pruning_info.json

{
  "base_model": "HuggingFaceTB/SmolVLM-500M-Instruct",
  "method": "layer_pruning_shortgpt_bi_score",
  "layers_removed": [
    14,
    15,
    16,
    17
  ],
  "n_layers_original": 32,
  "n_layers_remaining": 28,
  "params_before_M": 507.5,
  "params_after_M": 468.2,
  "param_reduction_pct": 7.7,
  "benchmarks": {
    "vqav2": {
      "accuracy": 0.28,
      "avg_latency_s": 0.3832,
      "peak_memory_mb": 2409.1,
      "avg_memory_mb": 2398.6,
      "throughput_sps": 2.607,
      "avg_power_w": 113.3,
      "avg_gpu_util_pct": 19.3,
      "n_samples": 50,
      "n_evaluated": 50,
      "n_skipped": 0,
      "all_failed": false,
      "zero_accuracy_warning": false,
      "metrics": {
        "exact_match": 0.28,
        "contains": 0.32,
        "token_f1": 0.2857,
        "bleu": 0.2833,
        "rouge_l": 0.2857
      }
    }
  }
}