trainer_state.json

{
  "best_metric": 1.4409527778625488,
  "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_logit_kd_data_aug_stsb/checkpoint-2518",
  "epoch": 6.0,
  "global_step": 15108,
  "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": 0.5057,
      "step": 2518
    },
    {
      "epoch": 1.0,
      "eval_combined_score": 0.17166336056520343,
      "eval_loss": 1.4409527778625488,
      "eval_pearson": 0.1663522931059537,
      "eval_runtime": 3.5524,
      "eval_samples_per_second": 422.25,
      "eval_spearmanr": 0.17697442802445312,
      "eval_steps_per_second": 3.378,
      "step": 2518
    },
    {
      "epoch": 2.0,
      "learning_rate": 4.8e-05,
      "loss": 0.2904,
      "step": 5036
    },
    {
      "epoch": 2.0,
      "eval_combined_score": 0.17197091361815903,
      "eval_loss": 1.5530579090118408,
      "eval_pearson": 0.1681095014225142,
      "eval_runtime": 3.5526,
      "eval_samples_per_second": 422.221,
      "eval_spearmanr": 0.17583232581380384,
      "eval_steps_per_second": 3.378,
      "step": 5036
    },
    {
      "epoch": 3.0,
      "learning_rate": 4.7e-05,
      "loss": 0.2164,
      "step": 7554
    },
    {
      "epoch": 3.0,
      "eval_combined_score": 0.1749035035726532,
      "eval_loss": 1.501253604888916,
      "eval_pearson": 0.1731732181456647,
      "eval_runtime": 3.5358,
      "eval_samples_per_second": 424.234,
      "eval_spearmanr": 0.1766337889996417,
      "eval_steps_per_second": 3.394,
      "step": 7554
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.1385,
      "step": 10072
    },
    {
      "epoch": 4.0,
      "eval_combined_score": 0.18374625206094036,
      "eval_loss": 1.4792943000793457,
      "eval_pearson": 0.185429786113599,
      "eval_runtime": 3.6273,
      "eval_samples_per_second": 413.536,
      "eval_spearmanr": 0.18206271800828172,
      "eval_steps_per_second": 3.308,
      "step": 10072
    },
    {
      "epoch": 5.0,
      "learning_rate": 4.5e-05,
      "loss": 0.0944,
      "step": 12590
    },
    {
      "epoch": 5.0,
      "eval_combined_score": 0.17174984728725837,
      "eval_loss": 1.5299888849258423,
      "eval_pearson": 0.16942220982783884,
      "eval_runtime": 3.532,
      "eval_samples_per_second": 424.685,
      "eval_spearmanr": 0.1740774847466779,
      "eval_steps_per_second": 3.397,
      "step": 12590
    },
    {
      "epoch": 6.0,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0682,
      "step": 15108
    },
    {
      "epoch": 6.0,
      "eval_combined_score": 0.16931633700280446,
      "eval_loss": 1.5758521556854248,
      "eval_pearson": 0.16949328163034338,
      "eval_runtime": 3.5289,
      "eval_samples_per_second": 425.062,
      "eval_spearmanr": 0.16913939237526554,
      "eval_steps_per_second": 3.4,
      "step": 15108
    },
    {
      "epoch": 6.0,
      "step": 15108,
      "total_flos": 6.062137602893414e+16,
      "train_loss": 0.21894432231487973,
      "train_runtime": 11653.2321,
      "train_samples_per_second": 1382.651,
      "train_steps_per_second": 10.804
    }
  ],
  "max_steps": 125900,
  "num_train_epochs": 50,
  "total_flos": 6.062137602893414e+16,
  "trial_name": null,
  "trial_params": null
}