trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 99,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.30303030303030304,
      "grad_norm": 8.750067710876465,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 3.3586,
      "step": 10
    },
    {
      "epoch": 0.6060606060606061,
      "grad_norm": 9.16357421875,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 3.332,
      "step": 20
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 8.869489669799805,
      "learning_rate": 3e-06,
      "loss": 3.3107,
      "step": 30
    },
    {
      "epoch": 1.2121212121212122,
      "grad_norm": 8.31998348236084,
      "learning_rate": 4.000000000000001e-06,
      "loss": 3.2271,
      "step": 40
    },
    {
      "epoch": 1.5151515151515151,
      "grad_norm": 8.316051483154297,
      "learning_rate": 5e-06,
      "loss": 3.0808,
      "step": 50
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 8.299915313720703,
      "learning_rate": 6e-06,
      "loss": 3.0593,
      "step": 60
    },
    {
      "epoch": 2.121212121212121,
      "grad_norm": 7.488762855529785,
      "learning_rate": 7.000000000000001e-06,
      "loss": 2.8781,
      "step": 70
    },
    {
      "epoch": 2.4242424242424243,
      "grad_norm": 8.20336627960205,
      "learning_rate": 8.000000000000001e-06,
      "loss": 2.731,
      "step": 80
    },
    {
      "epoch": 2.7272727272727275,
      "grad_norm": 9.079769134521484,
      "learning_rate": 9e-06,
      "loss": 2.5706,
      "step": 90
    }
  ],
  "logging_steps": 10,
  "max_steps": 99,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.20913507233792e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}