BennyDaBall/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-NVFP4
Same weights, converted vs DavidAU/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored. Same weights stored in another dtype (BF16 → T40/BF16), plus 63 new tensors.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/161