{
  "best_metric": 0.6402289867401123,
  "best_model_checkpoint": "nllb_200_distilled_1.3B_ENtoFO_bsz_64_epochs_10lr0.0001/checkpoint-7500",
  "epoch": 2.635885382917266,
  "eval_steps": 500,
  "global_step": 7500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.035145138438896883,
      "grad_norm": 0.7466627955436707,
      "learning_rate": 2e-05,
      "loss": 1.2984,
      "step": 100
    },
    {
      "epoch": 0.07029027687779377,
      "grad_norm": 0.8146196603775024,
      "learning_rate": 4e-05,
      "loss": 0.9825,
      "step": 200
    },
    {
      "epoch": 0.10543541531669065,
      "grad_norm": 0.8888100981712341,
      "learning_rate": 6e-05,
      "loss": 0.9552,
      "step": 300
    },
    {
      "epoch": 0.14058055375558753,
      "grad_norm": 0.7786069512367249,
      "learning_rate": 8e-05,
      "loss": 0.9249,
      "step": 400
    },
    {
      "epoch": 0.17572569219448442,
      "grad_norm": 0.7248039245605469,
      "learning_rate": 0.0001,
      "loss": 0.9121,
      "step": 500
    },
    {
      "epoch": 0.17572569219448442,
      "eval_bleu": 39.6289,
      "eval_chrf++": 58.7253,
      "eval_gen_len": 17.6964,
      "eval_loss": 0.8153083920478821,
      "eval_runtime": 3428.0218,
      "eval_samples_per_second": 2.136,
      "eval_steps_per_second": 1.068,
      "step": 500
    },
    {
      "epoch": 0.2108708306333813,
      "grad_norm": 0.6579689979553223,
      "learning_rate": 9.964221824686941e-05,
      "loss": 0.9033,
      "step": 600
    },
    {
      "epoch": 0.24601596907227818,
      "grad_norm": 0.7566829919815063,
      "learning_rate": 9.928443649373882e-05,
      "loss": 0.89,
      "step": 700
    },
    {
      "epoch": 0.28116110751117507,
      "grad_norm": 0.750449001789093,
      "learning_rate": 9.892665474060824e-05,
      "loss": 0.8709,
      "step": 800
    },
    {
      "epoch": 0.31630624595007195,
      "grad_norm": 0.6458595395088196,
      "learning_rate": 9.856887298747764e-05,
      "loss": 0.855,
      "step": 900
    },
    {
      "epoch": 0.35145138438896883,
      "grad_norm": 0.6504934430122375,
      "learning_rate": 9.821109123434705e-05,
      "loss": 0.8558,
      "step": 1000
    },
    {
      "epoch": 0.35145138438896883,
      "eval_bleu": 41.3868,
      "eval_chrf++": 60.2285,
      "eval_gen_len": 17.8309,
      "eval_loss": 0.7573392987251282,
      "eval_runtime": 3438.4799,
      "eval_samples_per_second": 2.129,
      "eval_steps_per_second": 1.065,
      "step": 1000
    },
    {
      "epoch": 0.3865965228278657,
      "grad_norm": 0.7932488918304443,
      "learning_rate": 9.785330948121646e-05,
      "loss": 0.8455,
      "step": 1100
    },
    {
      "epoch": 0.4217416612667626,
      "grad_norm": 0.7324668765068054,
      "learning_rate": 9.749552772808587e-05,
      "loss": 0.8363,
      "step": 1200
    },
    {
      "epoch": 0.4568867997056595,
      "grad_norm": 0.7606379985809326,
      "learning_rate": 9.713774597495528e-05,
      "loss": 0.816,
      "step": 1300
    },
    {
      "epoch": 0.49203193814455637,
      "grad_norm": 0.5376294851303101,
      "learning_rate": 9.677996422182469e-05,
      "loss": 0.8058,
      "step": 1400
    },
    {
      "epoch": 0.5271770765834533,
      "grad_norm": 0.903544008731842,
      "learning_rate": 9.64221824686941e-05,
      "loss": 0.8106,
      "step": 1500
    },
    {
      "epoch": 0.5271770765834533,
      "eval_bleu": 41.8145,
      "eval_chrf++": 60.6355,
      "eval_gen_len": 17.7981,
      "eval_loss": 0.7231032252311707,
      "eval_runtime": 3422.1307,
      "eval_samples_per_second": 2.139,
      "eval_steps_per_second": 1.07,
      "step": 1500
    },
    {
      "epoch": 0.5623222150223501,
      "grad_norm": 0.6048879027366638,
      "learning_rate": 9.606440071556351e-05,
      "loss": 0.7833,
      "step": 1600
    },
    {
      "epoch": 0.597467353461247,
      "grad_norm": 0.6456710696220398,
      "learning_rate": 9.570661896243292e-05,
      "loss": 0.7913,
      "step": 1700
    },
    {
      "epoch": 0.6326124919001439,
      "grad_norm": 0.6244344711303711,
      "learning_rate": 9.534883720930233e-05,
      "loss": 0.7991,
      "step": 1800
    },
    {
      "epoch": 0.6677576303390408,
      "grad_norm": 0.6789991855621338,
      "learning_rate": 9.499105545617174e-05,
      "loss": 0.7764,
      "step": 1900
    },
    {
      "epoch": 0.7029027687779377,
      "grad_norm": 0.7175530195236206,
      "learning_rate": 9.463327370304115e-05,
      "loss": 0.782,
      "step": 2000
    },
    {
      "epoch": 0.7029027687779377,
      "eval_bleu": 42.7477,
      "eval_chrf++": 61.3889,
      "eval_gen_len": 17.7144,
      "eval_loss": 0.6981974244117737,
      "eval_runtime": 3442.4579,
      "eval_samples_per_second": 2.127,
      "eval_steps_per_second": 1.063,
      "step": 2000
    },
    {
      "epoch": 0.7380479072168346,
      "grad_norm": 0.5626235604286194,
      "learning_rate": 9.427549194991055e-05,
      "loss": 0.7717,
      "step": 2100
    },
    {
      "epoch": 0.7731930456557314,
      "grad_norm": 0.6301568746566772,
      "learning_rate": 9.391771019677997e-05,
      "loss": 0.7697,
      "step": 2200
    },
    {
      "epoch": 0.8083381840946283,
      "grad_norm": 0.5354183316230774,
      "learning_rate": 9.355992844364938e-05,
      "loss": 0.7659,
      "step": 2300
    },
    {
      "epoch": 0.8434833225335252,
      "grad_norm": 0.6053968071937561,
      "learning_rate": 9.320214669051879e-05,
      "loss": 0.7588,
      "step": 2400
    },
    {
      "epoch": 0.8786284609724221,
      "grad_norm": 0.593941330909729,
      "learning_rate": 9.284436493738819e-05,
      "loss": 0.7488,
      "step": 2500
    },
    {
      "epoch": 0.8786284609724221,
      "eval_bleu": 43.5398,
      "eval_chrf++": 61.9091,
      "eval_gen_len": 17.8354,
      "eval_loss": 0.6773364543914795,
      "eval_runtime": 3438.2409,
      "eval_samples_per_second": 2.129,
      "eval_steps_per_second": 1.065,
      "step": 2500
    },
    {
      "epoch": 0.913773599411319,
      "grad_norm": 0.5975068211555481,
      "learning_rate": 9.248658318425761e-05,
      "loss": 0.7571,
      "step": 2600
    },
    {
      "epoch": 0.9489187378502159,
      "grad_norm": 0.5520368218421936,
      "learning_rate": 9.212880143112702e-05,
      "loss": 0.7398,
      "step": 2700
    },
    {
      "epoch": 0.9840638762891127,
      "grad_norm": 0.5748527646064758,
      "learning_rate": 9.177101967799643e-05,
      "loss": 0.7455,
      "step": 2800
    },
    {
      "epoch": 1.0192090147280095,
      "grad_norm": 0.6073290705680847,
      "learning_rate": 9.141323792486583e-05,
      "loss": 0.6714,
      "step": 2900
    },
    {
      "epoch": 1.0543541531669065,
      "grad_norm": 0.6117560863494873,
      "learning_rate": 9.105545617173525e-05,
      "loss": 0.6163,
      "step": 3000
    },
    {
      "epoch": 1.0543541531669065,
      "eval_bleu": 43.8509,
      "eval_chrf++": 62.1772,
      "eval_gen_len": 17.8905,
      "eval_loss": 0.6679165959358215,
      "eval_runtime": 3455.7087,
      "eval_samples_per_second": 2.119,
      "eval_steps_per_second": 1.059,
      "step": 3000
    },
    {
      "epoch": 1.0894992916058033,
      "grad_norm": 0.5141863822937012,
      "learning_rate": 9.069767441860465e-05,
      "loss": 0.6151,
      "step": 3100
    },
    {
      "epoch": 1.1246444300447003,
      "grad_norm": 0.5742676258087158,
      "learning_rate": 9.033989266547407e-05,
      "loss": 0.6236,
      "step": 3200
    },
    {
      "epoch": 1.159789568483597,
      "grad_norm": 0.5287165641784668,
      "learning_rate": 8.998211091234347e-05,
      "loss": 0.6249,
      "step": 3300
    },
    {
      "epoch": 1.194934706922494,
      "grad_norm": 0.5631851553916931,
      "learning_rate": 8.962432915921289e-05,
      "loss": 0.6229,
      "step": 3400
    },
    {
      "epoch": 1.2300798453613908,
      "grad_norm": 0.5420289039611816,
      "learning_rate": 8.926654740608229e-05,
      "loss": 0.6103,
      "step": 3500
    },
    {
      "epoch": 1.2300798453613908,
      "eval_bleu": 44.0794,
      "eval_chrf++": 62.3925,
      "eval_gen_len": 17.7716,
      "eval_loss": 0.6606557369232178,
      "eval_runtime": 3446.6004,
      "eval_samples_per_second": 2.124,
      "eval_steps_per_second": 1.062,
      "step": 3500
    },
    {
      "epoch": 1.2652249838002878,
      "grad_norm": 0.576799750328064,
      "learning_rate": 8.890876565295171e-05,
      "loss": 0.6076,
      "step": 3600
    },
    {
      "epoch": 1.3003701222391846,
      "grad_norm": 0.4976769983768463,
      "learning_rate": 8.855098389982111e-05,
      "loss": 0.6084,
      "step": 3700
    },
    {
      "epoch": 1.3355152606780816,
      "grad_norm": 0.5788149237632751,
      "learning_rate": 8.819320214669053e-05,
      "loss": 0.6121,
      "step": 3800
    },
    {
      "epoch": 1.3706603991169783,
      "grad_norm": 0.5019386410713196,
      "learning_rate": 8.783542039355993e-05,
      "loss": 0.6172,
      "step": 3900
    },
    {
      "epoch": 1.4058055375558753,
      "grad_norm": 0.5911116600036621,
      "learning_rate": 8.747763864042933e-05,
      "loss": 0.6037,
      "step": 4000
    },
    {
      "epoch": 1.4058055375558753,
      "eval_bleu": 44.0555,
      "eval_chrf++": 62.3227,
      "eval_gen_len": 17.7384,
      "eval_loss": 0.6558951139450073,
      "eval_runtime": 3448.9332,
      "eval_samples_per_second": 2.123,
      "eval_steps_per_second": 1.061,
      "step": 4000
    },
    {
      "epoch": 1.4409506759947721,
      "grad_norm": 0.5771309733390808,
      "learning_rate": 8.711985688729875e-05,
      "loss": 0.6067,
      "step": 4100
    },
    {
      "epoch": 1.476095814433669,
      "grad_norm": 0.6425995230674744,
      "learning_rate": 8.676207513416817e-05,
      "loss": 0.6097,
      "step": 4200
    },
    {
      "epoch": 1.5112409528725659,
      "grad_norm": 0.5513516664505005,
      "learning_rate": 8.640429338103757e-05,
      "loss": 0.6084,
      "step": 4300
    },
    {
      "epoch": 1.5463860913114629,
      "grad_norm": 0.5649448037147522,
      "learning_rate": 8.604651162790697e-05,
      "loss": 0.6026,
      "step": 4400
    },
    {
      "epoch": 1.5815312297503596,
      "grad_norm": 0.6140225529670715,
      "learning_rate": 8.568872987477639e-05,
      "loss": 0.6074,
      "step": 4500
    },
    {
      "epoch": 1.5815312297503596,
      "eval_bleu": 43.6701,
      "eval_chrf++": 62.1631,
      "eval_gen_len": 17.6893,
      "eval_loss": 0.6474220156669617,
      "eval_runtime": 3434.705,
      "eval_samples_per_second": 2.131,
      "eval_steps_per_second": 1.066,
      "step": 4500
    },
    {
      "epoch": 1.6166763681892564,
      "grad_norm": 0.5787105560302734,
      "learning_rate": 8.533094812164581e-05,
      "loss": 0.6167,
      "step": 4600
    },
    {
      "epoch": 1.6518215066281534,
      "grad_norm": 0.5308918356895447,
      "learning_rate": 8.497316636851521e-05,
      "loss": 0.5936,
      "step": 4700
    },
    {
      "epoch": 1.6869666450670504,
      "grad_norm": 0.5810381770133972,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.6074,
      "step": 4800
    },
    {
      "epoch": 1.7221117835059472,
      "grad_norm": 0.6304420232772827,
      "learning_rate": 8.425760286225403e-05,
      "loss": 0.5995,
      "step": 4900
    },
    {
      "epoch": 1.757256921944844,
      "grad_norm": 0.6314705610275269,
      "learning_rate": 8.389982110912343e-05,
      "loss": 0.6064,
      "step": 5000
    },
    {
      "epoch": 1.757256921944844,
      "eval_bleu": 44.2084,
      "eval_chrf++": 62.587,
      "eval_gen_len": 17.7677,
      "eval_loss": 0.6390019059181213,
      "eval_runtime": 3462.9771,
      "eval_samples_per_second": 2.114,
      "eval_steps_per_second": 1.057,
      "step": 5000
    },
    {
      "epoch": 1.792402060383741,
      "grad_norm": 0.5468895435333252,
      "learning_rate": 8.354203935599285e-05,
      "loss": 0.6014,
      "step": 5100
    },
    {
      "epoch": 1.827547198822638,
      "grad_norm": 0.564479649066925,
      "learning_rate": 8.318425760286225e-05,
      "loss": 0.5984,
      "step": 5200
    },
    {
      "epoch": 1.8626923372615347,
      "grad_norm": 0.6068270206451416,
      "learning_rate": 8.282647584973167e-05,
      "loss": 0.5995,
      "step": 5300
    },
    {
      "epoch": 1.8978374757004315,
      "grad_norm": 0.5385338068008423,
      "learning_rate": 8.246869409660107e-05,
      "loss": 0.6023,
      "step": 5400
    },
    {
      "epoch": 1.9329826141393285,
      "grad_norm": 0.5280515551567078,
      "learning_rate": 8.211091234347049e-05,
      "loss": 0.5969,
      "step": 5500
    },
    {
      "epoch": 1.9329826141393285,
      "eval_bleu": 44.5351,
      "eval_chrf++": 62.758,
      "eval_gen_len": 17.7998,
      "eval_loss": 0.6315435767173767,
      "eval_runtime": 3432.7408,
      "eval_samples_per_second": 2.133,
      "eval_steps_per_second": 1.066,
      "step": 5500
    },
    {
      "epoch": 1.9681277525782255,
      "grad_norm": 0.5488588213920593,
      "learning_rate": 8.17531305903399e-05,
      "loss": 0.5845,
      "step": 5600
    },
    {
      "epoch": 2.0032728910171222,
      "grad_norm": 0.49017634987831116,
      "learning_rate": 8.139534883720931e-05,
      "loss": 0.59,
      "step": 5700
    },
    {
      "epoch": 2.038418029456019,
      "grad_norm": 0.5274912118911743,
      "learning_rate": 8.103756708407871e-05,
      "loss": 0.4667,
      "step": 5800
    },
    {
      "epoch": 2.0735631678949162,
      "grad_norm": 1.4353556632995605,
      "learning_rate": 8.067978533094812e-05,
      "loss": 0.4706,
      "step": 5900
    },
    {
      "epoch": 2.108708306333813,
      "grad_norm": 0.5296390056610107,
      "learning_rate": 8.032200357781753e-05,
      "loss": 0.4697,
      "step": 6000
    },
    {
      "epoch": 2.108708306333813,
      "eval_bleu": 44.1342,
      "eval_chrf++": 62.5401,
      "eval_gen_len": 17.7677,
      "eval_loss": 0.6582108736038208,
      "eval_runtime": 3425.6987,
      "eval_samples_per_second": 2.137,
      "eval_steps_per_second": 1.069,
      "step": 6000
    },
    {
      "epoch": 2.14385344477271,
      "grad_norm": 0.6416345238685608,
      "learning_rate": 7.996422182468695e-05,
      "loss": 0.4882,
      "step": 6100
    },
    {
      "epoch": 2.1789985832116066,
      "grad_norm": 0.5234227180480957,
      "learning_rate": 7.960644007155635e-05,
      "loss": 0.4835,
      "step": 6200
    },
    {
      "epoch": 2.2141437216505038,
      "grad_norm": 0.4757489860057831,
      "learning_rate": 7.924865831842576e-05,
      "loss": 0.4771,
      "step": 6300
    },
    {
      "epoch": 2.2492888600894005,
      "grad_norm": 0.5438205599784851,
      "learning_rate": 7.889087656529517e-05,
      "loss": 0.4829,
      "step": 6400
    },
    {
      "epoch": 2.2844339985282973,
      "grad_norm": 0.5392005443572998,
      "learning_rate": 7.853309481216459e-05,
      "loss": 0.474,
      "step": 6500
    },
    {
      "epoch": 2.2844339985282973,
      "eval_bleu": 44.2923,
      "eval_chrf++": 62.5586,
      "eval_gen_len": 17.7492,
      "eval_loss": 0.6481789350509644,
      "eval_runtime": 3422.6338,
      "eval_samples_per_second": 2.139,
      "eval_steps_per_second": 1.07,
      "step": 6500
    },
    {
      "epoch": 2.319579136967194,
      "grad_norm": 0.5046759843826294,
      "learning_rate": 7.8175313059034e-05,
      "loss": 0.4802,
      "step": 6600
    },
    {
      "epoch": 2.354724275406091,
      "grad_norm": 0.49111250042915344,
      "learning_rate": 7.78175313059034e-05,
      "loss": 0.4916,
      "step": 6700
    },
    {
      "epoch": 2.389869413844988,
      "grad_norm": 0.5712496638298035,
      "learning_rate": 7.745974955277281e-05,
      "loss": 0.4845,
      "step": 6800
    },
    {
      "epoch": 2.425014552283885,
      "grad_norm": 0.6314510703086853,
      "learning_rate": 7.710196779964223e-05,
      "loss": 0.484,
      "step": 6900
    },
    {
      "epoch": 2.4601596907227816,
      "grad_norm": 0.6166778802871704,
      "learning_rate": 7.674418604651163e-05,
      "loss": 0.488,
      "step": 7000
    },
    {
      "epoch": 2.4601596907227816,
      "eval_bleu": 44.7709,
      "eval_chrf++": 62.9298,
      "eval_gen_len": 17.7547,
      "eval_loss": 0.6452430486679077,
      "eval_runtime": 3438.1607,
      "eval_samples_per_second": 2.129,
      "eval_steps_per_second": 1.065,
      "step": 7000
    },
    {
      "epoch": 2.495304829161679,
      "grad_norm": 0.5143587589263916,
      "learning_rate": 7.638640429338104e-05,
      "loss": 0.4875,
      "step": 7100
    },
    {
      "epoch": 2.5304499676005756,
      "grad_norm": 0.5172815322875977,
      "learning_rate": 7.602862254025045e-05,
      "loss": 0.4805,
      "step": 7200
    },
    {
      "epoch": 2.5655951060394724,
      "grad_norm": 0.49376818537712097,
      "learning_rate": 7.567084078711986e-05,
      "loss": 0.488,
      "step": 7300
    },
    {
      "epoch": 2.600740244478369,
      "grad_norm": 0.5714296102523804,
      "learning_rate": 7.531305903398927e-05,
      "loss": 0.4893,
      "step": 7400
    },
    {
      "epoch": 2.635885382917266,
      "grad_norm": 0.47455132007598877,
      "learning_rate": 7.495527728085868e-05,
      "loss": 0.4767,
      "step": 7500
    },
    {
      "epoch": 2.635885382917266,
      "eval_bleu": 44.8961,
      "eval_chrf++": 63.0641,
      "eval_gen_len": 17.7427,
      "eval_loss": 0.6402289867401123,
      "eval_runtime": 3447.7653,
      "eval_samples_per_second": 2.123,
      "eval_steps_per_second": 1.062,
      "step": 7500
    }
  ],
  "logging_steps": 100,
  "max_steps": 28450,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 1500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.268447098428719e+18,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}