checkpoint-50/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.0,
      "grad_norm": 0.8825879693031311,
      "learning_rate": 8.2e-05,
      "loss": 1.669,
      "step": 10
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.5425239205360413,
      "learning_rate": 6.2e-05,
      "loss": 1.4806,
      "step": 20
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.5552927851676941,
      "learning_rate": 4.2e-05,
      "loss": 1.2466,
      "step": 30
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.46205174922943115,
      "learning_rate": 2.2000000000000003e-05,
      "loss": 1.0795,
      "step": 40
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.4868693947792053,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 1.0003,
      "step": 50
    }
  ],
  "logging_steps": 10,
  "max_steps": 50,
  "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": 1.70759651328e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}