Goekdeniz-Guelmez/JOSIE-4B-Thinking
Full fine-tune vs Goekdeniz-Guelmez/Qwen3-4B-Instruct-2507-gabliterated. Fine-tune: 100% of the weights changed, across 36 of 36 layers.
Tensors changed: Embeddings 1/1, Norms 141/145, MLP 108/108, Attention 144/144
2 derivatives checked
- mradermacher/JOSIE-4B-Thinking-i1-GGUF: Quantization
- mradermacher/JOSIE-4B-Thinking-GGUF: Same weights, converted