katanemo/Arch-Function-Chat-3B
Full fine-tune vs Qwen/Qwen2.5-Coder-3B-Instruct. Fine-tune: 100% of the weights changed, across 36 of 36 layers.
Tensors changed: Embeddings 1/1, Norms 70/73, MLP 108/108, Attention 252/252
4 derivatives checked
- bartowski/katanemo_Arch-Function-Chat-3B-GGUF: Same weights, converted
- mradermacher/Arch-Function-Chat-3B-i1-GGUF: Quantization
- mradermacher/Arch-Function-Chat-3B-GGUF: Same weights, converted
- ysn-rfd/Arch-Function-Chat-3B-GGUF: Quantization