Intellectual hub of the topic
AI: Events
Technical context • Infrastructure
AMD demonstrates how to set up LLM training on GPU clusters so that failures are handled automatically, eliminating the need for manual intervention.
AI: Events
Development
The Photoroom team shares how they managed to train their own image generation model in just 24 hours and what the results were.
AI: Events
Research
Sakana AI has proposed a method to instantly update the knowledge of language models without costly retraining – by generating adapters directly from text.
We explain how the Mixture of Experts architecture works – an approach that makes models smarter without making them 'think' harder.
AI: Events
Development
AMD has introduced LuminaSFT, an approach that uses synthetic data to fine-tune small language models and achieve surprisingly high performance.
AMD has released JAX-AITER, a library of pre-built, optimized computational blocks for developing large AI models on AMD GPUs using the JAX framework.
AI: Events
Technical context • Infrastructure
AMD has unveiled Primus, an implementation of pipeline parallelism for large model training that eliminates idle time and flexibly adapts to various tasks.
AI: Events
Development
Researchers from Tencent have developed a tool that helps to precisely identify where failures occur during reinforcement learning model training.
AI: Events
Development
Allen AI has introduced Olmix, an open-source framework for data mixing in the language model training process, including pre-training, instruction tuning, and alignment.
Don’t miss a single experiment!
Subscribe to our Telegram channel —
we regularly post announcements of new books, articles, and interviews.