Pinned Loading

1. LMLMAudit LMLMAudit Public

   A causal audit of forgetting and knowledge externalization in Limited Memory Language Models.

   Python

2. RetrievalAugmentedGeneration RetrievalAugmentedGeneration Public

   BM25 + LLM + extractive fallback for short-answer QA over a UC Berkeley EECS web crawl, instrumented to measure where each component actually breaks.

   Python

3. andersvestrum/llumnix_sim andersvestrum/llumnix_sim Public

   Forked from microsoft/vidur

   A large-scale simulation framework for LLM inference

   Python

4. RISCVProcessorFPGA RISCVProcessorFPGA Public

   Four-stage pipelined RV32I/RV32F RISC-V SoC for the PYNQ-Z1, with a custom FPU, UART boot, and on-chip memories. Simulation to bitstream.

   Verilog 1

5. ForgeLM ForgeLM Public

   A transformer language model built from scratch, from byte-level BPE tokenization through pretraining and QA fine-tuning.

   Python 1

6. abhip02/VerifAI abhip02/VerifAI Public

   Forked from BerkeleyLearnVerify/VerifAI

   Fork for Compositional Analysis research. VerifAI is a software toolkit for the formal design and analysis of systems that include artificial intelligence (AI) and machine learning (ML) components.

   Python