mradermacher/Marco-LLM-AR-V3-GGUF
Same weights, converted vs ATH-MaaS/Marco-LLM-AR-V4. Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Norms 0/57, Attention 0/196, MLP 0/84, lm_head 0/1, Embeddings 0/1
Same weights, converted vs ATH-MaaS/Marco-LLM-AR-V4. Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Norms 0/57, Attention 0/196, MLP 0/84, lm_head 0/1, Embeddings 0/1