DevQuasar/tencent.HY-MT1.5-7B-GGUF
Same weights, converted vs tencent/HY-MT1.5-7B. Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: MLP 0/96, Attention 0/128, Norms 0/129, Embeddings 0/1
Same weights, converted vs tencent/HY-MT1.5-7B. Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: MLP 0/96, Attention 0/128, Norms 0/129, Embeddings 0/1