mradermacher/SmolVLM2-2.2B-Instruct-Agentic-GUI-GGUF
Full fine-tune (fewer layers) vs smolagents/SmolVLM2-2.2B-Instruct-Agentic-GUI. mradermacher/SmolVLM2-2.2B-Instruct-Agentic-GUI-GGUF matches part of smolagents/SmolVLM2-2.2B-Instruct-Agentic-GUI (219 of its 657 tensors; the rest of smolagents/SmolVLM2-2.2B-Instruct-Agentic-GUI has no counterpart). On the shared part: mradermacher/SmolVLM2-2.2B-Instruct-Agentic-GUI-GGUF is a GGUF (F16) of different weights. Fine-tune: 100% of the weights changed, across 24 of 24 layers.
Tensors changed: lm_head 1/1, Embeddings 1/1, Norms 49/49, MLP 72/72, Attention 96/96