Quiz · 5 questions
🤗 Hugging Face Stack
The open-source path with TRL v1.0
Level 0Observer
0 XP0/43 lessons0/11 achievements
0/120 XP to next level120 XP to go0% complete
Quiz
01Which TRL class is the recommended way to do supervised fine-tuning in 2025–2026?
02What does packing=True do in SFTConfig?
03Which GGUF quantization format is the most-used balance of quality and size?
04Which function prepares a quantized model for QLoRA training?
05Why is
dataset_kwargs={'assistant_only_loss': True} recommended in modern SFTTrainer setups?Comments 0
🔔 Reply notifications (sign in)Sign in — Please sign in to comment.
No comments yet — be the first.