{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.8095238095238093,
  "eval_steps": 500,
  "global_step": 40,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.19047619047619047,
      "grad_norm": 1.0338395833969116,
      "learning_rate": 0.00019,
      "loss": 9.5594,
      "step": 2
    },
    {
      "epoch": 0.38095238095238093,
      "grad_norm": 1.250166654586792,
      "learning_rate": 0.00018,
      "loss": 9.2309,
      "step": 4
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 1.9235066175460815,
      "learning_rate": 0.00017,
      "loss": 8.8061,
      "step": 6
    },
    {
      "epoch": 0.7619047619047619,
      "grad_norm": 2.2285144329071045,
      "learning_rate": 0.00016,
      "loss": 8.2859,
      "step": 8
    },
    {
      "epoch": 0.9523809523809523,
      "grad_norm": 1.434334397315979,
      "learning_rate": 0.00015000000000000001,
      "loss": 7.9618,
      "step": 10
    },
    {
      "epoch": 1.1428571428571428,
      "grad_norm": 1.1094856262207031,
      "learning_rate": 0.00014,
      "loss": 7.7526,
      "step": 12
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 0.9047002196311951,
      "learning_rate": 0.00013000000000000002,
      "loss": 7.5692,
      "step": 14
    },
    {
      "epoch": 1.5238095238095237,
      "grad_norm": 1.0015796422958374,
      "learning_rate": 0.00012,
      "loss": 7.4736,
      "step": 16
    },
    {
      "epoch": 1.7142857142857144,
      "grad_norm": 1.095115065574646,
      "learning_rate": 0.00011000000000000002,
      "loss": 7.487,
      "step": 18
    },
    {
      "epoch": 1.9047619047619047,
      "grad_norm": 0.8204529285430908,
      "learning_rate": 0.0001,
      "loss": 7.4386,
      "step": 20
    },
    {
      "epoch": 2.0952380952380953,
      "grad_norm": 0.8284938931465149,
      "learning_rate": 9e-05,
      "loss": 7.446,
      "step": 22
    },
    {
      "epoch": 2.2857142857142856,
      "grad_norm": 0.8024454712867737,
      "learning_rate": 8e-05,
      "loss": 7.3784,
      "step": 24
    },
    {
      "epoch": 2.4761904761904763,
      "grad_norm": 1.7449926137924194,
      "learning_rate": 7e-05,
      "loss": 7.2682,
      "step": 26
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.6010074019432068,
      "learning_rate": 6e-05,
      "loss": 7.317,
      "step": 28
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 0.7939924597740173,
      "learning_rate": 5e-05,
      "loss": 7.3338,
      "step": 30
    },
    {
      "epoch": 3.0476190476190474,
      "grad_norm": 0.5959632992744446,
      "learning_rate": 4e-05,
      "loss": 7.2549,
      "step": 32
    },
    {
      "epoch": 3.238095238095238,
      "grad_norm": 0.780635416507721,
      "learning_rate": 3e-05,
      "loss": 7.2918,
      "step": 34
    },
    {
      "epoch": 3.4285714285714284,
      "grad_norm": 0.7039668560028076,
      "learning_rate": 2e-05,
      "loss": 7.3645,
      "step": 36
    },
    {
      "epoch": 3.619047619047619,
      "grad_norm": 0.5156592726707458,
      "learning_rate": 1e-05,
      "loss": 7.2357,
      "step": 38
    },
    {
      "epoch": 3.8095238095238093,
      "grad_norm": 0.4115481376647949,
      "learning_rate": 0.0,
      "loss": 7.2353,
      "step": 40
    },
    {
      "epoch": 3.8095238095238093,
      "step": 40,
      "total_flos": 200497171101768.0,
      "train_loss": 7.734543776512146,
      "train_runtime": 169.1152,
      "train_samples_per_second": 3.974,
      "train_steps_per_second": 0.237
    }
  ],
  "logging_steps": 2,
  "max_steps": 40,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 200497171101768.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}