Built a GPT-3 bot that lets people with no accounting knowledge generate financial statements 📈 👀
Here it is creating balance sheets by turning everyday language into Python code:
A new method was able to delete 40% of LLM layers with no drop in accuracy.
This makes them mich cheaper and faster.
The method combines pruning, quantization and PEFT.
They tested this across various open source models.
Each family of models had a maximum amount of layers
I still primarily use text with GPT-4 but recently started using voice in a more conversational context. It’s been surprisingly good. It picks up on the intent and details quite well (but doesn’t always get the exact words right) and sounds pretty human-like.
This Post is from an account that no longer exists. Learn more
