Megatron, every mention
3 scenes, the whole family · ← back to Megatron
tap a year for its mentions
Verbatim, from the transcripts: the passages where Megatron comes up
[Paper Club] Upcycling Large Language Models into Mixture of Experts
- ▶ 6:58 Ethan He So, um, now, uh, let me introduce a Magtron Core MOE. 6 times in the scene
- ▶ 14:41 Ethan He yeah, you can import Megatron core as a library and just use it in your network, but I, I think there are some caveats.
- ▶ 38:22 Ethan He I, I put the paper links, uh, Megatron Core MOE GitHub there, and, uh, Nemo GitHub, VRVPRA high-level training interface.