Fine-tuning LLMs 30x Faster - Unsloth AI's Approach

Fine-tuning LLMs 30x Faster - Unsloth AI's Approach

Aleksa Gordić - The AI Epiphany via YouTube Direct link

01:00:00 - Reproducibility in ML research

10 of 10

10 of 10

01:00:00 - Reproducibility in ML research

Class Central Classrooms beta

YouTube playlists curated by Class Central.

Classroom Contents

Fine-tuning LLMs 30x Faster - Unsloth AI's Approach

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

  1. 1 00:00:00 - Intro
  2. 2 00:01:40 - Hyperstack GPUs! sponsored
  3. 3 00:03:21 - About Daniel, getting started with ML
  4. 4 00:06:30 - Using Desmos to check grads
  5. 5 00:11:00 - Deep dive into Gemma bugs
  6. 6 00:38:00 - approximate GELU bug
  7. 7 00:50:00 - What are the bottlenecks to speeding up the runtime?
  8. 8 00:54:15 - comparison with llm.c?
  9. 9 00:58:30 - Is multi-GPU coming to Unsloth? :
  10. 10 01:00:00 - Reproducibility in ML research

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.