lewiswatson/Frame2KG-SmolVLM-2.2b-JSON
LoRA-style (fewer layers) vs HuggingFaceTB/SmolVLM2-2.2B-Instruct. lewiswatson/Frame2KG-SmolVLM-2.2b-JSON matches part of HuggingFaceTB/SmolVLM2-2.2B-Instruct (219 of its 657 tensors; the rest of HuggingFaceTB/SmolVLM2-2.2B-Instruct has no counterpart). On the shared part: lewiswatson/Frame2KG-SmolVLM-2.2b-JSON is a GGUF (F16) of different weights. 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 24 of 24 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/49, MLP 72/72, Attention 96/96, lm_head 0/1