LLM Fine-Tuning
Cluster focuses on discussions about fine-tuning large language models (LLMs), including techniques, use cases, tools, alternatives like in-context learning or LoRA, costs, and comparisons to prompting or retraining.
Activity Over Time
Top Contributors
Keywords
Sample Comments
Did you try fine tuning the LLMs?
Don't forget the ability to finetune the LLM.
Can't we just finetune the model based on the LLM's output? Has anyone tried it?
Have you had any success finetuning models? What did you do?
What do you think the main use case for fine tuning small language models is?
No mention of training / fine tuning it through transformers?
is there a well-established tool-chain for finetuning these models?
There are LLM finetunes which do this, it is very far from watertight.
A possible alternative to fine-tuning is in-context learning, especially if you are using a model with long context where you can provide a lot of examples. Models can do one/few-shot learning, but in-context learning improves the more examples you give. You could experiment cheaply with Claude Haiku to see if this works for you.
In the interest of transparency you should update your post with the model you fine tuned, it matters.