{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.999912914743534,
  "global_step": 31578,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17,
      "learning_rate": 4.9419279907084783e-05,
      "loss": 0.0381,
      "step": 500
    },
    {
      "epoch": 0.35,
      "learning_rate": 4.883855981416957e-05,
      "loss": 0.0241,
      "step": 1000
    },
    {
      "epoch": 0.52,
      "learning_rate": 4.825783972125435e-05,
      "loss": 0.022,
      "step": 1500
    },
    {
      "epoch": 0.7,
      "learning_rate": 4.767711962833915e-05,
      "loss": 0.0207,
      "step": 2000
    },
    {
      "epoch": 0.87,
      "learning_rate": 4.709639953542393e-05,
      "loss": 0.0199,
      "step": 2500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9726582157682948,
      "eval_f1_HIRIQ": 0.7791055764551668,
      "eval_f1_HOLAM": 0.6713693875265927,
      "eval_f1_KUBUTZ": 0.7119160348309771,
      "eval_f1_PATACH": 0.851339663936741,
      "eval_f1_SHADDA": 0.8292565235143462,
      "eval_f1_SHVA": 0.8089031685917613,
      "eval_f1_TSERE": 0.7485289696344972,
      "eval_loss": 0.013756499625742435,
      "eval_macro_f1": 0.7714884749271546,
      "eval_micro_f1": 0.8119978846741942,
      "eval_runtime": 51.0176,
      "eval_samples_per_second": 750.232,
      "eval_steps_per_second": 7.507,
      "step": 2870
    },
    {
      "epoch": 1.05,
      "learning_rate": 4.6515679442508716e-05,
      "loss": 0.0189,
      "step": 3000
    },
    {
      "epoch": 1.22,
      "learning_rate": 4.59349593495935e-05,
      "loss": 0.0177,
      "step": 3500
    },
    {
      "epoch": 1.39,
      "learning_rate": 4.5354239256678285e-05,
      "loss": 0.0174,
      "step": 4000
    },
    {
      "epoch": 1.57,
      "learning_rate": 4.4773519163763066e-05,
      "loss": 0.0172,
      "step": 4500
    },
    {
      "epoch": 1.74,
      "learning_rate": 4.4192799070847854e-05,
      "loss": 0.0167,
      "step": 5000
    },
    {
      "epoch": 1.92,
      "learning_rate": 4.361207897793264e-05,
      "loss": 0.0166,
      "step": 5500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9761089815978242,
      "eval_f1_HIRIQ": 0.8106474124231328,
      "eval_f1_HOLAM": 0.7086283185840708,
      "eval_f1_KUBUTZ": 0.7472128945601075,
      "eval_f1_PATACH": 0.8701410253039187,
      "eval_f1_SHADDA": 0.8522099185169709,
      "eval_f1_SHVA": 0.8329891346774823,
      "eval_f1_TSERE": 0.7836281031964952,
      "eval_loss": 0.012357393279671669,
      "eval_macro_f1": 0.8007795438945969,
      "eval_micro_f1": 0.8363289062525998,
      "eval_runtime": 52.8127,
      "eval_samples_per_second": 724.731,
      "eval_steps_per_second": 7.252,
      "step": 5741
    },
    {
      "epoch": 2.09,
      "learning_rate": 4.303135888501742e-05,
      "loss": 0.0157,
      "step": 6000
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.245063879210221e-05,
      "loss": 0.0151,
      "step": 6500
    },
    {
      "epoch": 2.44,
      "learning_rate": 4.186991869918699e-05,
      "loss": 0.015,
      "step": 7000
    },
    {
      "epoch": 2.61,
      "learning_rate": 4.128919860627178e-05,
      "loss": 0.0148,
      "step": 7500
    },
    {
      "epoch": 2.79,
      "learning_rate": 4.070847851335656e-05,
      "loss": 0.0148,
      "step": 8000
    },
    {
      "epoch": 2.96,
      "learning_rate": 4.012775842044135e-05,
      "loss": 0.0147,
      "step": 8500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9780662605569959,
      "eval_f1_HIRIQ": 0.8294876571557304,
      "eval_f1_HOLAM": 0.7245056654076871,
      "eval_f1_KUBUTZ": 0.7711184824965379,
      "eval_f1_PATACH": 0.8810396563640023,
      "eval_f1_SHADDA": 0.8664105216491798,
      "eval_f1_SHVA": 0.8441380510188548,
      "eval_f1_TSERE": 0.7970168612191959,
      "eval_loss": 0.011536195874214172,
      "eval_macro_f1": 0.8162452707587411,
      "eval_micro_f1": 0.8498705967528521,
      "eval_runtime": 54.0777,
      "eval_samples_per_second": 707.778,
      "eval_steps_per_second": 7.082,
      "step": 8612
    },
    {
      "epoch": 3.14,
      "learning_rate": 3.9547038327526136e-05,
      "loss": 0.0137,
      "step": 9000
    },
    {
      "epoch": 3.31,
      "learning_rate": 3.8966318234610924e-05,
      "loss": 0.0134,
      "step": 9500
    },
    {
      "epoch": 3.48,
      "learning_rate": 3.8385598141695705e-05,
      "loss": 0.0135,
      "step": 10000
    },
    {
      "epoch": 3.66,
      "learning_rate": 3.780487804878049e-05,
      "loss": 0.0135,
      "step": 10500
    },
    {
      "epoch": 3.83,
      "learning_rate": 3.7224157955865274e-05,
      "loss": 0.0132,
      "step": 11000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9792273312868799,
      "eval_f1_HIRIQ": 0.8373507327062703,
      "eval_f1_HOLAM": 0.7442917433178227,
      "eval_f1_KUBUTZ": 0.7827533265097235,
      "eval_f1_PATACH": 0.8867928661237257,
      "eval_f1_SHADDA": 0.8725204092940786,
      "eval_f1_SHVA": 0.8508812530764879,
      "eval_f1_TSERE": 0.801735428945645,
      "eval_loss": 0.010950990952551365,
      "eval_macro_f1": 0.8251893942819649,
      "eval_micro_f1": 0.8567964095958783,
      "eval_runtime": 52.4162,
      "eval_samples_per_second": 730.213,
      "eval_steps_per_second": 7.307,
      "step": 11483
    },
    {
      "epoch": 4.01,
      "learning_rate": 3.664343786295006e-05,
      "loss": 0.0131,
      "step": 11500
    },
    {
      "epoch": 4.18,
      "learning_rate": 3.606271777003484e-05,
      "loss": 0.0122,
      "step": 12000
    },
    {
      "epoch": 4.35,
      "learning_rate": 3.548199767711963e-05,
      "loss": 0.0122,
      "step": 12500
    },
    {
      "epoch": 4.53,
      "learning_rate": 3.490127758420442e-05,
      "loss": 0.0123,
      "step": 13000
    },
    {
      "epoch": 4.7,
      "learning_rate": 3.43205574912892e-05,
      "loss": 0.0122,
      "step": 13500
    },
    {
      "epoch": 4.88,
      "learning_rate": 3.373983739837399e-05,
      "loss": 0.0122,
      "step": 14000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9799653269288884,
      "eval_f1_HIRIQ": 0.8439961236895427,
      "eval_f1_HOLAM": 0.7499999999999999,
      "eval_f1_KUBUTZ": 0.7890396727485228,
      "eval_f1_PATACH": 0.8902453557194496,
      "eval_f1_SHADDA": 0.8791103582869978,
      "eval_f1_SHVA": 0.8568982423275623,
      "eval_f1_TSERE": 0.8086938522045125,
      "eval_loss": 0.010681645944714546,
      "eval_macro_f1": 0.8311405149966554,
      "eval_micro_f1": 0.8620291101643662,
      "eval_runtime": 49.3354,
      "eval_samples_per_second": 775.812,
      "eval_steps_per_second": 7.763,
      "step": 14353
    },
    {
      "epoch": 5.05,
      "learning_rate": 3.315911730545877e-05,
      "loss": 0.012,
      "step": 14500
    },
    {
      "epoch": 5.23,
      "learning_rate": 3.2578397212543556e-05,
      "loss": 0.011,
      "step": 15000
    },
    {
      "epoch": 5.4,
      "learning_rate": 3.199767711962834e-05,
      "loss": 0.0114,
      "step": 15500
    },
    {
      "epoch": 5.57,
      "learning_rate": 3.1416957026713125e-05,
      "loss": 0.0112,
      "step": 16000
    },
    {
      "epoch": 5.75,
      "learning_rate": 3.083623693379791e-05,
      "loss": 0.0112,
      "step": 16500
    },
    {
      "epoch": 5.92,
      "learning_rate": 3.0255516840882698e-05,
      "loss": 0.0112,
      "step": 17000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9805248675902215,
      "eval_f1_HIRIQ": 0.8487574675668987,
      "eval_f1_HOLAM": 0.7595126122274476,
      "eval_f1_KUBUTZ": 0.797991035951556,
      "eval_f1_PATACH": 0.8928041018387551,
      "eval_f1_SHADDA": 0.8846275811897255,
      "eval_f1_SHVA": 0.8607482938928253,
      "eval_f1_TSERE": 0.8122762990085161,
      "eval_loss": 0.010463288053870201,
      "eval_macro_f1": 0.8366739130965322,
      "eval_micro_f1": 0.8659091900525827,
      "eval_runtime": 49.2152,
      "eval_samples_per_second": 777.707,
      "eval_steps_per_second": 7.782,
      "step": 17224
    },
    {
      "epoch": 6.1,
      "learning_rate": 2.9674796747967482e-05,
      "loss": 0.0107,
      "step": 17500
    },
    {
      "epoch": 6.27,
      "learning_rate": 2.9094076655052267e-05,
      "loss": 0.0105,
      "step": 18000
    },
    {
      "epoch": 6.44,
      "learning_rate": 2.851335656213705e-05,
      "loss": 0.0103,
      "step": 18500
    },
    {
      "epoch": 6.62,
      "learning_rate": 2.7932636469221835e-05,
      "loss": 0.0106,
      "step": 19000
    },
    {
      "epoch": 6.79,
      "learning_rate": 2.735191637630662e-05,
      "loss": 0.0105,
      "step": 19500
    },
    {
      "epoch": 6.97,
      "learning_rate": 2.6771196283391408e-05,
      "loss": 0.0104,
      "step": 20000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9811321234870294,
      "eval_f1_HIRIQ": 0.8550038223948401,
      "eval_f1_HOLAM": 0.763222830168437,
      "eval_f1_KUBUTZ": 0.8037383177570094,
      "eval_f1_PATACH": 0.8963010344751529,
      "eval_f1_SHADDA": 0.8854757929883139,
      "eval_f1_SHVA": 0.8649184871804224,
      "eval_f1_TSERE": 0.8171559407964675,
      "eval_loss": 0.010495145805180073,
      "eval_macro_f1": 0.8408308893943776,
      "eval_micro_f1": 0.8700826091477137,
      "eval_runtime": 52.1218,
      "eval_samples_per_second": 734.337,
      "eval_steps_per_second": 7.348,
      "step": 20095
    },
    {
      "epoch": 7.14,
      "learning_rate": 2.6190476190476192e-05,
      "loss": 0.0098,
      "step": 20500
    },
    {
      "epoch": 7.32,
      "learning_rate": 2.5609756097560977e-05,
      "loss": 0.0097,
      "step": 21000
    },
    {
      "epoch": 7.49,
      "learning_rate": 2.502903600464576e-05,
      "loss": 0.0097,
      "step": 21500
    },
    {
      "epoch": 7.66,
      "learning_rate": 2.4448315911730546e-05,
      "loss": 0.0098,
      "step": 22000
    },
    {
      "epoch": 7.84,
      "learning_rate": 2.3867595818815333e-05,
      "loss": 0.0097,
      "step": 22500
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.981536112480715,
      "eval_f1_HIRIQ": 0.8594682202461781,
      "eval_f1_HOLAM": 0.7635878626208735,
      "eval_f1_KUBUTZ": 0.8044461430664374,
      "eval_f1_PATACH": 0.8982397971622214,
      "eval_f1_SHADDA": 0.888189855746859,
      "eval_f1_SHVA": 0.8671149394978173,
      "eval_f1_TSERE": 0.8201351021404737,
      "eval_loss": 0.010434958152472973,
      "eval_macro_f1": 0.8430259886401229,
      "eval_micro_f1": 0.8725397588417411,
      "eval_runtime": 50.6576,
      "eval_samples_per_second": 755.562,
      "eval_steps_per_second": 7.561,
      "step": 22966
    },
    {
      "epoch": 8.01,
      "learning_rate": 2.3286875725900118e-05,
      "loss": 0.0096,
      "step": 23000
    },
    {
      "epoch": 8.19,
      "learning_rate": 2.2706155632984902e-05,
      "loss": 0.009,
      "step": 23500
    },
    {
      "epoch": 8.36,
      "learning_rate": 2.2125435540069687e-05,
      "loss": 0.0091,
      "step": 24000
    },
    {
      "epoch": 8.53,
      "learning_rate": 2.1544715447154475e-05,
      "loss": 0.0091,
      "step": 24500
    },
    {
      "epoch": 8.71,
      "learning_rate": 2.096399535423926e-05,
      "loss": 0.0092,
      "step": 25000
    },
    {
      "epoch": 8.88,
      "learning_rate": 2.038327526132404e-05,
      "loss": 0.0091,
      "step": 25500
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9816429946081784,
      "eval_f1_HIRIQ": 0.8617756604590732,
      "eval_f1_HOLAM": 0.7760372565622354,
      "eval_f1_KUBUTZ": 0.8087252634284247,
      "eval_f1_PATACH": 0.8981484778351434,
      "eval_f1_SHADDA": 0.8896060289509028,
      "eval_f1_SHVA": 0.8676933023775238,
      "eval_f1_TSERE": 0.8195844976309076,
      "eval_loss": 0.010549969039857388,
      "eval_macro_f1": 0.8459386410348871,
      "eval_micro_f1": 0.8734815959940515,
      "eval_runtime": 50.289,
      "eval_samples_per_second": 761.101,
      "eval_steps_per_second": 7.616,
      "step": 25836
    },
    {
      "epoch": 9.06,
      "learning_rate": 1.9802555168408828e-05,
      "loss": 0.0088,
      "step": 26000
    },
    {
      "epoch": 9.23,
      "learning_rate": 1.9221835075493612e-05,
      "loss": 0.0085,
      "step": 26500
    },
    {
      "epoch": 9.41,
      "learning_rate": 1.8641114982578397e-05,
      "loss": 0.0086,
      "step": 27000
    },
    {
      "epoch": 9.58,
      "learning_rate": 1.806039488966318e-05,
      "loss": 0.0085,
      "step": 27500
    },
    {
      "epoch": 9.75,
      "learning_rate": 1.747967479674797e-05,
      "loss": 0.0086,
      "step": 28000
    },
    {
      "epoch": 9.93,
      "learning_rate": 1.6898954703832754e-05,
      "loss": 0.0086,
      "step": 28500
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.9820323509296518,
      "eval_f1_HIRIQ": 0.8644184686296934,
      "eval_f1_HOLAM": 0.7807301793043854,
      "eval_f1_KUBUTZ": 0.8107445035070613,
      "eval_f1_PATACH": 0.9007614706453949,
      "eval_f1_SHADDA": 0.8914874936954775,
      "eval_f1_SHVA": 0.8700223746362294,
      "eval_f1_TSERE": 0.8237858288440186,
      "eval_loss": 0.010514745488762856,
      "eval_macro_f1": 0.8488500456088944,
      "eval_micro_f1": 0.876108330623615,
      "eval_runtime": 51.4328,
      "eval_samples_per_second": 744.175,
      "eval_steps_per_second": 7.447,
      "step": 28707
    },
    {
      "epoch": 10.1,
      "learning_rate": 1.6318234610917538e-05,
      "loss": 0.0083,
      "step": 29000
    },
    {
      "epoch": 10.28,
      "learning_rate": 1.5737514518002326e-05,
      "loss": 0.008,
      "step": 29500
    },
    {
      "epoch": 10.45,
      "learning_rate": 1.5156794425087109e-05,
      "loss": 0.0082,
      "step": 30000
    },
    {
      "epoch": 10.62,
      "learning_rate": 1.4576074332171893e-05,
      "loss": 0.0081,
      "step": 30500
    },
    {
      "epoch": 10.8,
      "learning_rate": 1.3995354239256678e-05,
      "loss": 0.0081,
      "step": 31000
    },
    {
      "epoch": 10.97,
      "learning_rate": 1.3414634146341466e-05,
      "loss": 0.0081,
      "step": 31500
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.9822652012787683,
      "eval_f1_HIRIQ": 0.8650168041552093,
      "eval_f1_HOLAM": 0.7809975223526877,
      "eval_f1_KUBUTZ": 0.8156180840973759,
      "eval_f1_PATACH": 0.9017915187001648,
      "eval_f1_SHADDA": 0.893456266013353,
      "eval_f1_SHVA": 0.8708873379860418,
      "eval_f1_TSERE": 0.824460194494808,
      "eval_loss": 0.010741644538939,
      "eval_macro_f1": 0.85031824682852,
      "eval_micro_f1": 0.8772879261954959,
      "eval_runtime": 49.6631,
      "eval_samples_per_second": 770.693,
      "eval_steps_per_second": 7.712,
      "step": 31578
    }
  ],
  "max_steps": 43050,
  "num_train_epochs": 15,
  "total_flos": 1.540842233529006e+17,
  "trial_name": null,
  "trial_params": null
}