{
  "best_global_step": 2000,
  "best_metric": 0.33797791600227356,
  "best_model_checkpoint": "Assignment4_Distilled_ModernBERT/run-3/checkpoint-2000",
  "epoch": 4.1928721174004195,
  "eval_steps": 100,
  "global_step": 2000,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.20964360587002095,
      "grad_norm": 9.00671100616455,
      "learning_rate": 5.982303202015292e-05,
      "loss": 5.2947,
      "step": 100
    },
    {
      "epoch": 0.20964360587002095,
      "eval_accuracy": 0.7319354838709677,
      "eval_loss": 3.0118744373321533,
      "eval_runtime": 24.6185,
      "eval_samples_per_second": 125.921,
      "eval_steps_per_second": 15.76,
      "step": 100
    },
    {
      "epoch": 0.4192872117400419,
      "grad_norm": 9.74803352355957,
      "learning_rate": 5.9287097231582686e-05,
      "loss": 2.1846,
      "step": 200
    },
    {
      "epoch": 0.4192872117400419,
      "eval_accuracy": 0.9187096774193548,
      "eval_loss": 1.5048528909683228,
      "eval_runtime": 24.5137,
      "eval_samples_per_second": 126.46,
      "eval_steps_per_second": 15.828,
      "step": 200
    },
    {
      "epoch": 0.6289308176100629,
      "grad_norm": 2.313565731048584,
      "learning_rate": 5.8398628698129085e-05,
      "loss": 1.2384,
      "step": 300
    },
    {
      "epoch": 0.6289308176100629,
      "eval_accuracy": 0.9496774193548387,
      "eval_loss": 0.9920670986175537,
      "eval_runtime": 24.4854,
      "eval_samples_per_second": 126.606,
      "eval_steps_per_second": 15.846,
      "step": 300
    },
    {
      "epoch": 0.8385744234800838,
      "grad_norm": 6.652754306793213,
      "learning_rate": 5.71683210658621e-05,
      "loss": 0.8701,
      "step": 400
    },
    {
      "epoch": 0.8385744234800838,
      "eval_accuracy": 0.9574193548387097,
      "eval_loss": 0.8129040002822876,
      "eval_runtime": 24.4845,
      "eval_samples_per_second": 126.611,
      "eval_steps_per_second": 15.847,
      "step": 400
    },
    {
      "epoch": 1.0482180293501049,
      "grad_norm": 6.9664835929870605,
      "learning_rate": 5.561098375592486e-05,
      "loss": 0.6485,
      "step": 500
    },
    {
      "epoch": 1.0482180293501049,
      "eval_accuracy": 0.964516129032258,
      "eval_loss": 0.6442162990570068,
      "eval_runtime": 24.8893,
      "eval_samples_per_second": 124.552,
      "eval_steps_per_second": 15.589,
      "step": 500
    },
    {
      "epoch": 1.2578616352201257,
      "grad_norm": 1.474057912826538,
      "learning_rate": 5.3745362701025445e-05,
      "loss": 0.464,
      "step": 600
    },
    {
      "epoch": 1.2578616352201257,
      "eval_accuracy": 0.9658064516129032,
      "eval_loss": 0.5599253177642822,
      "eval_runtime": 25.4446,
      "eval_samples_per_second": 121.833,
      "eval_steps_per_second": 15.249,
      "step": 600
    },
    {
      "epoch": 1.4675052410901468,
      "grad_norm": 1.928739309310913,
      "learning_rate": 5.1593914697473044e-05,
      "loss": 0.4216,
      "step": 700
    },
    {
      "epoch": 1.4675052410901468,
      "eval_accuracy": 0.9648387096774194,
      "eval_loss": 0.5214470624923706,
      "eval_runtime": 25.3207,
      "eval_samples_per_second": 122.43,
      "eval_steps_per_second": 15.323,
      "step": 700
    },
    {
      "epoch": 1.6771488469601676,
      "grad_norm": 0.9871991276741028,
      "learning_rate": 4.918253708892097e-05,
      "loss": 0.3627,
      "step": 800
    },
    {
      "epoch": 1.6771488469601676,
      "eval_accuracy": 0.9664516129032258,
      "eval_loss": 0.48103204369544983,
      "eval_runtime": 24.854,
      "eval_samples_per_second": 124.728,
      "eval_steps_per_second": 15.611,
      "step": 800
    },
    {
      "epoch": 1.8867924528301887,
      "grad_norm": 3.7150447368621826,
      "learning_rate": 4.654025603565916e-05,
      "loss": 0.3641,
      "step": 900
    },
    {
      "epoch": 1.8867924528301887,
      "eval_accuracy": 0.9716129032258064,
      "eval_loss": 0.4549710154533386,
      "eval_runtime": 29.4013,
      "eval_samples_per_second": 105.438,
      "eval_steps_per_second": 13.197,
      "step": 900
    },
    {
      "epoch": 2.0964360587002098,
      "grad_norm": 1.0180896520614624,
      "learning_rate": 4.3698877121811395e-05,
      "loss": 0.3114,
      "step": 1000
    },
    {
      "epoch": 2.0964360587002098,
      "eval_accuracy": 0.97,
      "eval_loss": 0.42610129714012146,
      "eval_runtime": 28.5631,
      "eval_samples_per_second": 108.532,
      "eval_steps_per_second": 13.584,
      "step": 1000
    },
    {
      "epoch": 2.3060796645702304,
      "grad_norm": 0.8552842140197754,
      "learning_rate": 4.069260250613759e-05,
      "loss": 0.2549,
      "step": 1100
    },
    {
      "epoch": 2.3060796645702304,
      "eval_accuracy": 0.97,
      "eval_loss": 0.4119759500026703,
      "eval_runtime": 24.7756,
      "eval_samples_per_second": 125.123,
      "eval_steps_per_second": 15.661,
      "step": 1100
    },
    {
      "epoch": 2.5157232704402515,
      "grad_norm": 1.075974464416504,
      "learning_rate": 3.755761922486161e-05,
      "loss": 0.2496,
      "step": 1200
    },
    {
      "epoch": 2.5157232704402515,
      "eval_accuracy": 0.9703225806451613,
      "eval_loss": 0.3999555706977844,
      "eval_runtime": 24.7123,
      "eval_samples_per_second": 125.444,
      "eval_steps_per_second": 15.701,
      "step": 1200
    },
    {
      "epoch": 2.7253668763102725,
      "grad_norm": 0.907684326171875,
      "learning_rate": 3.433166360219327e-05,
      "loss": 0.2276,
      "step": 1300
    },
    {
      "epoch": 2.7253668763102725,
      "eval_accuracy": 0.9709677419354839,
      "eval_loss": 0.3826442062854767,
      "eval_runtime": 24.7314,
      "eval_samples_per_second": 125.347,
      "eval_steps_per_second": 15.689,
      "step": 1300
    },
    {
      "epoch": 2.9350104821802936,
      "grad_norm": 0.8015017509460449,
      "learning_rate": 3.1053567011808615e-05,
      "loss": 0.2209,
      "step": 1400
    },
    {
      "epoch": 2.9350104821802936,
      "eval_accuracy": 0.9712903225806452,
      "eval_loss": 0.37278029322624207,
      "eval_runtime": 24.835,
      "eval_samples_per_second": 124.824,
      "eval_steps_per_second": 15.623,
      "step": 1400
    },
    {
      "epoch": 3.1446540880503147,
      "grad_norm": 0.8900270462036133,
      "learning_rate": 2.776278845703461e-05,
      "loss": 0.1986,
      "step": 1500
    },
    {
      "epoch": 3.1446540880503147,
      "eval_accuracy": 0.9696774193548388,
      "eval_loss": 0.35799843072891235,
      "eval_runtime": 24.6469,
      "eval_samples_per_second": 125.776,
      "eval_steps_per_second": 15.742,
      "step": 1500
    },
    {
      "epoch": 3.3542976939203353,
      "grad_norm": 0.7829304337501526,
      "learning_rate": 2.44989395961494e-05,
      "loss": 0.185,
      "step": 1600
    },
    {
      "epoch": 3.3542976939203353,
      "eval_accuracy": 0.9716129032258064,
      "eval_loss": 0.3530636131763458,
      "eval_runtime": 24.982,
      "eval_samples_per_second": 124.089,
      "eval_steps_per_second": 15.531,
      "step": 1600
    },
    {
      "epoch": 3.5639412997903563,
      "grad_norm": 0.6377654671669006,
      "learning_rate": 2.130130793014917e-05,
      "loss": 0.1777,
      "step": 1700
    },
    {
      "epoch": 3.5639412997903563,
      "eval_accuracy": 0.97,
      "eval_loss": 0.35114002227783203,
      "eval_runtime": 25.0031,
      "eval_samples_per_second": 123.985,
      "eval_steps_per_second": 15.518,
      "step": 1700
    },
    {
      "epoch": 3.7735849056603774,
      "grad_norm": 0.7244699597358704,
      "learning_rate": 1.8208383892451047e-05,
      "loss": 0.175,
      "step": 1800
    },
    {
      "epoch": 3.7735849056603774,
      "eval_accuracy": 0.9719354838709677,
      "eval_loss": 0.34239596128463745,
      "eval_runtime": 25.13,
      "eval_samples_per_second": 123.358,
      "eval_steps_per_second": 15.44,
      "step": 1800
    },
    {
      "epoch": 3.9832285115303985,
      "grad_norm": 0.7429079413414001,
      "learning_rate": 1.525739753303325e-05,
      "loss": 0.1747,
      "step": 1900
    },
    {
      "epoch": 3.9832285115303985,
      "eval_accuracy": 0.9712903225806452,
      "eval_loss": 0.3403235673904419,
      "eval_runtime": 26.1406,
      "eval_samples_per_second": 118.59,
      "eval_steps_per_second": 14.843,
      "step": 1900
    },
    {
      "epoch": 4.1928721174004195,
      "grad_norm": 0.5660148859024048,
      "learning_rate": 1.2483870374024092e-05,
      "loss": 0.1555,
      "step": 2000
    },
    {
      "epoch": 4.1928721174004195,
      "eval_accuracy": 0.9709677419354839,
      "eval_loss": 0.33797791600227356,
      "eval_runtime": 31.949,
      "eval_samples_per_second": 97.03,
      "eval_steps_per_second": 12.144,
      "step": 2000
    }
  ],
  "logging_steps": 100,
  "max_steps": 2862,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 896601881570736.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.41727432608641335,
    "num_train_epochs": 6,
    "temperature": 10
  }
}