checkpoint-1722/trainer_state.json

{
  "best_global_step": 1722,
  "best_metric": 0.44515112042427063,
  "best_model_checkpoint": "siglip2-finetune-full/checkpoint-1722",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1722,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.646481178396072,
      "eval_loss": 0.8569742441177368,
      "eval_model_preparation_time": 0.0022,
      "eval_runtime": 120.9539,
      "eval_samples_per_second": 75.773,
      "eval_steps_per_second": 9.475,
      "step": 287
    },
    {
      "epoch": 1.7421602787456445,
      "grad_norm": 29.628196716308594,
      "learning_rate": 1.4617224880382776e-05,
      "loss": 1.0903,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6763775231860338,
      "eval_loss": 0.7689628005027771,
      "eval_model_preparation_time": 0.0022,
      "eval_runtime": 120.8745,
      "eval_samples_per_second": 75.822,
      "eval_steps_per_second": 9.481,
      "step": 574
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7290780141843972,
      "eval_loss": 0.6629450917243958,
      "eval_model_preparation_time": 0.0022,
      "eval_runtime": 121.8989,
      "eval_samples_per_second": 75.185,
      "eval_steps_per_second": 9.401,
      "step": 861
    },
    {
      "epoch": 3.484320557491289,
      "grad_norm": 19.66444969177246,
      "learning_rate": 8.636363636363637e-06,
      "loss": 0.7613,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7742498636115658,
      "eval_loss": 0.552396297454834,
      "eval_model_preparation_time": 0.0022,
      "eval_runtime": 121.0546,
      "eval_samples_per_second": 75.71,
      "eval_steps_per_second": 9.467,
      "step": 1148
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8007637752318604,
      "eval_loss": 0.4922109544277191,
      "eval_model_preparation_time": 0.0022,
      "eval_runtime": 121.4614,
      "eval_samples_per_second": 75.456,
      "eval_steps_per_second": 9.435,
      "step": 1435
    },
    {
      "epoch": 5.2264808362369335,
      "grad_norm": 19.047819137573242,
      "learning_rate": 2.6555023923444976e-06,
      "loss": 0.5947,
      "step": 1500
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8224768139661757,
      "eval_loss": 0.44515112042427063,
      "eval_model_preparation_time": 0.0022,
      "eval_runtime": 121.6668,
      "eval_samples_per_second": 75.329,
      "eval_steps_per_second": 9.419,
      "step": 1722
    }
  ],
  "logging_steps": 500,
  "max_steps": 1722,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "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": 4.6059438704851354e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}