Minthy/ToriiGate-0.5
Full fine-tune vs Qwen/Qwen3.5-4B-Base. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 4/4, MLP 193/196, Attention 279/320, Norms 130/203
4 derivatives checked
- DraconicDragon/ToriiGate-0.5-GGUF: Same weights, converted
- SleepVeryHard/ToriiGate-0.5_GGUF: Same weights, converted
- mradermacher/ToriiGate-0.5-i1-GGUF: Quantization
- mradermacher/ToriiGate-0.5-GGUF: Same weights, converted