trainer_state.json

{
  "best_global_step": 264,
  "best_metric": 0.43877333402633667,
  "best_model_checkpoint": "bert_base_km_20_v2_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": 6.834911823272705,
      "learning_rate": 4.9e-05,
      "loss": 0.4155,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7981651376146789,
      "eval_loss": 0.43877333402633667,
      "eval_runtime": 0.579,
      "eval_samples_per_second": 1505.957,
      "eval_steps_per_second": 6.908,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 8.833080291748047,
      "learning_rate": 4.8e-05,
      "loss": 0.2288,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7912844036697247,
      "eval_loss": 0.5246180295944214,
      "eval_runtime": 0.5879,
      "eval_samples_per_second": 1483.241,
      "eval_steps_per_second": 6.804,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 6.723768711090088,
      "learning_rate": 4.7e-05,
      "loss": 0.1668,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7935779816513762,
      "eval_loss": 0.6083675026893616,
      "eval_runtime": 0.5769,
      "eval_samples_per_second": 1511.529,
      "eval_steps_per_second": 6.934,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 10.352004051208496,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.1248,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7947247706422018,
      "eval_loss": 0.6475391387939453,
      "eval_runtime": 0.5851,
      "eval_samples_per_second": 1490.38,
      "eval_steps_per_second": 6.837,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 10.663488388061523,
      "learning_rate": 4.5e-05,
      "loss": 0.0928,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7798165137614679,
      "eval_loss": 0.7931265234947205,
      "eval_runtime": 0.572,
      "eval_samples_per_second": 1524.394,
      "eval_steps_per_second": 6.993,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 10.436763763427734,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0692,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7924311926605505,
      "eval_loss": 0.876919150352478,
      "eval_runtime": 0.5731,
      "eval_samples_per_second": 1521.593,
      "eval_steps_per_second": 6.98,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "step": 1584,
      "total_flos": 5.316079940232192e+16,
      "train_loss": 0.182965358098348,
      "train_runtime": 663.2246,
      "train_samples_per_second": 5077.39,
      "train_steps_per_second": 19.903
    }
  ],
  "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
}