LLM fine-tuning labs
Eight GPU labs from data preparation to a model you trained yourself.
Fine-tuning changes a model's weights on your own data. Parameter-efficient methods such as LoRA (training small low-rank adapter matrices instead of the full model) and QLoRA (LoRA on a 4-bit quantized base) make that possible on a single GPU, which is why they dominate both production practice and the NVIDIA NCP-GENL and NCA-GENL blueprints. The concepts are simple to state and easy to get subtly wrong: the wrong learning rate, a leaking eval set, forgetting that the adapter rank trades capacity for memory.
This collection is ordered the way a real project runs. Prepare and deduplicate the data and train a BPE tokenizer; fine-tune Llama 3 8B with LoRA and then QLoRA on an instruction set; align a model with Direct Preference Optimization using TRL; domain-adapt GPT-2 to Python and measure catastrophic forgetting; train a small GPT-style model from zero on TinyStories. Two labs cover the parts teams skip: an evaluation lab (perplexity, why BLEU misses paraphrase, side-by-side comparison, an LLM judge) and a reproducibility lab that flips PyTorch determinism flags until same-seed runs match bit for bit. A final lab applies LoRA to Stable Diffusion, which shows the technique is not specific to text.
Every lab runs on a live NVIDIA GPU pod inside the browser (Jupyter or terminal), takes 40 to 55 minutes, and is checked step by step.
The labs, in order
All labs- 1GPU Pro# dedup + tokenize$ filter_lang("en")raw ............ 52k → 18ktokenizer ...... BPE 32ktrain_ready .... 16,420ncp-genl · data-prepIntermediate
Data Preparation for LLM Training
Build a real pretraining/instruction data pipeline: load a raw corpus, apply quality filters, deduplicate, train a BPE tokenizer, and batch-validate on GPU. This is the unglamorous work that actually decides how good your model will be.
ncp-genlnca-genlncp-ads45 min5 stepsLaunch - 2GPU Pro# lora · r=16 · alpha=32$ trainer.train()trainable ....... 0.8%eval_loss ....... 1.24perplexity ...... 3.45ncp-genl · fine-tune-llm-lora-jupyterIntermediate
Fine-Tune an LLM with LoRA and QLoRA (Jupyter)
Fine-tune Meta Llama 3 8B on a custom instruction dataset using LoRA and QLoRA. Learn parameter-efficient fine-tuning from data preparation through evaluation — the #1 most demanded AI skill.
ncp-genlncp-adsnca-genl45 min7 stepsLaunch - 3GPU Proncp-genl · dpo-alignmentAdvanced
RLHF & DPO Alignment
Run real Direct Preference Optimization on a small language model with TRL's DPOTrainer. Capture a baseline, build a preference dataset, train, and measurably shift the model's behavior in four steps.
ncp-genlncp-adsnca-genl55 min4 stepsLaunch - 4GPU Proncp-genl · continued-pretrainingAdvanced
Continued Pre-Training: Adapt a Pretrained LM to a New Domain
Take GPT-2 and domain-adapt it to Python code in 150 steps, measuring both the gain on code and the cost in catastrophic forgetting on English. The exact recipe behind Code Llama, BloombergGPT, and every domain-specialized LLM of the last three years.
ncp-genlncp-adsnca-genl45 min4 stepsLaunch - 5GPU Proncp-genl · train-slmAdvanced
Train a Small Language Model from Scratch
Train a real GPT-style language model from zero on TinyStories: tokenize, wire up the optimizer and LR schedule, run the training loop with validation perplexity, and generate coherent text from your own weights. End-to-end pretraining in minutes on one GPU.
ncp-genlncp-adsnca-genl55 min4 stepsLaunch - 6GPU Proncp-genl · evaluationIntermediate
Evaluation & Benchmarking LLMs
Four evaluation lenses in one lab: compute real perplexity, expose BLEU's blindness to paraphrase, run side-by-side model comparisons, and build an LLM-as-judge harness with position-bias detection.
ncp-genlnca-genlncp-ads45 min4 stepsLaunch - 7GPU Pro# determinism flags$ torch.use_deterministic_…seed ............ 42cudnn.benchmark . Falsereproducible .... TRUEncp-genl · reproducible-trainingIntermediate
Reproducible Training: The Flags, The Cost, The Artifacts
Measure the non-determinism noise floor in default PyTorch, flip every determinism flag until same-seed runs match bit-for-bit, quantify the perf cost, and capture a content-addressable training config that makes a run reproducible forever.
ncp-genlncp-aioncp-ads40 min4 stepsLaunch - 8GPU Pronca-genm · sd-loraIntermediate
Fine-Tune Stable Diffusion with LoRA: Custom Text-to-Image
Load Stable Diffusion, attach LoRA adapters to the U-Net's attention layers, run a tiny overfit training loop, and generate with the adapted weights to prove that a few million trainable parameters actually move pixels.
nca-genmncp-genl45 min4 stepsLaunch
Before you start
- PyTorch basics: tensors, a training loop, what a loss curve is
- The idea of a tokenizer and of train/validation splits
- No fine-tuning experience required; the LoRA lab starts from a base model download
Every lab in this collection, plus every practice test on the platform
$29.99 per month or $290 per year. Cancel any time. Free labs need only an account.
Certifications these labs prepare you for
Frequently asked questions
Each GPU lab provisions a dedicated NVIDIA GPU pod for your session. You work in Jupyter or a terminal in the browser; there is nothing to install locally and the environment is torn down when you finish.
Fine-tuning, PEFT/LoRA/QLoRA, data preparation, evaluation and distributed-training concepts are weighted domains on NVIDIA NCP-GENL and NCA-GENL. On our practice data, fine-tuning is the weakest NCP-GENL topic, which is why the collection front-loads it.
Roughly six to seven hours across eight labs of 40 to 55 minutes. Labs are independent, so you can do the LoRA lab alone if that is what you need.
The GPU labs are part of Preporato Pro ($29.99 per month or $290 per year), together with every practice test and every hosted lab. Individual lab pages show the full brief before you subscribe.