Pinned Loading

1. GWSysSec/unsaferust-benchmark-artifact GWSysSec/unsaferust-benchmark-artifact Public

   LLVM

2. W2S_Safety W2S_Safety Public

   Weak to Strong Alignment implementation for OpenAi GPT-4o. Safety evaluation for Weak+Strong model over baselines and considerations for better benchmarking frameworks

   Jupyter Notebook 1

3. memory-llama memory-llama Public

   Adding short-term memory to a local Ollama chat bot using model-tools. Using agentic protocols to give a locally run LLMs pseudo increase in their context window

   Python

4. FRecogRPI FRecogRPI Public

   Face Recognition Algorithm for the Raspberry Pi in C. OpenMp used to parallelize code to decrease load times. Libraries are rework of Raspicam libraries provided by the Pi Foundation.

   C 1

5. vulong2505/attention-visualizer vulong2505/attention-visualizer Public

   Visualizing BERT-base-cased's attention heads on their subject-verb relationships through pattern analysis and BertViz visualization tools. This project explores how different attention heads captu…

   Jupyter Notebook