mradermacher/gemma2-9b-cpt-sahabatai-v1-instruct-GGUF

Same weights, converted vs GoToCompany/gemma2-9b-cpt-sahabatai-v1-instruct. Same weights stored in another dtype (BF16 → F16/F32).

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

On the Hub · Downloads on Model Pulse