2410.21680

N/A

# link_1b90g52z1kw.pdf ## Metadata - PDFFormatVersion=1.5 - IsLinearized=false - IsAcroFormPresent=false - IsXFAPresent=false - IsCollectionPresent=false - IsSignaturesPresent=false - CreationDate=D:20250210011133Z - Creator=LaTeX with hyperref - ModDate=D:20250210011133Z - Custom.PTEX.Fullbanner=This is pdfTeX, Version 3.141592653-2.6-1.40.25 (TeX Live 2023) kpathsea version 6.3.5 - Producer=pdfTeX-1.40.25 - Trapped=False ## Contents ### Page 1 2025 IEEE International Symposium on High-Performance Computer Architecture (HPCA)Revisiting Reliability in Large-Scale Machine Learning Research C

saved by

• x

related reading

• [2510.13786] The Art of Scaling Reinforcement Learning Compute for LLMsarxiv.org
• arxiv.org/pdf/2512.24880#page=3.56arxiv.org
• Multi-Datacenter Training: OpenAI's Ambitious Plan To Beat Google's Infrastructuresemianalysis.com
• 2403.09611.pdfarxiv.org
• From bare metal to a 70B model: infrastructure set-up and scripts - Imbueimbue.com
• [2410.05686] Deep Learning and Machine Learning with GPGPU and CUDA: Unlocking the Power of Parallel Computingarxiv.org
• 2409.02908arxiv.org
• IsoCompute Playbook: Optimally Scaling Sampling Compute for RL Training of LLMscompute-optimal-rl-llm-scaling.github.io
• 2405.01741arxiv.org
• 2502.11089arxiv.org
• 2404.17625arxiv.org
• [2602.05910] Chunky Post-Training: Data Driven Failures of Generalizationarxiv.org