trainer_state.json

{
  "best_global_step": 34,
  "best_metric": 0.6152296662330627,
  "best_model_checkpoint": "distilbert_rand_100_v1_cola/checkpoint-34",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 204,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.9416478872299194,
      "learning_rate": 4.9e-05,
      "loss": 0.6123,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6152296662330627,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.4443,
      "eval_samples_per_second": 2347.569,
      "eval_steps_per_second": 11.254,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.6123756170272827,
      "learning_rate": 4.8e-05,
      "loss": 0.5921,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.615736722946167,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.4639,
      "eval_samples_per_second": 2248.185,
      "eval_steps_per_second": 10.777,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.6057207584381104,
      "learning_rate": 4.7e-05,
      "loss": 0.5443,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6864812970161438,
      "eval_loss": 0.6180893778800964,
      "eval_matthews_correlation": 0.030962605224010897,
      "eval_runtime": 0.4771,
      "eval_samples_per_second": 2185.999,
      "eval_steps_per_second": 10.479,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.3933920860290527,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.4936,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6251198649406433,
      "eval_loss": 0.7308949828147888,
      "eval_matthews_correlation": 0.11488429746491452,
      "eval_runtime": 0.4344,
      "eval_samples_per_second": 2400.987,
      "eval_steps_per_second": 11.51,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.6650655269622803,
      "learning_rate": 4.5e-05,
      "loss": 0.4244,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6500479578971863,
      "eval_loss": 0.7332515120506287,
      "eval_matthews_correlation": 0.1287063919725079,
      "eval_runtime": 0.4509,
      "eval_samples_per_second": 2312.973,
      "eval_steps_per_second": 11.088,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.0416698455810547,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.3743,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6308724880218506,
      "eval_loss": 0.827803909778595,
      "eval_matthews_correlation": 0.11791046144111551,
      "eval_runtime": 0.4367,
      "eval_samples_per_second": 2388.279,
      "eval_steps_per_second": 11.449,
      "step": 204
    },
    {
      "epoch": 6.0,
      "step": 204,
      "total_flos": 3398186177722368.0,
      "train_loss": 0.506822478537466,
      "train_runtime": 57.6156,
      "train_samples_per_second": 7420.739,
      "train_steps_per_second": 29.506
    }
  ],
  "logging_steps": 1,
  "max_steps": 1700,
  "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": 3398186177722368.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}