Joseph717171's picture
Update README.md
5cdb018 verified

Custom GGUF quants of Google’s gemma-2-2b-it, where the Output Tensors are quantized to Q8_0 or kept at F32 while the Embeddings are kept at F32. 🧠🔥🚀

Notes: Great SMOL LLM for on-device inference for mobile devices. 😋