Mihir Patel
Mihir Patel's posts
- Research
Training MoEs at Scale with PyTorch and Databricks
<1 min read - Research
Building DBRX-class Custom LLMs with Databricks Training
9 min read - Research
Bringing MegaBlocks to Databricks
4 min read - Research
Turbocharged Training: Optimizing the Databricks Stack With FP8
4 min read - Research
How We Trained Stable Diffusion for Less than $50k (Part 3)
6 min read - Research
Training Stable Diffusion from Scratch for <$50k with MosaicML (Part 2)
6 min read - Research
Farewell, CUDA OOM: Automatic Gradient Accumulation
5 min read
Get the latest posts in your inbox
Subscribe to our blog and get the latest posts delivered to your inbox.