mradermacher/DrMedra4B-GGUF

Same weights, converted (fewer layers) vs drwlf/DrMedra4B. mradermacher/DrMedra4B-GGUF matches part of drwlf/DrMedra4B (444 of its 883 tensors; the rest of drwlf/DrMedra4B has no counterpart). On the shared part: Same weights stored in another dtype (F32 → F16/F32).

Tensors changed: Embeddings 0/1, Norms 0/205, MLP 0/102, Attention 0/136

On the Hub · Downloads on Model Pulse