checkpoint-4000/trainer_state.json

{
  "best_metric": 0.7458393573760986,
  "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l57-l/checkpoint-4000",
  "epoch": 0.42662116040955633,
  "eval_steps": 500,
  "global_step": 4000,
  "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
    },
    {
      "epoch": 0.21331058020477817,
      "grad_norm": 0.01787523180246353,
      "learning_rate": 4.6444823663253695e-07,
      "loss": 0.1889,
      "step": 2000
    },
    {
      "epoch": 0.21331058020477817,
      "eval_loss": 0.8620074987411499,
      "eval_runtime": 63.2824,
      "eval_samples_per_second": 15.597,
      "eval_steps_per_second": 1.959,
      "step": 2000
    },
    {
      "epoch": 0.2666382252559727,
      "grad_norm": 118.66907501220703,
      "learning_rate": 4.5556029579067116e-07,
      "loss": 0.2071,
      "step": 2500
    },
    {
      "epoch": 0.2666382252559727,
      "eval_loss": 0.791786253452301,
      "eval_runtime": 63.7932,
      "eval_samples_per_second": 15.472,
      "eval_steps_per_second": 1.944,
      "step": 2500
    },
    {
      "epoch": 0.3199658703071672,
      "grad_norm": 7.21087323185543e-09,
      "learning_rate": 4.4667235494880547e-07,
      "loss": 0.1588,
      "step": 3000
    },
    {
      "epoch": 0.3199658703071672,
      "eval_loss": 0.7657254338264465,
      "eval_runtime": 62.4738,
      "eval_samples_per_second": 15.799,
      "eval_steps_per_second": 1.985,
      "step": 3000
    },
    {
      "epoch": 0.37329351535836175,
      "grad_norm": 0.0014836661284789443,
      "learning_rate": 4.377844141069397e-07,
      "loss": 0.1718,
      "step": 3500
    },
    {
      "epoch": 0.37329351535836175,
      "eval_loss": 0.7610095739364624,
      "eval_runtime": 63.9118,
      "eval_samples_per_second": 15.443,
      "eval_steps_per_second": 1.94,
      "step": 3500
    },
    {
      "epoch": 0.42662116040955633,
      "grad_norm": 3.2739710807800293,
      "learning_rate": 4.2889647326507393e-07,
      "loss": 0.1113,
      "step": 4000
    },
    {
      "epoch": 0.42662116040955633,
      "eval_loss": 0.7458393573760986,
      "eval_runtime": 62.1803,
      "eval_samples_per_second": 15.873,
      "eval_steps_per_second": 1.994,
      "step": 4000
    }
  ],
  "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": 1439752921742520.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}