tencent/WeMM-Embedding-4B
Full fine-tune vs Qwen/Qwen3.5-4B. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: MLP 148/196, Attention 249/320, Norms 107/203, Embeddings 1/3
2 derivatives checked
- TuTuCSF/WeMM-Embedding-4B-GGUF: Same weights, converted
- seamon67/WeMM-Embedding-4B-GGUF: Quantization