mzoelfakar/Al-Khwarizmi-3B

Full fine-tune vs HuggingFaceTB/SmolLM3-3B-Base. Fine-tune: 100% of the weights changed, across 36 of 36 layers.

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

3 derivatives checked

• mradermacher/Al-Khwarizmi-3B-i1-GGUF: Quantization
• mzoelfakar/Al-Khwarizmi-3B-GGUF: Same weights, converted
• mradermacher/Al-Khwarizmi-3B-GGUF: LoRA-style

On the Hub · Downloads on Model Pulse