trainer_state.json

{
  "best_metric": 0.5071280002593994,
  "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_qqp_128/checkpoint-11372",
  "epoch": 9.0,
  "global_step": 25587,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 4.9e-05,
      "loss": 0.6507,
      "step": 2843
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6318327974276527,
      "eval_combined_score": 0.3159163987138264,
      "eval_f1": 0.0,
      "eval_loss": 0.6496740579605103,
      "eval_runtime": 70.2237,
      "eval_samples_per_second": 575.732,
      "eval_steps_per_second": 4.5,
      "step": 2843
    },
    {
      "epoch": 2.0,
      "learning_rate": 4.8e-05,
      "loss": 0.6311,
      "step": 5686
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7259460796438288,
      "eval_combined_score": 0.6440872239388642,
      "eval_f1": 0.5622283682338997,
      "eval_loss": 0.5445142984390259,
      "eval_runtime": 70.5366,
      "eval_samples_per_second": 573.178,
      "eval_steps_per_second": 4.48,
      "step": 5686
    },
    {
      "epoch": 3.0,
      "learning_rate": 4.7e-05,
      "loss": 0.5153,
      "step": 8529
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7493198120207767,
      "eval_combined_score": 0.6692568011504313,
      "eval_f1": 0.5891937902800859,
      "eval_loss": 0.5152718424797058,
      "eval_runtime": 70.4498,
      "eval_samples_per_second": 573.884,
      "eval_steps_per_second": 4.485,
      "step": 8529
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.4912,
      "step": 11372
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7568142468464012,
      "eval_combined_score": 0.6964812852172944,
      "eval_f1": 0.6361483235881874,
      "eval_loss": 0.5071280002593994,
      "eval_runtime": 70.7466,
      "eval_samples_per_second": 571.476,
      "eval_steps_per_second": 4.467,
      "step": 11372
    },
    {
      "epoch": 5.0,
      "learning_rate": 4.5e-05,
      "loss": 0.4805,
      "step": 14215
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6318327974276527,
      "eval_combined_score": 0.3159163987138264,
      "eval_f1": 0.0,
      "eval_loss": NaN,
      "eval_runtime": 70.2577,
      "eval_samples_per_second": 575.453,
      "eval_steps_per_second": 4.498,
      "step": 14215
    },
    {
      "epoch": 6.0,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0,
      "step": 17058
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6318327974276527,
      "eval_combined_score": 0.3159163987138264,
      "eval_f1": 0.0,
      "eval_loss": NaN,
      "eval_runtime": 70.2761,
      "eval_samples_per_second": 575.302,
      "eval_steps_per_second": 4.497,
      "step": 17058
    },
    {
      "epoch": 7.0,
      "learning_rate": 4.3e-05,
      "loss": 0.0,
      "step": 19901
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6318327974276527,
      "eval_combined_score": 0.3159163987138264,
      "eval_f1": 0.0,
      "eval_loss": NaN,
      "eval_runtime": 69.9273,
      "eval_samples_per_second": 578.172,
      "eval_steps_per_second": 4.519,
      "step": 19901
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.2e-05,
      "loss": 0.0,
      "step": 22744
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6318327974276527,
      "eval_combined_score": 0.3159163987138264,
      "eval_f1": 0.0,
      "eval_loss": NaN,
      "eval_runtime": 70.3246,
      "eval_samples_per_second": 574.906,
      "eval_steps_per_second": 4.493,
      "step": 22744
    },
    {
      "epoch": 9.0,
      "learning_rate": 4.1e-05,
      "loss": 0.0,
      "step": 25587
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6318327974276527,
      "eval_combined_score": 0.3159163987138264,
      "eval_f1": 0.0,
      "eval_loss": NaN,
      "eval_runtime": 70.2338,
      "eval_samples_per_second": 575.649,
      "eval_steps_per_second": 4.499,
      "step": 25587
    },
    {
      "epoch": 9.0,
      "step": 25587,
      "total_flos": 7.817367256458854e+16,
      "train_loss": 0.30765416115544514,
      "train_runtime": 17142.0128,
      "train_samples_per_second": 1061.27,
      "train_steps_per_second": 8.292
    }
  ],
  "max_steps": 142150,
  "num_train_epochs": 50,
  "total_flos": 7.817367256458854e+16,
  "trial_name": null,
  "trial_params": null
}