mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-GGUF

Same weights, converted (fewer layers) vs DavidAU/Gemma-3-4b-it-Uncensored-DBL-X. mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-GGUF matches part of DavidAU/Gemma-3-4b-it-Uncensored-DBL-X (496 of its 935 tensors; the rest of DavidAU/Gemma-3-4b-it-Uncensored-DBL-X has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).

Tensors changed: Embeddings 0/1, Norms 0/229, MLP 0/114, Attention 0/152

On the Hub · Downloads on Model Pulse