DevQuasar/nvidia.AceReason-Nemotron-14B-GGUF
Same weights, converted vs nvidia/AceReason-Nemotron-14B. Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/97, MLP 0/144, Attention 0/336, lm_head 0/1
Same weights, converted vs nvidia/AceReason-Nemotron-14B. Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/97, MLP 0/144, Attention 0/336, lm_head 0/1