trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 3600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.25,
      "grad_norm": 149.0,
      "learning_rate": 1e-05,
      "loss": 1.8187,
      "step": 500
    },
    {
      "epoch": 2.5,
      "grad_norm": 56.5,
      "learning_rate": 1e-05,
      "loss": 1.6052,
      "step": 1000
    },
    {
      "epoch": 3.75,
      "grad_norm": 520.0,
      "learning_rate": 1e-05,
      "loss": 1.4004,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "grad_norm": 92.0,
      "learning_rate": 1e-05,
      "loss": 0.2656,
      "step": 2000
    },
    {
      "epoch": 6.25,
      "grad_norm": 208.0,
      "learning_rate": 1e-05,
      "loss": -0.6543,
      "step": 2500
    },
    {
      "epoch": 7.5,
      "grad_norm": 87.0,
      "learning_rate": 1e-05,
      "loss": -0.9582,
      "step": 3000
    },
    {
      "epoch": 8.75,
      "grad_norm": 29.875,
      "learning_rate": 1e-05,
      "loss": -0.8479,
      "step": 3500
    }
  ],
  "logging_steps": 500,
  "max_steps": 4000,
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}