mradermacher/Orin-Instruct-Alpaca-JP-GGUF
Same weights, converted (fewer layers) vs MakiAi/Orin-Instruct-Alpaca-JP. mradermacher/Orin-Instruct-Alpaca-JP-GGUF matches part of MakiAi/Orin-Instruct-Alpaca-JP (809 of its 1676 tensors; the rest of MakiAi/Orin-Instruct-Alpaca-JP has no counterpart). On the shared part: Same weights stored in another dtype (BF16/F16 → F16/F32), plus 37 new tensors.
Tensors changed: Embeddings 0/1, other 0/246, Norms 0/317, MLP 0/105, Attention 0/140