trainer_state.json

{
  "best_metric": 0.8797919434527426,
  "best_model_checkpoint": "result/clhaif-roberta-large-refse",
  "epoch": 0.6963788300835655,
  "global_step": 750,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12,
      "eval_avg_sts": 0.8064963302246256,
      "eval_sickr_spearman": 0.762082012728511,
      "eval_stsb_spearman": 0.8509106477207403,
      "step": 125
    },
    {
      "epoch": 0.23,
      "eval_avg_sts": 0.8452781032001684,
      "eval_sickr_spearman": 0.8198200556739618,
      "eval_stsb_spearman": 0.8707361507263751,
      "step": 250
    },
    {
      "epoch": 0.35,
      "eval_avg_sts": 0.8271010163591597,
      "eval_sickr_spearman": 0.7789582685609077,
      "eval_stsb_spearman": 0.8752437641574118,
      "step": 375
    },
    {
      "epoch": 0.46,
      "learning_rate": 4.226245744351594e-05,
      "loss": 7.5855,
      "step": 500
    },
    {
      "epoch": 0.46,
      "eval_avg_sts": 0.8293696081597453,
      "eval_sickr_spearman": 0.7859766211341461,
      "eval_stsb_spearman": 0.8727625951853443,
      "step": 500
    },
    {
      "epoch": 0.58,
      "eval_avg_sts": 0.838054856261987,
      "eval_sickr_spearman": 0.8025079656271157,
      "eval_stsb_spearman": 0.8736017468968582,
      "step": 625
    },
    {
      "epoch": 0.7,
      "eval_avg_sts": 0.8370776376044988,
      "eval_sickr_spearman": 0.794363331756255,
      "eval_stsb_spearman": 0.8797919434527426,
      "step": 750
    }
  ],
  "max_steps": 3231,
  "num_train_epochs": 3,
  "total_flos": 0,
  "trial_name": null,
  "trial_params": null
}