{
  "best_metric": 0.6904399394989014,
  "best_model_checkpoint": "nllb_200_distilled_600M_ENtoFO_bsz_64_epochs_10lr7e-05/checkpoint-16500",
  "epoch": 5.798947842417985,
  "eval_steps": 500,
  "global_step": 16500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 1.4e-05,
      "loss": 1.6553,
      "step": 100
    },
    {
      "epoch": 0.07,
      "learning_rate": 2.8e-05,
      "loss": 1.3075,
      "step": 200
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.2e-05,
      "loss": 1.2606,
      "step": 300
    },
    {
      "epoch": 0.14,
      "learning_rate": 5.6e-05,
      "loss": 1.2085,
      "step": 400
    },
    {
      "epoch": 0.18,
      "learning_rate": 7e-05,
      "loss": 1.1956,
      "step": 500
    },
    {
      "epoch": 0.18,
      "eval_bleu": 36.3211,
      "eval_chrf++": 55.6095,
      "eval_gen_len": 17.5404,
      "eval_loss": 1.01542067527771,
      "eval_runtime": 1596.0286,
      "eval_samples_per_second": 4.587,
      "eval_steps_per_second": 2.294,
      "step": 500
    },
    {
      "epoch": 0.21,
      "learning_rate": 6.974955277280858e-05,
      "loss": 1.1487,
      "step": 600
    },
    {
      "epoch": 0.25,
      "learning_rate": 6.949910554561716e-05,
      "loss": 1.1406,
      "step": 700
    },
    {
      "epoch": 0.28,
      "learning_rate": 6.924865831842576e-05,
      "loss": 1.1197,
      "step": 800
    },
    {
      "epoch": 0.32,
      "learning_rate": 6.899821109123434e-05,
      "loss": 1.0739,
      "step": 900
    },
    {
      "epoch": 0.35,
      "learning_rate": 6.874776386404293e-05,
      "loss": 1.0817,
      "step": 1000
    },
    {
      "epoch": 0.35,
      "eval_bleu": 37.9118,
      "eval_chrf++": 57.1775,
      "eval_gen_len": 17.6499,
      "eval_loss": 0.9288437962532043,
      "eval_runtime": 1599.8831,
      "eval_samples_per_second": 4.576,
      "eval_steps_per_second": 2.288,
      "step": 1000
    },
    {
      "epoch": 0.39,
      "learning_rate": 6.849731663685151e-05,
      "loss": 1.0784,
      "step": 1100
    },
    {
      "epoch": 0.42,
      "learning_rate": 6.824686940966009e-05,
      "loss": 1.0538,
      "step": 1200
    },
    {
      "epoch": 0.46,
      "learning_rate": 6.799642218246869e-05,
      "loss": 1.0431,
      "step": 1300
    },
    {
      "epoch": 0.49,
      "learning_rate": 6.774597495527727e-05,
      "loss": 1.0347,
      "step": 1400
    },
    {
      "epoch": 0.53,
      "learning_rate": 6.749552772808586e-05,
      "loss": 1.0162,
      "step": 1500
    },
    {
      "epoch": 0.53,
      "eval_bleu": 39.3818,
      "eval_chrf++": 58.3559,
      "eval_gen_len": 17.6375,
      "eval_loss": 0.8730005025863647,
      "eval_runtime": 1591.8771,
      "eval_samples_per_second": 4.599,
      "eval_steps_per_second": 2.3,
      "step": 1500
    },
    {
      "epoch": 0.56,
      "learning_rate": 6.724508050089444e-05,
      "loss": 0.9995,
      "step": 1600
    },
    {
      "epoch": 0.6,
      "learning_rate": 6.699463327370304e-05,
      "loss": 0.987,
      "step": 1700
    },
    {
      "epoch": 0.63,
      "learning_rate": 6.674418604651162e-05,
      "loss": 1.0009,
      "step": 1800
    },
    {
      "epoch": 0.67,
      "learning_rate": 6.64937388193202e-05,
      "loss": 0.9843,
      "step": 1900
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.624329159212879e-05,
      "loss": 0.974,
      "step": 2000
    },
    {
      "epoch": 0.7,
      "eval_bleu": 40.0752,
      "eval_chrf++": 58.9288,
      "eval_gen_len": 17.741,
      "eval_loss": 0.8381767272949219,
      "eval_runtime": 1584.9965,
      "eval_samples_per_second": 4.619,
      "eval_steps_per_second": 2.31,
      "step": 2000
    },
    {
      "epoch": 0.74,
      "learning_rate": 6.599284436493739e-05,
      "loss": 0.9649,
      "step": 2100
    },
    {
      "epoch": 0.77,
      "learning_rate": 6.574239713774597e-05,
      "loss": 0.9511,
      "step": 2200
    },
    {
      "epoch": 0.81,
      "learning_rate": 6.549194991055455e-05,
      "loss": 0.9689,
      "step": 2300
    },
    {
      "epoch": 0.84,
      "learning_rate": 6.524150268336314e-05,
      "loss": 0.9512,
      "step": 2400
    },
    {
      "epoch": 0.88,
      "learning_rate": 6.499105545617173e-05,
      "loss": 0.9513,
      "step": 2500
    },
    {
      "epoch": 0.88,
      "eval_bleu": 40.6133,
      "eval_chrf++": 59.5241,
      "eval_gen_len": 17.7642,
      "eval_loss": 0.8131038546562195,
      "eval_runtime": 1608.2394,
      "eval_samples_per_second": 4.552,
      "eval_steps_per_second": 2.276,
      "step": 2500
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.474060822898032e-05,
      "loss": 0.9476,
      "step": 2600
    },
    {
      "epoch": 0.95,
      "learning_rate": 6.44901610017889e-05,
      "loss": 0.9202,
      "step": 2700
    },
    {
      "epoch": 0.98,
      "learning_rate": 6.423971377459748e-05,
      "loss": 0.9423,
      "step": 2800
    },
    {
      "epoch": 1.02,
      "learning_rate": 6.398926654740608e-05,
      "loss": 0.8949,
      "step": 2900
    },
    {
      "epoch": 1.05,
      "learning_rate": 6.373881932021467e-05,
      "loss": 0.8405,
      "step": 3000
    },
    {
      "epoch": 1.05,
      "eval_bleu": 40.8822,
      "eval_chrf++": 59.9339,
      "eval_gen_len": 17.7189,
      "eval_loss": 0.7998338937759399,
      "eval_runtime": 1592.0637,
      "eval_samples_per_second": 4.598,
      "eval_steps_per_second": 2.3,
      "step": 3000
    },
    {
      "epoch": 1.09,
      "learning_rate": 6.348837209302325e-05,
      "loss": 0.8431,
      "step": 3100
    },
    {
      "epoch": 1.12,
      "learning_rate": 6.323792486583183e-05,
      "loss": 0.8383,
      "step": 3200
    },
    {
      "epoch": 1.16,
      "learning_rate": 6.298747763864043e-05,
      "loss": 0.838,
      "step": 3300
    },
    {
      "epoch": 1.19,
      "learning_rate": 6.273703041144901e-05,
      "loss": 0.8527,
      "step": 3400
    },
    {
      "epoch": 1.23,
      "learning_rate": 6.24865831842576e-05,
      "loss": 0.8252,
      "step": 3500
    },
    {
      "epoch": 1.23,
      "eval_bleu": 41.6082,
      "eval_chrf++": 60.3254,
      "eval_gen_len": 17.7662,
      "eval_loss": 0.7859154939651489,
      "eval_runtime": 1613.9109,
      "eval_samples_per_second": 4.536,
      "eval_steps_per_second": 2.268,
      "step": 3500
    },
    {
      "epoch": 1.27,
      "learning_rate": 6.22361359570662e-05,
      "loss": 0.836,
      "step": 3600
    },
    {
      "epoch": 1.3,
      "learning_rate": 6.198568872987478e-05,
      "loss": 0.8274,
      "step": 3700
    },
    {
      "epoch": 1.34,
      "learning_rate": 6.173524150268336e-05,
      "loss": 0.8257,
      "step": 3800
    },
    {
      "epoch": 1.37,
      "learning_rate": 6.148479427549194e-05,
      "loss": 0.8301,
      "step": 3900
    },
    {
      "epoch": 1.41,
      "learning_rate": 6.123434704830053e-05,
      "loss": 0.8235,
      "step": 4000
    },
    {
      "epoch": 1.41,
      "eval_bleu": 41.9785,
      "eval_chrf++": 60.615,
      "eval_gen_len": 17.6996,
      "eval_loss": 0.7718562483787537,
      "eval_runtime": 1591.3834,
      "eval_samples_per_second": 4.6,
      "eval_steps_per_second": 2.301,
      "step": 4000
    },
    {
      "epoch": 1.44,
      "learning_rate": 6.098389982110912e-05,
      "loss": 0.8335,
      "step": 4100
    },
    {
      "epoch": 1.48,
      "learning_rate": 6.073345259391771e-05,
      "loss": 0.8184,
      "step": 4200
    },
    {
      "epoch": 1.51,
      "learning_rate": 6.048300536672629e-05,
      "loss": 0.8139,
      "step": 4300
    },
    {
      "epoch": 1.55,
      "learning_rate": 6.0232558139534877e-05,
      "loss": 0.8121,
      "step": 4400
    },
    {
      "epoch": 1.58,
      "learning_rate": 5.998211091234346e-05,
      "loss": 0.8174,
      "step": 4500
    },
    {
      "epoch": 1.58,
      "eval_bleu": 41.9942,
      "eval_chrf++": 60.7015,
      "eval_gen_len": 17.7548,
      "eval_loss": 0.7601388692855835,
      "eval_runtime": 1610.8686,
      "eval_samples_per_second": 4.545,
      "eval_steps_per_second": 2.273,
      "step": 4500
    },
    {
      "epoch": 1.62,
      "learning_rate": 5.973166368515206e-05,
      "loss": 0.8232,
      "step": 4600
    },
    {
      "epoch": 1.65,
      "learning_rate": 5.948121645796064e-05,
      "loss": 0.8099,
      "step": 4700
    },
    {
      "epoch": 1.69,
      "learning_rate": 5.9230769230769225e-05,
      "loss": 0.8154,
      "step": 4800
    },
    {
      "epoch": 1.72,
      "learning_rate": 5.898032200357781e-05,
      "loss": 0.818,
      "step": 4900
    },
    {
      "epoch": 1.76,
      "learning_rate": 5.87298747763864e-05,
      "loss": 0.7992,
      "step": 5000
    },
    {
      "epoch": 1.76,
      "eval_bleu": 42.3622,
      "eval_chrf++": 61.0481,
      "eval_gen_len": 17.8145,
      "eval_loss": 0.7486168146133423,
      "eval_runtime": 1597.5591,
      "eval_samples_per_second": 4.583,
      "eval_steps_per_second": 2.292,
      "step": 5000
    },
    {
      "epoch": 1.79,
      "learning_rate": 5.847942754919499e-05,
      "loss": 0.8087,
      "step": 5100
    },
    {
      "epoch": 1.83,
      "learning_rate": 5.822898032200357e-05,
      "loss": 0.7958,
      "step": 5200
    },
    {
      "epoch": 1.86,
      "learning_rate": 5.7978533094812156e-05,
      "loss": 0.8022,
      "step": 5300
    },
    {
      "epoch": 1.9,
      "learning_rate": 5.7728085867620747e-05,
      "loss": 0.7876,
      "step": 5400
    },
    {
      "epoch": 1.93,
      "learning_rate": 5.747763864042934e-05,
      "loss": 0.7915,
      "step": 5500
    },
    {
      "epoch": 1.93,
      "eval_bleu": 42.4851,
      "eval_chrf++": 61.1145,
      "eval_gen_len": 17.7756,
      "eval_loss": 0.7351738810539246,
      "eval_runtime": 1598.7533,
      "eval_samples_per_second": 4.579,
      "eval_steps_per_second": 2.29,
      "step": 5500
    },
    {
      "epoch": 1.97,
      "learning_rate": 5.722719141323792e-05,
      "loss": 0.7795,
      "step": 5600
    },
    {
      "epoch": 2.0,
      "learning_rate": 5.6976744186046504e-05,
      "loss": 0.8015,
      "step": 5700
    },
    {
      "epoch": 2.04,
      "learning_rate": 5.6726296958855094e-05,
      "loss": 0.7219,
      "step": 5800
    },
    {
      "epoch": 2.07,
      "learning_rate": 5.647584973166368e-05,
      "loss": 0.7231,
      "step": 5900
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.622540250447227e-05,
      "loss": 0.718,
      "step": 6000
    },
    {
      "epoch": 2.11,
      "eval_bleu": 42.5957,
      "eval_chrf++": 61.1828,
      "eval_gen_len": 17.7144,
      "eval_loss": 0.7349444627761841,
      "eval_runtime": 1593.2386,
      "eval_samples_per_second": 4.595,
      "eval_steps_per_second": 2.298,
      "step": 6000
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.597495527728085e-05,
      "loss": 0.7155,
      "step": 6100
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.572450805008944e-05,
      "loss": 0.7222,
      "step": 6200
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.5474060822898026e-05,
      "loss": 0.7113,
      "step": 6300
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.522361359570661e-05,
      "loss": 0.7067,
      "step": 6400
    },
    {
      "epoch": 2.28,
      "learning_rate": 5.497316636851521e-05,
      "loss": 0.714,
      "step": 6500
    },
    {
      "epoch": 2.28,
      "eval_bleu": 43.1947,
      "eval_chrf++": 61.6389,
      "eval_gen_len": 17.7485,
      "eval_loss": 0.7279652953147888,
      "eval_runtime": 1602.3425,
      "eval_samples_per_second": 4.569,
      "eval_steps_per_second": 2.285,
      "step": 6500
    },
    {
      "epoch": 2.32,
      "learning_rate": 5.472271914132379e-05,
      "loss": 0.7284,
      "step": 6600
    },
    {
      "epoch": 2.35,
      "learning_rate": 5.4472271914132374e-05,
      "loss": 0.7106,
      "step": 6700
    },
    {
      "epoch": 2.39,
      "learning_rate": 5.422182468694096e-05,
      "loss": 0.7226,
      "step": 6800
    },
    {
      "epoch": 2.43,
      "learning_rate": 5.3971377459749555e-05,
      "loss": 0.7151,
      "step": 6900
    },
    {
      "epoch": 2.46,
      "learning_rate": 5.372093023255814e-05,
      "loss": 0.7242,
      "step": 7000
    },
    {
      "epoch": 2.46,
      "eval_bleu": 43.0217,
      "eval_chrf++": 61.4,
      "eval_gen_len": 17.7472,
      "eval_loss": 0.7255465984344482,
      "eval_runtime": 1596.9259,
      "eval_samples_per_second": 4.584,
      "eval_steps_per_second": 2.293,
      "step": 7000
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.347048300536672e-05,
      "loss": 0.7115,
      "step": 7100
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.3220035778175306e-05,
      "loss": 0.6996,
      "step": 7200
    },
    {
      "epoch": 2.57,
      "learning_rate": 5.296958855098389e-05,
      "loss": 0.7226,
      "step": 7300
    },
    {
      "epoch": 2.6,
      "learning_rate": 5.2719141323792486e-05,
      "loss": 0.7023,
      "step": 7400
    },
    {
      "epoch": 2.64,
      "learning_rate": 5.246869409660107e-05,
      "loss": 0.7035,
      "step": 7500
    },
    {
      "epoch": 2.64,
      "eval_bleu": 42.9886,
      "eval_chrf++": 61.5585,
      "eval_gen_len": 17.7513,
      "eval_loss": 0.7192216515541077,
      "eval_runtime": 1608.427,
      "eval_samples_per_second": 4.552,
      "eval_steps_per_second": 2.276,
      "step": 7500
    },
    {
      "epoch": 2.67,
      "learning_rate": 5.2218246869409654e-05,
      "loss": 0.7175,
      "step": 7600
    },
    {
      "epoch": 2.71,
      "learning_rate": 5.1967799642218244e-05,
      "loss": 0.7164,
      "step": 7700
    },
    {
      "epoch": 2.74,
      "learning_rate": 5.171735241502683e-05,
      "loss": 0.703,
      "step": 7800
    },
    {
      "epoch": 2.78,
      "learning_rate": 5.146690518783542e-05,
      "loss": 0.7067,
      "step": 7900
    },
    {
      "epoch": 2.81,
      "learning_rate": 5.1216457960644e-05,
      "loss": 0.7048,
      "step": 8000
    },
    {
      "epoch": 2.81,
      "eval_bleu": 42.9399,
      "eval_chrf++": 61.4851,
      "eval_gen_len": 17.7067,
      "eval_loss": 0.7168448567390442,
      "eval_runtime": 1552.4929,
      "eval_samples_per_second": 4.716,
      "eval_steps_per_second": 2.358,
      "step": 8000
    },
    {
      "epoch": 2.85,
      "learning_rate": 5.096601073345259e-05,
      "loss": 0.7127,
      "step": 8100
    },
    {
      "epoch": 2.88,
      "learning_rate": 5.0715563506261176e-05,
      "loss": 0.7091,
      "step": 8200
    },
    {
      "epoch": 2.92,
      "learning_rate": 5.046511627906976e-05,
      "loss": 0.7122,
      "step": 8300
    },
    {
      "epoch": 2.95,
      "learning_rate": 5.021466905187835e-05,
      "loss": 0.6949,
      "step": 8400
    },
    {
      "epoch": 2.99,
      "learning_rate": 4.996422182468694e-05,
      "loss": 0.685,
      "step": 8500
    },
    {
      "epoch": 2.99,
      "eval_bleu": 43.114,
      "eval_chrf++": 61.6028,
      "eval_gen_len": 17.844,
      "eval_loss": 0.7094260454177856,
      "eval_runtime": 1415.2591,
      "eval_samples_per_second": 5.173,
      "eval_steps_per_second": 2.587,
      "step": 8500
    },
    {
      "epoch": 3.02,
      "learning_rate": 4.9713774597495524e-05,
      "loss": 0.6618,
      "step": 8600
    },
    {
      "epoch": 3.06,
      "learning_rate": 4.946332737030411e-05,
      "loss": 0.6417,
      "step": 8700
    },
    {
      "epoch": 3.09,
      "learning_rate": 4.92128801431127e-05,
      "loss": 0.65,
      "step": 8800
    },
    {
      "epoch": 3.13,
      "learning_rate": 4.896243291592129e-05,
      "loss": 0.6375,
      "step": 8900
    },
    {
      "epoch": 3.16,
      "learning_rate": 4.871198568872987e-05,
      "loss": 0.632,
      "step": 9000
    },
    {
      "epoch": 3.16,
      "eval_bleu": 43.3779,
      "eval_chrf++": 61.8915,
      "eval_gen_len": 17.7121,
      "eval_loss": 0.7186790108680725,
      "eval_runtime": 1408.2967,
      "eval_samples_per_second": 5.198,
      "eval_steps_per_second": 2.6,
      "step": 9000
    },
    {
      "epoch": 3.2,
      "learning_rate": 4.8461538461538455e-05,
      "loss": 0.6434,
      "step": 9100
    },
    {
      "epoch": 3.23,
      "learning_rate": 4.821109123434704e-05,
      "loss": 0.6354,
      "step": 9200
    },
    {
      "epoch": 3.27,
      "learning_rate": 4.7960644007155636e-05,
      "loss": 0.6374,
      "step": 9300
    },
    {
      "epoch": 3.3,
      "learning_rate": 4.771019677996422e-05,
      "loss": 0.6478,
      "step": 9400
    },
    {
      "epoch": 3.34,
      "learning_rate": 4.74597495527728e-05,
      "loss": 0.6444,
      "step": 9500
    },
    {
      "epoch": 3.34,
      "eval_bleu": 43.0761,
      "eval_chrf++": 61.6092,
      "eval_gen_len": 17.7518,
      "eval_loss": 0.7161450982093811,
      "eval_runtime": 1412.5727,
      "eval_samples_per_second": 5.183,
      "eval_steps_per_second": 2.592,
      "step": 9500
    },
    {
      "epoch": 3.37,
      "learning_rate": 4.720930232558139e-05,
      "loss": 0.6358,
      "step": 9600
    },
    {
      "epoch": 3.41,
      "learning_rate": 4.695885509838998e-05,
      "loss": 0.6502,
      "step": 9700
    },
    {
      "epoch": 3.44,
      "learning_rate": 4.670840787119857e-05,
      "loss": 0.6376,
      "step": 9800
    },
    {
      "epoch": 3.48,
      "learning_rate": 4.645796064400715e-05,
      "loss": 0.638,
      "step": 9900
    },
    {
      "epoch": 3.51,
      "learning_rate": 4.6207513416815735e-05,
      "loss": 0.6302,
      "step": 10000
    },
    {
      "epoch": 3.51,
      "eval_bleu": 43.4763,
      "eval_chrf++": 61.8105,
      "eval_gen_len": 17.7754,
      "eval_loss": 0.7070448398590088,
      "eval_runtime": 1409.243,
      "eval_samples_per_second": 5.195,
      "eval_steps_per_second": 2.598,
      "step": 10000
    },
    {
      "epoch": 3.55,
      "learning_rate": 4.5957066189624325e-05,
      "loss": 0.632,
      "step": 10100
    },
    {
      "epoch": 3.58,
      "learning_rate": 4.570661896243291e-05,
      "loss": 0.6364,
      "step": 10200
    },
    {
      "epoch": 3.62,
      "learning_rate": 4.54561717352415e-05,
      "loss": 0.6466,
      "step": 10300
    },
    {
      "epoch": 3.66,
      "learning_rate": 4.520572450805009e-05,
      "loss": 0.6373,
      "step": 10400
    },
    {
      "epoch": 3.69,
      "learning_rate": 4.495527728085867e-05,
      "loss": 0.6478,
      "step": 10500
    },
    {
      "epoch": 3.69,
      "eval_bleu": 43.725,
      "eval_chrf++": 62.0616,
      "eval_gen_len": 17.788,
      "eval_loss": 0.705007016658783,
      "eval_runtime": 1414.2328,
      "eval_samples_per_second": 5.177,
      "eval_steps_per_second": 2.589,
      "step": 10500
    },
    {
      "epoch": 3.73,
      "learning_rate": 4.470483005366726e-05,
      "loss": 0.6516,
      "step": 10600
    },
    {
      "epoch": 3.76,
      "learning_rate": 4.445438282647585e-05,
      "loss": 0.6334,
      "step": 10700
    },
    {
      "epoch": 3.8,
      "learning_rate": 4.420393559928444e-05,
      "loss": 0.6542,
      "step": 10800
    },
    {
      "epoch": 3.83,
      "learning_rate": 4.395348837209302e-05,
      "loss": 0.646,
      "step": 10900
    },
    {
      "epoch": 3.87,
      "learning_rate": 4.3703041144901605e-05,
      "loss": 0.6374,
      "step": 11000
    },
    {
      "epoch": 3.87,
      "eval_bleu": 43.7206,
      "eval_chrf++": 62.1048,
      "eval_gen_len": 17.7229,
      "eval_loss": 0.6963800191879272,
      "eval_runtime": 1413.4358,
      "eval_samples_per_second": 5.18,
      "eval_steps_per_second": 2.59,
      "step": 11000
    },
    {
      "epoch": 3.9,
      "learning_rate": 4.345259391771019e-05,
      "loss": 0.6416,
      "step": 11100
    },
    {
      "epoch": 3.94,
      "learning_rate": 4.3202146690518786e-05,
      "loss": 0.6249,
      "step": 11200
    },
    {
      "epoch": 3.97,
      "learning_rate": 4.295169946332737e-05,
      "loss": 0.638,
      "step": 11300
    },
    {
      "epoch": 4.01,
      "learning_rate": 4.270125223613595e-05,
      "loss": 0.6352,
      "step": 11400
    },
    {
      "epoch": 4.04,
      "learning_rate": 4.2450805008944536e-05,
      "loss": 0.5804,
      "step": 11500
    },
    {
      "epoch": 4.04,
      "eval_bleu": 43.8669,
      "eval_chrf++": 62.1364,
      "eval_gen_len": 17.7865,
      "eval_loss": 0.7024260759353638,
      "eval_runtime": 1410.9857,
      "eval_samples_per_second": 5.189,
      "eval_steps_per_second": 2.595,
      "step": 11500
    },
    {
      "epoch": 4.08,
      "learning_rate": 4.220035778175313e-05,
      "loss": 0.5799,
      "step": 11600
    },
    {
      "epoch": 4.11,
      "learning_rate": 4.194991055456172e-05,
      "loss": 0.5852,
      "step": 11700
    },
    {
      "epoch": 4.15,
      "learning_rate": 4.16994633273703e-05,
      "loss": 0.5801,
      "step": 11800
    },
    {
      "epoch": 4.18,
      "learning_rate": 4.1449016100178884e-05,
      "loss": 0.5945,
      "step": 11900
    },
    {
      "epoch": 4.22,
      "learning_rate": 4.1198568872987475e-05,
      "loss": 0.5919,
      "step": 12000
    },
    {
      "epoch": 4.22,
      "eval_bleu": 43.6775,
      "eval_chrf++": 61.9586,
      "eval_gen_len": 17.8369,
      "eval_loss": 0.7032192945480347,
      "eval_runtime": 1420.6308,
      "eval_samples_per_second": 5.153,
      "eval_steps_per_second": 2.577,
      "step": 12000
    },
    {
      "epoch": 4.25,
      "learning_rate": 4.094812164579606e-05,
      "loss": 0.5928,
      "step": 12100
    },
    {
      "epoch": 4.29,
      "learning_rate": 4.069767441860465e-05,
      "loss": 0.5923,
      "step": 12200
    },
    {
      "epoch": 4.32,
      "learning_rate": 4.044722719141323e-05,
      "loss": 0.5943,
      "step": 12300
    },
    {
      "epoch": 4.36,
      "learning_rate": 4.019677996422182e-05,
      "loss": 0.5905,
      "step": 12400
    },
    {
      "epoch": 4.39,
      "learning_rate": 3.9946332737030406e-05,
      "loss": 0.5879,
      "step": 12500
    },
    {
      "epoch": 4.39,
      "eval_bleu": 43.9642,
      "eval_chrf++": 62.2021,
      "eval_gen_len": 17.7873,
      "eval_loss": 0.7024480104446411,
      "eval_runtime": 1420.746,
      "eval_samples_per_second": 5.153,
      "eval_steps_per_second": 2.577,
      "step": 12500
    },
    {
      "epoch": 4.43,
      "learning_rate": 3.9695885509839e-05,
      "loss": 0.5855,
      "step": 12600
    },
    {
      "epoch": 4.46,
      "learning_rate": 3.944543828264758e-05,
      "loss": 0.5855,
      "step": 12700
    },
    {
      "epoch": 4.5,
      "learning_rate": 3.919499105545617e-05,
      "loss": 0.5833,
      "step": 12800
    },
    {
      "epoch": 4.53,
      "learning_rate": 3.8944543828264754e-05,
      "loss": 0.5767,
      "step": 12900
    },
    {
      "epoch": 4.57,
      "learning_rate": 3.869409660107334e-05,
      "loss": 0.5858,
      "step": 13000
    },
    {
      "epoch": 4.57,
      "eval_bleu": 44.027,
      "eval_chrf++": 62.2226,
      "eval_gen_len": 17.7678,
      "eval_loss": 0.6992958784103394,
      "eval_runtime": 1410.8973,
      "eval_samples_per_second": 5.189,
      "eval_steps_per_second": 2.595,
      "step": 13000
    },
    {
      "epoch": 4.6,
      "learning_rate": 3.8443649373881935e-05,
      "loss": 0.5775,
      "step": 13100
    },
    {
      "epoch": 4.64,
      "learning_rate": 3.819320214669052e-05,
      "loss": 0.5825,
      "step": 13200
    },
    {
      "epoch": 4.67,
      "learning_rate": 3.79427549194991e-05,
      "loss": 0.5851,
      "step": 13300
    },
    {
      "epoch": 4.71,
      "learning_rate": 3.7692307692307686e-05,
      "loss": 0.5913,
      "step": 13400
    },
    {
      "epoch": 4.74,
      "learning_rate": 3.744186046511627e-05,
      "loss": 0.5877,
      "step": 13500
    },
    {
      "epoch": 4.74,
      "eval_bleu": 44.1426,
      "eval_chrf++": 62.3429,
      "eval_gen_len": 17.7805,
      "eval_loss": 0.6957116723060608,
      "eval_runtime": 1414.8882,
      "eval_samples_per_second": 5.174,
      "eval_steps_per_second": 2.587,
      "step": 13500
    },
    {
      "epoch": 4.78,
      "learning_rate": 3.719141323792487e-05,
      "loss": 0.5857,
      "step": 13600
    },
    {
      "epoch": 4.81,
      "learning_rate": 3.694096601073345e-05,
      "loss": 0.5776,
      "step": 13700
    },
    {
      "epoch": 4.85,
      "learning_rate": 3.6690518783542034e-05,
      "loss": 0.5985,
      "step": 13800
    },
    {
      "epoch": 4.89,
      "learning_rate": 3.644007155635062e-05,
      "loss": 0.5855,
      "step": 13900
    },
    {
      "epoch": 4.92,
      "learning_rate": 3.618962432915921e-05,
      "loss": 0.5895,
      "step": 14000
    },
    {
      "epoch": 4.92,
      "eval_bleu": 44.2097,
      "eval_chrf++": 62.4158,
      "eval_gen_len": 17.7713,
      "eval_loss": 0.6944009065628052,
      "eval_runtime": 1417.3776,
      "eval_samples_per_second": 5.165,
      "eval_steps_per_second": 2.583,
      "step": 14000
    },
    {
      "epoch": 4.96,
      "learning_rate": 3.59391771019678e-05,
      "loss": 0.581,
      "step": 14100
    },
    {
      "epoch": 4.99,
      "learning_rate": 3.568872987477638e-05,
      "loss": 0.5835,
      "step": 14200
    },
    {
      "epoch": 5.03,
      "learning_rate": 3.543828264758497e-05,
      "loss": 0.5512,
      "step": 14300
    },
    {
      "epoch": 5.06,
      "learning_rate": 3.5187835420393556e-05,
      "loss": 0.5324,
      "step": 14400
    },
    {
      "epoch": 5.1,
      "learning_rate": 3.4937388193202146e-05,
      "loss": 0.5381,
      "step": 14500
    },
    {
      "epoch": 5.1,
      "eval_bleu": 43.9778,
      "eval_chrf++": 62.2087,
      "eval_gen_len": 17.8153,
      "eval_loss": 0.7013605833053589,
      "eval_runtime": 1419.0259,
      "eval_samples_per_second": 5.159,
      "eval_steps_per_second": 2.58,
      "step": 14500
    },
    {
      "epoch": 5.13,
      "learning_rate": 3.468694096601073e-05,
      "loss": 0.5364,
      "step": 14600
    },
    {
      "epoch": 5.17,
      "learning_rate": 3.443649373881932e-05,
      "loss": 0.543,
      "step": 14700
    },
    {
      "epoch": 5.2,
      "learning_rate": 3.4186046511627904e-05,
      "loss": 0.5432,
      "step": 14800
    },
    {
      "epoch": 5.24,
      "learning_rate": 3.3935599284436494e-05,
      "loss": 0.542,
      "step": 14900
    },
    {
      "epoch": 5.27,
      "learning_rate": 3.368515205724508e-05,
      "loss": 0.5385,
      "step": 15000
    },
    {
      "epoch": 5.27,
      "eval_bleu": 44.1326,
      "eval_chrf++": 62.3372,
      "eval_gen_len": 17.8174,
      "eval_loss": 0.7036887407302856,
      "eval_runtime": 1418.4921,
      "eval_samples_per_second": 5.161,
      "eval_steps_per_second": 2.581,
      "step": 15000
    },
    {
      "epoch": 5.31,
      "learning_rate": 3.343470483005367e-05,
      "loss": 0.5467,
      "step": 15100
    },
    {
      "epoch": 5.34,
      "learning_rate": 3.318425760286225e-05,
      "loss": 0.5466,
      "step": 15200
    },
    {
      "epoch": 5.38,
      "learning_rate": 3.2933810375670835e-05,
      "loss": 0.5439,
      "step": 15300
    },
    {
      "epoch": 5.41,
      "learning_rate": 3.2683363148479426e-05,
      "loss": 0.5403,
      "step": 15400
    },
    {
      "epoch": 5.45,
      "learning_rate": 3.243291592128801e-05,
      "loss": 0.5481,
      "step": 15500
    },
    {
      "epoch": 5.45,
      "eval_bleu": 44.053,
      "eval_chrf++": 62.418,
      "eval_gen_len": 17.7614,
      "eval_loss": 0.6976599097251892,
      "eval_runtime": 1417.644,
      "eval_samples_per_second": 5.164,
      "eval_steps_per_second": 2.582,
      "step": 15500
    },
    {
      "epoch": 5.48,
      "learning_rate": 3.21824686940966e-05,
      "loss": 0.5445,
      "step": 15600
    },
    {
      "epoch": 5.52,
      "learning_rate": 3.1932021466905183e-05,
      "loss": 0.5518,
      "step": 15700
    },
    {
      "epoch": 5.55,
      "learning_rate": 3.1681574239713774e-05,
      "loss": 0.5451,
      "step": 15800
    },
    {
      "epoch": 5.59,
      "learning_rate": 3.143112701252236e-05,
      "loss": 0.5387,
      "step": 15900
    },
    {
      "epoch": 5.62,
      "learning_rate": 3.118067978533094e-05,
      "loss": 0.5473,
      "step": 16000
    },
    {
      "epoch": 5.62,
      "eval_bleu": 44.2406,
      "eval_chrf++": 62.4882,
      "eval_gen_len": 17.8263,
      "eval_loss": 0.6982511878013611,
      "eval_runtime": 1415.9788,
      "eval_samples_per_second": 5.17,
      "eval_steps_per_second": 2.585,
      "step": 16000
    },
    {
      "epoch": 5.66,
      "learning_rate": 3.093023255813953e-05,
      "loss": 0.5483,
      "step": 16100
    },
    {
      "epoch": 5.69,
      "learning_rate": 3.0679785330948115e-05,
      "loss": 0.537,
      "step": 16200
    },
    {
      "epoch": 5.73,
      "learning_rate": 3.0429338103756705e-05,
      "loss": 0.5479,
      "step": 16300
    },
    {
      "epoch": 5.76,
      "learning_rate": 3.0178890876565292e-05,
      "loss": 0.5445,
      "step": 16400
    },
    {
      "epoch": 5.8,
      "learning_rate": 2.992844364937388e-05,
      "loss": 0.5466,
      "step": 16500
    },
    {
      "epoch": 5.8,
      "eval_bleu": 44.2692,
      "eval_chrf++": 62.4172,
      "eval_gen_len": 17.7783,
      "eval_loss": 0.6904399394989014,
      "eval_runtime": 1415.8621,
      "eval_samples_per_second": 5.171,
      "eval_steps_per_second": 2.586,
      "step": 16500
    }
  ],
  "logging_steps": 100,
  "max_steps": 28450,
  "num_train_epochs": 10,
  "save_steps": 1500,
  "total_flos": 2.2884516243072614e+18,
  "trial_name": null,
  "trial_params": null
}