{
  "best_global_step": 2000,
  "best_metric": 0.37039539217948914,
  "best_model_checkpoint": "Assignment4_Distilled_ModernBERT/run-0/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": 13.524835586547852,
      "learning_rate": 5.982303202015292e-05,
      "loss": 7.1794,
      "step": 100
    },
    {
      "epoch": 0.20964360587002095,
      "eval_accuracy": 0.7551612903225806,
      "eval_loss": 3.656074285507202,
      "eval_runtime": 27.0212,
      "eval_samples_per_second": 114.725,
      "eval_steps_per_second": 14.359,
      "step": 100
    },
    {
      "epoch": 0.4192872117400419,
      "grad_norm": 7.738944053649902,
      "learning_rate": 5.9287097231582686e-05,
      "loss": 2.6619,
      "step": 200
    },
    {
      "epoch": 0.4192872117400419,
      "eval_accuracy": 0.9035483870967742,
      "eval_loss": 1.7683824300765991,
      "eval_runtime": 24.4159,
      "eval_samples_per_second": 126.967,
      "eval_steps_per_second": 15.891,
      "step": 200
    },
    {
      "epoch": 0.6289308176100629,
      "grad_norm": 5.298769950866699,
      "learning_rate": 5.8398628698129085e-05,
      "loss": 1.4594,
      "step": 300
    },
    {
      "epoch": 0.6289308176100629,
      "eval_accuracy": 0.9451612903225807,
      "eval_loss": 1.1510082483291626,
      "eval_runtime": 24.2312,
      "eval_samples_per_second": 127.934,
      "eval_steps_per_second": 16.012,
      "step": 300
    },
    {
      "epoch": 0.8385744234800838,
      "grad_norm": 7.6655049324035645,
      "learning_rate": 5.71683210658621e-05,
      "loss": 1.0566,
      "step": 400
    },
    {
      "epoch": 0.8385744234800838,
      "eval_accuracy": 0.9551612903225807,
      "eval_loss": 0.9296484589576721,
      "eval_runtime": 24.2932,
      "eval_samples_per_second": 127.608,
      "eval_steps_per_second": 15.972,
      "step": 400
    },
    {
      "epoch": 1.0482180293501049,
      "grad_norm": 14.238273620605469,
      "learning_rate": 5.561098375592486e-05,
      "loss": 0.8012,
      "step": 500
    },
    {
      "epoch": 1.0482180293501049,
      "eval_accuracy": 0.96,
      "eval_loss": 0.7582064270973206,
      "eval_runtime": 24.3409,
      "eval_samples_per_second": 127.358,
      "eval_steps_per_second": 15.94,
      "step": 500
    },
    {
      "epoch": 1.2578616352201257,
      "grad_norm": 2.1679141521453857,
      "learning_rate": 5.3745362701025445e-05,
      "loss": 0.5647,
      "step": 600
    },
    {
      "epoch": 1.2578616352201257,
      "eval_accuracy": 0.9661290322580646,
      "eval_loss": 0.6347441077232361,
      "eval_runtime": 25.9162,
      "eval_samples_per_second": 119.616,
      "eval_steps_per_second": 14.971,
      "step": 600
    },
    {
      "epoch": 1.4675052410901468,
      "grad_norm": 3.5543723106384277,
      "learning_rate": 5.1593914697473044e-05,
      "loss": 0.5087,
      "step": 700
    },
    {
      "epoch": 1.4675052410901468,
      "eval_accuracy": 0.9593548387096774,
      "eval_loss": 0.6069635152816772,
      "eval_runtime": 24.4377,
      "eval_samples_per_second": 126.853,
      "eval_steps_per_second": 15.877,
      "step": 700
    },
    {
      "epoch": 1.6771488469601676,
      "grad_norm": 2.1209726333618164,
      "learning_rate": 4.918253708892097e-05,
      "loss": 0.4459,
      "step": 800
    },
    {
      "epoch": 1.6771488469601676,
      "eval_accuracy": 0.9619354838709677,
      "eval_loss": 0.5518479347229004,
      "eval_runtime": 24.5853,
      "eval_samples_per_second": 126.092,
      "eval_steps_per_second": 15.782,
      "step": 800
    },
    {
      "epoch": 1.8867924528301887,
      "grad_norm": 7.709955215454102,
      "learning_rate": 4.654025603565916e-05,
      "loss": 0.4246,
      "step": 900
    },
    {
      "epoch": 1.8867924528301887,
      "eval_accuracy": 0.9693548387096774,
      "eval_loss": 0.5302325487136841,
      "eval_runtime": 24.3067,
      "eval_samples_per_second": 127.537,
      "eval_steps_per_second": 15.963,
      "step": 900
    },
    {
      "epoch": 2.0964360587002098,
      "grad_norm": 4.7413787841796875,
      "learning_rate": 4.3698877121811395e-05,
      "loss": 0.3684,
      "step": 1000
    },
    {
      "epoch": 2.0964360587002098,
      "eval_accuracy": 0.967741935483871,
      "eval_loss": 0.4881534278392792,
      "eval_runtime": 24.3233,
      "eval_samples_per_second": 127.45,
      "eval_steps_per_second": 15.952,
      "step": 1000
    },
    {
      "epoch": 2.3060796645702304,
      "grad_norm": 1.3662949800491333,
      "learning_rate": 4.069260250613759e-05,
      "loss": 0.2996,
      "step": 1100
    },
    {
      "epoch": 2.3060796645702304,
      "eval_accuracy": 0.9696774193548388,
      "eval_loss": 0.4698588252067566,
      "eval_runtime": 26.4097,
      "eval_samples_per_second": 117.381,
      "eval_steps_per_second": 14.692,
      "step": 1100
    },
    {
      "epoch": 2.5157232704402515,
      "grad_norm": 2.2415976524353027,
      "learning_rate": 3.755761922486161e-05,
      "loss": 0.3004,
      "step": 1200
    },
    {
      "epoch": 2.5157232704402515,
      "eval_accuracy": 0.9690322580645161,
      "eval_loss": 0.4617905616760254,
      "eval_runtime": 24.5627,
      "eval_samples_per_second": 126.208,
      "eval_steps_per_second": 15.796,
      "step": 1200
    },
    {
      "epoch": 2.7253668763102725,
      "grad_norm": 1.3881778717041016,
      "learning_rate": 3.433166360219327e-05,
      "loss": 0.2721,
      "step": 1300
    },
    {
      "epoch": 2.7253668763102725,
      "eval_accuracy": 0.9696774193548388,
      "eval_loss": 0.4343847632408142,
      "eval_runtime": 24.5311,
      "eval_samples_per_second": 126.37,
      "eval_steps_per_second": 15.817,
      "step": 1300
    },
    {
      "epoch": 2.9350104821802936,
      "grad_norm": 1.2978781461715698,
      "learning_rate": 3.1053567011808615e-05,
      "loss": 0.2591,
      "step": 1400
    },
    {
      "epoch": 2.9350104821802936,
      "eval_accuracy": 0.97,
      "eval_loss": 0.418453186750412,
      "eval_runtime": 24.631,
      "eval_samples_per_second": 125.858,
      "eval_steps_per_second": 15.752,
      "step": 1400
    },
    {
      "epoch": 3.1446540880503147,
      "grad_norm": 1.3791582584381104,
      "learning_rate": 2.776278845703461e-05,
      "loss": 0.2313,
      "step": 1500
    },
    {
      "epoch": 3.1446540880503147,
      "eval_accuracy": 0.9693548387096774,
      "eval_loss": 0.40362659096717834,
      "eval_runtime": 24.4944,
      "eval_samples_per_second": 126.56,
      "eval_steps_per_second": 15.84,
      "step": 1500
    },
    {
      "epoch": 3.3542976939203353,
      "grad_norm": 1.2525160312652588,
      "learning_rate": 2.44989395961494e-05,
      "loss": 0.2179,
      "step": 1600
    },
    {
      "epoch": 3.3542976939203353,
      "eval_accuracy": 0.9706451612903226,
      "eval_loss": 0.3979440927505493,
      "eval_runtime": 26.4009,
      "eval_samples_per_second": 117.42,
      "eval_steps_per_second": 14.696,
      "step": 1600
    },
    {
      "epoch": 3.5639412997903563,
      "grad_norm": 0.8602608442306519,
      "learning_rate": 2.130130793014917e-05,
      "loss": 0.2074,
      "step": 1700
    },
    {
      "epoch": 3.5639412997903563,
      "eval_accuracy": 0.9690322580645161,
      "eval_loss": 0.3927501142024994,
      "eval_runtime": 24.7417,
      "eval_samples_per_second": 125.295,
      "eval_steps_per_second": 15.682,
      "step": 1700
    },
    {
      "epoch": 3.7735849056603774,
      "grad_norm": 0.9303811192512512,
      "learning_rate": 1.8208383892451047e-05,
      "loss": 0.204,
      "step": 1800
    },
    {
      "epoch": 3.7735849056603774,
      "eval_accuracy": 0.97,
      "eval_loss": 0.3800348937511444,
      "eval_runtime": 24.5036,
      "eval_samples_per_second": 126.512,
      "eval_steps_per_second": 15.834,
      "step": 1800
    },
    {
      "epoch": 3.9832285115303985,
      "grad_norm": 2.1014115810394287,
      "learning_rate": 1.525739753303325e-05,
      "loss": 0.202,
      "step": 1900
    },
    {
      "epoch": 3.9832285115303985,
      "eval_accuracy": 0.9709677419354839,
      "eval_loss": 0.37354010343551636,
      "eval_runtime": 24.7707,
      "eval_samples_per_second": 125.148,
      "eval_steps_per_second": 15.664,
      "step": 1900
    },
    {
      "epoch": 4.1928721174004195,
      "grad_norm": 0.7623395919799805,
      "learning_rate": 1.2483870374024092e-05,
      "loss": 0.1776,
      "step": 2000
    },
    {
      "epoch": 4.1928721174004195,
      "eval_accuracy": 0.9703225806451613,
      "eval_loss": 0.37039539217948914,
      "eval_runtime": 24.6812,
      "eval_samples_per_second": 125.602,
      "eval_steps_per_second": 15.72,
      "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": 0.43742261757664425,
    "num_train_epochs": 6,
    "temperature": 7
  }
}