tomaarsen/Qwen3-VL-Embedding-2B-vdr
Partial fine-tune vs Qwen/Qwen3-VL-Embedding-2B. Partial change: 85% of the weights changed (attention, embeddings, mlp, norms), in 28 of 28 layers.
Tensors changed: Embeddings 3/4, Norms 137/217, MLP 180/196, Attention 208/208