trainer_state.json

{
  "best_metric": 0.5875912408759124,
  "best_model_checkpoint": "UL_interior_classification/checkpoint-91",
  "epoch": 6.867924528301887,
  "eval_steps": 500,
  "global_step": 91,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.7547169811320755,
      "grad_norm": 4.003274440765381,
      "learning_rate": 5e-05,
      "loss": 2.7547,
      "step": 10
    },
    {
      "epoch": 0.9811320754716981,
      "eval_accuracy": 0.3284671532846715,
      "eval_loss": 2.342195510864258,
      "eval_runtime": 10.3898,
      "eval_samples_per_second": 26.372,
      "eval_steps_per_second": 0.866,
      "step": 13
    },
    {
      "epoch": 1.509433962264151,
      "grad_norm": 2.726027011871338,
      "learning_rate": 4.3827160493827164e-05,
      "loss": 1.7119,
      "step": 20
    },
    {
      "epoch": 1.9622641509433962,
      "eval_accuracy": 0.49635036496350365,
      "eval_loss": 1.8849732875823975,
      "eval_runtime": 9.8675,
      "eval_samples_per_second": 27.768,
      "eval_steps_per_second": 0.912,
      "step": 26
    },
    {
      "epoch": 2.2641509433962264,
      "grad_norm": 2.424898624420166,
      "learning_rate": 3.7654320987654326e-05,
      "loss": 1.249,
      "step": 30
    },
    {
      "epoch": 2.9433962264150946,
      "eval_accuracy": 0.5291970802919708,
      "eval_loss": 1.5653367042541504,
      "eval_runtime": 10.0672,
      "eval_samples_per_second": 27.217,
      "eval_steps_per_second": 0.894,
      "step": 39
    },
    {
      "epoch": 3.018867924528302,
      "grad_norm": 6.8397016525268555,
      "learning_rate": 3.148148148148148e-05,
      "loss": 1.0982,
      "step": 40
    },
    {
      "epoch": 3.7735849056603774,
      "grad_norm": 2.1824584007263184,
      "learning_rate": 2.5308641975308646e-05,
      "loss": 0.8838,
      "step": 50
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5693430656934306,
      "eval_loss": 1.36745285987854,
      "eval_runtime": 10.0511,
      "eval_samples_per_second": 27.261,
      "eval_steps_per_second": 0.895,
      "step": 53
    },
    {
      "epoch": 4.528301886792453,
      "grad_norm": 1.9526816606521606,
      "learning_rate": 1.91358024691358e-05,
      "loss": 0.8896,
      "step": 60
    },
    {
      "epoch": 4.981132075471698,
      "eval_accuracy": 0.5802919708029197,
      "eval_loss": 1.2906520366668701,
      "eval_runtime": 11.107,
      "eval_samples_per_second": 24.669,
      "eval_steps_per_second": 0.81,
      "step": 66
    },
    {
      "epoch": 5.283018867924528,
      "grad_norm": 2.0880227088928223,
      "learning_rate": 1.2962962962962962e-05,
      "loss": 0.7262,
      "step": 70
    },
    {
      "epoch": 5.962264150943396,
      "eval_accuracy": 0.5802919708029197,
      "eval_loss": 1.2625008821487427,
      "eval_runtime": 10.0312,
      "eval_samples_per_second": 27.315,
      "eval_steps_per_second": 0.897,
      "step": 79
    },
    {
      "epoch": 6.037735849056604,
      "grad_norm": 1.8009761571884155,
      "learning_rate": 6.790123456790123e-06,
      "loss": 0.7061,
      "step": 80
    },
    {
      "epoch": 6.7924528301886795,
      "grad_norm": 1.9284666776657104,
      "learning_rate": 6.17283950617284e-07,
      "loss": 0.6817,
      "step": 90
    },
    {
      "epoch": 6.867924528301887,
      "eval_accuracy": 0.5875912408759124,
      "eval_loss": 1.251747727394104,
      "eval_runtime": 9.9685,
      "eval_samples_per_second": 27.487,
      "eval_steps_per_second": 0.903,
      "step": 91
    },
    {
      "epoch": 6.867924528301887,
      "step": 91,
      "total_flos": 8.883485181176095e+17,
      "train_loss": 1.1828759214380284,
      "train_runtime": 1039.6912,
      "train_samples_per_second": 11.21,
      "train_steps_per_second": 0.088
    }
  ],
  "logging_steps": 10,
  "max_steps": 91,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 8.883485181176095e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}