urufura/NVIDIA-Nemotron-Nano-9B-v2-Japanese
Full fine-tune vs nvidia/NVIDIA-Nemotron-Nano-9B-v2. Fine-tune: 100% of the weights changed, across 56 of 56 layers.
Tensors changed: Embeddings 1/1, other 189/189, Norms 84/84, MLP 50/50, Attention 16/16, lm_head 1/1
2 derivatives checked
- mradermacher/NVIDIA-Nemotron-Nano-9B-v2-Japanese-i1-GGUF: Quantization
- mradermacher/NVIDIA-Nemotron-Nano-9B-v2-Japanese-GGUF: Same weights, converted
2 other repos hold byte-for-byte identical weight files.