{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0032224280995730285,
  "eval_steps": 5,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0001611214049786514,
      "grad_norm": 1.0841774940490723,
      "learning_rate": 1e-05,
      "loss": 0.6736,
      "step": 1
    },
    {
      "epoch": 0.0001611214049786514,
      "eval_loss": 0.7637655735015869,
      "eval_runtime": 145.2772,
      "eval_samples_per_second": 71.959,
      "eval_steps_per_second": 35.979,
      "step": 1
    },
    {
      "epoch": 0.0003222428099573028,
      "grad_norm": 1.151691198348999,
      "learning_rate": 2e-05,
      "loss": 0.7124,
      "step": 2
    },
    {
      "epoch": 0.00048336421493595425,
      "grad_norm": 1.1659541130065918,
      "learning_rate": 3e-05,
      "loss": 0.7709,
      "step": 3
    },
    {
      "epoch": 0.0006444856199146056,
      "grad_norm": 1.1289387941360474,
      "learning_rate": 4e-05,
      "loss": 0.7062,
      "step": 4
    },
    {
      "epoch": 0.0008056070248932571,
      "grad_norm": 1.1830432415008545,
      "learning_rate": 5e-05,
      "loss": 0.6925,
      "step": 5
    },
    {
      "epoch": 0.0008056070248932571,
      "eval_loss": 0.6898554563522339,
      "eval_runtime": 74.4675,
      "eval_samples_per_second": 140.383,
      "eval_steps_per_second": 70.192,
      "step": 5
    },
    {
      "epoch": 0.0009667284298719085,
      "grad_norm": 1.3095643520355225,
      "learning_rate": 6e-05,
      "loss": 0.7211,
      "step": 6
    },
    {
      "epoch": 0.00112784983485056,
      "grad_norm": 1.2183135747909546,
      "learning_rate": 7e-05,
      "loss": 0.6138,
      "step": 7
    },
    {
      "epoch": 0.0012889712398292113,
      "grad_norm": 1.2897781133651733,
      "learning_rate": 8e-05,
      "loss": 0.5534,
      "step": 8
    },
    {
      "epoch": 0.0014500926448078628,
      "grad_norm": 1.2032550573349,
      "learning_rate": 9e-05,
      "loss": 0.5131,
      "step": 9
    },
    {
      "epoch": 0.0016112140497865142,
      "grad_norm": 1.0904922485351562,
      "learning_rate": 0.0001,
      "loss": 0.4467,
      "step": 10
    },
    {
      "epoch": 0.0016112140497865142,
      "eval_loss": 0.3901079595088959,
      "eval_runtime": 74.5268,
      "eval_samples_per_second": 140.272,
      "eval_steps_per_second": 70.136,
      "step": 10
    },
    {
      "epoch": 0.0017723354547651655,
      "grad_norm": 1.0180318355560303,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.3632,
      "step": 11
    },
    {
      "epoch": 0.001933456859743817,
      "grad_norm": 0.8413439989089966,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.2812,
      "step": 12
    },
    {
      "epoch": 0.0020945782647224685,
      "grad_norm": 0.7293118238449097,
      "learning_rate": 7.938926261462366e-05,
      "loss": 0.1871,
      "step": 13
    },
    {
      "epoch": 0.00225569966970112,
      "grad_norm": 0.953554630279541,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.2257,
      "step": 14
    },
    {
      "epoch": 0.002416821074679771,
      "grad_norm": 0.8305586576461792,
      "learning_rate": 5e-05,
      "loss": 0.1871,
      "step": 15
    },
    {
      "epoch": 0.002416821074679771,
      "eval_loss": 0.16657570004463196,
      "eval_runtime": 74.6167,
      "eval_samples_per_second": 140.103,
      "eval_steps_per_second": 70.051,
      "step": 15
    },
    {
      "epoch": 0.0025779424796584225,
      "grad_norm": 0.7156417369842529,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.1565,
      "step": 16
    },
    {
      "epoch": 0.002739063884637074,
      "grad_norm": 0.6914603114128113,
      "learning_rate": 2.061073738537635e-05,
      "loss": 0.1457,
      "step": 17
    },
    {
      "epoch": 0.0029001852896157255,
      "grad_norm": 0.6389317512512207,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.1318,
      "step": 18
    },
    {
      "epoch": 0.003061306694594377,
      "grad_norm": 0.7102766036987305,
      "learning_rate": 2.4471741852423237e-06,
      "loss": 0.1477,
      "step": 19
    },
    {
      "epoch": 0.0032224280995730285,
      "grad_norm": 0.5916241407394409,
      "learning_rate": 0.0,
      "loss": 0.1028,
      "step": 20
    },
    {
      "epoch": 0.0032224280995730285,
      "eval_loss": 0.13295121490955353,
      "eval_runtime": 74.6245,
      "eval_samples_per_second": 140.088,
      "eval_steps_per_second": 70.044,
      "step": 20
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1924364123504640.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}