mradermacher/Nidum-gemma-3-4b-it-Uncensored-GGUF

Same weights, converted (fewer layers) vs lemuralabs/Gemma-3-4B-it-Uncensored. mradermacher/Nidum-gemma-3-4b-it-Uncensored-GGUF matches part of lemuralabs/Gemma-3-4B-it-Uncensored (444 of its 883 tensors; the rest of lemuralabs/Gemma-3-4B-it-Uncensored has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).

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

On the Hub · Downloads on Model Pulse