trainer_state.json

{
  "best_metric": 0.3312614858150482,
  "best_model_checkpoint": "bert_base_lda_20_v1_book_sst2/checkpoint-264",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1584,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 6.884793281555176,
      "learning_rate": 4.9e-05,
      "loss": 0.3964,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8704128440366973,
      "eval_loss": 0.3312614858150482,
      "eval_runtime": 0.5784,
      "eval_samples_per_second": 1507.629,
      "eval_steps_per_second": 6.916,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 13.503787994384766,
      "learning_rate": 4.8e-05,
      "loss": 0.1876,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8841743119266054,
      "eval_loss": 0.3377615511417389,
      "eval_runtime": 0.5699,
      "eval_samples_per_second": 1530.053,
      "eval_steps_per_second": 7.019,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.8918938636779785,
      "learning_rate": 4.7e-05,
      "loss": 0.1202,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8669724770642202,
      "eval_loss": 0.40393030643463135,
      "eval_runtime": 0.573,
      "eval_samples_per_second": 1521.855,
      "eval_steps_per_second": 6.981,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.135744333267212,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.0867,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8910550458715596,
      "eval_loss": 0.34491220116615295,
      "eval_runtime": 0.5842,
      "eval_samples_per_second": 1492.685,
      "eval_steps_per_second": 6.847,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 15.718326568603516,
      "learning_rate": 4.5e-05,
      "loss": 0.0667,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8818807339449541,
      "eval_loss": 0.3893832564353943,
      "eval_runtime": 0.5858,
      "eval_samples_per_second": 1488.608,
      "eval_steps_per_second": 6.828,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 6.3708720207214355,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0527,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8830275229357798,
      "eval_loss": 0.4835165739059448,
      "eval_runtime": 0.5758,
      "eval_samples_per_second": 1514.436,
      "eval_steps_per_second": 6.947,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "step": 1584,
      "total_flos": 5.316079940232192e+16,
      "train_loss": 0.15173852744728628,
      "train_runtime": 674.7007,
      "train_samples_per_second": 4991.028,
      "train_steps_per_second": 19.564
    }
  ],
  "logging_steps": 1,
  "max_steps": 13200,
  "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": 5.316079940232192e+16,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}