p-e-r-e-g-r-i-n-e/Qwen3.5-27B-heretic-RYS-XL

Partial fine-tune vs coder3101/Qwen3.5-27B-heretic. Partial change: 42% of the weights changed (attention, mlp, norms), in 30 of 64 layers.

Tensors changed: lm_head 0/1, Embeddings 0/4, Norms 98/319, Attention 208/556, MLP 90/304

2 derivatives checked

• mradermacher/Qwen3.5-27B-heretic-RYS-XL-i1-GGUF: Quantization
• mradermacher/Qwen3.5-27B-heretic-RYS-XL-GGUF: Quantization

On the Hub · Downloads on Model Pulse