zake7749/Llama-3.2-1B-it-chinese-kyara
Full fine-tune vs meta-llama/Llama-3.2-1B-Instruct. Fine-tune: 100% of the weights changed, across 16 of 16 layers.
Tensors changed: Embeddings 1/1, Norms 9/33, MLP 48/48, Attention 64/64
2 derivatives checked
- mradermacher/Llama-3.2-1B-it-chinese-kyara-i1-GGUF: Quantization
- mradermacher/Llama-3.2-1B-it-chinese-kyara-GGUF: Same weights, converted