QuietImpostor/OpenELM-3B-Instruct-GGUFs
Same weights, converted vs apple/OpenELM-3B-Instruct. Same weights stored in another dtype (BF16 → F16/F32), plus 72 new tensors.
Tensors changed: Norms 0/145, Attention 0/72, MLP 0/36, Embeddings 0/1
Same weights, converted vs apple/OpenELM-3B-Instruct. Same weights stored in another dtype (BF16 → F16/F32), plus 72 new tensors.
Tensors changed: Norms 0/145, Attention 0/72, MLP 0/36, Embeddings 0/1