youngseok12/HyperCLOVA-X-SEED-Think-14B-minimal-sft-71875-71949-correct-540
LoRA-style vs naver-hyperclovax/HyperCLOVAX-SEED-Think-14B. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of q_proj, v_proj changed, in 38 of 38 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/153, MLP 0/114, Attention 76/152, lm_head 0/1