HuggingFaceTB/FineMath-Llama-3B
Full fine-tune vs meta-llama/Llama-3.2-3B. 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
5 derivatives checked
- bartowski/FineMath-Llama-3B-GGUF: Same weights, converted
- mradermacher/FineMath-Llama-3B-i1-GGUF: Quantization
- mradermacher/FineMath-Llama-3B-GGUF: Same weights, converted
- DevQuasar-7/HuggingFaceTB.FineMath-Llama-3B-GGUF: Same weights, converted
- lkevinzc/Llama-3.2-3B-NuminaQA: Full fine-tune
2 other repos hold byte-for-byte identical weight files.