Pretraining

As explained in How Large Language Models Work

Pretraining is the initial training of a model on broad data before adapting it to particular tasks or applications.

Pretraining is a model’s first, larger round of training on broad data. It builds a general-purpose model before fine-tuning adapts that model to a particular task. section 5.0

Broad data gives the model information that a narrowly designed dataset could never anticipate. Web pages and other general documents can provide wide-ranging knowledge, while later training makes the model more useful for a specific purpose. section 5.1

The training pipeline separates general capability from task-specific behavior:

• Curate training data before changing the model’s parameters. section 5.0
• Pretrain on a very large collection to produce a base model or foundation model. section 5.0
• Fine-tune on a much smaller, task-related dataset to constrain the model’s behavior. section 5.1
• Add post-training code when outputs require further control. section 5.0

A car assistant

A car-sales dataset might cover vehicles but miss questions about sports equipment, cleaning, or accessibility. Broad pretraining can supply related information; fine-tuning can then shape the model into a more useful car salesperson. section 5.1

Figure 5.1 One may intervene to change or constrain an LLM’s behavior in four places. The two stages of model training are shown in the middle of the diagram, where the model’s parameters are altered. On the left, one could also alter the training data before model training. On the right, one could intercept the model outputs after model training and write code to handle specific situations.

Use the pipeline diagram to compare the four intervention points: data curation, pretraining, fine-tuning, and output-handling code. section 5.0

How How Large Language Models Work covers it

How deep it goes

Explains the training pipeline and rationale for broad data, but stops short of coding an LLM from scratch.

Examples

Uses web pages, general documents, and a car-salesperson scenario involving sports equipment and accessibility.

What you can do after

Understand how broad pretraining and narrower fine-tuning divide general knowledge from task-specific behavior.

How it connects

Links pretraining to transformers, fine-tuning, foundation models, base models, and training-data curation.

liveBookmanning.com

Read How Large Language Models Work on liveBook Get the book at manning.com

Also explained in

• 

  liveBookmanning.com
  Deep Learning with R, Third EditionUnsupervised pretraining is training a model on input data without manually supplied labels before adapting it to downstream tasks.
• 

  liveBookmanning.com
  Deep Learning with Python, Third EditionThe initial large-scale training phase used to learn broad language representations before task-specific fine-tuning.
• 

  liveBookmanning.com
  Build a Large Language Model (From Scratch)The initial training phase in which a language model learns general patterns from a large text corpus before task-specific adaptation.
• 

  liveBookmanning.com
  Time Series Forecasting Using Foundation ModelsPretraining is initial training of a model on a large dataset to learn general patterns and relationships before applying it to specific tasks.
• 

  liveBookmanning.com
  Inside Deep Learning: Math, Algorithms, ModelsA strategy of training a model on a source dataset before adapting it to a downstream task.
• 

  liveBookmanning.com
  Build a Reasoning Model (From Scratch)The initial training stage in which a language model learns general statistical structure from large amounts of data before task-specific adaptation.
• 

  liveBookmanning.com
  The Art of AI Product DevelopmentPretraining is the initial training of a model on broad data before adapting it to specific tasks or domains.
• 

  liveBookmanning.com
  Introduction to Generative AI, Second EditionThe initial training phase in which a model learns general representations or predictive behavior from a large corpus before task-specific adaptation.
• 

  liveBookmanning.com
  LLMs in ProductionPretraining is the initial training of a model on a large, broad dataset before adapting it to a specific task or domain.
• 

  liveBookmanning.com
  Sutskever's ListPretraining is the initial training of a model on a broad dataset before task-specific adaptation or evaluation.
• 

  liveBookmanning.com
  Data Without LabelsPretraining is an initial model-training stage that learns general patterns or representations before adaptation to a specific task.
• 

  liveBookmanning.com
  Build a Website with ChatGPTPretraining is the initial training of a model on large amounts of data to learn general patterns before task-specific use.
• 

  liveBookmanning.com
  Reinforcement Learning from Human FeedbackPretraining is the initial large-scale optimization of a model on broad data, commonly through next-token prediction.

Compare how 14 books explain pretraining →

Related concepts

• foundational modelproduces · A base or foundation model is a pretrained model that provides learned parameters or capabilities as a starting point for reuse or adaptation to downstream tasks.
• pretrained modelproduces · A pretrained model is a machine-learning model whose parameters were learned through prior training before it is used or adapted for a subsequent task.
• next-token predictionuses · Next-token prediction is the task or training objective of estimating the token that follows a sequence, given its preceding tokens and any conditioning information.
• datasetuses · A dataset is an organized collection of related data items, records, or observations treated as a unit for storage, processing, analysis, or computation.
• zero-shot learningused by · Zero-shot learning is a setting in which a model performs a task or predicts categories without task-specific examples during training, adaptation, or inference, relying on learned knowledge and available task information.
• similarity searchused by · Similarity search is a retrieval technique that finds or ranks items by comparing their representations with a query according to a similarity or distance measure.

This explainer was generated by Manning's knowledge graph from the text of How Large Language Models Work. The section links point to the parts of the book it draws on.