{
  "best_global_step": 30,
  "best_metric": 0.31936541199684143,
  "best_model_checkpoint": "/tmp/checkpoints_/job_2cd1f599-85bf-4aa0-a3cf-f84970e4883a/checkpoint-30",
  "epoch": 0.6666666666666666,
  "eval_steps": 10,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.022222222222222223,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 4.059433460235596,
      "step": 1
    },
    {
      "epoch": 0.044444444444444446,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 3.624906063079834,
      "step": 2
    },
    {
      "epoch": 0.06666666666666667,
      "grad_norm": 12.42263412475586,
      "learning_rate": 0.0,
      "loss": 4.020474910736084,
      "step": 3
    },
    {
      "epoch": 0.08888888888888889,
      "grad_norm": 13.56965446472168,
      "learning_rate": 1e-05,
      "loss": 3.645026922225952,
      "step": 4
    },
    {
      "epoch": 0.1111111111111111,
      "grad_norm": 13.113579750061035,
      "learning_rate": 2e-05,
      "loss": 4.158425331115723,
      "step": 5
    },
    {
      "epoch": 0.13333333333333333,
      "grad_norm": 12.33988094329834,
      "learning_rate": 3e-05,
      "loss": 3.8069803714752197,
      "step": 6
    },
    {
      "epoch": 0.15555555555555556,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 3.160885810852051,
      "step": 7
    },
    {
      "epoch": 0.17777777777777778,
      "grad_norm": 12.848119735717773,
      "learning_rate": 4e-05,
      "loss": 3.30314302444458,
      "step": 8
    },
    {
      "epoch": 0.2,
      "grad_norm": 11.116307258605957,
      "learning_rate": 5e-05,
      "loss": 2.979320526123047,
      "step": 9
    },
    {
      "epoch": 0.2222222222222222,
      "grad_norm": 9.405905723571777,
      "learning_rate": 4.961538461538462e-05,
      "loss": 2.488369941711426,
      "step": 10
    },
    {
      "epoch": 0.2222222222222222,
      "eval_loss": 2.172542095184326,
      "eval_runtime": 1.4397,
      "eval_samples_per_second": 13.892,
      "eval_steps_per_second": 3.473,
      "step": 10
    },
    {
      "epoch": 0.24444444444444444,
      "grad_norm": 8.405779838562012,
      "learning_rate": 4.923076923076924e-05,
      "loss": 2.295246124267578,
      "step": 11
    },
    {
      "epoch": 0.26666666666666666,
      "grad_norm": 7.363292217254639,
      "learning_rate": 4.884615384615385e-05,
      "loss": 2.2567129135131836,
      "step": 12
    },
    {
      "epoch": 0.28888888888888886,
      "grad_norm": 7.1502790451049805,
      "learning_rate": 4.846153846153846e-05,
      "loss": 1.823792815208435,
      "step": 13
    },
    {
      "epoch": 0.3111111111111111,
      "grad_norm": 6.763433933258057,
      "learning_rate": 4.8076923076923084e-05,
      "loss": 1.5706839561462402,
      "step": 14
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 6.633345127105713,
      "learning_rate": 4.76923076923077e-05,
      "loss": 1.5136253833770752,
      "step": 15
    },
    {
      "epoch": 0.35555555555555557,
      "grad_norm": 6.369776248931885,
      "learning_rate": 4.730769230769231e-05,
      "loss": 1.6171478033065796,
      "step": 16
    },
    {
      "epoch": 0.37777777777777777,
      "grad_norm": 6.243237018585205,
      "learning_rate": 4.692307692307693e-05,
      "loss": 1.0360909700393677,
      "step": 17
    },
    {
      "epoch": 0.4,
      "grad_norm": 7.544528484344482,
      "learning_rate": 4.653846153846154e-05,
      "loss": 1.2657012939453125,
      "step": 18
    },
    {
      "epoch": 0.4222222222222222,
      "grad_norm": 5.462157726287842,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.9959335327148438,
      "step": 19
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 5.940985679626465,
      "learning_rate": 4.576923076923077e-05,
      "loss": 1.0224928855895996,
      "step": 20
    },
    {
      "epoch": 0.4444444444444444,
      "eval_loss": 0.6980130076408386,
      "eval_runtime": 0.746,
      "eval_samples_per_second": 26.808,
      "eval_steps_per_second": 6.702,
      "step": 20
    },
    {
      "epoch": 0.4666666666666667,
      "grad_norm": 6.72797155380249,
      "learning_rate": 4.538461538461539e-05,
      "loss": 0.8094964623451233,
      "step": 21
    },
    {
      "epoch": 0.4888888888888889,
      "grad_norm": 3.994920492172241,
      "learning_rate": 4.5e-05,
      "loss": 0.6374857425689697,
      "step": 22
    },
    {
      "epoch": 0.5111111111111111,
      "grad_norm": 4.263758659362793,
      "learning_rate": 4.461538461538462e-05,
      "loss": 0.5003803968429565,
      "step": 23
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 5.624288558959961,
      "learning_rate": 4.423076923076923e-05,
      "loss": 0.5343151688575745,
      "step": 24
    },
    {
      "epoch": 0.5555555555555556,
      "grad_norm": 4.008190631866455,
      "learning_rate": 4.384615384615385e-05,
      "loss": 0.6797525882720947,
      "step": 25
    },
    {
      "epoch": 0.5777777777777777,
      "grad_norm": 4.962954521179199,
      "learning_rate": 4.346153846153846e-05,
      "loss": 0.8150541186332703,
      "step": 26
    },
    {
      "epoch": 0.6,
      "grad_norm": 3.9157612323760986,
      "learning_rate": 4.3076923076923084e-05,
      "loss": 0.5314605832099915,
      "step": 27
    },
    {
      "epoch": 0.6222222222222222,
      "grad_norm": 6.704800128936768,
      "learning_rate": 4.269230769230769e-05,
      "loss": 0.4506711959838867,
      "step": 28
    },
    {
      "epoch": 0.6444444444444445,
      "grad_norm": 3.85221004486084,
      "learning_rate": 4.230769230769231e-05,
      "loss": 0.6027767658233643,
      "step": 29
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 3.5128495693206787,
      "learning_rate": 4.192307692307693e-05,
      "loss": 0.44177567958831787,
      "step": 30
    },
    {
      "epoch": 0.6666666666666666,
      "eval_loss": 0.31936541199684143,
      "eval_runtime": 0.7586,
      "eval_samples_per_second": 26.364,
      "eval_steps_per_second": 6.591,
      "step": 30
    }
  ],
  "logging_steps": 1,
  "max_steps": 135,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 27328409944320.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}