KordAI/Typhoon-Gemma3-KordTranslate-EN-TH-4B
LoRA-style vs typhoon-ai/typhoon2.1-gemma3-4b. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj changed, in 34 of 34 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/205, MLP 102/102, Attention 136/136
1 derivatives checked
- KordAI/Typhoon-Gemma3-KordTranslate-EN-TH-4B-GGUF: Same weights, converted