trainer_state.json

{
  "best_global_step": 10,
  "best_metric": 0.6856103539466858,
  "best_model_checkpoint": "distilbert_rand_50_v1_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": 0.8824705481529236,
      "learning_rate": 4.9e-05,
      "loss": 0.6953,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5667870036101083,
      "eval_loss": 0.6856103539466858,
      "eval_runtime": 0.1255,
      "eval_samples_per_second": 2206.464,
      "eval_steps_per_second": 15.931,
      "step": 10
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.0209002494812012,
      "learning_rate": 4.8e-05,
      "loss": 0.6788,
      "step": 20
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5090252707581228,
      "eval_loss": 0.7198992967605591,
      "eval_runtime": 0.1263,
      "eval_samples_per_second": 2194.059,
      "eval_steps_per_second": 15.842,
      "step": 20
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.9179524183273315,
      "learning_rate": 4.7e-05,
      "loss": 0.6353,
      "step": 30
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5306859205776173,
      "eval_loss": 0.769457221031189,
      "eval_runtime": 0.1264,
      "eval_samples_per_second": 2190.943,
      "eval_steps_per_second": 15.819,
      "step": 30
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.7931067943572998,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.5298,
      "step": 40
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.9473313093185425,
      "eval_runtime": 0.1298,
      "eval_samples_per_second": 2134.345,
      "eval_steps_per_second": 15.41,
      "step": 40
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.034212112426758,
      "learning_rate": 4.5e-05,
      "loss": 0.3829,
      "step": 50
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5126353790613718,
      "eval_loss": 1.1310056447982788,
      "eval_runtime": 0.1266,
      "eval_samples_per_second": 2187.174,
      "eval_steps_per_second": 15.792,
      "step": 50
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.213277816772461,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.2531,
      "step": 60
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5306859205776173,
      "eval_loss": 1.3281751871109009,
      "eval_runtime": 0.1319,
      "eval_samples_per_second": 2099.669,
      "eval_steps_per_second": 15.16,
      "step": 60
    },
    {
      "epoch": 6.0,
      "step": 60,
      "total_flos": 989531467960320.0,
      "train_loss": 0.5291796922683716,
      "train_runtime": 22.0246,
      "train_samples_per_second": 5652.772,
      "train_steps_per_second": 22.702
    }
  ],
  "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": 989531467960320.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}