{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "grad_norm": 2.1588942222092173,
      "learning_rate": 5e-06,
      "loss": 1.21,
      "step": 10
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.5217404430350148,
      "learning_rate": 5e-06,
      "loss": 1.0846,
      "step": 20
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.5500543610331259,
      "learning_rate": 5e-06,
      "loss": 1.0531,
      "step": 30
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.3628030482205304,
      "learning_rate": 5e-06,
      "loss": 1.0295,
      "step": 40
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.4313705936329209,
      "learning_rate": 5e-06,
      "loss": 1.0065,
      "step": 50
    },
    {
      "epoch": 0.6,
      "grad_norm": 1.3903725450273834,
      "learning_rate": 5e-06,
      "loss": 0.9979,
      "step": 60
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.0515922083105305,
      "learning_rate": 5e-06,
      "loss": 0.9822,
      "step": 70
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.7986272446943243,
      "learning_rate": 5e-06,
      "loss": 0.9709,
      "step": 80
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5959851594356917,
      "learning_rate": 5e-06,
      "loss": 0.9671,
      "step": 90
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.6602684302756796,
      "learning_rate": 5e-06,
      "loss": 0.9551,
      "step": 100
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.9628956317901611,
      "eval_runtime": 10.1031,
      "eval_samples_per_second": 265.563,
      "eval_steps_per_second": 1.089,
      "step": 100
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.6125962802138768,
      "learning_rate": 5e-06,
      "loss": 0.9153,
      "step": 110
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.769098378899703,
      "learning_rate": 5e-06,
      "loss": 0.9161,
      "step": 120
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.5913464424447598,
      "learning_rate": 5e-06,
      "loss": 0.9052,
      "step": 130
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.6372987877644805,
      "learning_rate": 5e-06,
      "loss": 0.9066,
      "step": 140
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.5028280444009187,
      "learning_rate": 5e-06,
      "loss": 0.9017,
      "step": 150
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.5192521745339289,
      "learning_rate": 5e-06,
      "loss": 0.8985,
      "step": 160
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.7069503869754896,
      "learning_rate": 5e-06,
      "loss": 0.9017,
      "step": 170
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.9508492067333021,
      "learning_rate": 5e-06,
      "loss": 0.8965,
      "step": 180
    },
    {
      "epoch": 1.9,
      "grad_norm": 0.43433200716899767,
      "learning_rate": 5e-06,
      "loss": 0.897,
      "step": 190
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.4981549070149512,
      "learning_rate": 5e-06,
      "loss": 0.8934,
      "step": 200
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.9496641159057617,
      "eval_runtime": 10.0777,
      "eval_samples_per_second": 266.231,
      "eval_steps_per_second": 1.092,
      "step": 200
    },
    {
      "epoch": 2.1,
      "grad_norm": 0.6575120061353793,
      "learning_rate": 5e-06,
      "loss": 0.8388,
      "step": 210
    },
    {
      "epoch": 2.2,
      "grad_norm": 0.6846093193242142,
      "learning_rate": 5e-06,
      "loss": 0.8356,
      "step": 220
    },
    {
      "epoch": 2.3,
      "grad_norm": 0.710451571620139,
      "learning_rate": 5e-06,
      "loss": 0.838,
      "step": 230
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.4914600404862552,
      "learning_rate": 5e-06,
      "loss": 0.8426,
      "step": 240
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.4780619064954024,
      "learning_rate": 5e-06,
      "loss": 0.845,
      "step": 250
    },
    {
      "epoch": 2.6,
      "grad_norm": 0.5079666890561413,
      "learning_rate": 5e-06,
      "loss": 0.8307,
      "step": 260
    },
    {
      "epoch": 2.7,
      "grad_norm": 0.5689090956024323,
      "learning_rate": 5e-06,
      "loss": 0.8391,
      "step": 270
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.5601234349656946,
      "learning_rate": 5e-06,
      "loss": 0.8426,
      "step": 280
    },
    {
      "epoch": 2.9,
      "grad_norm": 0.5256031398066893,
      "learning_rate": 5e-06,
      "loss": 0.8447,
      "step": 290
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.7628735928141229,
      "learning_rate": 5e-06,
      "loss": 0.841,
      "step": 300
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.9554556012153625,
      "eval_runtime": 9.9342,
      "eval_samples_per_second": 270.078,
      "eval_steps_per_second": 1.107,
      "step": 300
    },
    {
      "epoch": 3.0,
      "step": 300,
      "total_flos": 502511173632000.0,
      "train_loss": 0.9228995005289714,
      "train_runtime": 2189.1071,
      "train_samples_per_second": 69.856,
      "train_steps_per_second": 0.137
    }
  ],
  "logging_steps": 10,
  "max_steps": 300,
  "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": 502511173632000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}