checkpoint-1500/trainer_state.json

{
  "best_metric": 0.8716810345649719,
  "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l57-l/checkpoint-1500",
  "epoch": 0.2763448784082535,
  "eval_steps": 500,
  "global_step": 1500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09211495946941783,
      "grad_norm": 3.083087682723999,
      "learning_rate": 4.84647506755097e-07,
      "loss": 0.2708,
      "step": 500
    },
    {
      "epoch": 0.09211495946941783,
      "eval_loss": 1.0631108283996582,
      "eval_runtime": 74.1964,
      "eval_samples_per_second": 16.268,
      "eval_steps_per_second": 2.035,
      "step": 500
    },
    {
      "epoch": 0.18422991893883567,
      "grad_norm": 7.726487159729004,
      "learning_rate": 4.69295013510194e-07,
      "loss": 0.251,
      "step": 1000
    },
    {
      "epoch": 0.18422991893883567,
      "eval_loss": 0.942773699760437,
      "eval_runtime": 74.6047,
      "eval_samples_per_second": 16.179,
      "eval_steps_per_second": 2.024,
      "step": 1000
    },
    {
      "epoch": 0.2763448784082535,
      "grad_norm": 0.02718021161854267,
      "learning_rate": 4.5394252026529107e-07,
      "loss": 0.235,
      "step": 1500
    },
    {
      "epoch": 0.2763448784082535,
      "eval_loss": 0.8716810345649719,
      "eval_runtime": 75.03,
      "eval_samples_per_second": 16.087,
      "eval_steps_per_second": 2.013,
      "step": 1500
    }
  ],
  "logging_steps": 500,
  "max_steps": 16284,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 540357043592520.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}