mradermacher/Gemma2-9b-WangchanLIONv2-instruct-GGUF

Same weights, converted vs aisingapore/Gemma2-9b-WangchanLIONv2-instruct. Same weights stored in another dtype (BF16 → F16/F32).

Tensors changed: Embeddings 0/1, Attention 0/168, Norms 0/169, MLP 0/126

On the Hub · Downloads on Model Pulse