trainer_state.json

{
  "best_metric": 0.7299185991287231,
  "best_model_checkpoint": "distilbert_lda_5_v1_wnli/checkpoint-6",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 21,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.3966708183288574,
      "learning_rate": 4.9e-05,
      "loss": 0.7316,
      "step": 3
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.30985915492957744,
      "eval_loss": 0.744938313961029,
      "eval_runtime": 0.0425,
      "eval_samples_per_second": 1670.119,
      "eval_steps_per_second": 23.523,
      "step": 3
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.5810788869857788,
      "learning_rate": 4.8e-05,
      "loss": 0.7137,
      "step": 6
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.4507042253521127,
      "eval_loss": 0.7299185991287231,
      "eval_runtime": 0.0402,
      "eval_samples_per_second": 1767.018,
      "eval_steps_per_second": 24.888,
      "step": 6
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.5314455032348633,
      "learning_rate": 4.7e-05,
      "loss": 0.7026,
      "step": 9
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.29577464788732394,
      "eval_loss": 0.7607282996177673,
      "eval_runtime": 0.0414,
      "eval_samples_per_second": 1715.106,
      "eval_steps_per_second": 24.156,
      "step": 9
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.062108278274536,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.6937,
      "step": 12
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.49295774647887325,
      "eval_loss": 0.7552266120910645,
      "eval_runtime": 0.0406,
      "eval_samples_per_second": 1747.237,
      "eval_steps_per_second": 24.609,
      "step": 12
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.7301479578018188,
      "learning_rate": 4.5e-05,
      "loss": 0.7014,
      "step": 15
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.2112676056338028,
      "eval_loss": 0.7826256155967712,
      "eval_runtime": 0.0424,
      "eval_samples_per_second": 1674.505,
      "eval_steps_per_second": 23.585,
      "step": 15
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.8389469385147095,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.6909,
      "step": 18
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.23943661971830985,
      "eval_loss": 0.8128851652145386,
      "eval_runtime": 0.0401,
      "eval_samples_per_second": 1771.918,
      "eval_steps_per_second": 24.957,
      "step": 18
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.9974502325057983,
      "learning_rate": 4.3e-05,
      "loss": 0.6894,
      "step": 21
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.2112676056338028,
      "eval_loss": 0.8085389733314514,
      "eval_runtime": 0.0406,
      "eval_samples_per_second": 1747.114,
      "eval_steps_per_second": 24.607,
      "step": 21
    },
    {
      "epoch": 7.0,
      "step": 21,
      "total_flos": 294408793512960.0,
      "train_loss": 0.7033271676018125,
      "train_runtime": 17.0611,
      "train_samples_per_second": 1860.963,
      "train_steps_per_second": 8.792
    }
  ],
  "logging_steps": 1,
  "max_steps": 150,
  "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": 294408793512960.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}