Insights, tips, and product updates

Ensuring the safety of large language models (LLMs) across languages is crucial as AI becomes more integrated into our lives. This article presents our red teaming study at Kili Technology, evaluating LLM vulnerabilities against adversarial prompts in both English and French. Our findings reveal critical insights into multilingual weaknesses and highlight the need for improved safety measures in AI systems.

This article explores the latest developments in AI safety, including regulatory frameworks, risk management strategies, and technical approaches to mitigate potential harms. From red teaming and toxicity detection to reinforcement learning from human feedback, we delve into the multifaceted efforts to create AI systems that are not only powerful, but also reliable, ethical, and aligned with human values.

Explore the latest techniques in red teaming large language models (LLMs) and crafting adversarial prompts. Learn how to evaluate and improve AI safety through automated testing methods, benchmark datasets, and robust defense strategies.

Discover how to surpass frontier LLM performance using Reinforcement Learning from Human Feedback (RLHF) with this recap of our latest webinar featuring Adaptive ML.

The successful implementation of SLMs requires careful planning, from dataset preparation to continuous evaluation and monitoring. Organizations must adopt a proactive approach to manage these models, ensuring they continually meet the evolving demands and maintain alignment with business goals.

Explore SmolLM, a compact yet powerful language model challenging the notion that bigger is always better in AI. Learn how its meticulously curated datasets and efficient design deliver high performance with lower resource demands, making it ideal for applications in education, coding, and customer support.

Explore how GPT-4o Mini delivers advanced AI capabilities at a reduced cost. Learn about its key features, performance comparisons, and real-world applications, making it a top choice for businesses seeking efficient and cost-effective AI solutions.

Llama 3.1 is Meta's latest flagship language model, boasting an impressive 405 billion parameters. This article dives into how the model was trained and fine-tuned, and pulls out new insights for domain-specific LLMs.

Imagine a Michelin-starred restaurant. The success of each dish relies on three key elements: premium ingredients (Food), the right tools (Tool), and a skilled team (Team).

Learn more about DBRX's model training, focus on data quality, and evaluation, showcasing how high-performing enterprise-level large language models are made.

This article explores the best practices and insights from FineWeb's documentation, offering a comprehensive guide for anyone seeking to build high quality datasets for their language models.

Generative AI (GenAI) is transforming industries by significantly boosting productivity and efficiency. This article explores GenAI's impact, emphasizing the crucial role of high-quality datasets in its success. We discuss challenges in creating these datasets and propose innovative solutions to enhance data processes, ensuring optimal performance of GenAI models.

Gemma 2 is an advanced AI model from Google, featuring diverse data integration, rigorous preprocessing, and comprehensive evaluations to ensure high performance, safety, and ethical behavior. Learn how Gemma 2 balances cutting-edge capabilities with robust security and ethical standards.

To ensure widespread adoption and long-term value delivery of your RAG application, following best practices when deploying such use cases is essential. By doing so, we can maximize the potential value that they offer.

Fine-tuning and aligning LLMs can be a challenging and complex process. In this article, we hope to clarify and structure this journey based on our experience with clients and existing examples.

Assessing and benchmarking LLMs makes it easier for data science teams to select the right model and develop a strategy to adapt it faster. Here's a guide to building an LLM evaluation dataset.

Learn how to fine-tune large language models (LLMs) for specific tasks using Kili Technology. This tutorial provides a step-by-step guide and example on fine-tuning OpenAI models to categorize news articles into predefined categories.

Discover examples and techniques for developing domain-specific LLMs (Large Language Models) in this informative guide.

We share 9 open-sourced datasets used for training LLMs, and the key steps to data preprocessing.

Reinforcement learning from human feedback (RLHF) allows large language models to understand human instruction naturally. This approach allows LLMs with lesser parameters to perform better. It also ensures the model avoids producing dangerous behavior. Still, RLHF is an evolving process, with challenges researchers must overcome.