PinnedInData Science CollectivebyPere Martra·Apr 7, 2025From Biased to Balanced: Visualizing and Fixing Bias in Transformer ModelsIn this article, we’ll explore how internal activations in transformer models reveal hidden biases, and how these insights can be used to…A response icon3A response icon3
InLevel Up CodingbyPere Martra·Feb 26Fairness Pruning: Finding Demographic Bias Neurons in LLMs (And What Happens When You Remove Them)Zeroing just 5 neurons is enough to reshape how an LLM responds to demographic prompts
InData Science CollectivebyPere Martra·Feb 19What Transformer Blocks in Your LLM Aren’t Doing Anything?Every time you run inference on an LLM, you’re paying for the whole model. But for most real-world tasks, a significant chunk of that model…A response icon1A response icon1
InGoPenAIbyPere Martra·Feb 1Rearchitecting LLMs. From API Consumer to Model Architect.Bridging the GAP between research and engineering, presenting and explaining the techniques used to create the model families of the main…A response icon1A response icon1
InTDS ArchivebyPere Martra·Nov 27, 2024How to Prune LLaMA 3.2 and Similar Large Language ModelsThis article presents a structured pruning technique for state-of-the-art models, that uses a GLU architecture, enabling the creation of…A response icon1A response icon1
InTowards AIbyPere Martra·Oct 7, 2024Evaluating a RAG Solution with GiskardGiskard is an open-source project dedicated to providing a framework for the evaluation of different machine-learning models. Here, you…
InTowards AIbyPere Martra·Sep 25, 2024Tracing a LLM Agent with LangSmithWhat happens inside an agent? Learn how to use LangSmith to dive into the inner workings of your agent.
InTowards AIbyPere Martra·Sep 16, 2024How to Create a Medical Agent / RAG System.Learn how to build a ReAct agent powered by a RAG system using LangChain, ChromaDB, and OpenAI, with a user-friendly Gradio interface.A response icon3A response icon3
InLevel Up CodingbyPere Martra·Aug 27, 2024Influencing a Large Language Model response with in-context learning.A simple introduction to prompt Engineering with OpenAI. Some differences between gpt-3.5 and gpt-4o.
InTowards AIbyPere Martra·Aug 14, 2024First NL2SQL Chat with OpenAI and Gradio.You are going to create a simple Natural Language to SQL translator using models from OpenAI and create the interface with Gradio.