{
  "best_global_step": 2000,
  "best_metric": 0.33800485730171204,
  "best_model_checkpoint": "Assignment4_Distilled_ModernBERT/run-4/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": 11.635396957397461,
      "learning_rate": 5.974527645836366e-05,
      "loss": 6.4293,
      "step": 100
    },
    {
      "epoch": 0.20964360587002095,
      "eval_accuracy": 0.7254838709677419,
      "eval_loss": 3.423161268234253,
      "eval_runtime": 24.7311,
      "eval_samples_per_second": 125.348,
      "eval_steps_per_second": 15.689,
      "step": 100
    },
    {
      "epoch": 0.4192872117400419,
      "grad_norm": 7.08547830581665,
      "learning_rate": 5.897521628664326e-05,
      "loss": 2.4649,
      "step": 200
    },
    {
      "epoch": 0.4192872117400419,
      "eval_accuracy": 0.9012903225806451,
      "eval_loss": 1.6392749547958374,
      "eval_runtime": 24.334,
      "eval_samples_per_second": 127.394,
      "eval_steps_per_second": 15.945,
      "step": 200
    },
    {
      "epoch": 0.6289308176100629,
      "grad_norm": 2.8314898014068604,
      "learning_rate": 5.7703135417326854e-05,
      "loss": 1.364,
      "step": 300
    },
    {
      "epoch": 0.6289308176100629,
      "eval_accuracy": 0.9361290322580645,
      "eval_loss": 1.0801855325698853,
      "eval_runtime": 24.4718,
      "eval_samples_per_second": 126.676,
      "eval_steps_per_second": 15.855,
      "step": 300
    },
    {
      "epoch": 0.8385744234800838,
      "grad_norm": 11.189908981323242,
      "learning_rate": 5.595107375234632e-05,
      "loss": 0.9328,
      "step": 400
    },
    {
      "epoch": 0.8385744234800838,
      "eval_accuracy": 0.9506451612903226,
      "eval_loss": 0.8573997616767883,
      "eval_runtime": 24.7134,
      "eval_samples_per_second": 125.438,
      "eval_steps_per_second": 15.7,
      "step": 400
    },
    {
      "epoch": 1.0482180293501049,
      "grad_norm": 9.433755874633789,
      "learning_rate": 5.3749387276298395e-05,
      "loss": 0.708,
      "step": 500
    },
    {
      "epoch": 1.0482180293501049,
      "eval_accuracy": 0.9564516129032258,
      "eval_loss": 0.6847298741340637,
      "eval_runtime": 24.9978,
      "eval_samples_per_second": 124.011,
      "eval_steps_per_second": 15.521,
      "step": 500
    },
    {
      "epoch": 1.2578616352201257,
      "grad_norm": 1.7648718357086182,
      "learning_rate": 5.113622211274119e-05,
      "loss": 0.4702,
      "step": 600
    },
    {
      "epoch": 1.2578616352201257,
      "eval_accuracy": 0.9616129032258065,
      "eval_loss": 0.5767695307731628,
      "eval_runtime": 25.0749,
      "eval_samples_per_second": 123.63,
      "eval_steps_per_second": 15.474,
      "step": 600
    },
    {
      "epoch": 1.4675052410901468,
      "grad_norm": 2.782994031906128,
      "learning_rate": 4.815685360959153e-05,
      "loss": 0.4294,
      "step": 700
    },
    {
      "epoch": 1.4675052410901468,
      "eval_accuracy": 0.9583870967741935,
      "eval_loss": 0.5501284599304199,
      "eval_runtime": 24.8053,
      "eval_samples_per_second": 124.973,
      "eval_steps_per_second": 15.642,
      "step": 700
    },
    {
      "epoch": 1.6771488469601676,
      "grad_norm": 1.1726200580596924,
      "learning_rate": 4.486290190454053e-05,
      "loss": 0.3644,
      "step": 800
    },
    {
      "epoch": 1.6771488469601676,
      "eval_accuracy": 0.964516129032258,
      "eval_loss": 0.48550063371658325,
      "eval_runtime": 25.3011,
      "eval_samples_per_second": 122.524,
      "eval_steps_per_second": 15.335,
      "step": 800
    },
    {
      "epoch": 1.8867924528301887,
      "grad_norm": 11.229223251342773,
      "learning_rate": 4.13114375614956e-05,
      "loss": 0.3547,
      "step": 900
    },
    {
      "epoch": 1.8867924528301887,
      "eval_accuracy": 0.9664516129032258,
      "eval_loss": 0.46239548921585083,
      "eval_runtime": 25.1571,
      "eval_samples_per_second": 123.226,
      "eval_steps_per_second": 15.423,
      "step": 900
    },
    {
      "epoch": 2.0964360587002098,
      "grad_norm": 1.082562804222107,
      "learning_rate": 3.756399277367126e-05,
      "loss": 0.2973,
      "step": 1000
    },
    {
      "epoch": 2.0964360587002098,
      "eval_accuracy": 0.967741935483871,
      "eval_loss": 0.4247872829437256,
      "eval_runtime": 31.4174,
      "eval_samples_per_second": 98.671,
      "eval_steps_per_second": 12.35,
      "step": 1000
    },
    {
      "epoch": 2.3060796645702304,
      "grad_norm": 1.0132005214691162,
      "learning_rate": 3.36854952650912e-05,
      "loss": 0.2349,
      "step": 1100
    },
    {
      "epoch": 2.3060796645702304,
      "eval_accuracy": 0.9703225806451613,
      "eval_loss": 0.3974890410900116,
      "eval_runtime": 29.8794,
      "eval_samples_per_second": 103.75,
      "eval_steps_per_second": 12.986,
      "step": 1100
    },
    {
      "epoch": 2.5157232704402515,
      "grad_norm": 1.2016338109970093,
      "learning_rate": 2.9743143361580517e-05,
      "loss": 0.2317,
      "step": 1200
    },
    {
      "epoch": 2.5157232704402515,
      "eval_accuracy": 0.9683870967741935,
      "eval_loss": 0.3919287621974945,
      "eval_runtime": 26.2852,
      "eval_samples_per_second": 117.937,
      "eval_steps_per_second": 14.761,
      "step": 1200
    },
    {
      "epoch": 2.7253668763102725,
      "grad_norm": 0.9690835475921631,
      "learning_rate": 2.5805241721616645e-05,
      "loss": 0.2045,
      "step": 1300
    },
    {
      "epoch": 2.7253668763102725,
      "eval_accuracy": 0.97,
      "eval_loss": 0.37409406900405884,
      "eval_runtime": 25.1471,
      "eval_samples_per_second": 123.275,
      "eval_steps_per_second": 15.429,
      "step": 1300
    },
    {
      "epoch": 2.9350104821802936,
      "grad_norm": 0.7975201606750488,
      "learning_rate": 2.1940017899009786e-05,
      "loss": 0.197,
      "step": 1400
    },
    {
      "epoch": 2.9350104821802936,
      "eval_accuracy": 0.9690322580645161,
      "eval_loss": 0.36775872111320496,
      "eval_runtime": 24.7763,
      "eval_samples_per_second": 125.12,
      "eval_steps_per_second": 15.66,
      "step": 1400
    },
    {
      "epoch": 3.1446540880503147,
      "grad_norm": 0.850993275642395,
      "learning_rate": 1.8214440241488357e-05,
      "loss": 0.1785,
      "step": 1500
    },
    {
      "epoch": 3.1446540880503147,
      "eval_accuracy": 0.9696774193548388,
      "eval_loss": 0.35433417558670044,
      "eval_runtime": 24.6341,
      "eval_samples_per_second": 125.842,
      "eval_steps_per_second": 15.751,
      "step": 1500
    },
    {
      "epoch": 3.3542976939203353,
      "grad_norm": 0.8356935977935791,
      "learning_rate": 1.4693057606119536e-05,
      "loss": 0.1645,
      "step": 1600
    },
    {
      "epoch": 3.3542976939203353,
      "eval_accuracy": 0.9703225806451613,
      "eval_loss": 0.35072386264801025,
      "eval_runtime": 24.97,
      "eval_samples_per_second": 124.149,
      "eval_steps_per_second": 15.539,
      "step": 1600
    },
    {
      "epoch": 3.5639412997903563,
      "grad_norm": 0.5763674974441528,
      "learning_rate": 1.143688099449935e-05,
      "loss": 0.1576,
      "step": 1700
    },
    {
      "epoch": 3.5639412997903563,
      "eval_accuracy": 0.9680645161290322,
      "eval_loss": 0.3471537232398987,
      "eval_runtime": 24.9021,
      "eval_samples_per_second": 124.488,
      "eval_steps_per_second": 15.581,
      "step": 1700
    },
    {
      "epoch": 3.7735849056603774,
      "grad_norm": 0.7285761833190918,
      "learning_rate": 8.502326484350608e-06,
      "loss": 0.1553,
      "step": 1800
    },
    {
      "epoch": 3.7735849056603774,
      "eval_accuracy": 0.9693548387096774,
      "eval_loss": 0.3421308100223541,
      "eval_runtime": 25.0889,
      "eval_samples_per_second": 123.561,
      "eval_steps_per_second": 15.465,
      "step": 1800
    },
    {
      "epoch": 3.9832285115303985,
      "grad_norm": 0.9058928489685059,
      "learning_rate": 5.940237772153215e-06,
      "loss": 0.155,
      "step": 1900
    },
    {
      "epoch": 3.9832285115303985,
      "eval_accuracy": 0.97,
      "eval_loss": 0.3395131826400757,
      "eval_runtime": 25.8266,
      "eval_samples_per_second": 120.031,
      "eval_steps_per_second": 15.023,
      "step": 1900
    },
    {
      "epoch": 4.1928721174004195,
      "grad_norm": 0.5501458048820496,
      "learning_rate": 3.795005262102136e-06,
      "loss": 0.1413,
      "step": 2000
    },
    {
      "epoch": 4.1928721174004195,
      "eval_accuracy": 0.9696774193548388,
      "eval_loss": 0.33800485730171204,
      "eval_runtime": 28.266,
      "eval_samples_per_second": 109.673,
      "eval_steps_per_second": 13.727,
      "step": 2000
    }
  ],
  "logging_steps": 100,
  "max_steps": 2385,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 0.5611126477660928,
    "num_train_epochs": 5,
    "temperature": 7
  }
}