Fine-tuning
22 moments from 19 videos · 49:36 min total
Learn how to adapt large models to custom domains. This playlist highlights technical segments focusing on dataset preparation, LoRA, and efficient optimization strategies.
Identifying when narrow tasks require custom model fine-tuning
Determining when to transition from prompt engineering and retrieval-augmented generation to supervised fine-tuning reduces latency and API dependency.
Trade-offs between proprietary, open weight, and fine-tuned models
Navigating model selection requires balancing organizational control, hosting responsibilities, and specific task complexities.
Implementing simple supervised fine-tuning with the Unsloth library
Specific training prompts and the Unsloth library streamline the process of fine-tuning quantized models.
Benchmark limitations and final recommendations for fine-tuning models
Despite evaluation limits, supervised fine-tuning offers clear return on investment for highly specialized sub-agent tasks.
Methods for supervised fine-tuning and reinforcement learning
Data distillation enables supervised fine-tuning while GRPO offers reinforcement learning with verifiable rewards.
Comparing fine-tuning to database retrieval and grounding
Grounding language models via accessible database context offers superior security and efficiency compared to full fine-tuning.
Enhancing models with retrieval augmented generation and fine-tuning
Indexing proprietary documentation for retrieval prevents model hallucinations, while permanent fine-tuning adapts generic models to domain-specific terminology.
Cost considerations of fine-tuning large language models
Fine-tuning requires massive datasets and costly specialized hardware which proves impractical for most organizational use cases.
Instruction fine-tuning and human feedback reinforcement
Models programmatically align toward desired communication metrics after analyzing complex behavioral outcomes guided exclusively via human rewards.
Designing private fine-tuning models for local codebases
Enterprise engineering demands customizable training models that learn from local mistakes without exposing private intellectual property.
Fine-tuning and serving custom AI models
Automation pipelines process specific datasets to refine output behaviors and publish the resulting formats as queryable network endpoints.
Customizing foundation models for specific software domains
Enabling developers to fine-tune base open implementations into specialized variants for targeted tasks.
Fine-tuning image generation models with specific references
Extracting specific artistic styles without massive compute requires training textual inversion techniques on small curated datasets.
Mitigating context window limits with prompt engineering
Refining model inputs to produce targeted responses without relying on extensive fine-tuning.
Adopting fine-tuned task oriented AI models
Small fine-tuned deployment models securely solve predefined functions without demanding the overhead of sprawling general systems.
Demystifying retrieval augmented generation and fine-tuning models
Explaining established architecture patterns helps stakeholders understand when to use context retrieval versus model fine-tuning.
Building and fine-tuning models with the NeMo framework
Leveraging containerized microservices operating on scalable clusters facilitates robust pre-training and custom proprietary data model fine-tuning.
Fine-tuning BERT models for audience sentiment analysis classification
Training standard language representations on domain-specific event text improves sentiment scoring accuracy for non-standard sentences.
Fine-tuning models to answer questions and execute tasks
Supervised learning on prompt and answer pairs transforms a text completion engine into a responsive assistant.
Fine-tuning models and publishing structured brand messaging
How automating message coordination across various platforms builds consistent brand context for artificial intelligence.
Cost-effective LoRA fine-tuning workflows on local inference hardware
Training targeted style representations locally on graphical processors drastically reduces ongoing operational costs.
Preparing training datasets and executing fast model fine-tuning
Providing paired positive and negative examples accelerates deployment without requiring massive datasets or specialized infrastructure.
Your mix. Instantly.
More mixes
AI & Machine Learning
AI coding agents
AI & Machine Learning
MCP (Model Context Protocol)
AI & Machine Learning
RAG
Languages & Frameworks
TypeScript
Languages & Frameworks
React
Languages & Frameworks
Java
Cloud, Platform & Data
Platform engineering
Cloud, Platform & Data
Observability
Cloud, Platform & Data
Infrastructure as code
Architecture, Quality & Security
System architecture
Architecture, Quality & Security
Code review
Architecture, Quality & Security
Software supply chain security
Leadership, Career & Community
Engineering strategy
Leadership, Career & Community
Developer experience
Leadership, Career & Community