second-state/Yi-34Bx2-MoE-60B-GGUF
Same weights, converted vs cloudyu/Yi-34Bx2-MoE-60B. Same weights stored in another dtype (BF16 → F32), plus 180 new tensors.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/121, Attention 0/240, other 0/60
Same weights, converted vs cloudyu/Yi-34Bx2-MoE-60B. Same weights stored in another dtype (BF16 → F32), plus 180 new tensors.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/121, Attention 0/240, other 0/60