Qiaowenshu/Tongda1-1.5B-BKI
LoRA-style vs Qwen/Qwen2-1.5B-Instruct. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of q_proj, v_proj changed, in 28 of 28 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/57, MLP 0/84, Attention 56/196
2 derivatives checked
- mradermacher/Tongda1-BKI-1.5b-GGUF: LoRA-style
- mradermacher/Tongda1-1.5B-BKI-GGUF: Same weights, converted