{
  "best_global_step": 7670,
  "best_metric": 0.8055477738380432,
  "best_model_checkpoint": "bert_base_rand_20_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": 1.2353837490081787,
      "learning_rate": 4.9e-05,
      "loss": 0.9776,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5755476311767702,
      "eval_loss": 0.9095600843429565,
      "eval_runtime": 6.2444,
      "eval_samples_per_second": 1571.82,
      "eval_steps_per_second": 6.246,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.7086997032165527,
      "learning_rate": 4.8e-05,
      "loss": 0.8756,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6042791645440652,
      "eval_loss": 0.8629811406135559,
      "eval_runtime": 6.2245,
      "eval_samples_per_second": 1576.83,
      "eval_steps_per_second": 6.266,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.6689685583114624,
      "learning_rate": 4.7e-05,
      "loss": 0.7956,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6348446255731024,
      "eval_loss": 0.8167165517807007,
      "eval_runtime": 6.2259,
      "eval_samples_per_second": 1576.472,
      "eval_steps_per_second": 6.264,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.151376962661743,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.7175,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6471726948548141,
      "eval_loss": 0.8117183446884155,
      "eval_runtime": 6.2292,
      "eval_samples_per_second": 1575.652,
      "eval_steps_per_second": 6.261,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.3178389072418213,
      "learning_rate": 4.5e-05,
      "loss": 0.6438,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.658889454915945,
      "eval_loss": 0.8055477738380432,
      "eval_runtime": 6.2237,
      "eval_samples_per_second": 1577.032,
      "eval_steps_per_second": 6.266,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.5540552139282227,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.5715,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6651044319918492,
      "eval_loss": 0.8539048433303833,
      "eval_runtime": 6.2171,
      "eval_samples_per_second": 1578.707,
      "eval_steps_per_second": 6.273,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.5792064666748047,
      "learning_rate": 4.3e-05,
      "loss": 0.4957,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6585838003056547,
      "eval_loss": 0.9527242183685303,
      "eval_runtime": 6.205,
      "eval_samples_per_second": 1581.777,
      "eval_steps_per_second": 6.285,
      "step": 10738
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.162525177001953,
      "learning_rate": 4.2e-05,
      "loss": 0.4267,
      "step": 12272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6547121752419766,
      "eval_loss": 0.9706308841705322,
      "eval_runtime": 6.2206,
      "eval_samples_per_second": 1577.826,
      "eval_steps_per_second": 6.27,
      "step": 12272
    },
    {
      "epoch": 9.0,
      "grad_norm": 4.91460657119751,
      "learning_rate": 4.1e-05,
      "loss": 0.362,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6468670402445237,
      "eval_loss": 1.1230700016021729,
      "eval_runtime": 6.2048,
      "eval_samples_per_second": 1581.841,
      "eval_steps_per_second": 6.285,
      "step": 13806
    },
    {
      "epoch": 10.0,
      "grad_norm": 4.130703926086426,
      "learning_rate": 4e-05,
      "loss": 0.3054,
      "step": 15340
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.6572592969943963,
      "eval_loss": 1.1829255819320679,
      "eval_runtime": 6.2212,
      "eval_samples_per_second": 1577.676,
      "eval_steps_per_second": 6.269,
      "step": 15340
    },
    {
      "epoch": 10.0,
      "step": 15340,
      "total_flos": 5.166258268431053e+17,
      "train_loss": 0.6171293813215551,
      "train_runtime": 6430.1974,
      "train_samples_per_second": 3053.577,
      "train_steps_per_second": 11.928
    }
  ],
  "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": 5.166258268431053e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}