ihopper/ko-gemma-7b-sft-dpo-v1.0
LoRA-style vs google/gemma-7b-it. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of q_proj, v_proj changed, in 28 of 28 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/57, MLP 0/84, Attention 56/112
2 derivatives checked
- mradermacher/ko-gemma-7b-sft-dpo-v1.0-i1-GGUF: Quantization
- mradermacher/ko-gemma-7b-sft-dpo-v1.0-GGUF: Same weights, converted