{
  "best_metric": 0.7247127294540405,
  "best_model_checkpoint": "bert_tiny_lda_20_v1_mnli/checkpoint-6136",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 13806,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.368187427520752,
      "learning_rate": 4.9e-05,
      "loss": 0.9696,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6101884870096791,
      "eval_loss": 0.8634526133537292,
      "eval_runtime": 3.0133,
      "eval_samples_per_second": 3257.2,
      "eval_steps_per_second": 12.943,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.6427279710769653,
      "learning_rate": 4.8e-05,
      "loss": 0.8307,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.650127356087621,
      "eval_loss": 0.7848993539810181,
      "eval_runtime": 2.9642,
      "eval_samples_per_second": 3311.169,
      "eval_steps_per_second": 13.157,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.6854288578033447,
      "learning_rate": 4.7e-05,
      "loss": 0.7523,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6728476821192053,
      "eval_loss": 0.7466796040534973,
      "eval_runtime": 3.0301,
      "eval_samples_per_second": 3239.134,
      "eval_steps_per_second": 12.871,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.0163750648498535,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.6962,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6861946001018848,
      "eval_loss": 0.7247127294540405,
      "eval_runtime": 2.956,
      "eval_samples_per_second": 3320.356,
      "eval_steps_per_second": 13.193,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.9129326343536377,
      "learning_rate": 4.5e-05,
      "loss": 0.6472,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6956698930208864,
      "eval_loss": 0.7247539162635803,
      "eval_runtime": 3.0432,
      "eval_samples_per_second": 3225.221,
      "eval_steps_per_second": 12.815,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.0681214332580566,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.6032,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6984207845134998,
      "eval_loss": 0.7455375790596008,
      "eval_runtime": 3.0056,
      "eval_samples_per_second": 3265.623,
      "eval_steps_per_second": 12.976,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.9462851285934448,
      "learning_rate": 4.3e-05,
      "loss": 0.5606,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6987264391237901,
      "eval_loss": 0.7510461807250977,
      "eval_runtime": 3.0013,
      "eval_samples_per_second": 3270.26,
      "eval_steps_per_second": 12.994,
      "step": 10738
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.2662479877471924,
      "learning_rate": 4.2e-05,
      "loss": 0.5204,
      "step": 12272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.691492613346918,
      "eval_loss": 0.7848535776138306,
      "eval_runtime": 2.9767,
      "eval_samples_per_second": 3297.3,
      "eval_steps_per_second": 13.102,
      "step": 12272
    },
    {
      "epoch": 9.0,
      "grad_norm": 2.846130847930908,
      "learning_rate": 4.1e-05,
      "loss": 0.4808,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6962812022414672,
      "eval_loss": 0.8428312540054321,
      "eval_runtime": 2.9346,
      "eval_samples_per_second": 3344.624,
      "eval_steps_per_second": 13.29,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "step": 13806,
      "total_flos": 9.268524907513344e+16,
      "train_loss": 0.6734655303711928,
      "train_runtime": 1930.2365,
      "train_samples_per_second": 10172.38,
      "train_steps_per_second": 39.736
    }
  ],
  "logging_steps": 1,
  "max_steps": 76700,
  "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": 9.268524907513344e+16,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}