run-1/checkpoint-3000/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.04289544235925,
  "eval_steps": 500,
  "global_step": 3000,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5766007304191589,
      "eval_loss": 0.6931423544883728,
      "eval_runtime": 13.5721,
      "eval_samples_per_second": 219.789,
      "eval_steps_per_second": 13.778,
      "step": 373
    },
    {
      "epoch": 1.3404825737265416,
      "grad_norm": 0.6603747606277466,
      "learning_rate": 4.647975235317143e-05,
      "loss": 0.695,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5092189311981201,
      "eval_loss": 0.693084716796875,
      "eval_runtime": 18.8977,
      "eval_samples_per_second": 157.85,
      "eval_steps_per_second": 9.895,
      "step": 746
    },
    {
      "epoch": 2.680965147453083,
      "grad_norm": 0.5473846197128296,
      "learning_rate": 3.9284744248965325e-05,
      "loss": 0.6953,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.48139455914497375,
      "eval_loss": 0.6931472420692444,
      "eval_runtime": 13.4613,
      "eval_samples_per_second": 221.598,
      "eval_steps_per_second": 13.892,
      "step": 1119
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5511230230331421,
      "eval_loss": 0.6931468844413757,
      "eval_runtime": 13.5694,
      "eval_samples_per_second": 219.832,
      "eval_steps_per_second": 13.781,
      "step": 1492
    },
    {
      "epoch": 4.021447721179625,
      "grad_norm": 0.5608623027801514,
      "learning_rate": 3.2089736144759215e-05,
      "loss": 0.6945,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5102245807647705,
      "eval_loss": 0.6931470632553101,
      "eval_runtime": 18.1489,
      "eval_samples_per_second": 164.362,
      "eval_steps_per_second": 10.304,
      "step": 1865
    },
    {
      "epoch": 5.361930294906166,
      "grad_norm": 0.44889602065086365,
      "learning_rate": 2.4894728040553116e-05,
      "loss": 0.6947,
      "step": 2000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5665437579154968,
      "eval_loss": 0.6931470632553101,
      "eval_runtime": 17.2033,
      "eval_samples_per_second": 173.397,
      "eval_steps_per_second": 10.87,
      "step": 2238
    },
    {
      "epoch": 6.702412868632708,
      "grad_norm": 0.5518374443054199,
      "learning_rate": 1.7699719936347013e-05,
      "loss": 0.6944,
      "step": 2500
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5749245882034302,
      "eval_loss": 0.6931470036506653,
      "eval_runtime": 16.9359,
      "eval_samples_per_second": 176.135,
      "eval_steps_per_second": 11.042,
      "step": 2611
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5772712230682373,
      "eval_loss": 0.6931470036506653,
      "eval_runtime": 17.9386,
      "eval_samples_per_second": 166.29,
      "eval_steps_per_second": 10.424,
      "step": 2984
    },
    {
      "epoch": 8.04289544235925,
      "grad_norm": 0.553704023361206,
      "learning_rate": 1.050471183214091e-05,
      "loss": 0.6941,
      "step": 3000
    }
  ],
  "logging_steps": 500,
  "max_steps": 3730,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4358935077786780.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 5.367476045737753e-05,
    "per_device_train_batch_size": 32
  }
}