Fine-tuning LLMs

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)

Fine-tuning T5 LLM for Text Generation: Complete Tutorial w/ free COLAB #coding

Discover AI · beginner hands-on

→ Fine-tune LLMs for downstream tasks→ Use COLAB for LLM development→ Implement text summarization with LLMs

Train image classifier using transfer learning - Fine-tuning MobileNet with Keras

deeplizard · advanced hands-on

→ Fine-tune pre-trained models for custom tasks→ Adapt models to new datasets

LLM Fine-tuning: Two Crucial Tips for New Models - LLama 2

Discover AI · advanced hands-on

→ Fine-tune a LLama 2 model on a custom dataset→ Optimize LLM workflow

Fine-tuning with TensorFlow

HuggingFace · beginner hands-on

→ Fine-tune Transformers models→ Use TensorFlow and Keras for LLMs

MedAI: Vision Language Models & Fine-Tuning (KnowAda)

Discover AI · advanced hands-on

→ Fine-tune a vision language model using KnowAda→ Mitigate hallucinations in multimodal models

Robust Fine-Tuning of Zero-Shot Models

Connor Shorten · advanced hands-on

→ Fine-tune pre-trained models for specific datasets→ Improve in-distribution accuracy

Finetune LLMs on MonsterAPI using Monster Tuner

MonsterAPI · intermediate hands-on

→ Fine-tune LLMs for specific use-cases→ Improve LLM performance using Monster Tuner

Flan-T5 Model Fine-tuning: Advanced Techniques for Professionals

Discover AI · advanced hands-on

→ Fine-tune Flan-T5 models for specific NLP tasks→ Implement task-specific data sets in real-time

NeurIPS Hacker Cup AI: FineTuning

Weights & Biases · advanced hands-on

→ Fine-tune large language models with constrained resources→ Optimize memory usage for LLMs

Model Optimization in Microsoft Foundry: Supervised Fine-Tuning

Microsoft Developer · advanced hands-on

→ Apply Supervised Fine-Tuning→ Use Fine-Tuning for Specific Tasks→ Reduce Token Usage

Read (10 articles)