samaritan-ai/LightOnOCR-2-1B-sam-44-mss-alb-GGUF
LoRA-style (fewer layers) vs lightonai/LightOnOCR-2-1B-base. samaritan-ai/LightOnOCR-2-1B-sam-44-mss-alb-GGUF matches part of lightonai/LightOnOCR-2-1B-base (310 of its 532 tensors; the rest of lightonai/LightOnOCR-2-1B-base has no counterpart). On the shared part: samaritan-ai/LightOnOCR-2-1B-sam-44-mss-alb-GGUF 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, o_proj, up_proj changed, in 28 of 28 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/113, MLP 84/84, Attention 28/112