criemen

6 ms·

1. 91.▲
   by criemen 1y ago

   This desk (used to?) fit the budget, for example: https://www.architonic.com/en/p/holmris-b8-milk-classic-1070... Essentially, you pay a lot for fancy design.
2. 92.▲
   by criemen 1y ago

   > Indeed perhaps it's important to include a high quality corpus in pre training? I doubt anyone wants to train an LLM from scratch for EDA. That does sound reasonable to me. The main problem is that you (at least for software) can&
3. 93.▲
   by criemen 1y ago

   > Maybe those tools should explicitly confirm executing every external command This wouldn't work - it's not external commands that's the problem, it's arbitrary code that's being executed. That code has access t
4. 94.▲
   by criemen 1y ago

   All links render as blue strike-through line in Firefox (underline in Chrome), hurting legibility :(
5. 95.▲
   by criemen 1y ago

   Presumably because the layoff is targeted at a specific (set of) teams in customer service, rather than a 1% haircut across the company?
6. 96.▲
   by criemen 1y ago

   How much would it cost at list-price API pricing?
7. 97.▲
   by criemen 1y ago

   Running evals aren't the problem, the problem is acquiring or building a high-quality, non-contaminated dataset. https://arxiv.org/abs/2506.12286 makes a very compelling case that swebench (and in extension, anyth
8. 98.▲
   by criemen 1y ago

   Visa and Mastercard have a net profit margin in the order of 50%. Do you think that's justified for the service they're providing, and is indicative of a healthy market/competition?
9. 99.▲
   by criemen 1y ago

   Thanks for the reminder! https://www.youtube.com/watch?v=TjDEsGZLbio
10. 100.▲
    by criemen 1y ago

    Perplexities isn't bad? Although I lack the OpenAI subscription to compare.
11. 101.▲
    by criemen 1y ago

    I totally get it. Due to my work, I mostly keep up with new model releases, but the pace is not sustainable for individuals, or the industry. I'm hoping that model releases (and the entire development speed of the field) will slow down
12. 102.▲
    by criemen 1y ago

    I'd expect that's manageable by some sort of agent-of-agent pattern. You have a high-level planning instance that calls upon fresh LLM instances (new context window!) for executing more targeted tasks or bug-fixes. Currently, an L
13. 103.▲
    by criemen 1y ago

    My understanding is that this is essentially how RLHF works, and it doesn't scale. As you run RL for longer, the model will learn how to cheat the imperfections of the grader, instead of getting better at the task at hand. Therefore, t
14. 104.▲
    by criemen 1y ago

    They have a point about RLs increasing importance. From my outsider perspective, all major advances in model capabilities in the last period of time come from RL, so it's natural to expect that we can "milk" RL more for perfo
15. 105.▲
    by criemen 1y ago

    Thanks a lot for putting this together! I have a question. In https://github.com/bentoml/llm-inference-in-production/blob/... , you have a single picture that defines TTFT and ITL. That does not match my under
16. 106.▲
    by criemen 1y ago

    And on the topic of prefill: Do you know what the role of GPUs is vs. in inference?
17. 107.▲
    by criemen 1y ago

    Thanks for writing the article! I didn't quite get Note that during the prefill phase, all prompt tokens from a request can be processed in one batch. This is possible because the query (Q) tensors, calculated from the tokens immediat