Running 2.41k 2.41k The Ultra-Scale Playbook π The ultimate guide to training LLM on large GPU Clusters
Running 61 61 Scaling FineWeb to 1000+ languages: Step 1: finding signal in 100s of evaluation tasks π Evaluate multilingual models using FineTasks
Running 106 106 TxT360: Trillion Extracted Text π Create a large, deduplicated dataset for LLM pre-training
Running 116 116 Open-LLM performances are plateauing, letβs make the leaderboard steep again π Update leaderboard for fair model evaluation