The easiest LLM Fine Tuning UI just Landed! π
Now, ANYONE can fine-tune (almost) any LLM available on Hugging Face Hub by just uploading a CSV and choosing the parameters and by a single click of a button! π₯ Here's how you can do it: 1/N
Now, ANYONE can fine-tune (almost) any LLM available on Hugging Face Hub by just uploading a CSV and choosing the parameters and by a single click of a button! π₯ Here's how you can do it: 1/N
First, you need a huggingface account! If you dont have one, create one: hf.co.
Once your account is setup, click this link: huggingface.co
You can choose any name for the space 2/N
Once your account is setup, click this link: huggingface.co
You can choose any name for the space 2/N
huggingface.co/new-space?tempβ¦
Hugging Face β The AI community building the future.
Weβre on a journey to advance and democratize artificial intelligence through open source and open s...
hf.co
Hugging Face β The AI community building the future.
Weβre on a journey to advance and democratize artificial intelligence through open source and open s...
After that, enter your huggingface write token. you can find/create your write token here: huggingface.co and enter task as "LLM" (without quotes). Make sure to keep your space private! 3/N
Space will be created in a few seconds. And you will get the screen below. Now, all you need to do is upload your dataset in proper format, choose the column mapping, choose hyperparameters, the model and the backend. You can add as many jobs as you like! 4/N
Your training spaces will be visible in your account and you can monitor live logs. The spaces will shut themselves down on completion. So, you pay only for the training time! π₯ Models will be saved to your account as private repositories! 5/N
How much does it cost? You will be billed per minute for the training. Here's how much the backends cost right now. 6/N
If you face any issues or have feature requests, check out the github repository βοΈ 7/N
github.com
github.com
Loading suggestions...