I’m building a trip itinerary for upcoming travel. It’s helpful but very iterative. Gemma QAT has weird memory

That’s funny. When I first starred using an LLM, I threw a list of cities at it and asked for the shortest trip though all of them ( traveling salesman problem) and it was so bad I immediately knew not to trust answers from then on

Oh no… Ansible is on my “to learn” list to help with various deployments. Is it that bad? Do you have you’d recommend instead?

this is the most responsible idea. i love it

that’s a great idea re: music. i use a ton of Chromecast audios and, when Google “forgot to update the certs” I was SOL

How can one reproduce this?

It’s not often discussed that Microsoft, Amazon, and other ‘very cool’ tech companies still practice this 80s firing the bottom 10% bull.. It has nothing with being a low performer, but everything to do with culling jobs indiscriminately and creating an in/out culture. Managers are forced to place employees into bottom performance slots, sometimes arbitrarily, so that teams/groups/departments fit the bell curve nicely. Then, people get fired, bonuses go out, and everything is peachy at the end of the fiscal yesr.

Let that stew for a few years and these behemoths are factories of pet projects and crunch culture. Your long-timers are either golden performers who’ve seen massive churn and project turbulance OR they’re climbers, willing to step on top of anyone or burn any project to climb the ladder or at least keep their spots.

Thanks. I’ll try it out!

https://pytorch.org/blog/quantization-aware-training/

I had heard of it but I’m not aware of public models implementing this