Popular repositories Loading

1. cs336-alignment-rl cs336-alignment-rl Public

   GRPO, off-policy policy-gradient variants, DPO, SFT, and safety/MMLU evaluation utilities.

   Python 1

2. cs128-cpp-machine-problems cs128-cpp-machine-problems Public

   C++ machine problems portfolio from CS128

   C++ 1

3. cs336-systems cs336-systems Public

   FlashAttention, DDP, optimizer sharding, and FSDP implementations for efficient Transformer training.

   Python

4. cs336-data-pipeline cs336-data-pipeline Public

   Common Crawl extraction, filtering, PII masking, toxicity filtering, deduplication, and tokenization for LM training.

   Python

5. cs336-scaling-laws cs336-scaling-laws Public

   IsoFLOPs scaling-law analysis and experiment infrastructure for compute-optimal language model training.

   Python

6. toy_multi_agent toy_multi_agent Public

   A reliability-first LangGraph and MCP multi-agent system evaluated on a balanced GAIA benchmark sample.

   Python