Tag
#finetunning
4 posts
- Nov 7, 2024 · Essay · 1 min
Step-by-Step Optimization of LLM by OpenAI
OpenAI's step-by-step scheme allows for excellent results by progressing from simple methods to more complex ones.
- Apr 20, 2024 · Essay · 1 min
Errors and Self-Correction in LLMs
Unlike humans, LLMs don’t correct themselves on the fly. It’s interesting to consider how to train a model for self-correction.
- Apr 1, 2024 · Essay · 1 min
Open Source LLMs and Open Weights LLMs Are Not the Same
Open Source LLMs and Open Weights LLMs are not the same. There aren't many Open Source models on the market.
- May 10, 2023 · Essay · 1 min
What Works Better: Prompting or Finetuning?
Andrej Karpathy's thoughts on comparing prompting and finetuning for training LLMs.