ydau
Joined 8 years ago
Karma
62
Submissions
39
Member since
Oct 4
Recent submissions
Comment · 3 years ago
Yes! It’s coming to all regions except Utah in a couple weeks. Utah will be a bit longer (couple months)
Comment · 3 years ago
Hey! FYI, we are hoping to roll out a fix for the PyTorch issue tomorrow (I’m one of the founders of Lambda).
Also, that’s good feedback on the CPUs bottlenecking. I’ll let our HPC hardware team know about this.
We are also looking into GPU direct storage to help resolve this:
Comment · 6 years ago
You might also enjoy “Ted the Caver”
Comment · 6 years ago
GPT-3 was another buckets worth of evidence in favor of the scaling hypothesis. Performance kept improving (and cost to train kept increasing) as more parameters were added. Even with 175 billion parameters, the performance had not yet plateaued. One take-away is that throwing a lot of compute at the problem helps tremendously :).
You can read more about GPT-3 here: https://lambdalabs.com/blog/gpt-3/
Comment · 6 years ago
Happy to add any demos / papers to this post that you think are worth linking.