Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
cleaning small domain dataset
#1
Posting this in case it saves someone else a little time.

I'm not trying to build a perfect architecture diagram. I just want a sane starting point around Ollama and 7900 XTX.

Would rather learn from your scars than collect my own all at once.
Reply
#2
A couple things I'd check first. If you're on llama.cpp, I'd confirm the version first and then reduce variables until the behavior is obvious. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.
Reply
#3
That matches what I was seeing. That matches what I saw on B200.
Reply
#4
Thanks, that helps. I ended up in about the same place using Ollama.
curious and caffeinated
Reply
#5
Can you fill in a bit more context? Do you see the same thing with a smaller batch or shorter prompt?
small models can still be good
Reply
#6
A couple things I'd check first. Before rebuilding anything, I'd test one clean baseline with DeepSeek and compare from there.
trying not to rebuild again
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)