mradermacher/gemma-3-4b-it-PaperWitch-heresy-GGUF

Same weights, converted (fewer layers) vs MuXodious/gemma-3-4b-it-PaperWitch-heresy. mradermacher/gemma-3-4b-it-PaperWitch-heresy-GGUF matches part of MuXodious/gemma-3-4b-it-PaperWitch-heresy (444 of its 883 tensors; the rest of MuXodious/gemma-3-4b-it-PaperWitch-heresy 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