{
  "best_metric": 0.15075480937957764,
  "best_model_checkpoint": "/home/paperspace/microsoft_deberta-v3-base_cpt_64aa255e/checkpoint-9820",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 9820,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "learning_rate": 9.709048588885654e-07,
      "loss": 0.5526,
      "step": 500
    },
    {
      "epoch": 0.41,
      "learning_rate": 9.418097177771311e-07,
      "loss": 0.2449,
      "step": 1000
    },
    {
      "epoch": 0.61,
      "learning_rate": 9.127145766656967e-07,
      "loss": 0.1637,
      "step": 1500
    },
    {
      "epoch": 0.81,
      "learning_rate": 8.836194355542625e-07,
      "loss": 0.1627,
      "step": 2000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9574516238542492,
      "eval_f1": 0.9576402605174089,
      "eval_loss": 0.15390127897262573,
      "eval_precision": 0.9583932523988209,
      "eval_recall": 0.9574516238542492,
      "eval_runtime": 11.3772,
      "eval_samples_per_second": 776.73,
      "eval_steps_per_second": 48.606,
      "step": 2455
    },
    {
      "epoch": 1.02,
      "learning_rate": 8.54524294442828e-07,
      "loss": 0.1536,
      "step": 2500
    },
    {
      "epoch": 1.22,
      "learning_rate": 8.254291533313936e-07,
      "loss": 0.1497,
      "step": 3000
    },
    {
      "epoch": 1.43,
      "learning_rate": 7.963340122199592e-07,
      "loss": 0.1478,
      "step": 3500
    },
    {
      "epoch": 1.63,
      "learning_rate": 7.672388711085248e-07,
      "loss": 0.1452,
      "step": 4000
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.381437299970904e-07,
      "loss": 0.149,
      "step": 4500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9596016747765078,
      "eval_f1": 0.9597684084567597,
      "eval_loss": 0.15375804901123047,
      "eval_precision": 0.9604205146694305,
      "eval_recall": 0.9596016747765078,
      "eval_runtime": 11.4943,
      "eval_samples_per_second": 768.813,
      "eval_steps_per_second": 48.111,
      "step": 4910
    },
    {
      "epoch": 2.04,
      "learning_rate": 7.090485888856561e-07,
      "loss": 0.1415,
      "step": 5000
    },
    {
      "epoch": 2.24,
      "learning_rate": 6.799534477742217e-07,
      "loss": 0.1358,
      "step": 5500
    },
    {
      "epoch": 2.44,
      "learning_rate": 6.508583066627873e-07,
      "loss": 0.1382,
      "step": 6000
    },
    {
      "epoch": 2.65,
      "learning_rate": 6.217631655513529e-07,
      "loss": 0.1289,
      "step": 6500
    },
    {
      "epoch": 2.85,
      "learning_rate": 5.926680244399185e-07,
      "loss": 0.1329,
      "step": 7000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9591490324770849,
      "eval_f1": 0.9593624018161547,
      "eval_loss": 0.16056260466575623,
      "eval_precision": 0.9603821831580927,
      "eval_recall": 0.9591490324770849,
      "eval_runtime": 11.3768,
      "eval_samples_per_second": 776.759,
      "eval_steps_per_second": 48.608,
      "step": 7365
    },
    {
      "epoch": 3.05,
      "learning_rate": 5.63572883328484e-07,
      "loss": 0.1321,
      "step": 7500
    },
    {
      "epoch": 3.26,
      "learning_rate": 5.344777422170497e-07,
      "loss": 0.1324,
      "step": 8000
    },
    {
      "epoch": 3.46,
      "learning_rate": 5.053826011056153e-07,
      "loss": 0.1299,
      "step": 8500
    },
    {
      "epoch": 3.67,
      "learning_rate": 4.7628745999418097e-07,
      "loss": 0.136,
      "step": 9000
    },
    {
      "epoch": 3.87,
      "learning_rate": 4.4719231888274657e-07,
      "loss": 0.1308,
      "step": 9500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9608464410999208,
      "eval_f1": 0.9610329418449746,
      "eval_loss": 0.15075480937957764,
      "eval_precision": 0.9618808519987467,
      "eval_recall": 0.9608464410999208,
      "eval_runtime": 11.2225,
      "eval_samples_per_second": 787.436,
      "eval_steps_per_second": 49.276,
      "step": 9820
    }
  ],
  "logging_steps": 500,
  "max_steps": 17185,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 3808161445168620.0,
  "trial_name": null,
  "trial_params": null
}