checkpoint-1500/trainer_state.json

{
  "best_metric": 1.2872973680496216,
  "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e4l58-l/checkpoint-1500",
  "epoch": 0.27629397679130596,
  "eval_steps": 500,
  "global_step": 1500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09209799226376865,
      "grad_norm": 0.0014697719598188996,
      "learning_rate": 4.884877509670289e-08,
      "loss": 0.4588,
      "step": 500
    },
    {
      "epoch": 0.09209799226376865,
      "eval_loss": 1.4194883108139038,
      "eval_runtime": 78.7703,
      "eval_samples_per_second": 15.31,
      "eval_steps_per_second": 1.917,
      "step": 500
    },
    {
      "epoch": 0.1841959845275373,
      "grad_norm": 301.76043701171875,
      "learning_rate": 4.769755019340578e-08,
      "loss": 0.4255,
      "step": 1000
    },
    {
      "epoch": 0.1841959845275373,
      "eval_loss": 1.3416553735733032,
      "eval_runtime": 77.0051,
      "eval_samples_per_second": 15.661,
      "eval_steps_per_second": 1.961,
      "step": 1000
    },
    {
      "epoch": 0.27629397679130596,
      "grad_norm": 35.28853225708008,
      "learning_rate": 4.654632529010867e-08,
      "loss": 0.3724,
      "step": 1500
    },
    {
      "epoch": 0.27629397679130596,
      "eval_loss": 1.2872973680496216,
      "eval_runtime": 77.5439,
      "eval_samples_per_second": 15.552,
      "eval_steps_per_second": 1.947,
      "step": 1500
    }
  ],
  "logging_steps": 500,
  "max_steps": 21716,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "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
}