{
  "best_global_step": 7670,
  "best_metric": 0.8617857098579407,
  "best_model_checkpoint": "tiny_bert_rand_50_v2_mnli/checkpoint-7670",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 15340,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.2021228075027466,
      "learning_rate": 4.9e-05,
      "loss": 0.9887,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5562913907284768,
      "eval_loss": 0.9252273440361023,
      "eval_runtime": 2.8614,
      "eval_samples_per_second": 3430.107,
      "eval_steps_per_second": 13.63,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.3760755062103271,
      "learning_rate": 4.8e-05,
      "loss": 0.9027,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5815588385124809,
      "eval_loss": 0.8902696967124939,
      "eval_runtime": 2.7981,
      "eval_samples_per_second": 3507.682,
      "eval_steps_per_second": 13.938,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.5229641199111938,
      "learning_rate": 4.7e-05,
      "loss": 0.8517,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6066225165562914,
      "eval_loss": 0.8663865327835083,
      "eval_runtime": 2.8132,
      "eval_samples_per_second": 3488.892,
      "eval_steps_per_second": 13.863,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.7879924774169922,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.8069,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6093734080489047,
      "eval_loss": 0.8639984130859375,
      "eval_runtime": 2.823,
      "eval_samples_per_second": 3476.774,
      "eval_steps_per_second": 13.815,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.73756742477417,
      "learning_rate": 4.5e-05,
      "loss": 0.7659,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6140601120733571,
      "eval_loss": 0.8617857098579407,
      "eval_runtime": 2.7858,
      "eval_samples_per_second": 3523.184,
      "eval_steps_per_second": 13.999,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.468418598175049,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.7247,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6189505858380031,
      "eval_loss": 0.8869032263755798,
      "eval_runtime": 2.7964,
      "eval_samples_per_second": 3509.916,
      "eval_steps_per_second": 13.947,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.2936809062957764,
      "learning_rate": 4.3e-05,
      "loss": 0.6864,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.624350483953133,
      "eval_loss": 0.8765807151794434,
      "eval_runtime": 2.8222,
      "eval_samples_per_second": 3477.815,
      "eval_steps_per_second": 13.819,
      "step": 10738
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.6551053524017334,
      "learning_rate": 4.2e-05,
      "loss": 0.6468,
      "step": 12272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6184411614875192,
      "eval_loss": 0.9364959001541138,
      "eval_runtime": 2.7836,
      "eval_samples_per_second": 3526.01,
      "eval_steps_per_second": 14.011,
      "step": 12272
    },
    {
      "epoch": 9.0,
      "grad_norm": 2.8846616744995117,
      "learning_rate": 4.1e-05,
      "loss": 0.6076,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6210901681100357,
      "eval_loss": 0.9490213394165039,
      "eval_runtime": 2.9527,
      "eval_samples_per_second": 3324.092,
      "eval_steps_per_second": 13.208,
      "step": 13806
    },
    {
      "epoch": 10.0,
      "grad_norm": 3.2129249572753906,
      "learning_rate": 4e-05,
      "loss": 0.5709,
      "step": 15340
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.6202750891492613,
      "eval_loss": 0.9911266565322876,
      "eval_runtime": 2.7898,
      "eval_samples_per_second": 3518.165,
      "eval_steps_per_second": 13.979,
      "step": 15340
    },
    {
      "epoch": 10.0,
      "step": 15340,
      "total_flos": 1.029836100834816e+17,
      "train_loss": 0.7552322765683568,
      "train_runtime": 2078.3043,
      "train_samples_per_second": 9447.654,
      "train_steps_per_second": 36.905
    }
  ],
  "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": 1.029836100834816e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}