trainer_state.json

{
  "best_metric": 2.2568883895874023,
  "best_model_checkpoint": "tiny_bert_rand_50_v1_stsb/checkpoint-23",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 138,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 6.64505672454834,
      "learning_rate": 4.9e-05,
      "loss": 3.4713,
      "step": 23
    },
    {
      "epoch": 1.0,
      "eval_combined_score": 0.11346061932108288,
      "eval_loss": 2.2568883895874023,
      "eval_pearson": 0.11480426487286079,
      "eval_runtime": 0.6984,
      "eval_samples_per_second": 2147.658,
      "eval_spearmanr": 0.11211697376930498,
      "eval_steps_per_second": 8.591,
      "step": 23
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.127440452575684,
      "learning_rate": 4.8e-05,
      "loss": 2.0275,
      "step": 46
    },
    {
      "epoch": 2.0,
      "eval_combined_score": 0.08747432980447115,
      "eval_loss": 2.6621973514556885,
      "eval_pearson": 0.09797904028841499,
      "eval_runtime": 0.6495,
      "eval_samples_per_second": 2309.362,
      "eval_spearmanr": 0.0769696193205273,
      "eval_steps_per_second": 9.237,
      "step": 46
    },
    {
      "epoch": 3.0,
      "grad_norm": 6.796934604644775,
      "learning_rate": 4.7e-05,
      "loss": 1.8912,
      "step": 69
    },
    {
      "epoch": 3.0,
      "eval_combined_score": 0.1679648923583315,
      "eval_loss": 2.4156179428100586,
      "eval_pearson": 0.1744431045329717,
      "eval_runtime": 0.5709,
      "eval_samples_per_second": 2627.537,
      "eval_spearmanr": 0.16148668018369133,
      "eval_steps_per_second": 10.51,
      "step": 69
    },
    {
      "epoch": 4.0,
      "grad_norm": 7.0087785720825195,
      "learning_rate": 4.600000000000001e-05,
      "loss": 1.6966,
      "step": 92
    },
    {
      "epoch": 4.0,
      "eval_combined_score": 0.211070610209586,
      "eval_loss": 2.5946474075317383,
      "eval_pearson": 0.21027160965287675,
      "eval_runtime": 0.5525,
      "eval_samples_per_second": 2714.697,
      "eval_spearmanr": 0.2118696107662953,
      "eval_steps_per_second": 10.859,
      "step": 92
    },
    {
      "epoch": 5.0,
      "grad_norm": 17.69513702392578,
      "learning_rate": 4.5e-05,
      "loss": 1.4665,
      "step": 115
    },
    {
      "epoch": 5.0,
      "eval_combined_score": 0.231848067216258,
      "eval_loss": 2.562957286834717,
      "eval_pearson": 0.22792827229288598,
      "eval_runtime": 0.5572,
      "eval_samples_per_second": 2692.232,
      "eval_spearmanr": 0.23576786213963,
      "eval_steps_per_second": 10.769,
      "step": 115
    },
    {
      "epoch": 6.0,
      "grad_norm": 6.968521595001221,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 1.2375,
      "step": 138
    },
    {
      "epoch": 6.0,
      "eval_combined_score": 0.259776621258329,
      "eval_loss": 2.49996018409729,
      "eval_pearson": 0.2583030902504291,
      "eval_runtime": 0.5756,
      "eval_samples_per_second": 2605.763,
      "eval_spearmanr": 0.26125015226622894,
      "eval_steps_per_second": 10.423,
      "step": 138
    },
    {
      "epoch": 6.0,
      "step": 138,
      "total_flos": 904528923227136.0,
      "train_loss": 1.9650979664014734,
      "train_runtime": 31.6691,
      "train_samples_per_second": 9076.676,
      "train_steps_per_second": 36.313
    }
  ],
  "logging_steps": 1,
  "max_steps": 1150,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "total_flos": 904528923227136.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}