{
  "best_metric": 0.6857091188430786,
  "best_model_checkpoint": "tiny_bert_rand_50_v1_rte/checkpoint-30",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 80,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.4516654312610626,
      "learning_rate": 4.9e-05,
      "loss": 0.6992,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5631768953068592,
      "eval_loss": 0.6891781091690063,
      "eval_runtime": 0.1139,
      "eval_samples_per_second": 2431.038,
      "eval_steps_per_second": 17.553,
      "step": 10
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.5379452705383301,
      "learning_rate": 4.8e-05,
      "loss": 0.6919,
      "step": 20
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5379061371841155,
      "eval_loss": 0.6864846348762512,
      "eval_runtime": 0.1259,
      "eval_samples_per_second": 2200.333,
      "eval_steps_per_second": 15.887,
      "step": 20
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.4263160526752472,
      "learning_rate": 4.7e-05,
      "loss": 0.6803,
      "step": 30
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5595667870036101,
      "eval_loss": 0.6857091188430786,
      "eval_runtime": 0.122,
      "eval_samples_per_second": 2270.843,
      "eval_steps_per_second": 16.396,
      "step": 30
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.8484940528869629,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.6477,
      "step": 40
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.555956678700361,
      "eval_loss": 0.7186089754104614,
      "eval_runtime": 0.1098,
      "eval_samples_per_second": 2523.528,
      "eval_steps_per_second": 18.22,
      "step": 40
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.037821650505066,
      "learning_rate": 4.5e-05,
      "loss": 0.5941,
      "step": 50
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5523465703971119,
      "eval_loss": 0.7786869406700134,
      "eval_runtime": 0.1209,
      "eval_samples_per_second": 2290.916,
      "eval_steps_per_second": 16.541,
      "step": 50
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.3715161085128784,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.5146,
      "step": 60
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5415162454873647,
      "eval_loss": 0.8585395216941833,
      "eval_runtime": 0.122,
      "eval_samples_per_second": 2270.705,
      "eval_steps_per_second": 16.395,
      "step": 60
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.5981643199920654,
      "learning_rate": 4.3e-05,
      "loss": 0.3902,
      "step": 70
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.516245487364621,
      "eval_loss": 1.0804373025894165,
      "eval_runtime": 0.1155,
      "eval_samples_per_second": 2399.17,
      "eval_steps_per_second": 17.323,
      "step": 70
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.5909316539764404,
      "learning_rate": 4.2e-05,
      "loss": 0.2783,
      "step": 80
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 1.2460827827453613,
      "eval_runtime": 0.1156,
      "eval_samples_per_second": 2395.702,
      "eval_steps_per_second": 17.297,
      "step": 80
    },
    {
      "epoch": 8.0,
      "step": 80,
      "total_flos": 522373675991040.0,
      "train_loss": 0.5620430141687394,
      "train_runtime": 20.1921,
      "train_samples_per_second": 6165.772,
      "train_steps_per_second": 24.762
    }
  ],
  "logging_steps": 1,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "total_flos": 522373675991040.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}