Fine-tuning LLMs
Fine-tune open-source LLMs with LoRA/QLoRA for custom tasks and domain adaptation.
0%
Confidence · no data yet
After this skill you can…
- Prepare fine-tuning datasets
- Run LoRA/QLoRA training with Unsloth or HF Trainer
- Evaluate and merge adapters
- Push models to Hugging Face Hub
Prerequisites
Watch (10 videos)
Meet Cosmos 3: Our Latest Frontier Model for Physical AI
→ Fine-tune Cosmos 3 for specific tasks→ Optimize model performance
How to turn general AI into an industry expert! (2-minute AI with Google)
→ Fine-tune pre-trained models for industry-specific problems→ Improve model performance on specialized tasks
LoRA Deep Dive: Rank, Alpha, and Dropout Explained (With Code)
→ Fine-tune large language models efficiently→ Implement LoRA with code
Smaller, faster, smarter: Distilling models with fine‑tuning | DEM322
→ Fine-tune large language models→ Optimize model performance for specific tasks→ Use supervised fine-tuning for improved accuracy
New Google Gemini Update is INSANE!
→ Fine-tune AI models for tasks like Sudoku and code generation
Freezing vs. Unfreezing Layers: The Ultimate Guide to Transfer Learning
→ Optimize model performance→ Improve compute efficiency→ Enhance model accuracy
High Entropy & KL Divergence Token Masking for SFT
→ Fine-tune LLMs with high entropy and KL divergence token masking→ Optimize model performance with selective fine-tuning→ Apply entropy-KL divergence-based token masking for improved results
Frontier Tuning: Microsoft Build 2026
→ Fine-tune pre-trained models for custom data→ Optimize model performance for specific tasks
Chapter 7: LLM Finetuning Explained: LoRA, PEFT & When to Fine-Tune
→ Fine-tune LLMs→ Adapt foundation models→ Improve model performance
Mastering LLM Fine Tuning
→ Fine-tune LLMs for specific tasks→ Improve model performance with LoRA→ Apply parameter efficient fine-tuning techniques
DeepCamp AI