Isrark371/

ds

	Blind Spots of SmolLM2-1.7B




	
	Model Tested

SmolLM2-1.7Bhttps://huggingface.co/HuggingFaceTB/SmolLM2-1.7B

This dataset contains examples where the tested model produced incorrect predictions. Each example includes:

input prompt expected correct answer model output

The model struggles with:

logical reasoning arithmetic calculations factual knowledge trick questions common sense reasoning