last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0004034980174280105,
  "eval_steps": 10,
  "global_step": 13,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.103830903292389e-05,
      "eval_loss": 3.7120916843414307,
      "eval_runtime": 1355.8503,
      "eval_samples_per_second": 10.006,
      "eval_steps_per_second": 5.003,
      "step": 1
    },
    {
      "epoch": 0.00015519154516461942,
      "grad_norm": 2.1552507877349854,
      "learning_rate": 5e-05,
      "loss": 3.9595,
      "step": 5
    },
    {
      "epoch": 0.00031038309032923884,
      "grad_norm": 3.6430537700653076,
      "learning_rate": 0.0001,
      "loss": 3.5018,
      "step": 10
    },
    {
      "epoch": 0.00031038309032923884,
      "eval_loss": 3.1252193450927734,
      "eval_runtime": 1356.105,
      "eval_samples_per_second": 10.004,
      "eval_steps_per_second": 5.002,
      "step": 10
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 13,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4071873105100800.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}