{
  "best_metric": 4.9799017906188965,
  "best_model_checkpoint": "/home/datta0/models/lora_final/gemma-2-9b_pct_reverse_r32/checkpoint-8",
  "epoch": 0.020603621730382295,
  "eval_steps": 8,
  "global_step": 8,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002575452716297787,
      "grad_norm": 10.302180290222168,
      "learning_rate": 1.25e-05,
      "loss": 2.1766,
      "step": 1
    },
    {
      "epoch": 0.010301810865191148,
      "grad_norm": 7.941323280334473,
      "learning_rate": 5e-05,
      "loss": 2.2338,
      "step": 4
    },
    {
      "epoch": 0.020603621730382295,
      "grad_norm": 11.684494018554688,
      "learning_rate": 0.0001,
      "loss": 2.5978,
      "step": 8
    },
    {
      "epoch": 0.020603621730382295,
      "eval_loss": 4.9799017906188965,
      "eval_runtime": 400.2297,
      "eval_samples_per_second": 0.612,
      "eval_steps_per_second": 0.612,
      "step": 8
    }
  ],
  "logging_steps": 4,
  "max_steps": 388,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 8,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.212444633759744e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}