{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9959017484099317,
  "eval_steps": 5000,
  "global_step": 150000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 4.9833674854299174e-05,
      "loss": 1.2201,
      "step": 500
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.9667349708598347e-05,
      "loss": 1.1313,
      "step": 1000
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.950102456289752e-05,
      "loss": 1.1352,
      "step": 1500
    },
    {
      "epoch": 0.03,
      "learning_rate": 4.933469941719669e-05,
      "loss": 1.0939,
      "step": 2000
    },
    {
      "epoch": 0.03,
      "learning_rate": 4.916837427149586e-05,
      "loss": 1.0539,
      "step": 2500
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.9002049125795035e-05,
      "loss": 1.0473,
      "step": 3000
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.8835723980094214e-05,
      "loss": 1.0461,
      "step": 3500
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.866939883439338e-05,
      "loss": 0.9995,
      "step": 4000
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.850307368869255e-05,
      "loss": 1.0156,
      "step": 4500
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.833674854299173e-05,
      "loss": 1.0091,
      "step": 5000
    },
    {
      "epoch": 0.07,
      "eval_bleu": 39.1306,
      "eval_chrf++": 61.4553,
      "eval_gen_len": 24.5812,
      "eval_loss": 0.983640730381012,
      "eval_runtime": 4879.4604,
      "eval_samples_per_second": 4.77,
      "eval_spbleu": 46.5203,
      "eval_steps_per_second": 0.954,
      "eval_ter": 49.0585,
      "step": 5000
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.8170423397290895e-05,
      "loss": 1.0139,
      "step": 5500
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.800409825159007e-05,
      "loss": 0.9984,
      "step": 6000
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.7837773105889246e-05,
      "loss": 0.9854,
      "step": 6500
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.767144796018841e-05,
      "loss": 1.0016,
      "step": 7000
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.750512281448759e-05,
      "loss": 0.9851,
      "step": 7500
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.733879766878676e-05,
      "loss": 0.9736,
      "step": 8000
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.717247252308593e-05,
      "loss": 0.968,
      "step": 8500
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.700614737738511e-05,
      "loss": 0.9772,
      "step": 9000
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.683982223168428e-05,
      "loss": 0.9452,
      "step": 9500
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.6673497085983444e-05,
      "loss": 0.9593,
      "step": 10000
    },
    {
      "epoch": 0.13,
      "eval_bleu": 40.7812,
      "eval_chrf++": 62.5078,
      "eval_gen_len": 24.4548,
      "eval_loss": 0.9384894967079163,
      "eval_runtime": 4848.287,
      "eval_samples_per_second": 4.8,
      "eval_spbleu": 48.2302,
      "eval_steps_per_second": 0.96,
      "eval_ter": 47.6761,
      "step": 10000
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.650717194028262e-05,
      "loss": 0.9495,
      "step": 10500
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.6340846794581795e-05,
      "loss": 0.9684,
      "step": 11000
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.617452164888097e-05,
      "loss": 0.9496,
      "step": 11500
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.600819650318014e-05,
      "loss": 0.9417,
      "step": 12000
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.584187135747931e-05,
      "loss": 0.9415,
      "step": 12500
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.5675546211778483e-05,
      "loss": 0.9325,
      "step": 13000
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.5509221066077656e-05,
      "loss": 0.9488,
      "step": 13500
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.534289592037683e-05,
      "loss": 0.9233,
      "step": 14000
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.5176570774676e-05,
      "loss": 0.9541,
      "step": 14500
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.501024562897517e-05,
      "loss": 0.9716,
      "step": 15000
    },
    {
      "epoch": 0.2,
      "eval_bleu": 41.9435,
      "eval_chrf++": 63.2793,
      "eval_gen_len": 24.3083,
      "eval_loss": 0.9115902781486511,
      "eval_runtime": 4836.6682,
      "eval_samples_per_second": 4.812,
      "eval_spbleu": 49.3278,
      "eval_steps_per_second": 0.962,
      "eval_ter": 46.5599,
      "step": 15000
    },
    {
      "epoch": 0.21,
      "learning_rate": 4.4843920483274344e-05,
      "loss": 0.9314,
      "step": 15500
    },
    {
      "epoch": 0.21,
      "learning_rate": 4.4677595337573516e-05,
      "loss": 0.9492,
      "step": 16000
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.451127019187269e-05,
      "loss": 0.9195,
      "step": 16500
    },
    {
      "epoch": 0.23,
      "learning_rate": 4.434494504617187e-05,
      "loss": 0.924,
      "step": 17000
    },
    {
      "epoch": 0.23,
      "learning_rate": 4.417861990047103e-05,
      "loss": 0.8895,
      "step": 17500
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.4012294754770204e-05,
      "loss": 0.9303,
      "step": 18000
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.384596960906938e-05,
      "loss": 0.9499,
      "step": 18500
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.3679644463368555e-05,
      "loss": 0.9335,
      "step": 19000
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.351331931766772e-05,
      "loss": 0.9238,
      "step": 19500
    },
    {
      "epoch": 0.27,
      "learning_rate": 4.33469941719669e-05,
      "loss": 0.9204,
      "step": 20000
    },
    {
      "epoch": 0.27,
      "eval_bleu": 42.615,
      "eval_chrf++": 63.6457,
      "eval_gen_len": 24.1677,
      "eval_loss": 0.8911730647087097,
      "eval_runtime": 4794.0613,
      "eval_samples_per_second": 4.855,
      "eval_spbleu": 50.0126,
      "eval_steps_per_second": 0.971,
      "eval_ter": 45.765,
      "step": 20000
    },
    {
      "epoch": 0.27,
      "learning_rate": 4.318066902626607e-05,
      "loss": 0.9066,
      "step": 20500
    },
    {
      "epoch": 0.28,
      "learning_rate": 4.301434388056524e-05,
      "loss": 0.8974,
      "step": 21000
    },
    {
      "epoch": 0.29,
      "learning_rate": 4.2848018734864416e-05,
      "loss": 0.9196,
      "step": 21500
    },
    {
      "epoch": 0.29,
      "learning_rate": 4.268169358916359e-05,
      "loss": 0.9046,
      "step": 22000
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.251536844346276e-05,
      "loss": 0.8821,
      "step": 22500
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.234904329776193e-05,
      "loss": 0.9095,
      "step": 23000
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.2182718152061104e-05,
      "loss": 0.8814,
      "step": 23500
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.2016393006360276e-05,
      "loss": 0.9074,
      "step": 24000
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.185006786065945e-05,
      "loss": 0.9086,
      "step": 24500
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.168374271495862e-05,
      "loss": 0.87,
      "step": 25000
    },
    {
      "epoch": 0.33,
      "eval_bleu": 43.372,
      "eval_chrf++": 64.3092,
      "eval_gen_len": 24.2173,
      "eval_loss": 0.8829550743103027,
      "eval_runtime": 4787.4055,
      "eval_samples_per_second": 4.862,
      "eval_spbleu": 50.8714,
      "eval_steps_per_second": 0.972,
      "eval_ter": 45.1953,
      "step": 25000
    },
    {
      "epoch": 0.34,
      "learning_rate": 4.151741756925779e-05,
      "loss": 0.8688,
      "step": 25500
    },
    {
      "epoch": 0.35,
      "learning_rate": 4.1351092423556965e-05,
      "loss": 0.8861,
      "step": 26000
    },
    {
      "epoch": 0.35,
      "learning_rate": 4.1184767277856137e-05,
      "loss": 0.854,
      "step": 26500
    },
    {
      "epoch": 0.36,
      "learning_rate": 4.101844213215531e-05,
      "loss": 0.9359,
      "step": 27000
    },
    {
      "epoch": 0.37,
      "learning_rate": 4.085211698645448e-05,
      "loss": 0.8724,
      "step": 27500
    },
    {
      "epoch": 0.37,
      "learning_rate": 4.068579184075366e-05,
      "loss": 0.8954,
      "step": 28000
    },
    {
      "epoch": 0.38,
      "learning_rate": 4.0519466695052825e-05,
      "loss": 0.9009,
      "step": 28500
    },
    {
      "epoch": 0.39,
      "learning_rate": 4.0353141549352e-05,
      "loss": 0.889,
      "step": 29000
    },
    {
      "epoch": 0.39,
      "learning_rate": 4.0186816403651176e-05,
      "loss": 0.8782,
      "step": 29500
    },
    {
      "epoch": 0.4,
      "learning_rate": 4.002049125795034e-05,
      "loss": 0.8891,
      "step": 30000
    },
    {
      "epoch": 0.4,
      "eval_bleu": 43.9744,
      "eval_chrf++": 64.7903,
      "eval_gen_len": 24.3772,
      "eval_loss": 0.8677447438240051,
      "eval_runtime": 4855.7738,
      "eval_samples_per_second": 4.793,
      "eval_spbleu": 51.5389,
      "eval_steps_per_second": 0.959,
      "eval_ter": 44.8206,
      "step": 30000
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.985416611224951e-05,
      "loss": 0.8785,
      "step": 30500
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.968784096654869e-05,
      "loss": 0.8603,
      "step": 31000
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.952151582084786e-05,
      "loss": 0.8513,
      "step": 31500
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.9355190675147036e-05,
      "loss": 0.8737,
      "step": 32000
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.918886552944621e-05,
      "loss": 0.8808,
      "step": 32500
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.9022540383745374e-05,
      "loss": 0.8564,
      "step": 33000
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.885621523804455e-05,
      "loss": 0.8816,
      "step": 33500
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.8689890092343725e-05,
      "loss": 0.8593,
      "step": 34000
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.852356494664289e-05,
      "loss": 0.8382,
      "step": 34500
    },
    {
      "epoch": 0.47,
      "learning_rate": 3.835723980094207e-05,
      "loss": 0.876,
      "step": 35000
    },
    {
      "epoch": 0.47,
      "eval_bleu": 44.1888,
      "eval_chrf++": 64.937,
      "eval_gen_len": 24.3824,
      "eval_loss": 0.853570818901062,
      "eval_runtime": 4869.588,
      "eval_samples_per_second": 4.779,
      "eval_spbleu": 51.71,
      "eval_steps_per_second": 0.956,
      "eval_ter": 44.5618,
      "step": 35000
    },
    {
      "epoch": 0.47,
      "learning_rate": 3.819091465524124e-05,
      "loss": 0.8818,
      "step": 35500
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.802458950954041e-05,
      "loss": 0.8723,
      "step": 36000
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.7858264363839585e-05,
      "loss": 0.8363,
      "step": 36500
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.769193921813876e-05,
      "loss": 0.8441,
      "step": 37000
    },
    {
      "epoch": 0.5,
      "learning_rate": 3.752561407243793e-05,
      "loss": 0.8483,
      "step": 37500
    },
    {
      "epoch": 0.51,
      "learning_rate": 3.73592889267371e-05,
      "loss": 0.8321,
      "step": 38000
    },
    {
      "epoch": 0.51,
      "learning_rate": 3.7192963781036273e-05,
      "loss": 0.8286,
      "step": 38500
    },
    {
      "epoch": 0.52,
      "learning_rate": 3.7026638635335446e-05,
      "loss": 0.8525,
      "step": 39000
    },
    {
      "epoch": 0.53,
      "learning_rate": 3.686031348963462e-05,
      "loss": 0.8697,
      "step": 39500
    },
    {
      "epoch": 0.53,
      "learning_rate": 3.669398834393379e-05,
      "loss": 0.8636,
      "step": 40000
    },
    {
      "epoch": 0.53,
      "eval_bleu": 44.7097,
      "eval_chrf++": 65.2018,
      "eval_gen_len": 24.2135,
      "eval_loss": 0.8435378670692444,
      "eval_runtime": 4812.8015,
      "eval_samples_per_second": 4.836,
      "eval_spbleu": 52.2257,
      "eval_steps_per_second": 0.967,
      "eval_ter": 43.745,
      "step": 40000
    },
    {
      "epoch": 0.54,
      "learning_rate": 3.652766319823296e-05,
      "loss": 0.873,
      "step": 40500
    },
    {
      "epoch": 0.55,
      "learning_rate": 3.6361338052532134e-05,
      "loss": 0.8743,
      "step": 41000
    },
    {
      "epoch": 0.55,
      "learning_rate": 3.619501290683131e-05,
      "loss": 0.86,
      "step": 41500
    },
    {
      "epoch": 0.56,
      "learning_rate": 3.602868776113048e-05,
      "loss": 0.8492,
      "step": 42000
    },
    {
      "epoch": 0.57,
      "learning_rate": 3.586236261542965e-05,
      "loss": 0.8454,
      "step": 42500
    },
    {
      "epoch": 0.57,
      "learning_rate": 3.569603746972883e-05,
      "loss": 0.8348,
      "step": 43000
    },
    {
      "epoch": 0.58,
      "learning_rate": 3.5529712324027994e-05,
      "loss": 0.8528,
      "step": 43500
    },
    {
      "epoch": 0.59,
      "learning_rate": 3.5363387178327166e-05,
      "loss": 0.8358,
      "step": 44000
    },
    {
      "epoch": 0.59,
      "learning_rate": 3.5197062032626345e-05,
      "loss": 0.8293,
      "step": 44500
    },
    {
      "epoch": 0.6,
      "learning_rate": 3.503073688692551e-05,
      "loss": 0.8501,
      "step": 45000
    },
    {
      "epoch": 0.6,
      "eval_bleu": 45.0613,
      "eval_chrf++": 65.5511,
      "eval_gen_len": 24.2012,
      "eval_loss": 0.8350074887275696,
      "eval_runtime": 4817.2319,
      "eval_samples_per_second": 4.831,
      "eval_spbleu": 52.5625,
      "eval_steps_per_second": 0.966,
      "eval_ter": 43.4063,
      "step": 45000
    },
    {
      "epoch": 0.61,
      "learning_rate": 3.486441174122469e-05,
      "loss": 0.8366,
      "step": 45500
    },
    {
      "epoch": 0.61,
      "learning_rate": 3.469808659552386e-05,
      "loss": 0.8342,
      "step": 46000
    },
    {
      "epoch": 0.62,
      "learning_rate": 3.453176144982303e-05,
      "loss": 0.8239,
      "step": 46500
    },
    {
      "epoch": 0.63,
      "learning_rate": 3.4365436304122206e-05,
      "loss": 0.8663,
      "step": 47000
    },
    {
      "epoch": 0.63,
      "learning_rate": 3.419911115842138e-05,
      "loss": 0.8041,
      "step": 47500
    },
    {
      "epoch": 0.64,
      "learning_rate": 3.403278601272055e-05,
      "loss": 0.8368,
      "step": 48000
    },
    {
      "epoch": 0.65,
      "learning_rate": 3.386646086701972e-05,
      "loss": 0.8106,
      "step": 48500
    },
    {
      "epoch": 0.65,
      "learning_rate": 3.3700135721318894e-05,
      "loss": 0.836,
      "step": 49000
    },
    {
      "epoch": 0.66,
      "learning_rate": 3.3533810575618066e-05,
      "loss": 0.8145,
      "step": 49500
    },
    {
      "epoch": 0.67,
      "learning_rate": 3.336748542991724e-05,
      "loss": 0.8417,
      "step": 50000
    },
    {
      "epoch": 0.67,
      "eval_bleu": 45.3651,
      "eval_chrf++": 65.7396,
      "eval_gen_len": 24.2033,
      "eval_loss": 0.8263786435127258,
      "eval_runtime": 4807.1892,
      "eval_samples_per_second": 4.841,
      "eval_spbleu": 52.8998,
      "eval_steps_per_second": 0.968,
      "eval_ter": 43.3251,
      "step": 50000
    },
    {
      "epoch": 0.67,
      "learning_rate": 3.320116028421641e-05,
      "loss": 0.7953,
      "step": 50500
    },
    {
      "epoch": 0.68,
      "learning_rate": 3.303483513851558e-05,
      "loss": 0.8218,
      "step": 51000
    },
    {
      "epoch": 0.69,
      "learning_rate": 3.2868509992814755e-05,
      "loss": 0.81,
      "step": 51500
    },
    {
      "epoch": 0.69,
      "learning_rate": 3.270218484711393e-05,
      "loss": 0.822,
      "step": 52000
    },
    {
      "epoch": 0.7,
      "learning_rate": 3.25358597014131e-05,
      "loss": 0.8174,
      "step": 52500
    },
    {
      "epoch": 0.71,
      "learning_rate": 3.236953455571227e-05,
      "loss": 0.8136,
      "step": 53000
    },
    {
      "epoch": 0.71,
      "learning_rate": 3.220320941001144e-05,
      "loss": 0.8669,
      "step": 53500
    },
    {
      "epoch": 0.72,
      "learning_rate": 3.2036884264310615e-05,
      "loss": 0.8195,
      "step": 54000
    },
    {
      "epoch": 0.73,
      "learning_rate": 3.187055911860979e-05,
      "loss": 0.825,
      "step": 54500
    },
    {
      "epoch": 0.73,
      "learning_rate": 3.1704233972908966e-05,
      "loss": 0.8083,
      "step": 55000
    },
    {
      "epoch": 0.73,
      "eval_bleu": 45.8888,
      "eval_chrf++": 66.0994,
      "eval_gen_len": 24.2883,
      "eval_loss": 0.8205437660217285,
      "eval_runtime": 4815.6563,
      "eval_samples_per_second": 4.833,
      "eval_spbleu": 53.2813,
      "eval_steps_per_second": 0.967,
      "eval_ter": 42.9175,
      "step": 55000
    },
    {
      "epoch": 0.74,
      "learning_rate": 3.153790882720814e-05,
      "loss": 0.8357,
      "step": 55500
    },
    {
      "epoch": 0.75,
      "learning_rate": 3.13715836815073e-05,
      "loss": 0.8097,
      "step": 56000
    },
    {
      "epoch": 0.75,
      "learning_rate": 3.120525853580648e-05,
      "loss": 0.7932,
      "step": 56500
    },
    {
      "epoch": 0.76,
      "learning_rate": 3.1038933390105654e-05,
      "loss": 0.8225,
      "step": 57000
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.087260824440482e-05,
      "loss": 0.7973,
      "step": 57500
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.0706283098704e-05,
      "loss": 0.8379,
      "step": 58000
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.053995795300317e-05,
      "loss": 0.8085,
      "step": 58500
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.037363280730234e-05,
      "loss": 0.8528,
      "step": 59000
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.0207307661601515e-05,
      "loss": 0.8262,
      "step": 59500
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.0040982515900683e-05,
      "loss": 0.7964,
      "step": 60000
    },
    {
      "epoch": 0.8,
      "eval_bleu": 46.137,
      "eval_chrf++": 66.311,
      "eval_gen_len": 24.357,
      "eval_loss": 0.8131713271141052,
      "eval_runtime": 4833.9641,
      "eval_samples_per_second": 4.815,
      "eval_spbleu": 53.6989,
      "eval_steps_per_second": 0.963,
      "eval_ter": 42.7845,
      "step": 60000
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.987465737019986e-05,
      "loss": 0.8137,
      "step": 60500
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.970833222449903e-05,
      "loss": 0.8046,
      "step": 61000
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.95420070787982e-05,
      "loss": 0.8496,
      "step": 61500
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.9375681933097375e-05,
      "loss": 0.8115,
      "step": 62000
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.9209356787396547e-05,
      "loss": 0.8222,
      "step": 62500
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.9043031641695716e-05,
      "loss": 0.8067,
      "step": 63000
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.887670649599489e-05,
      "loss": 0.7789,
      "step": 63500
    },
    {
      "epoch": 0.85,
      "learning_rate": 2.8710381350294063e-05,
      "loss": 0.8369,
      "step": 64000
    },
    {
      "epoch": 0.86,
      "learning_rate": 2.854405620459324e-05,
      "loss": 0.8139,
      "step": 64500
    },
    {
      "epoch": 0.86,
      "learning_rate": 2.837773105889241e-05,
      "loss": 0.7947,
      "step": 65000
    },
    {
      "epoch": 0.86,
      "eval_bleu": 46.3,
      "eval_chrf++": 66.3209,
      "eval_gen_len": 24.3462,
      "eval_loss": 0.8046720623970032,
      "eval_runtime": 4808.297,
      "eval_samples_per_second": 4.84,
      "eval_spbleu": 53.7389,
      "eval_steps_per_second": 0.968,
      "eval_ter": 42.7587,
      "step": 65000
    },
    {
      "epoch": 0.87,
      "learning_rate": 2.821140591319158e-05,
      "loss": 0.7977,
      "step": 65500
    },
    {
      "epoch": 0.88,
      "learning_rate": 2.8045080767490755e-05,
      "loss": 0.809,
      "step": 66000
    },
    {
      "epoch": 0.88,
      "learning_rate": 2.7878755621789927e-05,
      "loss": 0.8344,
      "step": 66500
    },
    {
      "epoch": 0.89,
      "learning_rate": 2.7712430476089096e-05,
      "loss": 0.7839,
      "step": 67000
    },
    {
      "epoch": 0.9,
      "learning_rate": 2.754610533038827e-05,
      "loss": 0.7905,
      "step": 67500
    },
    {
      "epoch": 0.9,
      "learning_rate": 2.7379780184687444e-05,
      "loss": 0.806,
      "step": 68000
    },
    {
      "epoch": 0.91,
      "learning_rate": 2.7213455038986612e-05,
      "loss": 0.7844,
      "step": 68500
    },
    {
      "epoch": 0.92,
      "learning_rate": 2.7047129893285788e-05,
      "loss": 0.7724,
      "step": 69000
    },
    {
      "epoch": 0.92,
      "learning_rate": 2.688080474758496e-05,
      "loss": 0.8193,
      "step": 69500
    },
    {
      "epoch": 0.93,
      "learning_rate": 2.6714479601884135e-05,
      "loss": 0.7972,
      "step": 70000
    },
    {
      "epoch": 0.93,
      "eval_bleu": 46.7802,
      "eval_chrf++": 66.6403,
      "eval_gen_len": 24.1858,
      "eval_loss": 0.7967858910560608,
      "eval_runtime": 4793.9096,
      "eval_samples_per_second": 4.855,
      "eval_spbleu": 54.0456,
      "eval_steps_per_second": 0.971,
      "eval_ter": 42.2931,
      "step": 70000
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.6548154456183304e-05,
      "loss": 0.8105,
      "step": 70500
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.6381829310482476e-05,
      "loss": 0.7997,
      "step": 71000
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.621550416478165e-05,
      "loss": 0.8022,
      "step": 71500
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.604917901908082e-05,
      "loss": 0.8455,
      "step": 72000
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.5882853873379992e-05,
      "loss": 0.7892,
      "step": 72500
    },
    {
      "epoch": 0.97,
      "learning_rate": 2.5716528727679168e-05,
      "loss": 0.764,
      "step": 73000
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.5550203581978337e-05,
      "loss": 0.7833,
      "step": 73500
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.5383878436277515e-05,
      "loss": 0.7769,
      "step": 74000
    },
    {
      "epoch": 0.99,
      "learning_rate": 2.5217553290576684e-05,
      "loss": 0.7938,
      "step": 74500
    },
    {
      "epoch": 1.0,
      "learning_rate": 2.5051228144875856e-05,
      "loss": 0.8199,
      "step": 75000
    },
    {
      "epoch": 1.0,
      "eval_bleu": 47.0207,
      "eval_chrf++": 66.8796,
      "eval_gen_len": 24.2671,
      "eval_loss": 0.792296826839447,
      "eval_runtime": 4833.1796,
      "eval_samples_per_second": 4.815,
      "eval_spbleu": 54.3927,
      "eval_steps_per_second": 0.963,
      "eval_ter": 42.1693,
      "step": 75000
    },
    {
      "epoch": 1.0,
      "learning_rate": 2.4884902999175028e-05,
      "loss": 0.6904,
      "step": 75500
    },
    {
      "epoch": 1.01,
      "learning_rate": 2.47185778534742e-05,
      "loss": 0.6305,
      "step": 76000
    },
    {
      "epoch": 1.02,
      "learning_rate": 2.4552252707773372e-05,
      "loss": 0.5988,
      "step": 76500
    },
    {
      "epoch": 1.02,
      "learning_rate": 2.4385927562072548e-05,
      "loss": 0.6285,
      "step": 77000
    },
    {
      "epoch": 1.03,
      "learning_rate": 2.4219602416371717e-05,
      "loss": 0.6293,
      "step": 77500
    },
    {
      "epoch": 1.04,
      "learning_rate": 2.405327727067089e-05,
      "loss": 0.6043,
      "step": 78000
    },
    {
      "epoch": 1.04,
      "learning_rate": 2.3886952124970064e-05,
      "loss": 0.6415,
      "step": 78500
    },
    {
      "epoch": 1.05,
      "learning_rate": 2.3720626979269233e-05,
      "loss": 0.6285,
      "step": 79000
    },
    {
      "epoch": 1.06,
      "learning_rate": 2.355430183356841e-05,
      "loss": 0.6169,
      "step": 79500
    },
    {
      "epoch": 1.06,
      "learning_rate": 2.338797668786758e-05,
      "loss": 0.6288,
      "step": 80000
    },
    {
      "epoch": 1.06,
      "eval_bleu": 47.092,
      "eval_chrf++": 66.9397,
      "eval_gen_len": 24.3014,
      "eval_loss": 0.8010306358337402,
      "eval_runtime": 4827.2842,
      "eval_samples_per_second": 4.821,
      "eval_spbleu": 54.5163,
      "eval_steps_per_second": 0.964,
      "eval_ter": 41.7379,
      "step": 80000
    },
    {
      "epoch": 1.07,
      "learning_rate": 2.3221651542166753e-05,
      "loss": 0.6062,
      "step": 80500
    },
    {
      "epoch": 1.08,
      "learning_rate": 2.3055326396465925e-05,
      "loss": 0.6193,
      "step": 81000
    },
    {
      "epoch": 1.08,
      "learning_rate": 2.2889001250765097e-05,
      "loss": 0.6205,
      "step": 81500
    },
    {
      "epoch": 1.09,
      "learning_rate": 2.272267610506427e-05,
      "loss": 0.6073,
      "step": 82000
    },
    {
      "epoch": 1.1,
      "learning_rate": 2.255635095936344e-05,
      "loss": 0.5999,
      "step": 82500
    },
    {
      "epoch": 1.1,
      "learning_rate": 2.2390025813662613e-05,
      "loss": 0.6397,
      "step": 83000
    },
    {
      "epoch": 1.11,
      "learning_rate": 2.2223700667961785e-05,
      "loss": 0.6124,
      "step": 83500
    },
    {
      "epoch": 1.12,
      "learning_rate": 2.205737552226096e-05,
      "loss": 0.6026,
      "step": 84000
    },
    {
      "epoch": 1.12,
      "learning_rate": 2.1891050376560133e-05,
      "loss": 0.6123,
      "step": 84500
    },
    {
      "epoch": 1.13,
      "learning_rate": 2.17247252308593e-05,
      "loss": 0.6313,
      "step": 85000
    },
    {
      "epoch": 1.13,
      "eval_bleu": 47.321,
      "eval_chrf++": 67.0811,
      "eval_gen_len": 24.2745,
      "eval_loss": 0.7992239594459534,
      "eval_runtime": 4828.2975,
      "eval_samples_per_second": 4.82,
      "eval_spbleu": 54.7759,
      "eval_steps_per_second": 0.964,
      "eval_ter": 41.5623,
      "step": 85000
    },
    {
      "epoch": 1.14,
      "learning_rate": 2.1558400085158477e-05,
      "loss": 0.5976,
      "step": 85500
    },
    {
      "epoch": 1.14,
      "learning_rate": 2.139207493945765e-05,
      "loss": 0.6151,
      "step": 86000
    },
    {
      "epoch": 1.15,
      "learning_rate": 2.122574979375682e-05,
      "loss": 0.607,
      "step": 86500
    },
    {
      "epoch": 1.16,
      "learning_rate": 2.1059424648055993e-05,
      "loss": 0.6202,
      "step": 87000
    },
    {
      "epoch": 1.16,
      "learning_rate": 2.0893099502355165e-05,
      "loss": 0.6135,
      "step": 87500
    },
    {
      "epoch": 1.17,
      "learning_rate": 2.0726774356654337e-05,
      "loss": 0.6156,
      "step": 88000
    },
    {
      "epoch": 1.18,
      "learning_rate": 2.056044921095351e-05,
      "loss": 0.5956,
      "step": 88500
    },
    {
      "epoch": 1.18,
      "learning_rate": 2.039412406525268e-05,
      "loss": 0.6202,
      "step": 89000
    },
    {
      "epoch": 1.19,
      "learning_rate": 2.0227798919551854e-05,
      "loss": 0.6041,
      "step": 89500
    },
    {
      "epoch": 1.2,
      "learning_rate": 2.006147377385103e-05,
      "loss": 0.5869,
      "step": 90000
    },
    {
      "epoch": 1.2,
      "eval_bleu": 47.6385,
      "eval_chrf++": 67.2633,
      "eval_gen_len": 24.3011,
      "eval_loss": 0.7958626747131348,
      "eval_runtime": 4842.841,
      "eval_samples_per_second": 4.806,
      "eval_spbleu": 54.9942,
      "eval_steps_per_second": 0.961,
      "eval_ter": 41.356,
      "step": 90000
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.9895148628150198e-05,
      "loss": 0.6183,
      "step": 90500
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.972882348244937e-05,
      "loss": 0.617,
      "step": 91000
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.9562498336748545e-05,
      "loss": 0.5859,
      "step": 91500
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.9396173191047717e-05,
      "loss": 0.6061,
      "step": 92000
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.9229848045346886e-05,
      "loss": 0.5971,
      "step": 92500
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.906352289964606e-05,
      "loss": 0.6302,
      "step": 93000
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.8897197753945234e-05,
      "loss": 0.5907,
      "step": 93500
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.8730872608244406e-05,
      "loss": 0.6167,
      "step": 94000
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.8564547462543578e-05,
      "loss": 0.5929,
      "step": 94500
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.839822231684275e-05,
      "loss": 0.5778,
      "step": 95000
    },
    {
      "epoch": 1.26,
      "eval_bleu": 47.8127,
      "eval_chrf++": 67.3458,
      "eval_gen_len": 24.2553,
      "eval_loss": 0.7928242683410645,
      "eval_runtime": 4822.252,
      "eval_samples_per_second": 4.826,
      "eval_spbleu": 55.1003,
      "eval_steps_per_second": 0.965,
      "eval_ter": 41.2697,
      "step": 95000
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.8231897171141922e-05,
      "loss": 0.6058,
      "step": 95500
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.8065572025441097e-05,
      "loss": 0.6104,
      "step": 96000
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.7899246879740266e-05,
      "loss": 0.5958,
      "step": 96500
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.7732921734039438e-05,
      "loss": 0.5968,
      "step": 97000
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.7566596588338614e-05,
      "loss": 0.6034,
      "step": 97500
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.7400271442637782e-05,
      "loss": 0.6189,
      "step": 98000
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.7233946296936958e-05,
      "loss": 0.6028,
      "step": 98500
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.706762115123613e-05,
      "loss": 0.6044,
      "step": 99000
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.6901296005535302e-05,
      "loss": 0.6021,
      "step": 99500
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.6734970859834474e-05,
      "loss": 0.6055,
      "step": 100000
    },
    {
      "epoch": 1.33,
      "eval_bleu": 47.9968,
      "eval_chrf++": 67.5009,
      "eval_gen_len": 24.343,
      "eval_loss": 0.7921544313430786,
      "eval_runtime": 4846.7187,
      "eval_samples_per_second": 4.802,
      "eval_spbleu": 55.2772,
      "eval_steps_per_second": 0.96,
      "eval_ter": 41.328,
      "step": 100000
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.6568645714133646e-05,
      "loss": 0.6328,
      "step": 100500
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.6402320568432818e-05,
      "loss": 0.6082,
      "step": 101000
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.6235995422731994e-05,
      "loss": 0.6217,
      "step": 101500
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.6069670277031162e-05,
      "loss": 0.6225,
      "step": 102000
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.5903345131330335e-05,
      "loss": 0.6126,
      "step": 102500
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.573701998562951e-05,
      "loss": 0.6208,
      "step": 103000
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.5570694839928682e-05,
      "loss": 0.6068,
      "step": 103500
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.540436969422785e-05,
      "loss": 0.6062,
      "step": 104000
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.5238044548527025e-05,
      "loss": 0.6063,
      "step": 104500
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.5071719402826198e-05,
      "loss": 0.6136,
      "step": 105000
    },
    {
      "epoch": 1.4,
      "eval_bleu": 48.1927,
      "eval_chrf++": 67.6488,
      "eval_gen_len": 24.2325,
      "eval_loss": 0.7866764664649963,
      "eval_runtime": 4833.6742,
      "eval_samples_per_second": 4.815,
      "eval_spbleu": 55.5056,
      "eval_steps_per_second": 0.963,
      "eval_ter": 40.8666,
      "step": 105000
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.490539425712537e-05,
      "loss": 0.5853,
      "step": 105500
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.4739069111424541e-05,
      "loss": 0.6171,
      "step": 106000
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.4572743965723715e-05,
      "loss": 0.6224,
      "step": 106500
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.4406418820022888e-05,
      "loss": 0.6133,
      "step": 107000
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.4240093674322059e-05,
      "loss": 0.6056,
      "step": 107500
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.4073768528621231e-05,
      "loss": 0.6023,
      "step": 108000
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.3907443382920405e-05,
      "loss": 0.6124,
      "step": 108500
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.3741118237219577e-05,
      "loss": 0.5926,
      "step": 109000
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.3574793091518747e-05,
      "loss": 0.6123,
      "step": 109500
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.3408467945817921e-05,
      "loss": 0.5983,
      "step": 110000
    },
    {
      "epoch": 1.46,
      "eval_bleu": 48.2871,
      "eval_chrf++": 67.7183,
      "eval_gen_len": 24.2734,
      "eval_loss": 0.7818799018859863,
      "eval_runtime": 4831.9282,
      "eval_samples_per_second": 4.817,
      "eval_spbleu": 55.6111,
      "eval_steps_per_second": 0.963,
      "eval_ter": 40.8669,
      "step": 110000
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.3242142800117093e-05,
      "loss": 0.6285,
      "step": 110500
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.3075817654416267e-05,
      "loss": 0.5953,
      "step": 111000
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.2909492508715437e-05,
      "loss": 0.6105,
      "step": 111500
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.2743167363014611e-05,
      "loss": 0.6008,
      "step": 112000
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.2576842217313783e-05,
      "loss": 0.615,
      "step": 112500
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.2410517071612955e-05,
      "loss": 0.6252,
      "step": 113000
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.2244191925912129e-05,
      "loss": 0.5896,
      "step": 113500
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.20778667802113e-05,
      "loss": 0.5892,
      "step": 114000
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.1911541634510473e-05,
      "loss": 0.5956,
      "step": 114500
    },
    {
      "epoch": 1.53,
      "learning_rate": 1.1745216488809645e-05,
      "loss": 0.6225,
      "step": 115000
    },
    {
      "epoch": 1.53,
      "eval_bleu": 48.489,
      "eval_chrf++": 67.8465,
      "eval_gen_len": 24.3063,
      "eval_loss": 0.7788467407226562,
      "eval_runtime": 4828.3962,
      "eval_samples_per_second": 4.82,
      "eval_spbleu": 55.8079,
      "eval_steps_per_second": 0.964,
      "eval_ter": 40.8166,
      "step": 115000
    },
    {
      "epoch": 1.54,
      "learning_rate": 1.1578891343108817e-05,
      "loss": 0.5987,
      "step": 115500
    },
    {
      "epoch": 1.54,
      "learning_rate": 1.141256619740799e-05,
      "loss": 0.5966,
      "step": 116000
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.1246241051707161e-05,
      "loss": 0.6044,
      "step": 116500
    },
    {
      "epoch": 1.56,
      "learning_rate": 1.1079915906006334e-05,
      "loss": 0.6069,
      "step": 117000
    },
    {
      "epoch": 1.56,
      "learning_rate": 1.0913590760305506e-05,
      "loss": 0.586,
      "step": 117500
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.074726561460468e-05,
      "loss": 0.6076,
      "step": 118000
    },
    {
      "epoch": 1.58,
      "learning_rate": 1.0580940468903852e-05,
      "loss": 0.5896,
      "step": 118500
    },
    {
      "epoch": 1.58,
      "learning_rate": 1.0414615323203024e-05,
      "loss": 0.5918,
      "step": 119000
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.0248290177502196e-05,
      "loss": 0.5885,
      "step": 119500
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.0081965031801368e-05,
      "loss": 0.5937,
      "step": 120000
    },
    {
      "epoch": 1.6,
      "eval_bleu": 48.6385,
      "eval_chrf++": 67.9368,
      "eval_gen_len": 24.2573,
      "eval_loss": 0.7750729322433472,
      "eval_runtime": 4834.5977,
      "eval_samples_per_second": 4.814,
      "eval_spbleu": 55.9507,
      "eval_steps_per_second": 0.963,
      "eval_ter": 40.604,
      "step": 120000
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.91563988610054e-06,
      "loss": 0.5941,
      "step": 120500
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.749314740399714e-06,
      "loss": 0.5856,
      "step": 121000
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.582989594698886e-06,
      "loss": 0.5904,
      "step": 121500
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.416664448998058e-06,
      "loss": 0.5872,
      "step": 122000
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.25033930329723e-06,
      "loss": 0.6085,
      "step": 122500
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.084014157596404e-06,
      "loss": 0.6025,
      "step": 123000
    },
    {
      "epoch": 1.64,
      "learning_rate": 8.917689011895574e-06,
      "loss": 0.6026,
      "step": 123500
    },
    {
      "epoch": 1.65,
      "learning_rate": 8.751363866194748e-06,
      "loss": 0.5826,
      "step": 124000
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.58503872049392e-06,
      "loss": 0.5989,
      "step": 124500
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.418713574793092e-06,
      "loss": 0.5934,
      "step": 125000
    },
    {
      "epoch": 1.66,
      "eval_bleu": 48.7432,
      "eval_chrf++": 67.9418,
      "eval_gen_len": 24.2522,
      "eval_loss": 0.7731887698173523,
      "eval_runtime": 4826.8535,
      "eval_samples_per_second": 4.822,
      "eval_spbleu": 56.0071,
      "eval_steps_per_second": 0.964,
      "eval_ter": 40.5337,
      "step": 125000
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.252388429092264e-06,
      "loss": 0.5797,
      "step": 125500
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.086063283391436e-06,
      "loss": 0.5713,
      "step": 126000
    },
    {
      "epoch": 1.68,
      "learning_rate": 7.919738137690608e-06,
      "loss": 0.5993,
      "step": 126500
    },
    {
      "epoch": 1.69,
      "learning_rate": 7.75341299198978e-06,
      "loss": 0.5953,
      "step": 127000
    },
    {
      "epoch": 1.7,
      "learning_rate": 7.587087846288954e-06,
      "loss": 0.5866,
      "step": 127500
    },
    {
      "epoch": 1.7,
      "learning_rate": 7.4207627005881254e-06,
      "loss": 0.5942,
      "step": 128000
    },
    {
      "epoch": 1.71,
      "learning_rate": 7.254437554887298e-06,
      "loss": 0.6301,
      "step": 128500
    },
    {
      "epoch": 1.72,
      "learning_rate": 7.0881124091864705e-06,
      "loss": 0.6143,
      "step": 129000
    },
    {
      "epoch": 1.72,
      "learning_rate": 6.921787263485643e-06,
      "loss": 0.5774,
      "step": 129500
    },
    {
      "epoch": 1.73,
      "learning_rate": 6.7554621177848155e-06,
      "loss": 0.5974,
      "step": 130000
    },
    {
      "epoch": 1.73,
      "eval_bleu": 48.9376,
      "eval_chrf++": 68.1416,
      "eval_gen_len": 24.3125,
      "eval_loss": 0.7730206847190857,
      "eval_runtime": 4839.9438,
      "eval_samples_per_second": 4.809,
      "eval_spbleu": 56.1898,
      "eval_steps_per_second": 0.962,
      "eval_ter": 40.479,
      "step": 130000
    },
    {
      "epoch": 1.74,
      "learning_rate": 6.589136972083988e-06,
      "loss": 0.5838,
      "step": 130500
    },
    {
      "epoch": 1.74,
      "learning_rate": 6.42281182638316e-06,
      "loss": 0.5732,
      "step": 131000
    },
    {
      "epoch": 1.75,
      "learning_rate": 6.2564866806823334e-06,
      "loss": 0.5922,
      "step": 131500
    },
    {
      "epoch": 1.76,
      "learning_rate": 6.090161534981505e-06,
      "loss": 0.5852,
      "step": 132000
    },
    {
      "epoch": 1.76,
      "learning_rate": 5.923836389280677e-06,
      "loss": 0.5853,
      "step": 132500
    },
    {
      "epoch": 1.77,
      "learning_rate": 5.75751124357985e-06,
      "loss": 0.6043,
      "step": 133000
    },
    {
      "epoch": 1.78,
      "learning_rate": 5.591186097879022e-06,
      "loss": 0.6143,
      "step": 133500
    },
    {
      "epoch": 1.78,
      "learning_rate": 5.424860952178194e-06,
      "loss": 0.5775,
      "step": 134000
    },
    {
      "epoch": 1.79,
      "learning_rate": 5.258535806477367e-06,
      "loss": 0.5763,
      "step": 134500
    },
    {
      "epoch": 1.8,
      "learning_rate": 5.092210660776539e-06,
      "loss": 0.5944,
      "step": 135000
    },
    {
      "epoch": 1.8,
      "eval_bleu": 49.0549,
      "eval_chrf++": 68.2099,
      "eval_gen_len": 24.3198,
      "eval_loss": 0.7679416537284851,
      "eval_runtime": 4837.1327,
      "eval_samples_per_second": 4.812,
      "eval_spbleu": 56.3421,
      "eval_steps_per_second": 0.962,
      "eval_ter": 40.2163,
      "step": 135000
    },
    {
      "epoch": 1.8,
      "learning_rate": 4.925885515075712e-06,
      "loss": 0.6115,
      "step": 135500
    },
    {
      "epoch": 1.81,
      "learning_rate": 4.759560369374884e-06,
      "loss": 0.5966,
      "step": 136000
    },
    {
      "epoch": 1.82,
      "learning_rate": 4.593235223674056e-06,
      "loss": 0.5748,
      "step": 136500
    },
    {
      "epoch": 1.82,
      "learning_rate": 4.426910077973229e-06,
      "loss": 0.573,
      "step": 137000
    },
    {
      "epoch": 1.83,
      "learning_rate": 4.260584932272401e-06,
      "loss": 0.5958,
      "step": 137500
    },
    {
      "epoch": 1.84,
      "learning_rate": 4.094259786571573e-06,
      "loss": 0.577,
      "step": 138000
    },
    {
      "epoch": 1.84,
      "learning_rate": 3.927934640870746e-06,
      "loss": 0.5966,
      "step": 138500
    },
    {
      "epoch": 1.85,
      "learning_rate": 3.761609495169918e-06,
      "loss": 0.5884,
      "step": 139000
    },
    {
      "epoch": 1.86,
      "learning_rate": 3.5952843494690906e-06,
      "loss": 0.5924,
      "step": 139500
    },
    {
      "epoch": 1.86,
      "learning_rate": 3.4289592037682623e-06,
      "loss": 0.5753,
      "step": 140000
    },
    {
      "epoch": 1.86,
      "eval_bleu": 49.1791,
      "eval_chrf++": 68.2787,
      "eval_gen_len": 24.3117,
      "eval_loss": 0.7665663361549377,
      "eval_runtime": 4836.1543,
      "eval_samples_per_second": 4.813,
      "eval_spbleu": 56.4922,
      "eval_steps_per_second": 0.963,
      "eval_ter": 40.1748,
      "step": 140000
    },
    {
      "epoch": 1.87,
      "learning_rate": 3.262634058067435e-06,
      "loss": 0.5892,
      "step": 140500
    },
    {
      "epoch": 1.88,
      "learning_rate": 3.0963089123666073e-06,
      "loss": 0.5891,
      "step": 141000
    },
    {
      "epoch": 1.88,
      "learning_rate": 2.92998376666578e-06,
      "loss": 0.5821,
      "step": 141500
    },
    {
      "epoch": 1.89,
      "learning_rate": 2.7636586209649524e-06,
      "loss": 0.5701,
      "step": 142000
    },
    {
      "epoch": 1.9,
      "learning_rate": 2.5973334752641244e-06,
      "loss": 0.5632,
      "step": 142500
    },
    {
      "epoch": 1.9,
      "learning_rate": 2.4310083295632965e-06,
      "loss": 0.588,
      "step": 143000
    },
    {
      "epoch": 1.91,
      "learning_rate": 2.264683183862469e-06,
      "loss": 0.604,
      "step": 143500
    },
    {
      "epoch": 1.92,
      "learning_rate": 2.0983580381616416e-06,
      "loss": 0.5766,
      "step": 144000
    },
    {
      "epoch": 1.92,
      "learning_rate": 1.932032892460814e-06,
      "loss": 0.595,
      "step": 144500
    },
    {
      "epoch": 1.93,
      "learning_rate": 1.7657077467599864e-06,
      "loss": 0.5907,
      "step": 145000
    },
    {
      "epoch": 1.93,
      "eval_bleu": 49.2939,
      "eval_chrf++": 68.3612,
      "eval_gen_len": 24.3154,
      "eval_loss": 0.7650372982025146,
      "eval_runtime": 4848.1649,
      "eval_samples_per_second": 4.801,
      "eval_spbleu": 56.5766,
      "eval_steps_per_second": 0.96,
      "eval_ter": 40.0916,
      "step": 145000
    },
    {
      "epoch": 1.94,
      "learning_rate": 1.5993826010591587e-06,
      "loss": 0.6039,
      "step": 145500
    },
    {
      "epoch": 1.94,
      "learning_rate": 1.433057455358331e-06,
      "loss": 0.5995,
      "step": 146000
    },
    {
      "epoch": 1.95,
      "learning_rate": 1.2667323096575033e-06,
      "loss": 0.5899,
      "step": 146500
    },
    {
      "epoch": 1.96,
      "learning_rate": 1.1004071639566756e-06,
      "loss": 0.5666,
      "step": 147000
    },
    {
      "epoch": 1.96,
      "learning_rate": 9.340820182558481e-07,
      "loss": 0.5925,
      "step": 147500
    },
    {
      "epoch": 1.97,
      "learning_rate": 7.677568725550204e-07,
      "loss": 0.6101,
      "step": 148000
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.014317268541928e-07,
      "loss": 0.6,
      "step": 148500
    },
    {
      "epoch": 1.98,
      "learning_rate": 4.351065811533651e-07,
      "loss": 0.5961,
      "step": 149000
    },
    {
      "epoch": 1.99,
      "learning_rate": 2.687814354525375e-07,
      "loss": 0.5851,
      "step": 149500
    },
    {
      "epoch": 2.0,
      "learning_rate": 1.0245628975170984e-07,
      "loss": 0.5811,
      "step": 150000
    },
    {
      "epoch": 2.0,
      "eval_bleu": 49.3031,
      "eval_chrf++": 68.3562,
      "eval_gen_len": 24.3172,
      "eval_loss": 0.7639886140823364,
      "eval_runtime": 4840.3648,
      "eval_samples_per_second": 4.808,
      "eval_spbleu": 56.6,
      "eval_steps_per_second": 0.962,
      "eval_ter": 40.1023,
      "step": 150000
    }
  ],
  "logging_steps": 500,
  "max_steps": 150308,
  "num_train_epochs": 2,
  "save_steps": 5000,
  "total_flos": 6.383678253160202e+17,
  "trial_name": null,
  "trial_params": null
}