trainer_state.json

{
  "best_metric": 2.256788492202759,
  "best_model_checkpoint": "distilbert_lda_5_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": 6.748433589935303,
      "learning_rate": 0.00098,
      "loss": 6.3938,
      "step": 23
    },
    {
      "epoch": 1.0,
      "eval_combined_score": NaN,
      "eval_loss": 2.256788492202759,
      "eval_pearson": NaN,
      "eval_runtime": 0.6384,
      "eval_samples_per_second": 2349.63,
      "eval_spearmanr": NaN,
      "eval_steps_per_second": 9.399,
      "step": 23
    },
    {
      "epoch": 2.0,
      "grad_norm": 18.032630920410156,
      "learning_rate": 0.00096,
      "loss": 2.2292,
      "step": 46
    },
    {
      "epoch": 2.0,
      "eval_combined_score": NaN,
      "eval_loss": 2.2729742527008057,
      "eval_pearson": NaN,
      "eval_runtime": 0.636,
      "eval_samples_per_second": 2358.504,
      "eval_spearmanr": NaN,
      "eval_steps_per_second": 9.434,
      "step": 46
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.141530990600586,
      "learning_rate": 0.00094,
      "loss": 2.2087,
      "step": 69
    },
    {
      "epoch": 3.0,
      "eval_combined_score": -0.03073418981930874,
      "eval_loss": 2.3179895877838135,
      "eval_pearson": -0.030123554701908854,
      "eval_runtime": 0.6514,
      "eval_samples_per_second": 2302.65,
      "eval_spearmanr": -0.03134482493670863,
      "eval_steps_per_second": 9.211,
      "step": 69
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.1647350788116455,
      "learning_rate": 0.00092,
      "loss": 2.2043,
      "step": 92
    },
    {
      "epoch": 4.0,
      "eval_combined_score": 0.009516354467436434,
      "eval_loss": 2.654869794845581,
      "eval_pearson": 0.007963371211781475,
      "eval_runtime": 0.6535,
      "eval_samples_per_second": 2295.416,
      "eval_spearmanr": 0.011069337723091394,
      "eval_steps_per_second": 9.182,
      "step": 92
    },
    {
      "epoch": 5.0,
      "grad_norm": 14.778202056884766,
      "learning_rate": 0.0009000000000000001,
      "loss": 2.2273,
      "step": 115
    },
    {
      "epoch": 5.0,
      "eval_combined_score": 0.0019123765153302998,
      "eval_loss": 2.615886688232422,
      "eval_pearson": 0.0017619346721126088,
      "eval_runtime": 0.652,
      "eval_samples_per_second": 2300.625,
      "eval_spearmanr": 0.002062818358547991,
      "eval_steps_per_second": 9.202,
      "step": 115
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.5717430114746094,
      "learning_rate": 0.00088,
      "loss": 2.1946,
      "step": 138
    },
    {
      "epoch": 6.0,
      "eval_combined_score": NaN,
      "eval_loss": 2.3650248050689697,
      "eval_pearson": NaN,
      "eval_runtime": 0.6498,
      "eval_samples_per_second": 2308.569,
      "eval_spearmanr": NaN,
      "eval_steps_per_second": 9.234,
      "step": 138
    },
    {
      "epoch": 6.0,
      "step": 138,
      "total_flos": 2284624480859136.0,
      "train_loss": 2.9096765656402144,
      "train_runtime": 46.2023,
      "train_samples_per_second": 6221.552,
      "train_steps_per_second": 24.891
    }
  ],
  "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": 2284624480859136.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}