{
  "best_global_step": 23,
  "best_metric": 2.2278292179107666,
  "best_model_checkpoint": "tiny_bert_rand_10_v2_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": 9.029034614562988,
      "learning_rate": 4.9e-05,
      "loss": 3.1623,
      "step": 23
    },
    {
      "epoch": 1.0,
      "eval_combined_score": 0.10656599648165108,
      "eval_loss": 2.2278292179107666,
      "eval_pearson": 0.10286103935922952,
      "eval_runtime": 0.4487,
      "eval_samples_per_second": 3342.959,
      "eval_spearmanr": 0.11027095360407262,
      "eval_steps_per_second": 13.372,
      "step": 23
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.678215980529785,
      "learning_rate": 4.8e-05,
      "loss": 2.0283,
      "step": 46
    },
    {
      "epoch": 2.0,
      "eval_combined_score": 0.06990612797290352,
      "eval_loss": 2.643486261367798,
      "eval_pearson": 0.08081103017697638,
      "eval_runtime": 0.4948,
      "eval_samples_per_second": 3031.727,
      "eval_spearmanr": 0.05900122576883067,
      "eval_steps_per_second": 12.127,
      "step": 46
    },
    {
      "epoch": 3.0,
      "grad_norm": 5.944765567779541,
      "learning_rate": 4.7e-05,
      "loss": 1.8414,
      "step": 69
    },
    {
      "epoch": 3.0,
      "eval_combined_score": 0.1860037955190617,
      "eval_loss": 2.458224058151245,
      "eval_pearson": 0.18957216659773407,
      "eval_runtime": 0.5114,
      "eval_samples_per_second": 2933.331,
      "eval_spearmanr": 0.18243542444038938,
      "eval_steps_per_second": 11.733,
      "step": 69
    },
    {
      "epoch": 4.0,
      "grad_norm": 6.3005547523498535,
      "learning_rate": 4.600000000000001e-05,
      "loss": 1.6228,
      "step": 92
    },
    {
      "epoch": 4.0,
      "eval_combined_score": 0.24407603269699163,
      "eval_loss": 2.470104455947876,
      "eval_pearson": 0.2440971760192733,
      "eval_runtime": 0.4499,
      "eval_samples_per_second": 3333.776,
      "eval_spearmanr": 0.24405488937470993,
      "eval_steps_per_second": 13.335,
      "step": 92
    },
    {
      "epoch": 5.0,
      "grad_norm": 9.786120414733887,
      "learning_rate": 4.5e-05,
      "loss": 1.3269,
      "step": 115
    },
    {
      "epoch": 5.0,
      "eval_combined_score": 0.26661013917539245,
      "eval_loss": 2.4297025203704834,
      "eval_pearson": 0.26780639111822985,
      "eval_runtime": 0.4808,
      "eval_samples_per_second": 3119.606,
      "eval_spearmanr": 0.2654138872325551,
      "eval_steps_per_second": 12.478,
      "step": 115
    },
    {
      "epoch": 6.0,
      "grad_norm": 14.39130973815918,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 1.0809,
      "step": 138
    },
    {
      "epoch": 6.0,
      "eval_combined_score": 0.29916237306072646,
      "eval_loss": 2.4509849548339844,
      "eval_pearson": 0.29832754761397473,
      "eval_runtime": 0.4707,
      "eval_samples_per_second": 3186.462,
      "eval_spearmanr": 0.29999719850747825,
      "eval_steps_per_second": 12.746,
      "step": 138
    },
    {
      "epoch": 6.0,
      "step": 138,
      "total_flos": 904528923227136.0,
      "train_loss": 1.8437776703765427,
      "train_runtime": 25.7958,
      "train_samples_per_second": 11143.266,
      "train_steps_per_second": 44.581
    }
  ],
  "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": 904528923227136.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}