trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 10240,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.48828125,
      "grad_norm": 14.348142623901367,
      "learning_rate": 2e-05,
      "loss": 0.718,
      "step": 500
    },
    {
      "epoch": 0.9765625,
      "grad_norm": 12.407145500183105,
      "learning_rate": 2e-05,
      "loss": 0.6899,
      "step": 1000
    },
    {
      "epoch": 1.46484375,
      "grad_norm": 9.278654098510742,
      "learning_rate": 2e-05,
      "loss": 0.3397,
      "step": 1500
    },
    {
      "epoch": 1.953125,
      "grad_norm": 8.198233604431152,
      "learning_rate": 2e-05,
      "loss": 0.3419,
      "step": 2000
    },
    {
      "epoch": 2.44140625,
      "grad_norm": 4.592423915863037,
      "learning_rate": 2e-05,
      "loss": 0.1915,
      "step": 2500
    },
    {
      "epoch": 2.9296875,
      "grad_norm": 8.893723487854004,
      "learning_rate": 2e-05,
      "loss": 0.1922,
      "step": 3000
    },
    {
      "epoch": 3.41796875,
      "grad_norm": 8.089862823486328,
      "learning_rate": 2e-05,
      "loss": 0.1263,
      "step": 3500
    },
    {
      "epoch": 3.90625,
      "grad_norm": 6.3830485343933105,
      "learning_rate": 2e-05,
      "loss": 0.121,
      "step": 4000
    },
    {
      "epoch": 4.39453125,
      "grad_norm": 7.365416049957275,
      "learning_rate": 2e-05,
      "loss": 0.0897,
      "step": 4500
    },
    {
      "epoch": 4.8828125,
      "grad_norm": 7.9196600914001465,
      "learning_rate": 2e-05,
      "loss": 0.087,
      "step": 5000
    },
    {
      "epoch": 5.37109375,
      "grad_norm": 3.7600364685058594,
      "learning_rate": 2e-05,
      "loss": 0.0665,
      "step": 5500
    },
    {
      "epoch": 5.859375,
      "grad_norm": 4.597715854644775,
      "learning_rate": 2e-05,
      "loss": 0.0639,
      "step": 6000
    },
    {
      "epoch": 6.34765625,
      "grad_norm": 4.1985182762146,
      "learning_rate": 2e-05,
      "loss": 0.0543,
      "step": 6500
    },
    {
      "epoch": 6.8359375,
      "grad_norm": 5.26850700378418,
      "learning_rate": 2e-05,
      "loss": 0.0506,
      "step": 7000
    },
    {
      "epoch": 7.32421875,
      "grad_norm": 3.8958706855773926,
      "learning_rate": 2e-05,
      "loss": 0.049,
      "step": 7500
    },
    {
      "epoch": 7.8125,
      "grad_norm": 6.896234035491943,
      "learning_rate": 2e-05,
      "loss": 0.0451,
      "step": 8000
    },
    {
      "epoch": 8.30078125,
      "grad_norm": 3.6532912254333496,
      "learning_rate": 2e-05,
      "loss": 0.0431,
      "step": 8500
    },
    {
      "epoch": 8.7890625,
      "grad_norm": 4.900750160217285,
      "learning_rate": 2e-05,
      "loss": 0.0402,
      "step": 9000
    },
    {
      "epoch": 9.27734375,
      "grad_norm": 1.2737749814987183,
      "learning_rate": 2e-05,
      "loss": 0.0367,
      "step": 9500
    },
    {
      "epoch": 9.765625,
      "grad_norm": 5.822772026062012,
      "learning_rate": 2e-05,
      "loss": 0.0391,
      "step": 10000
    }
  ],
  "logging_steps": 500,
  "max_steps": 10240,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3193555713038746e+17,
  "train_batch_size": 3,
  "trial_name": null,
  "trial_params": null
}