Hacker News: l1n

New comment by l1n in "MegaTrain: Full Precision Training of 100B+ Parameter LLMs on a Single GPU"

l1n — Wed, 08 Apr 2026 13:24:37 +0000

Seems similar to Microsoft DeepSpeed.

Anthropic expands partnership with Google and Broadcom for next-gen compute

l1n — Mon, 06 Apr 2026 21:52:08 +0000

Article URL: https://www.anthropic.com/news/google-broadcom-partnership-compute

Comments URL: https://news.ycombinator.com/item?id=47667717

Points: 283

# Comments: 126

New comment by l1n in "'Project Hail Mary' Crosses $300M in Sales to Become Amazon/MGM's Highest-Gross"

l1n — Sun, 29 Mar 2026 22:40:58 +0000

Watched it twice, cried both times. Probably could have had a few minutes cut while still tight if you care about that kind of thing, recommend watching in IMAX since it's a space movie. One notable annoying science inaccuracy that felt particularly blatent was jurer gur pragevshtr jnf haonynaprq sbe ab ernfba but that's forgivable.

New comment by l1n in "Speed up responses with fast mode"

l1n — Sat, 07 Feb 2026 19:19:08 +0000

2.5x faster or so (https://x.com/claudeai/status/2020207322124132504).

New comment by l1n in "List animals until failure"

l1n — Sun, 01 Feb 2026 15:44:48 +0000

https://rose.systems/blog/list-animals-until-failure on how it's made!

List animals until failure

l1n — Sun, 01 Feb 2026 01:03:23 +0000

Article URL: https://rose.systems/animalist/

Comments URL: https://news.ycombinator.com/item?id=46842603

Points: 363

# Comments: 187

New comment by l1n in "The Adolescence of Technology"

l1n — Mon, 26 Jan 2026 18:43:11 +0000

> Here's something Sonnet 4 did last year

Hate to be that gal but a lot has changed in the past year

New comment by l1n in "Elevated errors across many models"

l1n — Mon, 15 Dec 2025 01:00:20 +0000

We're considering this!

New comment by l1n in "Elevated errors across many models"

l1n — Sun, 14 Dec 2025 23:45:15 +0000

Also an engineer on this incident. This was a network routing misconfiguration - an overlapping route advertisement caused traffic to some of our inference backends to be blackholed. Detection took longer than we’d like (about 75 minutes from impact to identification), and some of our normal mitigation paths didn’t work as expected during the incident.

The bad route has been removed and service is restored. We’re doing a full review internally with a focus on synthetic monitoring and better visibility into high-impact infrastructure changes to catch these faster in the future.

New comment by l1n in "Claude Opus 4.5"

l1n — Mon, 24 Nov 2025 19:40:17 +0000

15$/Megatoken in, 75$/Megatoken out

New comment by l1n in "Major Bitcoin mining firm pivoting to AI"

l1n — Sun, 16 Nov 2025 17:12:22 +0000

Was the talk recorded? Couldn't find a link.

New comment by l1n in "Data centers contribute to high prices as energy bills electrify local politics"

l1n — Sat, 01 Nov 2025 16:39:25 +0000

hm, I think this contradicts the results from https://www.sciencedirect.com/science/article/pii/S104061902...

New comment by l1n in "Claude Sonnet 4.5"

l1n — Mon, 29 Sep 2025 17:28:32 +0000

https://docs.anthropic.com/en/api/openai-sdk

New comment by l1n in "A postmortem of three recent issues"

l1n — Sun, 21 Sep 2025 22:57:06 +0000

> so tough luck if you signed up on iOS via that

FYI that you can sign in with your private relay email address. It's annoying though.

New comment by l1n in "A postmortem of three recent issues"

l1n — Thu, 18 Sep 2025 10:10:58 +0000

> Claude et al offering a cheaper off peak plan We do offer Batch Processing today - https://docs.claude.com/en/docs/build-with-claude/batch-proc...

New comment by l1n in "A postmortem of three recent issues"

l1n — Thu, 18 Sep 2025 10:08:14 +0000

The load balancer container is provided to Bedrock, since it's part of our overall LLM serving system, but they run it.

New comment by l1n in "A postmortem of three recent issues"

l1n — Wed, 17 Sep 2025 22:53:30 +0000

(Anthropic employee, speaking in a personal capacity)

> I’m pretty surprised that Anthropic can directly impact the infra for AWS Bedrock as this article suggests.

We don't directly manage AWS Bedrock deployments today, those are managed by AWS.

> I can’t imagine the average person equates hitting this button with forfeiting their privacy.

We specify

> Submitting this report will send the entire current conversation to Anthropic for future improvements to our models.

in the thumbs down modal. Is there a straightforward way to improve this copy?

New comment by l1n in "Language models pack billions of concepts into 12k dimensions"

l1n — Mon, 15 Sep 2025 15:33:59 +0000

There are several [replications and comments from reviewers](https://transformer-circuits.pub/2022/toy_model/index.html#c...), so yes, it is.

New comment by l1n in "Show HN: 1 min workouts for people who sit all day"

l1n — Sat, 24 May 2025 21:33:54 +0000

Would love to see this for Android!

New comment by l1n in "Claude 4"

l1n — Thu, 22 May 2025 19:57:40 +0000

1P = Anthropic's first party API, e.g. not through Bedrock or Vertex