checkpoint-1500/trainer_state.json

{
  "best_metric": 0.9099885821342468,
  "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l57-l/checkpoint-1500",
  "epoch": 0.1599829351535836,
  "eval_steps": 500,
  "global_step": 1500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05332764505119454,
      "grad_norm": 414.62841796875,
      "learning_rate": 4.911120591581342e-07,
      "loss": 0.3812,
      "step": 500
    },
    {
      "epoch": 0.05332764505119454,
      "eval_loss": 1.1163015365600586,
      "eval_runtime": 61.8962,
      "eval_samples_per_second": 15.946,
      "eval_steps_per_second": 2.003,
      "step": 500
    },
    {
      "epoch": 0.10665529010238908,
      "grad_norm": 34.08418655395508,
      "learning_rate": 4.822241183162685e-07,
      "loss": 0.2683,
      "step": 1000
    },
    {
      "epoch": 0.10665529010238908,
      "eval_loss": 0.9684447646141052,
      "eval_runtime": 61.821,
      "eval_samples_per_second": 15.965,
      "eval_steps_per_second": 2.006,
      "step": 1000
    },
    {
      "epoch": 0.1599829351535836,
      "grad_norm": 350.4574890136719,
      "learning_rate": 4.733361774744027e-07,
      "loss": 0.2119,
      "step": 1500
    },
    {
      "epoch": 0.1599829351535836,
      "eval_loss": 0.9099885821342468,
      "eval_runtime": 61.9606,
      "eval_samples_per_second": 15.929,
      "eval_steps_per_second": 2.001,
      "step": 1500
    }
  ],
  "logging_steps": 500,
  "max_steps": 28128,
  "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
}