eomcaleb/weight-space-bench
Benchmark: can an LLM make a language model better at math by editing its weights directly? (Qwen3.5-0.8B, MATH-500/GSM8K, via OpenRouter)
GitHub repository with 15 stars and 2 forks.
Language: Python
Topics: interpretability, llm-benchmark, openrouter, qwen, weight-editing