Abdllahd/OMEGA-V21-Full-Merged-GGUF
Partial fine-tune (fewer layers) vs OBLITERATUS/gemma-4-E4B-it-OBLITERATED. Abdllahd/OMEGA-V21-Full-Merged-GGUF matches part of OBLITERATUS/gemma-4-E4B-it-OBLITERATED (623 of its 2130 tensors; the rest of OBLITERATUS/gemma-4-E4B-it-OBLITERATED has no counterpart). On the shared part: Abdllahd/OMEGA-V21-Full-Merged-GGUF is a GGUF (F16) of different weights. Partial change: 61% of the weights changed (attention, embeddings, mlp), in 42 of 42 layers.
Tensors changed: Embeddings 1/2, Norms 0/278, MLP 126/126, other 0/85, Attention 132/132