PinnedReweighting: Refining AI with Precision and EfficiencyArtificial intelligence is as much about innovation in data management as it is about algorithmic advancement. “Reweighting,” a novel…Nov 10, 2023Nov 10, 2023
PinnedLlama-2, Mo’ LoraMeta just released LLaMA-2, a transformer trained on 2 TRILLION tokens of natural language data! And many are itching to be some of the…Jul 20, 20231Jul 20, 20231
PinnedSmall-Scale Home Evaluation of Google’s New Optimizer “Lion”I evaluate “Lion” a neural network optimizer produced by Google via an Evolutionary Algorithm.Feb 16, 20231Feb 16, 20231
PinnedInverse Scaling in Sentiment AnalysisI wanted to see how well conditional generative models could be adapted, without training, to classification tasks. I was inspired by a…Feb 13, 2023Feb 13, 2023
XPapers — Paying ‘Attention’ to your Dataset MixtureA simple method by which you can optimize the ratios of domains in your training dataset for your large-language-models.Oct 30, 2023Oct 30, 2023
Vicuña, an Instruct-GPT-J Model Trained on Free Hardware; an EvaluationSparks of Normal Language Modelling, Excruciating experiments with my pal VMar 26, 20231Mar 26, 20231
RLHF Automatic Prompt Engineer for Stable Diffusion 2Using Reinforcement Learning from Human Feedback to create the ultimate automatic prompt engineer for Stable Diffusion 2Dec 1, 2022Dec 1, 2022
An Intro to Transformers, from a TransformerThis text was a conversation with OpenAI’s ChatGPT curated and edited by Crumb to create a simple intro to BERT for classification and…Dec 1, 2022Dec 1, 2022
My 2022 Code Journey RecapThings I’ve made this year. I’m writing this so I can have my own catalog of important code that I’d want to come back to. To either revise…Nov 5, 2022Nov 5, 2022