unsloth/Qwen3-Embedding-0.6B
LoRA-style vs Qwen/Qwen3-0.6B-Base. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj changed, in 28 of 28 layers; norms, biases and embeddings are untouched.
Tensors changed: Norms 0/113, MLP 84/84, Attention 112/112
3 derivatives checked
- ENOSYS/Qwen3-Embedding-0.6B-250-v1-GGUF: Quantization
- lastmass/Qwen3-Embedding-Medical-0.6B: LoRA-style
- Casual-Autopsy/QwEmotion-RP-Embed-0.6B-v0.1: LoRA-style
38 other repos hold byte-for-byte identical weight files.