Do not have time for README but you know. Uncensored, geometric mean between harmful and harmless. 3/100 refusals based on mlabonne dataset. its good.

Downloads last month: 300

Safetensors

Model size

3B params

Tensor type

BF16

Model tree for s3nh/SmolLLM-3B-Uncensored

Quantizations

2 models