Columbidae/gemma-3-27b-half-full
Same weights, repacked vs google/gemma-3-27b-it. Same weights in different files (re-saved, re-sharded or new metadata).
Tensors changed: Embeddings 0/4, Norms 0/484, MLP 0/294, Attention 0/464, other 0/1
2 derivatives checked
- mradermacher/gemma-3-27b-half-full-i1-GGUF: Quantization (fewer layers)
- mradermacher/gemma-3-27b-half-full-GGUF: Quantization (fewer layers)
157 other repos hold byte-for-byte identical weight files.