{
  "best_global_step": 3000,
  "best_metric": 1.5481228828430176,
  "best_model_checkpoint": "./models/checkpoint-3000",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 9405,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1595150741745095,
      "grad_norm": 6.21875,
      "learning_rate": 4.73471557682084e-05,
      "loss": 1.6533,
      "step": 500
    },
    {
      "epoch": 0.1595150741745095,
      "eval_loss": 1.670200228691101,
      "eval_runtime": 1.8707,
      "eval_samples_per_second": 53.457,
      "eval_steps_per_second": 13.364,
      "step": 500
    },
    {
      "epoch": 0.319030148349019,
      "grad_norm": 6.5,
      "learning_rate": 4.468899521531101e-05,
      "loss": 1.5746,
      "step": 1000
    },
    {
      "epoch": 0.319030148349019,
      "eval_loss": 1.6311461925506592,
      "eval_runtime": 1.8699,
      "eval_samples_per_second": 53.48,
      "eval_steps_per_second": 13.37,
      "step": 1000
    },
    {
      "epoch": 0.4785452225235285,
      "grad_norm": 6.90625,
      "learning_rate": 4.203083466241361e-05,
      "loss": 1.5591,
      "step": 1500
    },
    {
      "epoch": 0.4785452225235285,
      "eval_loss": 1.6131033897399902,
      "eval_runtime": 1.8703,
      "eval_samples_per_second": 53.466,
      "eval_steps_per_second": 13.367,
      "step": 1500
    },
    {
      "epoch": 0.638060296698038,
      "grad_norm": 6.0,
      "learning_rate": 3.9372674109516216e-05,
      "loss": 1.5301,
      "step": 2000
    },
    {
      "epoch": 0.638060296698038,
      "eval_loss": 1.582728624343872,
      "eval_runtime": 1.8741,
      "eval_samples_per_second": 53.358,
      "eval_steps_per_second": 13.34,
      "step": 2000
    },
    {
      "epoch": 0.7975753708725475,
      "grad_norm": 6.09375,
      "learning_rate": 3.671451355661882e-05,
      "loss": 1.5222,
      "step": 2500
    },
    {
      "epoch": 0.7975753708725475,
      "eval_loss": 1.5706853866577148,
      "eval_runtime": 1.8744,
      "eval_samples_per_second": 53.351,
      "eval_steps_per_second": 13.338,
      "step": 2500
    },
    {
      "epoch": 0.957090445047057,
      "grad_norm": 6.1875,
      "learning_rate": 3.405635300372142e-05,
      "loss": 1.4987,
      "step": 3000
    },
    {
      "epoch": 0.957090445047057,
      "eval_loss": 1.5481228828430176,
      "eval_runtime": 1.875,
      "eval_samples_per_second": 53.335,
      "eval_steps_per_second": 13.334,
      "step": 3000
    },
    {
      "epoch": 1.116446004147392,
      "grad_norm": 6.21875,
      "learning_rate": 3.1398192450824035e-05,
      "loss": 1.3145,
      "step": 3500
    },
    {
      "epoch": 1.116446004147392,
      "eval_loss": 1.565281629562378,
      "eval_runtime": 1.8772,
      "eval_samples_per_second": 53.272,
      "eval_steps_per_second": 13.318,
      "step": 3500
    },
    {
      "epoch": 1.2759610783219015,
      "grad_norm": 7.28125,
      "learning_rate": 2.8740031897926635e-05,
      "loss": 1.2554,
      "step": 4000
    },
    {
      "epoch": 1.2759610783219015,
      "eval_loss": 1.5643690824508667,
      "eval_runtime": 1.8766,
      "eval_samples_per_second": 53.287,
      "eval_steps_per_second": 13.322,
      "step": 4000
    },
    {
      "epoch": 1.435476152496411,
      "grad_norm": 7.0,
      "learning_rate": 2.608187134502924e-05,
      "loss": 1.2597,
      "step": 4500
    },
    {
      "epoch": 1.435476152496411,
      "eval_loss": 1.562727451324463,
      "eval_runtime": 1.8896,
      "eval_samples_per_second": 52.92,
      "eval_steps_per_second": 13.23,
      "step": 4500
    },
    {
      "epoch": 1.5949912266709205,
      "grad_norm": 6.84375,
      "learning_rate": 2.3423710792131844e-05,
      "loss": 1.2496,
      "step": 5000
    },
    {
      "epoch": 1.5949912266709205,
      "eval_loss": 1.560770869255066,
      "eval_runtime": 1.8771,
      "eval_samples_per_second": 53.273,
      "eval_steps_per_second": 13.318,
      "step": 5000
    },
    {
      "epoch": 1.75450630084543,
      "grad_norm": 6.90625,
      "learning_rate": 2.076555023923445e-05,
      "loss": 1.2422,
      "step": 5500
    },
    {
      "epoch": 1.75450630084543,
      "eval_loss": 1.5582979917526245,
      "eval_runtime": 1.8771,
      "eval_samples_per_second": 53.274,
      "eval_steps_per_second": 13.319,
      "step": 5500
    },
    {
      "epoch": 1.9140213750199395,
      "grad_norm": 7.4375,
      "learning_rate": 1.8107389686337057e-05,
      "loss": 1.2432,
      "step": 6000
    },
    {
      "epoch": 1.9140213750199395,
      "eval_loss": 1.5532492399215698,
      "eval_runtime": 1.8789,
      "eval_samples_per_second": 53.222,
      "eval_steps_per_second": 13.306,
      "step": 6000
    },
    {
      "epoch": 2.0733769341202746,
      "grad_norm": 7.5,
      "learning_rate": 1.544922913343966e-05,
      "loss": 1.1977,
      "step": 6500
    },
    {
      "epoch": 2.0733769341202746,
      "eval_loss": 1.5878651142120361,
      "eval_runtime": 1.8781,
      "eval_samples_per_second": 53.244,
      "eval_steps_per_second": 13.311,
      "step": 6500
    },
    {
      "epoch": 2.232892008294784,
      "grad_norm": 7.53125,
      "learning_rate": 1.2791068580542265e-05,
      "loss": 1.15,
      "step": 7000
    },
    {
      "epoch": 2.232892008294784,
      "eval_loss": 1.5915130376815796,
      "eval_runtime": 1.8844,
      "eval_samples_per_second": 53.068,
      "eval_steps_per_second": 13.267,
      "step": 7000
    },
    {
      "epoch": 2.392407082469293,
      "grad_norm": 7.6875,
      "learning_rate": 1.0132908027644871e-05,
      "loss": 1.1389,
      "step": 7500
    },
    {
      "epoch": 2.392407082469293,
      "eval_loss": 1.5931323766708374,
      "eval_runtime": 1.8778,
      "eval_samples_per_second": 53.255,
      "eval_steps_per_second": 13.314,
      "step": 7500
    },
    {
      "epoch": 2.551922156643803,
      "grad_norm": 7.4375,
      "learning_rate": 7.474747474747475e-06,
      "loss": 1.1464,
      "step": 8000
    },
    {
      "epoch": 2.551922156643803,
      "eval_loss": 1.5903747081756592,
      "eval_runtime": 1.8791,
      "eval_samples_per_second": 53.216,
      "eval_steps_per_second": 13.304,
      "step": 8000
    },
    {
      "epoch": 2.711437230818312,
      "grad_norm": 7.46875,
      "learning_rate": 4.816586921850081e-06,
      "loss": 1.1446,
      "step": 8500
    },
    {
      "epoch": 2.711437230818312,
      "eval_loss": 1.590154767036438,
      "eval_runtime": 1.8787,
      "eval_samples_per_second": 53.23,
      "eval_steps_per_second": 13.307,
      "step": 8500
    },
    {
      "epoch": 2.870952304992822,
      "grad_norm": 8.125,
      "learning_rate": 2.158426368952685e-06,
      "loss": 1.1444,
      "step": 9000
    },
    {
      "epoch": 2.870952304992822,
      "eval_loss": 1.5894922018051147,
      "eval_runtime": 1.8789,
      "eval_samples_per_second": 53.221,
      "eval_steps_per_second": 13.305,
      "step": 9000
    }
  ],
  "logging_steps": 500,
  "max_steps": 9405,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 2.035763059359744e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}