{
  "best_metric": 0.07832363992929459,
  "best_model_checkpoint": "/storage/BioM-ELECTRA-Large-SQuAD2_carDB_5e_neg_lg_SGD/checkpoint-3204",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 8010,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3121098626716604,
      "grad_norm": 3.2046010494232178,
      "learning_rate": 4.690387016229713e-05,
      "loss": 0.1789,
      "step": 500
    },
    {
      "epoch": 0.6242197253433208,
      "grad_norm": 1.161939263343811,
      "learning_rate": 4.378277153558052e-05,
      "loss": 0.1113,
      "step": 1000
    },
    {
      "epoch": 0.9363295880149812,
      "grad_norm": 2.1341910362243652,
      "learning_rate": 4.066167290886392e-05,
      "loss": 0.1054,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_accuracy_score": 0.9690997325938397,
      "eval_f1": 0.7794316644113668,
      "eval_loss": 0.08875501155853271,
      "eval_precision": 0.7747886241352806,
      "eval_recall": 0.7841306884480747,
      "eval_runtime": 4.093,
      "eval_samples_per_second": 348.153,
      "eval_steps_per_second": 43.733,
      "step": 1602
    },
    {
      "epoch": 1.2484394506866416,
      "grad_norm": 0.97569340467453,
      "learning_rate": 3.754057428214732e-05,
      "loss": 0.079,
      "step": 2000
    },
    {
      "epoch": 1.5605493133583022,
      "grad_norm": 0.8073092699050903,
      "learning_rate": 3.441947565543071e-05,
      "loss": 0.0685,
      "step": 2500
    },
    {
      "epoch": 1.8726591760299627,
      "grad_norm": 2.6909968852996826,
      "learning_rate": 3.129837702871411e-05,
      "loss": 0.0671,
      "step": 3000
    },
    {
      "epoch": 2.0,
      "eval_accuracy_score": 0.973506982271962,
      "eval_f1": 0.814260389994091,
      "eval_loss": 0.07832363992929459,
      "eval_precision": 0.8248204309656824,
      "eval_recall": 0.8039673278879813,
      "eval_runtime": 4.1937,
      "eval_samples_per_second": 339.796,
      "eval_steps_per_second": 42.683,
      "step": 3204
    },
    {
      "epoch": 2.184769038701623,
      "grad_norm": 0.14074726402759552,
      "learning_rate": 2.818352059925094e-05,
      "loss": 0.0534,
      "step": 3500
    },
    {
      "epoch": 2.4968789013732833,
      "grad_norm": 0.8002931475639343,
      "learning_rate": 2.5062421972534333e-05,
      "loss": 0.043,
      "step": 4000
    },
    {
      "epoch": 2.808988764044944,
      "grad_norm": 1.326310157775879,
      "learning_rate": 2.194132334581773e-05,
      "loss": 0.0433,
      "step": 4500
    },
    {
      "epoch": 3.0,
      "eval_accuracy_score": 0.9748935327324948,
      "eval_f1": 0.8186646433990896,
      "eval_loss": 0.09053385257720947,
      "eval_precision": 0.7989633469085524,
      "eval_recall": 0.839362115908207,
      "eval_runtime": 4.09,
      "eval_samples_per_second": 348.41,
      "eval_steps_per_second": 43.765,
      "step": 4806
    },
    {
      "epoch": 3.1210986267166043,
      "grad_norm": 0.9473728537559509,
      "learning_rate": 1.8820224719101125e-05,
      "loss": 0.0318,
      "step": 5000
    },
    {
      "epoch": 3.4332084893882646,
      "grad_norm": 0.391970157623291,
      "learning_rate": 1.569912609238452e-05,
      "loss": 0.0228,
      "step": 5500
    },
    {
      "epoch": 3.7453183520599254,
      "grad_norm": 5.073329448699951,
      "learning_rate": 1.258426966292135e-05,
      "loss": 0.0224,
      "step": 6000
    },
    {
      "epoch": 4.0,
      "eval_accuracy_score": 0.9751906506883232,
      "eval_f1": 0.8262295081967214,
      "eval_loss": 0.09821399301290512,
      "eval_precision": 0.819433817903596,
      "eval_recall": 0.8331388564760793,
      "eval_runtime": 4.1062,
      "eval_samples_per_second": 347.04,
      "eval_steps_per_second": 43.593,
      "step": 6408
    },
    {
      "epoch": 4.057428214731585,
      "grad_norm": 0.055937759578228,
      "learning_rate": 9.463171036204745e-06,
      "loss": 0.0204,
      "step": 6500
    },
    {
      "epoch": 4.369538077403246,
      "grad_norm": 0.6860196590423584,
      "learning_rate": 6.342072409488139e-06,
      "loss": 0.0116,
      "step": 7000
    },
    {
      "epoch": 4.681647940074907,
      "grad_norm": 0.6975510716438293,
      "learning_rate": 3.2209737827715358e-06,
      "loss": 0.011,
      "step": 7500
    },
    {
      "epoch": 4.9937578027465666,
      "grad_norm": 0.33860084414482117,
      "learning_rate": 9.987515605493134e-08,
      "loss": 0.0091,
      "step": 8000
    },
    {
      "epoch": 5.0,
      "eval_accuracy_score": 0.9755125284738041,
      "eval_f1": 0.8289676425269646,
      "eval_loss": 0.11285488307476044,
      "eval_precision": 0.8210606638687524,
      "eval_recall": 0.8370283936211591,
      "eval_runtime": 4.1141,
      "eval_samples_per_second": 346.37,
      "eval_steps_per_second": 43.509,
      "step": 8010
    }
  ],
  "logging_steps": 500,
  "max_steps": 8010,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 1.064813249753856e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}