{
  "best_metric": 0.6503068804740906,
  "best_model_checkpoint": "tiny_bert_rand_100_v1_qnli/checkpoint-410",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 2460,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.4773553609848022,
      "learning_rate": 4.9e-05,
      "loss": 0.6657,
      "step": 410
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.61925681859784,
      "eval_loss": 0.6503068804740906,
      "eval_runtime": 1.9736,
      "eval_samples_per_second": 2768.051,
      "eval_steps_per_second": 11.147,
      "step": 410
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.6680256128311157,
      "learning_rate": 4.8e-05,
      "loss": 0.6367,
      "step": 820
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6212703642687168,
      "eval_loss": 0.6538947820663452,
      "eval_runtime": 1.9585,
      "eval_samples_per_second": 2789.444,
      "eval_steps_per_second": 11.233,
      "step": 820
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.4384894371032715,
      "learning_rate": 4.7e-05,
      "loss": 0.5927,
      "step": 1230
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6128500823723229,
      "eval_loss": 0.6680212616920471,
      "eval_runtime": 2.1951,
      "eval_samples_per_second": 2488.7,
      "eval_steps_per_second": 10.022,
      "step": 1230
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.7538044452667236,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.531,
      "step": 1640
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6115687351272194,
      "eval_loss": 0.7184457182884216,
      "eval_runtime": 1.9482,
      "eval_samples_per_second": 2804.177,
      "eval_steps_per_second": 11.293,
      "step": 1640
    },
    {
      "epoch": 5.0,
      "grad_norm": 5.5930867195129395,
      "learning_rate": 4.5e-05,
      "loss": 0.4631,
      "step": 2050
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6058941973274757,
      "eval_loss": 0.8040358424186707,
      "eval_runtime": 1.9417,
      "eval_samples_per_second": 2813.526,
      "eval_steps_per_second": 11.33,
      "step": 2050
    },
    {
      "epoch": 6.0,
      "grad_norm": 5.124334335327148,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.3972,
      "step": 2460
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6069924949661358,
      "eval_loss": 0.9130538702011108,
      "eval_runtime": 1.9358,
      "eval_samples_per_second": 2822.07,
      "eval_steps_per_second": 11.365,
      "step": 2460
    },
    {
      "epoch": 6.0,
      "step": 2460,
      "total_flos": 1.6480417453111296e+16,
      "train_loss": 0.5477293495240251,
      "train_runtime": 433.0019,
      "train_samples_per_second": 12094.98,
      "train_steps_per_second": 47.344
    }
  ],
  "logging_steps": 1,
  "max_steps": 20500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "total_flos": 1.6480417453111296e+16,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}