katanemo/Arch-Function-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 73/73, MLP 108/108, Attention 252/252
6 derivatives checked
- bartowski/Arch-Function-3B-GGUF: Full fine-tune
- mradermacher/Arch-Function-3B-i1-GGUF: Quantization
- katanemo/Arch-Function-3B.gguf: Quantization
- mradermacher/Arch-Function-3B-GGUF: Full fine-tune
- DevQuasar-3/katanemo.Arch-Function-3B-GGUF: Quantization
- MaziyarPanahi/Arch-Function-3B-GGUF: Full fine-tune