C.W.K.
Stream
Quiz · 5 questions

🎯 Trainer + PEFT

SFT, LoRA, QLoRA, DPO

Level 0Scout
0 XP0/50 lessons0/10 achievements
0/120 XP to next level120 XP to go0% complete

Quiz

01What is the effective batch size when per_device_train_batch_size=4, gradient_accumulation_steps=8, on 2 GPUs?
02Why use DataCollatorForCompletionOnlyLM for instruction tuning?
03When should you fine-tune instead of using RAG?
04What does QLoRA quantize?
05When should you train with LoRA but ship merged?
Spotted a bug or have feedback on this page?Report an Issue

Comments 0

🔔 Reply notifications (sign in)
Sign inPlease sign in to comment.

No comments yet — be the first.