deepsweet/Qwen3.8-27B-DFlash2-FP16
Same weights, converted vs z-lab/Qwen3.8-27B-DFlash2. Same weights stored in another dtype (BF16 → F16).
Tensors changed: other 0/4, Norms 0/22, Attention 0/30, MLP 0/25
Same weights, converted vs z-lab/Qwen3.8-27B-DFlash2. Same weights stored in another dtype (BF16 → F16).
Tensors changed: other 0/4, Norms 0/22, Attention 0/30, MLP 0/25