "memory_allocated (GB)": 19.37,
      "step": 150,
      "total_memory_available (GB)": 93.74
    },
    {
      "epoch": 5.0,
      "max_memory_allocated (GB)": 83.39,
      "memory_allocated (GB)": 19.37,
      "step": 155,
      "total_flos": 3.0973626685587456e+16,
      "total_memory_available (GB)": 93.74,
      "train_loss": 0.07280745967741936,
      "train_runtime": 346.0342,
      "train_samples_per_second": 638.618,
      "train_steps_per_second": 2.022
    }
  ],
  "max_steps": 155,
  "num_train_epochs": 5,
  "total_flos": 3.0973626685587456e+16,
  "trial_name": null,
  "trial_params": null
}