run-3/checkpoint-576/trainer_state.json

{
  "best_metric": 0.8416256812127808,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-stsb/run-3/checkpoint-576",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 576,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 1.0159441232681274,
      "eval_pearson": 0.7754149460856693,
      "eval_runtime": 0.9276,
      "eval_samples_per_second": 1617.073,
      "eval_spearmanr": 0.7770625188634587,
      "eval_steps_per_second": 101.337,
      "step": 144
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.7027656435966492,
      "eval_pearson": 0.8294526923897652,
      "eval_runtime": 1.1203,
      "eval_samples_per_second": 1338.921,
      "eval_spearmanr": 0.8239919782128128,
      "eval_steps_per_second": 83.906,
      "step": 288
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.6792955994606018,
      "eval_pearson": 0.8371525320180763,
      "eval_runtime": 1.1241,
      "eval_samples_per_second": 1334.415,
      "eval_spearmanr": 0.8358915789303897,
      "eval_steps_per_second": 83.623,
      "step": 432
    },
    {
      "epoch": 3.47,
      "grad_norm": 24.244155883789062,
      "learning_rate": 1.052427584566897e-05,
      "loss": 0.9809,
      "step": 500
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.6611331105232239,
      "eval_pearson": 0.8416256812127808,
      "eval_runtime": 1.1415,
      "eval_samples_per_second": 1314.012,
      "eval_spearmanr": 0.8371352595017586,
      "eval_steps_per_second": 82.345,
      "step": 576
    }
  ],
  "logging_steps": 500,
  "max_steps": 720,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 22488785838402.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 3.444308458582572e-05,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 4,
    "seed": 38
  }
}