trainer_state.json

{
  "best_global_step": 264,
  "best_metric": 0.4426547884941101,
  "best_model_checkpoint": "bert_base_km_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": 7.741626739501953,
      "learning_rate": 4.9e-05,
      "loss": 0.3933,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8027522935779816,
      "eval_loss": 0.4426547884941101,
      "eval_runtime": 0.5773,
      "eval_samples_per_second": 1510.606,
      "eval_steps_per_second": 6.929,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.120843887329102,
      "learning_rate": 4.8e-05,
      "loss": 0.2191,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.801605504587156,
      "eval_loss": 0.5448803901672363,
      "eval_runtime": 0.5815,
      "eval_samples_per_second": 1499.484,
      "eval_steps_per_second": 6.878,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.026700973510742,
      "learning_rate": 4.7e-05,
      "loss": 0.1553,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7821100917431193,
      "eval_loss": 0.5948361754417419,
      "eval_runtime": 0.5667,
      "eval_samples_per_second": 1538.771,
      "eval_steps_per_second": 7.059,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 9.058463096618652,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.1123,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8142201834862385,
      "eval_loss": 0.7206083536148071,
      "eval_runtime": 0.5655,
      "eval_samples_per_second": 1542.078,
      "eval_steps_per_second": 7.074,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 12.111141204833984,
      "learning_rate": 4.5e-05,
      "loss": 0.0838,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.801605504587156,
      "eval_loss": 0.6722702383995056,
      "eval_runtime": 0.5592,
      "eval_samples_per_second": 1559.454,
      "eval_steps_per_second": 7.153,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 12.46601390838623,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0619,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.805045871559633,
      "eval_loss": 0.7456474900245667,
      "eval_runtime": 0.5677,
      "eval_samples_per_second": 1536.107,
      "eval_steps_per_second": 7.046,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "step": 1584,
      "total_flos": 5.316079940232192e+16,
      "train_loss": 0.17093069264383026,
      "train_runtime": 662.6134,
      "train_samples_per_second": 5082.073,
      "train_steps_per_second": 19.921
    }
  ],
  "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
}