{
  "best_metric": 0.1455519050359726,
  "best_model_checkpoint": "for_test25/checkpoint-2000",
  "epoch": 4.7281323877068555,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2364066193853428,
      "grad_norm": 60.020355224609375,
      "learning_rate": 2.4e-05,
      "loss": 9.8742,
      "step": 100
    },
    {
      "epoch": 0.4728132387706856,
      "grad_norm": 4.358085632324219,
      "learning_rate": 4.9e-05,
      "loss": 4.8037,
      "step": 200
    },
    {
      "epoch": 0.7092198581560284,
      "grad_norm": 10.840340614318848,
      "learning_rate": 7.4e-05,
      "loss": 4.1932,
      "step": 300
    },
    {
      "epoch": 0.9456264775413712,
      "grad_norm": 22.108749389648438,
      "learning_rate": 9.900000000000001e-05,
      "loss": 3.9676,
      "step": 400
    },
    {
      "epoch": 1.1820330969267139,
      "grad_norm": 12.02284049987793,
      "learning_rate": 9.74934725848564e-05,
      "loss": 3.8466,
      "step": 500
    },
    {
      "epoch": 1.1820330969267139,
      "eval_PER": 0.9633641181124875,
      "eval_loss": 4.664893627166748,
      "eval_runtime": 420.6139,
      "eval_samples_per_second": 11.84,
      "eval_steps_per_second": 0.742,
      "learning_rate": 9.74934725848564e-05,
      "step": 500
    },
    {
      "epoch": 1.4184397163120568,
      "grad_norm": 10.632979393005371,
      "learning_rate": 9.488250652741515e-05,
      "loss": 3.6831,
      "step": 600
    },
    {
      "epoch": 1.6548463356973995,
      "grad_norm": 3.2810802459716797,
      "learning_rate": 9.22715404699739e-05,
      "loss": 3.7744,
      "step": 700
    },
    {
      "epoch": 1.8912529550827424,
      "grad_norm": 1.6217520236968994,
      "learning_rate": 8.966057441253264e-05,
      "loss": 3.4184,
      "step": 800
    },
    {
      "epoch": 2.127659574468085,
      "grad_norm": 12.222310066223145,
      "learning_rate": 8.704960835509139e-05,
      "loss": 3.7714,
      "step": 900
    },
    {
      "epoch": 2.3640661938534278,
      "grad_norm": 6.44509220123291,
      "learning_rate": 8.443864229765015e-05,
      "loss": 3.4804,
      "step": 1000
    },
    {
      "epoch": 2.3640661938534278,
      "eval_PER": 0.9612962852146624,
      "eval_loss": 3.0293383598327637,
      "eval_runtime": 424.1378,
      "eval_samples_per_second": 11.741,
      "eval_steps_per_second": 0.736,
      "learning_rate": 8.443864229765015e-05,
      "step": 1000
    },
    {
      "epoch": 2.6004728132387704,
      "grad_norm": 8.923386573791504,
      "learning_rate": 8.182767624020888e-05,
      "loss": 3.3555,
      "step": 1100
    },
    {
      "epoch": 2.8368794326241136,
      "grad_norm": 0.9465563297271729,
      "learning_rate": 7.921671018276762e-05,
      "loss": 2.2104,
      "step": 1200
    },
    {
      "epoch": 3.0732860520094563,
      "grad_norm": 0.8739014863967896,
      "learning_rate": 7.660574412532637e-05,
      "loss": 1.4531,
      "step": 1300
    },
    {
      "epoch": 3.309692671394799,
      "grad_norm": 0.7739846110343933,
      "learning_rate": 7.399477806788511e-05,
      "loss": 1.4319,
      "step": 1400
    },
    {
      "epoch": 3.546099290780142,
      "grad_norm": 3.5161337852478027,
      "learning_rate": 7.138381201044386e-05,
      "loss": 1.2535,
      "step": 1500
    },
    {
      "epoch": 3.546099290780142,
      "eval_PER": 0.10295114994115652,
      "eval_loss": 0.18029829859733582,
      "eval_runtime": 420.7727,
      "eval_samples_per_second": 11.835,
      "eval_steps_per_second": 0.741,
      "learning_rate": 7.138381201044386e-05,
      "step": 1500
    },
    {
      "epoch": 3.7825059101654848,
      "grad_norm": 2.1900475025177,
      "learning_rate": 6.877284595300262e-05,
      "loss": 1.1847,
      "step": 1600
    },
    {
      "epoch": 4.0189125295508275,
      "grad_norm": 1.6488945484161377,
      "learning_rate": 6.616187989556137e-05,
      "loss": 1.166,
      "step": 1700
    },
    {
      "epoch": 4.25531914893617,
      "grad_norm": 1.0157734155654907,
      "learning_rate": 6.355091383812011e-05,
      "loss": 1.6181,
      "step": 1800
    },
    {
      "epoch": 4.491725768321513,
      "grad_norm": 0.4091039299964905,
      "learning_rate": 6.093994778067885e-05,
      "loss": 1.4641,
      "step": 1900
    },
    {
      "epoch": 4.7281323877068555,
      "grad_norm": 0.38830655813217163,
      "learning_rate": 5.8328981723237596e-05,
      "loss": 1.304,
      "step": 2000
    },
    {
      "epoch": 4.7281323877068555,
      "eval_PER": 0.09349463759732445,
      "eval_loss": 0.1455519050359726,
      "eval_runtime": 417.7255,
      "eval_samples_per_second": 11.922,
      "eval_steps_per_second": 0.747,
      "learning_rate": 5.8328981723237596e-05,
      "step": 2000
    }
  ],
  "logging_steps": 100,
  "max_steps": 4230,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 1000,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "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": 1.1795037641339611e+19,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}