trainer_state.json

{
  "best_metric": 0.8799816370010376,
  "best_model_checkpoint": "bert_tiny_olda_book_10_v1_stsb/checkpoint-115",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 230,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 11.179266929626465,
      "learning_rate": 4.9e-05,
      "loss": 3.0028,
      "step": 23
    },
    {
      "epoch": 1.0,
      "eval_combined_score": 0.16405123813632694,
      "eval_loss": 2.521865129470825,
      "eval_pearson": 0.16436790479245705,
      "eval_runtime": 0.3317,
      "eval_samples_per_second": 4521.686,
      "eval_spearmanr": 0.16373457148019682,
      "eval_steps_per_second": 18.087,
      "step": 23
    },
    {
      "epoch": 2.0,
      "grad_norm": 13.332984924316406,
      "learning_rate": 4.8e-05,
      "loss": 1.7825,
      "step": 46
    },
    {
      "epoch": 2.0,
      "eval_combined_score": 0.6438387957292973,
      "eval_loss": 1.7353206872940063,
      "eval_pearson": 0.6315278911071724,
      "eval_runtime": 0.3645,
      "eval_samples_per_second": 4115.581,
      "eval_spearmanr": 0.6561497003514221,
      "eval_steps_per_second": 16.462,
      "step": 46
    },
    {
      "epoch": 3.0,
      "grad_norm": 14.926468849182129,
      "learning_rate": 4.7e-05,
      "loss": 1.2017,
      "step": 69
    },
    {
      "epoch": 3.0,
      "eval_combined_score": 0.7305982694859476,
      "eval_loss": 1.1421477794647217,
      "eval_pearson": 0.7242637175185412,
      "eval_runtime": 0.3335,
      "eval_samples_per_second": 4498.087,
      "eval_spearmanr": 0.736932821453354,
      "eval_steps_per_second": 17.992,
      "step": 69
    },
    {
      "epoch": 4.0,
      "grad_norm": 7.934566497802734,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.8992,
      "step": 92
    },
    {
      "epoch": 4.0,
      "eval_combined_score": 0.7613350326098534,
      "eval_loss": 1.0969609022140503,
      "eval_pearson": 0.7549617016126466,
      "eval_runtime": 0.3463,
      "eval_samples_per_second": 4331.381,
      "eval_spearmanr": 0.7677083636070602,
      "eval_steps_per_second": 17.326,
      "step": 92
    },
    {
      "epoch": 5.0,
      "grad_norm": 6.861032962799072,
      "learning_rate": 4.5e-05,
      "loss": 0.6849,
      "step": 115
    },
    {
      "epoch": 5.0,
      "eval_combined_score": 0.7896674148128824,
      "eval_loss": 0.8799816370010376,
      "eval_pearson": 0.7898520417474288,
      "eval_runtime": 0.3331,
      "eval_samples_per_second": 4502.838,
      "eval_spearmanr": 0.7894827878783358,
      "eval_steps_per_second": 18.011,
      "step": 115
    },
    {
      "epoch": 6.0,
      "grad_norm": 10.177314758300781,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.5834,
      "step": 138
    },
    {
      "epoch": 6.0,
      "eval_combined_score": 0.7971524650979971,
      "eval_loss": 0.8917778134346008,
      "eval_pearson": 0.7964934549644671,
      "eval_runtime": 0.3323,
      "eval_samples_per_second": 4513.871,
      "eval_spearmanr": 0.7978114752315271,
      "eval_steps_per_second": 18.055,
      "step": 138
    },
    {
      "epoch": 7.0,
      "grad_norm": 6.432939052581787,
      "learning_rate": 4.3e-05,
      "loss": 0.4852,
      "step": 161
    },
    {
      "epoch": 7.0,
      "eval_combined_score": 0.7956582562495537,
      "eval_loss": 0.975614070892334,
      "eval_pearson": 0.7947805354894133,
      "eval_runtime": 0.3317,
      "eval_samples_per_second": 4522.73,
      "eval_spearmanr": 0.796535977009694,
      "eval_steps_per_second": 18.091,
      "step": 161
    },
    {
      "epoch": 8.0,
      "grad_norm": 14.945619583129883,
      "learning_rate": 4.2e-05,
      "loss": 0.4346,
      "step": 184
    },
    {
      "epoch": 8.0,
      "eval_combined_score": 0.7863762406665387,
      "eval_loss": 0.8956815600395203,
      "eval_pearson": 0.7867420462420818,
      "eval_runtime": 0.3364,
      "eval_samples_per_second": 4458.854,
      "eval_spearmanr": 0.7860104350909956,
      "eval_steps_per_second": 17.835,
      "step": 184
    },
    {
      "epoch": 9.0,
      "grad_norm": 15.58797550201416,
      "learning_rate": 4.1e-05,
      "loss": 0.3871,
      "step": 207
    },
    {
      "epoch": 9.0,
      "eval_combined_score": 0.7890857599874309,
      "eval_loss": 0.9086082577705383,
      "eval_pearson": 0.7899781947887469,
      "eval_runtime": 0.3317,
      "eval_samples_per_second": 4521.966,
      "eval_spearmanr": 0.788193325186115,
      "eval_steps_per_second": 18.088,
      "step": 207
    },
    {
      "epoch": 10.0,
      "grad_norm": 11.382484436035156,
      "learning_rate": 4e-05,
      "loss": 0.3449,
      "step": 230
    },
    {
      "epoch": 10.0,
      "eval_combined_score": 0.7886464144879618,
      "eval_loss": 1.0219333171844482,
      "eval_pearson": 0.7874362728579339,
      "eval_runtime": 0.3322,
      "eval_samples_per_second": 4515.484,
      "eval_spearmanr": 0.7898565561179897,
      "eval_steps_per_second": 18.062,
      "step": 230
    },
    {
      "epoch": 10.0,
      "step": 230,
      "total_flos": 1507548205378560.0,
      "train_loss": 0.9806491292041281,
      "train_runtime": 32.0715,
      "train_samples_per_second": 8962.796,
      "train_steps_per_second": 35.857
    }
  ],
  "logging_steps": 1,
  "max_steps": 1150,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 5
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1507548205378560.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}