{
  "best_metric": 0.3463841676712036,
  "best_model_checkpoint": "/kaggle/working/hubert-base960-agu-amharic/checkpoint-1500",
  "epoch": 3.9893617021276597,
  "eval_steps": 500,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.13297872340425532,
      "grad_norm": 3.2855160236358643,
      "learning_rate": 9.73404255319149e-06,
      "loss": 1.4894,
      "step": 100
    },
    {
      "epoch": 0.26595744680851063,
      "grad_norm": 14.008848190307617,
      "learning_rate": 9.470744680851066e-06,
      "loss": 1.1355,
      "step": 200
    },
    {
      "epoch": 0.39893617021276595,
      "grad_norm": 5.032841205596924,
      "learning_rate": 9.204787234042554e-06,
      "loss": 0.7905,
      "step": 300
    },
    {
      "epoch": 0.5319148936170213,
      "grad_norm": 6.931905746459961,
      "learning_rate": 8.938829787234043e-06,
      "loss": 0.6621,
      "step": 400
    },
    {
      "epoch": 0.6648936170212766,
      "grad_norm": 5.367379188537598,
      "learning_rate": 8.675531914893619e-06,
      "loss": 0.502,
      "step": 500
    },
    {
      "epoch": 0.6648936170212766,
      "eval_accuracy": 0.8707025647163391,
      "eval_loss": 0.4583953320980072,
      "eval_runtime": 60.5694,
      "eval_samples_per_second": 22.09,
      "eval_steps_per_second": 2.774,
      "step": 500
    },
    {
      "epoch": 0.7978723404255319,
      "grad_norm": 11.638015747070312,
      "learning_rate": 8.412234042553192e-06,
      "loss": 0.4986,
      "step": 600
    },
    {
      "epoch": 0.9308510638297872,
      "grad_norm": 21.74812126159668,
      "learning_rate": 8.14627659574468e-06,
      "loss": 0.4095,
      "step": 700
    },
    {
      "epoch": 1.0638297872340425,
      "grad_norm": 13.848979949951172,
      "learning_rate": 7.880319148936171e-06,
      "loss": 0.3986,
      "step": 800
    },
    {
      "epoch": 1.196808510638298,
      "grad_norm": 18.350629806518555,
      "learning_rate": 7.61436170212766e-06,
      "loss": 0.3427,
      "step": 900
    },
    {
      "epoch": 1.3297872340425532,
      "grad_norm": 8.771578788757324,
      "learning_rate": 7.348404255319149e-06,
      "loss": 0.3239,
      "step": 1000
    },
    {
      "epoch": 1.3297872340425532,
      "eval_accuracy": 0.9013453125953674,
      "eval_loss": 0.3637482821941376,
      "eval_runtime": 60.6579,
      "eval_samples_per_second": 22.058,
      "eval_steps_per_second": 2.77,
      "step": 1000
    },
    {
      "epoch": 1.4627659574468086,
      "grad_norm": 20.235013961791992,
      "learning_rate": 7.0824468085106394e-06,
      "loss": 0.2807,
      "step": 1100
    },
    {
      "epoch": 1.5957446808510638,
      "grad_norm": 2.6776235103607178,
      "learning_rate": 6.816489361702127e-06,
      "loss": 0.2626,
      "step": 1200
    },
    {
      "epoch": 1.728723404255319,
      "grad_norm": 1.6725181341171265,
      "learning_rate": 6.550531914893618e-06,
      "loss": 0.2898,
      "step": 1300
    },
    {
      "epoch": 1.8617021276595744,
      "grad_norm": 10.663431167602539,
      "learning_rate": 6.284574468085107e-06,
      "loss": 0.2627,
      "step": 1400
    },
    {
      "epoch": 1.9946808510638299,
      "grad_norm": 34.09892654418945,
      "learning_rate": 6.018617021276596e-06,
      "loss": 0.2329,
      "step": 1500
    },
    {
      "epoch": 1.9946808510638299,
      "eval_accuracy": 0.9147982001304626,
      "eval_loss": 0.3463841676712036,
      "eval_runtime": 60.4518,
      "eval_samples_per_second": 22.133,
      "eval_steps_per_second": 2.779,
      "step": 1500
    },
    {
      "epoch": 2.127659574468085,
      "grad_norm": 0.13858869671821594,
      "learning_rate": 5.752659574468086e-06,
      "loss": 0.1826,
      "step": 1600
    },
    {
      "epoch": 2.2606382978723403,
      "grad_norm": 12.44304084777832,
      "learning_rate": 5.4867021276595745e-06,
      "loss": 0.2282,
      "step": 1700
    },
    {
      "epoch": 2.393617021276596,
      "grad_norm": 47.84838104248047,
      "learning_rate": 5.220744680851064e-06,
      "loss": 0.1477,
      "step": 1800
    },
    {
      "epoch": 2.526595744680851,
      "grad_norm": 9.419229507446289,
      "learning_rate": 4.954787234042554e-06,
      "loss": 0.204,
      "step": 1900
    },
    {
      "epoch": 2.6595744680851063,
      "grad_norm": 10.298186302185059,
      "learning_rate": 4.6888297872340425e-06,
      "loss": 0.1768,
      "step": 2000
    },
    {
      "epoch": 2.6595744680851063,
      "eval_accuracy": 0.9125560522079468,
      "eval_loss": 0.40557414293289185,
      "eval_runtime": 60.529,
      "eval_samples_per_second": 22.105,
      "eval_steps_per_second": 2.776,
      "step": 2000
    },
    {
      "epoch": 2.7925531914893615,
      "grad_norm": 8.407756805419922,
      "learning_rate": 4.422872340425532e-06,
      "loss": 0.1877,
      "step": 2100
    },
    {
      "epoch": 2.925531914893617,
      "grad_norm": 0.09165091067552567,
      "learning_rate": 4.156914893617022e-06,
      "loss": 0.1581,
      "step": 2200
    },
    {
      "epoch": 3.0585106382978724,
      "grad_norm": 17.642484664916992,
      "learning_rate": 3.890957446808511e-06,
      "loss": 0.1732,
      "step": 2300
    },
    {
      "epoch": 3.1914893617021276,
      "grad_norm": 19.95751953125,
      "learning_rate": 3.625e-06,
      "loss": 0.1504,
      "step": 2400
    },
    {
      "epoch": 3.324468085106383,
      "grad_norm": 0.0460813082754612,
      "learning_rate": 3.3590425531914896e-06,
      "loss": 0.1315,
      "step": 2500
    },
    {
      "epoch": 3.324468085106383,
      "eval_accuracy": 0.9162929654121399,
      "eval_loss": 0.38025063276290894,
      "eval_runtime": 60.4863,
      "eval_samples_per_second": 22.121,
      "eval_steps_per_second": 2.777,
      "step": 2500
    },
    {
      "epoch": 3.4574468085106385,
      "grad_norm": 0.5341168642044067,
      "learning_rate": 3.0957446808510637e-06,
      "loss": 0.1281,
      "step": 2600
    },
    {
      "epoch": 3.5904255319148937,
      "grad_norm": 0.07712464034557343,
      "learning_rate": 2.8297872340425537e-06,
      "loss": 0.1329,
      "step": 2700
    },
    {
      "epoch": 3.723404255319149,
      "grad_norm": 0.05446462705731392,
      "learning_rate": 2.563829787234043e-06,
      "loss": 0.137,
      "step": 2800
    },
    {
      "epoch": 3.8563829787234045,
      "grad_norm": 84.07832336425781,
      "learning_rate": 2.297872340425532e-06,
      "loss": 0.1207,
      "step": 2900
    },
    {
      "epoch": 3.9893617021276597,
      "grad_norm": 37.90822219848633,
      "learning_rate": 2.0319148936170213e-06,
      "loss": 0.1383,
      "step": 3000
    },
    {
      "epoch": 3.9893617021276597,
      "eval_accuracy": 0.9297459125518799,
      "eval_loss": 0.34964442253112793,
      "eval_runtime": 60.8105,
      "eval_samples_per_second": 22.003,
      "eval_steps_per_second": 2.763,
      "step": 3000
    },
    {
      "epoch": 3.9893617021276597,
      "step": 3000,
      "total_flos": 1.8714792090048768e+18,
      "train_loss": 0.34261955579121905,
      "train_runtime": 3211.0561,
      "train_samples_per_second": 18.713,
      "train_steps_per_second": 1.171
    }
  ],
  "logging_steps": 100,
  "max_steps": 3760,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.8714792090048768e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}