2025-12-05
The idea is WILD. One idea would be to ask the Claude Code to figure out the hyperparameters that best train a regime of models. Another one would be to work on distillation of bigger models. The ideas are endless here (really cool work)
Hugging Face
Hugging Face details how it used its new tool, Skills, to fine tune an LLM using Claude, including for writing scripts, submitting jobs to cloud GPUs, and more
We gave Claude the ability to fine-tune language models using a new tool called Hugging Face Skills. X: @ben_burtenshaw , @donvito , @ben_burtenshaw , @arig23498 , and @ben_burtens...