Fine-tune Google's Gemma 3

Re:Fine-tune Google's Gemma 3 by 317070 on Thursday March 20, 2025 @04:44PM (#43428720)

I take my quip both ways, so I would wager that even with finetuning, these models are only 1 generation ahead in esoteric language performance and therefore _still not very good_. Am I correct?

[ Parent | on HN ]

Re:Fine-tune Google's Gemma 3 by fragmede on Friday March 21, 2025 @05:44AM (#43433657)

Wanting it to be bad reeks of copium.

[ Parent | on HN ]

Re:Fine-tune Google's Gemma 3 by 317070 on Saturday March 22, 2025 @10:18AM (#43445765)

Why would I want it to be bad? I'm afraid I don't understand what you mean.

[ Parent | on HN ]

Re:Fine-tune Google's Gemma 3 by fragmede on Sunday March 23, 2025 @06:12AM (#43451972)

you wrote, emphatically, that it would be "still not very good". Why do you believe that it would be still not very good after training on a specific problem? LLMs aren't able to do things outside their training data, as vast as it is, but if it's in it's training data, why are you emphatic that it's still not very good? If I ask it to make something that it just needs to copy out sample code of, it would be pretty good at that one very specific task to me.

[ Parent | on HN ]