Stress-testing LLM benchmarks through dataset forensics and controlled perturbations
-
Updated
Sep 11, 2026 - Python
#
mmlu-benchmark
Here are 3 public repositories matching this topic...
A production-grade LLM Evaluation & Benchmarking Framework for systematic model auditing. Features parallel benchmarking, fairness/bias detection, MMLU integration, and a real-time analytics dashboard powered by React and FastAPI.
benchmarking performance-analysis ai-safety mlops fastapi fairness-ml research-tools prompt-engineering generative-ai llm-evaluation react-v19 model-auditing mmlu-benchmark
-
Updated
Apr 14, 2026 - Python
FSDP fine-tuning of Qwen2.5-7B on MMLU auxiliary_train with 3.6x inference throughput optimization (1.8x PagedAttention + continuous batching, 2.0x tensor parallelism) validated on 2-node H200 cluster; reproducible scaling path for multi-GPU HPC deployments.
-
Updated
May 25, 2026 - Python
Add this topic to your repo
To associate your repository with the mmlu-benchmark topic, visit your repo's landing page and select "manage topics."