Yuma42/Llama3.1-DeluXeOne-8B
Partial fine-tune vs dphn/Dolphin-X1-8B. Partial change: 69% of the weights changed (attention, embeddings, lm_head, mlp, norms), in 32 of 32 layers.
Tensors changed: Embeddings 1/1, Norms 39/65, MLP 60/96, Attention 89/128, lm_head 1/1
2 derivatives checked
- mradermacher/Llama3.1-DeluXeOne-8B-i1-GGUF: Quantization
- mradermacher/Llama3.1-DeluXeOne-8B-GGUF: Same weights, converted