HuggingFaceTB/SmolLM3-3B-GSM8K-SFT

Full fine-tune vs HuggingFaceTB/SmolLM3-3B-Base. Fine-tune: 91% of the weights changed, across 36 of 36 layers.

Tensors changed: Embeddings 0/1, Norms 11/73, MLP 108/108, Attention 144/144

2 derivatives checked

• mradermacher/SmolLM3-3B-GSM8K-SFT-i1-GGUF: Quantization
• mradermacher/SmolLM3-3B-GSM8K-SFT-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse