h100_20251117_222425/experiment_metadata.json

{
  "gpu_type": "h100",
  "timestamp": "20251117_222425",
  "model_name": "meta-llama/Llama-3.1-8B-Instruct",
  "n_prompts": 100,
  "max_ctx_len": 512,
  "max_tokens": 500,
  "python_version": "3.12.3",
  "torch_version": "2.9.0+cu128",
  "cuda_available": true,
  "cuda_version": "12.8",
  "gpu_name": "NVIDIA H100 NVL"
}