Fresher / New Grad

target_modules in a PEFT config: what does this list control?

Should the system message contribute to the SFT loss?

Step versus epoch in a training run: define the relationship.

ShareGPT format: what data shape does it use?

Fill in the SFT JSONL: the three required fields of a typical chat example.

Define supervised fine-tuning (SFT) in one breath.

save_strategy='steps' vs save_strategy='epoch': what behaviour changes?

Roles in a chat training example: pick the three that are standard.

Reward model in RLHF: what does it score, and on what data does it train?

Rank r in LoRA: what does the number actually count?

Pretraining vs fine-tuning: two key differences in data and objective.

Preference pair: name the two pieces and what the label expresses.

PEFT: unpack the acronym and the core problem it tackles.

Paged optimizer in QLoRA, what gets paged, and to where?

Pick the sensible `num_train_epochs` for a typical instruction-SFT run on a curated dataset

When does a PEFT config actually need `modules_to_save` set, and what goes in it?

Modern LLM fine-tuning defaults to causal LM or masked LM, pick one and say why

After calling `merge_and_unload()` on a PEFT model, what does the returned object look like?

`max_seq_len` controls one bound, what happens to examples that exceed it?

Why ramp the learning rate during warmup instead of starting at the peak value?

Which loss function does supervised fine-tuning actually minimize?

In LoRA, where in the forward pass does `lora_dropout` apply its mask?

Gradient norm clipping, name the failure mode it exists to prevent

Gradient checkpointing, what does it trade, and roughly by how much?

Compute the effective batch from per-device batch, accum steps, and GPU count