wxzhang/Fast-dLLM-v2-7B-dTRPO-Lite
LoRA-style vs Efficient-Large-Model/Fast_dLLM_v2_7B. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, up_proj changed, in 28 of 28 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/57, MLP 84/84, Attention 0/196, lm_head 0/1