mrq f4f435d7f5 when you already had these ideas to stabilize training but you just ignored them		2025-02-27 23:39:20 -06:00
..
ext	made muon actually work by actually utilizing param groups (thanks APOLLO for reminding me this is the sane way to handle this split)	2025-02-26 10:39:13 -06:00
__init__.py
distributed.py
io.py
ml.py	borrowed muon since it might better work under deepspeed and not require cruft (even though it really does not like the masked-NAR, also make the masked-NAR faux-causal since it might better help out for cfg.model.version >= 7	2025-02-23 17:23:24 -06:00
pattern.py
sampler.py
trainer.py	when you already had these ideas to stabilize training but you just ignored them	2025-02-27 23:39:20 -06:00
utils.py	added muon optimizer through kludge hacks because it necessitates a second optimizer in tandum that seems to only sometimes work with deepspeed	2025-02-23 11:22:13 -06:00