{
  "best_metric": 0.15070733428001404,
  "best_model_checkpoint": "checkpoints_orfeo/checkpoint-65079",
  "epoch": 9.0,
  "global_step": 65079,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "learning_rate": 1.996542663532015e-05,
      "loss": 1.5229,
      "step": 500
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.99308532706403e-05,
      "loss": 0.7443,
      "step": 1000
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.989627990596045e-05,
      "loss": 0.5936,
      "step": 1500
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.986177568800996e-05,
      "loss": 0.508,
      "step": 2000
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.982720232333011e-05,
      "loss": 0.4597,
      "step": 2500
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.9792628958650257e-05,
      "loss": 0.4274,
      "step": 3000
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.9758055593970406e-05,
      "loss": 0.3898,
      "step": 3500
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.9723482229290554e-05,
      "loss": 0.3645,
      "step": 4000
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.9688978011340064e-05,
      "loss": 0.3486,
      "step": 4500
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.9654404646660216e-05,
      "loss": 0.337,
      "step": 5000
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.9619831281980364e-05,
      "loss": 0.3175,
      "step": 5500
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.9585257917300513e-05,
      "loss": 0.3015,
      "step": 6000
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.955068455262066e-05,
      "loss": 0.2914,
      "step": 6500
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.9516111187940813e-05,
      "loss": 0.2843,
      "step": 7000
    },
    {
      "epoch": 1.0,
      "eval_bleu": 83.675,
      "eval_gen_len": 14.2653,
      "eval_loss": 0.24304039776325226,
      "eval_runtime": 556.3105,
      "eval_samples_per_second": 51.762,
      "eval_steps_per_second": 1.618,
      "step": 7231
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.948160696999032e-05,
      "loss": 0.2721,
      "step": 7500
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.944703360531047e-05,
      "loss": 0.2533,
      "step": 8000
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.941246024063062e-05,
      "loss": 0.2493,
      "step": 8500
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.937788687595077e-05,
      "loss": 0.2444,
      "step": 9000
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.9343313511270917e-05,
      "loss": 0.2432,
      "step": 9500
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.930874014659107e-05,
      "loss": 0.2329,
      "step": 10000
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.9274166781911217e-05,
      "loss": 0.2318,
      "step": 10500
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.9239662563960727e-05,
      "loss": 0.2296,
      "step": 11000
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.9205089199280876e-05,
      "loss": 0.2297,
      "step": 11500
    },
    {
      "epoch": 1.66,
      "learning_rate": 1.9170515834601024e-05,
      "loss": 0.2226,
      "step": 12000
    },
    {
      "epoch": 1.73,
      "learning_rate": 1.9135942469921173e-05,
      "loss": 0.2214,
      "step": 12500
    },
    {
      "epoch": 1.8,
      "learning_rate": 1.9101369105241325e-05,
      "loss": 0.2193,
      "step": 13000
    },
    {
      "epoch": 1.87,
      "learning_rate": 1.9066795740561473e-05,
      "loss": 0.2141,
      "step": 13500
    },
    {
      "epoch": 1.94,
      "learning_rate": 1.903222237588162e-05,
      "loss": 0.2121,
      "step": 14000
    },
    {
      "epoch": 2.0,
      "eval_bleu": 86.0981,
      "eval_gen_len": 14.1714,
      "eval_loss": 0.19072701036930084,
      "eval_runtime": 554.6671,
      "eval_samples_per_second": 51.916,
      "eval_steps_per_second": 1.623,
      "step": 14462
    },
    {
      "epoch": 2.01,
      "learning_rate": 1.899764901120177e-05,
      "loss": 0.207,
      "step": 14500
    },
    {
      "epoch": 2.07,
      "learning_rate": 1.896314479325128e-05,
      "loss": 0.1895,
      "step": 15000
    },
    {
      "epoch": 2.14,
      "learning_rate": 1.892857142857143e-05,
      "loss": 0.1903,
      "step": 15500
    },
    {
      "epoch": 2.21,
      "learning_rate": 1.889399806389158e-05,
      "loss": 0.1883,
      "step": 16000
    },
    {
      "epoch": 2.28,
      "learning_rate": 1.885942469921173e-05,
      "loss": 0.1911,
      "step": 16500
    },
    {
      "epoch": 2.35,
      "learning_rate": 1.8824851334531877e-05,
      "loss": 0.1912,
      "step": 17000
    },
    {
      "epoch": 2.42,
      "learning_rate": 1.8790277969852026e-05,
      "loss": 0.1856,
      "step": 17500
    },
    {
      "epoch": 2.49,
      "learning_rate": 1.8755773751901536e-05,
      "loss": 0.1836,
      "step": 18000
    },
    {
      "epoch": 2.56,
      "learning_rate": 1.8721200387221684e-05,
      "loss": 0.1842,
      "step": 18500
    },
    {
      "epoch": 2.63,
      "learning_rate": 1.8686627022541836e-05,
      "loss": 0.1789,
      "step": 19000
    },
    {
      "epoch": 2.7,
      "learning_rate": 1.8652053657861984e-05,
      "loss": 0.1826,
      "step": 19500
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.8617480293182133e-05,
      "loss": 0.1793,
      "step": 20000
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.8582976075231643e-05,
      "loss": 0.1747,
      "step": 20500
    },
    {
      "epoch": 2.9,
      "learning_rate": 1.854840271055179e-05,
      "loss": 0.1725,
      "step": 21000
    },
    {
      "epoch": 2.97,
      "learning_rate": 1.851382934587194e-05,
      "loss": 0.1768,
      "step": 21500
    },
    {
      "epoch": 3.0,
      "eval_bleu": 87.1351,
      "eval_gen_len": 14.2257,
      "eval_loss": 0.17269779741764069,
      "eval_runtime": 554.1036,
      "eval_samples_per_second": 51.969,
      "eval_steps_per_second": 1.624,
      "step": 21693
    },
    {
      "epoch": 3.04,
      "learning_rate": 1.847925598119209e-05,
      "loss": 0.1662,
      "step": 22000
    },
    {
      "epoch": 3.11,
      "learning_rate": 1.844468261651224e-05,
      "loss": 0.164,
      "step": 22500
    },
    {
      "epoch": 3.18,
      "learning_rate": 1.841010925183239e-05,
      "loss": 0.1614,
      "step": 23000
    },
    {
      "epoch": 3.25,
      "learning_rate": 1.837553588715254e-05,
      "loss": 0.1604,
      "step": 23500
    },
    {
      "epoch": 3.32,
      "learning_rate": 1.834096252247269e-05,
      "loss": 0.155,
      "step": 24000
    },
    {
      "epoch": 3.39,
      "learning_rate": 1.8306389157792837e-05,
      "loss": 0.1574,
      "step": 24500
    },
    {
      "epoch": 3.46,
      "learning_rate": 1.8271884939842347e-05,
      "loss": 0.1593,
      "step": 25000
    },
    {
      "epoch": 3.53,
      "learning_rate": 1.82373115751625e-05,
      "loss": 0.1573,
      "step": 25500
    },
    {
      "epoch": 3.6,
      "learning_rate": 1.8202738210482648e-05,
      "loss": 0.1561,
      "step": 26000
    },
    {
      "epoch": 3.66,
      "learning_rate": 1.8168164845802796e-05,
      "loss": 0.1544,
      "step": 26500
    },
    {
      "epoch": 3.73,
      "learning_rate": 1.8133591481122945e-05,
      "loss": 0.1562,
      "step": 27000
    },
    {
      "epoch": 3.8,
      "learning_rate": 1.809908726317245e-05,
      "loss": 0.1596,
      "step": 27500
    },
    {
      "epoch": 3.87,
      "learning_rate": 1.806458304522196e-05,
      "loss": 0.1513,
      "step": 28000
    },
    {
      "epoch": 3.94,
      "learning_rate": 1.8030009680542113e-05,
      "loss": 0.1576,
      "step": 28500
    },
    {
      "epoch": 4.0,
      "eval_bleu": 87.7394,
      "eval_gen_len": 14.2568,
      "eval_loss": 0.1625259518623352,
      "eval_runtime": 551.0375,
      "eval_samples_per_second": 52.258,
      "eval_steps_per_second": 1.633,
      "step": 28924
    },
    {
      "epoch": 4.01,
      "learning_rate": 1.799543631586226e-05,
      "loss": 0.1499,
      "step": 29000
    },
    {
      "epoch": 4.08,
      "learning_rate": 1.796086295118241e-05,
      "loss": 0.1397,
      "step": 29500
    },
    {
      "epoch": 4.15,
      "learning_rate": 1.792635873323192e-05,
      "loss": 0.1391,
      "step": 30000
    },
    {
      "epoch": 4.22,
      "learning_rate": 1.7891785368552068e-05,
      "loss": 0.1415,
      "step": 30500
    },
    {
      "epoch": 4.29,
      "learning_rate": 1.7857212003872216e-05,
      "loss": 0.1384,
      "step": 31000
    },
    {
      "epoch": 4.36,
      "learning_rate": 1.7822638639192368e-05,
      "loss": 0.1397,
      "step": 31500
    },
    {
      "epoch": 4.43,
      "learning_rate": 1.7788065274512517e-05,
      "loss": 0.1431,
      "step": 32000
    },
    {
      "epoch": 4.49,
      "learning_rate": 1.7753561056562027e-05,
      "loss": 0.1398,
      "step": 32500
    },
    {
      "epoch": 4.56,
      "learning_rate": 1.7718987691882175e-05,
      "loss": 0.1387,
      "step": 33000
    },
    {
      "epoch": 4.63,
      "learning_rate": 1.7684414327202324e-05,
      "loss": 0.142,
      "step": 33500
    },
    {
      "epoch": 4.7,
      "learning_rate": 1.7649840962522472e-05,
      "loss": 0.1393,
      "step": 34000
    },
    {
      "epoch": 4.77,
      "learning_rate": 1.7615267597842624e-05,
      "loss": 0.1381,
      "step": 34500
    },
    {
      "epoch": 4.84,
      "learning_rate": 1.7580694233162772e-05,
      "loss": 0.141,
      "step": 35000
    },
    {
      "epoch": 4.91,
      "learning_rate": 1.7546120868482924e-05,
      "loss": 0.1369,
      "step": 35500
    },
    {
      "epoch": 4.98,
      "learning_rate": 1.7511547503803073e-05,
      "loss": 0.1361,
      "step": 36000
    },
    {
      "epoch": 5.0,
      "eval_bleu": 88.1298,
      "eval_gen_len": 14.2962,
      "eval_loss": 0.1563238501548767,
      "eval_runtime": 552.3181,
      "eval_samples_per_second": 52.137,
      "eval_steps_per_second": 1.629,
      "step": 36155
    },
    {
      "epoch": 5.05,
      "learning_rate": 1.747697413912322e-05,
      "loss": 0.1286,
      "step": 36500
    },
    {
      "epoch": 5.12,
      "learning_rate": 1.7442469921172728e-05,
      "loss": 0.1253,
      "step": 37000
    },
    {
      "epoch": 5.19,
      "learning_rate": 1.740789655649288e-05,
      "loss": 0.1278,
      "step": 37500
    },
    {
      "epoch": 5.26,
      "learning_rate": 1.7373323191813028e-05,
      "loss": 0.1258,
      "step": 38000
    },
    {
      "epoch": 5.32,
      "learning_rate": 1.733874982713318e-05,
      "loss": 0.127,
      "step": 38500
    },
    {
      "epoch": 5.39,
      "learning_rate": 1.7304245609182686e-05,
      "loss": 0.1242,
      "step": 39000
    },
    {
      "epoch": 5.46,
      "learning_rate": 1.7269672244502838e-05,
      "loss": 0.1224,
      "step": 39500
    },
    {
      "epoch": 5.53,
      "learning_rate": 1.7235098879822987e-05,
      "loss": 0.1267,
      "step": 40000
    },
    {
      "epoch": 5.6,
      "learning_rate": 1.7200594661872493e-05,
      "loss": 0.1275,
      "step": 40500
    },
    {
      "epoch": 5.67,
      "learning_rate": 1.716602129719264e-05,
      "loss": 0.1277,
      "step": 41000
    },
    {
      "epoch": 5.74,
      "learning_rate": 1.7131447932512794e-05,
      "loss": 0.1285,
      "step": 41500
    },
    {
      "epoch": 5.81,
      "learning_rate": 1.7096874567832945e-05,
      "loss": 0.126,
      "step": 42000
    },
    {
      "epoch": 5.88,
      "learning_rate": 1.7062301203153094e-05,
      "loss": 0.1246,
      "step": 42500
    },
    {
      "epoch": 5.95,
      "learning_rate": 1.7027727838473242e-05,
      "loss": 0.1256,
      "step": 43000
    },
    {
      "epoch": 6.0,
      "eval_bleu": 88.3004,
      "eval_gen_len": 14.2442,
      "eval_loss": 0.15259724855422974,
      "eval_runtime": 550.8856,
      "eval_samples_per_second": 52.272,
      "eval_steps_per_second": 1.634,
      "step": 43386
    },
    {
      "epoch": 6.02,
      "learning_rate": 1.699315447379339e-05,
      "loss": 0.1258,
      "step": 43500
    },
    {
      "epoch": 6.08,
      "learning_rate": 1.695858110911354e-05,
      "loss": 0.112,
      "step": 44000
    },
    {
      "epoch": 6.15,
      "learning_rate": 1.692407689116305e-05,
      "loss": 0.1127,
      "step": 44500
    },
    {
      "epoch": 6.22,
      "learning_rate": 1.68895035264832e-05,
      "loss": 0.1139,
      "step": 45000
    },
    {
      "epoch": 6.29,
      "learning_rate": 1.685493016180335e-05,
      "loss": 0.1163,
      "step": 45500
    },
    {
      "epoch": 6.36,
      "learning_rate": 1.6820356797123498e-05,
      "loss": 0.1157,
      "step": 46000
    },
    {
      "epoch": 6.43,
      "learning_rate": 1.6785783432443647e-05,
      "loss": 0.1142,
      "step": 46500
    },
    {
      "epoch": 6.5,
      "learning_rate": 1.67512100677638e-05,
      "loss": 0.1123,
      "step": 47000
    },
    {
      "epoch": 6.57,
      "learning_rate": 1.6716636703083947e-05,
      "loss": 0.1135,
      "step": 47500
    },
    {
      "epoch": 6.64,
      "learning_rate": 1.6682063338404095e-05,
      "loss": 0.1146,
      "step": 48000
    },
    {
      "epoch": 6.71,
      "learning_rate": 1.6647489973724244e-05,
      "loss": 0.1159,
      "step": 48500
    },
    {
      "epoch": 6.78,
      "learning_rate": 1.6612916609044392e-05,
      "loss": 0.1157,
      "step": 49000
    },
    {
      "epoch": 6.85,
      "learning_rate": 1.6578343244364544e-05,
      "loss": 0.1148,
      "step": 49500
    },
    {
      "epoch": 6.91,
      "learning_rate": 1.6543769879684693e-05,
      "loss": 0.1139,
      "step": 50000
    },
    {
      "epoch": 6.98,
      "learning_rate": 1.6509265661734203e-05,
      "loss": 0.1164,
      "step": 50500
    },
    {
      "epoch": 7.0,
      "eval_bleu": 88.4796,
      "eval_gen_len": 14.2928,
      "eval_loss": 0.15147347748279572,
      "eval_runtime": 552.236,
      "eval_samples_per_second": 52.144,
      "eval_steps_per_second": 1.63,
      "step": 50617
    },
    {
      "epoch": 7.05,
      "learning_rate": 1.647469229705435e-05,
      "loss": 0.1055,
      "step": 51000
    },
    {
      "epoch": 7.12,
      "learning_rate": 1.64401189323745e-05,
      "loss": 0.1028,
      "step": 51500
    },
    {
      "epoch": 7.19,
      "learning_rate": 1.640561471442401e-05,
      "loss": 0.1055,
      "step": 52000
    },
    {
      "epoch": 7.26,
      "learning_rate": 1.6371041349744158e-05,
      "loss": 0.105,
      "step": 52500
    },
    {
      "epoch": 7.33,
      "learning_rate": 1.6336537131793668e-05,
      "loss": 0.1052,
      "step": 53000
    },
    {
      "epoch": 7.4,
      "learning_rate": 1.6301963767113816e-05,
      "loss": 0.1063,
      "step": 53500
    },
    {
      "epoch": 7.47,
      "learning_rate": 1.6267390402433968e-05,
      "loss": 0.1048,
      "step": 54000
    },
    {
      "epoch": 7.54,
      "learning_rate": 1.6232817037754117e-05,
      "loss": 0.1073,
      "step": 54500
    },
    {
      "epoch": 7.61,
      "learning_rate": 1.6198243673074265e-05,
      "loss": 0.106,
      "step": 55000
    },
    {
      "epoch": 7.68,
      "learning_rate": 1.6163670308394413e-05,
      "loss": 0.1036,
      "step": 55500
    },
    {
      "epoch": 7.74,
      "learning_rate": 1.6129096943714565e-05,
      "loss": 0.1057,
      "step": 56000
    },
    {
      "epoch": 7.81,
      "learning_rate": 1.6094592725764072e-05,
      "loss": 0.1051,
      "step": 56500
    },
    {
      "epoch": 7.88,
      "learning_rate": 1.6060019361084224e-05,
      "loss": 0.1057,
      "step": 57000
    },
    {
      "epoch": 7.95,
      "learning_rate": 1.6025445996404372e-05,
      "loss": 0.1051,
      "step": 57500
    },
    {
      "epoch": 8.0,
      "eval_bleu": 88.7364,
      "eval_gen_len": 14.3991,
      "eval_loss": 0.1514560878276825,
      "eval_runtime": 559.4528,
      "eval_samples_per_second": 51.472,
      "eval_steps_per_second": 1.609,
      "step": 57848
    },
    {
      "epoch": 8.02,
      "learning_rate": 1.599087263172452e-05,
      "loss": 0.103,
      "step": 58000
    },
    {
      "epoch": 8.09,
      "learning_rate": 1.595629926704467e-05,
      "loss": 0.0925,
      "step": 58500
    },
    {
      "epoch": 8.16,
      "learning_rate": 1.592172590236482e-05,
      "loss": 0.0942,
      "step": 59000
    },
    {
      "epoch": 8.23,
      "learning_rate": 1.588715253768497e-05,
      "loss": 0.0965,
      "step": 59500
    },
    {
      "epoch": 8.3,
      "learning_rate": 1.585264831973448e-05,
      "loss": 0.0952,
      "step": 60000
    },
    {
      "epoch": 8.37,
      "learning_rate": 1.5818074955054628e-05,
      "loss": 0.0967,
      "step": 60500
    },
    {
      "epoch": 8.44,
      "learning_rate": 1.5783501590374776e-05,
      "loss": 0.0957,
      "step": 61000
    },
    {
      "epoch": 8.51,
      "learning_rate": 1.5748928225694925e-05,
      "loss": 0.0938,
      "step": 61500
    },
    {
      "epoch": 8.57,
      "learning_rate": 1.5714354861015077e-05,
      "loss": 0.0981,
      "step": 62000
    },
    {
      "epoch": 8.64,
      "learning_rate": 1.5679781496335225e-05,
      "loss": 0.1008,
      "step": 62500
    },
    {
      "epoch": 8.71,
      "learning_rate": 1.5645208131655374e-05,
      "loss": 0.0987,
      "step": 63000
    },
    {
      "epoch": 8.78,
      "learning_rate": 1.5610703913704883e-05,
      "loss": 0.0985,
      "step": 63500
    },
    {
      "epoch": 8.85,
      "learning_rate": 1.5576130549025032e-05,
      "loss": 0.097,
      "step": 64000
    },
    {
      "epoch": 8.92,
      "learning_rate": 1.554155718434518e-05,
      "loss": 0.0985,
      "step": 64500
    },
    {
      "epoch": 8.99,
      "learning_rate": 1.5506983819665332e-05,
      "loss": 0.0966,
      "step": 65000
    },
    {
      "epoch": 9.0,
      "eval_bleu": 88.826,
      "eval_gen_len": 14.3624,
      "eval_loss": 0.15070733428001404,
      "eval_runtime": 553.1909,
      "eval_samples_per_second": 52.054,
      "eval_steps_per_second": 1.627,
      "step": 65079
    }
  ],
  "max_steps": 289240,
  "num_train_epochs": 40,
  "total_flos": 2.004229422148485e+17,
  "trial_name": null,
  "trial_params": null
}