bunnycore/Smol-Llama-3.2-3B
Full fine-tune vs HuggingFaceTB/finemath-ablation-infiwebmath. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 57/57, MLP 84/84, Attention 112/112
2 derivatives checked
- mradermacher/Smol-Llama-3.2-3B-i1-GGUF: Quantization
- mradermacher/Smol-Llama-3.2-3B-GGUF: Same weights, converted