trainer_state.json

{
  "best_global_step": 264,
  "best_metric": 0.44483500719070435,
  "best_model_checkpoint": "distilbert_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": 5.650020122528076,
      "learning_rate": 4.9e-05,
      "loss": 0.4147,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8038990825688074,
      "eval_loss": 0.44483500719070435,
      "eval_runtime": 0.3679,
      "eval_samples_per_second": 2370.256,
      "eval_steps_per_second": 10.873,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 11.352852821350098,
      "learning_rate": 4.8e-05,
      "loss": 0.2287,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7993119266055045,
      "eval_loss": 0.5161756873130798,
      "eval_runtime": 0.3767,
      "eval_samples_per_second": 2314.884,
      "eval_steps_per_second": 10.619,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.9486942291259766,
      "learning_rate": 4.7e-05,
      "loss": 0.1685,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7947247706422018,
      "eval_loss": 0.5433734059333801,
      "eval_runtime": 0.367,
      "eval_samples_per_second": 2376.236,
      "eval_steps_per_second": 10.9,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.108094692230225,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.1285,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.801605504587156,
      "eval_loss": 0.6299866437911987,
      "eval_runtime": 0.3652,
      "eval_samples_per_second": 2387.846,
      "eval_steps_per_second": 10.953,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.361087799072266,
      "learning_rate": 4.5e-05,
      "loss": 0.0951,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7947247706422018,
      "eval_loss": 0.7009722590446472,
      "eval_runtime": 0.3691,
      "eval_samples_per_second": 2362.341,
      "eval_steps_per_second": 10.836,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.255533695220947,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0742,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7935779816513762,
      "eval_loss": 0.8115570545196533,
      "eval_runtime": 0.3639,
      "eval_samples_per_second": 2396.05,
      "eval_steps_per_second": 10.991,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "step": 1584,
      "total_flos": 2.676464049624883e+16,
      "train_loss": 0.1849725451132264,
      "train_runtime": 377.4684,
      "train_samples_per_second": 8921.144,
      "train_steps_per_second": 34.97
    }
  ],
  "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": 2.676464049624883e+16,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}