Vineeth147

5 ms·

1. 1.▲
   A coding agent in 5 files, plus a skeptic that catches fake fixes(github.com)

   2 points by Vineeth147 2mo ago|0 comments
2. 2.▲
   by Vineeth147 2mo ago

   Done — the why section is now basically that comment, first person, no filler.
3. 3.▲
   by Vineeth147 2mo ago

   I fine-tune small models with limited resources, but it's hard to tell which speedup claims are real. Each method, like DoRA, rsLoRA, Unsloth, or packing tricks, uses different models, data, and hardware in their reports. This makes it
4. 4.▲
   by Vineeth147 2mo ago

   That's fair on both points. Much of this was built with AI, but the runs and numbers are real. They are also reproducible, so I would prefer to be judged on that. And yes, using a single model and task can lead to overfitting. The plan
5. 5.▲
   LoRA Speedrun – a public wall-clock leaderboard for fine-tuning techniques(github.com)

   145 points by Vineeth147 2mo ago|29 comments