trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.271259147110775,
  "eval_steps": 500,
  "global_step": 9001,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.19,
      "learning_rate": 4.8952813525107246e-05,
      "loss": 2.8381,
      "step": 500
    },
    {
      "epoch": 0.38,
      "learning_rate": 4.790142148204223e-05,
      "loss": 2.3291,
      "step": 1000
    },
    {
      "epoch": 0.47,
      "learning_rate": 4.7377273205087775e-05,
      "loss": 1.8496,
      "step": 1500
    },
    {
      "epoch": 0.63,
      "learning_rate": 4.650303094011704e-05,
      "loss": 2.0352,
      "step": 2000
    },
    {
      "epoch": 0.63,
      "learning_rate": 4.650096728067962e-05,
      "loss": 1.2524,
      "step": 2500
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.580003925196961e-05,
      "loss": 1.9539,
      "step": 3000
    },
    {
      "epoch": 0.88,
      "learning_rate": 4.510051307931702e-05,
      "loss": 1.8607,
      "step": 3500
    },
    {
      "epoch": 1.01,
      "learning_rate": 4.4399585050607007e-05,
      "loss": 1.7809,
      "step": 4000
    },
    {
      "epoch": 1.14,
      "learning_rate": 4.369865702189699e-05,
      "loss": 1.297,
      "step": 4500
    },
    {
      "epoch": 1.26,
      "learning_rate": 4.299772899318698e-05,
      "loss": 1.3119,
      "step": 5000
    },
    {
      "epoch": 1.39,
      "learning_rate": 4.229820282053439e-05,
      "loss": 1.343,
      "step": 5500
    },
    {
      "epoch": 1.51,
      "learning_rate": 4.1597274791824374e-05,
      "loss": 1.2962,
      "step": 6000
    },
    {
      "epoch": 1.64,
      "learning_rate": 4.0897748619171785e-05,
      "loss": 1.2929,
      "step": 6500
    },
    {
      "epoch": 1.77,
      "learning_rate": 4.019682059046178e-05,
      "loss": 1.2726,
      "step": 7000
    },
    {
      "epoch": 1.89,
      "learning_rate": 3.949589256175176e-05,
      "loss": 1.2686,
      "step": 7500
    },
    {
      "epoch": 2.02,
      "learning_rate": 3.879496453304175e-05,
      "loss": 1.1702,
      "step": 8000
    },
    {
      "epoch": 2.14,
      "learning_rate": 3.8094036504331735e-05,
      "loss": 0.8145,
      "step": 8500
    },
    {
      "epoch": 2.27,
      "learning_rate": 3.739310847562173e-05,
      "loss": 0.8225,
      "step": 9000
    },
    {
      "epoch": 2.27,
      "step": 9001,
      "total_flos": 5.460691370980147e+16,
      "train_loss": 5.4469778463318936e-05,
      "train_runtime": 1.6018,
      "train_samples_per_second": 89898.478,
      "train_steps_per_second": 5618.655
    }
  ],
  "logging_steps": 500,
  "max_steps": 9000,
  "num_train_epochs": 3,
  "save_steps": 1000,
  "total_flos": 5.460691370980147e+16,
  "trial_name": null,
  "trial_params": null
}