{
  "best_metric": 0.8553540706634521,
  "best_model_checkpoint": "/data/kabanda/result_en_fr/checkpoint-70000",
  "epoch": 3.464660463274599,
  "global_step": 70000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 4.9876262126311625e-05,
      "loss": 5.2926,
      "step": 500
    },
    {
      "epoch": 0.02,
      "eval_bleu": 4.1544,
      "eval_gen_len": 19.7843,
      "eval_loss": 4.638627529144287,
      "eval_runtime": 1532.3805,
      "eval_samples_per_second": 4.661,
      "eval_steps_per_second": 0.389,
      "step": 500
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.975252425262325e-05,
      "loss": 4.2117,
      "step": 1000
    },
    {
      "epoch": 0.05,
      "eval_bleu": 5.0944,
      "eval_gen_len": 25.8076,
      "eval_loss": 3.982415199279785,
      "eval_runtime": 2306.4168,
      "eval_samples_per_second": 3.097,
      "eval_steps_per_second": 0.258,
      "step": 1000
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.962878637893486e-05,
      "loss": 3.6754,
      "step": 1500
    },
    {
      "epoch": 0.07,
      "eval_bleu": 9.7913,
      "eval_gen_len": 18.5118,
      "eval_loss": 3.528294801712036,
      "eval_runtime": 1473.4816,
      "eval_samples_per_second": 4.848,
      "eval_steps_per_second": 0.404,
      "step": 1500
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.9505048505246485e-05,
      "loss": 3.2658,
      "step": 2000
    },
    {
      "epoch": 0.1,
      "eval_bleu": 11.9905,
      "eval_gen_len": 18.3529,
      "eval_loss": 3.1733100414276123,
      "eval_runtime": 1495.811,
      "eval_samples_per_second": 4.775,
      "eval_steps_per_second": 0.398,
      "step": 2000
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.938131063155811e-05,
      "loss": 2.9666,
      "step": 2500
    },
    {
      "epoch": 0.12,
      "eval_bleu": 14.6761,
      "eval_gen_len": 17.3686,
      "eval_loss": 2.8982934951782227,
      "eval_runtime": 1369.1732,
      "eval_samples_per_second": 5.217,
      "eval_steps_per_second": 0.435,
      "step": 2500
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.925757275786973e-05,
      "loss": 2.7349,
      "step": 3000
    },
    {
      "epoch": 0.15,
      "eval_bleu": 16.3199,
      "eval_gen_len": 17.9567,
      "eval_loss": 2.693359613418579,
      "eval_runtime": 1446.9239,
      "eval_samples_per_second": 4.937,
      "eval_steps_per_second": 0.412,
      "step": 3000
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.913383488418135e-05,
      "loss": 2.5024,
      "step": 3500
    },
    {
      "epoch": 0.17,
      "eval_bleu": 18.1705,
      "eval_gen_len": 18.0659,
      "eval_loss": 2.521655321121216,
      "eval_runtime": 1477.4963,
      "eval_samples_per_second": 4.835,
      "eval_steps_per_second": 0.403,
      "step": 3500
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.9010097010492975e-05,
      "loss": 2.3656,
      "step": 4000
    },
    {
      "epoch": 0.2,
      "eval_bleu": 19.9036,
      "eval_gen_len": 17.349,
      "eval_loss": 2.388259172439575,
      "eval_runtime": 1363.4598,
      "eval_samples_per_second": 5.239,
      "eval_steps_per_second": 0.437,
      "step": 4000
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.888635913680459e-05,
      "loss": 2.2392,
      "step": 4500
    },
    {
      "epoch": 0.22,
      "eval_bleu": 21.2018,
      "eval_gen_len": 17.771,
      "eval_loss": 2.2573838233947754,
      "eval_runtime": 1439.0867,
      "eval_samples_per_second": 4.964,
      "eval_steps_per_second": 0.414,
      "step": 4500
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.876262126311621e-05,
      "loss": 2.0961,
      "step": 5000
    },
    {
      "epoch": 0.25,
      "eval_bleu": 22.5424,
      "eval_gen_len": 17.0724,
      "eval_loss": 2.15091609954834,
      "eval_runtime": 1354.3029,
      "eval_samples_per_second": 5.274,
      "eval_steps_per_second": 0.44,
      "step": 5000
    },
    {
      "epoch": 0.27,
      "learning_rate": 4.8638883389427836e-05,
      "loss": 1.9984,
      "step": 5500
    },
    {
      "epoch": 0.27,
      "eval_bleu": 22.8122,
      "eval_gen_len": 18.099,
      "eval_loss": 2.0877768993377686,
      "eval_runtime": 1398.4571,
      "eval_samples_per_second": 5.108,
      "eval_steps_per_second": 0.426,
      "step": 5500
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.851514551573946e-05,
      "loss": 1.9071,
      "step": 6000
    },
    {
      "epoch": 0.3,
      "eval_bleu": 25.4086,
      "eval_gen_len": 17.1431,
      "eval_loss": 1.975221872329712,
      "eval_runtime": 1364.3455,
      "eval_samples_per_second": 5.235,
      "eval_steps_per_second": 0.437,
      "step": 6000
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.839140764205108e-05,
      "loss": 1.8419,
      "step": 6500
    },
    {
      "epoch": 0.32,
      "eval_bleu": 26.1858,
      "eval_gen_len": 17.5429,
      "eval_loss": 1.896142601966858,
      "eval_runtime": 1384.4266,
      "eval_samples_per_second": 5.16,
      "eval_steps_per_second": 0.431,
      "step": 6500
    },
    {
      "epoch": 0.35,
      "learning_rate": 4.82676697683627e-05,
      "loss": 1.7696,
      "step": 7000
    },
    {
      "epoch": 0.35,
      "eval_bleu": 26.7992,
      "eval_gen_len": 17.5017,
      "eval_loss": 1.835010051727295,
      "eval_runtime": 1380.8355,
      "eval_samples_per_second": 5.173,
      "eval_steps_per_second": 0.432,
      "step": 7000
    },
    {
      "epoch": 0.37,
      "learning_rate": 4.8143931894674325e-05,
      "loss": 1.709,
      "step": 7500
    },
    {
      "epoch": 0.37,
      "eval_bleu": 28.566,
      "eval_gen_len": 17.306,
      "eval_loss": 1.754841923713684,
      "eval_runtime": 1381.6803,
      "eval_samples_per_second": 5.17,
      "eval_steps_per_second": 0.431,
      "step": 7500
    },
    {
      "epoch": 0.4,
      "learning_rate": 4.802019402098595e-05,
      "loss": 1.6455,
      "step": 8000
    },
    {
      "epoch": 0.4,
      "eval_bleu": 28.9268,
      "eval_gen_len": 17.2583,
      "eval_loss": 1.7277344465255737,
      "eval_runtime": 1383.5791,
      "eval_samples_per_second": 5.163,
      "eval_steps_per_second": 0.431,
      "step": 8000
    },
    {
      "epoch": 0.42,
      "learning_rate": 4.789645614729757e-05,
      "loss": 1.5882,
      "step": 8500
    },
    {
      "epoch": 0.42,
      "eval_bleu": 29.6493,
      "eval_gen_len": 17.3461,
      "eval_loss": 1.6740621328353882,
      "eval_runtime": 1382.0901,
      "eval_samples_per_second": 5.168,
      "eval_steps_per_second": 0.431,
      "step": 8500
    },
    {
      "epoch": 0.45,
      "learning_rate": 4.777271827360919e-05,
      "loss": 1.5327,
      "step": 9000
    },
    {
      "epoch": 0.45,
      "eval_bleu": 30.4256,
      "eval_gen_len": 17.1523,
      "eval_loss": 1.6332658529281616,
      "eval_runtime": 1358.5029,
      "eval_samples_per_second": 5.258,
      "eval_steps_per_second": 0.439,
      "step": 9000
    },
    {
      "epoch": 0.47,
      "learning_rate": 4.7648980399920815e-05,
      "loss": 1.4922,
      "step": 9500
    },
    {
      "epoch": 0.47,
      "eval_bleu": 31.3386,
      "eval_gen_len": 17.1382,
      "eval_loss": 1.5828430652618408,
      "eval_runtime": 1370.5696,
      "eval_samples_per_second": 5.212,
      "eval_steps_per_second": 0.435,
      "step": 9500
    },
    {
      "epoch": 0.49,
      "learning_rate": 4.752524252623243e-05,
      "loss": 1.4468,
      "step": 10000
    },
    {
      "epoch": 0.49,
      "eval_bleu": 32.089,
      "eval_gen_len": 17.0442,
      "eval_loss": 1.5548750162124634,
      "eval_runtime": 1350.7323,
      "eval_samples_per_second": 5.288,
      "eval_steps_per_second": 0.441,
      "step": 10000
    },
    {
      "epoch": 0.52,
      "learning_rate": 4.740150465254405e-05,
      "loss": 1.4125,
      "step": 10500
    },
    {
      "epoch": 0.52,
      "eval_bleu": 32.4293,
      "eval_gen_len": 16.9427,
      "eval_loss": 1.52163827419281,
      "eval_runtime": 1355.6836,
      "eval_samples_per_second": 5.269,
      "eval_steps_per_second": 0.44,
      "step": 10500
    },
    {
      "epoch": 0.54,
      "learning_rate": 4.7277766778855676e-05,
      "loss": 1.411,
      "step": 11000
    },
    {
      "epoch": 0.54,
      "eval_bleu": 33.0258,
      "eval_gen_len": 16.7907,
      "eval_loss": 1.4887491464614868,
      "eval_runtime": 1353.9121,
      "eval_samples_per_second": 5.276,
      "eval_steps_per_second": 0.44,
      "step": 11000
    },
    {
      "epoch": 0.57,
      "learning_rate": 4.71540289051673e-05,
      "loss": 1.3452,
      "step": 11500
    },
    {
      "epoch": 0.57,
      "eval_bleu": 33.4866,
      "eval_gen_len": 16.9954,
      "eval_loss": 1.4632903337478638,
      "eval_runtime": 1372.2724,
      "eval_samples_per_second": 5.205,
      "eval_steps_per_second": 0.434,
      "step": 11500
    },
    {
      "epoch": 0.59,
      "learning_rate": 4.703029103147892e-05,
      "loss": 1.3275,
      "step": 12000
    },
    {
      "epoch": 0.59,
      "eval_bleu": 34.0579,
      "eval_gen_len": 16.8965,
      "eval_loss": 1.4243967533111572,
      "eval_runtime": 1354.7775,
      "eval_samples_per_second": 5.272,
      "eval_steps_per_second": 0.44,
      "step": 12000
    },
    {
      "epoch": 0.62,
      "learning_rate": 4.6906553157790536e-05,
      "loss": 1.3073,
      "step": 12500
    },
    {
      "epoch": 0.62,
      "eval_bleu": 34.2442,
      "eval_gen_len": 16.8726,
      "eval_loss": 1.405735969543457,
      "eval_runtime": 1344.8184,
      "eval_samples_per_second": 5.311,
      "eval_steps_per_second": 0.443,
      "step": 12500
    },
    {
      "epoch": 0.64,
      "learning_rate": 4.678281528410216e-05,
      "loss": 1.283,
      "step": 13000
    },
    {
      "epoch": 0.64,
      "eval_bleu": 35.3879,
      "eval_gen_len": 17.1285,
      "eval_loss": 1.3807507753372192,
      "eval_runtime": 1352.5706,
      "eval_samples_per_second": 5.281,
      "eval_steps_per_second": 0.441,
      "step": 13000
    },
    {
      "epoch": 0.67,
      "learning_rate": 4.665907741041378e-05,
      "loss": 1.2526,
      "step": 13500
    },
    {
      "epoch": 0.67,
      "eval_bleu": 35.286,
      "eval_gen_len": 17.0997,
      "eval_loss": 1.3556008338928223,
      "eval_runtime": 1360.8433,
      "eval_samples_per_second": 5.249,
      "eval_steps_per_second": 0.438,
      "step": 13500
    },
    {
      "epoch": 0.69,
      "learning_rate": 4.65353395367254e-05,
      "loss": 1.2355,
      "step": 14000
    },
    {
      "epoch": 0.69,
      "eval_bleu": 35.5153,
      "eval_gen_len": 16.7456,
      "eval_loss": 1.3322172164916992,
      "eval_runtime": 1342.5622,
      "eval_samples_per_second": 5.32,
      "eval_steps_per_second": 0.444,
      "step": 14000
    },
    {
      "epoch": 0.72,
      "learning_rate": 4.6411601663037026e-05,
      "loss": 1.2072,
      "step": 14500
    },
    {
      "epoch": 0.72,
      "eval_bleu": 35.9173,
      "eval_gen_len": 16.9327,
      "eval_loss": 1.3275119066238403,
      "eval_runtime": 1338.7699,
      "eval_samples_per_second": 5.335,
      "eval_steps_per_second": 0.445,
      "step": 14500
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.628786378934865e-05,
      "loss": 1.195,
      "step": 15000
    },
    {
      "epoch": 0.74,
      "eval_bleu": 35.7681,
      "eval_gen_len": 16.8524,
      "eval_loss": 1.3128494024276733,
      "eval_runtime": 1345.0353,
      "eval_samples_per_second": 5.311,
      "eval_steps_per_second": 0.443,
      "step": 15000
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.6164125915660264e-05,
      "loss": 1.1689,
      "step": 15500
    },
    {
      "epoch": 0.77,
      "eval_bleu": 36.7489,
      "eval_gen_len": 16.8296,
      "eval_loss": 1.2844336032867432,
      "eval_runtime": 1344.8016,
      "eval_samples_per_second": 5.312,
      "eval_steps_per_second": 0.443,
      "step": 15500
    },
    {
      "epoch": 0.79,
      "learning_rate": 4.6040388041971886e-05,
      "loss": 1.1573,
      "step": 16000
    },
    {
      "epoch": 0.79,
      "eval_bleu": 37.2278,
      "eval_gen_len": 16.8921,
      "eval_loss": 1.2794346809387207,
      "eval_runtime": 1341.0354,
      "eval_samples_per_second": 5.326,
      "eval_steps_per_second": 0.444,
      "step": 16000
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.591665016828351e-05,
      "loss": 1.1307,
      "step": 16500
    },
    {
      "epoch": 0.82,
      "eval_bleu": 36.748,
      "eval_gen_len": 16.5867,
      "eval_loss": 1.2566970586776733,
      "eval_runtime": 1321.9093,
      "eval_samples_per_second": 5.404,
      "eval_steps_per_second": 0.451,
      "step": 16500
    },
    {
      "epoch": 0.84,
      "learning_rate": 4.579291229459513e-05,
      "loss": 1.1222,
      "step": 17000
    },
    {
      "epoch": 0.84,
      "eval_bleu": 37.2793,
      "eval_gen_len": 16.838,
      "eval_loss": 1.2408839464187622,
      "eval_runtime": 1339.7522,
      "eval_samples_per_second": 5.332,
      "eval_steps_per_second": 0.445,
      "step": 17000
    },
    {
      "epoch": 0.87,
      "learning_rate": 4.5669174420906754e-05,
      "loss": 1.1129,
      "step": 17500
    },
    {
      "epoch": 0.87,
      "eval_bleu": 37.5801,
      "eval_gen_len": 16.6506,
      "eval_loss": 1.2285243272781372,
      "eval_runtime": 1332.5485,
      "eval_samples_per_second": 5.36,
      "eval_steps_per_second": 0.447,
      "step": 17500
    },
    {
      "epoch": 0.89,
      "learning_rate": 4.5545436547218376e-05,
      "loss": 1.11,
      "step": 18000
    },
    {
      "epoch": 0.89,
      "eval_bleu": 37.3726,
      "eval_gen_len": 17.1085,
      "eval_loss": 1.2697083950042725,
      "eval_runtime": 1412.2634,
      "eval_samples_per_second": 5.058,
      "eval_steps_per_second": 0.422,
      "step": 18000
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.542169867352999e-05,
      "loss": 1.0779,
      "step": 18500
    },
    {
      "epoch": 0.92,
      "eval_bleu": 38.4744,
      "eval_gen_len": 16.9744,
      "eval_loss": 1.2056454420089722,
      "eval_runtime": 1349.6935,
      "eval_samples_per_second": 5.292,
      "eval_steps_per_second": 0.442,
      "step": 18500
    },
    {
      "epoch": 0.94,
      "learning_rate": 4.5297960799841614e-05,
      "loss": 1.0651,
      "step": 19000
    },
    {
      "epoch": 0.94,
      "eval_bleu": 38.3271,
      "eval_gen_len": 17.006,
      "eval_loss": 1.1996458768844604,
      "eval_runtime": 1369.5468,
      "eval_samples_per_second": 5.216,
      "eval_steps_per_second": 0.435,
      "step": 19000
    },
    {
      "epoch": 0.97,
      "learning_rate": 4.5174222926153237e-05,
      "loss": 1.0647,
      "step": 19500
    },
    {
      "epoch": 0.97,
      "eval_bleu": 38.6809,
      "eval_gen_len": 16.7178,
      "eval_loss": 1.1885197162628174,
      "eval_runtime": 1332.8156,
      "eval_samples_per_second": 5.359,
      "eval_steps_per_second": 0.447,
      "step": 19500
    },
    {
      "epoch": 0.99,
      "learning_rate": 4.505048505246486e-05,
      "loss": 1.0463,
      "step": 20000
    },
    {
      "epoch": 0.99,
      "eval_bleu": 39.0708,
      "eval_gen_len": 16.8376,
      "eval_loss": 1.1725914478302002,
      "eval_runtime": 1343.3869,
      "eval_samples_per_second": 5.317,
      "eval_steps_per_second": 0.444,
      "step": 20000
    },
    {
      "epoch": 1.01,
      "learning_rate": 4.492674717877648e-05,
      "loss": 0.9704,
      "step": 20500
    },
    {
      "epoch": 1.01,
      "eval_bleu": 39.6273,
      "eval_gen_len": 16.8159,
      "eval_loss": 1.158150315284729,
      "eval_runtime": 1348.8721,
      "eval_samples_per_second": 5.296,
      "eval_steps_per_second": 0.442,
      "step": 20500
    },
    {
      "epoch": 1.04,
      "learning_rate": 4.4803009305088104e-05,
      "loss": 0.8936,
      "step": 21000
    },
    {
      "epoch": 1.04,
      "eval_bleu": 39.3436,
      "eval_gen_len": 16.6532,
      "eval_loss": 1.1593575477600098,
      "eval_runtime": 1336.9289,
      "eval_samples_per_second": 5.343,
      "eval_steps_per_second": 0.446,
      "step": 21000
    },
    {
      "epoch": 1.06,
      "learning_rate": 4.4679271431399726e-05,
      "loss": 0.8839,
      "step": 21500
    },
    {
      "epoch": 1.06,
      "eval_bleu": 39.8191,
      "eval_gen_len": 16.4936,
      "eval_loss": 1.1491360664367676,
      "eval_runtime": 1334.0963,
      "eval_samples_per_second": 5.354,
      "eval_steps_per_second": 0.447,
      "step": 21500
    },
    {
      "epoch": 1.09,
      "learning_rate": 4.455553355771135e-05,
      "loss": 0.8823,
      "step": 22000
    },
    {
      "epoch": 1.09,
      "eval_bleu": 40.01,
      "eval_gen_len": 16.6578,
      "eval_loss": 1.14137601852417,
      "eval_runtime": 1346.5689,
      "eval_samples_per_second": 5.305,
      "eval_steps_per_second": 0.443,
      "step": 22000
    },
    {
      "epoch": 1.11,
      "learning_rate": 4.443179568402297e-05,
      "loss": 0.8898,
      "step": 22500
    },
    {
      "epoch": 1.11,
      "eval_bleu": 39.7192,
      "eval_gen_len": 16.8047,
      "eval_loss": 1.125238299369812,
      "eval_runtime": 1349.9067,
      "eval_samples_per_second": 5.291,
      "eval_steps_per_second": 0.442,
      "step": 22500
    },
    {
      "epoch": 1.14,
      "learning_rate": 4.4308057810334594e-05,
      "loss": 0.8783,
      "step": 23000
    },
    {
      "epoch": 1.14,
      "eval_bleu": 39.9004,
      "eval_gen_len": 16.7007,
      "eval_loss": 1.1242141723632812,
      "eval_runtime": 1331.8043,
      "eval_samples_per_second": 5.363,
      "eval_steps_per_second": 0.448,
      "step": 23000
    },
    {
      "epoch": 1.16,
      "learning_rate": 4.418431993664621e-05,
      "loss": 0.869,
      "step": 23500
    },
    {
      "epoch": 1.16,
      "eval_bleu": 40.6549,
      "eval_gen_len": 16.8279,
      "eval_loss": 1.1088123321533203,
      "eval_runtime": 1337.7469,
      "eval_samples_per_second": 5.34,
      "eval_steps_per_second": 0.446,
      "step": 23500
    },
    {
      "epoch": 1.19,
      "learning_rate": 4.406058206295783e-05,
      "loss": 0.8602,
      "step": 24000
    },
    {
      "epoch": 1.19,
      "eval_bleu": 40.9968,
      "eval_gen_len": 16.7487,
      "eval_loss": 1.100970983505249,
      "eval_runtime": 1347.265,
      "eval_samples_per_second": 5.302,
      "eval_steps_per_second": 0.442,
      "step": 24000
    },
    {
      "epoch": 1.21,
      "learning_rate": 4.3936844189269454e-05,
      "loss": 0.8617,
      "step": 24500
    },
    {
      "epoch": 1.21,
      "eval_bleu": 40.6003,
      "eval_gen_len": 16.544,
      "eval_loss": 1.1008094549179077,
      "eval_runtime": 1323.7915,
      "eval_samples_per_second": 5.396,
      "eval_steps_per_second": 0.45,
      "step": 24500
    },
    {
      "epoch": 1.24,
      "learning_rate": 4.3813106315581076e-05,
      "loss": 0.8571,
      "step": 25000
    },
    {
      "epoch": 1.24,
      "eval_bleu": 41.2081,
      "eval_gen_len": 16.8625,
      "eval_loss": 1.0926491022109985,
      "eval_runtime": 1338.7092,
      "eval_samples_per_second": 5.336,
      "eval_steps_per_second": 0.445,
      "step": 25000
    },
    {
      "epoch": 1.26,
      "learning_rate": 4.36893684418927e-05,
      "loss": 0.845,
      "step": 25500
    },
    {
      "epoch": 1.26,
      "eval_bleu": 41.0718,
      "eval_gen_len": 16.6553,
      "eval_loss": 1.088665246963501,
      "eval_runtime": 1330.6533,
      "eval_samples_per_second": 5.368,
      "eval_steps_per_second": 0.448,
      "step": 25500
    },
    {
      "epoch": 1.29,
      "learning_rate": 4.356563056820432e-05,
      "loss": 0.8533,
      "step": 26000
    },
    {
      "epoch": 1.29,
      "eval_bleu": 41.2884,
      "eval_gen_len": 16.7981,
      "eval_loss": 1.0817835330963135,
      "eval_runtime": 1339.6246,
      "eval_samples_per_second": 5.332,
      "eval_steps_per_second": 0.445,
      "step": 26000
    },
    {
      "epoch": 1.31,
      "learning_rate": 4.344189269451594e-05,
      "loss": 0.8486,
      "step": 26500
    },
    {
      "epoch": 1.31,
      "eval_bleu": 40.9701,
      "eval_gen_len": 16.9712,
      "eval_loss": 1.0806611776351929,
      "eval_runtime": 1354.0651,
      "eval_samples_per_second": 5.275,
      "eval_steps_per_second": 0.44,
      "step": 26500
    },
    {
      "epoch": 1.34,
      "learning_rate": 4.331815482082756e-05,
      "loss": 0.8441,
      "step": 27000
    },
    {
      "epoch": 1.34,
      "eval_bleu": 41.3608,
      "eval_gen_len": 16.6277,
      "eval_loss": 1.0752187967300415,
      "eval_runtime": 1330.2452,
      "eval_samples_per_second": 5.37,
      "eval_steps_per_second": 0.448,
      "step": 27000
    },
    {
      "epoch": 1.36,
      "learning_rate": 4.319441694713918e-05,
      "loss": 0.8531,
      "step": 27500
    },
    {
      "epoch": 1.36,
      "eval_bleu": 41.5215,
      "eval_gen_len": 16.8085,
      "eval_loss": 1.066485047340393,
      "eval_runtime": 1334.1554,
      "eval_samples_per_second": 5.354,
      "eval_steps_per_second": 0.447,
      "step": 27500
    },
    {
      "epoch": 1.39,
      "learning_rate": 4.3070679073450804e-05,
      "loss": 0.8371,
      "step": 28000
    },
    {
      "epoch": 1.39,
      "eval_bleu": 41.9045,
      "eval_gen_len": 16.736,
      "eval_loss": 1.0553510189056396,
      "eval_runtime": 1333.8696,
      "eval_samples_per_second": 5.355,
      "eval_steps_per_second": 0.447,
      "step": 28000
    },
    {
      "epoch": 1.41,
      "learning_rate": 4.294694119976243e-05,
      "loss": 0.8171,
      "step": 28500
    },
    {
      "epoch": 1.41,
      "eval_bleu": 42.0877,
      "eval_gen_len": 16.7124,
      "eval_loss": 1.0496976375579834,
      "eval_runtime": 1332.4047,
      "eval_samples_per_second": 5.361,
      "eval_steps_per_second": 0.447,
      "step": 28500
    },
    {
      "epoch": 1.44,
      "learning_rate": 4.282320332607405e-05,
      "loss": 0.8359,
      "step": 29000
    },
    {
      "epoch": 1.44,
      "eval_bleu": 42.1081,
      "eval_gen_len": 16.6317,
      "eval_loss": 1.0442363023757935,
      "eval_runtime": 1329.2047,
      "eval_samples_per_second": 5.374,
      "eval_steps_per_second": 0.448,
      "step": 29000
    },
    {
      "epoch": 1.46,
      "learning_rate": 4.2699465452385665e-05,
      "loss": 0.8097,
      "step": 29500
    },
    {
      "epoch": 1.46,
      "eval_bleu": 42.3305,
      "eval_gen_len": 16.5241,
      "eval_loss": 1.0467731952667236,
      "eval_runtime": 1325.978,
      "eval_samples_per_second": 5.387,
      "eval_steps_per_second": 0.449,
      "step": 29500
    },
    {
      "epoch": 1.48,
      "learning_rate": 4.257572757869729e-05,
      "loss": 0.8151,
      "step": 30000
    },
    {
      "epoch": 1.48,
      "eval_bleu": 42.208,
      "eval_gen_len": 16.5507,
      "eval_loss": 1.0358787775039673,
      "eval_runtime": 1325.8985,
      "eval_samples_per_second": 5.387,
      "eval_steps_per_second": 0.45,
      "step": 30000
    },
    {
      "epoch": 1.51,
      "learning_rate": 4.245198970500891e-05,
      "loss": 0.8008,
      "step": 30500
    },
    {
      "epoch": 1.51,
      "eval_bleu": 43.0639,
      "eval_gen_len": 16.7362,
      "eval_loss": 1.0226542949676514,
      "eval_runtime": 1337.8285,
      "eval_samples_per_second": 5.339,
      "eval_steps_per_second": 0.445,
      "step": 30500
    },
    {
      "epoch": 1.53,
      "learning_rate": 4.232825183132053e-05,
      "loss": 0.8105,
      "step": 31000
    },
    {
      "epoch": 1.53,
      "eval_bleu": 42.7617,
      "eval_gen_len": 16.4645,
      "eval_loss": 1.016375184059143,
      "eval_runtime": 1327.3901,
      "eval_samples_per_second": 5.381,
      "eval_steps_per_second": 0.449,
      "step": 31000
    },
    {
      "epoch": 1.56,
      "learning_rate": 4.2204513957632154e-05,
      "loss": 0.8108,
      "step": 31500
    },
    {
      "epoch": 1.56,
      "eval_bleu": 42.9124,
      "eval_gen_len": 16.8116,
      "eval_loss": 1.0157625675201416,
      "eval_runtime": 1343.0655,
      "eval_samples_per_second": 5.318,
      "eval_steps_per_second": 0.444,
      "step": 31500
    },
    {
      "epoch": 1.58,
      "learning_rate": 4.208077608394378e-05,
      "loss": 0.8052,
      "step": 32000
    },
    {
      "epoch": 1.58,
      "eval_bleu": 42.8819,
      "eval_gen_len": 16.6151,
      "eval_loss": 1.0091865062713623,
      "eval_runtime": 1330.5695,
      "eval_samples_per_second": 5.368,
      "eval_steps_per_second": 0.448,
      "step": 32000
    },
    {
      "epoch": 1.61,
      "learning_rate": 4.195703821025539e-05,
      "loss": 0.8024,
      "step": 32500
    },
    {
      "epoch": 1.61,
      "eval_bleu": 43.0788,
      "eval_gen_len": 16.7483,
      "eval_loss": 1.0076141357421875,
      "eval_runtime": 1332.4446,
      "eval_samples_per_second": 5.361,
      "eval_steps_per_second": 0.447,
      "step": 32500
    },
    {
      "epoch": 1.63,
      "learning_rate": 4.1833300336567015e-05,
      "loss": 0.7971,
      "step": 33000
    },
    {
      "epoch": 1.63,
      "eval_bleu": 43.6296,
      "eval_gen_len": 16.7942,
      "eval_loss": 0.9935155510902405,
      "eval_runtime": 1351.7288,
      "eval_samples_per_second": 5.284,
      "eval_steps_per_second": 0.441,
      "step": 33000
    },
    {
      "epoch": 1.66,
      "learning_rate": 4.170956246287864e-05,
      "loss": 0.8006,
      "step": 33500
    },
    {
      "epoch": 1.66,
      "eval_bleu": 43.9739,
      "eval_gen_len": 16.7378,
      "eval_loss": 0.991138219833374,
      "eval_runtime": 1346.9798,
      "eval_samples_per_second": 5.303,
      "eval_steps_per_second": 0.442,
      "step": 33500
    },
    {
      "epoch": 1.68,
      "learning_rate": 4.158582458919026e-05,
      "loss": 0.7855,
      "step": 34000
    },
    {
      "epoch": 1.68,
      "eval_bleu": 43.5366,
      "eval_gen_len": 16.5919,
      "eval_loss": 0.991809070110321,
      "eval_runtime": 1337.0859,
      "eval_samples_per_second": 5.342,
      "eval_steps_per_second": 0.446,
      "step": 34000
    },
    {
      "epoch": 1.71,
      "learning_rate": 4.146208671550188e-05,
      "loss": 0.7786,
      "step": 34500
    },
    {
      "epoch": 1.71,
      "eval_bleu": 43.8212,
      "eval_gen_len": 16.5705,
      "eval_loss": 0.9883202314376831,
      "eval_runtime": 1322.7907,
      "eval_samples_per_second": 5.4,
      "eval_steps_per_second": 0.451,
      "step": 34500
    },
    {
      "epoch": 1.73,
      "learning_rate": 4.1338348841813505e-05,
      "loss": 0.7877,
      "step": 35000
    },
    {
      "epoch": 1.73,
      "eval_bleu": 43.7342,
      "eval_gen_len": 16.7165,
      "eval_loss": 0.9775007963180542,
      "eval_runtime": 1338.9365,
      "eval_samples_per_second": 5.335,
      "eval_steps_per_second": 0.445,
      "step": 35000
    },
    {
      "epoch": 1.76,
      "learning_rate": 4.121461096812513e-05,
      "loss": 0.7832,
      "step": 35500
    },
    {
      "epoch": 1.76,
      "eval_bleu": 43.8512,
      "eval_gen_len": 16.712,
      "eval_loss": 0.9759693145751953,
      "eval_runtime": 1341.6352,
      "eval_samples_per_second": 5.324,
      "eval_steps_per_second": 0.444,
      "step": 35500
    },
    {
      "epoch": 1.78,
      "learning_rate": 4.109087309443675e-05,
      "loss": 0.7812,
      "step": 36000
    },
    {
      "epoch": 1.78,
      "eval_bleu": 44.606,
      "eval_gen_len": 16.7719,
      "eval_loss": 0.9725255370140076,
      "eval_runtime": 1344.5964,
      "eval_samples_per_second": 5.312,
      "eval_steps_per_second": 0.443,
      "step": 36000
    },
    {
      "epoch": 1.81,
      "learning_rate": 4.096713522074837e-05,
      "loss": 0.766,
      "step": 36500
    },
    {
      "epoch": 1.81,
      "eval_bleu": 44.1586,
      "eval_gen_len": 16.568,
      "eval_loss": 0.9698735475540161,
      "eval_runtime": 1331.8223,
      "eval_samples_per_second": 5.363,
      "eval_steps_per_second": 0.448,
      "step": 36500
    },
    {
      "epoch": 1.83,
      "learning_rate": 4.0843397347059994e-05,
      "loss": 0.7688,
      "step": 37000
    },
    {
      "epoch": 1.83,
      "eval_bleu": 44.5813,
      "eval_gen_len": 16.7396,
      "eval_loss": 0.9606902599334717,
      "eval_runtime": 1334.582,
      "eval_samples_per_second": 5.352,
      "eval_steps_per_second": 0.447,
      "step": 37000
    },
    {
      "epoch": 1.86,
      "learning_rate": 4.071965947337161e-05,
      "loss": 0.7727,
      "step": 37500
    },
    {
      "epoch": 1.86,
      "eval_bleu": 44.1615,
      "eval_gen_len": 16.707,
      "eval_loss": 0.962368369102478,
      "eval_runtime": 1341.1556,
      "eval_samples_per_second": 5.326,
      "eval_steps_per_second": 0.444,
      "step": 37500
    },
    {
      "epoch": 1.88,
      "learning_rate": 4.059592159968323e-05,
      "loss": 0.7645,
      "step": 38000
    },
    {
      "epoch": 1.88,
      "eval_bleu": 44.1652,
      "eval_gen_len": 16.5927,
      "eval_loss": 0.9592034220695496,
      "eval_runtime": 1336.4184,
      "eval_samples_per_second": 5.345,
      "eval_steps_per_second": 0.446,
      "step": 38000
    },
    {
      "epoch": 1.91,
      "learning_rate": 4.0472183725994855e-05,
      "loss": 0.7638,
      "step": 38500
    },
    {
      "epoch": 1.91,
      "eval_bleu": 44.4663,
      "eval_gen_len": 16.6448,
      "eval_loss": 0.9551029205322266,
      "eval_runtime": 1335.4927,
      "eval_samples_per_second": 5.349,
      "eval_steps_per_second": 0.446,
      "step": 38500
    },
    {
      "epoch": 1.93,
      "learning_rate": 4.034844585230648e-05,
      "loss": 0.7544,
      "step": 39000
    },
    {
      "epoch": 1.93,
      "eval_bleu": 44.3807,
      "eval_gen_len": 16.7161,
      "eval_loss": 0.9564537405967712,
      "eval_runtime": 1335.4847,
      "eval_samples_per_second": 5.349,
      "eval_steps_per_second": 0.446,
      "step": 39000
    },
    {
      "epoch": 1.96,
      "learning_rate": 4.02247079786181e-05,
      "loss": 0.7513,
      "step": 39500
    },
    {
      "epoch": 1.96,
      "eval_bleu": 44.9465,
      "eval_gen_len": 16.6132,
      "eval_loss": 0.9481050372123718,
      "eval_runtime": 1327.2386,
      "eval_samples_per_second": 5.382,
      "eval_steps_per_second": 0.449,
      "step": 39500
    },
    {
      "epoch": 1.98,
      "learning_rate": 4.010097010492972e-05,
      "loss": 0.7563,
      "step": 40000
    },
    {
      "epoch": 1.98,
      "eval_bleu": 45.5396,
      "eval_gen_len": 16.8183,
      "eval_loss": 0.939414381980896,
      "eval_runtime": 1341.9443,
      "eval_samples_per_second": 5.323,
      "eval_steps_per_second": 0.444,
      "step": 40000
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.997723223124134e-05,
      "loss": 0.7206,
      "step": 40500
    },
    {
      "epoch": 2.0,
      "eval_bleu": 44.6231,
      "eval_gen_len": 16.5586,
      "eval_loss": 0.9512418508529663,
      "eval_runtime": 1333.952,
      "eval_samples_per_second": 5.355,
      "eval_steps_per_second": 0.447,
      "step": 40500
    },
    {
      "epoch": 2.03,
      "learning_rate": 3.985349435755296e-05,
      "loss": 0.5864,
      "step": 41000
    },
    {
      "epoch": 2.03,
      "eval_bleu": 45.4858,
      "eval_gen_len": 16.784,
      "eval_loss": 0.9353885650634766,
      "eval_runtime": 1336.3848,
      "eval_samples_per_second": 5.345,
      "eval_steps_per_second": 0.446,
      "step": 41000
    },
    {
      "epoch": 2.05,
      "learning_rate": 3.972975648386458e-05,
      "loss": 0.5916,
      "step": 41500
    },
    {
      "epoch": 2.05,
      "eval_bleu": 45.1704,
      "eval_gen_len": 16.6795,
      "eval_loss": 0.9400736689567566,
      "eval_runtime": 1337.6062,
      "eval_samples_per_second": 5.34,
      "eval_steps_per_second": 0.446,
      "step": 41500
    },
    {
      "epoch": 2.08,
      "learning_rate": 3.9606018610176205e-05,
      "loss": 0.5907,
      "step": 42000
    },
    {
      "epoch": 2.08,
      "eval_bleu": 45.0619,
      "eval_gen_len": 16.6905,
      "eval_loss": 0.9472643733024597,
      "eval_runtime": 1332.9894,
      "eval_samples_per_second": 5.359,
      "eval_steps_per_second": 0.447,
      "step": 42000
    },
    {
      "epoch": 2.1,
      "learning_rate": 3.948228073648783e-05,
      "loss": 0.584,
      "step": 42500
    },
    {
      "epoch": 2.1,
      "eval_bleu": 45.072,
      "eval_gen_len": 16.7739,
      "eval_loss": 0.943854808807373,
      "eval_runtime": 1336.6571,
      "eval_samples_per_second": 5.344,
      "eval_steps_per_second": 0.446,
      "step": 42500
    },
    {
      "epoch": 2.13,
      "learning_rate": 3.935854286279945e-05,
      "loss": 0.5979,
      "step": 43000
    },
    {
      "epoch": 2.13,
      "eval_bleu": 45.1267,
      "eval_gen_len": 16.9086,
      "eval_loss": 0.9410138726234436,
      "eval_runtime": 1338.7702,
      "eval_samples_per_second": 5.335,
      "eval_steps_per_second": 0.445,
      "step": 43000
    },
    {
      "epoch": 2.15,
      "learning_rate": 3.9234804989111066e-05,
      "loss": 0.5929,
      "step": 43500
    },
    {
      "epoch": 2.15,
      "eval_bleu": 45.3207,
      "eval_gen_len": 16.6486,
      "eval_loss": 0.9384570121765137,
      "eval_runtime": 1333.0456,
      "eval_samples_per_second": 5.358,
      "eval_steps_per_second": 0.447,
      "step": 43500
    },
    {
      "epoch": 2.18,
      "learning_rate": 3.911106711542269e-05,
      "loss": 0.588,
      "step": 44000
    },
    {
      "epoch": 2.18,
      "eval_bleu": 45.3945,
      "eval_gen_len": 16.7284,
      "eval_loss": 0.9332572817802429,
      "eval_runtime": 1330.0518,
      "eval_samples_per_second": 5.37,
      "eval_steps_per_second": 0.448,
      "step": 44000
    },
    {
      "epoch": 2.2,
      "learning_rate": 3.898732924173431e-05,
      "loss": 0.594,
      "step": 44500
    },
    {
      "epoch": 2.2,
      "eval_bleu": 45.3986,
      "eval_gen_len": 16.6814,
      "eval_loss": 0.9292554259300232,
      "eval_runtime": 1334.5426,
      "eval_samples_per_second": 5.352,
      "eval_steps_per_second": 0.447,
      "step": 44500
    },
    {
      "epoch": 2.23,
      "learning_rate": 3.886359136804593e-05,
      "loss": 0.5983,
      "step": 45000
    },
    {
      "epoch": 2.23,
      "eval_bleu": 45.6357,
      "eval_gen_len": 16.617,
      "eval_loss": 0.920947253704071,
      "eval_runtime": 1332.827,
      "eval_samples_per_second": 5.359,
      "eval_steps_per_second": 0.447,
      "step": 45000
    },
    {
      "epoch": 2.25,
      "learning_rate": 3.8739853494357555e-05,
      "loss": 0.6063,
      "step": 45500
    },
    {
      "epoch": 2.25,
      "eval_bleu": 45.8697,
      "eval_gen_len": 16.7532,
      "eval_loss": 0.9304359555244446,
      "eval_runtime": 1340.1007,
      "eval_samples_per_second": 5.33,
      "eval_steps_per_second": 0.445,
      "step": 45500
    },
    {
      "epoch": 2.28,
      "learning_rate": 3.861611562066917e-05,
      "loss": 0.6004,
      "step": 46000
    },
    {
      "epoch": 2.28,
      "eval_bleu": 45.8707,
      "eval_gen_len": 16.6399,
      "eval_loss": 0.9174883961677551,
      "eval_runtime": 1333.508,
      "eval_samples_per_second": 5.357,
      "eval_steps_per_second": 0.447,
      "step": 46000
    },
    {
      "epoch": 2.3,
      "learning_rate": 3.8492377746980794e-05,
      "loss": 0.5951,
      "step": 46500
    },
    {
      "epoch": 2.3,
      "eval_bleu": 45.0087,
      "eval_gen_len": 16.5349,
      "eval_loss": 0.9316369295120239,
      "eval_runtime": 1331.8005,
      "eval_samples_per_second": 5.363,
      "eval_steps_per_second": 0.448,
      "step": 46500
    },
    {
      "epoch": 2.33,
      "learning_rate": 3.8368639873292416e-05,
      "loss": 0.5908,
      "step": 47000
    },
    {
      "epoch": 2.33,
      "eval_bleu": 45.485,
      "eval_gen_len": 16.5745,
      "eval_loss": 0.9216477870941162,
      "eval_runtime": 1328.0985,
      "eval_samples_per_second": 5.378,
      "eval_steps_per_second": 0.449,
      "step": 47000
    },
    {
      "epoch": 2.35,
      "learning_rate": 3.824490199960404e-05,
      "loss": 0.5847,
      "step": 47500
    },
    {
      "epoch": 2.35,
      "eval_bleu": 45.9099,
      "eval_gen_len": 16.6137,
      "eval_loss": 0.9153968691825867,
      "eval_runtime": 1331.7962,
      "eval_samples_per_second": 5.363,
      "eval_steps_per_second": 0.448,
      "step": 47500
    },
    {
      "epoch": 2.38,
      "learning_rate": 3.812116412591566e-05,
      "loss": 0.5897,
      "step": 48000
    },
    {
      "epoch": 2.38,
      "eval_bleu": 46.0344,
      "eval_gen_len": 16.5282,
      "eval_loss": 0.9067392945289612,
      "eval_runtime": 1320.4869,
      "eval_samples_per_second": 5.409,
      "eval_steps_per_second": 0.451,
      "step": 48000
    },
    {
      "epoch": 2.4,
      "learning_rate": 3.799742625222728e-05,
      "loss": 0.5951,
      "step": 48500
    },
    {
      "epoch": 2.4,
      "eval_bleu": 45.9988,
      "eval_gen_len": 16.6501,
      "eval_loss": 0.9071055054664612,
      "eval_runtime": 1333.3527,
      "eval_samples_per_second": 5.357,
      "eval_steps_per_second": 0.447,
      "step": 48500
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.7873688378538906e-05,
      "loss": 0.5781,
      "step": 49000
    },
    {
      "epoch": 2.43,
      "eval_bleu": 46.0018,
      "eval_gen_len": 16.7336,
      "eval_loss": 0.9073439240455627,
      "eval_runtime": 1334.0382,
      "eval_samples_per_second": 5.354,
      "eval_steps_per_second": 0.447,
      "step": 49000
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.774995050485053e-05,
      "loss": 0.5934,
      "step": 49500
    },
    {
      "epoch": 2.45,
      "eval_bleu": 46.5706,
      "eval_gen_len": 16.8471,
      "eval_loss": 0.9027701616287231,
      "eval_runtime": 1336.5927,
      "eval_samples_per_second": 5.344,
      "eval_steps_per_second": 0.446,
      "step": 49500
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.762621263116215e-05,
      "loss": 0.5883,
      "step": 50000
    },
    {
      "epoch": 2.47,
      "eval_bleu": 46.6303,
      "eval_gen_len": 16.8614,
      "eval_loss": 0.9047964215278625,
      "eval_runtime": 1340.8517,
      "eval_samples_per_second": 5.327,
      "eval_steps_per_second": 0.444,
      "step": 50000
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.750247475747377e-05,
      "loss": 0.5964,
      "step": 50500
    },
    {
      "epoch": 2.5,
      "eval_bleu": 46.087,
      "eval_gen_len": 16.6417,
      "eval_loss": 0.8980411887168884,
      "eval_runtime": 1332.5501,
      "eval_samples_per_second": 5.36,
      "eval_steps_per_second": 0.447,
      "step": 50500
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.7378736883785395e-05,
      "loss": 0.5924,
      "step": 51000
    },
    {
      "epoch": 2.52,
      "eval_bleu": 47.0224,
      "eval_gen_len": 16.6821,
      "eval_loss": 0.9017878770828247,
      "eval_runtime": 1333.0812,
      "eval_samples_per_second": 5.358,
      "eval_steps_per_second": 0.447,
      "step": 51000
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.725499901009701e-05,
      "loss": 0.5866,
      "step": 51500
    },
    {
      "epoch": 2.55,
      "eval_bleu": 46.244,
      "eval_gen_len": 16.7493,
      "eval_loss": 0.8992940187454224,
      "eval_runtime": 1343.2842,
      "eval_samples_per_second": 5.318,
      "eval_steps_per_second": 0.444,
      "step": 51500
    },
    {
      "epoch": 2.57,
      "learning_rate": 3.7131261136408633e-05,
      "loss": 0.5764,
      "step": 52000
    },
    {
      "epoch": 2.57,
      "eval_bleu": 46.3888,
      "eval_gen_len": 16.6049,
      "eval_loss": 0.9019588828086853,
      "eval_runtime": 1338.9531,
      "eval_samples_per_second": 5.335,
      "eval_steps_per_second": 0.445,
      "step": 52000
    },
    {
      "epoch": 2.6,
      "learning_rate": 3.7007523262720256e-05,
      "loss": 0.5865,
      "step": 52500
    },
    {
      "epoch": 2.6,
      "eval_bleu": 46.8308,
      "eval_gen_len": 16.7173,
      "eval_loss": 0.8952421545982361,
      "eval_runtime": 1352.2564,
      "eval_samples_per_second": 5.282,
      "eval_steps_per_second": 0.441,
      "step": 52500
    },
    {
      "epoch": 2.62,
      "learning_rate": 3.688378538903188e-05,
      "loss": 0.6016,
      "step": 53000
    },
    {
      "epoch": 2.62,
      "eval_bleu": 47.0928,
      "eval_gen_len": 16.7008,
      "eval_loss": 0.8945097923278809,
      "eval_runtime": 1354.2335,
      "eval_samples_per_second": 5.275,
      "eval_steps_per_second": 0.44,
      "step": 53000
    },
    {
      "epoch": 2.65,
      "learning_rate": 3.67600475153435e-05,
      "loss": 0.5902,
      "step": 53500
    },
    {
      "epoch": 2.65,
      "eval_bleu": 46.6548,
      "eval_gen_len": 16.7439,
      "eval_loss": 0.8932007551193237,
      "eval_runtime": 1382.9624,
      "eval_samples_per_second": 5.165,
      "eval_steps_per_second": 0.431,
      "step": 53500
    },
    {
      "epoch": 2.67,
      "learning_rate": 3.663630964165512e-05,
      "loss": 0.587,
      "step": 54000
    },
    {
      "epoch": 2.67,
      "eval_bleu": 46.9113,
      "eval_gen_len": 16.6783,
      "eval_loss": 0.8879196047782898,
      "eval_runtime": 1386.1138,
      "eval_samples_per_second": 5.153,
      "eval_steps_per_second": 0.43,
      "step": 54000
    },
    {
      "epoch": 2.7,
      "learning_rate": 3.651257176796674e-05,
      "loss": 0.5968,
      "step": 54500
    },
    {
      "epoch": 2.7,
      "eval_bleu": 47.0957,
      "eval_gen_len": 16.5933,
      "eval_loss": 0.8864062428474426,
      "eval_runtime": 1383.0254,
      "eval_samples_per_second": 5.165,
      "eval_steps_per_second": 0.431,
      "step": 54500
    },
    {
      "epoch": 2.72,
      "learning_rate": 3.638883389427836e-05,
      "loss": 0.5853,
      "step": 55000
    },
    {
      "epoch": 2.72,
      "eval_bleu": 47.1709,
      "eval_gen_len": 16.6479,
      "eval_loss": 0.8836300373077393,
      "eval_runtime": 1390.1185,
      "eval_samples_per_second": 5.138,
      "eval_steps_per_second": 0.429,
      "step": 55000
    },
    {
      "epoch": 2.75,
      "learning_rate": 3.6265096020589984e-05,
      "loss": 0.5943,
      "step": 55500
    },
    {
      "epoch": 2.75,
      "eval_bleu": 47.184,
      "eval_gen_len": 16.6594,
      "eval_loss": 0.8828004598617554,
      "eval_runtime": 1374.4327,
      "eval_samples_per_second": 5.197,
      "eval_steps_per_second": 0.434,
      "step": 55500
    },
    {
      "epoch": 2.77,
      "learning_rate": 3.6141358146901606e-05,
      "loss": 0.5891,
      "step": 56000
    },
    {
      "epoch": 2.77,
      "eval_bleu": 46.7804,
      "eval_gen_len": 16.7312,
      "eval_loss": 0.8724841475486755,
      "eval_runtime": 1340.042,
      "eval_samples_per_second": 5.33,
      "eval_steps_per_second": 0.445,
      "step": 56000
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.601762027321323e-05,
      "loss": 0.585,
      "step": 56500
    },
    {
      "epoch": 2.8,
      "eval_bleu": 46.8152,
      "eval_gen_len": 16.6263,
      "eval_loss": 0.8797315955162048,
      "eval_runtime": 1333.5752,
      "eval_samples_per_second": 5.356,
      "eval_steps_per_second": 0.447,
      "step": 56500
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.5893882399524844e-05,
      "loss": 0.5889,
      "step": 57000
    },
    {
      "epoch": 2.82,
      "eval_bleu": 47.3674,
      "eval_gen_len": 16.7459,
      "eval_loss": 0.8741356134414673,
      "eval_runtime": 1336.3135,
      "eval_samples_per_second": 5.345,
      "eval_steps_per_second": 0.446,
      "step": 57000
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.577014452583647e-05,
      "loss": 0.5841,
      "step": 57500
    },
    {
      "epoch": 2.85,
      "eval_bleu": 46.9029,
      "eval_gen_len": 16.742,
      "eval_loss": 0.875153124332428,
      "eval_runtime": 1335.6996,
      "eval_samples_per_second": 5.348,
      "eval_steps_per_second": 0.446,
      "step": 57500
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.564640665214809e-05,
      "loss": 0.5786,
      "step": 58000
    },
    {
      "epoch": 2.87,
      "eval_bleu": 47.4711,
      "eval_gen_len": 16.6132,
      "eval_loss": 0.8668653964996338,
      "eval_runtime": 1333.4019,
      "eval_samples_per_second": 5.357,
      "eval_steps_per_second": 0.447,
      "step": 58000
    },
    {
      "epoch": 2.9,
      "learning_rate": 3.552266877845971e-05,
      "loss": 0.5784,
      "step": 58500
    },
    {
      "epoch": 2.9,
      "eval_bleu": 47.3049,
      "eval_gen_len": 16.6745,
      "eval_loss": 0.8657991290092468,
      "eval_runtime": 1339.6294,
      "eval_samples_per_second": 5.332,
      "eval_steps_per_second": 0.445,
      "step": 58500
    },
    {
      "epoch": 2.92,
      "learning_rate": 3.5398930904771334e-05,
      "loss": 0.5807,
      "step": 59000
    },
    {
      "epoch": 2.92,
      "eval_bleu": 47.2634,
      "eval_gen_len": 16.6962,
      "eval_loss": 0.8614685535430908,
      "eval_runtime": 1334.3541,
      "eval_samples_per_second": 5.353,
      "eval_steps_per_second": 0.447,
      "step": 59000
    },
    {
      "epoch": 2.94,
      "learning_rate": 3.5275193031082956e-05,
      "loss": 0.5856,
      "step": 59500
    },
    {
      "epoch": 2.94,
      "eval_bleu": 47.8743,
      "eval_gen_len": 16.7721,
      "eval_loss": 0.8591042757034302,
      "eval_runtime": 1335.7672,
      "eval_samples_per_second": 5.347,
      "eval_steps_per_second": 0.446,
      "step": 59500
    },
    {
      "epoch": 2.97,
      "learning_rate": 3.515145515739457e-05,
      "loss": 0.58,
      "step": 60000
    },
    {
      "epoch": 2.97,
      "eval_bleu": 47.6119,
      "eval_gen_len": 16.6251,
      "eval_loss": 0.8595740795135498,
      "eval_runtime": 1332.2154,
      "eval_samples_per_second": 5.362,
      "eval_steps_per_second": 0.447,
      "step": 60000
    },
    {
      "epoch": 2.99,
      "learning_rate": 3.5027717283706194e-05,
      "loss": 0.5802,
      "step": 60500
    },
    {
      "epoch": 2.99,
      "eval_bleu": 47.703,
      "eval_gen_len": 16.6114,
      "eval_loss": 0.8570582270622253,
      "eval_runtime": 1325.8364,
      "eval_samples_per_second": 5.388,
      "eval_steps_per_second": 0.45,
      "step": 60500
    },
    {
      "epoch": 3.02,
      "learning_rate": 3.490397941001782e-05,
      "loss": 0.4491,
      "step": 61000
    },
    {
      "epoch": 3.02,
      "eval_bleu": 47.8822,
      "eval_gen_len": 16.732,
      "eval_loss": 0.859774649143219,
      "eval_runtime": 1336.5318,
      "eval_samples_per_second": 5.344,
      "eval_steps_per_second": 0.446,
      "step": 61000
    },
    {
      "epoch": 3.04,
      "learning_rate": 3.478024153632944e-05,
      "loss": 0.4305,
      "step": 61500
    },
    {
      "epoch": 3.04,
      "eval_bleu": 47.6096,
      "eval_gen_len": 16.5932,
      "eval_loss": 0.8659536242485046,
      "eval_runtime": 1328.147,
      "eval_samples_per_second": 5.378,
      "eval_steps_per_second": 0.449,
      "step": 61500
    },
    {
      "epoch": 3.07,
      "learning_rate": 3.465650366264107e-05,
      "loss": 0.4268,
      "step": 62000
    },
    {
      "epoch": 3.07,
      "eval_bleu": 48.0044,
      "eval_gen_len": 16.7596,
      "eval_loss": 0.8622444868087769,
      "eval_runtime": 1332.1214,
      "eval_samples_per_second": 5.362,
      "eval_steps_per_second": 0.447,
      "step": 62000
    },
    {
      "epoch": 3.09,
      "learning_rate": 3.4532765788952684e-05,
      "loss": 0.4274,
      "step": 62500
    },
    {
      "epoch": 3.09,
      "eval_bleu": 47.8272,
      "eval_gen_len": 16.6352,
      "eval_loss": 0.8669659495353699,
      "eval_runtime": 1330.5387,
      "eval_samples_per_second": 5.369,
      "eval_steps_per_second": 0.448,
      "step": 62500
    },
    {
      "epoch": 3.12,
      "learning_rate": 3.4409027915264307e-05,
      "loss": 0.4383,
      "step": 63000
    },
    {
      "epoch": 3.12,
      "eval_bleu": 47.8411,
      "eval_gen_len": 16.6331,
      "eval_loss": 0.8617204427719116,
      "eval_runtime": 1339.6282,
      "eval_samples_per_second": 5.332,
      "eval_steps_per_second": 0.445,
      "step": 63000
    },
    {
      "epoch": 3.14,
      "learning_rate": 3.428529004157593e-05,
      "loss": 0.4339,
      "step": 63500
    },
    {
      "epoch": 3.14,
      "eval_bleu": 48.232,
      "eval_gen_len": 16.839,
      "eval_loss": 0.8655442595481873,
      "eval_runtime": 1339.0945,
      "eval_samples_per_second": 5.334,
      "eval_steps_per_second": 0.445,
      "step": 63500
    },
    {
      "epoch": 3.17,
      "learning_rate": 3.416155216788755e-05,
      "loss": 0.4361,
      "step": 64000
    },
    {
      "epoch": 3.17,
      "eval_bleu": 47.978,
      "eval_gen_len": 16.7057,
      "eval_loss": 0.8637909889221191,
      "eval_runtime": 1337.5265,
      "eval_samples_per_second": 5.34,
      "eval_steps_per_second": 0.446,
      "step": 64000
    },
    {
      "epoch": 3.19,
      "learning_rate": 3.4037814294199174e-05,
      "loss": 0.4373,
      "step": 64500
    },
    {
      "epoch": 3.19,
      "eval_bleu": 48.0663,
      "eval_gen_len": 16.652,
      "eval_loss": 0.8616685271263123,
      "eval_runtime": 1344.0854,
      "eval_samples_per_second": 5.314,
      "eval_steps_per_second": 0.443,
      "step": 64500
    },
    {
      "epoch": 3.22,
      "learning_rate": 3.3914076420510796e-05,
      "loss": 0.43,
      "step": 65000
    },
    {
      "epoch": 3.22,
      "eval_bleu": 47.8077,
      "eval_gen_len": 16.7621,
      "eval_loss": 0.8689031004905701,
      "eval_runtime": 1338.3656,
      "eval_samples_per_second": 5.337,
      "eval_steps_per_second": 0.445,
      "step": 65000
    },
    {
      "epoch": 3.24,
      "learning_rate": 3.379033854682241e-05,
      "loss": 0.4325,
      "step": 65500
    },
    {
      "epoch": 3.24,
      "eval_bleu": 47.9434,
      "eval_gen_len": 16.6489,
      "eval_loss": 0.870697021484375,
      "eval_runtime": 1329.6325,
      "eval_samples_per_second": 5.372,
      "eval_steps_per_second": 0.448,
      "step": 65500
    },
    {
      "epoch": 3.27,
      "learning_rate": 3.3666600673134034e-05,
      "loss": 0.4409,
      "step": 66000
    },
    {
      "epoch": 3.27,
      "eval_bleu": 48.2346,
      "eval_gen_len": 16.6951,
      "eval_loss": 0.871823787689209,
      "eval_runtime": 1334.2329,
      "eval_samples_per_second": 5.354,
      "eval_steps_per_second": 0.447,
      "step": 66000
    },
    {
      "epoch": 3.29,
      "learning_rate": 3.354286279944566e-05,
      "loss": 0.4407,
      "step": 66500
    },
    {
      "epoch": 3.29,
      "eval_bleu": 48.0356,
      "eval_gen_len": 16.7549,
      "eval_loss": 0.8691708445549011,
      "eval_runtime": 1345.0681,
      "eval_samples_per_second": 5.311,
      "eval_steps_per_second": 0.443,
      "step": 66500
    },
    {
      "epoch": 3.32,
      "learning_rate": 3.341912492575728e-05,
      "loss": 0.444,
      "step": 67000
    },
    {
      "epoch": 3.32,
      "eval_bleu": 48.295,
      "eval_gen_len": 16.6175,
      "eval_loss": 0.8652631044387817,
      "eval_runtime": 1333.9567,
      "eval_samples_per_second": 5.355,
      "eval_steps_per_second": 0.447,
      "step": 67000
    },
    {
      "epoch": 3.34,
      "learning_rate": 3.32953870520689e-05,
      "loss": 0.4403,
      "step": 67500
    },
    {
      "epoch": 3.34,
      "eval_bleu": 48.9108,
      "eval_gen_len": 16.7733,
      "eval_loss": 0.8616347312927246,
      "eval_runtime": 1343.5713,
      "eval_samples_per_second": 5.316,
      "eval_steps_per_second": 0.444,
      "step": 67500
    },
    {
      "epoch": 3.37,
      "learning_rate": 3.317164917838052e-05,
      "loss": 0.4383,
      "step": 68000
    },
    {
      "epoch": 3.37,
      "eval_bleu": 48.7693,
      "eval_gen_len": 16.8501,
      "eval_loss": 0.8552740812301636,
      "eval_runtime": 1342.1211,
      "eval_samples_per_second": 5.322,
      "eval_steps_per_second": 0.444,
      "step": 68000
    },
    {
      "epoch": 3.39,
      "learning_rate": 3.304791130469214e-05,
      "loss": 0.4354,
      "step": 68500
    },
    {
      "epoch": 3.39,
      "eval_bleu": 48.6325,
      "eval_gen_len": 16.7572,
      "eval_loss": 0.8553012013435364,
      "eval_runtime": 1337.7018,
      "eval_samples_per_second": 5.34,
      "eval_steps_per_second": 0.446,
      "step": 68500
    },
    {
      "epoch": 3.42,
      "learning_rate": 3.292417343100376e-05,
      "loss": 0.4316,
      "step": 69000
    },
    {
      "epoch": 3.42,
      "eval_bleu": 48.7044,
      "eval_gen_len": 16.7796,
      "eval_loss": 0.8564686179161072,
      "eval_runtime": 1336.7625,
      "eval_samples_per_second": 5.344,
      "eval_steps_per_second": 0.446,
      "step": 69000
    },
    {
      "epoch": 3.44,
      "learning_rate": 3.2800435557315385e-05,
      "loss": 0.4465,
      "step": 69500
    },
    {
      "epoch": 3.44,
      "eval_bleu": 48.5715,
      "eval_gen_len": 16.6865,
      "eval_loss": 0.8632897138595581,
      "eval_runtime": 1336.3648,
      "eval_samples_per_second": 5.345,
      "eval_steps_per_second": 0.446,
      "step": 69500
    },
    {
      "epoch": 3.46,
      "learning_rate": 3.267669768362701e-05,
      "loss": 0.4412,
      "step": 70000
    },
    {
      "epoch": 3.46,
      "eval_bleu": 48.18,
      "eval_gen_len": 16.6885,
      "eval_loss": 0.8553540706634521,
      "eval_runtime": 1331.2155,
      "eval_samples_per_second": 5.366,
      "eval_steps_per_second": 0.448,
      "step": 70000
    }
  ],
  "max_steps": 202040,
  "num_train_epochs": 10,
  "total_flos": 2.2754599026819072e+17,
  "trial_name": null,
  "trial_params": null
}