issai/Qwen3.5-9B-Base-Kazakh
Full fine-tune (fewer layers) vs Qwen/Qwen3.5-9B-Base. issai/Qwen3.5-9B-Base-Kazakh matches part of Qwen/Qwen3.5-9B-Base (425 of its 775 tensors; the rest of Qwen/Qwen3.5-9B-Base has no counterpart). On the shared part: Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: MLP 96/96, Attention 211/224, Norms 94/105
2 derivatives checked
- issai/Qwen3.5-9B-Kazakh: Full fine-tune (fewer layers)
- mradermacher/Qwen3.5-9B-Base-Kazakh-GGUF: Same weights, converted