{
  "best_metric": 1.1251049041748047,
  "best_model_checkpoint": "./mbartLarge_koja_mid2_run1/checkpoint-22708",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 68124,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 4.977981328166285e-05,
      "loss": 1.9241,
      "step": 500
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.95596265633257e-05,
      "loss": 1.6494,
      "step": 1000
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.933943984498855e-05,
      "loss": 1.5434,
      "step": 1500
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.91192531266514e-05,
      "loss": 1.4836,
      "step": 2000
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.889906640831425e-05,
      "loss": 1.4499,
      "step": 2500
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.86788796899771e-05,
      "loss": 1.3934,
      "step": 3000
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.845869297163995e-05,
      "loss": 1.3857,
      "step": 3500
    },
    {
      "epoch": 0.35,
      "learning_rate": 4.82385062533028e-05,
      "loss": 1.3396,
      "step": 4000
    },
    {
      "epoch": 0.4,
      "learning_rate": 4.801831953496565e-05,
      "loss": 1.3322,
      "step": 4500
    },
    {
      "epoch": 0.44,
      "learning_rate": 4.77981328166285e-05,
      "loss": 1.3106,
      "step": 5000
    },
    {
      "epoch": 0.48,
      "learning_rate": 4.757794609829135e-05,
      "loss": 1.2924,
      "step": 5500
    },
    {
      "epoch": 0.53,
      "learning_rate": 4.73577593799542e-05,
      "loss": 1.2834,
      "step": 6000
    },
    {
      "epoch": 0.57,
      "learning_rate": 4.713757266161705e-05,
      "loss": 1.2791,
      "step": 6500
    },
    {
      "epoch": 0.62,
      "learning_rate": 4.69173859432799e-05,
      "loss": 1.255,
      "step": 7000
    },
    {
      "epoch": 0.66,
      "learning_rate": 4.669719922494275e-05,
      "loss": 1.2573,
      "step": 7500
    },
    {
      "epoch": 0.7,
      "learning_rate": 4.64770125066056e-05,
      "loss": 1.2382,
      "step": 8000
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.625682578826845e-05,
      "loss": 1.223,
      "step": 8500
    },
    {
      "epoch": 0.79,
      "learning_rate": 4.6036639069931303e-05,
      "loss": 1.2145,
      "step": 9000
    },
    {
      "epoch": 0.84,
      "learning_rate": 4.5816452351594153e-05,
      "loss": 1.219,
      "step": 9500
    },
    {
      "epoch": 0.88,
      "learning_rate": 4.5596265633257004e-05,
      "loss": 1.2125,
      "step": 10000
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.5376078914919854e-05,
      "loss": 1.1996,
      "step": 10500
    },
    {
      "epoch": 0.97,
      "learning_rate": 4.5155892196582704e-05,
      "loss": 1.1823,
      "step": 11000
    },
    {
      "epoch": 1.0,
      "eval_bleu": 29.4501,
      "eval_gen_len": 18.8118,
      "eval_loss": 1.1695398092269897,
      "eval_runtime": 1716.1373,
      "eval_samples_per_second": 13.231,
      "eval_steps_per_second": 0.827,
      "step": 11354
    },
    {
      "epoch": 1.01,
      "learning_rate": 4.4935705478245554e-05,
      "loss": 1.1596,
      "step": 11500
    },
    {
      "epoch": 1.06,
      "learning_rate": 4.4715518759908404e-05,
      "loss": 1.0777,
      "step": 12000
    },
    {
      "epoch": 1.1,
      "learning_rate": 4.4495332041571254e-05,
      "loss": 1.0658,
      "step": 12500
    },
    {
      "epoch": 1.14,
      "learning_rate": 4.427514532323411e-05,
      "loss": 1.0296,
      "step": 13000
    },
    {
      "epoch": 1.19,
      "learning_rate": 4.4054958604896954e-05,
      "loss": 1.0269,
      "step": 13500
    },
    {
      "epoch": 1.23,
      "learning_rate": 4.3834771886559804e-05,
      "loss": 1.0192,
      "step": 14000
    },
    {
      "epoch": 1.28,
      "learning_rate": 4.3614585168222654e-05,
      "loss": 0.9946,
      "step": 14500
    },
    {
      "epoch": 1.32,
      "learning_rate": 4.3394398449885504e-05,
      "loss": 0.9996,
      "step": 15000
    },
    {
      "epoch": 1.37,
      "learning_rate": 4.3174211731548354e-05,
      "loss": 0.975,
      "step": 15500
    },
    {
      "epoch": 1.41,
      "learning_rate": 4.295402501321121e-05,
      "loss": 0.9874,
      "step": 16000
    },
    {
      "epoch": 1.45,
      "learning_rate": 4.2733838294874054e-05,
      "loss": 0.9683,
      "step": 16500
    },
    {
      "epoch": 1.5,
      "learning_rate": 4.2513651576536904e-05,
      "loss": 0.9675,
      "step": 17000
    },
    {
      "epoch": 1.54,
      "learning_rate": 4.2293464858199754e-05,
      "loss": 0.969,
      "step": 17500
    },
    {
      "epoch": 1.59,
      "learning_rate": 4.2073278139862604e-05,
      "loss": 0.9548,
      "step": 18000
    },
    {
      "epoch": 1.63,
      "learning_rate": 4.1853091421525454e-05,
      "loss": 0.9563,
      "step": 18500
    },
    {
      "epoch": 1.67,
      "learning_rate": 4.163290470318831e-05,
      "loss": 0.9536,
      "step": 19000
    },
    {
      "epoch": 1.72,
      "learning_rate": 4.1412717984851155e-05,
      "loss": 0.941,
      "step": 19500
    },
    {
      "epoch": 1.76,
      "learning_rate": 4.1192531266514005e-05,
      "loss": 0.9285,
      "step": 20000
    },
    {
      "epoch": 1.81,
      "learning_rate": 4.0972344548176855e-05,
      "loss": 0.9303,
      "step": 20500
    },
    {
      "epoch": 1.85,
      "learning_rate": 4.0752157829839705e-05,
      "loss": 0.9428,
      "step": 21000
    },
    {
      "epoch": 1.89,
      "learning_rate": 4.0531971111502555e-05,
      "loss": 0.9359,
      "step": 21500
    },
    {
      "epoch": 1.94,
      "learning_rate": 4.031178439316541e-05,
      "loss": 0.9302,
      "step": 22000
    },
    {
      "epoch": 1.98,
      "learning_rate": 4.0091597674828255e-05,
      "loss": 0.9207,
      "step": 22500
    },
    {
      "epoch": 2.0,
      "eval_bleu": 30.842,
      "eval_gen_len": 18.0892,
      "eval_loss": 1.1251049041748047,
      "eval_runtime": 1535.7063,
      "eval_samples_per_second": 14.786,
      "eval_steps_per_second": 0.925,
      "step": 22708
    },
    {
      "epoch": 2.03,
      "learning_rate": 3.9871410956491105e-05,
      "loss": 0.8788,
      "step": 23000
    },
    {
      "epoch": 2.07,
      "learning_rate": 3.9651224238153955e-05,
      "loss": 0.8367,
      "step": 23500
    },
    {
      "epoch": 2.11,
      "learning_rate": 3.9431037519816805e-05,
      "loss": 0.8196,
      "step": 24000
    },
    {
      "epoch": 2.16,
      "learning_rate": 3.921085080147966e-05,
      "loss": 0.7967,
      "step": 24500
    },
    {
      "epoch": 2.2,
      "learning_rate": 3.8990664083142505e-05,
      "loss": 0.7956,
      "step": 25000
    },
    {
      "epoch": 2.25,
      "learning_rate": 3.8770477364805355e-05,
      "loss": 0.7843,
      "step": 25500
    },
    {
      "epoch": 2.29,
      "learning_rate": 3.8550290646468205e-05,
      "loss": 0.7767,
      "step": 26000
    },
    {
      "epoch": 2.33,
      "learning_rate": 3.8330103928131055e-05,
      "loss": 0.7668,
      "step": 26500
    },
    {
      "epoch": 2.38,
      "learning_rate": 3.8109917209793905e-05,
      "loss": 0.754,
      "step": 27000
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.788973049145676e-05,
      "loss": 0.7671,
      "step": 27500
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.7669543773119605e-05,
      "loss": 0.7569,
      "step": 28000
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.7449357054782456e-05,
      "loss": 0.7553,
      "step": 28500
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.7229170336445306e-05,
      "loss": 0.7526,
      "step": 29000
    },
    {
      "epoch": 2.6,
      "learning_rate": 3.7008983618108156e-05,
      "loss": 0.7427,
      "step": 29500
    },
    {
      "epoch": 2.64,
      "learning_rate": 3.6788796899771006e-05,
      "loss": 0.7407,
      "step": 30000
    },
    {
      "epoch": 2.69,
      "learning_rate": 3.656861018143386e-05,
      "loss": 0.7418,
      "step": 30500
    },
    {
      "epoch": 2.73,
      "learning_rate": 3.6348423463096706e-05,
      "loss": 0.7301,
      "step": 31000
    },
    {
      "epoch": 2.77,
      "learning_rate": 3.6128236744759556e-05,
      "loss": 0.7218,
      "step": 31500
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.5908050026422406e-05,
      "loss": 0.7232,
      "step": 32000
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.5687863308085256e-05,
      "loss": 0.7375,
      "step": 32500
    },
    {
      "epoch": 2.91,
      "learning_rate": 3.5467676589748106e-05,
      "loss": 0.7261,
      "step": 33000
    },
    {
      "epoch": 2.95,
      "learning_rate": 3.524748987141096e-05,
      "loss": 0.7197,
      "step": 33500
    },
    {
      "epoch": 2.99,
      "learning_rate": 3.5027303153073806e-05,
      "loss": 0.7127,
      "step": 34000
    },
    {
      "epoch": 3.0,
      "eval_bleu": 31.2642,
      "eval_gen_len": 18.1188,
      "eval_loss": 1.168749451637268,
      "eval_runtime": 1511.0096,
      "eval_samples_per_second": 15.028,
      "eval_steps_per_second": 0.94,
      "step": 34062
    },
    {
      "epoch": 3.04,
      "learning_rate": 3.4807116434736656e-05,
      "loss": 0.6673,
      "step": 34500
    },
    {
      "epoch": 3.08,
      "learning_rate": 3.4586929716399506e-05,
      "loss": 0.6346,
      "step": 35000
    },
    {
      "epoch": 3.13,
      "learning_rate": 3.4366742998062356e-05,
      "loss": 0.6203,
      "step": 35500
    },
    {
      "epoch": 3.17,
      "learning_rate": 3.4146556279725206e-05,
      "loss": 0.605,
      "step": 36000
    },
    {
      "epoch": 3.21,
      "learning_rate": 3.392636956138806e-05,
      "loss": 0.5996,
      "step": 36500
    },
    {
      "epoch": 3.26,
      "learning_rate": 3.3706182843050906e-05,
      "loss": 0.5942,
      "step": 37000
    },
    {
      "epoch": 3.3,
      "learning_rate": 3.3485996124713757e-05,
      "loss": 0.5911,
      "step": 37500
    },
    {
      "epoch": 3.35,
      "learning_rate": 3.3265809406376607e-05,
      "loss": 0.5741,
      "step": 38000
    },
    {
      "epoch": 3.39,
      "learning_rate": 3.304562268803946e-05,
      "loss": 0.5807,
      "step": 38500
    },
    {
      "epoch": 3.43,
      "learning_rate": 3.2825435969702313e-05,
      "loss": 0.5777,
      "step": 39000
    },
    {
      "epoch": 3.48,
      "learning_rate": 3.2605249251365164e-05,
      "loss": 0.5823,
      "step": 39500
    },
    {
      "epoch": 3.52,
      "learning_rate": 3.238506253302801e-05,
      "loss": 0.5698,
      "step": 40000
    },
    {
      "epoch": 3.57,
      "learning_rate": 3.216487581469086e-05,
      "loss": 0.5727,
      "step": 40500
    },
    {
      "epoch": 3.61,
      "learning_rate": 3.194468909635371e-05,
      "loss": 0.5648,
      "step": 41000
    },
    {
      "epoch": 3.66,
      "learning_rate": 3.172450237801656e-05,
      "loss": 0.5641,
      "step": 41500
    },
    {
      "epoch": 3.7,
      "learning_rate": 3.1504315659679414e-05,
      "loss": 0.559,
      "step": 42000
    },
    {
      "epoch": 3.74,
      "learning_rate": 3.1284128941342264e-05,
      "loss": 0.5487,
      "step": 42500
    },
    {
      "epoch": 3.79,
      "learning_rate": 3.106394222300511e-05,
      "loss": 0.5521,
      "step": 43000
    },
    {
      "epoch": 3.83,
      "learning_rate": 3.084375550466796e-05,
      "loss": 0.5509,
      "step": 43500
    },
    {
      "epoch": 3.88,
      "learning_rate": 3.062356878633081e-05,
      "loss": 0.5557,
      "step": 44000
    },
    {
      "epoch": 3.92,
      "learning_rate": 3.0403382067993657e-05,
      "loss": 0.551,
      "step": 44500
    },
    {
      "epoch": 3.96,
      "learning_rate": 3.018319534965651e-05,
      "loss": 0.5406,
      "step": 45000
    },
    {
      "epoch": 4.0,
      "eval_bleu": 30.9531,
      "eval_gen_len": 17.9958,
      "eval_loss": 1.2618861198425293,
      "eval_runtime": 1469.856,
      "eval_samples_per_second": 15.448,
      "eval_steps_per_second": 0.966,
      "step": 45416
    },
    {
      "epoch": 4.01,
      "learning_rate": 2.996300863131936e-05,
      "loss": 0.5371,
      "step": 45500
    },
    {
      "epoch": 4.05,
      "learning_rate": 2.974282191298221e-05,
      "loss": 0.4843,
      "step": 46000
    },
    {
      "epoch": 4.1,
      "learning_rate": 2.952263519464506e-05,
      "loss": 0.4779,
      "step": 46500
    },
    {
      "epoch": 4.14,
      "learning_rate": 2.9302448476307908e-05,
      "loss": 0.4551,
      "step": 47000
    },
    {
      "epoch": 4.18,
      "learning_rate": 2.9082261757970758e-05,
      "loss": 0.4512,
      "step": 47500
    },
    {
      "epoch": 4.23,
      "learning_rate": 2.886207503963361e-05,
      "loss": 0.4464,
      "step": 48000
    },
    {
      "epoch": 4.27,
      "learning_rate": 2.864188832129646e-05,
      "loss": 0.4317,
      "step": 48500
    },
    {
      "epoch": 4.32,
      "learning_rate": 2.842170160295931e-05,
      "loss": 0.4392,
      "step": 49000
    },
    {
      "epoch": 4.36,
      "learning_rate": 2.820151488462216e-05,
      "loss": 0.4232,
      "step": 49500
    },
    {
      "epoch": 4.4,
      "learning_rate": 2.7981328166285008e-05,
      "loss": 0.4341,
      "step": 50000
    },
    {
      "epoch": 4.45,
      "learning_rate": 2.7761141447947865e-05,
      "loss": 0.4245,
      "step": 50500
    },
    {
      "epoch": 4.49,
      "learning_rate": 2.754095472961071e-05,
      "loss": 0.4266,
      "step": 51000
    },
    {
      "epoch": 4.54,
      "learning_rate": 2.732076801127356e-05,
      "loss": 0.4199,
      "step": 51500
    },
    {
      "epoch": 4.58,
      "learning_rate": 2.710058129293641e-05,
      "loss": 0.4266,
      "step": 52000
    },
    {
      "epoch": 4.62,
      "learning_rate": 2.688039457459926e-05,
      "loss": 0.4191,
      "step": 52500
    },
    {
      "epoch": 4.67,
      "learning_rate": 2.6660207856262108e-05,
      "loss": 0.4155,
      "step": 53000
    },
    {
      "epoch": 4.71,
      "learning_rate": 2.6440021137924965e-05,
      "loss": 0.4058,
      "step": 53500
    },
    {
      "epoch": 4.76,
      "learning_rate": 2.6219834419587812e-05,
      "loss": 0.4079,
      "step": 54000
    },
    {
      "epoch": 4.8,
      "learning_rate": 2.5999647701250662e-05,
      "loss": 0.4041,
      "step": 54500
    },
    {
      "epoch": 4.84,
      "learning_rate": 2.5779460982913512e-05,
      "loss": 0.4099,
      "step": 55000
    },
    {
      "epoch": 4.89,
      "learning_rate": 2.5559274264576362e-05,
      "loss": 0.4107,
      "step": 55500
    },
    {
      "epoch": 4.93,
      "learning_rate": 2.533908754623921e-05,
      "loss": 0.4016,
      "step": 56000
    },
    {
      "epoch": 4.98,
      "learning_rate": 2.5118900827902065e-05,
      "loss": 0.4027,
      "step": 56500
    },
    {
      "epoch": 5.0,
      "eval_bleu": 30.7923,
      "eval_gen_len": 18.0582,
      "eval_loss": 1.3788937330245972,
      "eval_runtime": 1470.5701,
      "eval_samples_per_second": 15.441,
      "eval_steps_per_second": 0.966,
      "step": 56770
    },
    {
      "epoch": 5.02,
      "learning_rate": 2.4898714109564912e-05,
      "loss": 0.3814,
      "step": 57000
    },
    {
      "epoch": 5.06,
      "learning_rate": 2.4678527391227762e-05,
      "loss": 0.3509,
      "step": 57500
    },
    {
      "epoch": 5.11,
      "learning_rate": 2.4458340672890612e-05,
      "loss": 0.3365,
      "step": 58000
    },
    {
      "epoch": 5.15,
      "learning_rate": 2.4238153954553462e-05,
      "loss": 0.3269,
      "step": 58500
    },
    {
      "epoch": 5.2,
      "learning_rate": 2.4017967236216312e-05,
      "loss": 0.3289,
      "step": 59000
    },
    {
      "epoch": 5.24,
      "learning_rate": 2.3797780517879162e-05,
      "loss": 0.3147,
      "step": 59500
    },
    {
      "epoch": 5.28,
      "learning_rate": 2.3577593799542012e-05,
      "loss": 0.3133,
      "step": 60000
    },
    {
      "epoch": 5.33,
      "learning_rate": 2.3357407081204862e-05,
      "loss": 0.3149,
      "step": 60500
    },
    {
      "epoch": 5.37,
      "learning_rate": 2.3137220362867713e-05,
      "loss": 0.3028,
      "step": 61000
    },
    {
      "epoch": 5.42,
      "learning_rate": 2.2917033644530563e-05,
      "loss": 0.313,
      "step": 61500
    },
    {
      "epoch": 5.46,
      "learning_rate": 2.2696846926193413e-05,
      "loss": 0.3081,
      "step": 62000
    },
    {
      "epoch": 5.5,
      "learning_rate": 2.2476660207856263e-05,
      "loss": 0.3068,
      "step": 62500
    },
    {
      "epoch": 5.55,
      "learning_rate": 2.2256473489519113e-05,
      "loss": 0.3054,
      "step": 63000
    },
    {
      "epoch": 5.59,
      "learning_rate": 2.2036286771181963e-05,
      "loss": 0.3015,
      "step": 63500
    },
    {
      "epoch": 5.64,
      "learning_rate": 2.1816100052844813e-05,
      "loss": 0.3043,
      "step": 64000
    },
    {
      "epoch": 5.68,
      "learning_rate": 2.1595913334507663e-05,
      "loss": 0.3009,
      "step": 64500
    },
    {
      "epoch": 5.72,
      "learning_rate": 2.1375726616170513e-05,
      "loss": 0.2964,
      "step": 65000
    },
    {
      "epoch": 5.77,
      "learning_rate": 2.1155539897833363e-05,
      "loss": 0.2883,
      "step": 65500
    },
    {
      "epoch": 5.81,
      "learning_rate": 2.0935353179496213e-05,
      "loss": 0.2958,
      "step": 66000
    },
    {
      "epoch": 5.86,
      "learning_rate": 2.0715166461159063e-05,
      "loss": 0.2985,
      "step": 66500
    },
    {
      "epoch": 5.9,
      "learning_rate": 2.0494979742821913e-05,
      "loss": 0.2975,
      "step": 67000
    },
    {
      "epoch": 5.95,
      "learning_rate": 2.0274793024484763e-05,
      "loss": 0.288,
      "step": 67500
    },
    {
      "epoch": 5.99,
      "learning_rate": 2.0054606306147613e-05,
      "loss": 0.286,
      "step": 68000
    },
    {
      "epoch": 6.0,
      "eval_bleu": 30.9393,
      "eval_gen_len": 18.1183,
      "eval_loss": 1.4784362316131592,
      "eval_runtime": 1467.697,
      "eval_samples_per_second": 15.471,
      "eval_steps_per_second": 0.968,
      "step": 68124
    },
    {
      "epoch": 6.0,
      "step": 68124,
      "total_flos": 2.3622407943975076e+18,
      "train_loss": 0.7323888317612346,
      "train_runtime": 87093.7472,
      "train_samples_per_second": 20.858,
      "train_steps_per_second": 1.304
    }
  ],
  "logging_steps": 500,
  "max_steps": 113540,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 2.3622407943975076e+18,
  "trial_name": null,
  "trial_params": null
}