ramgpt/Jev_Qwen3.8-27B-EXL3-4.0bpw
Same weights, converted vs SargeDev/Jev_Qwen3.8-27B. Same weights stored in another dtype (BF16 → I16/BF16), plus 1604 new tensors.
Tensors changed: Embeddings 0/1, Norms 0/209, Attention 0/240
Same weights, converted vs SargeDev/Jev_Qwen3.8-27B. Same weights stored in another dtype (BF16 → I16/BF16), plus 1604 new tensors.
Tensors changed: Embeddings 0/1, Norms 0/209, Attention 0/240