{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 50,
  "global_step": 4254,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 9.998432847516064e-05,
      "loss": 1.7366,
      "step": 1
    },
    {
      "epoch": 0.02,
      "learning_rate": 9.921642375803167e-05,
      "loss": 1.6411,
      "step": 50
    },
    {
      "epoch": 0.02,
      "eval_bleu": 0.0,
      "eval_brevity_penalty": 0.0444548482935275,
      "eval_exact_match": 0.06640625,
      "eval_gen_len": 2.0,
      "eval_length_ratio": 0.24311490978157646,
      "eval_loss": 2.303663492202759,
      "eval_reference_length": 2106,
      "eval_runtime": 21.5251,
      "eval_samples_per_second": 23.786,
      "eval_steps_per_second": 1.487,
      "eval_translation_length": 512,
      "step": 50
    },
    {
      "epoch": 0.05,
      "learning_rate": 9.843284751606332e-05,
      "loss": 2.8505,
      "step": 100
    },
    {
      "epoch": 0.05,
      "eval_bleu": 0.0,
      "eval_brevity_penalty": 0.0444548482935275,
      "eval_exact_match": 0.068359375,
      "eval_gen_len": 2.0,
      "eval_length_ratio": 0.24311490978157646,
      "eval_loss": 2.0377542972564697,
      "eval_reference_length": 2106,
      "eval_runtime": 21.9349,
      "eval_samples_per_second": 23.342,
      "eval_steps_per_second": 1.459,
      "eval_translation_length": 512,
      "step": 100
    },
    {
      "epoch": 0.07,
      "learning_rate": 9.764927127409498e-05,
      "loss": 2.5985,
      "step": 150
    },
    {
      "epoch": 0.07,
      "eval_bleu": 0.0,
      "eval_brevity_penalty": 0.05906652611110617,
      "eval_exact_match": 0.060546875,
      "eval_gen_len": 2.115234375,
      "eval_length_ratio": 0.2611585944919278,
      "eval_loss": 1.8557136058807373,
      "eval_reference_length": 2106,
      "eval_runtime": 22.3363,
      "eval_samples_per_second": 22.922,
      "eval_steps_per_second": 1.433,
      "eval_translation_length": 550,
      "step": 150
    },
    {
      "epoch": 0.09,
      "learning_rate": 9.686569503212663e-05,
      "loss": 2.4089,
      "step": 200
    },
    {
      "epoch": 0.09,
      "eval_bleu": 0.011576539062082333,
      "eval_brevity_penalty": 0.0658189125921447,
      "eval_exact_match": 0.060546875,
      "eval_gen_len": 2.208984375,
      "eval_length_ratio": 0.26875593542260207,
      "eval_loss": 1.8144108057022095,
      "eval_reference_length": 2106,
      "eval_runtime": 23.7934,
      "eval_samples_per_second": 21.519,
      "eval_steps_per_second": 1.345,
      "eval_translation_length": 566,
      "step": 200
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.608211879015829e-05,
      "loss": 2.3867,
      "step": 250
    },
    {
      "epoch": 0.12,
      "eval_bleu": 0.06335053804735992,
      "eval_brevity_penalty": 0.40697312998195423,
      "eval_exact_match": 0.05859375,
      "eval_gen_len": 3.8125,
      "eval_length_ratio": 0.52659069325736,
      "eval_loss": 1.7400383949279785,
      "eval_reference_length": 2106,
      "eval_runtime": 29.4794,
      "eval_samples_per_second": 17.368,
      "eval_steps_per_second": 1.086,
      "eval_translation_length": 1109,
      "step": 250
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.529854254818994e-05,
      "loss": 2.199,
      "step": 300
    },
    {
      "epoch": 0.14,
      "eval_bleu": 0.1636518792509081,
      "eval_brevity_penalty": 0.7065164246788794,
      "eval_exact_match": 0.103515625,
      "eval_gen_len": 5.21875,
      "eval_length_ratio": 0.7421652421652422,
      "eval_loss": 1.551315188407898,
      "eval_reference_length": 2106,
      "eval_runtime": 32.1335,
      "eval_samples_per_second": 15.934,
      "eval_steps_per_second": 0.996,
      "eval_translation_length": 1563,
      "step": 300
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.45149663062216e-05,
      "loss": 2.1623,
      "step": 350
    },
    {
      "epoch": 0.16,
      "eval_bleu": 0.2018876786015886,
      "eval_brevity_penalty": 0.6032877660521909,
      "eval_exact_match": 0.171875,
      "eval_gen_len": 4.716796875,
      "eval_length_ratio": 0.664292497625831,
      "eval_loss": 1.3646727800369263,
      "eval_reference_length": 2106,
      "eval_runtime": 31.8812,
      "eval_samples_per_second": 16.06,
      "eval_steps_per_second": 1.004,
      "eval_translation_length": 1399,
      "step": 350
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.373139006425325e-05,
      "loss": 1.9871,
      "step": 400
    },
    {
      "epoch": 0.19,
      "eval_bleu": 0.3336189639058238,
      "eval_brevity_penalty": 0.9918951577772216,
      "eval_exact_match": 0.25,
      "eval_gen_len": 6.9140625,
      "eval_length_ratio": 0.9919278252611586,
      "eval_loss": 1.186793327331543,
      "eval_reference_length": 2106,
      "eval_runtime": 35.7547,
      "eval_samples_per_second": 14.32,
      "eval_steps_per_second": 0.895,
      "eval_translation_length": 2089,
      "step": 400
    },
    {
      "epoch": 0.21,
      "learning_rate": 9.294781382228491e-05,
      "loss": 1.7705,
      "step": 450
    },
    {
      "epoch": 0.21,
      "eval_bleu": 0.3747254791441505,
      "eval_brevity_penalty": 0.8750477201167791,
      "eval_exact_match": 0.298828125,
      "eval_gen_len": 6.1953125,
      "eval_length_ratio": 0.8822412155745489,
      "eval_loss": 1.119579553604126,
      "eval_reference_length": 2106,
      "eval_runtime": 34.9183,
      "eval_samples_per_second": 14.663,
      "eval_steps_per_second": 0.916,
      "eval_translation_length": 1858,
      "step": 450
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.216423758031657e-05,
      "loss": 1.589,
      "step": 500
    },
    {
      "epoch": 0.24,
      "eval_bleu": 0.39875210184754006,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.328125,
      "eval_gen_len": 7.37109375,
      "eval_length_ratio": 1.0802469135802468,
      "eval_loss": 1.0730499029159546,
      "eval_reference_length": 2106,
      "eval_runtime": 37.8703,
      "eval_samples_per_second": 13.52,
      "eval_steps_per_second": 0.845,
      "eval_translation_length": 2275,
      "step": 500
    },
    {
      "epoch": 0.26,
      "learning_rate": 9.138066133834823e-05,
      "loss": 1.5929,
      "step": 550
    },
    {
      "epoch": 0.26,
      "eval_bleu": 0.46535583502999767,
      "eval_brevity_penalty": 0.9715887436284736,
      "eval_exact_match": 0.357421875,
      "eval_gen_len": 6.783203125,
      "eval_length_ratio": 0.9719848053181387,
      "eval_loss": 0.9941992163658142,
      "eval_reference_length": 2106,
      "eval_runtime": 35.7255,
      "eval_samples_per_second": 14.331,
      "eval_steps_per_second": 0.896,
      "eval_translation_length": 2047,
      "step": 550
    },
    {
      "epoch": 0.28,
      "learning_rate": 9.059708509637988e-05,
      "loss": 1.5104,
      "step": 600
    },
    {
      "epoch": 0.28,
      "eval_bleu": 0.47603827067185095,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.3828125,
      "eval_gen_len": 7.125,
      "eval_length_ratio": 1.0261158594491928,
      "eval_loss": 0.9567559957504272,
      "eval_reference_length": 2106,
      "eval_runtime": 36.3056,
      "eval_samples_per_second": 14.103,
      "eval_steps_per_second": 0.881,
      "eval_translation_length": 2161,
      "step": 600
    },
    {
      "epoch": 0.31,
      "learning_rate": 8.981350885441154e-05,
      "loss": 1.382,
      "step": 650
    },
    {
      "epoch": 0.31,
      "eval_bleu": 0.4441517830638515,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.39453125,
      "eval_gen_len": 7.638671875,
      "eval_length_ratio": 1.1277302943969612,
      "eval_loss": 0.9603753089904785,
      "eval_reference_length": 2106,
      "eval_runtime": 37.6569,
      "eval_samples_per_second": 13.596,
      "eval_steps_per_second": 0.85,
      "eval_translation_length": 2375,
      "step": 650
    },
    {
      "epoch": 0.33,
      "learning_rate": 8.90299326124432e-05,
      "loss": 1.4739,
      "step": 700
    },
    {
      "epoch": 0.33,
      "eval_bleu": 0.5062868407195426,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.416015625,
      "eval_gen_len": 7.041015625,
      "eval_length_ratio": 1.0090218423551758,
      "eval_loss": 0.878071665763855,
      "eval_reference_length": 2106,
      "eval_runtime": 35.5071,
      "eval_samples_per_second": 14.42,
      "eval_steps_per_second": 0.901,
      "eval_translation_length": 2125,
      "step": 700
    },
    {
      "epoch": 0.35,
      "learning_rate": 8.824635637047485e-05,
      "loss": 1.3245,
      "step": 750
    },
    {
      "epoch": 0.35,
      "eval_bleu": 0.5210361604149488,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.419921875,
      "eval_gen_len": 7.25390625,
      "eval_length_ratio": 1.0574548907882242,
      "eval_loss": 0.9063434600830078,
      "eval_reference_length": 2106,
      "eval_runtime": 34.9391,
      "eval_samples_per_second": 14.654,
      "eval_steps_per_second": 0.916,
      "eval_translation_length": 2227,
      "step": 750
    },
    {
      "epoch": 0.38,
      "learning_rate": 8.746278012850651e-05,
      "loss": 1.3303,
      "step": 800
    },
    {
      "epoch": 0.38,
      "eval_bleu": 0.4770015676872273,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.3984375,
      "eval_gen_len": 7.439453125,
      "eval_length_ratio": 1.0826210826210827,
      "eval_loss": 0.9164894819259644,
      "eval_reference_length": 2106,
      "eval_runtime": 36.5235,
      "eval_samples_per_second": 14.018,
      "eval_steps_per_second": 0.876,
      "eval_translation_length": 2280,
      "step": 800
    },
    {
      "epoch": 0.4,
      "learning_rate": 8.667920388653816e-05,
      "loss": 1.2626,
      "step": 850
    },
    {
      "epoch": 0.4,
      "eval_bleu": 0.5256219338123865,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.40625,
      "eval_gen_len": 7.05859375,
      "eval_length_ratio": 1.0237416904083572,
      "eval_loss": 0.8638509511947632,
      "eval_reference_length": 2106,
      "eval_runtime": 34.7224,
      "eval_samples_per_second": 14.746,
      "eval_steps_per_second": 0.922,
      "eval_translation_length": 2156,
      "step": 850
    },
    {
      "epoch": 0.42,
      "learning_rate": 8.589562764456982e-05,
      "loss": 1.2751,
      "step": 900
    },
    {
      "epoch": 0.42,
      "eval_bleu": 0.5549630399715574,
      "eval_brevity_penalty": 0.995717351313512,
      "eval_exact_match": 0.421875,
      "eval_gen_len": 6.89453125,
      "eval_length_ratio": 0.9957264957264957,
      "eval_loss": 0.8271727561950684,
      "eval_reference_length": 2106,
      "eval_runtime": 36.4284,
      "eval_samples_per_second": 14.055,
      "eval_steps_per_second": 0.878,
      "eval_translation_length": 2097,
      "step": 900
    },
    {
      "epoch": 0.45,
      "learning_rate": 8.511205140260148e-05,
      "loss": 1.291,
      "step": 950
    },
    {
      "epoch": 0.45,
      "eval_bleu": 0.5601197353946883,
      "eval_brevity_penalty": 0.9754879045293495,
      "eval_exact_match": 0.447265625,
      "eval_gen_len": 6.837890625,
      "eval_length_ratio": 0.9757834757834758,
      "eval_loss": 0.827507734298706,
      "eval_reference_length": 2106,
      "eval_runtime": 35.2125,
      "eval_samples_per_second": 14.54,
      "eval_steps_per_second": 0.909,
      "eval_translation_length": 2055,
      "step": 950
    },
    {
      "epoch": 0.47,
      "learning_rate": 8.432847516063314e-05,
      "loss": 1.0992,
      "step": 1000
    },
    {
      "epoch": 0.47,
      "eval_bleu": 0.5469392101262274,
      "eval_brevity_penalty": 0.9538582733351182,
      "eval_exact_match": 0.4453125,
      "eval_gen_len": 6.73046875,
      "eval_length_ratio": 0.9548907882241215,
      "eval_loss": 0.8413195610046387,
      "eval_reference_length": 2106,
      "eval_runtime": 34.3944,
      "eval_samples_per_second": 14.886,
      "eval_steps_per_second": 0.93,
      "eval_translation_length": 2011,
      "step": 1000
    },
    {
      "epoch": 0.49,
      "learning_rate": 8.354489891866479e-05,
      "loss": 1.1555,
      "step": 1050
    },
    {
      "epoch": 0.49,
      "eval_bleu": 0.5455158722304639,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.466796875,
      "eval_gen_len": 7.326171875,
      "eval_length_ratio": 1.0607787274453941,
      "eval_loss": 0.8606413006782532,
      "eval_reference_length": 2106,
      "eval_runtime": 35.8914,
      "eval_samples_per_second": 14.265,
      "eval_steps_per_second": 0.892,
      "eval_translation_length": 2234,
      "step": 1050
    },
    {
      "epoch": 0.52,
      "learning_rate": 8.276132267669645e-05,
      "loss": 1.2137,
      "step": 1100
    },
    {
      "epoch": 0.52,
      "eval_bleu": 0.5620322190043872,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.447265625,
      "eval_gen_len": 7.017578125,
      "eval_length_ratio": 1.0118708452041785,
      "eval_loss": 0.8165176510810852,
      "eval_reference_length": 2106,
      "eval_runtime": 35.4665,
      "eval_samples_per_second": 14.436,
      "eval_steps_per_second": 0.902,
      "eval_translation_length": 2131,
      "step": 1100
    },
    {
      "epoch": 0.54,
      "learning_rate": 8.197774643472809e-05,
      "loss": 1.2334,
      "step": 1150
    },
    {
      "epoch": 0.54,
      "eval_bleu": 0.561256814480964,
      "eval_brevity_penalty": 0.9573296047876775,
      "eval_exact_match": 0.443359375,
      "eval_gen_len": 6.708984375,
      "eval_length_ratio": 0.9582146248812915,
      "eval_loss": 0.8095741271972656,
      "eval_reference_length": 2106,
      "eval_runtime": 34.1264,
      "eval_samples_per_second": 15.003,
      "eval_steps_per_second": 0.938,
      "eval_translation_length": 2018,
      "step": 1150
    },
    {
      "epoch": 0.56,
      "learning_rate": 8.119417019275976e-05,
      "loss": 1.2993,
      "step": 1200
    },
    {
      "epoch": 0.56,
      "eval_bleu": 0.588770590240424,
      "eval_brevity_penalty": 0.9808251378136823,
      "eval_exact_match": 0.451171875,
      "eval_gen_len": 6.87109375,
      "eval_length_ratio": 0.9810066476733144,
      "eval_loss": 0.7695479393005371,
      "eval_reference_length": 2106,
      "eval_runtime": 34.8276,
      "eval_samples_per_second": 14.701,
      "eval_steps_per_second": 0.919,
      "eval_translation_length": 2066,
      "step": 1200
    },
    {
      "epoch": 0.59,
      "learning_rate": 8.04105939507914e-05,
      "loss": 1.2508,
      "step": 1250
    },
    {
      "epoch": 0.59,
      "eval_bleu": 0.563750421831778,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.462890625,
      "eval_gen_len": 7.181640625,
      "eval_length_ratio": 1.047483380816714,
      "eval_loss": 0.7718470096588135,
      "eval_reference_length": 2106,
      "eval_runtime": 35.174,
      "eval_samples_per_second": 14.556,
      "eval_steps_per_second": 0.91,
      "eval_translation_length": 2206,
      "step": 1250
    },
    {
      "epoch": 0.61,
      "learning_rate": 7.962701770882307e-05,
      "loss": 1.0999,
      "step": 1300
    },
    {
      "epoch": 0.61,
      "eval_bleu": 0.5781765484721586,
      "eval_brevity_penalty": 0.9715887436284736,
      "eval_exact_match": 0.447265625,
      "eval_gen_len": 6.826171875,
      "eval_length_ratio": 0.9719848053181387,
      "eval_loss": 0.8051897883415222,
      "eval_reference_length": 2106,
      "eval_runtime": 34.2261,
      "eval_samples_per_second": 14.959,
      "eval_steps_per_second": 0.935,
      "eval_translation_length": 2047,
      "step": 1300
    },
    {
      "epoch": 0.63,
      "learning_rate": 7.884344146685473e-05,
      "loss": 1.1623,
      "step": 1350
    },
    {
      "epoch": 0.63,
      "eval_bleu": 0.5717608964815384,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.478515625,
      "eval_gen_len": 7.076171875,
      "eval_length_ratio": 1.017094017094017,
      "eval_loss": 0.7817097306251526,
      "eval_reference_length": 2106,
      "eval_runtime": 34.9684,
      "eval_samples_per_second": 14.642,
      "eval_steps_per_second": 0.915,
      "eval_translation_length": 2142,
      "step": 1350
    },
    {
      "epoch": 0.66,
      "learning_rate": 7.805986522488639e-05,
      "loss": 1.1405,
      "step": 1400
    },
    {
      "epoch": 0.66,
      "eval_bleu": 0.5807905073436569,
      "eval_brevity_penalty": 0.9774319412545756,
      "eval_exact_match": 0.46484375,
      "eval_gen_len": 6.876953125,
      "eval_length_ratio": 0.9776828110161444,
      "eval_loss": 0.8067466616630554,
      "eval_reference_length": 2106,
      "eval_runtime": 34.9137,
      "eval_samples_per_second": 14.665,
      "eval_steps_per_second": 0.917,
      "eval_translation_length": 2059,
      "step": 1400
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.727628898291805e-05,
      "loss": 1.0645,
      "step": 1450
    },
    {
      "epoch": 0.68,
      "eval_bleu": 0.6024093506065553,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.48046875,
      "eval_gen_len": 6.974609375,
      "eval_length_ratio": 1.0056980056980056,
      "eval_loss": 0.7935568690299988,
      "eval_reference_length": 2106,
      "eval_runtime": 35.237,
      "eval_samples_per_second": 14.53,
      "eval_steps_per_second": 0.908,
      "eval_translation_length": 2118,
      "step": 1450
    },
    {
      "epoch": 0.71,
      "learning_rate": 7.64927127409497e-05,
      "loss": 1.114,
      "step": 1500
    },
    {
      "epoch": 0.71,
      "eval_bleu": 0.5585335749210942,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.482421875,
      "eval_gen_len": 7.26953125,
      "eval_length_ratio": 1.0555555555555556,
      "eval_loss": 0.7591338157653809,
      "eval_reference_length": 2106,
      "eval_runtime": 35.9474,
      "eval_samples_per_second": 14.243,
      "eval_steps_per_second": 0.89,
      "eval_translation_length": 2223,
      "step": 1500
    },
    {
      "epoch": 0.73,
      "learning_rate": 7.570913649898136e-05,
      "loss": 0.9622,
      "step": 1550
    },
    {
      "epoch": 0.73,
      "eval_bleu": 0.6037520631327163,
      "eval_brevity_penalty": 0.996194105459075,
      "eval_exact_match": 0.517578125,
      "eval_gen_len": 6.9609375,
      "eval_length_ratio": 0.9962013295346629,
      "eval_loss": 0.7658359408378601,
      "eval_reference_length": 2106,
      "eval_runtime": 35.7224,
      "eval_samples_per_second": 14.333,
      "eval_steps_per_second": 0.896,
      "eval_translation_length": 2098,
      "step": 1550
    },
    {
      "epoch": 0.75,
      "learning_rate": 7.4925560257013e-05,
      "loss": 1.0521,
      "step": 1600
    },
    {
      "epoch": 0.75,
      "eval_bleu": 0.5825978988431111,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.509765625,
      "eval_gen_len": 7.033203125,
      "eval_length_ratio": 1.000474833808167,
      "eval_loss": 0.7511448860168457,
      "eval_reference_length": 2106,
      "eval_runtime": 35.8085,
      "eval_samples_per_second": 14.298,
      "eval_steps_per_second": 0.894,
      "eval_translation_length": 2107,
      "step": 1600
    },
    {
      "epoch": 0.78,
      "learning_rate": 7.414198401504467e-05,
      "loss": 1.0841,
      "step": 1650
    },
    {
      "epoch": 0.78,
      "eval_bleu": 0.5825352487207182,
      "eval_brevity_penalty": 0.9759742597265093,
      "eval_exact_match": 0.486328125,
      "eval_gen_len": 6.845703125,
      "eval_length_ratio": 0.976258309591643,
      "eval_loss": 0.7474812269210815,
      "eval_reference_length": 2106,
      "eval_runtime": 35.5909,
      "eval_samples_per_second": 14.386,
      "eval_steps_per_second": 0.899,
      "eval_translation_length": 2056,
      "step": 1650
    },
    {
      "epoch": 0.8,
      "learning_rate": 7.335840777307632e-05,
      "loss": 1.0913,
      "step": 1700
    },
    {
      "epoch": 0.8,
      "eval_bleu": 0.572534516070901,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.486328125,
      "eval_gen_len": 7.140625,
      "eval_length_ratio": 1.0332383665717,
      "eval_loss": 0.7539807558059692,
      "eval_reference_length": 2106,
      "eval_runtime": 36.2066,
      "eval_samples_per_second": 14.141,
      "eval_steps_per_second": 0.884,
      "eval_translation_length": 2176,
      "step": 1700
    },
    {
      "epoch": 0.82,
      "learning_rate": 7.257483153110798e-05,
      "loss": 1.0432,
      "step": 1750
    },
    {
      "epoch": 0.82,
      "eval_bleu": 0.5471774953144333,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.494140625,
      "eval_gen_len": 7.560546875,
      "eval_length_ratio": 1.1054131054131053,
      "eval_loss": 0.7697689533233643,
      "eval_reference_length": 2106,
      "eval_runtime": 36.5988,
      "eval_samples_per_second": 13.99,
      "eval_steps_per_second": 0.874,
      "eval_translation_length": 2328,
      "step": 1750
    },
    {
      "epoch": 0.85,
      "learning_rate": 7.179125528913963e-05,
      "loss": 1.0081,
      "step": 1800
    },
    {
      "epoch": 0.85,
      "eval_bleu": 0.5805134856584743,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.4921875,
      "eval_gen_len": 7.1171875,
      "eval_length_ratio": 1.0270655270655271,
      "eval_loss": 0.7475109696388245,
      "eval_reference_length": 2106,
      "eval_runtime": 35.6381,
      "eval_samples_per_second": 14.367,
      "eval_steps_per_second": 0.898,
      "eval_translation_length": 2163,
      "step": 1800
    },
    {
      "epoch": 0.87,
      "learning_rate": 7.10076790471713e-05,
      "loss": 1.0837,
      "step": 1850
    },
    {
      "epoch": 0.87,
      "eval_bleu": 0.565305888773277,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.498046875,
      "eval_gen_len": 7.259765625,
      "eval_length_ratio": 1.0517568850902184,
      "eval_loss": 0.7250198125839233,
      "eval_reference_length": 2106,
      "eval_runtime": 35.8261,
      "eval_samples_per_second": 14.291,
      "eval_steps_per_second": 0.893,
      "eval_translation_length": 2215,
      "step": 1850
    },
    {
      "epoch": 0.89,
      "learning_rate": 7.022410280520295e-05,
      "loss": 1.0115,
      "step": 1900
    },
    {
      "epoch": 0.89,
      "eval_bleu": 0.6107745307911603,
      "eval_brevity_penalty": 0.997146934891016,
      "eval_exact_match": 0.51171875,
      "eval_gen_len": 6.921875,
      "eval_length_ratio": 0.9971509971509972,
      "eval_loss": 0.7463927268981934,
      "eval_reference_length": 2106,
      "eval_runtime": 34.3792,
      "eval_samples_per_second": 14.893,
      "eval_steps_per_second": 0.931,
      "eval_translation_length": 2100,
      "step": 1900
    },
    {
      "epoch": 0.92,
      "learning_rate": 6.944052656323461e-05,
      "loss": 0.9624,
      "step": 1950
    },
    {
      "epoch": 0.92,
      "eval_bleu": 0.6074013490021136,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.515625,
      "eval_gen_len": 6.962890625,
      "eval_length_ratio": 1.002849002849003,
      "eval_loss": 0.7404871582984924,
      "eval_reference_length": 2106,
      "eval_runtime": 35.4273,
      "eval_samples_per_second": 14.452,
      "eval_steps_per_second": 0.903,
      "eval_translation_length": 2112,
      "step": 1950
    },
    {
      "epoch": 0.94,
      "learning_rate": 6.865695032126626e-05,
      "loss": 0.8905,
      "step": 2000
    },
    {
      "epoch": 0.94,
      "eval_bleu": 0.605227378439144,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.51171875,
      "eval_gen_len": 6.986328125,
      "eval_length_ratio": 1.0080721747388415,
      "eval_loss": 0.7403624653816223,
      "eval_reference_length": 2106,
      "eval_runtime": 35.8318,
      "eval_samples_per_second": 14.289,
      "eval_steps_per_second": 0.893,
      "eval_translation_length": 2123,
      "step": 2000
    },
    {
      "epoch": 0.96,
      "learning_rate": 6.787337407929793e-05,
      "loss": 1.0936,
      "step": 2050
    },
    {
      "epoch": 0.96,
      "eval_bleu": 0.613080912893354,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.52734375,
      "eval_gen_len": 7.005859375,
      "eval_length_ratio": 1.0061728395061729,
      "eval_loss": 0.7132320404052734,
      "eval_reference_length": 2106,
      "eval_runtime": 35.4858,
      "eval_samples_per_second": 14.428,
      "eval_steps_per_second": 0.902,
      "eval_translation_length": 2119,
      "step": 2050
    },
    {
      "epoch": 0.99,
      "learning_rate": 6.708979783732958e-05,
      "loss": 1.0418,
      "step": 2100
    },
    {
      "epoch": 0.99,
      "eval_bleu": 0.6157679585094484,
      "eval_brevity_penalty": 0.9909373387239356,
      "eval_exact_match": 0.5234375,
      "eval_gen_len": 6.91796875,
      "eval_length_ratio": 0.9909781576448243,
      "eval_loss": 0.7031872868537903,
      "eval_reference_length": 2106,
      "eval_runtime": 35.4742,
      "eval_samples_per_second": 14.433,
      "eval_steps_per_second": 0.902,
      "eval_translation_length": 2087,
      "step": 2100
    },
    {
      "epoch": 1.01,
      "learning_rate": 6.630622159536123e-05,
      "loss": 0.9631,
      "step": 2150
    },
    {
      "epoch": 1.01,
      "eval_bleu": 0.6228792131756815,
      "eval_brevity_penalty": 0.946880981878215,
      "eval_exact_match": 0.525390625,
      "eval_gen_len": 6.642578125,
      "eval_length_ratio": 0.9482431149097815,
      "eval_loss": 0.695751428604126,
      "eval_reference_length": 2106,
      "eval_runtime": 34.7409,
      "eval_samples_per_second": 14.738,
      "eval_steps_per_second": 0.921,
      "eval_translation_length": 1997,
      "step": 2150
    },
    {
      "epoch": 1.03,
      "learning_rate": 6.552264535339289e-05,
      "loss": 0.8085,
      "step": 2200
    },
    {
      "epoch": 1.03,
      "eval_bleu": 0.6272859670438183,
      "eval_brevity_penalty": 0.946880981878215,
      "eval_exact_match": 0.53125,
      "eval_gen_len": 6.650390625,
      "eval_length_ratio": 0.9482431149097815,
      "eval_loss": 0.7051365375518799,
      "eval_reference_length": 2106,
      "eval_runtime": 34.9034,
      "eval_samples_per_second": 14.669,
      "eval_steps_per_second": 0.917,
      "eval_translation_length": 1997,
      "step": 2200
    },
    {
      "epoch": 1.06,
      "learning_rate": 6.473906911142454e-05,
      "loss": 0.9716,
      "step": 2250
    },
    {
      "epoch": 1.06,
      "eval_bleu": 0.6325687827118958,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.533203125,
      "eval_gen_len": 6.943359375,
      "eval_length_ratio": 1.000474833808167,
      "eval_loss": 0.6994906067848206,
      "eval_reference_length": 2106,
      "eval_runtime": 35.1655,
      "eval_samples_per_second": 14.56,
      "eval_steps_per_second": 0.91,
      "eval_translation_length": 2107,
      "step": 2250
    },
    {
      "epoch": 1.08,
      "learning_rate": 6.39554928694562e-05,
      "loss": 0.8824,
      "step": 2300
    },
    {
      "epoch": 1.08,
      "eval_bleu": 0.6202715514980656,
      "eval_brevity_penalty": 0.9691442418767943,
      "eval_exact_match": 0.5234375,
      "eval_gen_len": 6.76953125,
      "eval_length_ratio": 0.9696106362773029,
      "eval_loss": 0.7111188173294067,
      "eval_reference_length": 2106,
      "eval_runtime": 35.1592,
      "eval_samples_per_second": 14.562,
      "eval_steps_per_second": 0.91,
      "eval_translation_length": 2042,
      "step": 2300
    },
    {
      "epoch": 1.1,
      "learning_rate": 6.317191662748786e-05,
      "loss": 0.7888,
      "step": 2350
    },
    {
      "epoch": 1.1,
      "eval_bleu": 0.6314970589714921,
      "eval_brevity_penalty": 0.9423710963415396,
      "eval_exact_match": 0.533203125,
      "eval_gen_len": 6.609375,
      "eval_length_ratio": 0.9439696106362773,
      "eval_loss": 0.6993675827980042,
      "eval_reference_length": 2106,
      "eval_runtime": 34.3211,
      "eval_samples_per_second": 14.918,
      "eval_steps_per_second": 0.932,
      "eval_translation_length": 1988,
      "step": 2350
    },
    {
      "epoch": 1.13,
      "learning_rate": 6.238834038551952e-05,
      "loss": 0.8409,
      "step": 2400
    },
    {
      "epoch": 1.13,
      "eval_bleu": 0.6283952758196866,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.54296875,
      "eval_gen_len": 6.984375,
      "eval_length_ratio": 1.0151946818613484,
      "eval_loss": 0.7133020162582397,
      "eval_reference_length": 2106,
      "eval_runtime": 35.2074,
      "eval_samples_per_second": 14.542,
      "eval_steps_per_second": 0.909,
      "eval_translation_length": 2138,
      "step": 2400
    },
    {
      "epoch": 1.15,
      "learning_rate": 6.160476414355117e-05,
      "loss": 0.8331,
      "step": 2450
    },
    {
      "epoch": 1.15,
      "eval_bleu": 0.6355611685013638,
      "eval_brevity_penalty": 0.9725649218577119,
      "eval_exact_match": 0.541015625,
      "eval_gen_len": 6.76953125,
      "eval_length_ratio": 0.9729344729344729,
      "eval_loss": 0.7177965044975281,
      "eval_reference_length": 2106,
      "eval_runtime": 35.1779,
      "eval_samples_per_second": 14.555,
      "eval_steps_per_second": 0.91,
      "eval_translation_length": 2049,
      "step": 2450
    },
    {
      "epoch": 1.18,
      "learning_rate": 6.0821187901582835e-05,
      "loss": 0.8205,
      "step": 2500
    },
    {
      "epoch": 1.18,
      "eval_bleu": 0.6392355323482277,
      "eval_brevity_penalty": 0.9740274534203012,
      "eval_exact_match": 0.53515625,
      "eval_gen_len": 6.80078125,
      "eval_length_ratio": 0.9743589743589743,
      "eval_loss": 0.7102757096290588,
      "eval_reference_length": 2106,
      "eval_runtime": 35.5038,
      "eval_samples_per_second": 14.421,
      "eval_steps_per_second": 0.901,
      "eval_translation_length": 2052,
      "step": 2500
    },
    {
      "epoch": 1.2,
      "learning_rate": 6.003761165961448e-05,
      "loss": 0.8214,
      "step": 2550
    },
    {
      "epoch": 1.2,
      "eval_bleu": 0.6337232384678544,
      "eval_brevity_penalty": 0.9711003070476397,
      "eval_exact_match": 0.544921875,
      "eval_gen_len": 6.759765625,
      "eval_length_ratio": 0.9715099715099715,
      "eval_loss": 0.7109658718109131,
      "eval_reference_length": 2106,
      "eval_runtime": 35.4746,
      "eval_samples_per_second": 14.433,
      "eval_steps_per_second": 0.902,
      "eval_translation_length": 2046,
      "step": 2550
    },
    {
      "epoch": 1.22,
      "learning_rate": 5.925403541764614e-05,
      "loss": 0.8607,
      "step": 2600
    },
    {
      "epoch": 1.22,
      "eval_bleu": 0.6413697910504058,
      "eval_brevity_penalty": 0.951371692779408,
      "eval_exact_match": 0.54296875,
      "eval_gen_len": 6.650390625,
      "eval_length_ratio": 0.9525166191832859,
      "eval_loss": 0.6937527656555176,
      "eval_reference_length": 2106,
      "eval_runtime": 35.0711,
      "eval_samples_per_second": 14.599,
      "eval_steps_per_second": 0.912,
      "eval_translation_length": 2006,
      "step": 2600
    },
    {
      "epoch": 1.25,
      "learning_rate": 5.8470459175677794e-05,
      "loss": 0.8754,
      "step": 2650
    },
    {
      "epoch": 1.25,
      "eval_bleu": 0.6345724305414163,
      "eval_brevity_penalty": 0.9880584183776264,
      "eval_exact_match": 0.521484375,
      "eval_gen_len": 6.88671875,
      "eval_length_ratio": 0.9881291547958214,
      "eval_loss": 0.7027400732040405,
      "eval_reference_length": 2106,
      "eval_runtime": 35.2724,
      "eval_samples_per_second": 14.516,
      "eval_steps_per_second": 0.907,
      "eval_translation_length": 2081,
      "step": 2650
    },
    {
      "epoch": 1.27,
      "learning_rate": 5.768688293370945e-05,
      "loss": 0.8625,
      "step": 2700
    },
    {
      "epoch": 1.27,
      "eval_bleu": 0.641086474299546,
      "eval_brevity_penalty": 0.9418688112227576,
      "eval_exact_match": 0.529296875,
      "eval_gen_len": 6.623046875,
      "eval_length_ratio": 0.9434947768281101,
      "eval_loss": 0.691420316696167,
      "eval_reference_length": 2106,
      "eval_runtime": 34.6599,
      "eval_samples_per_second": 14.772,
      "eval_steps_per_second": 0.923,
      "eval_translation_length": 1987,
      "step": 2700
    },
    {
      "epoch": 1.29,
      "learning_rate": 5.690330669174111e-05,
      "loss": 0.8625,
      "step": 2750
    },
    {
      "epoch": 1.29,
      "eval_bleu": 0.6428983658861092,
      "eval_brevity_penalty": 0.9543548830374949,
      "eval_exact_match": 0.537109375,
      "eval_gen_len": 6.693359375,
      "eval_length_ratio": 0.9553656220322887,
      "eval_loss": 0.7152715921401978,
      "eval_reference_length": 2106,
      "eval_runtime": 34.5929,
      "eval_samples_per_second": 14.801,
      "eval_steps_per_second": 0.925,
      "eval_translation_length": 2012,
      "step": 2750
    },
    {
      "epoch": 1.32,
      "learning_rate": 5.6119730449772766e-05,
      "loss": 0.8007,
      "step": 2800
    },
    {
      "epoch": 1.32,
      "eval_bleu": 0.6484293170697683,
      "eval_brevity_penalty": 0.9528643478208072,
      "eval_exact_match": 0.5546875,
      "eval_gen_len": 6.67578125,
      "eval_length_ratio": 0.9539411206077872,
      "eval_loss": 0.691640317440033,
      "eval_reference_length": 2106,
      "eval_runtime": 34.5133,
      "eval_samples_per_second": 14.835,
      "eval_steps_per_second": 0.927,
      "eval_translation_length": 2009,
      "step": 2800
    },
    {
      "epoch": 1.34,
      "learning_rate": 5.5336154207804424e-05,
      "loss": 0.8345,
      "step": 2850
    },
    {
      "epoch": 1.34,
      "eval_bleu": 0.6441657194783443,
      "eval_brevity_penalty": 0.9418688112227576,
      "eval_exact_match": 0.54296875,
      "eval_gen_len": 6.638671875,
      "eval_length_ratio": 0.9434947768281101,
      "eval_loss": 0.7108384370803833,
      "eval_reference_length": 2106,
      "eval_runtime": 34.9885,
      "eval_samples_per_second": 14.633,
      "eval_steps_per_second": 0.915,
      "eval_translation_length": 1987,
      "step": 2850
    },
    {
      "epoch": 1.36,
      "learning_rate": 5.455257796583608e-05,
      "loss": 0.8096,
      "step": 2900
    },
    {
      "epoch": 1.36,
      "eval_bleu": 0.628018545615884,
      "eval_brevity_penalty": 0.9938080693803775,
      "eval_exact_match": 0.537109375,
      "eval_gen_len": 6.912109375,
      "eval_length_ratio": 0.9938271604938271,
      "eval_loss": 0.7028455138206482,
      "eval_reference_length": 2106,
      "eval_runtime": 35.5719,
      "eval_samples_per_second": 14.393,
      "eval_steps_per_second": 0.9,
      "eval_translation_length": 2093,
      "step": 2900
    },
    {
      "epoch": 1.39,
      "learning_rate": 5.376900172386774e-05,
      "loss": 0.7349,
      "step": 2950
    },
    {
      "epoch": 1.39,
      "eval_bleu": 0.6373148854351528,
      "eval_brevity_penalty": 0.9822759172643585,
      "eval_exact_match": 0.541015625,
      "eval_gen_len": 6.853515625,
      "eval_length_ratio": 0.9824311490978158,
      "eval_loss": 0.7058258056640625,
      "eval_reference_length": 2106,
      "eval_runtime": 35.2471,
      "eval_samples_per_second": 14.526,
      "eval_steps_per_second": 0.908,
      "eval_translation_length": 2069,
      "step": 2950
    },
    {
      "epoch": 1.41,
      "learning_rate": 5.298542548189939e-05,
      "loss": 0.9361,
      "step": 3000
    },
    {
      "epoch": 1.41,
      "eval_bleu": 0.6287603417406736,
      "eval_brevity_penalty": 0.9923737263316806,
      "eval_exact_match": 0.53515625,
      "eval_gen_len": 6.89453125,
      "eval_length_ratio": 0.9924026590693258,
      "eval_loss": 0.6922191381454468,
      "eval_reference_length": 2106,
      "eval_runtime": 34.6615,
      "eval_samples_per_second": 14.771,
      "eval_steps_per_second": 0.923,
      "eval_translation_length": 2090,
      "step": 3000
    },
    {
      "epoch": 1.43,
      "learning_rate": 5.220184923993105e-05,
      "loss": 0.8269,
      "step": 3050
    },
    {
      "epoch": 1.43,
      "eval_bleu": 0.6439396809407835,
      "eval_brevity_penalty": 0.9642378110498295,
      "eval_exact_match": 0.548828125,
      "eval_gen_len": 6.740234375,
      "eval_length_ratio": 0.9648622981956315,
      "eval_loss": 0.6778197288513184,
      "eval_reference_length": 2106,
      "eval_runtime": 34.7891,
      "eval_samples_per_second": 14.717,
      "eval_steps_per_second": 0.92,
      "eval_translation_length": 2032,
      "step": 3050
    },
    {
      "epoch": 1.46,
      "learning_rate": 5.1418272997962704e-05,
      "loss": 0.7282,
      "step": 3100
    },
    {
      "epoch": 1.46,
      "eval_bleu": 0.642699995934985,
      "eval_brevity_penalty": 0.9904580880129722,
      "eval_exact_match": 0.55078125,
      "eval_gen_len": 6.9140625,
      "eval_length_ratio": 0.9905033238366572,
      "eval_loss": 0.685681164264679,
      "eval_reference_length": 2106,
      "eval_runtime": 35.0377,
      "eval_samples_per_second": 14.613,
      "eval_steps_per_second": 0.913,
      "eval_translation_length": 2086,
      "step": 3100
    },
    {
      "epoch": 1.48,
      "learning_rate": 5.063469675599436e-05,
      "loss": 0.7244,
      "step": 3150
    },
    {
      "epoch": 1.48,
      "eval_bleu": 0.6412084160600809,
      "eval_brevity_penalty": 0.9832419564425784,
      "eval_exact_match": 0.541015625,
      "eval_gen_len": 6.880859375,
      "eval_length_ratio": 0.98338081671415,
      "eval_loss": 0.6973811388015747,
      "eval_reference_length": 2106,
      "eval_runtime": 35.1389,
      "eval_samples_per_second": 14.571,
      "eval_steps_per_second": 0.911,
      "eval_translation_length": 2071,
      "step": 3150
    },
    {
      "epoch": 1.5,
      "learning_rate": 4.985112051402602e-05,
      "loss": 0.7876,
      "step": 3200
    },
    {
      "epoch": 1.5,
      "eval_bleu": 0.6514314556575306,
      "eval_brevity_penalty": 0.9947631636301487,
      "eval_exact_match": 0.544921875,
      "eval_gen_len": 6.9453125,
      "eval_length_ratio": 0.9947768281101614,
      "eval_loss": 0.6855226755142212,
      "eval_reference_length": 2106,
      "eval_runtime": 35.5075,
      "eval_samples_per_second": 14.42,
      "eval_steps_per_second": 0.901,
      "eval_translation_length": 2095,
      "step": 3200
    },
    {
      "epoch": 1.53,
      "learning_rate": 4.9067544272057677e-05,
      "loss": 0.8535,
      "step": 3250
    },
    {
      "epoch": 1.53,
      "eval_bleu": 0.6500634066257985,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.541015625,
      "eval_gen_len": 6.986328125,
      "eval_length_ratio": 1.0037986704653372,
      "eval_loss": 0.6744265556335449,
      "eval_reference_length": 2106,
      "eval_runtime": 36.0223,
      "eval_samples_per_second": 14.213,
      "eval_steps_per_second": 0.888,
      "eval_translation_length": 2114,
      "step": 3250
    },
    {
      "epoch": 1.55,
      "learning_rate": 4.8283968030089334e-05,
      "loss": 0.7917,
      "step": 3300
    },
    {
      "epoch": 1.55,
      "eval_bleu": 0.6499057150972741,
      "eval_brevity_penalty": 0.9985744834917494,
      "eval_exact_match": 0.533203125,
      "eval_gen_len": 6.986328125,
      "eval_length_ratio": 0.9985754985754985,
      "eval_loss": 0.672450065612793,
      "eval_reference_length": 2106,
      "eval_runtime": 35.9664,
      "eval_samples_per_second": 14.236,
      "eval_steps_per_second": 0.89,
      "eval_translation_length": 2103,
      "step": 3300
    },
    {
      "epoch": 1.57,
      "learning_rate": 4.7500391788120985e-05,
      "loss": 0.7538,
      "step": 3350
    },
    {
      "epoch": 1.57,
      "eval_bleu": 0.6310972587707046,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.5390625,
      "eval_gen_len": 7.095703125,
      "eval_length_ratio": 1.019468186134853,
      "eval_loss": 0.6634331941604614,
      "eval_reference_length": 2106,
      "eval_runtime": 35.9969,
      "eval_samples_per_second": 14.223,
      "eval_steps_per_second": 0.889,
      "eval_translation_length": 2147,
      "step": 3350
    },
    {
      "epoch": 1.6,
      "learning_rate": 4.671681554615264e-05,
      "loss": 0.8731,
      "step": 3400
    },
    {
      "epoch": 1.6,
      "eval_bleu": 0.6365367381748843,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.544921875,
      "eval_gen_len": 7.03125,
      "eval_length_ratio": 1.0113960113960114,
      "eval_loss": 0.695984959602356,
      "eval_reference_length": 2106,
      "eval_runtime": 35.7076,
      "eval_samples_per_second": 14.339,
      "eval_steps_per_second": 0.896,
      "eval_translation_length": 2130,
      "step": 3400
    },
    {
      "epoch": 1.62,
      "learning_rate": 4.59332393041843e-05,
      "loss": 0.749,
      "step": 3450
    },
    {
      "epoch": 1.62,
      "eval_bleu": 0.6227209715309775,
      "eval_brevity_penalty": 0.944377861736799,
      "eval_exact_match": 0.5390625,
      "eval_gen_len": 6.71484375,
      "eval_length_ratio": 0.9458689458689459,
      "eval_loss": 0.6684394478797913,
      "eval_reference_length": 2106,
      "eval_runtime": 35.0788,
      "eval_samples_per_second": 14.596,
      "eval_steps_per_second": 0.912,
      "eval_translation_length": 1992,
      "step": 3450
    },
    {
      "epoch": 1.65,
      "learning_rate": 4.514966306221596e-05,
      "loss": 0.8981,
      "step": 3500
    },
    {
      "epoch": 1.65,
      "eval_bleu": 0.6248917016218156,
      "eval_brevity_penalty": 0.997146934891016,
      "eval_exact_match": 0.5390625,
      "eval_gen_len": 6.947265625,
      "eval_length_ratio": 0.9971509971509972,
      "eval_loss": 0.6610915660858154,
      "eval_reference_length": 2106,
      "eval_runtime": 35.6448,
      "eval_samples_per_second": 14.364,
      "eval_steps_per_second": 0.898,
      "eval_translation_length": 2100,
      "step": 3500
    },
    {
      "epoch": 1.67,
      "learning_rate": 4.436608682024761e-05,
      "loss": 0.6892,
      "step": 3550
    },
    {
      "epoch": 1.67,
      "eval_bleu": 0.6171345792845108,
      "eval_brevity_penalty": 0.9827590514957918,
      "eval_exact_match": 0.5390625,
      "eval_gen_len": 6.861328125,
      "eval_length_ratio": 0.9829059829059829,
      "eval_loss": 0.6674365997314453,
      "eval_reference_length": 2106,
      "eval_runtime": 35.383,
      "eval_samples_per_second": 14.47,
      "eval_steps_per_second": 0.904,
      "eval_translation_length": 2070,
      "step": 3550
    },
    {
      "epoch": 1.69,
      "learning_rate": 4.3582510578279265e-05,
      "loss": 0.721,
      "step": 3600
    },
    {
      "epoch": 1.69,
      "eval_bleu": 0.6119684034182921,
      "eval_brevity_penalty": 0.9784025765941619,
      "eval_exact_match": 0.525390625,
      "eval_gen_len": 6.888671875,
      "eval_length_ratio": 0.9786324786324786,
      "eval_loss": 0.6938477754592896,
      "eval_reference_length": 2106,
      "eval_runtime": 35.4713,
      "eval_samples_per_second": 14.434,
      "eval_steps_per_second": 0.902,
      "eval_translation_length": 2061,
      "step": 3600
    },
    {
      "epoch": 1.72,
      "learning_rate": 4.279893433631092e-05,
      "loss": 0.8326,
      "step": 3650
    },
    {
      "epoch": 1.72,
      "eval_bleu": 0.6445182499922312,
      "eval_brevity_penalty": 0.9942857298792932,
      "eval_exact_match": 0.55078125,
      "eval_gen_len": 6.951171875,
      "eval_length_ratio": 0.9943019943019943,
      "eval_loss": 0.6575934886932373,
      "eval_reference_length": 2106,
      "eval_runtime": 35.1284,
      "eval_samples_per_second": 14.575,
      "eval_steps_per_second": 0.911,
      "eval_translation_length": 2094,
      "step": 3650
    },
    {
      "epoch": 1.74,
      "learning_rate": 4.201535809434258e-05,
      "loss": 0.7976,
      "step": 3700
    },
    {
      "epoch": 1.74,
      "eval_bleu": 0.6440453785850522,
      "eval_brevity_penalty": 0.9508736695620825,
      "eval_exact_match": 0.560546875,
      "eval_gen_len": 6.767578125,
      "eval_length_ratio": 0.9520417853751187,
      "eval_loss": 0.6605907678604126,
      "eval_reference_length": 2106,
      "eval_runtime": 35.1636,
      "eval_samples_per_second": 14.561,
      "eval_steps_per_second": 0.91,
      "eval_translation_length": 2005,
      "step": 3700
    },
    {
      "epoch": 1.76,
      "learning_rate": 4.123178185237424e-05,
      "loss": 0.7771,
      "step": 3750
    },
    {
      "epoch": 1.76,
      "eval_bleu": 0.644795546998577,
      "eval_brevity_penalty": 0.9523670318092967,
      "eval_exact_match": 0.560546875,
      "eval_gen_len": 6.724609375,
      "eval_length_ratio": 0.9534662867996201,
      "eval_loss": 0.6605848073959351,
      "eval_reference_length": 2106,
      "eval_runtime": 34.7093,
      "eval_samples_per_second": 14.751,
      "eval_steps_per_second": 0.922,
      "eval_translation_length": 2008,
      "step": 3750
    },
    {
      "epoch": 1.79,
      "learning_rate": 4.0448205610405895e-05,
      "loss": 0.856,
      "step": 3800
    },
    {
      "epoch": 1.79,
      "eval_bleu": 0.6464921288506784,
      "eval_brevity_penalty": 0.9817925536433939,
      "eval_exact_match": 0.556640625,
      "eval_gen_len": 6.865234375,
      "eval_length_ratio": 0.9819563152896487,
      "eval_loss": 0.6571842432022095,
      "eval_reference_length": 2106,
      "eval_runtime": 36.0118,
      "eval_samples_per_second": 14.218,
      "eval_steps_per_second": 0.889,
      "eval_translation_length": 2068,
      "step": 3800
    },
    {
      "epoch": 1.81,
      "learning_rate": 3.966462936843755e-05,
      "loss": 0.7667,
      "step": 3850
    },
    {
      "epoch": 1.81,
      "eval_bleu": 0.6443589957503941,
      "eval_brevity_penalty": 0.97694627798514,
      "eval_exact_match": 0.55859375,
      "eval_gen_len": 6.857421875,
      "eval_length_ratio": 0.9772079772079773,
      "eval_loss": 0.6501011252403259,
      "eval_reference_length": 2106,
      "eval_runtime": 36.8454,
      "eval_samples_per_second": 13.896,
      "eval_steps_per_second": 0.868,
      "eval_translation_length": 2058,
      "step": 3850
    },
    {
      "epoch": 1.83,
      "learning_rate": 3.888105312646921e-05,
      "loss": 0.7609,
      "step": 3900
    },
    {
      "epoch": 1.83,
      "eval_bleu": 0.6357951146509494,
      "eval_brevity_penalty": 0.9754879045293495,
      "eval_exact_match": 0.560546875,
      "eval_gen_len": 6.853515625,
      "eval_length_ratio": 0.9757834757834758,
      "eval_loss": 0.67498779296875,
      "eval_reference_length": 2106,
      "eval_runtime": 36.1747,
      "eval_samples_per_second": 14.154,
      "eval_steps_per_second": 0.885,
      "eval_translation_length": 2055,
      "step": 3900
    },
    {
      "epoch": 1.86,
      "learning_rate": 3.809747688450087e-05,
      "loss": 0.871,
      "step": 3950
    },
    {
      "epoch": 1.86,
      "eval_bleu": 0.6283900991765164,
      "eval_brevity_penalty": 0.9478805721747763,
      "eval_exact_match": 0.556640625,
      "eval_gen_len": 6.693359375,
      "eval_length_ratio": 0.9491927825261158,
      "eval_loss": 0.6474365592002869,
      "eval_reference_length": 2106,
      "eval_runtime": 35.4701,
      "eval_samples_per_second": 14.435,
      "eval_steps_per_second": 0.902,
      "eval_translation_length": 1999,
      "step": 3950
    },
    {
      "epoch": 1.88,
      "learning_rate": 3.731390064253252e-05,
      "loss": 0.8432,
      "step": 4000
    },
    {
      "epoch": 1.88,
      "eval_bleu": 0.6243444944082281,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.55859375,
      "eval_gen_len": 7.046875,
      "eval_length_ratio": 1.0147198480531814,
      "eval_loss": 0.6450670957565308,
      "eval_reference_length": 2106,
      "eval_runtime": 36.3555,
      "eval_samples_per_second": 14.083,
      "eval_steps_per_second": 0.88,
      "eval_translation_length": 2137,
      "step": 4000
    },
    {
      "epoch": 1.9,
      "learning_rate": 3.6530324400564175e-05,
      "loss": 0.7924,
      "step": 4050
    },
    {
      "epoch": 1.9,
      "eval_bleu": 0.6406129867109561,
      "eval_brevity_penalty": 1.0,
      "eval_exact_match": 0.55078125,
      "eval_gen_len": 6.98828125,
      "eval_length_ratio": 1.000474833808167,
      "eval_loss": 0.6552107930183411,
      "eval_reference_length": 2106,
      "eval_runtime": 35.9782,
      "eval_samples_per_second": 14.231,
      "eval_steps_per_second": 0.889,
      "eval_translation_length": 2107,
      "step": 4050
    },
    {
      "epoch": 1.93,
      "learning_rate": 3.574674815859583e-05,
      "loss": 0.8058,
      "step": 4100
    },
    {
      "epoch": 1.93,
      "eval_bleu": 0.6521608841609852,
      "eval_brevity_penalty": 0.9745145014799987,
      "eval_exact_match": 0.5546875,
      "eval_gen_len": 6.8515625,
      "eval_length_ratio": 0.9748338081671415,
      "eval_loss": 0.6530161499977112,
      "eval_reference_length": 2106,
      "eval_runtime": 36.0535,
      "eval_samples_per_second": 14.201,
      "eval_steps_per_second": 0.888,
      "eval_translation_length": 2053,
      "step": 4100
    },
    {
      "epoch": 1.95,
      "learning_rate": 3.496317191662749e-05,
      "loss": 0.7426,
      "step": 4150
    },
    {
      "epoch": 1.95,
      "eval_bleu": 0.6517932308118449,
      "eval_brevity_penalty": 0.9837246322096701,
      "eval_exact_match": 0.552734375,
      "eval_gen_len": 6.90625,
      "eval_length_ratio": 0.9838556505223172,
      "eval_loss": 0.6467949151992798,
      "eval_reference_length": 2106,
      "eval_runtime": 36.1369,
      "eval_samples_per_second": 14.168,
      "eval_steps_per_second": 0.886,
      "eval_translation_length": 2072,
      "step": 4150
    },
    {
      "epoch": 1.97,
      "learning_rate": 3.417959567465915e-05,
      "loss": 0.806,
      "step": 4200
    },
    {
      "epoch": 1.97,
      "eval_bleu": 0.6519640951085052,
      "eval_brevity_penalty": 0.9740274534203012,
      "eval_exact_match": 0.564453125,
      "eval_gen_len": 6.8515625,
      "eval_length_ratio": 0.9743589743589743,
      "eval_loss": 0.6484197378158569,
      "eval_reference_length": 2106,
      "eval_runtime": 35.8891,
      "eval_samples_per_second": 14.266,
      "eval_steps_per_second": 0.892,
      "eval_translation_length": 2052,
      "step": 4200
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.33960194326908e-05,
      "loss": 0.8135,
      "step": 4250
    },
    {
      "epoch": 2.0,
      "eval_bleu": 0.6506610495036242,
      "eval_brevity_penalty": 0.97694627798514,
      "eval_exact_match": 0.564453125,
      "eval_gen_len": 6.859375,
      "eval_length_ratio": 0.9772079772079773,
      "eval_loss": 0.6449689865112305,
      "eval_reference_length": 2106,
      "eval_runtime": 35.381,
      "eval_samples_per_second": 14.471,
      "eval_steps_per_second": 0.904,
      "eval_translation_length": 2058,
      "step": 4250
    }
  ],
  "logging_steps": 50,
  "max_steps": 6381,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 2.4719244569985024e+16,
  "trial_name": null,
  "trial_params": null
}