munkhbayar-batkhuu/gemma-3-4b-mongolian
Full fine-tune (fewer layers) vs google/gemma-3-4b-pt. munkhbayar-batkhuu/gemma-3-4b-mongolian matches part of google/gemma-3-4b-pt (444 of its 883 tensors; the rest of google/gemma-3-4b-pt has no counterpart). On the shared part: Fine-tune: 100% of the weights changed, across 34 of 34 layers.
Tensors changed: Embeddings 1/1, Norms 0/205, MLP 102/102, Attention 136/136
1 derivatives checked
- mradermacher/gemma-3-4b-mongolian-GGUF: Same weights, converted