trainer_state.json

{
  "best_global_step": 10,
  "best_metric": 0.6893051266670227,
  "best_model_checkpoint": "distilbert_rand_20_v2_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": 0.6134195923805237,
      "learning_rate": 4.9e-05,
      "loss": 0.6973,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5342960288808665,
      "eval_loss": 0.6893051266670227,
      "eval_runtime": 0.1548,
      "eval_samples_per_second": 1788.963,
      "eval_steps_per_second": 12.917,
      "step": 10
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.29136061668396,
      "learning_rate": 4.8e-05,
      "loss": 0.6835,
      "step": 20
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.6986406445503235,
      "eval_runtime": 0.1399,
      "eval_samples_per_second": 1979.855,
      "eval_steps_per_second": 14.295,
      "step": 20
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.7932871580123901,
      "learning_rate": 4.7e-05,
      "loss": 0.6373,
      "step": 30
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5126353790613718,
      "eval_loss": 0.7705271244049072,
      "eval_runtime": 0.1316,
      "eval_samples_per_second": 2104.872,
      "eval_steps_per_second": 15.198,
      "step": 30
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.2666196823120117,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.5447,
      "step": 40
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4981949458483754,
      "eval_loss": 0.8774519562721252,
      "eval_runtime": 0.1628,
      "eval_samples_per_second": 1701.384,
      "eval_steps_per_second": 12.284,
      "step": 40
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.704348564147949,
      "learning_rate": 4.5e-05,
      "loss": 0.4205,
      "step": 50
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.48014440433212996,
      "eval_loss": 1.098254680633545,
      "eval_runtime": 0.1289,
      "eval_samples_per_second": 2148.908,
      "eval_steps_per_second": 15.516,
      "step": 50
    },
    {
      "epoch": 6.0,
      "grad_norm": 8.498217582702637,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.2894,
      "step": 60
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5018050541516246,
      "eval_loss": 1.4304996728897095,
      "eval_runtime": 0.1341,
      "eval_samples_per_second": 2066.182,
      "eval_steps_per_second": 14.918,
      "step": 60
    },
    {
      "epoch": 6.0,
      "step": 60,
      "total_flos": 989531467960320.0,
      "train_loss": 0.5454349120457967,
      "train_runtime": 22.9872,
      "train_samples_per_second": 5416.059,
      "train_steps_per_second": 21.751
    }
  ],
  "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
}