{
  "best_global_step": 40,
  "best_metric": 0.9486833214759827,
  "best_model_checkpoint": "/data/uqlinh/tallrec/output-yelp-clean-7b-20260506/checkpoint-40",
  "epoch": 0.5688888888888889,
  "eval_steps": 20,
  "global_step": 40,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14222222222222222,
      "grad_norm": 2.703125,
      "learning_rate": 4.5e-06,
      "loss": 0.2308,
      "step": 10
    },
    {
      "epoch": 0.28444444444444444,
      "grad_norm": 1.6953125,
      "learning_rate": 9.5e-06,
      "loss": 0.0726,
      "step": 20
    },
    {
      "epoch": 0.28444444444444444,
      "eval_loss": 0.07094312459230423,
      "eval_runtime": 53.7702,
      "eval_samples_per_second": 37.195,
      "eval_steps_per_second": 2.325,
      "step": 20
    },
    {
      "epoch": 0.4266666666666667,
      "grad_norm": 1.171875,
      "learning_rate": 9.533678756476683e-06,
      "loss": 0.0714,
      "step": 30
    },
    {
      "epoch": 0.5688888888888889,
      "grad_norm": 0.578125,
      "learning_rate": 9.015544041450778e-06,
      "loss": 0.0697,
      "step": 40
    },
    {
      "epoch": 0.5688888888888889,
      "eval_loss": 0.071051687002182,
      "eval_runtime": 53.705,
      "eval_samples_per_second": 37.241,
      "eval_steps_per_second": 2.328,
      "step": 40
    }
  ],
  "logging_steps": 10,
  "max_steps": 213,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 20,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 20,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.223876583513129e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}