OdaxAI/DANTE-Mosaic-3.5B
Full fine-tune vs HuggingFaceTB/SmolLM3-3B. Fine-tune: 100% of the weights changed, across 36 of 36 layers.
Tensors changed: Embeddings 1/1, Norms 12/73, MLP 108/108, Attention 144/144
2 derivatives checked
- mradermacher/DANTE-Mosaic-3.5B-i1-GGUF: Quantization
- mradermacher/DANTE-Mosaic-3.5B-GGUF: Same weights, converted