trainer_state.json

{
  "best_metric": 3.464616060256958,
  "best_model_checkpoint": "clip-based-10k/checkpoint-939",
  "epoch": 3.0,
  "global_step": 939,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 3.4646265506744385,
      "eval_runtime": 124.5619,
      "eval_samples_per_second": 80.281,
      "eval_steps_per_second": 2.513,
      "step": 313
    },
    {
      "epoch": 1.6,
      "learning_rate": 2.3375931842385517e-05,
      "loss": 3.4664,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_loss": 3.46462345123291,
      "eval_runtime": 111.7715,
      "eval_samples_per_second": 89.468,
      "eval_steps_per_second": 2.8,
      "step": 626
    },
    {
      "epoch": 3.0,
      "eval_loss": 3.464616060256958,
      "eval_runtime": 111.8076,
      "eval_samples_per_second": 89.439,
      "eval_steps_per_second": 2.799,
      "step": 939
    },
    {
      "epoch": 3.0,
      "step": 939,
      "total_flos": 4005518400000000.0,
      "train_loss": 3.464635489466853,
      "train_runtime": 1342.7486,
      "train_samples_per_second": 22.342,
      "train_steps_per_second": 0.699
    }
  ],
  "max_steps": 939,
  "num_train_epochs": 3,
  "total_flos": 4005518400000000.0,
  "trial_name": null,
  "trial_params": null
}