{
  "best_metric": 0.4788999557495117,
  "best_model_checkpoint": "hBERTv2_new_pretrain_48_emb_com_sst2/checkpoint-2635",
  "epoch": 10.0,
  "global_step": 5270,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 3.9200000000000004e-05,
      "loss": 0.4943,
      "step": 527
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7454128440366973,
      "eval_loss": 0.5216304063796997,
      "eval_runtime": 3.8434,
      "eval_samples_per_second": 226.88,
      "eval_steps_per_second": 1.821,
      "step": 527
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.8400000000000005e-05,
      "loss": 0.3354,
      "step": 1054
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7912844036697247,
      "eval_loss": 0.5365985631942749,
      "eval_runtime": 3.8234,
      "eval_samples_per_second": 228.068,
      "eval_steps_per_second": 1.831,
      "step": 1054
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.76e-05,
      "loss": 0.274,
      "step": 1581
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7981651376146789,
      "eval_loss": 0.5090646743774414,
      "eval_runtime": 3.8273,
      "eval_samples_per_second": 227.835,
      "eval_steps_per_second": 1.829,
      "step": 1581
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.680000000000001e-05,
      "loss": 0.2347,
      "step": 2108
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7970183486238532,
      "eval_loss": 0.5886006355285645,
      "eval_runtime": 3.8334,
      "eval_samples_per_second": 227.472,
      "eval_steps_per_second": 1.826,
      "step": 2108
    },
    {
      "epoch": 5.0,
      "learning_rate": 3.6e-05,
      "loss": 0.2094,
      "step": 2635
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.805045871559633,
      "eval_loss": 0.4788999557495117,
      "eval_runtime": 3.8295,
      "eval_samples_per_second": 227.708,
      "eval_steps_per_second": 1.828,
      "step": 2635
    },
    {
      "epoch": 6.0,
      "learning_rate": 3.52e-05,
      "loss": 0.1944,
      "step": 3162
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7993119266055045,
      "eval_loss": 0.5025409460067749,
      "eval_runtime": 3.8292,
      "eval_samples_per_second": 227.726,
      "eval_steps_per_second": 1.828,
      "step": 3162
    },
    {
      "epoch": 7.0,
      "learning_rate": 3.44e-05,
      "loss": 0.1826,
      "step": 3689
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7901376146788991,
      "eval_loss": 0.6511347889900208,
      "eval_runtime": 3.836,
      "eval_samples_per_second": 227.32,
      "eval_steps_per_second": 1.825,
      "step": 3689
    },
    {
      "epoch": 8.0,
      "learning_rate": 3.3600000000000004e-05,
      "loss": 0.1642,
      "step": 4216
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7993119266055045,
      "eval_loss": 0.5241345763206482,
      "eval_runtime": 3.8355,
      "eval_samples_per_second": 227.352,
      "eval_steps_per_second": 1.825,
      "step": 4216
    },
    {
      "epoch": 9.0,
      "learning_rate": 3.28e-05,
      "loss": 0.1516,
      "step": 4743
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.801605504587156,
      "eval_loss": 0.6333720088005066,
      "eval_runtime": 3.8228,
      "eval_samples_per_second": 228.103,
      "eval_steps_per_second": 1.831,
      "step": 4743
    },
    {
      "epoch": 10.0,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 0.1462,
      "step": 5270
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.7912844036697247,
      "eval_loss": 0.6750393509864807,
      "eval_runtime": 3.857,
      "eval_samples_per_second": 226.084,
      "eval_steps_per_second": 1.815,
      "step": 5270
    },
    {
      "epoch": 10.0,
      "step": 5270,
      "total_flos": 9.865913541066752e+16,
      "train_loss": 0.23867457786140225,
      "train_runtime": 9155.0475,
      "train_samples_per_second": 367.824,
      "train_steps_per_second": 2.878
    }
  ],
  "max_steps": 26350,
  "num_train_epochs": 50,
  "total_flos": 9.865913541066752e+16,
  "trial_name": null,
  "trial_params": null
}