mradermacher/MedraN-E4B-Uncensored-EP7-GGUF
Same weights, converted (fewer layers) vs nicoboss/MedraN-E4B-Uncensored-EP7. mradermacher/MedraN-E4B-Uncensored-EP7-GGUF matches part of nicoboss/MedraN-E4B-Uncensored-EP7 (809 of its 1676 tensors; the rest of nicoboss/MedraN-E4B-Uncensored-EP7 has no counterpart). On the shared part: Same weights stored in another dtype (F16 → F16/F32), plus 37 new tensors.
Tensors changed: other 0/246, Norms 0/317, MLP 0/105, Attention 0/140, Embeddings 0/1