Fireside Alpha
@firesidealpha
Sachin Katti says OpenAI had Astra optimize its own inference on NVIDIA's Rubin chip, adding another 2x in 72 hours
"It's a testament to NVIDIA's hardware that out of the box we got a 3x improvement in throughput. We literally took the same code that we had optimized for
"It's a testament to NVIDIA's hardware that out of the box we got a 3x improvement in throughput. We literally took the same code that we had optimized for
0:48Fireside Alpha@firesidealpha · Sep 17OpenAI's Sachin Katti says the AI safety debate is an argument for more compute, not less, because alignment itself has to be trained
"If anything, I think it actually double-underscores the need for more compute."
"The way we will achieve safety and alignment for these very
"If anything, I think it actually double-underscores the need for more compute."
"The way we will achieve safety and alignment for these very
0:51 0 60