Yeah, if the benchmark is actually predictive of the tasks you have then it is trivial to conclude that the cheapest-per-benchmark-task model will be the cheapest one for your tasks…
Not even an average task. I can have a single task that I need to do and I could be choosing which model to use. The cheapest-per-benchmark-task model would be useless to me if it cannot do the task I need.
Cost per token doesn't say a lot, but "Cost per benchmark task" is also meaningless if your task is difficult enough that the cheaper model has no chance of cracking it.
This website was made by St. Jude Children's Research Hospital. AFAIK they are a non-profit which runs treatment clinical trials on children cancer patients and doesn't bill them for anything.
The course was mentioned in a recent whoishiring thread. Sounds like it could be a purposeful place to work.
I'm not affiliated in any way, just found it interesting.
> I have never managed to connect to LLM writing, it always comes off as shallow and vapid.
Me too, but I would be careful about being too dismissive, because I would totally bet that at some point the models will be able to write top tier stories.
And there will be people who will find those stories soulless purely based on their origin (which is completely fine!) and call them slop (which I feel hurts the language).
> and then later they can ban you which deprives you of being able to use what you bought, that should come with requirements that the company must provide full compensation of the purchase price.
Wouldn't this incentivize people to buy the game, play it and once they are done with it get banned to get a full refund?
Maybe not a major problem since almost no centralised multiplayer games are one-time purchases.
> we will be able to answer with 100% certainty that we are not alone or unique in the universe since we recreated life in the fucking petri dish
I don't think we are alone, but this is not logically sound. The conditions in the petri dish might be easily so special that their natural prevalence is < 1 per universe.
I co-created https://sugarlessapp.com/
I remain fascinated by llms & the diffusion models.
I use #FF4F00 for my topcolor.