checkpoint-1500/trainer_state.json

{
  "best_metric": 0.9395213723182678,
  "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l37-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": 8.437943458557129,
      "learning_rate": 2.907885040530582e-07,
      "loss": 0.3017,
      "step": 500
    },
    {
      "epoch": 0.09211495946941783,
      "eval_loss": 1.164457082748413,
      "eval_runtime": 75.2697,
      "eval_samples_per_second": 16.036,
      "eval_steps_per_second": 2.006,
      "step": 500
    },
    {
      "epoch": 0.18422991893883567,
      "grad_norm": 14.740598678588867,
      "learning_rate": 2.815770081061164e-07,
      "loss": 0.2899,
      "step": 1000
    },
    {
      "epoch": 0.18422991893883567,
      "eval_loss": 1.01627779006958,
      "eval_runtime": 76.4934,
      "eval_samples_per_second": 15.779,
      "eval_steps_per_second": 1.974,
      "step": 1000
    },
    {
      "epoch": 0.2763448784082535,
      "grad_norm": 0.17031528055667877,
      "learning_rate": 2.723655121591746e-07,
      "loss": 0.273,
      "step": 1500
    },
    {
      "epoch": 0.2763448784082535,
      "eval_loss": 0.9395213723182678,
      "eval_runtime": 76.6256,
      "eval_samples_per_second": 15.752,
      "eval_steps_per_second": 1.971,
      "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
}