DireDreadlord/GemCod-Topaz-270M
LoRA-style vs DireDreadlord/GemCod-Jade-270M. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of k_proj, o_proj, q_proj, v_proj changed, in 18 of 18 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/109, MLP 0/54, Attention 72/72
1 derivatives checked
- mradermacher/GemCod-Topaz-270M-GGUF: Same weights, converted