{
  "best_global_step": 4266,
  "best_metric": 0.3953316807746887,
  "best_model_checkpoint": "bert_base_rand_50_v2_qqp/checkpoint-4266",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 11376,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.5561025142669678,
      "learning_rate": 4.9e-05,
      "loss": 0.4729,
      "step": 1422
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7941380163245115,
      "eval_combined_score": 0.7416441715609485,
      "eval_f1": 0.6891503267973856,
      "eval_loss": 0.43494340777397156,
      "eval_runtime": 25.783,
      "eval_samples_per_second": 1568.086,
      "eval_steps_per_second": 6.128,
      "step": 1422
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.604124069213867,
      "learning_rate": 4.8e-05,
      "loss": 0.3717,
      "step": 2844
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8183032401681919,
      "eval_combined_score": 0.7890482913286189,
      "eval_f1": 0.7597933424890458,
      "eval_loss": 0.39567166566848755,
      "eval_runtime": 25.6617,
      "eval_samples_per_second": 1575.5,
      "eval_steps_per_second": 6.157,
      "step": 2844
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.7459185123443604,
      "learning_rate": 4.7e-05,
      "loss": 0.2951,
      "step": 4266
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.827652733118971,
      "eval_combined_score": 0.7986680403877396,
      "eval_f1": 0.7696833476565083,
      "eval_loss": 0.3953316807746887,
      "eval_runtime": 25.8037,
      "eval_samples_per_second": 1566.833,
      "eval_steps_per_second": 6.123,
      "step": 4266
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.060959339141846,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.2327,
      "step": 5688
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8347514222112293,
      "eval_combined_score": 0.7992742438968468,
      "eval_f1": 0.7637970655824642,
      "eval_loss": 0.4646459221839905,
      "eval_runtime": 25.6379,
      "eval_samples_per_second": 1576.964,
      "eval_steps_per_second": 6.163,
      "step": 5688
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.4146206378936768,
      "learning_rate": 4.5e-05,
      "loss": 0.1833,
      "step": 7110
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8385110066782092,
      "eval_combined_score": 0.8084103707459838,
      "eval_f1": 0.7783097348137584,
      "eval_loss": 0.4751463830471039,
      "eval_runtime": 25.5271,
      "eval_samples_per_second": 1583.807,
      "eval_steps_per_second": 6.19,
      "step": 7110
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.024450302124023,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.145,
      "step": 8532
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8344051446945338,
      "eval_combined_score": 0.8098179767036009,
      "eval_f1": 0.785230808712668,
      "eval_loss": 0.5040478706359863,
      "eval_runtime": 25.6967,
      "eval_samples_per_second": 1573.356,
      "eval_steps_per_second": 6.149,
      "step": 8532
    },
    {
      "epoch": 7.0,
      "grad_norm": 7.6686692237854,
      "learning_rate": 4.3e-05,
      "loss": 0.1174,
      "step": 9954
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8348256245362354,
      "eval_combined_score": 0.8105853231460292,
      "eval_f1": 0.7863450217558229,
      "eval_loss": 0.6121538877487183,
      "eval_runtime": 25.7118,
      "eval_samples_per_second": 1572.427,
      "eval_steps_per_second": 6.145,
      "step": 9954
    },
    {
      "epoch": 8.0,
      "grad_norm": 5.302459239959717,
      "learning_rate": 4.2e-05,
      "loss": 0.0944,
      "step": 11376
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8387830818698986,
      "eval_combined_score": 0.8108233485170666,
      "eval_f1": 0.7828636151642347,
      "eval_loss": 0.6166710257530212,
      "eval_runtime": 25.6523,
      "eval_samples_per_second": 1576.075,
      "eval_steps_per_second": 6.159,
      "step": 11376
    },
    {
      "epoch": 8.0,
      "step": 11376,
      "total_flos": 3.8292762019405824e+17,
      "train_loss": 0.2390665867157626,
      "train_runtime": 4931.5872,
      "train_samples_per_second": 3688.934,
      "train_steps_per_second": 14.417
    }
  ],
  "logging_steps": 1,
  "max_steps": 71100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 5
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.8292762019405824e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}