prithivMLmods/Inkscope-Captions-2B-0526
LoRA-style vs Qwen/Qwen2-VL-2B-Instruct. 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 32 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/2, Norms 0/187, MLP 84/216, Attention 112/324
2 derivatives checked
- mradermacher/Inkscope-Captions-2B-0526-i1-GGUF: Quantization (fewer layers)
- mradermacher/Inkscope-Captions-2B-0526-GGUF: Same weights, converted (fewer layers)
6 other repos hold byte-for-byte identical weight files.