12 Comments
User's avatar
Antonio Zarauz Moreno's avatar

¡Gracias Alberto! El token está eliminado desde que se grabó el vídeo; no obstante, te agradezco mucho el comentario 🙂

Servas Adolph's avatar

The Lab 0 is amazing; I can't wait to be a fine-tuned ninja.

Alberto López Casanova's avatar

Excited about this fine-tuning course! One note about the video: be careful, because even though Antonio’s HF Token is mostly covered, there’s a frame where it can still be seen.

Thank you both for your work! — greetings from Andalucía :)

Miguel Otero Pedrido's avatar

Gracias por el comentario Alberto!! Tenemos que ficharte para el team Security! Menudo ojo tienes jajaj

Adil's avatar
Feb 7Edited

yeaaah budddyyy let's get it 💪. Also, sponsorships! Neural Maze becoming a big player now!

Jorge Martínez Santiago's avatar

Question: Can the retrained model be exported, or does it only reside where it was trained?

Greetings from Vigo

Antonio Zarauz Moreno's avatar

Hola Jorge, las sesiones son efímeras en HF Jobs; por lo tanto, todos los entrenamientos necesitan terminar guardando en tu hub el modelo, o de lo contrario lo perderías.

The AI Architect's avatar

Super solid tech stack overview. The combination of Unsloth + Comet/Opik is really smart for bridgingtraining and evaluation. I've been using Opik for a few weeks and the batch eval workflow is clutch. One thing that stands out: running HF Jobs instead of local enviroment removes so much friction. Makes experimentation way more accessible.

Adil's avatar

bridge training?

Marcelo Acosta Cavalero's avatar

They meant bridging training and evaluation

Adil's avatar

That makes so much more sense!