anakin87/Llama-3-8b-ita-ties-pro
LoRA-style vs DeepMount00/Llama-3-8b-Ita. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj changed, in 32 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/65, MLP 19/96, Attention 109/128, lm_head 0/1
4 derivatives checked
- featherless-ai-quants/anakin87-Llama-3-8b-ita-ties-pro-GGUF: Quantization
- mradermacher/Llama-3-8b-ita-ties-pro-i1-GGUF: Quantization
- mradermacher/Llama-3-8b-ita-ties-pro-GGUF: Same weights, converted
- tensorblock/Llama-3-8b-ita-ties-pro-GGUF: Quantization