mradermacher/Desu1-roberta-amharic-embed-base-45k-GGUF

Same weights, converted vs Desalegnn/Desu1-roberta-amharic-embed-base-45k. Same weights stored in another dtype (F32 → F16/F32).

Tensors changed: Embeddings 0/3, Norms 0/48, Attention 0/96, other 0/48

On the Hub · Downloads on Model Pulse