{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.764705882352941,
  "eval_steps": 500,
  "global_step": 48,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "learning_rate": 9.989294616193018e-06,
      "loss": 2.3554,
      "step": 1
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.957224306869053e-06,
      "loss": 2.1848,
      "step": 2
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.903926402016153e-06,
      "loss": 1.9353,
      "step": 3
    },
    {
      "epoch": 0.31,
      "learning_rate": 9.829629131445342e-06,
      "loss": 1.6338,
      "step": 4
    },
    {
      "epoch": 0.39,
      "learning_rate": 9.73465064747553e-06,
      "loss": 1.5467,
      "step": 5
    },
    {
      "epoch": 0.47,
      "learning_rate": 9.619397662556434e-06,
      "loss": 1.6413,
      "step": 6
    },
    {
      "epoch": 0.55,
      "learning_rate": 9.484363707663443e-06,
      "loss": 1.5138,
      "step": 7
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.330127018922195e-06,
      "loss": 1.3815,
      "step": 8
    },
    {
      "epoch": 0.71,
      "learning_rate": 9.157348061512728e-06,
      "loss": 1.4179,
      "step": 9
    },
    {
      "epoch": 0.78,
      "learning_rate": 8.966766701456177e-06,
      "loss": 1.2832,
      "step": 10
    },
    {
      "epoch": 0.86,
      "learning_rate": 8.759199037394888e-06,
      "loss": 1.2613,
      "step": 11
    },
    {
      "epoch": 0.94,
      "learning_rate": 8.535533905932739e-06,
      "loss": 1.246,
      "step": 12
    },
    {
      "epoch": 1.02,
      "learning_rate": 8.296729075500345e-06,
      "loss": 1.3114,
      "step": 13
    },
    {
      "epoch": 1.1,
      "learning_rate": 8.043807145043604e-06,
      "loss": 1.1663,
      "step": 14
    },
    {
      "epoch": 1.18,
      "learning_rate": 7.777851165098012e-06,
      "loss": 1.1314,
      "step": 15
    },
    {
      "epoch": 1.25,
      "learning_rate": 7.500000000000001e-06,
      "loss": 1.0967,
      "step": 16
    },
    {
      "epoch": 1.33,
      "learning_rate": 7.211443451095007e-06,
      "loss": 1.1373,
      "step": 17
    },
    {
      "epoch": 1.41,
      "learning_rate": 6.913417161825449e-06,
      "loss": 1.0439,
      "step": 18
    },
    {
      "epoch": 1.49,
      "learning_rate": 6.607197326515808e-06,
      "loss": 1.0292,
      "step": 19
    },
    {
      "epoch": 1.57,
      "learning_rate": 6.294095225512604e-06,
      "loss": 1.0374,
      "step": 20
    },
    {
      "epoch": 1.65,
      "learning_rate": 5.975451610080643e-06,
      "loss": 1.0239,
      "step": 21
    },
    {
      "epoch": 1.73,
      "learning_rate": 5.65263096110026e-06,
      "loss": 0.9936,
      "step": 22
    },
    {
      "epoch": 1.8,
      "learning_rate": 5.327015646150716e-06,
      "loss": 1.0088,
      "step": 23
    },
    {
      "epoch": 1.88,
      "learning_rate": 5e-06,
      "loss": 1.0079,
      "step": 24
    },
    {
      "epoch": 1.96,
      "learning_rate": 4.672984353849285e-06,
      "loss": 1.0041,
      "step": 25
    },
    {
      "epoch": 2.04,
      "learning_rate": 4.347369038899744e-06,
      "loss": 0.9283,
      "step": 26
    },
    {
      "epoch": 2.12,
      "learning_rate": 4.02454838991936e-06,
      "loss": 0.9017,
      "step": 27
    },
    {
      "epoch": 2.2,
      "learning_rate": 3.705904774487396e-06,
      "loss": 0.9113,
      "step": 28
    },
    {
      "epoch": 2.27,
      "learning_rate": 3.3928026734841935e-06,
      "loss": 0.8751,
      "step": 29
    },
    {
      "epoch": 2.35,
      "learning_rate": 3.0865828381745515e-06,
      "loss": 0.8379,
      "step": 30
    },
    {
      "epoch": 2.43,
      "learning_rate": 2.7885565489049948e-06,
      "loss": 0.8646,
      "step": 31
    },
    {
      "epoch": 2.51,
      "learning_rate": 2.5000000000000015e-06,
      "loss": 0.8262,
      "step": 32
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.2221488349019903e-06,
      "loss": 0.8025,
      "step": 33
    },
    {
      "epoch": 2.67,
      "learning_rate": 1.956192854956397e-06,
      "loss": 0.8409,
      "step": 34
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.7032709244996559e-06,
      "loss": 0.8191,
      "step": 35
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.4644660940672628e-06,
      "loss": 0.8319,
      "step": 36
    },
    {
      "epoch": 2.9,
      "learning_rate": 1.2408009626051137e-06,
      "loss": 0.785,
      "step": 37
    },
    {
      "epoch": 2.98,
      "learning_rate": 1.0332332985438248e-06,
      "loss": 0.8475,
      "step": 38
    },
    {
      "epoch": 3.06,
      "learning_rate": 8.426519384872733e-07,
      "loss": 0.7509,
      "step": 39
    },
    {
      "epoch": 3.14,
      "learning_rate": 6.698729810778065e-07,
      "loss": 0.7825,
      "step": 40
    },
    {
      "epoch": 3.22,
      "learning_rate": 5.156362923365587e-07,
      "loss": 0.8104,
      "step": 41
    },
    {
      "epoch": 3.29,
      "learning_rate": 3.8060233744356634e-07,
      "loss": 0.7583,
      "step": 42
    },
    {
      "epoch": 3.37,
      "learning_rate": 2.653493525244721e-07,
      "loss": 0.7456,
      "step": 43
    },
    {
      "epoch": 3.45,
      "learning_rate": 1.7037086855465902e-07,
      "loss": 0.7221,
      "step": 44
    },
    {
      "epoch": 3.53,
      "learning_rate": 9.607359798384785e-08,
      "loss": 0.7034,
      "step": 45
    },
    {
      "epoch": 3.61,
      "learning_rate": 4.2775693130948094e-08,
      "loss": 0.7386,
      "step": 46
    },
    {
      "epoch": 3.69,
      "learning_rate": 1.0705383806982606e-08,
      "loss": 0.7359,
      "step": 47
    },
    {
      "epoch": 3.76,
      "learning_rate": 0.0,
      "loss": 0.7549,
      "step": 48
    },
    {
      "epoch": 3.76,
      "step": 48,
      "total_flos": 495910423363584.0,
      "train_loss": 1.082658023883899,
      "train_runtime": 22627.0931,
      "train_samples_per_second": 4.596,
      "train_steps_per_second": 0.002
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 48,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 495910423363584.0,
  "trial_name": null,
  "trial_params": null
}