How Replit Trained Their Own LLMs - LLM Bootcamp

How Replit Trained Their Own LLMs - LLM Bootcamp

The Full Stack via YouTube Direct link

Data Pipelines: Databricks & Hugging Face

3 of 10

3 of 10

Data Pipelines: Databricks & Hugging Face

Class Central Classrooms beta

YouTube videos curated by Class Central.

Classroom Contents

How Replit Trained Their Own LLMs - LLM Bootcamp

Automatically move to the next video in the Classroom when playback concludes

  1. 1 Why train your own LLMs?
  2. 2 The Modern LLM Stack
  3. 3 Data Pipelines: Databricks & Hugging Face
  4. 4 Preprocessing
  5. 5 Tokenizer Training
  6. 6 Running Training: MosaicML, Weights & Biases
  7. 7 Testing & Evaluation: HumanEval, Hugging Face
  8. 8 Deployment: FasterTransformer, Triton Server, k8s
  9. 9 Lessons learned: data-centrism, eval, and collaboration
  10. 10 What makes a good LLM engineer?

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Someone learning on their laptop while sitting on the floor.