papers
Publications (2)
cs.DC2026
Training LLMs with Fault Tolerant HSDP on 100,000 GPUs
Omkar Salpekar, Rohan Varma, Kenny Yu +20
Large-scale training systems typically use synchronous training, requiring all GPUs to be healthy simultaneously. In our experience training on O(100K) GPUs, synchronous training r…
cs.DM2014
A Generalization of the Fibonacci Word Fractal and the Fibonacci Snowflake
José L. RamÃrez, Gustavo N. Rubiano, Rodrigo de Castro
In this paper we introduce a family of infinite words that generalize the Fibonacci word and we study their combinatorial properties. Moreover, we associate to this family of words…