-
blog Public
A list of my open source contributions, and ML Qs and As
-
typed-argument-parser Public
Forked from swansonk14/typed-argument-parserTyped argument parser for Python
Python MIT License UpdatedJul 4, 2025 -
mpnrl Public
Multiple Positives and Negatives Ranking Loss
-
test_set_contamination Public
Forked from tatsu-lab/test_set_contaminationallows LoRA, and can input dtype. used for https://github.com/kddubey/pretrain-on-test/blob/main/analysis/contamination/test.ipynb
Python UpdatedDec 28, 2024 -
pretrain-on-test Public
Finetuning Mistral 7B 2,000 times + BERT and GPT-2 135,000 times, for science. Appeared at GenBench Workshop, EMNLP 2024
-
cappr Public
Completion After Prompt Probability. Make your LLM make a choice
-
backprompt Public
Lazy KV cache data structure for language models
-
-
MEME Public
Re-implementation and some initialization analysis of Multiple EM for Motif Elicitation (MEME).
Python MIT License UpdatedJul 10, 2023 -
ORIE4741Project Public
Forked from nickmorales/ORIE4741ProjectModel parts of the building permit approval process in San Francisco. Interesting parts I worked on: survivorship bias, multistage modeling, and data messiness.
Jupyter Notebook UpdatedApr 9, 2021