llama-3-70B-Instruct-abliterated-IQ2_M.gguf Benchmark | LLM Performance Benchmarks