son0fthematrix/qwen3-4b-thinking-2507-liberated-v7
Full fine-tune vs Qwen/Qwen3-4B-Thinking-2507. Fine-tune: 100% of the weights changed, across 36 of 36 layers.
Tensors changed: Embeddings 1/1, Norms 9/145, MLP 108/108, Attention 144/144
2 derivatives checked
- mradermacher/qwen3-4b-thinking-2507-liberated-v7-i1-GGUF: Quantization
- mradermacher/qwen3-4b-thinking-2507-liberated-v7-GGUF: Same weights, converted