Closed gorkem closed 2 months ago
gemma models are hitting https://github.com/ggerganov/llama.cpp/issues/7923
Included in this: we should not forget to update modelkit references where appropriate (e.g. the fine-tuning sample is on jozu.ml but references llama-3 on ghcr.io
fine-tuning sample is from my local and not produced by the the factory.
Migrate the ModelKits that are hosted on ghcr to Jozu Hub.