anakin87/Llama-3-8b-ita-slerp
Partial fine-tune vs DeepMount00/Llama-3-8b-Ita. Partial change: 20% of the weights changed (attention, embeddings, mlp, norms), in 31 of 32 layers.
Tensors changed: Embeddings 1/1, Norms 6/65, MLP 5/96, Attention 95/128, lm_head 0/1
1 derivatives checked
- featherless-ai-quants/anakin87-Llama-3-8b-ita-slerp-GGUF: Quantization