T145/ZEUS-8B-V16

Full fine-tune vs VAGOsolutions/Llama-3.1-SauerkrautLM-8b-Instruct. Fine-tune: 92% of the weights changed, across 32 of 32 layers.

Tensors changed: lm_head 1/1, Embeddings 0/1, Norms 11/65, MLP 94/96, Attention 125/128

2 derivatives checked

• mradermacher/ZEUS-8B-V16-i1-GGUF: Quantization
• mradermacher/ZEUS-8B-V16-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse