trainer_state.json

{
  "best_global_step": 10,
  "best_metric": 0.6894178986549377,
  "best_model_checkpoint": "bert_base_rand_50_v2_rte/checkpoint-10",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 60,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.888877511024475,
      "learning_rate": 4.9e-05,
      "loss": 0.7076,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5379061371841155,
      "eval_loss": 0.6894178986549377,
      "eval_runtime": 0.1945,
      "eval_samples_per_second": 1424.455,
      "eval_steps_per_second": 10.285,
      "step": 10
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.6448814868927002,
      "learning_rate": 4.8e-05,
      "loss": 0.6783,
      "step": 20
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5234657039711191,
      "eval_loss": 0.7175478339195251,
      "eval_runtime": 0.1941,
      "eval_samples_per_second": 1426.904,
      "eval_steps_per_second": 10.303,
      "step": 20
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.2871806621551514,
      "learning_rate": 4.7e-05,
      "loss": 0.6307,
      "step": 30
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5342960288808665,
      "eval_loss": 0.7913293838500977,
      "eval_runtime": 0.192,
      "eval_samples_per_second": 1442.551,
      "eval_steps_per_second": 10.416,
      "step": 30
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.8225648403167725,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.5137,
      "step": 40
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5090252707581228,
      "eval_loss": 1.0964628458023071,
      "eval_runtime": 0.2079,
      "eval_samples_per_second": 1332.391,
      "eval_steps_per_second": 9.62,
      "step": 40
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.495018482208252,
      "learning_rate": 4.5e-05,
      "loss": 0.372,
      "step": 50
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.48014440433212996,
      "eval_loss": 1.204417109489441,
      "eval_runtime": 0.1939,
      "eval_samples_per_second": 1428.734,
      "eval_steps_per_second": 10.316,
      "step": 50
    },
    {
      "epoch": 6.0,
      "grad_norm": 10.629965782165527,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.2719,
      "step": 60
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.48375451263537905,
      "eval_loss": 1.6170828342437744,
      "eval_runtime": 0.205,
      "eval_samples_per_second": 1351.207,
      "eval_steps_per_second": 9.756,
      "step": 60
    },
    {
      "epoch": 6.0,
      "step": 60,
      "total_flos": 1965439583539200.0,
      "train_loss": 0.5290212949117025,
      "train_runtime": 38.9028,
      "train_samples_per_second": 3200.288,
      "train_steps_per_second": 12.853
    }
  ],
  "logging_steps": 1,
  "max_steps": 500,
  "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": 1965439583539200.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}