AdamW, every mention
3 scenes across 2 shows · ← back to AdamW
tap a year for its mentions
the a16z Podcast 3
Latent Space 3
every year every show Latent Space 3 the a16z Podcast 3
Verbatim, from the transcripts: passages where AdamW comes up on Latent Space, the a16z Podcast
⚡ Open Model Pretraining Masterclass — Elie Bakouch, HuggingFace SmolLM 3, FineWeb, FinePDF
- ▶ 11:08 unnamed speaker Adam W. 3 times in the scene
The Quest for Community-Trained Open Source AI Models
- ▶ 35:07 Jeff Schmidt So, like, at AMW, like, it just works for all networks. 2 times in the scene
- ▶ 57:31 unnamed speaker So basically, uh, if you think about the usual training of Adam W is that you have all of these GPUs that have a copy of the model, like simply just, let's say the simplest scenario.