run-4/checkpoint-1576/trainer_state.json

{
  "best_metric": 0.7722345747637577,
  "best_model_checkpoint": "tinybert-TG-HS-HX-parentpretrained\\run-4\\checkpoint-1182",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 1576,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 2.1327047597286744e-06,
      "loss": 0.3205,
      "step": 197
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7181767648693719,
      "eval_loss": 0.3056386411190033,
      "eval_runtime": 125.6972,
      "eval_samples_per_second": 57.249,
      "eval_steps_per_second": 0.453,
      "step": 197
    },
    {
      "epoch": 2.0,
      "learning_rate": 1.8280326511960069e-06,
      "loss": 0.3065,
      "step": 394
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7493051695386326,
      "eval_loss": 0.2966243028640747,
      "eval_runtime": 125.6168,
      "eval_samples_per_second": 57.285,
      "eval_steps_per_second": 0.454,
      "step": 394
    },
    {
      "epoch": 3.0,
      "learning_rate": 1.523360542663339e-06,
      "loss": 0.3012,
      "step": 591
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7602834908282379,
      "eval_loss": 0.292987197637558,
      "eval_runtime": 125.5971,
      "eval_samples_per_second": 57.294,
      "eval_steps_per_second": 0.454,
      "step": 591
    },
    {
      "epoch": 4.0,
      "learning_rate": 1.2186884341306712e-06,
      "loss": 0.2993,
      "step": 788
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7659811006114509,
      "eval_loss": 0.2911587357521057,
      "eval_runtime": 125.8135,
      "eval_samples_per_second": 57.196,
      "eval_steps_per_second": 0.453,
      "step": 788
    },
    {
      "epoch": 5.0,
      "learning_rate": 9.140163255980034e-07,
      "loss": 0.2978,
      "step": 985
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7695942190105615,
      "eval_loss": 0.2902683913707733,
      "eval_runtime": 125.6067,
      "eval_samples_per_second": 57.29,
      "eval_steps_per_second": 0.454,
      "step": 985
    },
    {
      "epoch": 6.0,
      "learning_rate": 6.093442170653356e-07,
      "loss": 0.2964,
      "step": 1182
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7722345747637577,
      "eval_loss": 0.2897145748138428,
      "eval_runtime": 125.7458,
      "eval_samples_per_second": 57.227,
      "eval_steps_per_second": 0.453,
      "step": 1182
    },
    {
      "epoch": 7.0,
      "learning_rate": 3.046721085326678e-07,
      "loss": 0.2959,
      "step": 1379
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7718176764869372,
      "eval_loss": 0.28957682847976685,
      "eval_runtime": 125.8315,
      "eval_samples_per_second": 57.188,
      "eval_steps_per_second": 0.453,
      "step": 1379
    },
    {
      "epoch": 8.0,
      "learning_rate": 0.0,
      "loss": 0.2961,
      "step": 1576
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7722345747637577,
      "eval_loss": 0.2894397974014282,
      "eval_runtime": 125.7279,
      "eval_samples_per_second": 57.235,
      "eval_steps_per_second": 0.453,
      "step": 1576
    }
  ],
  "logging_steps": 500,
  "max_steps": 1576,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 353904216589332.0,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.4889425582875847,
    "learning_rate": 2.4373768682613424e-06,
    "num_train_epochs": 8,
    "temperature": 30
  }
}