Multi Step Reasoning Search With Iterative Refinement

via “reasoning and step-by-step problem decomposition”

text-generation model by undefined. 94,68,562 downloads.

Unique: Emergent chain-of-thought capability from instruction tuning on reasoning datasets; no explicit reasoning module or symbolic engine — reasoning emerges from learned token prediction patterns that favor intermediate explanation tokens, making it lightweight but probabilistic

vs others: Provides transparent reasoning comparable to GPT-4 on simple problems but with full local control; outperforms Mistral-7B on reasoning tasks due to instruction tuning, but lacks the formal verification and symbolic reasoning of specialized tools like Wolfram Alpha