trainer_state.json

{
  "best_global_step": 1218,
  "best_metric": 0.541471444081587,
  "best_model_checkpoint": "outputs/vit5/checkpoint-1218",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 1218,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9942528735632183,
      "grad_norm": 58.274559020996094,
      "learning_rate": 6.88e-06,
      "loss": 1.9914,
      "step": 173
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.3206997084548105,
      "eval_loss": 1.7322026491165161,
      "eval_macro_f1": 0.14803455259181977,
      "eval_runtime": 2.6376,
      "eval_samples_per_second": 260.08,
      "eval_steps_per_second": 8.341,
      "step": 174
    },
    {
      "epoch": 1.9885057471264367,
      "grad_norm": 18.986042022705078,
      "learning_rate": 1.38e-05,
      "loss": 1.6725,
      "step": 346
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.521865889212828,
      "eval_loss": 1.3959580659866333,
      "eval_macro_f1": 0.3094694125241264,
      "eval_runtime": 2.6511,
      "eval_samples_per_second": 258.757,
      "eval_steps_per_second": 8.298,
      "step": 348
    },
    {
      "epoch": 2.9827586206896552,
      "grad_norm": 15.519329071044922,
      "learning_rate": 1.999994401896218e-05,
      "loss": 1.2986,
      "step": 519
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5714285714285714,
      "eval_loss": 1.1622083187103271,
      "eval_macro_f1": 0.434315370257316,
      "eval_runtime": 2.6503,
      "eval_samples_per_second": 258.838,
      "eval_steps_per_second": 8.301,
      "step": 522
    },
    {
      "epoch": 3.9770114942528734,
      "grad_norm": 33.60322189331055,
      "learning_rate": 1.9993697433304146e-05,
      "loss": 1.0153,
      "step": 692
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5626822157434402,
      "eval_loss": 1.1611000299453735,
      "eval_macro_f1": 0.48082332945066447,
      "eval_runtime": 2.6538,
      "eval_samples_per_second": 258.499,
      "eval_steps_per_second": 8.29,
      "step": 696
    },
    {
      "epoch": 4.971264367816092,
      "grad_norm": 14.145572662353516,
      "learning_rate": 1.9977115923137912e-05,
      "loss": 0.7927,
      "step": 865
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6034985422740525,
      "eval_loss": 1.1552627086639404,
      "eval_macro_f1": 0.52788220890471,
      "eval_runtime": 2.6498,
      "eval_samples_per_second": 258.886,
      "eval_steps_per_second": 8.302,
      "step": 870
    },
    {
      "epoch": 5.9655172413793105,
      "grad_norm": 19.387939453125,
      "learning_rate": 1.995021663613649e-05,
      "loss": 0.5873,
      "step": 1038
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5860058309037901,
      "eval_loss": 1.2505557537078857,
      "eval_macro_f1": 0.5207841516451528,
      "eval_runtime": 2.6525,
      "eval_samples_per_second": 258.626,
      "eval_steps_per_second": 8.294,
      "step": 1044
    },
    {
      "epoch": 6.959770114942529,
      "grad_norm": 12.481695175170898,
      "learning_rate": 1.991302739004225e-05,
      "loss": 0.4349,
      "step": 1211
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.607871720116618,
      "eval_loss": 1.3485385179519653,
      "eval_macro_f1": 0.541471444081587,
      "eval_runtime": 2.653,
      "eval_samples_per_second": 258.571,
      "eval_steps_per_second": 8.292,
      "step": 1218
    }
  ],
  "logging_steps": 173,
  "max_steps": 17400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 100,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8246611698717840.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}