ssmits/Llama-3.1-Nemotron-92B-Instruct-HF-early
Partial fine-tune vs nvidia/Llama-3.1-Nemotron-70B-Instruct-HF. Partial change: 85% of the weights changed (attention, mlp, norms), in 70 of 80 layers.
Tensors changed: Embeddings 0/1, Norms 140/161, MLP 210/240, Attention 280/320, lm_head 0/1
4 derivatives checked
- mradermacher/Llama-3.1-Nemotron-92B-Instruct-HF-early-i1-GGUF: Quantization
- mradermacher/Llama-3.1-Nemotron-93B-Instruct-HF-early-i1-GGUF: Quantization
- mradermacher/Llama-3.1-Nemotron-93B-Instruct-HF-early-GGUF: Quantization
- mradermacher/Llama-3.1-Nemotron-92B-Instruct-HF-early-GGUF: Quantization
1 other repos hold byte-for-byte identical weight files.