{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "global_step": 27925,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09,
      "learning_rate": 3e-05,
      "loss": 3.5105,
      "step": 500
    },
    {
      "epoch": 0.18,
      "learning_rate": 2.945305378304467e-05,
      "loss": 1.6044,
      "step": 1000
    },
    {
      "epoch": 0.27,
      "learning_rate": 2.8906107566089334e-05,
      "loss": 1.5561,
      "step": 1500
    },
    {
      "epoch": 0.36,
      "learning_rate": 2.8359161349134002e-05,
      "loss": 1.5065,
      "step": 2000
    },
    {
      "epoch": 0.45,
      "learning_rate": 2.781221513217867e-05,
      "loss": 1.4607,
      "step": 2500
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.726526891522334e-05,
      "loss": 1.4575,
      "step": 3000
    },
    {
      "epoch": 0.63,
      "learning_rate": 2.6718322698268003e-05,
      "loss": 1.432,
      "step": 3500
    },
    {
      "epoch": 0.72,
      "learning_rate": 2.617137648131267e-05,
      "loss": 1.4269,
      "step": 4000
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.562443026435734e-05,
      "loss": 1.4204,
      "step": 4500
    },
    {
      "epoch": 0.9,
      "learning_rate": 2.5077484047402008e-05,
      "loss": 1.4037,
      "step": 5000
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.4530537830446673e-05,
      "loss": 1.4064,
      "step": 5500
    },
    {
      "epoch": 1.0,
      "eval_bleu": 0.8646,
      "eval_gen_len": 16.6,
      "eval_loss": 1.450329303741455,
      "eval_runtime": 368.5927,
      "eval_samples_per_second": 56.358,
      "eval_steps_per_second": 0.882,
      "step": 5585
    },
    {
      "epoch": 1.07,
      "learning_rate": 2.398359161349134e-05,
      "loss": 1.3601,
      "step": 6000
    },
    {
      "epoch": 1.16,
      "learning_rate": 2.343664539653601e-05,
      "loss": 1.3596,
      "step": 6500
    },
    {
      "epoch": 1.25,
      "learning_rate": 2.2889699179580678e-05,
      "loss": 1.3655,
      "step": 7000
    },
    {
      "epoch": 1.34,
      "learning_rate": 2.2342752962625342e-05,
      "loss": 1.326,
      "step": 7500
    },
    {
      "epoch": 1.43,
      "learning_rate": 2.179580674567001e-05,
      "loss": 1.3327,
      "step": 8000
    },
    {
      "epoch": 1.52,
      "learning_rate": 2.124886052871468e-05,
      "loss": 1.3367,
      "step": 8500
    },
    {
      "epoch": 1.61,
      "learning_rate": 2.0701914311759347e-05,
      "loss": 1.3305,
      "step": 9000
    },
    {
      "epoch": 1.7,
      "learning_rate": 2.0154968094804012e-05,
      "loss": 1.3218,
      "step": 9500
    },
    {
      "epoch": 1.79,
      "learning_rate": 1.960802187784868e-05,
      "loss": 1.3302,
      "step": 10000
    },
    {
      "epoch": 1.88,
      "learning_rate": 1.906107566089335e-05,
      "loss": 1.3173,
      "step": 10500
    },
    {
      "epoch": 1.97,
      "learning_rate": 1.8514129443938013e-05,
      "loss": 1.3306,
      "step": 11000
    },
    {
      "epoch": 2.0,
      "eval_bleu": 0.8302,
      "eval_gen_len": 16.9,
      "eval_loss": 1.422479271888733,
      "eval_runtime": 368.3493,
      "eval_samples_per_second": 56.395,
      "eval_steps_per_second": 0.882,
      "step": 11170
    },
    {
      "epoch": 2.06,
      "learning_rate": 1.7967183226982678e-05,
      "loss": 1.2841,
      "step": 11500
    },
    {
      "epoch": 2.15,
      "learning_rate": 1.7420237010027346e-05,
      "loss": 1.287,
      "step": 12000
    },
    {
      "epoch": 2.24,
      "learning_rate": 1.6873290793072015e-05,
      "loss": 1.2772,
      "step": 12500
    },
    {
      "epoch": 2.33,
      "learning_rate": 1.632634457611668e-05,
      "loss": 1.2802,
      "step": 13000
    },
    {
      "epoch": 2.42,
      "learning_rate": 1.5779398359161348e-05,
      "loss": 1.27,
      "step": 13500
    },
    {
      "epoch": 2.51,
      "learning_rate": 1.5232452142206018e-05,
      "loss": 1.273,
      "step": 14000
    },
    {
      "epoch": 2.6,
      "learning_rate": 1.4685505925250684e-05,
      "loss": 1.2888,
      "step": 14500
    },
    {
      "epoch": 2.69,
      "learning_rate": 1.4138559708295352e-05,
      "loss": 1.2767,
      "step": 15000
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.3591613491340019e-05,
      "loss": 1.2713,
      "step": 15500
    },
    {
      "epoch": 2.86,
      "learning_rate": 1.3044667274384685e-05,
      "loss": 1.253,
      "step": 16000
    },
    {
      "epoch": 2.95,
      "learning_rate": 1.2497721057429352e-05,
      "loss": 1.262,
      "step": 16500
    },
    {
      "epoch": 3.0,
      "eval_bleu": 1.0163,
      "eval_gen_len": 17.1,
      "eval_loss": 1.4149495363235474,
      "eval_runtime": 367.931,
      "eval_samples_per_second": 56.459,
      "eval_steps_per_second": 0.883,
      "step": 16755
    },
    {
      "epoch": 3.04,
      "learning_rate": 1.195077484047402e-05,
      "loss": 1.2486,
      "step": 17000
    },
    {
      "epoch": 3.13,
      "learning_rate": 1.1403828623518687e-05,
      "loss": 1.2413,
      "step": 17500
    },
    {
      "epoch": 3.22,
      "learning_rate": 1.0856882406563355e-05,
      "loss": 1.2174,
      "step": 18000
    },
    {
      "epoch": 3.31,
      "learning_rate": 1.0309936189608021e-05,
      "loss": 1.2258,
      "step": 18500
    },
    {
      "epoch": 3.4,
      "learning_rate": 9.76298997265269e-06,
      "loss": 1.2378,
      "step": 19000
    },
    {
      "epoch": 3.49,
      "learning_rate": 9.216043755697356e-06,
      "loss": 1.2286,
      "step": 19500
    },
    {
      "epoch": 3.58,
      "learning_rate": 8.669097538742024e-06,
      "loss": 1.2317,
      "step": 20000
    },
    {
      "epoch": 3.67,
      "learning_rate": 8.122151321786691e-06,
      "loss": 1.2386,
      "step": 20500
    },
    {
      "epoch": 3.76,
      "learning_rate": 7.575205104831359e-06,
      "loss": 1.22,
      "step": 21000
    },
    {
      "epoch": 3.85,
      "learning_rate": 7.028258887876026e-06,
      "loss": 1.2252,
      "step": 21500
    },
    {
      "epoch": 3.94,
      "learning_rate": 6.481312670920693e-06,
      "loss": 1.2175,
      "step": 22000
    },
    {
      "epoch": 4.0,
      "eval_bleu": 0.9445,
      "eval_gen_len": 17.2,
      "eval_loss": 1.416054606437683,
      "eval_runtime": 367.7817,
      "eval_samples_per_second": 56.482,
      "eval_steps_per_second": 0.884,
      "step": 22340
    },
    {
      "epoch": 4.03,
      "learning_rate": 5.9343664539653605e-06,
      "loss": 1.2158,
      "step": 22500
    },
    {
      "epoch": 4.12,
      "learning_rate": 5.387420237010028e-06,
      "loss": 1.1916,
      "step": 23000
    },
    {
      "epoch": 4.21,
      "learning_rate": 4.8404740200546945e-06,
      "loss": 1.2079,
      "step": 23500
    },
    {
      "epoch": 4.3,
      "learning_rate": 4.293527803099362e-06,
      "loss": 1.2124,
      "step": 24000
    },
    {
      "epoch": 4.39,
      "learning_rate": 3.7465815861440297e-06,
      "loss": 1.2006,
      "step": 24500
    },
    {
      "epoch": 4.48,
      "learning_rate": 3.199635369188696e-06,
      "loss": 1.2121,
      "step": 25000
    },
    {
      "epoch": 4.57,
      "learning_rate": 2.6526891522333636e-06,
      "loss": 1.2022,
      "step": 25500
    },
    {
      "epoch": 4.66,
      "learning_rate": 2.105742935278031e-06,
      "loss": 1.2029,
      "step": 26000
    },
    {
      "epoch": 4.74,
      "learning_rate": 1.5587967183226983e-06,
      "loss": 1.2008,
      "step": 26500
    },
    {
      "epoch": 4.83,
      "learning_rate": 1.0118505013673655e-06,
      "loss": 1.2193,
      "step": 27000
    },
    {
      "epoch": 4.92,
      "learning_rate": 4.6490428441203284e-07,
      "loss": 1.2108,
      "step": 27500
    },
    {
      "epoch": 5.0,
      "eval_bleu": 1.0114,
      "eval_gen_len": 17.3,
      "eval_loss": 1.4185353517532349,
      "eval_runtime": 367.8873,
      "eval_samples_per_second": 56.466,
      "eval_steps_per_second": 0.883,
      "step": 27925
    },
    {
      "epoch": 5.0,
      "step": 27925,
      "total_flos": 5.320769336064e+17,
      "train_loss": 1.3384765100436438,
      "train_runtime": 44633.7325,
      "train_samples_per_second": 40.04,
      "train_steps_per_second": 0.626
    }
  ],
  "max_steps": 27925,
  "num_train_epochs": 5,
  "total_flos": 5.320769336064e+17,
  "trial_name": null,
  "trial_params": null
}