trainer_state.json

{
  "best_global_step": 264,
  "best_metric": 0.4303896725177765,
  "best_model_checkpoint": "bert_base_rand_10_v1_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": 8.92810344696045,
      "learning_rate": 4.9e-05,
      "loss": 0.386,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8038990825688074,
      "eval_loss": 0.4303896725177765,
      "eval_runtime": 0.5775,
      "eval_samples_per_second": 1509.935,
      "eval_steps_per_second": 6.926,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 10.491100311279297,
      "learning_rate": 4.8e-05,
      "loss": 0.2197,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8061926605504587,
      "eval_loss": 0.5060594081878662,
      "eval_runtime": 0.5726,
      "eval_samples_per_second": 1522.813,
      "eval_steps_per_second": 6.985,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.117835998535156,
      "learning_rate": 4.7e-05,
      "loss": 0.1621,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8061926605504587,
      "eval_loss": 0.5004409551620483,
      "eval_runtime": 0.5619,
      "eval_samples_per_second": 1551.767,
      "eval_steps_per_second": 7.118,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.461856842041016,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.1243,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8107798165137615,
      "eval_loss": 0.5717501640319824,
      "eval_runtime": 0.5637,
      "eval_samples_per_second": 1546.994,
      "eval_steps_per_second": 7.096,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 13.303771018981934,
      "learning_rate": 4.5e-05,
      "loss": 0.0976,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8119266055045872,
      "eval_loss": 0.6290585398674011,
      "eval_runtime": 0.5619,
      "eval_samples_per_second": 1551.885,
      "eval_steps_per_second": 7.119,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 6.121748447418213,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0756,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8107798165137615,
      "eval_loss": 0.7312332987785339,
      "eval_runtime": 0.5591,
      "eval_samples_per_second": 1559.678,
      "eval_steps_per_second": 7.154,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "step": 1584,
      "total_flos": 5.316079940232192e+16,
      "train_loss": 0.17754850604317404,
      "train_runtime": 663.2768,
      "train_samples_per_second": 5076.991,
      "train_steps_per_second": 19.901
    }
  ],
  "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
}