last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 2.0,
      "grad_norm": NaN,
      "learning_rate": 8.888888888888889e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 3.0,
      "grad_norm": NaN,
      "learning_rate": 7.777777777777778e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 4.0,
      "grad_norm": NaN,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 5.0,
      "grad_norm": NaN,
      "learning_rate": 5.555555555555556e-05,
      "loss": 3.0425,
      "step": 5
    },
    {
      "epoch": 6.0,
      "grad_norm": NaN,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 7.0,
      "grad_norm": NaN,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 8.0,
      "grad_norm": NaN,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 9.0,
      "grad_norm": NaN,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 10.0,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 10
    }
  ],
  "logging_steps": 100,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "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": 4662133875351552.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}