DeepAuto-AI/ldm_soup_Llama-3.1-8B-Inst
Partial fine-tune vs meta-llama/Llama-3.1-8B-Instruct. Partial change: 24% of the weights changed (attention, mlp, norms), in 28 of 32 layers.
Tensors changed: Embeddings 0/1, Norms 14/65, MLP 27/96, Attention 32/128, lm_head 0/1
5 derivatives checked
- mradermacher/ldm_soup_Llama-3.1-8B-Inst-i1-GGUF: Quantization
- featherless-ai-quants/DeepAutoAI-ldm_soup_Llama-3.1-8B-Inst-GGUF: Quantization
- mradermacher/ldm_soup_Llama-3.1-8B-Inst-GGUF: Same weights, converted
- featherless-ai-quants/DeepAuto-AI-ldm_soup_Llama-3.1-8B-Inst-GGUF: Quantization
- tensorblock/ldm_soup_Llama-3.1-8B-Inst-GGUF: Quantization