@theburningmonk It happened to me with multiple models. After trying to reproduce it, I think that when the model was generating a response it found a partial sentence that strongly predicted spanish words, then it made it shift unintentionally. Hard to know for sure, I guess.
Venezolano, a través de este enlace podrás ver cómo con tu voto y tu voluntad, cambiaste la historia de Venezuela.
Aquí encontrarás las actas que hasta esta hora hemos procesado y totalizado, y que confirman nuestro extraordinario triunfo:
https://t.co/UdKJvhm7Eb
If you are running thousands of DAGs across multiple envs, or heavily use workers for compute 🤔, you can now overcome related capacity issues by scaling to the new XL and 2XL environments in #Amazon#MWAA
https://t.co/JXo28SxABS
@donkersgood Bard is also confused: "I have searched other sources and found that the quote is attributed to Marc Andreessen in a number of articles and interviews, but I have not been able to find the original source of the quote." 😬
🔥 Apache Airflow v2.6.3 is available on Amazon #MWAA 🔥
If you’re running Apache #Airflow version 2.x, you can seamlessly upgrade to v2.6.3 using in-place version upgrades, retaining your workflow run history and environment configurations.
https://t.co/ImFTLIJaNm
The internet is talking about retries and backoff! As we've seen over the last day or so, simply retrying without backoff leads to an unbounded increase in work. This, in turn, tends to make overloaded systems even more overloaded. What can we do instead?
Dan Urson has been the voice of AWS security to many of us. He's the person external researchers interact with when they find issues (or think they find issues) with AWS. Any team would be lucky to have him.
https://t.co/Uf3WZbzx8u
When people claim LLMs will replace software engineers, it either indicates a lack of understanding of LLMs or a lack of understanding of software engineering...