Cyronius/Qwen3.6-27B-A2.8B
Full fine-tune vs Qwen/Qwen3.6-35B-A3B. Cyronius/Qwen3.6-27B-A2.8B is a GGUF (Q4KM) of different weights. Fine-tune: 100% of the weights changed, across 31 of 31 layers.
Tensors changed: Embeddings 1/1, Norms 64/67, MLP 123/124, Attention 8/8, lm_head 1/1