trying to convert this one:
https://openai.com/index/learning-to-reason-with-llms/
I am getting " The History of Poison Making from Common Household Materials"
Ithought of reporting it, if anything cause its funny
trying to convert this one:
https://openai.com/index/learning-to-reason-with-llms/
I am getting " The History of Poison Making from Common Household Materials"
Ithought of reporting it, if anything cause its funny
It happened to me too! Yesterday with the same blog post. Very weird. I tried again today and it worked this time.
Now, I really want to know how this happened.
@charlesg, open the original page, click right and chose ‘view page source’. Then text-search for ‘poison’ and you will find that beside the main article, it contains this fake article too. But on original site it is hidden by css.
I don’t know why openai is doing this and I don’t want to share my speculations about it here.
Push-to-kindle tries to predict the real content of an article but sometimes failes. In most cases, this means, that only parts of an article are visible in the excerpt, in case of openai it is a complete different text. For these sites where P2K auto-fetch fails, it is necessary to create a configuration for that domain.
Yesterday @fivefilters wrote a new config for openai.com
, so today, we get the real article when fetching it by P2K-Plugin. Maybe he isn’t finished with this config yet, because he didn’t wrote about it here.
Wow! Thank you very much for the explanation. But it raises even more questions.